linux_dsm_epyc7002/kernel/sched
Linus Torvalds c40f7d74c7 sched/fair: Fix infinite loop in update_blocked_averages() by reverting a9e7f6544b
Zhipeng Xie, Xie XiuQi and Sargun Dhillon reported lockups in the
scheduler under high loads, starting at around the v4.18 time frame,
and Zhipeng Xie tracked it down to bugs in the rq->leaf_cfs_rq_list
manipulation.

Do a (manual) revert of:

  a9e7f6544b ("sched/fair: Fix O(nr_cgroups) in load balance path")

It turns out that the list_del_leaf_cfs_rq() introduced by this commit
is a surprising property that was not considered in followup commits
such as:

  9c2791f936 ("sched/fair: Fix hierarchical order in rq->leaf_cfs_rq_list")

As Vincent Guittot explains:

 "I think that there is a bigger problem with commit a9e7f6544b and
  cfs_rq throttling:

  Let take the example of the following topology TG2 --> TG1 --> root:

   1) The 1st time a task is enqueued, we will add TG2 cfs_rq then TG1
      cfs_rq to leaf_cfs_rq_list and we are sure to do the whole branch in
      one path because it has never been used and can't be throttled so
      tmp_alone_branch will point to leaf_cfs_rq_list at the end.

   2) Then TG1 is throttled

   3) and we add TG3 as a new child of TG1.

   4) The 1st enqueue of a task on TG3 will add TG3 cfs_rq just before TG1
      cfs_rq and tmp_alone_branch will stay  on rq->leaf_cfs_rq_list.

  With commit a9e7f6544b, we can del a cfs_rq from rq->leaf_cfs_rq_list.
  So if the load of TG1 cfs_rq becomes NULL before step 2) above, TG1
  cfs_rq is removed from the list.
  Then at step 4), TG3 cfs_rq is added at the beginning of rq->leaf_cfs_rq_list
  but tmp_alone_branch still points to TG3 cfs_rq because its throttled
  parent can't be enqueued when the lock is released.
  tmp_alone_branch doesn't point to rq->leaf_cfs_rq_list whereas it should.

  So if TG3 cfs_rq is removed or destroyed before tmp_alone_branch
  points on another TG cfs_rq, the next TG cfs_rq that will be added,
  will be linked outside rq->leaf_cfs_rq_list - which is bad.

  In addition, we can break the ordering of the cfs_rq in
  rq->leaf_cfs_rq_list but this ordering is used to update and
  propagate the update from leaf down to root."

Instead of trying to work through all these cases and trying to reproduce
the very high loads that produced the lockup to begin with, simplify
the code temporarily by reverting a9e7f6544b - which change was clearly
not thought through completely.

This (hopefully) gives us a kernel that doesn't lock up so people
can continue to enjoy their holidays without worrying about regressions. ;-)

[ mingo: Wrote changelog, fixed weird spelling in code comment while at it. ]

