Below is a patch from android kernel that maintains a histogram of
suspend times. Please review and provide feedback.

Statistices on the time spent in suspend are kept in
/sys/kernel/debug/sleep_time.

Cc: Android Kernel Team <kernel-team@android.com>
Cc: Colin Cross <ccross@android.com>
Cc: Todd Poynor <toddpoynor@google.com>
Cc: San Mehat <san@google.com>
Cc: Benoit Goby <benoit@android.com>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Colin Cross <ccross@android.com>
Signed-off-by: Todd Poynor <toddpoynor@google.com>
[zoran.markovic@linaro.org: Re-formatted suspend time table to better
fit expected values. Moved accounting of suspend time into timekeeping
core. Removed CONFIG_SUSPEND_TIME flag and made the feature conditional
on CONFIG_DEBUG_FS. Changed the file name to sleep_time to better fit
terminology in timekeeping core. Changed seq_printf to seq_puts. Tweaked
commit message]
Signed-off-by: Zoran Markovic <zoran.markovic@linaro.org>
Signed-off-by: John Stultz <john.stultz@linaro.org>

Add support for clocks CLOCK_REALTIME_ALARM and CLOCK_BOOTTIME_ALARM,
thereby enabling wakeup alarm timers via file descriptors.
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>

Add functions needed for hooking up alarmtimer to timerfd:

* alarm_restart: Similar to hrtimer_restart, restart an alarmtimer after
  the expires time has already been updated (as with alarm_forward).

* alarm_forward_now: Similar to hrtimer_forward_now, move the expires
  time forward to an interval from the current time of the associated clock.

* alarm_start_relative: Start an alarmtimer with an expires time relative to
  the current time of the associated clock.

* alarm_expires_remaining: Similar to hrtimer_expires_remaining, return the
  amount of time remaining until alarm expiry.
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>

All the virtualized platforms (KVM, lguest and Xen) have persistent
wallclocks that have more than one second of precision.

read_persistent_wallclock() and update_persistent_wallclock() allow
for nanosecond precision but their implementation on x86 with
x86_platform.get/set_wallclock() only allows for one second precision.
This means guests may see a wallclock time that is off by up to 1
second.

Make set_wallclock() and get_wallclock() take a struct timespec
parameter (which allows for nanosecond precision) so KVM and Xen
guests may start with a more accurate wallclock time and a Xen dom0
can maintain a more accurate wallclock for guests.
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>

The time.h header seems not to be used by current code.
Removing this include allows the driver to build on other
architecture that do not have this header.

Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Jamie Iles <jamie@jamieiles.com>
Cc: Dinh Nguyen <dinguyen@altera.com>
Acked-by: Jamie Iles <jamie@jamieiles.com>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
[tweaked commit message and header]
Signed-off-by: John Stultz <john.stultz@linaro.org>

It seems we made a mistake when creating dw_apb_timer_of.c:
picoxcell sched_clock had parts that were not related to
dw_apb_timer, yet we moved them to dw_apb_timer_of, and tried to
use them on socfpga.

This results in system where user/system time is not measured
properly, as demonstrated by

    time dd if=/dev/urandom of=/dev/zero bs=100000 count=100

So this patch switches sched_clock to hardware that exists on both
platforms, and adds missing of_node_put() in dw_apb_timer_init().
Signed-off-by: Pavel Machek <pavel@denx.de>
Acked-by: Jamie Iles <jamie@jamieiles.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>

The function is currently mainly used in the networking code and
if others start using it, they must check the result, otherwise
it cannot be determined if the timespec conversion suceeded.
Currently no user lacks this check, but make future users aware of
a possible misusage.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>

We've got the macro NSEC_PER_USEC defined in header file
include/linux/time.h. To make the code decent, this patch
replaces the immediate number 1000 to convert bewteen a
time value in microseconds and one in nanoseconds with the
macro NSEC_PER_USEC.
Signed-off-by: Liu Ying <Ying.Liu@freescale.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Daniel Borkmann <dborkman@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Signed-off-by: John Stultz <john.stultz@linaro.org>

commit 7eaeb343 (clocksource: Provide unbind interface in sysfs)
implemented clocksource_select_fallback() which is not defined for
CONFIG_ARCH_USES_GETTIMEOFFSET=y. Add an empty inline function for
that.
Reported-by: Ingo Molnar <mingo@kernel.org>
Reported-by: fengguang.wu@intel.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

