Re: [Intel-gfx] [PATCH v2 3/7] drm/i915/execlists: Make submission tasklet hardirq safe

Quoting Chris Wilson (2018-05-08 21:59:20) > Quoting Tvrtko Ursulin (2018-05-08 18:45:44) > > > > On 07/05/2018 14:57, Chris Wilson wrote: > > > Prepare to allow the execlists submission to be run from underneath a > > > hardirq timer context (and not just the current softirq context) as is > > >

Re: [Intel-gfx] [PATCH v2 3/7] drm/i915/execlists: Make submission tasklet hardirq safe

Quoting Tvrtko Ursulin (2018-05-08 18:45:44) > > On 07/05/2018 14:57, Chris Wilson wrote: > > Prepare to allow the execlists submission to be run from underneath a > > hardirq timer context (and not just the current softirq context) as is > > required for fast preemption resets and context

Re: [Intel-gfx] [PATCH v2 3/7] drm/i915/execlists: Make submission tasklet hardirq safe

On 07/05/2018 14:57, Chris Wilson wrote: Prepare to allow the execlists submission to be run from underneath a hardirq timer context (and not just the current softirq context) as is required for fast preemption resets and context switches. Signed-off-by: Chris Wilson

Re: [Intel-gfx] [PATCH v2 3/7] drm/i915/execlists: Make submission tasklet hardirq safe

On 07/05/2018 14:57, Chris Wilson wrote: Prepare to allow the execlists submission to be run from underneath a hardirq timer context (and not just the current softirq context) as is required for fast preemption resets and context switches. Signed-off-by: Chris Wilson

Re: [Intel-gfx] [PATCH v2 3/7] drm/i915/execlists: Make submission tasklet hardirq safe

Quoting Tvrtko Ursulin (2018-05-08 12:38:06) > > On 08/05/2018 12:05, Chris Wilson wrote: > > Quoting Tvrtko Ursulin (2018-05-08 11:56:37) > >> > >> On 08/05/2018 11:24, Chris Wilson wrote: > >>> Quoting Tvrtko Ursulin (2018-05-08 11:10:41) > On 07/05/2018 14:57, Chris Wilson wrote: > >

Re: [Intel-gfx] [PATCH v2 3/7] drm/i915/execlists: Make submission tasklet hardirq safe

On 08/05/2018 12:05, Chris Wilson wrote: Quoting Tvrtko Ursulin (2018-05-08 11:56:37) On 08/05/2018 11:24, Chris Wilson wrote: Quoting Tvrtko Ursulin (2018-05-08 11:10:41) On 07/05/2018 14:57, Chris Wilson wrote: @@ -744,13 +748,25 @@ static void execlists_dequeue(struct intel_engine_cs

Re: [Intel-gfx] [PATCH v2 3/7] drm/i915/execlists: Make submission tasklet hardirq safe

Quoting Tvrtko Ursulin (2018-05-08 11:56:37) > > On 08/05/2018 11:24, Chris Wilson wrote: > > Quoting Tvrtko Ursulin (2018-05-08 11:10:41) > >> On 07/05/2018 14:57, Chris Wilson wrote: > >>> @@ -744,13 +748,25 @@ static void execlists_dequeue(struct > >>> intel_engine_cs *engine) > >>>/*

Re: [Intel-gfx] [PATCH v2 3/7] drm/i915/execlists: Make submission tasklet hardirq safe

On 08/05/2018 11:24, Chris Wilson wrote: Quoting Tvrtko Ursulin (2018-05-08 11:10:41) On 07/05/2018 14:57, Chris Wilson wrote: @@ -744,13 +748,25 @@ static void execlists_dequeue(struct intel_engine_cs *engine) /* We must always keep the beast fed if we have work piled up */

Re: [Intel-gfx] [PATCH v2 3/7] drm/i915/execlists: Make submission tasklet hardirq safe

Quoting Tvrtko Ursulin (2018-05-08 11:10:41) > On 07/05/2018 14:57, Chris Wilson wrote: > > @@ -744,13 +748,25 @@ static void execlists_dequeue(struct intel_engine_cs > > *engine) > > /* We must always keep the beast fed if we have work piled up */ > > GEM_BUG_ON(execlists->first &&

Re: [Intel-gfx] [PATCH v2 3/7] drm/i915/execlists: Make submission tasklet hardirq safe

On 07/05/2018 14:57, Chris Wilson wrote: Prepare to allow the execlists submission to be run from underneath a hardirq timer context (and not just the current softirq context) as is required for fast preemption resets and context switches. Signed-off-by: Chris Wilson

[Intel-gfx] [PATCH v2 3/7] drm/i915/execlists: Make submission tasklet hardirq safe

Prepare to allow the execlists submission to be run from underneath a hardirq timer context (and not just the current softirq context) as is required for fast preemption resets and context switches. Signed-off-by: Chris Wilson --- drivers/gpu/drm/i915/intel_lrc.c | 42