linux.git/kernel, branch v3.10.27

sched: Guarantee new group-entities always have weight

2014-01-15T23:28:54+00:00

commit 0ac9b1c21874d2490331233b3242085f8151e166 upstream.

Currently, group entity load-weights are initialized to zero. This
admits some races with respect to the first time they are re-weighted in
earlty use. ( Let g[x] denote the se for "g" on cpu "x". )

Suppose that we have root->a and that a enters a throttled state,
immediately followed by a[0]->t1 (the only task running on cpu[0])
blocking:

  put_prev_task(group_cfs_rq(a[0]), t1)
  put_prev_entity(..., t1)
  check_cfs_rq_runtime(group_cfs_rq(a[0]))
  throttle_cfs_rq(group_cfs_rq(a[0]))

Then, before unthrottling occurs, let a[0]->b[0]->t2 wake for the first
time:

  enqueue_task_fair(rq[0], t2)
  enqueue_entity(group_cfs_rq(b[0]), t2)
  enqueue_entity_load_avg(group_cfs_rq(b[0]), t2)
  account_entity_enqueue(group_cfs_ra(b[0]), t2)
  update_cfs_shares(group_cfs_rq(b[0]))
  < skipped because b is part of a throttled hierarchy >
  enqueue_entity(group_cfs_rq(a[0]), b[0])
  ...

We now have b[0] enqueued, yet group_cfs_rq(a[0])->load.weight == 0
which violates invariants in several code-paths. Eliminate the
possibility of this by initializing group entity weight.

Signed-off-by: Paul Turner 
Signed-off-by: Peter Zijlstra 
Link: http://lkml.kernel.org/r/20131016181627.22647.47543.stgit@sword-of-the-dawn.mtv.corp.google.com
Signed-off-by: Ingo Molnar 
Cc: Chris J Arges 
Signed-off-by: Greg Kroah-Hartman

sched: Fix hrtimer_cancel()/rq->lock deadlock

2014-01-15T23:28:54+00:00

commit 927b54fccbf04207ec92f669dce6806848cbec7d upstream.

__start_cfs_bandwidth calls hrtimer_cancel while holding rq->lock,
waiting for the hrtimer to finish. However, if sched_cfs_period_timer
runs for another loop iteration, the hrtimer can attempt to take
rq->lock, resulting in deadlock.

Fix this by ensuring that cfs_b->timer_active is cleared only if the
_latest_ call to do_sched_cfs_period_timer is returning as idle. Then
__start_cfs_bandwidth can just call hrtimer_try_to_cancel and wait for
that to succeed or timer_active == 1.

Signed-off-by: Ben Segall 
Signed-off-by: Peter Zijlstra 
Cc: pjt@google.com
Link: http://lkml.kernel.org/r/20131016181622.22647.16643.stgit@sword-of-the-dawn.mtv.corp.google.com
Signed-off-by: Ingo Molnar 
Cc: Chris J Arges 
Signed-off-by: Greg Kroah-Hartman

sched: Fix cfs_bandwidth misuse of hrtimer_expires_remaining

2014-01-15T23:28:54+00:00

commit db06e78cc13d70f10877e0557becc88ab3ad2be8 upstream.

hrtimer_expires_remaining does not take internal hrtimer locks and thus
must be guarded against concurrent __hrtimer_start_range_ns (but
returning HRTIMER_RESTART is safe). Use cfs_b->lock to make it safe.

Signed-off-by: Ben Segall 
Signed-off-by: Peter Zijlstra 
Cc: pjt@google.com
Link: http://lkml.kernel.org/r/20131016181617.22647.73829.stgit@sword-of-the-dawn.mtv.corp.google.com
Signed-off-by: Ingo Molnar 
Cc: Chris J Arges 
Signed-off-by: Greg Kroah-Hartman

sched: Fix race on toggling cfs_bandwidth_used

2014-01-15T23:28:54+00:00

commit 1ee14e6c8cddeeb8a490d7b54cd9016e4bb900b4 upstream.

When we transition cfs_bandwidth_used to false, any currently
throttled groups will incorrectly return false from cfs_rq_throttled.
While tg_set_cfs_bandwidth will unthrottle them eventually, currently
running code (including at least dequeue_task_fair and
distribute_cfs_runtime) will cause errors.

Fix this by turning off cfs_bandwidth_used only after unthrottling all
cfs_rqs.

Tested: toggle bandwidth back and forth on a loaded cgroup. Caused
crashes in minutes without the patch, hasn't crashed with it.

