[Intel-gfx] [PATCH 1/2] drm/i915: Pull the unconditional GPU cache invalidation into request construction

Chris Wilson chris at chris-wilson.co.uk
Mon Nov 20 15:58:46 UTC 2017


Quoting Mika Kuoppala (2017-11-20 15:26:57)
> Chris Wilson <chris at chris-wilson.co.uk> writes:
> 
> > As the request now may implicitly invoke a context-switch, we should
> > follow that with a GPU TLB invalidation. Also even before using GGTT, we
> 
> s/follow/preample? As it is the context-switch you are preampling.
> 
> Also point that the request allocation will invote a context-switch
> but only after next patch.
> 
> So with some confusion untangled on these first lines,
> 
> Reviewed-by: Mika Kuoppala <mika.kuoppala at linux.intel.com>

Rewrote the changelog so that the effect of this plus the next patch is
clearer and pushed. Thanks for the review,

Now let's remove some modparams!
-Chris


More information about the Intel-gfx mailing list