Re: [Intel-gfx] [CI 1/3] drm/i915/gt: Skip rmw for masked registers

[Mika Kuoppala <mika.kuoppala@linux.intel.com>]

Chris Wilson <chris@chris-wilson.co.uk> writes:

> Quoting Mika Kuoppala (2020-01-31 11:51:44)
>> Chris Wilson <chris@chris-wilson.co.uk> writes:
>> 
>> > A masked register does not need rmw to update, and it is best not to use
>> > such a sequence.
>> >
>> > Reported-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
>> > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
>> > Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
>> > Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>> > ---
>> >  drivers/gpu/drm/i915/gt/intel_workarounds.c | 32 ++++++++++++++-------
>> >  1 file changed, 21 insertions(+), 11 deletions(-)
>> >
>> > diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
>> > index 5a7db279f702..e4c2b6d42f46 100644
>> > --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
>> > +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
>> > @@ -116,7 +116,8 @@ static void _wa_add(struct i915_wa_list *wal, const struct i915_wa *wa)
>> >               } else {
>> >                       wa_ = &wal->list[mid];
>> >  
>> > -                     if ((wa->mask & ~wa_->mask) == 0) {
>> > +                     if ((wa->mask | wa_->mask) &&
>> 
>> Don't we want to discard if someone tries to demote a masked
>> one into a plain?
>
> That should throw the error, right?
>
> If either used a mask and now we don't, then 0 & x == 0 => DRM_ERROR.

Yes, it will throw the error. My mistake.

But if we have a mask, we should not allow nonmasked additions
aswell? So mask == 0 would always be masked register and you
can't mix.

-Mika
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx