* pm-sleep:
PM / hibernate: Call platform_leave() in suspend path too
PM / Sleep: Add macro to define common late/early system PM callbacks
PM / hibernate: export hibernation_set_ops
* pm-runtime:
PM / Runtime: Implement the pm_generic_runtime functions for CONFIG_PM
PM / Runtime: Add second macro for definition of runtime PM callbacks
* pm-cpufreq: (28 commits)
cpufreq: speedstep: remove unused speedstep_get_state
intel_pstate: Add X86_FEATURE_APERFMPERF to cpu match parameters.
speedstep-smi: enable interrupts when waiting
powernow-k6: reorder frequencies
powernow-k6: correctly initialize default parameters
powernow-k6: disable cache when changing frequency
Documentation: add ABI entry for intel_pstate
cpufreq: exynos: Convert exynos-cpufreq to platform driver
cpufreq: Make sure CPU is running on a freq from freq-table
cpufreq: Mark ARM drivers with CPUFREQ_NEED_INITIAL_FREQ_CHECK flag
cpufreq: imx6q: add of_init_opp_table
cpufreq: imx6q-cpufreq driver is reused on i.MX6 series SoCs
cpufreq: imx6q: correct VDDSOC/PU voltage scaling when cpufreq is changed
cpufreq: send new set of notification for transition failures
cpufreq: Introduce cpufreq_notify_post_transition()
cpufreq: exynos5250: Set APLL rate using CCF API
cpufreq: Fix timer/workqueue corruption by protecting reading governor_enabled
cpufreq: s3c24xx: Staticize local variable
cpufreq: s3c2440: Staticize local variables
cpufreq: s3c2440: Remove hardware.h inclusion
...
* pm-cpuidle:
cpuidle: remove state_count field from struct cpuidle_device
intel_idle: use the common cpuidle_[un]register() routines
intel_idle: remove superfluous dev->state_count initialization
intel_idle: do C1E promotion disable quirk for hotplugged CPUs
ACPI / cpuidle: remove dev->state_count setting
ACPI / cpuidle: fix max idle state handling with hotplug CPU support
POWERPC: pseries: cpuidle: use the common cpuidle_[un]register() routines
POWERPC: pseries: cpuidle: remove superfluous dev->state_count initialization
ARM: EXYNOS: cpuidle: fix AFTR mode check
* pnp:
PNP: Mark the function pnp_build_option() as static in resource.c
PNP / card: add missing put_device() call
PNPACPI: check return value of pnp_add_device()
* acpi-cleanup: (22 commits)
ACPI / tables: Return proper error codes from acpi_table_parse() and fix comment.
ACPI / tables: Check if id is NULL in acpi_table_parse()
ACPI / proc: Include appropriate header file in proc.c
ACPI / EC: Remove unused functions and add prototype declaration in internal.h
ACPI / dock: Include appropriate header file in dock.c
ACPI / PCI: Include appropriate header file in pci_link.c
ACPI / PCI: Include appropriate header file in pci_slot.c
ACPI / EC: Mark the function acpi_ec_add_debugfs() as static in ec_sys.c
ACPI / NVS: Include appropriate header file in nvs.c
ACPI / OSL: Mark the function acpi_table_checksum() as static
ACPI / processor: initialize a variable to silence compiler warning
ACPI / processor: use ACPI_COMPANION() to get ACPI device
ACPI: correct minor typos
ACPI / sleep: Drop redundant acpi_disabled check
ACPI / dock: Drop redundant acpi_disabled check
ACPI / table: Replace '1' with specific error return values
ACPI: remove trailing whitespace
ACPI / IBFT: Fix incorrect <acpi/acpi.h> inclusion in iSCSI boot firmware module
ACPI / i915: Fix incorrect <acpi/acpi.h> inclusions via <linux/acpi_io.h>
SFI / ACPI: Fix warnings reported during builds with W=1
...
The only caller of speedstep_get_state() was removed in commit d4019f0a92ab
("cpufreq: move freq change notifications to cpufreq core"). So building
speedstep-smi.o now triggers a GCC warning:
drivers/cpufreq/speedstep-smi.c:148:12: warning: 'speedstep_get_state' defined but not used [-Wunused-function]
Remove this unused function.
Signed-off-by: Paul Bolle <pebolle@tiscali.nl> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Dirk Brandewie [Mon, 6 Jan 2014 18:59:16 +0000 (10:59 -0800)]
intel_pstate: Add X86_FEATURE_APERFMPERF to cpu match parameters.
KVM environments do not support APERF/MPERF MSRs. intel_pstate cannot
operate without these registers.
The previous validity checks in intel_pstate_msrs_not_valid() are
insufficent in nested KVMs.
References: https://bugzilla.redhat.com/show_bug.cgi?id=1046317 Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com> Cc: All applicable <stable@vger.kernel.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Mikulas Patocka [Thu, 12 Dec 2013 00:55:40 +0000 (19:55 -0500)]
speedstep-smi: enable interrupts when waiting
On Dell Latitude C600 laptop with Pentium 3 850MHz processor, the
speedstep-smi driver sometimes loads and sometimes doesn't load with
"change to state X failed" message.
I found out that we need to enable interrupts while waiting. When we
enable interrupts, the blockage that prevents frequency transition
resolves and the transition is possible. With disabled interrupts, the
blockage doesn't resolve (no matter how long do we wait).
This patch enables interrupts in the function speedstep_set_state that can
be called with disabled interrupts. However, this function is called with
disabled interrupts only from speedstep_get_freqs, so it shouldn't cause
any problem.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
The powernow-k6 driver used to read the initial multiplier from the
powernow register. However, there is a problem with this:
* If there was a frequency transition before, the multiplier read from the
register corresponds to the current multiplier.
* If there was no frequency transition since reset, the field in the
register always reads as zero, regardless of the current multiplier that
is set using switches on the mainboard and that the CPU is running at.
The zero value corresponds to multiplier 4.5, so as a consequence, the
powernow-k6 driver always assumes multiplier 4.5.
For example, if we have 550MHz CPU with bus frequency 100MHz and
multiplier 5.5, the powernow-k6 driver thinks that the multiplier is 4.5
and bus frequency is 122MHz. The powernow-k6 driver then sets the
multiplier to 4.5, underclocking the CPU to 450MHz, but reports the
current frequency as 550MHz.
There is no reliable way how to read the initial multiplier. I modified
the driver so that it contains a table of known frequencies (based on
parameters of existing CPUs and some common overclocking schemes) and sets
the multiplier according to the frequency. If the frequency is unknown
(because of unusual overclocking or underclocking), the user must supply
the bus speed and maximum multiplier as module parameters.