Signed-off-by: Ben Segall 
Signed-off-by: Peter Zijlstra 
Cc: pjt@google.com
Link: http://lkml.kernel.org/r/20131016181611.22647.80365.stgit@sword-of-the-dawn.mtv.corp.google.com
Signed-off-by: Ingo Molnar 
Cc: Chris J Arges 
Signed-off-by: Greg Kroah-Hartman

mm: fix TLB flush race between migration, and change_protection_range

2014-01-09T20:24:23+00:00

commit 20841405940e7be0617612d521e206e4b6b325db upstream.

There are a few subtle races, between change_protection_range (used by
mprotect and change_prot_numa) on one side, and NUMA page migration and
compaction on the other side.

The basic race is that there is a time window between when the PTE gets
made non-present (PROT_NONE or NUMA), and the TLB is flushed.

During that time, a CPU may continue writing to the page.

This is fine most of the time, however compaction or the NUMA migration
code may come in, and migrate the page away.

When that happens, the CPU may continue writing, through the cached
translation, to what is no longer the current memory location of the
process.

This only affects x86, which has a somewhat optimistic pte_accessible.
All other architectures appear to be safe, and will either always flush,
or flush whenever there is a valid mapping, even with no permissions
(SPARC).

The basic race looks like this:

CPU A			CPU B			CPU C

						load TLB entry
make entry PTE/PMD_NUMA
			fault on entry
						read/write old page
			start migrating page
			change PTE/PMD to new page
						read/write old page [*]
flush TLB
						reload TLB from new entry
						read/write new page
						lose data

[*] the old page may belong to a new user at this point!

The obvious fix is to flush remote TLB entries, by making sure that
pte_accessible aware of the fact that PROT_NONE and PROT_NUMA memory may
still be accessible if there is a TLB flush pending for the mm.

This should fix both NUMA migration and compaction.

[mgorman@suse.de: fix build]
Signed-off-by: Rik van Riel 
Signed-off-by: Mel Gorman 
Cc: Alex Thorlton 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

sched: fix the theoretical signal_wake_up() vs schedule() race

2014-01-09T20:24:23+00:00

commit e0acd0a68ec7dbf6b7a81a87a867ebd7ac9b76c4 upstream.

This is only theoretical, but after try_to_wake_up(p) was changed
to check p->state under p->pi_lock the code like

	__set_current_state(TASK_INTERRUPTIBLE);
	schedule();

can miss a signal. This is the special case of wait-for-condition,
it relies on try_to_wake_up/schedule interaction and thus it does
not need mb() between __set_current_state() and if(signal_pending).

However, this __set_current_state() can move into the critical
section protected by rq->lock, now that try_to_wake_up() takes
another lock we need to ensure that it can't be reordered with
"if (signal_pending(current))" check inside that section.

The patch is actually one-liner, it simply adds smp_wmb() before
spin_lock_irq(rq->lock). This is what try_to_wake_up() already
does by the same reason.

We turn this wmb() into the new helper, smp_mb__before_spinlock(),
for better documentation and to allow the architectures to change
the default implementation.

While at it, kill smp_mb__after_lock(), it has no callers.

Perhaps we can also add smp_mb__before/after_spinunlock() for
prepare_to_wait().

Signed-off-by: Oleg Nesterov 
Acked-by: Peter Zijlstra 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

libata, freezer: avoid block device removal while system is frozen

2014-01-09T20:24:23+00:00

commit 85fbd722ad0f5d64d1ad15888cd1eb2188bfb557 upstream.

Freezable kthreads and workqueues are fundamentally problematic in
that they effectively introduce a big kernel lock widely used in the
kernel and have already been the culprit of several deadlock
scenarios.  This is the latest occurrence.

During resume, libata rescans all the ports and revalidates all
pre-existing devices.  If it determines that a device has gone
missing, the device is removed from the system which involves
invalidating block device and flushing bdi while holding driver core
layer locks.  Unfortunately, this can race with the rest of device
resume.  Because freezable kthreads and workqueues are thawed after
device resume is complete and block device removal depends on
freezable workqueues and kthreads (e.g. bdi_wq, jbd2) to make
progress, this can lead to deadlock - block device removal can't
proceed because kthreads are frozen and kthreads can't be thawed
because device resume is blocked behind block device removal.

