[Intel-gfx] [PATCH 10/34] drm/i915: Remove GPU reset dependence on struct_mutex

[Chris Wilson]

Quoting Chris Wilson (2019-01-24 12:50:30)
> Quoting Mika Kuoppala (2019-01-24 12:06:01)
> > Chris Wilson <chris at chris-wilson.co.uk> writes:
> > > -     /*
> > > -      * We want to perform per-engine reset from atomic context (e.g.
> > > -      * softirq), which imposes the constraint that we cannot sleep.
> > > -      * However, experience suggests that spending a bit of time waiting
> > > -      * for a reset helps in various cases, so for a full-device reset
> > > -      * we apply the opposite rule and wait if we want to. As we should
> > > -      * always follow up a failed per-engine reset with a full device reset,
> > > -      * being a little faster, stricter and more error prone for the
> > > -      * atomic case seems an acceptable compromise.
> > > -      *
> > > -      * Unfortunately this leads to a bimodal routine, when the goal was
> > > -      * to have a single reset function that worked for resetting any
> > > -      * number of engines simultaneously.
> > > -      */
> > > -     might_sleep_if(engine_mask == ALL_ENGINES);
> > 
> > Oh here it is. I was after this on atomic resets.
> 
> I was saying it didn't make sense to lift the restriction until we
> relied upon. Just because the code became safe doesn't mean it was then
> part of the API :)

Hmm, set-wedged is meant to be magical more or less now. That gives more
weight to the argument of making intel_gpu_reset() magical and removing
this comment earlier.
-Chris