Unbreak architectures which do not use clockevents, but require to
build some of the core timekeeping infrastructure
Reported-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

Provide a sysfs interface to allow unbinding of clockevent
devices. The device is unbound if it is unused or if there is a
replacement device available. Unbinding of broadcast devices is not
supported as we don't want to foster that nonsense. If no replacement
device is available the unbind returns -EBUSY. Unbind is available
from the kernel and through sysfs, which is necessary to drop the
module refcount.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130425143436.499216659@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

Split out the clockevent device selection logic. Preparatory patch to
allow unbinding active clockevent devices.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130425143436.431796247@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

Provide a simple sysfs interface for the clockevent devices. Show the
current active clockevent device.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130425143436.371634778@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

We want to be able to remove clockevent modules as well. Add a
refcount so we don't remove a module with an active clock event
device.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130425143436.307435149@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

No need to call another function and have duplicated cases.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130425143436.235746557@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

Now that the notifier chain is gone there are no other users and it's
pointless to nest tick_device_lock inside of clockevents_lock because
there is no other use case.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130425143436.162888472@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

7+ years and still a single user. Kill it.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130425143436.098520211@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

The unregister call can fail, if the clocksource is the current one
and there is no replacement clocksource available. It can also fail,
if the clocksource is the watchdog clocksource and I'm not going to
provide support for this.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130425143436.029915527@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

With the module refcount held for the current clocksource there is no
way to unload the module. 

Provide a sysfs interface which allows to unbind the clocksource. One
could argue that the clocksource override could be (ab)used to do so,
but the clocksource override cannot be used from the kernel itself,
while an unbind function can be used to programmatically check whether
a clocksource can be shutdown or not.

The unbind functionality uses the new skip current feature of
clocksource_select and verifies that a fallback clocksource has been
installed. If the clocksource which should be unbound is the current
clocksource and no fallback can be found, unbind returns -EBUSY.

This does not support the unbinding of a clocksource which is used as
the watchdog clocksource. No point in fostering crappy hardware.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130425143435.964218245@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

Split out the user string input for clocksource override. Preparatory
patch for unbind.

[ jstultz: Fix an off by one error ]
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130425143435.895851338@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

Preparatory patch for clocksource unbind support.

Split out code from clocksource_select and modify it, so it skips the
current clocksource on request and tries to find a fallback
clocksource. Convert all existing users. No functional change.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130425143435.834965397@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

Add a module refcount, so the current clocksource cannot be removed
unconditionally.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130425143435.762417789@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

timekeeping_notify() can fail due cs->enable() failure. Though the
caller does not notice and happily keeps the wrong clocksource as the
current one.

Let the caller know about failure, so the current clocksource will be
shown correctly in sysfs.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <john.stultz@linaro.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130425143435.696321912@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

If a clocksource has a (wrong) high rating, but can't be used as a
timebase for oneshot tick mode, it is unconditionally selected even
when the system is already in oneshot tick mode. This causes full
system failure.

Verify the clocksource selection against the oneshot mode.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <john.stultz@linaro.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20130425143435.635040849@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <john.stultz@linaro.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Acked-by: Jamie Iles <jamie@jamieiles.com>
Link: http://lkml.kernel.org/r/20130425143435.558006195@linutronix.deSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

Merge tag 'trace-fixes-v3.10-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace

Pull tracing fixes from Steven Rostedt:
 "This includes a fix to a memory leak when adding filters to traces.

  Also, Masami Hiramatsu fixed up some minor bugs that were discovered
  by sparse."