839a8e8660b6 ("writeback: replace custom worker pool implementation
with unbound workqueue") made this particular deadlock scenario more
visible but the underlying problem has always been there - the
original forker task and jbd2 are freezable too.  In fact, this is
highly likely just one of many possible deadlock scenarios given that
freezer behaves as a big kernel lock and we don't have any debug
mechanism around it.

I believe the right thing to do is getting rid of freezable kthreads
and workqueues.  This is something fundamentally broken.  For now,
implement a funny workaround in libata - just avoid doing block device
hot[un]plug while the system is frozen.  Kernel engineering at its
finest.  :(

v2: Add EXPORT_SYMBOL_GPL(pm_freezing) for cases where libata is built
    as a module.

v3: Comment updated and polling interval changed to 10ms as suggested
    by Rafael.

v4: Add #ifdef CONFIG_FREEZER around the hack as pm_freezing is not
    defined when FREEZER is not configured thus breaking build.
    Reported by kbuild test robot.

Signed-off-by: Tejun Heo 
Reported-by: Tomaž Šolc 
Reviewed-by: "Rafael J. Wysocki" 
Link: https://bugzilla.kernel.org/show_bug.cgi?id=62801
Link: http://lkml.kernel.org/r/20131213174932.GA27070@htj.dyndns.org
Cc: Greg Kroah-Hartman 
Cc: Len Brown 
Cc: Oleg Nesterov 
Cc: kbuild test robot 
Signed-off-by: Greg Kroah-Hartman

sched/rt: Fix rq's cpupri leak while enqueue/dequeue child RT entities

2014-01-09T20:24:21+00:00

commit 757dfcaa41844595964f1220f1d33182dae49976 upstream.

This patch touches the RT group scheduling case.

Functions inc_rt_prio_smp() and dec_rt_prio_smp() change (global) rq's
priority, while rt_rq passed to them may be not the top-level rt_rq.
This is wrong, because changing of priority on a child level does not
guarantee that the priority is the highest all over the rq. So, this
leak makes RT balancing unusable.

The short example: the task having the highest priority among all rq's
RT tasks (no one other task has the same priority) are waking on a
throttle rt_rq.  The rq's cpupri is set to the task's priority
equivalent, but real rq->rt.highest_prio.curr is less.

The patch below fixes the problem.

Signed-off-by: Kirill Tkhai 
Signed-off-by: Peter Zijlstra 
CC: Steven Rostedt 
Link: http://lkml.kernel.org/r/49231385567953@web4m.yandex.ru
Signed-off-by: Ingo Molnar 
Signed-off-by: Greg Kroah-Hartman

sched: numa: skip inaccessible VMAs

2014-01-09T20:24:21+00:00

commit 3c67f474558748b604e247d92b55dfe89654c81d upstream.

Inaccessible VMA should not be trapping NUMA hint faults. Skip them.

Signed-off-by: Mel Gorman 
Reviewed-by: Rik van Riel 
Cc: Alex Thorlton 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

ftrace: Initialize the ftrace profiler for each possible cpu

2014-01-09T20:24:20+00:00

commit c4602c1c818bd6626178d6d3fcc152d9f2f48ac0 upstream.

Ftrace currently initializes only the online CPUs. This implementation has
two problems:
- If we online a CPU after we enable the function profile, and then run the
  test, we will lose the trace information on that CPU.
  Steps to reproduce:
  # echo 0 > /sys/devices/system/cpu/cpu1/online
  # cd /tracing/
  # echo  >> set_ftrace_filter
  # echo 1 > function_profile_enabled
  # echo 1 > /sys/devices/system/cpu/cpu1/online
  # run test
- If we offline a CPU before we enable the function profile, we will not clear
  the trace information when we enable the function profile. It will trouble
  the users.
  Steps to reproduce:
  # cd /tracing/
  # echo  >> set_ftrace_filter
  # echo 1 > function_profile_enabled
  # run test
  # cat trace_stat/function*
  # echo 0 > /sys/devices/system/cpu/cpu1/online
  # echo 0 > function_profile_enabled
  # echo 1 > function_profile_enabled
  # cat trace_stat/function*
  # run test
  # cat trace_stat/function*

So it is better that we initialize the ftrace profiler for each possible cpu
every time we enable the function profile instead of just the online ones.

Link: http://lkml.kernel.org/r/1387178401-10619-1-git-send-email-miaox@cn.fujitsu.com

Signed-off-by: Miao Xie 
Signed-off-by: Steven Rostedt 
Signed-off-by: Greg Kroah-Hartman