This patch should be backported to all stable kernels. If it doesn't
apply cleanly, change it, or ask me to change it.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Mikulas Patocka [Thu, 12 Dec 2013 00:38:32 +0000 (19:38 -0500)]
powernow-k6: disable cache when changing frequency
I found out that a system with k6-3+ processor is unstable during network
server load. The system locks up or the network card stops receiving. The
reason for the instability is the CPU frequency scaling.
During frequency transition the processor is in "EPM Stop Grant" state.
The documentation says that the processor doesn't respond to inquiry
requests in this state. Consequently, coherency of processor caches and
bus master devices is not maintained, causing the system instability.
This patch flushes the cache during frequency transition. It fixes the
instability.
Other minor changes:
* u64 invalue changed to unsigned long because the variable is 32-bit
* move the logic to set the multiplier to a separate function
powernow_k6_set_cpu_multiplier
* preserve lower 5 bits of the powernow port instead of 4 (the voltage
field has 5 bits)
* mask interrupts when reading the multiplier, so that the port is not
open during other activity (running other kernel code with the port open
shouldn't cause any misbehavior, but we should better be safe and keep
the port closed)
This patch should be backported to all stable kernels. If it doesn't
apply cleanly, change it, or ask me to change it.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Add a Documentation/ABI entry for
/sys/devices/system/cpu/intel_pstate/max_perf_pct,
/sys/devices/system/cpu/intel_pstate/min_perf_pct, and
/sys/devices/system/cpu/intel_pstate/no_turbo.
Cc: Dirk Brandewie <dirk.brandewie@gmail.com> Cc: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Lukasz Majewski [Thu, 28 Nov 2013 12:42:42 +0000 (13:42 +0100)]
cpufreq: exynos: Convert exynos-cpufreq to platform driver
To make the driver multiplatform-friendly, unconditional initialization
in an initcall is replaced with a platform driver probed only if
respective platform device is registered.
Tested at: Exynos4210 (TRATS) and Exynos4412 (TRATS2)
Signed-off-by: Lukasz Majewski <l.majewski@samsung.com> Signed-off-by: Tomasz Figa <t.figa@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com> Reviewed-by: Sachin Kamat <sachin.kamat@linaro.org> Tested-by: Sachin Kamat <sachin.kamat@linaro.org> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Viresh Kumar [Tue, 3 Dec 2013 05:50:46 +0000 (11:20 +0530)]
cpufreq: Make sure CPU is running on a freq from freq-table
Sometimes boot loaders set CPU frequency to a value outside of frequency table
present with cpufreq core. In such cases CPU might be unstable if it has to run
on that frequency for long duration of time and so its better to set it to a
frequency which is specified in freq-table. This also makes cpufreq stats
inconsistent as cpufreq-stats would fail to register because current frequency
of CPU isn't found in freq-table.
Because we don't want this change to affect boot process badly, we go for the
next freq which is >= policy->cur ('cur' must be set by now, otherwise we will
end up setting freq to lowest of the table as 'cur' is initialized to zero).
In case current frequency doesn't match any frequency from freq-table, we throw
warnings to user, so that user can get this fixed in their bootloaders or
freq-tables.
Reported-by: Carlos Hernandez <ceh@ti.com> Reported-and-tested-by: Nishanth Menon <nm@ti.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Viresh Kumar [Tue, 3 Dec 2013 05:50:45 +0000 (11:20 +0530)]
cpufreq: Mark ARM drivers with CPUFREQ_NEED_INITIAL_FREQ_CHECK flag
Sometimes boot loaders set CPU frequency to a value outside of frequency table
present with cpufreq core. In such cases CPU might be unstable if it has to run
on that frequency for long duration of time and so its better to set it to a
frequency which is specified in frequency table.
On some systems we can't really say what frequency we're running at the moment
and so for these we shouldn't check if we are running at a frequency present in
frequency table. And so we really can't force this for all the cpufreq drivers.
Hence we are created another flag here: CPUFREQ_NEED_INITIAL_FREQ_CHECK that
will be marked by platforms which want to go for this check at boot time.
Initially this is done for all ARM platforms but others may follow if required.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Bjørn Mork [Wed, 4 Dec 2013 15:06:58 +0000 (16:06 +0100)]
PM / hibernate: Call platform_leave() in suspend path too
Since create_image() only executes platform_leave() if in_suspend is
not set, enable_nonboot_cpus() is run by it with EC transactions
blocked (on ACPI systems) in the image creation code path (that is,
for in_suspend set), which may cause CPU online to fail for the CPUs
in question. In particular, this causes the acpi_cpufreq driver's
initialization to fail for those CPUs on some systems with the
following dmesg:
To fix this problem, modify create_image() to execute platform_leave()
unconditionally. [rjw: This shouldn't lead to any significant side
effects on ACPI systems.]
Signed-off-by: Bjørn Mork <bjorn@mork.no>
[rjw: Changelog] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
This tool is designed to assist kernel and OS developers in optimizing
their linux stack's suspend/resume time. Using a kernel image built with a
few extra options enabled, the tool will execute a suspend and will
capture dmesg and ftrace data until resume is complete. This data is
transformed into a device timeline and a callgraph to give a quick and
detailed view of which devices and callbacks are taking the most time in
suspend/resume. The output is a single html file which can be viewed in
firefox or chrome.
References: https://01.org/suspendresume
Signed-off-by: Todd Brandt <todd.e.brandt@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
John Tobias [Fri, 20 Dec 2013 06:56:28 +0000 (22:56 -0800)]
cpufreq: imx6q: add of_init_opp_table
Add a routine check to see if the platform supplied the OPP table.
Incase there's no OPP table exist, it will try to initialise it.
It's been tested on iMX6SL board where the platform doesn't have
an OPP table.
Signed-off-by: John Tobias <john.tobias.ph@gmail.com> Acked-by: Shawn Guo <shawn.guo@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Shawn Guo [Fri, 20 Dec 2013 02:12:16 +0000 (10:12 +0800)]
cpufreq: imx6q-cpufreq driver is reused on i.MX6 series SoCs
The imx6q-cpufreq driver nowadays is not only running on imx6q but also
other i.MX6 series SoCs like imx6dl and imx6sl. Update Kconfig prompt
and help text to make it clear to users.
Signed-off-by: Shawn Guo <shawn.guo@linaro.org> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Anson Huang [Thu, 19 Dec 2013 14:16:47 +0000 (09:16 -0500)]
cpufreq: imx6q: correct VDDSOC/PU voltage scaling when cpufreq is changed
on i.MX6Q, cpu freq change need to follow below flows:
1. each setpoint has different VDDARM, VDDSOC/PU voltage, get the setpoint
table from dts;
2. when cpu freq is scaling up, need to increase VDDSOC/PU voltage before
VDDARM, if VDDPU is off, no need to change it;
3. when cpu freq is scaling down, need to decrease VDDARM voltage before
VDDSOC/PU, if VDDPU is off, no need to change it;
normally dts will pass vddsoc/pu freq/volt info to kernel, if not, will
use fixed value for vddsoc/pu voltage setting.
