drm/i915: Check execlist/ring status during hangcheck
authorChris Wilson <chris@chris-wilson.co.uk>
Fri, 21 Jul 2017 12:32:23 +0000 (13:32 +0100)
committerDaniel Vetter <daniel.vetter@ffwll.ch>
Thu, 27 Jul 2017 07:38:45 +0000 (09:38 +0200)
Before we declare an engine as idle, check if there are any pending
execlist context-switches and if the ring itself reports as idle.
Otherwise, we may be left in a situation where we miss a crucial
execlist event (or something more sinister) yet the requests complete.
Since the seqno write happens, we believe the engine to be truly idle.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Michel Thierry <michel.thierry@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170721123238.16428-5-chris@chris-wilson.co.uk
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
drivers/gpu/drm/i915/intel_hangcheck.c

index 9b0ece427bdc8aeeb2eb28745c81bd18066f90c4..d9d87d96fb69d1b67219b9c01b81da9656a3badc 100644 (file)
@@ -324,7 +324,7 @@ hangcheck_get_action(struct intel_engine_cs *engine,
        if (engine->hangcheck.seqno != hc->seqno)
                return ENGINE_ACTIVE_SEQNO;
 
-       if (i915_seqno_passed(hc->seqno, intel_engine_last_submit(engine)))
+       if (intel_engine_is_idle(engine))
                return ENGINE_IDLE;
 
        return engine_stuck(engine, hc->acthd);