Hi guys,
this hopefully addresses the issues pointed out in previous reviews,
the main difference is the st/mesa max samples chooser is correct,
which Ian pointed out.
Also r600g sample positions are now valid, my last attempt missed
some unsigned/signed issues.
Dave.
_
On Sat, Apr 6, 2013 at 8:05 PM, Dave Airlie wrote:
> I sent these before, and Marek reviewed them (the first 3 at least).
Oh calim pointed out I'd missed some comments from before, I've pushed
these to a branch with get_sample_position documented in p_context.h,
and switching to using unsigned sa
I sent these before, and Marek reviewed them (the first 3 at least).
But I'd like to have the r600g one looked at before I pushed it,
I rewrote it to properly return the sample positions from the
correct place.
One other issue is all the 2D MS Array sampler tests fail on r600g,
not sure why, I've