Signed-off-by: Anson Huang <b20788@freescale.com> Acked-by: Shawn Guo <shawn.guo@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
cpuidle: remove state_count field from struct cpuidle_device
dev->state_count is now always equal to drv->state_count so
it can be removed.
Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
intel_idle: use the common cpuidle_[un]register() routines
It is now possible to use the common cpuidle_[un]register() routines
(instead of open-coding them) so do it.
Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com> Cc: Len Brown <lenb@kernel.org> Reviewed-by: Daniel Lezcano <daniel.lezcano@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
intel_idle driver sets dev->state_count to drv->state_count so
the default dev->state_count initialization in cpuidle_enable_device()
(called from cpuidle_register_device()) can be used instead.
Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com> Cc: Len Brown <lenb@kernel.org> Reviewed-by: Daniel Lezcano <daniel.lezcano@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
intel_idle: do C1E promotion disable quirk for hotplugged CPUs
If the system is booted with some CPUs offline C1E promotion disable quirk
won't be applied because on_each_cpu() in intel_idle_cpuidle_driver_init()
operates only on online CPUs. Fix it by adding the C1E promotion disable
handling to intel_idle_cpu_init() (which is also called during CPU_ONLINE
operation).
Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com> Cc: Len Brown <lenb@kernel.org> Reviewed-by: Daniel Lezcano <daniel.lezcano@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
dev->state_count is now always equal to drv->state_count and
drv->state_count no longer can change during driver's lifetime so
the default dev->state_count initialization in cpuidle_enable_device()
(called from cpuidle_register_device()) can be used instead.
Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com> Cc: Len Brown <lenb@kernel.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
ACPI / cpuidle: fix max idle state handling with hotplug CPU support
acpi_processor_hotplug() calls acpi_processor_setup_cpuidle_cx()
without calling acpi_processor_setup_cpuidle_states() first so it
is possible that dev->state_count becomes different from
drv->state_count (in case of SMP system with unsupported C2/C3
states + enabled CPU hotplug and num_online_cpus() becoming > 1).
The driver code assumes that cpuidle core will handle such cases
but currently this is untrue (dev->state_count is used only for
handling cpuidle state sysfs entries and drv->state_count is used
for all other cases) and will not be fixed in the future as
dev->state_count is planned to be removed.
Fix the issue by checking for the max supported idle state in
C2/C3 state's ->enter handler (acpi_idle_enter_simple() for C2/C3
and acpi_idle_enter_bm() for C3 + bm_check flag set) and setting
the C1 state (instead of higher states) when needed.
Also remove no longer needed max idle state checks from
acpi_processor_setup_cpuidle_[states,cx]().
Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com> Cc: Len Brown <lenb@kernel.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
pseries cpuidle driver sets dev->state_count to drv->state_count so
the default dev->state_count initialization in cpuidle_enable_device()
(called from cpuidle_register_device()) can be used instead.
Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Deepthi Dharwar <deepthi@linux.vnet.ibm.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
The EXYNOS cpuidle driver code assumes that cpuidle core will handle
dev->state_count smaller than drv->state_count but currently this is
untrue (dev->state_count is used only for handling cpuidle state sysfs
entries and drv->state_count is used for all other cases) and will not
be fixed in the future as dev->state_count is planned to be removed.
Fix the issue by checking for the max supported idle state in AFTR
state's ->enter handler (exynos4_enter_lowpower()) and entering AFTR
mode only when cores other than CPU0 are offline.
Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com> Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org> Acked-by: Kukjin Kim <kgene.kim@samsung.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
tangchen [Mon, 6 Jan 2014 08:47:59 +0000 (16:47 +0800)]
ACPI / tables: Return proper error codes from acpi_table_parse() and fix comment.
The comment about return value of acpi_table_parse() is incorrect.
This patch fix it.
Since all callers only check if the function succeeded or not, this
patch simplifies the semantics by returning -errno for all failure
cases. This will also simply the comment.
As suggested by Toshi Kani <toshi.kani@hp.com>, also change the stub
in linux/acpi.h to return -ENODEV.
Signed-off-by: Tang Chen <tangchen@cn.fujitsu.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Viresh Kumar [Mon, 2 Dec 2013 05:34:13 +0000 (11:04 +0530)]
cpufreq: send new set of notification for transition failures
In the current code, if we fail during a frequency transition, we
simply send the POSTCHANGE notification with the old frequency. This
isn't enough.
One of the core users of these notifications is the code responsible
for keeping loops_per_jiffy aligned with frequency changes. And mostly
it is written as:
So, suppose we are changing to a higher frequency and failed during
transition, then following will happen:
- CPUFREQ_PRECHANGE notification with freq-new > freq-old
- CPUFREQ_POSTCHANGE notification with freq-new == freq-old
The first one will update loops_per_jiffy and second one will do
nothing. Even if we send the 2nd notification by exchanging values of
freq-new and old, some users of these notifications might get
unstable.
This can be fixed by simply calling cpufreq_notify_post_transition()
with error code and this routine will take care of sending
notifications in the correct order.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
[rjw: Folded 3 patches into one, rebased unicore2 changes] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
This introduces a new routine cpufreq_notify_post_transition() which
can be used to send POSTCHANGE notification for new freq with or
without both {PRE|POST}CHANGE notifications for last freq. This is
useful at multiple places, especially for sending transition failure
notifications.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Sachin Kamat [Tue, 24 Dec 2013 10:05:24 +0000 (15:35 +0530)]
cpufreq: exynos5250: Set APLL rate using CCF API
Use common clock framework (CCF) APIs to set the clock rates
instead of direct register manipulation. This now updates the
sysfs entry (cpuinfo_cur_freq) correctly which did not reflect
the correct value until now. While at it clean up the PLL s-div
parameter setting as it is handled by the PLL driver.
Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Reviewed-by: Lukasz Majewski <l.majewski@samsung.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Jane Li [Fri, 3 Jan 2014 09:17:41 +0000 (17:17 +0800)]
cpufreq: Fix timer/workqueue corruption by protecting reading governor_enabled
When a CPU is hot removed we'll cancel all the delayed work items via
gov_cancel_work(). Sometimes the delayed work function determines that
it should adjust the delay for all other CPUs that the policy is
managing. If this scenario occurs, the canceling CPU will cancel its own
work but queue up the other CPUs works to run.
