On 09/25/2014 09:00 AM, Matt Turner wrote:
> On Thu, Sep 25, 2014 at 8:25 AM, Ian Romanick wrote:
>> How did you test this? Do we have piglit execution tests that actually
>> hit this path? I'm assuming you didn't play Serious Sam 3 looking for
>> rendering errors. ;)
>
> I wrote the patch and
On Mon, Sep 8, 2014 at 12:21 PM, Matt Turner wrote:
> Helps a handful of programs in Serious Sam 3 that use do-while loops.
>
> instructions in affected programs: 16114 -> 16075 (-0.24%)
> ---
How about a review?
___
mesa-dev mailing list
mesa-dev@l
On Thu, Sep 25, 2014 at 8:25 AM, Ian Romanick wrote:
> How did you test this? Do we have piglit execution tests that actually
> hit this path? I'm assuming you didn't play Serious Sam 3 looking for
> rendering errors. ;)
I wrote the patch and initially missed the necessary predicate_inverse
bit
How did you test this? Do we have piglit execution tests that actually
hit this path? I'm assuming you didn't play Serious Sam 3 looking for
rendering errors. ;)
On 09/08/2014 12:21 PM, Matt Turner wrote:
> Helps a handful of programs in Serious Sam 3 that use do-while loops.
>
> instructions i
Helps a handful of programs in Serious Sam 3 that use do-while loops.
instructions in affected programs: 16114 -> 16075 (-0.24%)
---
This patch depends on the save-cfg series.
.../dri/i965/brw_fs_peephole_predicated_break.cpp | 37 ++
1 file changed, 37 insertions(+)
di
