Re: NULL ptr deref in perf/filter_match

[Peter Zijlstra <peterz@infradead.org>]

On Fri, Jul 29, 2016 at 11:41:11PM +0200, Vegard Nossum wrote:

> Digging a bit deeper into this, it seems the event itself is getting
> created by perf_event_open() and it gets added to the pmu_event_list
> through:
> 
> perf_event_open()
>  - perf_event_alloc()
>     - account_event()
>        - account_pmu_sb_event()
>           - attach_sb_event()
> 
> so at this point the event is being attached but its ->ctx is still
> NULL. It seems like ->ctx is set just a bit later in
> perf_event_open(), though.
> 
> But before that, __schedule() comes along and creates a stack trace
> similar to the one above:
> 
> __schedule()
>  - __perf_event_task_sched_out()
>    - perf_iterate_sb()
>      - perf_iterate_sb_cpu()
>         - event_filter_match()
>           - perf_cgroup_match()
>             - __get_cpu_context()
>               - (dereference ctx which is NULL)
> 
> So I guess the question is... should the event be attached (= put on
> the list) before ->ctx gets set? Or should the cgroup code check for a
> NULL ->ctx?

Does this fix it? Ordering is a bit of a mess, adding the events to the
list _after_ they've been installed has the risk of missing things I
think, nor does that result in particularly nice code.

Then again, this isn't pretty either.

---
 kernel/events/core.c | 23 ++++++++++++++++++-----
 1 file changed, 18 insertions(+), 5 deletions(-)

diff --git a/kernel/events/core.c b/kernel/events/core.c
index a19550d80ab1..87d02b8cb87e 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -1716,8 +1716,8 @@ static inline int pmu_filter_match(struct perf_event *event)
 static inline int
 event_filter_match(struct perf_event *event)
 {
-	return (event->cpu == -1 || event->cpu == smp_processor_id())
-	    && perf_cgroup_match(event) && pmu_filter_match(event);
+	return (event->cpu == -1 || event->cpu == smp_processor_id()) &&
+	       perf_cgroup_match(event) && pmu_filter_match(event);
 }
 
 static void
@@ -1737,8 +1737,8 @@ event_sched_out(struct perf_event *event,
 	 * maintained, otherwise bogus information is return
 	 * via read() for time_enabled, time_running:
 	 */
-	if (event->state == PERF_EVENT_STATE_INACTIVE
-	    && !event_filter_match(event)) {
+	if (event->state == PERF_EVENT_STATE_INACTIVE &&
+	    !event_filter_match(event)) {
 		delta = tstamp - event->tstamp_stopped;
 		event->tstamp_running += delta;
 		event->tstamp_stopped = tstamp;
@@ -2236,10 +2236,15 @@ perf_install_in_context(struct perf_event_context *ctx,
 
 	lockdep_assert_held(&ctx->mutex);
 
-	event->ctx = ctx;
 	if (event->cpu != -1)
 		event->cpu = cpu;
 
+	/*
+	 * Ensures that if we can observe event->ctx, both the event and ctx
+	 * will be 'complete'. See perf_iterate_sb_cpu().
+	 */
+	smp_store_release(&event->ctx, ctx);
+
 	if (!task) {
 		cpu_function_call(cpu, __perf_install_in_context, event);
 		return;
@@ -5969,6 +5974,14 @@ static void perf_iterate_sb_cpu(perf_iterate_f output, void *data)
 	struct perf_event *event;
 
 	list_for_each_entry_rcu(event, &pel->list, sb_list) {
+		/*
+		 * Skip events that are not fully formed yet; ensure that
+		 * if we observe event->ctx, both event and ctx will be
+		 * complete enough. See perf_install_in_context().
+		 */
+		if (!smp_load_acquire(&event->ctx))
+			continue;
+
 		if (event->state < PERF_EVENT_STATE_INACTIVE)
 			continue;
 		if (!event_filter_match(event))