Commit 3617f2 (cpufreq: Fix timer/workqueue corruption due to double
queueing) has tried to fix this, but reading governor_enabled is not
protected by cpufreq_governor_lock. Even though od_dbs_timer() checks
governor_enabled before gov_queue_work(), this scenario may occur. For
example:
CPU0 CPU1
---- ----
cpu_down()
... <work runs>
__cpufreq_remove_dev() od_dbs_timer()
__cpufreq_governor() policy->governor_enabled
policy->governor_enabled = false;
cpufreq_governor_dbs()
case CPUFREQ_GOV_STOP:
gov_cancel_work(dbs_data, policy);
cpu0 work is canceled
timer is canceled
cpu1 work is canceled
<waits for cpu1>
gov_queue_work(*, *, true);
cpu0 work queued
cpu1 work queued
cpu2 work queued
...
cpu1 work is canceled
cpu2 work is canceled
...
At the end of the GOV_STOP case cpu0 still has a work queued to
run although the code is expecting all of the works to be
canceled. __cpufreq_remove_dev() will then proceed to
re-initialize all the other CPUs works except for the CPU that is
going down. The CPUFREQ_GOV_START case in cpufreq_governor_dbs()
will trample over the queued work and debugobjects will spit out
a warning:
WARNING: at lib/debugobjects.c:260 debug_print_object+0x94/0xbc()
ODEBUG: init active (active state 0) object type: timer_list hint: delayed_work_timer_fn+0x0/0x14
Modules linked in:
CPU: 1 PID: 1205 Comm: sh Tainted: G W 3.10.0 #200
[<c01144f0>] (unwind_backtrace+0x0/0xf8) from [<c0111d98>] (show_stack+0x10/0x14)
[<c0111d98>] (show_stack+0x10/0x14) from [<c01272cc>] (warn_slowpath_common+0x4c/0x68)
[<c01272cc>] (warn_slowpath_common+0x4c/0x68) from [<c012737c>] (warn_slowpath_fmt+0x30/0x40)
[<c012737c>] (warn_slowpath_fmt+0x30/0x40) from [<c034c640>] (debug_print_object+0x94/0xbc)
[<c034c640>] (debug_print_object+0x94/0xbc) from [<c034c7f8>] (__debug_object_init+0xc8/0x3c0)
[<c034c7f8>] (__debug_object_init+0xc8/0x3c0) from [<c01360e0>] (init_timer_key+0x20/0x104)
[<c01360e0>] (init_timer_key+0x20/0x104) from [<c04872ac>] (cpufreq_governor_dbs+0x1dc/0x68c)
[<c04872ac>] (cpufreq_governor_dbs+0x1dc/0x68c) from [<c04833a8>] (__cpufreq_governor+0x80/0x1b0)
[<c04833a8>] (__cpufreq_governor+0x80/0x1b0) from [<c0483704>] (__cpufreq_remove_dev.isra.12+0x22c/0x380)
[<c0483704>] (__cpufreq_remove_dev.isra.12+0x22c/0x380) from [<c0692f38>] (cpufreq_cpu_callback+0x48/0x5c)
[<c0692f38>] (cpufreq_cpu_callback+0x48/0x5c) from [<c014fb40>] (notifier_call_chain+0x44/0x84)
[<c014fb40>] (notifier_call_chain+0x44/0x84) from [<c012ae44>] (__cpu_notify+0x2c/0x48)
[<c012ae44>] (__cpu_notify+0x2c/0x48) from [<c068dd40>] (_cpu_down+0x80/0x258)
[<c068dd40>] (_cpu_down+0x80/0x258) from [<c068df40>] (cpu_down+0x28/0x3c)
[<c068df40>] (cpu_down+0x28/0x3c) from [<c068e4c0>] (store_online+0x30/0x74)
[<c068e4c0>] (store_online+0x30/0x74) from [<c03a7308>] (dev_attr_store+0x18/0x24)
[<c03a7308>] (dev_attr_store+0x18/0x24) from [<c0256fe0>] (sysfs_write_file+0x100/0x180)
[<c0256fe0>] (sysfs_write_file+0x100/0x180) from [<c01fec9c>] (vfs_write+0xbc/0x184)
[<c01fec9c>] (vfs_write+0xbc/0x184) from [<c01ff034>] (SyS_write+0x40/0x68)
[<c01ff034>] (SyS_write+0x40/0x68) from [<c010e200>] (ret_fast_syscall+0x0/0x48)
In gov_queue_work(), lock cpufreq_governor_lock before gov_queue_work,
and unlock it after __gov_queue_work(). In this way, governor_enabled
is guaranteed not changed in gov_queue_work().
Signed-off-by: Jane Li <jiel@marvell.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Reviewed-by: Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Viresh Kumar [Fri, 3 Jan 2014 10:21:39 +0000 (15:51 +0530)]
cpufreq: arm-big-little: Make driver dependent on CONFIG_BIG_LITTLE
The arm_big_little cpufreq driver is only used by ARM bigLITTLE
platforms and hence must depend on CONFIG_BIG_LITTLE.
This was highlighted by Russell earlier when he reported this issue:
drivers/built-in.o: In function `bL_cpufreq_set_rate':
powercap_sys.c:(.text+0x5ed9a0): undefined reference to `bL_switch_request_cb'
Reported-by: Russell King <linux@arm.linux.org.uk> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Rashika Kheria [Sat, 14 Dec 2013 13:20:56 +0000 (18:50 +0530)]
PNP: Mark the function pnp_build_option() as static in resource.c
This patch marks the function pnp_build_option() as static in resource.c
because it is not used outside this file.
Thus, it also eliminates the following warning in resource.c:
drivers/pnp/resource.c:34:20: warning: no previous prototype for ‘pnp_build_option’ [-Wmissing-prototypes]
Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com> Reviewed-by: Josh Triplett <josh@joshtriplett.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Rashika [Tue, 17 Dec 2013 09:34:17 +0000 (15:04 +0530)]
ACPI / proc: Include appropriate header file in proc.c
Include appropriate header file internal.h in proc.c because function
acpi_sleep_proc_init() has its prototype declaration in internal.h.
This eliminates the following warning in proc.c:
drivers/acpi/proc.c:148:12: warning: no previous prototype for ‘acpi_sleep_proc_init’ [-Wmissing-prototypes]
Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com> Reviewed-by: Josh Triplett <josh@joshtriplett.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Rashika [Tue, 17 Dec 2013 09:32:14 +0000 (15:02 +0530)]
ACPI / EC: Remove unused functions and add prototype declaration in internal.h
Adds the prototype declarations of functions acpi_ec_add_query_handler()
and acpi_ec_remove_query_handler() in header file internal.h and removes
unused functions ec_burst_enable() and ec_burst_disable() in ec.c.
