Quoting Tvrtko Ursulin (2017-10-06 13:23:03)
>
> On 06/10/2017 12:56, Chris Wilson wrote:
> > If two nop's (requests in-flight following a wedged device) complete at
> > the same time, the global_seqno value written to the HWSP is undefined
> > as the two threads are not serialized.
> >
> > v2: Use irqsafe spinlock. We expect the callback may be called from
> > inside another irq spinlock, so we can't unconditionally restore irqs.
> >
> > Fixes: ce1135c7de64 ("drm/i915: Complete requests in nop_submit_request")
> > Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
> > Cc: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
> > Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx> #v1
> > ---
> > drivers/gpu/drm/i915/i915_gem.c | 7 ++++++-
> > 1 file changed, 6 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
> > index ab8c6946fea4..6a6974ed8f74 100644
> > --- a/drivers/gpu/drm/i915/i915_gem.c
> > +++ b/drivers/gpu/drm/i915/i915_gem.c
> > @@ -3014,10 +3014,15 @@ void i915_gem_reset_finish(struct drm_i915_private *dev_priv)
> >
> > static void nop_submit_request(struct drm_i915_gem_request *request)
> > {
> > + unsigned long flags;
> > +
> > GEM_BUG_ON(!i915_terminally_wedged(&request->i915->gpu_error));
> > dma_fence_set_error(&request->fence, -EIO);
> > - i915_gem_request_submit(request);
> > +
> > + spin_lock_irqsave(&request->engine->timeline->lock, flags);
> > + __i915_gem_request_submit(request);
> > intel_engine_init_global_seqno(request->engine, request->global_seqno);
> > + spin_unlock_irqrestore(&request->engine->timeline->lock, flags);
> > }
> >
> > static void engine_set_wedged(struct intel_engine_cs *engine)
> >
>
> Ooops..
>
> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
Thanks for asking the question that lead to the discovery of the race and
then reviewing the results! Pushed,
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
