drm/i915: Suppress busy status for engines if wedged

If the driver is wedged, HW state may be very inconsistent and
report that it is still busy, even though we have stopped using it. This
can lead to a double *ERROR* rather than a graceful cleanup after
wedging.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170330145041.9005-2-chris@chris-wilson.co.uk
This commit is contained in:
Chris Wilson 2017-03-30 15:50:37 +01:00
parent 2c170af76c
commit 8490ae207f
2 changed files with 10 additions and 3 deletions

View file

@ -3112,9 +3112,7 @@ i915_gem_idle_work_handler(struct work_struct *work)
* Wait for last execlists context complete, but bail out in case a
* new request is submitted.
*/
wait_for(READ_ONCE(dev_priv->gt.active_requests) ||
intel_engines_are_idle(dev_priv),
10);
wait_for(intel_engines_are_idle(dev_priv), 10);
if (READ_ONCE(dev_priv->gt.active_requests))
return;

View file

@ -1111,6 +1111,15 @@ bool intel_engines_are_idle(struct drm_i915_private *dev_priv)
struct intel_engine_cs *engine;
enum intel_engine_id id;
if (READ_ONCE(dev_priv->gt.active_requests))
return false;
/* If the driver is wedged, HW state may be very inconsistent and
* report that it is still busy, even though we have stopped using it.
*/
if (i915_terminally_wedged(&dev_priv->gpu_error))
return true;
for_each_engine(engine, dev_priv, id) {
if (!intel_engine_is_idle(engine))
return false;