[Intel-gfx] [PATCH 10/12] drm/i915/gt: Refactor l3cc/mocs availability

Chris Wilson chris at chris-wilson.co.uk
Tue Feb 18 18:33:46 UTC 2020


Quoting Brian Welty (2020-02-18 18:26:40)
> 
> On 2/18/2020 8:21 AM, Chris Wilson wrote:
> > @@ -223,9 +235,9 @@ static int check_mocs_engine(struct live_mocs *arg,
> >       /* Read the mocs tables back using SRM */
> >       offset = i915_ggtt_offset(vma);
> >       if (!err)
> > -             err = read_mocs_table(rq, &arg->table, &offset);
> > +             err = read_mocs_table(rq, &arg->mocs, &offset);
> >       if (!err && ce->engine->class == RENDER_CLASS)
> > -             err = read_l3cc_table(rq, &arg->table, &offset);
> > +             err = read_l3cc_table(rq, &arg->l3cc, &offset);
> 
> 
> Above functions will call read_regs().

And do nothing.

Perhaps using the flags for read_mocs_table() would be nice to eliminate
the HAS_GLOBAL_MOCS_REGISTERS(). Or creating global_mocs, engine_mocs,
render_l3cc. That might be overkill.
-Chris


More information about the Intel-gfx mailing list