Analyzed-by: Xie XiuQi <xiexiuqi@huawei.com>
Analyzed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reported-by: Zhipeng Xie <xiezhipeng1@huawei.com>
Reported-by: Sargun Dhillon <sargun@sargun.me>
Reported-by: Xie XiuQi <xiexiuqi@huawei.com>
Tested-by: Zhipeng Xie <xiezhipeng1@huawei.com>
Tested-by: Sargun Dhillon <sargun@sargun.me>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Vincent Guittot <vincent.guittot@linaro.org>
Cc: <stable@vger.kernel.org> # v4.13+
Cc: Bin Li <huawei.libin@huawei.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: a9e7f6544b ("sched/fair: Fix O(nr_cgroups) in load balance path")
Link: http://lkml.kernel.org/r/1545879866-27809-1-git-send-email-xiexiuqi@huawei.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2018-12-30 13:54:31 +01:00
..
autogroup.c sched/autogroup: Fix possible Spectre-v1 indexing for sched_prio_to_weight[] 2018-05-05 08:34:42 +02:00
autogroup.h sched/headers: Simplify and clean up header usage in the scheduler 2018-03-04 12:39:29 +01:00
clock.c sched/clock: Disable interrupts when calling generic_sched_clock_init() 2018-07-30 19:33:35 +02:00
completion.c sched/Documentation: Update wake_up() & co. memory-barrier guarantees 2018-07-17 09:30:34 +02:00
core.c Merge branch 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2018-12-26 14:56:10 -08:00
cpuacct.c sched/headers: Simplify and clean up header usage in the scheduler 2018-03-04 12:39:29 +01:00
cpudeadline.c sched/headers: Simplify and clean up header usage in the scheduler 2018-03-04 12:39:29 +01:00
cpudeadline.h sched/headers: Simplify and clean up header usage in the scheduler 2018-03-04 12:39:29 +01:00
cpufreq_schedutil.c Merge branch 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2018-12-26 14:56:10 -08:00
cpufreq.c sched/cpufreq: Add the SPDX tags 2018-12-11 11:35:25 +01:00
cpupri.c sched/headers: Simplify and clean up header usage in the scheduler 2018-03-04 12:39:29 +01:00
cpupri.h sched/headers: Simplify and clean up header usage in the scheduler 2018-03-04 12:39:29 +01:00
cputime.c sched: Fix various typos in comments 2018-12-03 11:55:42 +01:00
deadline.c sched/core: Remove unnecessary unlikely() in push_*_task() 2018-12-11 15:16:57 +01:00
debug.c sched/core: Create task_has_idle_policy() helper 2018-11-12 06:17:52 +01:00
fair.c sched/fair: Fix infinite loop in update_blocked_averages() by reverting a9e7f6544b 2018-12-30 13:54:31 +01:00
features.h sched/fair: Disable LB_BIAS by default 2018-10-02 09:45:01 +02:00
idle.c x86/stackprotector: Remove the call to boot_init_stack_canary() from cpu_startup_entry() 2018-10-22 04:07:24 +02:00
isolation.c sched: Fix various typos in comments 2018-12-03 11:55:42 +01:00
loadavg.c sched: loadavg: make calc_load_n() public 2018-10-26 16:26:32 -07:00
Makefile psi: pressure stall information for CPU, memory, and IO 2018-10-26 16:26:32 -07:00
membarrier.c sched/membarrier: synchronize_sched() with synchronize_rcu() 2018-11-27 09:21:43 -08:00
pelt.c sched/fair: Remove setting task's se->runnable_weight during PELT update 2018-10-02 09:45:03 +02:00
pelt.h sched/pelt: Fix warning and clean up IRQ PELT config 2018-10-02 09:45:00 +02:00
psi.c psi: make disabling/enabling easier for vendor kernels 2018-11-30 14:56:14 -08:00
rt.c sched/core: Remove unnecessary unlikely() in push_*_task() 2018-12-11 15:16:57 +01:00
sched-pelt.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
sched.h sched/fair: Add over-utilization/tipping point indicator 2018-12-11 15:17:01 +01:00
stats.c proc: introduce proc_create_seq{,_data} 2018-05-16 07:23:35 +02:00
stats.h psi: make disabling/enabling easier for vendor kernels 2018-11-30 14:56:14 -08:00
stop_task.c sched: Clean up and harmonize the coding style of the scheduler code base 2018-03-03 15:50:21 +01:00
swait.c sched/swait: Rename to exclusive 2018-06-20 11:35:56 +02:00
topology.c sched/toplogy: Introduce the 'sched_energy_present' static key 2018-12-11 15:17:01 +01:00
wait_bit.c sched/wait: Improve __var_waitqueue() code generation 2018-03-20 08:23:25 +01:00
wait.c sched/wait: assert the wait_queue_head lock is held in __wake_up_common 2018-08-22 10:52:47 -07:00