This eliminates the following warnings in ec.c:
drivers/acpi/ec.c:393:5: warning: no previous prototype for ‘ec_burst_enable’ [-Wmissing-prototypes]
drivers/acpi/ec.c:402:5: warning: no previous prototype for ‘ec_burst_disable’ [-Wmissing-prototypes]
drivers/acpi/ec.c:531:5: warning: no previous prototype for ‘acpi_ec_add_query_handler’ [-Wmissing-prototypes]
drivers/acpi/ec.c:552:6: warning: no previous prototype for ‘acpi_ec_remove_query_handler’ [-Wmissing-prototypes]
Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com> Reviewed-by: Josh Triplett <josh@joshtriplett.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Rashika [Tue, 17 Dec 2013 09:30:04 +0000 (15:00 +0530)]
ACPI / dock: Include appropriate header file in dock.c
Includes appropriate header file internal.h in dock.c because
function acpi_dock_init() has its prototype declaration in
internal.h.
This eliminates the following warning in dock.c:
drivers/acpi/dock.c:899:13: warning: no previous prototype for ‘acpi_dock_init’ [-Wmissing-prototypes]
Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com> Reviewed-by: Josh Triplett <josh@joshtriplett.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Rashika [Tue, 17 Dec 2013 09:28:31 +0000 (14:58 +0530)]
ACPI / PCI: Include appropriate header file in pci_link.c
Includes appropriate header file internal.h in pci_link.c
because function acpi_pci_link_init() has its prototype declaration in
internal.h.
This eliminates the following warning in pci_link.c:
drivers/acpi/pci_link.c:874:13: warning: no previous prototype for ‘acpi_pci_link_init’ [-Wmissing-prototypes]
Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com> Reviewed-by: Josh Triplett <josh@joshtriplett.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Rashika [Tue, 17 Dec 2013 09:18:19 +0000 (14:48 +0530)]
ACPI / PCI: Include appropriate header file in pci_slot.c
Includes appropriate header file linux/pci-acpi.h in pci_slot.c
because functions acpi_pci_slot_enumerate(), acpi_pci_slot_remove()
and acpi_pci_slot_init() have their prototype declaratons in
linux/pci-acpi.h.
This eliminates the following warnings in pci_slot.c:
drivers/acpi/pci_slot.c:162:6: warning: no previous prototype for ‘acpi_pci_slot_enumerate’ [-Wmissing-prototypes]
drivers/acpi/pci_slot.c:174:6: warning: no previous prototype for ‘acpi_pci_slot_remove’ [-Wmissing-prototypes]
drivers/acpi/pci_slot.c:215:13: warning: no previous prototype for ‘acpi_pci_slot_init’ [-Wmissing-prototypes]
Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com> Reviewed-by: Josh Triplett <josh@joshtriplett.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Rashika [Tue, 17 Dec 2013 09:16:41 +0000 (14:46 +0530)]
ACPI / EC: Mark the function acpi_ec_add_debugfs() as static in ec_sys.c
Mark the function acpi_ec_add_debugfs() as static in
ec_sys.c because it is not used outside this file.
This eliminates the following warning in ec_sys.c:
drivers/acpi/ec_sys.c:108:5: warning: no previous prototype for ‘acpi_ec_add_debugfs’ [-Wmissing-prototypes]
Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com> Reviewed-by: Josh Triplett <josh@joshtriplett.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Rashika [Tue, 17 Dec 2013 09:15:00 +0000 (14:45 +0530)]
ACPI / NVS: Include appropriate header file in nvs.c
Include header file internal.h in nvs.c because functions
suspend_nvs_free(), suspend_nvs_alloc(), suspend_nvs_save() and
suspend_nvs_restore() have their prototype declaration in internal.h.
This eliminates the following warnings in nvs.c:
drivers/acpi/nvs.c:128:6: warning: no previous prototype for ‘suspend_nvs_free’ [-Wmissing-prototypes]
drivers/acpi/nvs.c:152:5: warning: no previous prototype for ‘suspend_nvs_alloc’ [-Wmissing-prototypes]
drivers/acpi/nvs.c:169:5: warning: no previous prototype for ‘suspend_nvs_save’ [-Wmissing-prototypes]
drivers/acpi/nvs.c:201:6: warning: no previous prototype for ‘suspend_nvs_restore’ [-Wmissing-prototypes]
Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com> Reviewed-by: Josh Triplett <josh@joshtriplett.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Rashika [Tue, 17 Dec 2013 09:13:05 +0000 (14:43 +0530)]
ACPI / OSL: Mark the function acpi_table_checksum() as static
Marks the function acpi_table_checksum() as static in osl.c
because it is not used outside this file.
This eliminates the following warning in osl.c:
drivers/acpi/osl.c:547:11: warning: no previous prototype for
‘acpi_table_checksum’ [-Wmissing-prototypes]
Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com> Reviewed-by: Josh Triplett <josh@joshtriplett.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Jiang Liu [Thu, 19 Dec 2013 12:47:46 +0000 (20:47 +0800)]
ACPI / extlog: replace open-coded _DSM code with helper functions
Use helper functions to simplify _DSM related code in acpi_extlog driver.
Also mark initialization data and functions with __init and __initdata
to reduce memory footprint.
Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com> Tested-by: Chen, Gong <gong.chen@linux.intel.com> Reviewed-by: Chen, Gong <gong.chen@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Jiang Liu [Thu, 19 Dec 2013 12:38:22 +0000 (20:38 +0800)]
ACPI / nouveau: replace open-coded _DSM code with helper functions
Use helper functions to simplify _DSM related code in nouveau driver.
After analyzing the ACPI _DSM related code, I changed nouveau_optimus_dsm()
to expect a buffer and nouveau_dsm() to expect an integer only.
Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Jiang Liu [Thu, 19 Dec 2013 12:38:20 +0000 (20:38 +0800)]
ACPI / i915: replace open-coded _DSM code with helper functions
Use helper functions to simplify _DSM related code in i915 driver.
Function intel_dsm() is used to check functions supported by ACPI _DSM
method, but it has strange check for special value 0x80000002. After
digging into nouveau driver, I think the check is copied from nouveau
driver and is useless for i915 driver, so remove it.
Acked-by: Daniel Vetter <daniel@ffwll.ch> Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Jiang Liu [Thu, 19 Dec 2013 12:38:18 +0000 (20:38 +0800)]
ACPI / TPM: detect PPI features by checking availability of _DSM functions
Detecting physical presence interface features by checking availbility
of corresponding ACPI _DSM functions, it should be more accurate than
checking TPM version number.
Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Jiang Liu [Thu, 19 Dec 2013 12:38:17 +0000 (20:38 +0800)]
ACPI / TPM: replace open-coded _DSM code with helper functions
Use helper functions to simplify _DSM related code in TPM driver.
This patch also help to get rid of following warning messages:
[ 163.509575] ACPI Error: Incorrect return type [Buffer] requested [Package]
(20130517/nsxfeval-135)
But there is still an warning left.
