Signed-off-by: Vadim Girlin
---
src/gallium/drivers/r600/sb/sb_bc_parser.cpp | 3 ++-
src/gallium/drivers/r600/sb/sb_shader.cpp| 5 +
2 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/src/gallium/drivers/r600/sb/sb_bc_parser.cpp
b/src/gallium/drivers/r600/sb/sb_bc_parser.cpp
We can replace CNDxx with MOV (and possibly eliminate after
propagation) in following cases:
If src1 is equal to src2 in CNDxx instruction then the result doesn't
depend on condition and we can replace the instruction with
"MOV dst, src1".
If src0 is const then we can evaluate the condition at co
Use the same limit for kcache constants in alu group on r6xx as on other
chips (two const pairs). Relaxing this will require additional checks to
make sure that all 4 consts in the group come from 2 kcache sets (clause
limit), probably without noticeable improvements of shader performance.
Signed-
On Sat, May 4, 2013 at 12:14 AM, Matt Turner wrote:
> On Fri, May 3, 2013 at 10:42 PM, Chad Versace
> wrote:
>> +32% on GLBenchmark 2.5 Egypt Offscreen DXT1. Tested on Harris Beach GT3 with
>> Android.
>>
>> Please please please, someone run Piglit on this series. I currently have no
>> Haswell
On 05/04/2013 02:32 PM, gregory hainaut wrote:
On Sat, 04 May 2013 12:37:05 -0700
Kenneth Graunke wrote:
On 05/03/2013 10:44 AM, Gregory Hainaut wrote:
V2:
* Rename object
* Formatting improvement
---
src/mesa/main/pipelineobj.c | 77
+++
1 file
On Sat, 04 May 2013 12:37:05 -0700
Kenneth Graunke wrote:
> On 05/03/2013 10:44 AM, Gregory Hainaut wrote:
> > V2:
> > * Rename object
> > * Formatting improvement
> > ---
> > src/mesa/main/pipelineobj.c | 77
> > +++
> > 1 file changed, 77 insertions
This series, and the associated mesa changes are all:
Tested-By: Aaron Watry
--Aaron
On Fri, May 3, 2013 at 5:53 PM, Tom Stellard wrote:
> Hi,
>
> The attached patches modify the CodeEmitter to emit true ISA.
> Previously, we were prefixing all instructions with an instruction type
> byte.
>
>
On 05/03/2013 10:44 AM, Gregory Hainaut wrote:
V2:
* Rename object
* Formatting improvement
---
src/mesa/main/pipelineobj.c | 77 +++
1 file changed, 77 insertions(+)
diff --git a/src/mesa/main/pipelineobj.c b/src/mesa/main/pipelineobj.c
index d81bd0e
Hi,
Thank for doing this.
Patches 1 2 and 3 have my rb.
For patch 4:
>@@ -125,9 +106,7 @@ MCCodeEmitter *llvm::createR600MCCodeEmitter(const
>MCInstrInfo &MCII,
>
> void R600MCCodeEmitter::EncodeInstruction(const MCInst &MI, raw_ostream &OS,
>SmallVectorI
Hi,
I bisected a performance regression here:
https://bugs.freedesktop.org/show_bug.cgi?id=64213
Here's some benchmark numbers:
Core i5-3320M, Firefox 23 nightly build 2013-05-04
Epic Citadel 1920x1080 (browser window maximized, not game screen)
Windows 40.5 FPS
Linux mesa git: 18.9 FPS 2013-
On Fri, 3 May 2013 11:56:39 -0700
Matt Turner wrote:
> On Fri, May 3, 2013 at 10:44 AM, Gregory Hainaut
> wrote:
> > V2:
> > * Rename object
> > * Formatting improvement
> > ---
> > src/mesa/main/pipelineobj.c | 77
> > +++
> > 1 file changed, 77 inser
On Fri, 3 May 2013 12:04:48 -0700
Matt Turner wrote:
> On Fri, May 3, 2013 at 10:44 AM, Gregory Hainaut
> wrote:
> > ---
> > src/mesa/main/get.c |9 +
> > src/mesa/main/get_hash_params.py |3 +++
> > 2 files changed, 12 insertions(+)
> >
> > diff --git a/src/mesa/ma
Looks good AFAICT
- Original Message -
> From: Roland Scheidegger
>
> Adds the remaining integer opcodes, and some opcodes are moved to more
> appropriate places, along with getting rid of the (already nearly empty)
> ps_2_x section. Though the CAP bits for some of these are still a bit
On Fri, May 3, 2013 at 10:42 PM, Chad Versace
wrote:
> +32% on GLBenchmark 2.5 Egypt Offscreen DXT1. Tested on Harris Beach GT3 with
> Android.
>
> Please please please, someone run Piglit on this series. I currently have no
> Haswell hardware that is stable on non-Android Linux, and so can't run
14 matches
Mail list logo
