[Intel-gfx] [PATCH 2/2] drm/i915/fence: Separate timeout mechanism for awaiting on dma-fences

[Chris Wilson]

Quoting Chris Wilson (2018-01-15 10:15:45)
> Quoting Tvrtko Ursulin (2018-01-15 10:08:01)
> > If it compiles, and works, assuming we have tests cases which exercise 
> > both paths, then it is obviously fine.
> 
> The no-timeout variants are using for inter-engine signaling, the
> timeout variant for external fences and KMS. Yes, both are covered,
> though only the no-timeout variants are really stressed to see how
> timeouts are handled (i.e. GPU hang and reset). The actual timeout
> fences are only stressed in a token effort, and hard to distinguish from
> the multiple similar timeout mechanisms spread around the code.

I really need to beef up the sw-fence/dma-fence interop selftests.
-Chris