On 25 June 2018 at 16:24, Gert Wollny wrote:
> Am Montag, den 25.06.2018, 15:54 +1000 schrieb Dave Airlie:
>> /home/airlied/devel/piglit/bin/textureGather fs nonconst r 0 float
>> 2DArray repeat
>>
>> amongst others appears to regress with these two.
>
> I can confirm this. From which piglit subse
Am Montag, den 25.06.2018, 15:54 +1000 schrieb Dave Airlie:
> /home/airlied/devel/piglit/bin/textureGather fs nonconst r 0 float
> 2DArray repeat
>
> amongst others appears to regress with these two.
I can confirm this. From which piglit subset is this? I ran the shader
subset and didn't see this
/home/airlied/devel/piglit/bin/textureGather fs nonconst r 0 float
2DArray repeat
amongst others appears to regress with these two.
Dave.
On 22 June 2018 at 19:37, Gert Wollny wrote:
> these two patches correct the offets and the rounding modes for the
> index evaluation when accessing texture
https://bugs.freedesktop.org/show_bug.cgi?id=107022
Bug ID: 107022
Summary: [RADV] The Witcher 3: Trembling of trees
Product: Mesa
Version: git
Hardware: x86-64 (AMD64)
OS: Linux (All)
Status: NEW
Severity
https://bugs.freedesktop.org/show_bug.cgi?id=107022
zefkerri...@gmail.com changed:
What|Removed |Added
CC||zefkerri...@gmail.com
--
You ar
Hi Jason
Am So., 24. Juni 2018 um 22:14 Uhr schrieb Jason Ekstrand
:
>
> In general, drivers and passes shouldn't rely on the results of
> index_*_regs or index_ssa_defs after an optimization or lowering pass is
> run. We've considered the results of those functions to be too volatile to
> try an
fixes a couple of packed_pixel CTS tests. No regressions inside a CTS run.
Signed-off-by: Karol Herbst
---
src/gallium/drivers/nouveau/nv50/nv50_blit.h | 21 ++-
.../drivers/nouveau/nv50/nv50_surface.c | 14 +++--
2 files changed, 23 insertions(+), 12 deletions(-)
In general, drivers and passes shouldn't rely on the results of
index_*_regs or index_ssa_defs after an optimization or lowering pass is
run. We've considered the results of those functions to be too volatile to
try and keep up to date. Drivers or passes are expected to call the
indexing func
After calling nir_lower_regs_to_ssa(..) nir_function_impl's reg_alloc
counter still hold its old value. Just call nir_index_local_regs(..) to
set reg_alloc to the correct value - which should be 0.
Signed-off-by: Christian Gmeiner
---
src/compiler/nir/nir_lower_regs_to_ssa.c | 2 ++
1 file chang