[ 181.637366] ACPI Warning: \_SB_.IIO0.LPC0.TPM_._DSM: Argument #4 type
mismatch - Found [Buffer], ACPI requires [Package] (20130517/nsarguments-95)
Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Jiang Liu [Thu, 19 Dec 2013 12:38:14 +0000 (20:38 +0800)]
PCI / pci-label: treat PCI label with index 0 as valid label
Current pci-label driver detects ACPI label by checking label index
returned by ACPI _DSM method, and treats it as valid if label index
is positive. According to ACPI Firmware specification 3.1, zero is
also an valid label index. So change code to detect availability of
ACPI slot label by checking availaiblity of ACPI _DSM function for
PCI label.
Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Jiang Liu [Thu, 19 Dec 2013 12:38:10 +0000 (20:38 +0800)]
ACPI: introduce helper interfaces for _DSM method
There are several drivers making use of ACPI _DSM method to detect
and invoke device specific methods. Currently every driver has
implemented its private version to support ACPI _DSM method.
So this patch introduces three helper functions to support ACPI _DSM
method, which will be used to replace open-coded versions.
It helps to simplify code and improve code readability.
Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Jiang Liu [Thu, 19 Dec 2013 12:38:15 +0000 (20:38 +0800)]
ACPI / TPM: fix memory leak when walking ACPI namespace
In function ppi_callback(), memory allocated by acpi_get_name() will get
leaked when current device isn't the desired TPM device, so fix the
memory leak.
Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com> Cc: All applicable <stable@vger.kernel.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Alexander Mezin [Sat, 4 Jan 2014 02:07:59 +0000 (09:07 +0700)]
ACPI / AC: change notification handler type to ACPI_ALL_NOTIFY
With kernel 3.13rc5 there are no AC adapter notifications on my laptop.
Commit cc8ef5270734 "ACPI / AC: convert ACPI ac driver to platform bus"
changed the driver to listen to device notifications only. However, AML
code on my laptop notifies the driver with zero event.
This patch changes the driver to listen to all events again.
Fixes: cc8ef5270734 (ACPI / AC: convert ACPI ac driver to platform bus)
References: https://bugzilla.kernel.org/show_bug.cgi?id=67821 Signed-off-by: Alexander Mezin <mezin.alexander@gmail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Linus Torvalds [Fri, 3 Jan 2014 21:48:25 +0000 (13:48 -0800)]
Merge tag 'for-v3.13-fixes' of git://git.infradead.org/battery-2.6
Pull battery fixes from Anton Vorontsov:
"Two fixes:
- fix build error caused by max17042_battery conversion to the regmap
API.
- fix kernel oops when booting with wakeup_source_activate enabled"
* tag 'for-v3.13-fixes' of git://git.infradead.org/battery-2.6:
max17042_battery: Fix build errors caused by missing REGMAP_I2C config
power_supply: Fix Oops from NULL pointer dereference from wakeup_source_activate
Linus Torvalds [Fri, 3 Jan 2014 21:44:41 +0000 (13:44 -0800)]
Merge tag 'pm+acpi-3.13-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull ACPI and PM fixes and new device IDs from Rafael Wysocki:
"These commits, except for one, are regression fixes and the remaining
one fixes a divide error leading to a kernel panic. The majority of
the regressions fixed here were introduced during the 3.12 cycle, one
of them is from this cycle and one is older.
Specifics:
- VGA switcheroo was broken for some users as a result of the
ACPI-based PCI hotplug (ACPIPHP) changes in 3.12, because some
previously ignored hotplug events started to be handled. The fix
causes them to be ignored again.
- There are two more issues related to cpufreq's suspend/resume
handling changes from the 3.12 cycle addressed by Viresh Kumar's
fixes.
- intel_pstate triggers a divide error in a timer function if the
P-state information it needs is missing during initialization.
This leads to kernel panics on nested KVM clients and is fixed by
failing the initialization cleanly in those cases.
- PCI initalization code changes during the 3.9 cycle uncovered BIOS
issues related to ACPI wakeup notifications (some BIOSes send them
for devices that aren't supposed to support ACPI wakeup). Work
around them by installing an ACPI wakeup notify handler for all PCI
devices with ACPI support.
- The Calxeda cpuilde driver's probe function is tagged as __init,
which is incorrect and causes a section mismatch to occur during
build. Fix from Andre Przywara removes the __init tag from there.
- During the 3.12 cycle ACPIPHP started to print warnings about
missing _ADR for devices that legitimately don't have it. Fix from
Toshi Kani makes it only print the warnings where they make sense"
* tag 'pm+acpi-3.13-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
ACPIPHP / radeon / nouveau: Fix VGA switcheroo problem related to hotplug
intel_pstate: Fail initialization if P-state information is missing
ARM/cpuidle: remove __init tag from Calxeda cpuidle probe function
PCI / ACPI: Install wakeup notify handlers for all PCI devs with ACPI
cpufreq: preserve user_policy across suspend/resume
cpufreq: Clean up after a failing light-weight initialization
ACPI / PCI / hotplug: Avoid warning when _ADR not present
Linus Torvalds [Thu, 2 Jan 2014 22:40:38 +0000 (14:40 -0800)]
Merge branch 'akpm' (incoming from Andrew)
Merge patches from Andrew Morton:
"Ten fixes"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
epoll: do not take the nested ep->mtx on EPOLL_CTL_DEL
sh: add EXPORT_SYMBOL(min_low_pfn) and EXPORT_SYMBOL(max_low_pfn) to sh_ksyms_32.c
drivers/dma/ioat/dma.c: check DMA mapping error in ioat_dma_self_test()
mm/memory-failure.c: transfer page count from head page to tail page after split thp
MAINTAINERS: set up proper record for Xilinx Zynq
mm: remove bogus warning in copy_huge_pmd()
memcg: fix memcg_size() calculation
mm: fix use-after-free in sys_remap_file_pages
mm: munlock: fix deadlock in __munlock_pagevec()
mm: munlock: fix a bug where THP tail page is encountered
Jason Baron [Thu, 2 Jan 2014 20:58:54 +0000 (12:58 -0800)]
epoll: do not take the nested ep->mtx on EPOLL_CTL_DEL
The EPOLL_CTL_DEL path of epoll contains a classic, ab-ba deadlock.
That is, epoll_ctl(a, EPOLL_CTL_DEL, b, x), will deadlock with
epoll_ctl(b, EPOLL_CTL_DEL, a, x). The deadlock was introduced with
commmit 67347fe4e632 ("epoll: do not take global 'epmutex' for simple
topologies").
The acquistion of the ep->mtx for the destination 'ep' was added such
that a concurrent EPOLL_CTL_ADD operation would see the correct state of
the ep (Specifically, the check for '!list_empty(&f.file->f_ep_links')
However, by simply not acquiring the lock, we do not serialize behind
the ep->mtx from the add path, and thus may perform a full path check
when if we had waited a little longer it may not have been necessary.