* tag 'trace-fixes-v3.10-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  tracing/kprobes: Make print_*probe_event static
  tracing/kprobes: Fix a sparse warning for incorrect type in assignment
  tracing/kprobes: Use rcu_dereference_raw for tp->files
  tracing: Fix leaks of filter preds

Pull x86 fixes from Thomas Gleixner:

 - Fix for a CPU hot-add deadlock in microcode update code

 - Fix for idle consolidation fallout

 - Documentation update for initial kernel direct mapping

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/mm: Add missing comments for initial kernel direct mapping
  x86/microcode: Add local mutex to fix physical CPU hot-add deadlock
  x86: Fix idle consolidation fallout

Pull perf fixes from Thomas Gleixner:

 - Fix for a task exit cleanup race caused by a missing a preempt
   disable

 - Cleanup of the event notification functions with a massive reduction
   of duplicated code

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf: Factor out auxiliary events notification
  perf: Fix EXIT event notification

Pull timer fixes from Thomas Gleixner:

 - Cure for not using zalloc in the first place, which leads to random
   crashes with CPUMASK_OFF_STACK.

 - Revert a user space visible change which broke udev

 - Add a missing cpu_online early return introduced by the new full
   dyntick conversions

 - Plug a long standing race in the timer wheel cpu hotplug code.
   Sigh...

 - Cleanup NOHZ per cpu data on cpu down to prevent stale data on cpu
   up.

* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  time: Revert ALWAYS_USE_PERSISTENT_CLOCK compile time optimizaitons
  timer: Don't reinitialize the cpu base lock during CPU_UP_PREPARE
  tick: Don't invoke tick_nohz_stop_sched_tick() if the cpu is offline
  tick: Cleanup NOHZ per cpu data on cpu down
  tick: Use zalloc_cpumask_var for allocating offstack cpumasks

Pull core fixes from Thomas Gleixner:

 - Two fixlets for the fallout of the generic idle task conversion

 - Documentation update

* 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  rcu/idle: Wrap cpu-idle poll mode within rcu_idle_enter/exit
  idle: Fix hlt/nohlt command-line handling in new generic idle
  kthread: Document ways of reducing OS jitter due to per-CPU kthreads

Pull ARM fixes from Russell King:
 "A small number of fixes for stuff from the last merge window, and in
  one case (IRQ time accounting) the previous merge window."

* 'fixes' of git://git.linaro.org/people/rmk/linux-arm:
  ARM: 7720/1: ARM v6/v7 cmpxchg64 shouldn't clear upper 32 bits of the old/new value
  ARM: 7715/1: MCPM: adapt to GIC changes after upstream merge
  ARM: 7714/1: mmc: mmci: Ensure return value of regulator_enable() is checked
  ARM: 7712/1: Remove trailing whitespace in arch/arm/Makefile
  ARM: 7711/1: dove: fix Dove cpu type from V7 to PJ4
  ARM: finally enable IRQ time accounting config

Pull Ceph fixes from Sage Weil:
 "Yes, this is a much larger pull than I would like after -rc1.  There
  are a few things included:

   - a few fixes for leaks and incorrect assertions
   - a few patches fixing behavior when mapped images are resized
   - handling for cloned/layered images that are flattened out from
     underneath the client

  The last bit was non-trivial, and there is some code movement and
  associated cleanup mixed in.  This was ready and was meant to go in
  last week but I missed the boat on Friday.  My only excuse is that I
  was waiting for an all clear from the testing and there were many
  other shiny things to distract me.

  Strictly speaking, handling the flatten case isn't a regression and
  could wait, so if you like we can try to pull the series apart, but
  Alex and I would much prefer to have it all in as it is a case real
  users will hit with 3.10."

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client: (33 commits)
  rbd: re-submit flattened write request (part 2)
  rbd: re-submit write request for flattened clone
  rbd: re-submit read request for flattened clone
  rbd: detect when clone image is flattened
  rbd: reference count parent requests
  rbd: define parent image request routines
  rbd: define rbd_dev_unparent()
  rbd: don't release write request until necessary
  rbd: get parent info on refresh
  rbd: ignore zero-overlap parent
  rbd: support reading parent page data for writes
  rbd: fix parent request size assumption
  libceph: init sent and completed when starting
  rbd: kill rbd_img_request_get()
  rbd: only set up watch for mapped images
  rbd: set mapping read-only flag in rbd_add()
  rbd: support reading parent page data
  rbd: fix an incorrect assertion condition
  rbd: define rbd_dev_v2_header_info()
  rbd: get rid of trivial v1 header wrappers
  ...

According to sparse warning, print_*probe_event static because
those functions are not directly called from outside.

