linux-stable.git/kernel/sched, branch linux-5.14.y

Revert "sched: Add wrapper for get_wchan() to keep task blocked"

2021-11-18T13:01:34+00:00

This reverts commit e9ede14c116f1a6246eee89d320d60a90a86b5d5 which is
commit 42a20f86dc19f9282d974df0ba4d226c865ab9dd upstream.

It has been reported to be causing problems, and is being reworked
upstream and has been dropped from the current 5.15.y stable queue until
it gets resolved.

Reported-by: Chris Rankin 
Reported-by: Thorsten Leemhuis 
Link: https://lore.kernel.org/r/ed000478-2a60-0066-c337-a04bffc112b1@leemhuis.info
Cc: Stephen Rothwell 
Cc: Peter Zijlstra (Intel) 
Cc: Kees Cook 
Cc: Geert Uytterhoeven 
Cc: Russell King (Oracle) 
Cc: Mark Rutland 
Cc: Sasha Levin 
Signed-off-by: Greg Kroah-Hartman

sched: Add wrapper for get_wchan() to keep task blocked

2021-11-17T10:04:16+00:00

[ Upstream commit 42a20f86dc19f9282d974df0ba4d226c865ab9dd ]

Having a stable wchan means the process must be blocked and for it to
stay that way while performing stack unwinding.

Suggested-by: Peter Zijlstra 
Signed-off-by: Kees Cook 
Signed-off-by: Peter Zijlstra (Intel) 
Acked-by: Geert Uytterhoeven 
Acked-by: Russell King (Oracle)  [arm]
Tested-by: Mark Rutland  [arm64]
Link: https://lkml.kernel.org/r/20211008111626.332092234@infradead.org
Signed-off-by: Sasha Levin

kernel/sched: Fix sched_fork() access an invalid sched_task_group

2021-11-17T10:04:06+00:00

[ Upstream commit 4ef0c5c6b5ba1f38f0ea1cedad0cad722f00c14a ]

There is a small race between copy_process() and sched_fork()
where child->sched_task_group point to an already freed pointer.

	parent doing fork()      | someone moving the parent
				 | to another cgroup
  -------------------------------+-------------------------------
  copy_process()
      + dup_task_struct()<1>
				  parent move to another cgroup,
				  and free the old cgroup. <2>
      + sched_fork()
	+ __set_task_cpu()<3>
	+ task_fork_fair()
	  + sched_slice()<4>

In the worst case, this bug can lead to "use-after-free" and
cause panic as shown above:

  (1) parent copy its sched_task_group to child at <1>;

  (2) someone move the parent to another cgroup and free the old
      cgroup at <2>;

  (3) the sched_task_group and cfs_rq that belong to the old cgroup
      will be accessed at <3> and <4>, which cause a panic:

  [] BUG: unable to handle kernel NULL pointer dereference at 0000000000000000
  [] PGD 8000001fa0a86067 P4D 8000001fa0a86067 PUD 2029955067 PMD 0
  [] Oops: 0000 [#1] SMP PTI
  [] CPU: 7 PID: 648398 Comm: ebizzy Kdump: loaded Tainted: G           OE    --------- -  - 4.18.0.x86_64+ #1
  [] RIP: 0010:sched_slice+0x84/0xc0

  [] Call Trace:
  []  task_fork_fair+0x81/0x120
  []  sched_fork+0x132/0x240
  []  copy_process.part.5+0x675/0x20e0
  []  ? __handle_mm_fault+0x63f/0x690
  []  _do_fork+0xcd/0x3b0
  []  do_syscall_64+0x5d/0x1d0
  []  entry_SYSCALL_64_after_hwframe+0x65/0xca
  [] RIP: 0033:0x7f04418cd7e1

Between cgroup_can_fork() and cgroup_post_fork(), the cgroup
membership and thus sched_task_group can't change. So update child's
sched_task_group at sched_post_fork() and move task_fork() and
__set_task_cpu() (where accees the sched_task_group) from sched_fork()
to sched_post_fork().

Fixes: 8323f26ce342 ("sched: Fix race in task_group")
Signed-off-by: Zhang Qiao 
Signed-off-by: Peter Zijlstra (Intel) 
Acked-by: Tejun Heo 
Link: https://lkml.kernel.org/r/20210915064030.2231-1-zhangqiao22@huawei.com
Signed-off-by: Sasha Levin

sched/scs: Reset the shadow stack when idle_task_exit

2021-10-27T07:59:53+00:00

[ Upstream commit 63acd42c0d4942f74710b11c38602fb14dea7320 ]

Commit f1a0a376ca0c ("sched/core: Initialize the idle task with
preemption disabled") removed the init_idle() call from
idle_thread_get(). This was the sole call-path on hotplug that resets
the Shadow Call Stack (scs) Stack Pointer (sp).

Not resetting the scs-sp leads to scs overflow after enough hotplug
cycles. Therefore add an explicit scs_task_reset() to the hotplug code
to make sure the scs-sp does get reset on hotplug.

Fixes: f1a0a376ca0c ("sched/core: Initialize the idle task with preemption disabled")
Signed-off-by: Woody Lin 
[peterz: Changelog]
Signed-off-by: Peter Zijlstra (Intel) 
Reviewed-by: Valentin Schneider 
Link: https://lore.kernel.org/r/20211012083521.973587-1-woodylin@google.com
Signed-off-by: Sasha Levin

sched/fair: Null terminate buffer when updating tunable_scaling

2021-10-07T05:53:16+00:00

[ Upstream commit 703066188f63d66cc6b9d678e5b5ef1213c5938e ]

This patch null-terminates the temporary buffer in sched_scaling_write()
so kstrtouint() does not return failure and checks the value is valid.

Before:
  $ cat /sys/kernel/debug/sched/tunable_scaling
  1
  $ echo 0 > /sys/kernel/debug/sched/tunable_scaling
  -bash: echo: write error: Invalid argument
  $ cat /sys/kernel/debug/sched/tunable_scaling
  1

After:
  $ cat /sys/kernel/debug/sched/tunable_scaling
  1
  $ echo 0 > /sys/kernel/debug/sched/tunable_scaling
  $ cat /sys/kernel/debug/sched/tunable_scaling
  0
  $ echo 3 > /sys/kernel/debug/sched/tunable_scaling
  -bash: echo: write error: Invalid argument

Fixes: 8a99b6833c88 ("sched: Move SCHED_DEBUG sysctl to debugfs")
Signed-off-by: Mel Gorman 
Signed-off-by: Peter Zijlstra (Intel) 
Acked-by: Vincent Guittot 
Link: https://lore.kernel.org/r/20210927114635.GH3959@techsingularity.net
Signed-off-by: Sasha Levin

sched/fair: Add ancestors of unthrottled undecayed cfs_rq

2021-10-07T05:53:16+00:00

[ Upstream commit 2630cde26711dab0d0b56a8be1616475be646d13 ]

Since commit a7b359fc6a37 ("sched/fair: Correctly insert cfs_rq's to
list on unthrottle") we add cfs_rqs with no runnable tasks but not fully
decayed into the load (leaf) list. We may ignore adding some ancestors
and therefore breaking tmp_alone_branch invariant. This broke LTP test
cfs_bandwidth01 and it was partially fixed in commit fdaba61ef8a2
("sched/fair: Ensure that the CFS parent is added after unthrottling").

I noticed the named test still fails even with the fix (but with low
probability, 1 in ~1000 executions of the test). The reason is when
bailing out of unthrottle_cfs_rq early, we may miss adding ancestors of
the unthrottled cfs_rq, thus, not joining tmp_alone_branch properly.

Fix this by adding ancestors if we notice the unthrottled cfs_rq was
added to the load list.

Fixes: a7b359fc6a37 ("sched/fair: Correctly insert cfs_rq's to list on unthrottle")
Signed-off-by: Michal Koutný 
Signed-off-by: Peter Zijlstra (Intel) 
Reviewed-by: Vincent Guittot 
Reviewed-by: Odin Ugedal 
Link: https://lore.kernel.org/r/20210917153037.11176-1-mkoutny@suse.com
Signed-off-by: Sasha Levin

cpufreq: schedutil: Use kobject release() method to free sugov_tunables

2021-10-07T05:53:04+00:00

[ Upstream commit e5c6b312ce3cc97e90ea159446e6bfa06645364d ]

The struct sugov_tunables is protected by the kobject, so we can't free
it directly. Otherwise we would get a call trace like this:
  ODEBUG: free active (active state 0) object type: timer_list hint: delayed_work_timer_fn+0x0/0x30
  WARNING: CPU: 3 PID: 720 at lib/debugobjects.c:505 debug_print_object+0xb8/0x100
  Modules linked in:
  CPU: 3 PID: 720 Comm: a.sh Tainted: G        W         5.14.0-rc1-next-20210715-yocto-standard+ #507
  Hardware name: Marvell OcteonTX CN96XX board (DT)
  pstate: 40400009 (nZcv daif +PAN -UAO -TCO BTYPE=--)
  pc : debug_print_object+0xb8/0x100
  lr : debug_print_object+0xb8/0x100
  sp : ffff80001ecaf910
  x29: ffff80001ecaf910 x28: ffff00011b10b8d0 x27: ffff800011043d80
  x26: ffff00011a8f0000 x25: ffff800013cb3ff0 x24: 0000000000000000
  x23: ffff80001142aa68 x22: ffff800011043d80 x21: ffff00010de46f20
  x20: ffff800013c0c520 x19: ffff800011d8f5b0 x18: 0000000000000010
  x17: 6e6968207473696c x16: 5f72656d6974203a x15: 6570797420746365
  x14: 6a626f2029302065 x13: 303378302f307830 x12: 2b6e665f72656d69
  x11: ffff8000124b1560 x10: ffff800012331520 x9 : ffff8000100ca6b0
  x8 : 000000000017ffe8 x7 : c0000000fffeffff x6 : 0000000000000001
  x5 : ffff800011d8c000 x4 : ffff800011d8c740 x3 : 0000000000000000
  x2 : ffff0001108301c0 x1 : ab3c90eedf9c0f00 x0 : 0000000000000000
  Call trace:
   debug_print_object+0xb8/0x100
   __debug_check_no_obj_freed+0x1c0/0x230
   debug_check_no_obj_freed+0x20/0x88
   slab_free_freelist_hook+0x154/0x1c8
   kfree+0x114/0x5d0
   sugov_exit+0xbc/0xc0
   cpufreq_exit_governor+0x44/0x90
   cpufreq_set_policy+0x268/0x4a8
   store_scaling_governor+0xe0/0x128
   store+0xc0/0xf0
   sysfs_kf_write+0x54/0x80
   kernfs_fop_write_iter+0x128/0x1c0
   new_sync_write+0xf0/0x190
   vfs_write+0x2d4/0x478
   ksys_write+0x74/0x100
   __arm64_sys_write+0x24/0x30
   invoke_syscall.constprop.0+0x54/0xe0
   do_el0_svc+0x64/0x158
   el0_svc+0x2c/0xb0
   el0t_64_sync_handler+0xb0/0xb8
   el0t_64_sync+0x198/0x19c
  irq event stamp: 5518
  hardirqs last  enabled at (5517): [] console_unlock+0x554/0x6c8
  hardirqs last disabled at (5518): [] el1_dbg+0x28/0xa0
  softirqs last  enabled at (5504): [] __do_softirq+0x4d0/0x6c0
  softirqs last disabled at (5483): [] irq_exit+0x1b0/0x1b8

So split the original sugov_tunables_free() into two functions,
sugov_clear_global_tunables() is just used to clear the global_tunables
and the new sugov_tunables_free() is used as kobj_type::release to
release the sugov_tunables safely.