However, this is a transient state, and performing the full loop
checking in this case is not harmful.
The important point is that we wouldn't miss doing the full loop
checking when required, since EPOLL_CTL_ADD always locks any 'ep's that
its operating upon. The reason we don't need to do lock ordering in the
add path, is that we are already are holding the global 'epmutex'
whenever we do the double lock. Further, the original posting of this
patch, which was tested for the intended performance gains, did not
perform this additional locking.
Signed-off-by: Jason Baron <jbaron@akamai.com> Cc: Nathan Zimmer <nzimmer@sgi.com> Cc: Eric Wong <normalperson@yhbt.net> Cc: Nelson Elhage <nelhage@nelhage.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Davide Libenzi <davidel@xmailserver.org> Cc: "Paul E. McKenney" <paulmck@us.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
sh: add EXPORT_SYMBOL(min_low_pfn) and EXPORT_SYMBOL(max_low_pfn) to sh_ksyms_32.c
Min_low_pfn and max_low_pfn were used in pfn_valid macro if defined
CONFIG_FLATMEM. When the functions that use the pfn_valid is used in
driver module, max_low_pfn and min_low_pfn is to undefined, and fail to
build.
Naoya Horiguchi [Thu, 2 Jan 2014 20:58:51 +0000 (12:58 -0800)]
mm/memory-failure.c: transfer page count from head page to tail page after split thp
Memory failures on thp tail pages cause kernel panic like below:
mce: [Hardware Error]: Machine check events logged
MCE exception done on CPU 7
BUG: unable to handle kernel NULL pointer dereference at 0000000000000058
IP: [<ffffffff811b7cd1>] dequeue_hwpoisoned_huge_page+0x131/0x1e0
PGD bae42067 PUD ba47d067 PMD 0
Oops: 0000 [#1] SMP
...
CPU: 7 PID: 128 Comm: kworker/7:2 Tainted: G M O 3.13.0-rc4-131217-1558-00003-g83b7df08e462 #25
...
Call Trace:
me_huge_page+0x3e/0x50
memory_failure+0x4bb/0xc20
mce_process_work+0x3e/0x70
process_one_work+0x171/0x420
worker_thread+0x11b/0x3a0
? manage_workers.isra.25+0x2b0/0x2b0
kthread+0xe4/0x100
? kthread_create_on_node+0x190/0x190
ret_from_fork+0x7c/0xb0
? kthread_create_on_node+0x190/0x190
...
RIP dequeue_hwpoisoned_huge_page+0x131/0x1e0
CR2: 0000000000000058
The reasoning of this problem is shown below:
- when we have a memory error on a thp tail page, the memory error
handler grabs a refcount of the head page to keep the thp under us.
- Before unmapping the error page from processes, we split the thp,
where page refcounts of both of head/tail pages don't change.
- Then we call try_to_unmap() over the error page (which was a tail
page before). We didn't pin the error page to handle the memory error,
this error page is freed and removed from LRU list.
- We never have the error page on LRU list, so the first page state
check returns "unknown page," then we move to the second check
with the saved page flag.
- The saved page flag have PG_tail set, so the second page state check
returns "hugepage."
- We call me_huge_page() for freed error page, then we hit the above panic.
The root cause is that we didn't move refcount from the head page to the
tail page after split thp. So this patch suggests to do this.
This panic was introduced by commit 524fca1e73 ("HWPOISON: fix
misjudgement of page_action() for errors on mlocked pages"). Note that we
did have the same refcount problem before this commit, but it was just
ignored because we had only first page state check which returned "unknown
page." The commit changed the refcount problem from "doesn't work" to
"kernel panic."
This warning was introduced by "mm: numa: Avoid unnecessary disruption
of NUMA hinting during migration" for paranoia reasons but the warning
is bogus. I was thinking of parallel races between NUMA hinting faults
and forks but this warning would also be triggered by a parallel reclaim
splitting a THP during a fork. Remote the bogus warning.
Signed-off-by: Mel Gorman <mgorman@suse.de> Reported-by: Sasha Levin <sasha.levin@oracle.com> Cc: Alex Thorlton <athorlton@sgi.com> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Rik van Riel [Thu, 2 Jan 2014 20:58:46 +0000 (12:58 -0800)]
mm: fix use-after-free in sys_remap_file_pages
remap_file_pages calls mmap_region, which may merge the VMA with other
existing VMAs, and free "vma". This can lead to a use-after-free bug.
Avoid the bug by remembering vm_flags before calling mmap_region, and
not trying to dereference vma later.
Signed-off-by: Rik van Riel <riel@redhat.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Cc: PaX Team <pageexec@freemail.hu> Cc: Kees Cook <keescook@chromium.org> Cc: Michel Lespinasse <walken@google.com> Cc: Cyrill Gorcunov <gorcunov@openvz.org> Cc: Hugh Dickins <hughd@google.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Vlastimil Babka [Thu, 2 Jan 2014 20:58:44 +0000 (12:58 -0800)]
mm: munlock: fix deadlock in __munlock_pagevec()
Commit 7225522bb429 ("mm: munlock: batch non-THP page isolation and
munlock+putback using pagevec" introduced __munlock_pagevec() to speed
up munlock by holding lru_lock over multiple isolated pages. Pages that
fail to be isolated are put_page()d immediately, also within the lock.
This can lead to deadlock when __munlock_pagevec() becomes the holder of
the last page pin and put_page() leads to __page_cache_release() which
also locks lru_lock. The deadlock has been observed by Sasha Levin
using trinity.
This patch avoids the deadlock by deferring put_page() operations until
lru_lock is released. Another pagevec (which is also used by later
phases of the function is reused to gather the pages for put_page()
operation.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz> Reported-by: Sasha Levin <sasha.levin@oracle.com> Cc: Michel Lespinasse <walken@google.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Rik van Riel <riel@redhat.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Hugh Dickins <hughd@google.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Vlastimil Babka [Thu, 2 Jan 2014 20:58:43 +0000 (12:58 -0800)]
mm: munlock: fix a bug where THP tail page is encountered
Since commit ff6a6da60b89 ("mm: accelerate munlock() treatment of THP
pages") munlock skips tail pages of a munlocked THP page. However, when
the head page already has PageMlocked unset, it will not skip the tail
pages.
Commit 7225522bb429 ("mm: munlock: batch non-THP page isolation and
munlock+putback using pagevec") has added a PageTransHuge() check which
contains VM_BUG_ON(PageTail(page)). Sasha Levin found this triggered
using trinity, on the first tail page of a THP page without PageMlocked
flag.