Link: http://lkml.kernel.org/r/20130513115839.6545.83067.stgit@mhiramat-M0-7522

Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Tom Zanussi <tom.zanussi@intel.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

Fix a sparse warning about the rcu operated pointer is
defined without __rcu address space.

Link: http://lkml.kernel.org/r/20130513115837.6545.23322.stgit@mhiramat-M0-7522

Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Tom Zanussi <tom.zanussi@intel.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

Use rcu_dereference_raw() for accessing tp->files. Because the
write-side uses rcu_assign_pointer() for memory barrier,
the read-side also has to use rcu_dereference_raw() with
read memory barrier.

Link: http://lkml.kernel.org/r/20130513115834.6545.17022.stgit@mhiramat-M0-7522

Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Tom Zanussi <tom.zanussi@intel.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

Special preds are created when folding a series of preds that
can be done in serial. These are allocated in an ops field of
the pred structure. But they were never freed, causing memory
leaks.

This was discovered using the kmemleak checker:

unreferenced object 0xffff8800797fd5e0 (size 32):
  comm "swapper/0", pid 1, jiffies 4294690605 (age 104.608s)
  hex dump (first 32 bytes):
    00 00 01 00 03 00 05 00 07 00 09 00 0b 00 0d 00  ................
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<ffffffff814b52af>] kmemleak_alloc+0x73/0x98
    [<ffffffff8111ff84>] kmemleak_alloc_recursive.constprop.42+0x16/0x18
    [<ffffffff81120e68>] __kmalloc+0xd7/0x125
    [<ffffffff810d47eb>] kcalloc.constprop.24+0x2d/0x2f
    [<ffffffff810d4896>] fold_pred_tree_cb+0xa9/0xf4
    [<ffffffff810d3781>] walk_pred_tree+0x47/0xcc
    [<ffffffff810d5030>] replace_preds.isra.20+0x6f8/0x72f
    [<ffffffff810d50b5>] create_filter+0x4e/0x8b
    [<ffffffff81b1c30d>] ftrace_test_event_filter+0x5a/0x155
    [<ffffffff8100028d>] do_one_initcall+0xa0/0x137
    [<ffffffff81afbedf>] kernel_init_freeable+0x14d/0x1dc
    [<ffffffff814b24b7>] kernel_init+0xe/0xdb
    [<ffffffff814d539c>] ret_from_fork+0x7c/0xb0
    [<ffffffffffffffff>] 0xffffffffffffffff

Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: stable@vger.kernel.org # 2.6.39+
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

Kay Sievers noted that the ALWAYS_USE_PERSISTENT_CLOCK config,
which enables some minor compile time optimization to avoid
uncessary code in mostly the suspend/resume path could cause
problems for userland.

In particular, the dependency for RTC_HCTOSYS on
!ALWAYS_USE_PERSISTENT_CLOCK, which avoids setting the time
twice and simplifies suspend/resume, has the side effect
of causing the /sys/class/rtc/rtcN/hctosys flag to always be
zero, and this flag is commonly used by udev to setup the
/dev/rtc symlink to /dev/rtcN, which can cause pain for
older applications.

While the udev rules could use some work to be less fragile,
breaking userland should strongly be avoided. Additionally
the compile time optimizations are fairly minor, and the code
being optimized is likely to be reworked in the future, so
lets revert this change.
Reported-by: Kay Sievers <kay@vrfy.org>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: stable <stable@vger.kernel.org> #3.9
Cc: Feng Tang <feng.tang@intel.com>
Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
Link: http://lkml.kernel.org/r/1366828376-18124-1-git-send-email-john.stultz@linaro.orgSigned-off-by: Thomas Gleixner <tglx@linutronix.de>

Pull ext4 update from Ted Ts'o:
 "Fixed regressions (two stability regressions and a performance
  regression) introduced during the 3.10-rc1 merge window.

