perf: protect group_leader from races that cause ctx double-free

When moving a group_leader perf event from a software-context
to a hardware-context, there's a race in checking and
updating that context. The existing locking solution
doesn't work; note that it tries to grab a lock inside
the group_leader's context object, which you can only
get at by going through a pointer that should be protected
from these races. To avoid that problem, and to produce
a simple solution, we can just use a lock per group_leader
to protect all checks on the group_leader's context.
The new lock is grabbed and released when no context locks
are held.

RM-290

Bug: 30955111
Bug: 31095224
Change-Id: If37124c100ca6f4aa962559fba3bd5dbbec8e052

1 files changed, 6 insertions, 0 deletions

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 42dc246800c..82660a6e9df 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -757,6 +757,12 @@ struct perf_event {
 	int				nr_siblings;
 	int				group_flags;
 	struct perf_event		*group_leader;
+
+	/*
+	 * Protect the pmu, attributes and context of a group leader.
+	 * Note: does not protect the pointer to the group_leader.
+	 */
+	struct mutex			group_leader_mutex;
 	struct pmu			*pmu;
 
 	enum perf_event_active_state	state;