This patch fixes the issue by skipping tail pages also in the case when
PageMlocked flag is unset. There is still a possibility of race with
THP page split between clearing PageMlocked and determining how many
pages to skip. The race might result in former tail pages not being
skipped, which is however no longer a bug, as during the skip the
PageTail flags are cleared.
However this race also affects correctness of NR_MLOCK accounting, which
is to be fixed in a separate patch.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz> Reported-by: Sasha Levin <sasha.levin@oracle.com> Cc: Michel Lespinasse <walken@google.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Rik van Riel <riel@redhat.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Hugh Dickins <hughd@google.com> Cc: Bob Liu <bob.liu@oracle.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linus Torvalds [Thu, 2 Jan 2014 20:45:47 +0000 (12:45 -0800)]
Merge tag 'gfs2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-3.0-fixes
Pull GFS2 fixes from Steven Whitehouse:
"Here is a set of small fixes for GFS2. There is a fix to drop
s_umount which is copied in from the core vfs, two patches relate to a
hard to hit "use after free" and memory leak. Two patches related to
using DIO and buffered I/O on the same file to ensure correct
operation in relation to glock state changes. The final patch adds an
RCU read lock to ensure correct locking on an error path"
* tag 'gfs2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-3.0-fixes:
GFS2: Fix unsafe dereference in dump_holder()
GFS2: Wait for async DIO in glock state changes
GFS2: Fix incorrect invalidation for DIO/buffered I/O
GFS2: Fix slab memory leak in gfs2_bufdata
GFS2: Fix use-after-free race when calling gfs2_remove_from_ail
GFS2: don't hold s_umount over blkdev_put
Linus Torvalds [Thu, 2 Jan 2014 20:45:07 +0000 (12:45 -0800)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Pull s390 fixes from Martin Schwidefsky:
"Two small bug fixes and a follow-up to the CONFIG_NR_CPUS change.
A kernel compiled with CONFIG_NR_CPUS=256 will waste quite a bit of
memory for the per-cpu arrays. Under z/VM the maximum number of CPUs
is 64, the code now limits the possible cpu mask to 64 if running
under z/VM"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:
s390/pci: obtain function handle in hotplug notifier
s390/3270: fix allocation of tty3270_screen structure
s390/smp: improve setup of possible cpu mask
Jan Kiszka [Sat, 28 Dec 2013 15:31:52 +0000 (16:31 +0100)]
KVM: nVMX: Unconditionally uninit the MMU on nested vmexit
Three reasons for doing this: 1. arch.walk_mmu points to arch.mmu anyway
in case nested EPT wasn't in use. 2. this aligns VMX with SVM. But 3. is
most important: nested_cpu_has_ept(vmcs12) queries the VMCS page, and if
one guest VCPU manipulates the page of another VCPU in L2, we may be
fooled to skip over the nested_ept_uninit_mmu_context, leaving mmu in
nested state. That can crash the host later on if nested_ept_get_cr3 is
invoked while L1 already left vmxon and nested.current_vmcs12 became
NULL therefore.
Cc: stable@kernel.org Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Linus Torvalds [Wed, 1 Jan 2014 19:36:16 +0000 (11:36 -0800)]
Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
Pull radeon drm fixes from Dave Airlie:
"Just piping a bunch of fixes from pre-xmas from Alex for radeon, all
either fix bad hw setup issues or regressions"
* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
drm/radeon: Bump version for CIK DCE tiling fix
drm/radeon: set correct number of banks for CIK chips in DCE
drm/radeon: set correct pipe config for Hawaii in DCE
drm/radeon: expose render backend mask to the userspace
drm/radeon: fix render backend setup for SI and CIK
drm/radeon: 0x9649 is SUMO2 not SUMO
drm/radeon: fix UVD 256MB check
Dave Airlie [Wed, 1 Jan 2014 10:32:19 +0000 (20:32 +1000)]
Merge branch 'drm-fixes-3.13' of git://people.freedesktop.org/~agd5f/linux into drm-fixes
Radeon fixes, Christmas eve edition. Fix incorrect family for 0x9649
which lead to bogus rendering, tiling and RB fixes for SI and CIK,
and a UVD fix.
* 'drm-fixes-3.13' of git://people.freedesktop.org/~agd5f/linux:
drm/radeon: Bump version for CIK DCE tiling fix
drm/radeon: set correct number of banks for CIK chips in DCE
drm/radeon: set correct pipe config for Hawaii in DCE
drm/radeon: expose render backend mask to the userspace
drm/radeon: fix render backend setup for SI and CIK
drm/radeon: 0x9649 is SUMO2 not SUMO
drm/radeon: fix UVD 256MB check
Krzysztof Hałasa [Tue, 31 Dec 2013 10:51:16 +0000 (11:51 +0100)]
crypto: ixp4xx - Fix kernel compile error
drivers/crypto/ixp4xx_crypto.c: In function 'ixp_module_init':
drivers/crypto/ixp4xx_crypto.c:1419:2: error: 'dev' undeclared (first use in this function)
Now builds. Not tested on real hw.
Signed-off-by: Krzysztof Hałasa <khalasa@piap.pl> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* pm-cpufreq:
intel_pstate: Fail initialization if P-state information is missing
cpufreq: preserve user_policy across suspend/resume
cpufreq: Clean up after a failing light-weight initialization
* pm-cpuidle:
ARM/cpuidle: remove __init tag from Calxeda cpuidle probe function
Merge branches 'acpi-pci-pm' and 'acpi-pci-hotplug'
* acpi-pci-pm:
PCI / ACPI: Install wakeup notify handlers for all PCI devs with ACPI
* acpi-pci-hotplug:
ACPIPHP / radeon / nouveau: Fix VGA switcheroo problem related to hotplug
ACPI / PCI / hotplug: Avoid warning when _ADR not present
Linus Torvalds [Tue, 31 Dec 2013 20:19:30 +0000 (12:19 -0800)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
Pull input updates from Dmitry Torokhov:
"A fix for a panic in gpio-keys driver when set up with absolute
events, a fixup to the new zforce driver and a new keycode definition"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
Input: allocate absinfo data when setting ABS capability
Input: define KEY_WWAN for Wireless WAN
Input: zforce - fix possible driver hang during suspend
Linus Torvalds [Tue, 31 Dec 2013 20:17:14 +0000 (12:17 -0800)]
Merge branch 'for-linus' of git://git.samba.org/sfrench/cifs-2.6
Pull cifs fixes from Steve French:
"A few small cifs fixes including two for stable, and fixing a
regression introduced by the VFS change to file create"
* 'for-linus' of git://git.samba.org/sfrench/cifs-2.6:
cifs: set FILE_CREATED
cifs: We do not drop reference to tlink in CIFSCheckMFSymlink()
Add missing end of line termination to some cifs messages
projects / karo-tx-linux.git / log