perf/cgroups: Install cgroup events to correct cpuctx

cgroup events are always installed in the cpuctx. However, when it is not
installed via IPI, list_update_cgroup_event() adds it to cpuctx of current
CPU, which triggers list corruption:

  [] list_add double add: new=ffff888ff7cf0db0, prev=ffff888ff7ce82f0, next=ffff888ff7cf0db0.

To reproduce this, we can simply run:

  # perf stat -e cs -a &
  # perf stat -e cs -G anycgroup

Fix this by installing it to cpuctx that contains event->ctx, and the
proper cgrp_cpuctx_list.

Fixes: db0503e ("perf/core: Optimize perf_install_in_event()")
Suggested-by: Peter Zijlstra (Intel) <[email protected]>
Signed-off-by: Song Liu <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Cc: <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]

Author: liu-song-6

kernel/events/core.c

@@ -951,9 +951,9 @@ list_update_cgroup_event(struct perf_event *event,
 
 	/*
 	 * Because cgroup events are always per-cpu events,
-	 * this will always be called from the right CPU.
+	 * @ctx == &cpuctx->ctx.
 	 */
-	cpuctx = __get_cpu_context(ctx);
+	cpuctx = container_of(ctx, struct perf_cpu_context, ctx);
 
 	/*
 	 * Since setting cpuctx->cgrp is conditional on the current @cgrp
@@ -979,7 +979,8 @@ list_update_cgroup_event(struct perf_event *event,
 
 	cpuctx_entry = &cpuctx->cgrp_cpuctx_entry;
 	if (add)
-		list_add(cpuctx_entry, this_cpu_ptr(&cgrp_cpuctx_list));
+		list_add(cpuctx_entry,
+			 per_cpu_ptr(&cgrp_cpuctx_list, event->cpu));
 	else
 		list_del(cpuctx_entry);
 }