Fixes: 9bdcb44e391d ("cpufreq: schedutil: New governor based on scheduler utilization data")
Cc: 4.7+  # 4.7+
Signed-off-by: Kevin Hao 
Acked-by: Viresh Kumar 
Signed-off-by: Rafael J. Wysocki 
Signed-off-by: Sasha Levin

sched/idle: Make the idle timer expire in hard interrupt context

2021-09-26T12:10:25+00:00

[ Upstream commit 9848417926353daa59d2b05eb26e185063dbac6e ]

The intel powerclamp driver will setup a per-CPU worker with RT
priority. The worker will then invoke play_idle() in which it remains in
the idle poll loop until it is stopped by the timer it started earlier.

That timer needs to expire in hard interrupt context on PREEMPT_RT.
Otherwise the timer will expire in ksoftirqd as a SOFT timer but that task
won't be scheduled on the CPU because its priority is lower than the
priority of the worker which is in the idle loop.

Always expire the idle timer in hard interrupt context.

Reported-by: Thomas Gleixner 
Signed-off-by: Sebastian Andrzej Siewior 
Signed-off-by: Thomas Gleixner 
Link: https://lore.kernel.org/r/20210906113034.jgfxrjdvxnjqgtmc@linutronix.de
Signed-off-by: Sasha Levin

sched: Prevent balance_push() on remote runqueues

2021-09-18T11:43:24+00:00

commit 868ad33bfa3bf39960982682ad3a0f8ebda1656e upstream.

sched_setscheduler() and rt_mutex_setprio() invoke the run-queue balance
callback after changing priorities or the scheduling class of a task. The
run-queue for which the callback is invoked can be local or remote.

That's not a problem for the regular rq::push_work which is serialized with
a busy flag in the run-queue struct, but for the balance_push() work which
is only valid to be invoked on the outgoing CPU that's wrong. It not only
triggers the debug warning, but also leaves the per CPU variable push_work
unprotected, which can result in double enqueues on the stop machine list.

Remove the warning and validate that the function is invoked on the
outgoing CPU.

Fixes: ae7927023243 ("sched: Optimize finish_lock_switch()")
Reported-by: Sebastian Siewior 
Signed-off-by: Thomas Gleixner 
Signed-off-by: Peter Zijlstra (Intel) 
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/87zgt1hdw7.ffs@tglx
Signed-off-by: Greg Kroah-Hartman

sched: Fix UCLAMP_FLAG_IDLE setting

2021-09-15T08:02:07+00:00

[ Upstream commit ca4984a7dd863f3e1c0df775ae3e744bff24c303 ]

The UCLAMP_FLAG_IDLE flag is set on a runqueue when dequeueing the last
uclamp active task (that is, when buckets.tasks reaches 0 for all
buckets) to maintain the last uclamp.max and prevent blocked util from
suddenly becoming visible.

However, there is an asymmetry in how the flag is set and cleared which
can lead to having the flag set whilst there are active tasks on the rq.
Specifically, the flag is cleared in the uclamp_rq_inc() path, which is
called at enqueue time, but set in uclamp_rq_dec_id() which is called
both when dequeueing a task _and_ in the update_uclamp_active() path. As
a result, when both uclamp_rq_{dec,ind}_id() are called from
update_uclamp_active(), the flag ends up being set but not cleared,
hence leaving the runqueue in a broken state.

Fix this by clearing the flag in update_uclamp_active() as well.

Fixes: e496187da710 ("sched/uclamp: Enforce last task's UCLAMP_MAX")
Reported-by: Rick Yiu 
Signed-off-by: Quentin Perret 
Signed-off-by: Peter Zijlstra (Intel) 
Reviewed-by: Qais Yousef 
Tested-by: Dietmar Eggemann 
Link: https://lore.kernel.org/r/20210805102154.590709-2-qperret@google.com
Signed-off-by: Sasha Levin