system/linux
system/linux
1
0
Fork 0
mirror of https://github.com/torvalds/linux synced 2024-10-15 07:47:34 +00:00
Code Issues Packages Projects Releases Wiki Activity

perf/hw_breakpoint: Annotate tsk->perf_event_mutex vs ctx->mutex

Browse source 
Perf fuzzer gifted a lockdep splat:

  perf_event_init_context()
    mutex_lock(parent_ctx->mutex);			(B)
    inherit_task_group()
      inherit_group()
        inherit_event()
          perf_event_alloc()
            perf_try_init_event() := hw_breakpoint_event_init()
              register_perf_hw_breakpoint()
                mutex_lock(child->perf_event_mutex);	(A)

Which is against the normal (documented) order. Now, this is a false
positive in that child is not published yet, but also inherited events
never end up on ->perf_event_list.

Annotate this one away.

Fixes: 0912037fec ("perf/hw_breakpoint: Reduce contention with large number of tasks")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
This commit is contained in:
Peter Zijlstra 2022-10-04 12:20:39 +02:00
parent 7be51cc1c6
commit 82aad7ff7a
1 changed files with 11 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

12
kernel/events/hw_breakpoint.c
View file

@ -117,7 +117,17 @@ static struct mutex *bp_constraints_lock(struct perf_event *bp)
struct mutex *tsk_mtx = get_task_bps_mutex(bp);
if (tsk_mtx) {
mutex_lock(tsk_mtx);
/*
* Fully analogous to the perf_try_init_event() nesting
* argument in the comment near perf_event_ctx_lock_nested();
* this child->perf_event_mutex cannot ever deadlock against
* the parent->perf_event_mutex usage from
* perf_event_task_{en,dis}able().
*
* Specifically, inherited events will never occur on
* ->perf_event_list.
*/
mutex_lock_nested(tsk_mtx, SINGLE_DEPTH_NESTING);
percpu_down_read(&bp_cpuinfo_sem);
} else {
percpu_down_write(&bp_cpuinfo_sem);