Quoting Summers, Stuart (2019-07-08 22:11:15)
> On Fri, 2019-07-05 at 13:43 +0100, Chris Wilson wrote:
> > We now track features correctly instead of probing i915->engine[RCS0]
> > which is much more flexible and avoids any nasty surprises.
> >
> > Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
> > Cc: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
> > ---
> > drivers/gpu/drm/i915/gt/intel_engine_cs.c | 6 ------
> > 1 file changed, 6 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c
> > b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
> > index df5932f5f578..bdf279fa3b2e 100644
> > --- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c
> > +++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
> > @@ -448,12 +448,6 @@ int intel_engines_init_mmio(struct
> > drm_i915_private *i915)
> > if (WARN_ON(mask != engine_mask))
> > device_info->engine_mask = mask;
> >
> > - /* We always presume we have at least RCS available for later
> > probing */
> > - if (WARN_ON(!HAS_ENGINE(i915, RCS0))) {
> > - err = -ENODEV;
> > - goto cleanup;
> > - }
> > -
>
> Just going by the series here, we have quite a few other place we are
> touching RCS0 specifically during driver load. Do we really want to get
> rid of this? Or is there an alternative if RCS0 isn't present for some
> reason?
Outside of gvt/ (which I don't dare to try and verify), the only places
where we now mention RCS0 are in direct hw mappings to that engine
(e.g. interrupts and mmio setup). [Excluding selftests/ which are mostly
converted and really just a matter of generalising if applicable or
marking as "no really, this only applies to RCS0".] Assuming the other
couple of patches also land.
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