  Also included is a bug fix relating to allocating blocks after
  resizing an ext3 file system when using the ext4 file system driver"

* tag 'ext4_for_linus_stable' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  jbd,jbd2: fix oops in jbd2_journal_put_journal_head()
  ext4: revert "ext4: use io_end for multiple bios"
  ext4: limit group search loop for non-extent files
  ext4: fix fio regression

Pull workqueue fix from Tejun Heo:
 "A fix for a workqueue_congested() regression that broke fscache"

* 'for-3.10-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
  workqueue: workqueue_congested() shouldn't translate WORK_CPU_UNBOUND into node number

An inactive timer's base can refer to a offline cpu's base.

In the current code, cpu_base's lock is blindly reinitialized each
time a CPU is brought up. If a CPU is brought online during the period
that another thread is trying to modify an inactive timer on that CPU
with holding its timer base lock, then the lock will be reinitialized
under its feet. This leads to following SPIN_BUG().

<0> BUG: spinlock already unlocked on CPU#3, kworker/u:3/1466
<0> lock: 0xe3ebe000, .magic: dead4ead, .owner: kworker/u:3/1466, .owner_cpu: 1
<4> [<c0013dc4>] (unwind_backtrace+0x0/0x11c) from [<c026e794>] (do_raw_spin_unlock+0x40/0xcc)
<4> [<c026e794>] (do_raw_spin_unlock+0x40/0xcc) from [<c076c160>] (_raw_spin_unlock+0x8/0x30)
<4> [<c076c160>] (_raw_spin_unlock+0x8/0x30) from [<c009b858>] (mod_timer+0x294/0x310)
<4> [<c009b858>] (mod_timer+0x294/0x310) from [<c00a5e04>] (queue_delayed_work_on+0x104/0x120)
<4> [<c00a5e04>] (queue_delayed_work_on+0x104/0x120) from [<c04eae00>] (sdhci_msm_bus_voting+0x88/0x9c)
<4> [<c04eae00>] (sdhci_msm_bus_voting+0x88/0x9c) from [<c04d8780>] (sdhci_disable+0x40/0x48)
<4> [<c04d8780>] (sdhci_disable+0x40/0x48) from [<c04bf300>] (mmc_release_host+0x4c/0xb0)
<4> [<c04bf300>] (mmc_release_host+0x4c/0xb0) from [<c04c7aac>] (mmc_sd_detect+0x90/0xfc)
<4> [<c04c7aac>] (mmc_sd_detect+0x90/0xfc) from [<c04c2504>] (mmc_rescan+0x7c/0x2c4)
<4> [<c04c2504>] (mmc_rescan+0x7c/0x2c4) from [<c00a6a7c>] (process_one_work+0x27c/0x484)
<4> [<c00a6a7c>] (process_one_work+0x27c/0x484) from [<c00a6e94>] (worker_thread+0x210/0x3b0)
<4> [<c00a6e94>] (worker_thread+0x210/0x3b0) from [<c00aad9c>] (kthread+0x80/0x8c)
<4> [<c00aad9c>] (kthread+0x80/0x8c) from [<c000ea80>] (kernel_thread_exit+0x0/0x8)

As an example, this particular crash occurred when CPU #3 is executing
mod_timer() on an inactive timer whose base is refered to offlined CPU
#2.  The code locked the timer_base corresponding to CPU #2. Before it
could proceed, CPU #2 came online and reinitialized the spinlock
corresponding to its base. Thus now CPU #3 held a lock which was
reinitialized. When CPU #3 finally ended up unlocking the old cpu_base
corresponding to CPU #2, we hit the above SPIN_BUG().

CPU #0		CPU #3				       CPU #2
------		-------				       -------
.....		 ......				      <Offline>
		mod_timer()
		 lock_timer_base
		   spin_lock_irqsave(&base->lock)

cpu_up(2)	 .....				        ......
							init_timers_cpu()
....		 .....				    	spin_lock_init(&base->lock)
.....		   spin_unlock_irqrestore(&base->lock)  ......
		   <spin_bug>

Allocation of per_cpu timer vector bases is done only once under
"tvec_base_done[]" check. In the current code, spinlock_initialization
of base->lock isn't under this check. When a CPU is up each time the
base lock is reinitialized. Move base spinlock initialization under
the check.
Signed-off-by: Tirupathi Reddy <tirupath@codeaurora.org>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/1368520142-4136-1-git-send-email-tirupath@codeaurora.orgSigned-off-by: Thomas Gleixner <tglx@linutronix.de>