In the failure path, we call perf_detach_cgroup(), but we didn't
call perf_get_cgroup() prio to it.
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <
4D6F346E.
9070606@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
cgrp = container_of(css, struct perf_cgroup, css);
event->cgrp = cgrp;
+ /* must be done before we fput() the file */
+ perf_get_cgroup(event);
+
/*
* all events in a group must monitor
* the same cgroup because a task belongs
if (group_leader && group_leader->cgrp != cgrp) {
perf_detach_cgroup(event);
ret = -EINVAL;
- } else {
- /* must be done before we fput() the file */
- perf_get_cgroup(event);
}
out:
fput_light(file, fput_needed);