Thanks Roland and everyone else for the input. Based on the feedback
I'll include the fix in thew next round of deqp fixes.
Iago
On Thu, 2015-01-15 at 15:32 +0100, Roland Scheidegger wrote:
> Am 15.01.2015 um 10:05 schrieb Iago Toral:
> > Hi,
> >
> > We have 16 deqp tests that fail, at least on
On 01/16/2015 05:53 PM, Ian Romanick wrote:
> I also notice that the 12 vertex shaders no longer appear in the hurt
> list or the helped list. Ugh.
I think I botched my report.py changes. Now all vertex shaders show up
as "VS vec4", so results are not properly reported. Ugh^2.
On 01/16/2015 08:15 AM, Matt Turner wrote:
> From: Ian Romanick
>
> Some shaders end up with code that looks something like:
>
> if (some_condition)
> result = 1.0;
> else
> result = 0.0;
>
> This pass converts those if-statements to
>
> result = float(some_conditi
On Friday, January 16, 2015 04:54:44 PM Ian Romanick wrote:
> On 01/16/2015 04:25 PM, Matt Turner wrote:
> > On Thu, Jan 15, 2015 at 6:46 PM, Ian Romanick wrote:
> >> On 01/16/2015 08:15 AM, Matt Turner wrote:
> >>> From: Ian Romanick
> >>>
> >>> Some shaders end up with code that looks something
https://bugs.freedesktop.org/show_bug.cgi?id=88480
--- Comment #2 from Kenneth Graunke ---
Sounds good to me.
--
You are receiving this mail because:
You are the assignee for the bug.
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lis
https://bugs.freedesktop.org/show_bug.cgi?id=88480
--- Comment #1 from Connor Abbott ---
Created attachment 112325
--> https://bugs.freedesktop.org/attachment.cgi?id=112325&action=edit
public PGP key
--
You are receiving this mail because:
You are the assignee for the bug.
___
https://bugs.freedesktop.org/show_bug.cgi?id=88480
Bug ID: 88480
Summary: Requesting fdo account and commit accees to mesa
Product: Mesa
Version: unspecified
Hardware: Other
OS: All
Status: NEW
Severity: n
On Thursday, January 15, 2015 05:08:49 PM Brian Paul wrote:
>
> I have a question on behalf of a coworker. He's using Ubuntu 14.04
> (Mesa 10.1.3, kernel 3.13.0) on a Haswell system. But glxinfo does not
> list support for GL_ARB_transform_feedback[23]. It does list
> ARB_conservative_depth.
>
>
On 01/16/2015 04:25 PM, Matt Turner wrote:
> On Thu, Jan 15, 2015 at 6:46 PM, Ian Romanick wrote:
>> On 01/16/2015 08:15 AM, Matt Turner wrote:
>>> From: Ian Romanick
>>>
>>> Some shaders end up with code that looks something like:
>>>
>>> if (some_condition)
>>> result = 1.0;
>>>
On 01/16/2015 04:25 PM, Matt Turner wrote:
> On Thu, Jan 15, 2015 at 6:46 PM, Ian Romanick wrote:
>> On 01/16/2015 08:15 AM, Matt Turner wrote:
>>> From: Ian Romanick
>>>
>>> Some shaders end up with code that looks something like:
>>>
>>> if (some_condition)
>>> result = 1.0;
>>>
Reviewed-By: Jason Ekstrand
On Jan 15, 2015 9:20 PM, "Connor Abbott" wrote:
> Cc: Jason Ekstrand
> ---
> Seems like we missed fixing this issue during review. Jason, can you commit
> this?
>
> src/glsl/nir/nir_lower_vars_to_ssa.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff
On Thu, Jan 15, 2015 at 6:46 PM, Ian Romanick wrote:
> On 01/16/2015 08:15 AM, Matt Turner wrote:
>> From: Ian Romanick
>>
>> Some shaders end up with code that looks something like:
>>
>> if (some_condition)
>> result = 1.0;
>> else
>> result = 0.0;
>>
>> This pass conver
Cc: Jason Ekstrand
---
Seems like we missed fixing this issue during review. Jason, can you commit
this?
src/glsl/nir/nir_lower_vars_to_ssa.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/glsl/nir/nir_lower_vars_to_ssa.c
b/src/glsl/nir/nir_lower_vars_to_ssa.c
index e3
On 01/16/2015 08:15 AM, Matt Turner wrote:
> From: Ian Romanick
>
> Some shaders end up with code that looks something like:
>
> if (some_condition)
> result = 1.0;
> else
> result = 0.0;
>
> This pass converts those if-statements to
>
> result = float(some_conditi
On 16.01.2015 09:48, Tom Stellard wrote:
> - Use context global scratch buffers, one for each shader type.
AFAICT the code actually uses a single buffer for all shader types. As
we discussed before, that needs to be fixed.
> @@ -208,9 +208,15 @@ void radeon_shader_binary_free_relocs(struct
>
Instead of passing a pointer to the scratch buffer via user sgprs, we
now patch the shader with the buffer address using reloc information
from the LLVM generated ELF.
v2:
- Make sure not to break older LLVM.
---
src/gallium/drivers/radeonsi/si_compute.c | 42 +++
1
v2:
- Use strdup for copying reloc names.
- Free reloc memory.
---
src/gallium/drivers/radeon/r600_pipe_common.h | 8
src/gallium/drivers/radeon/radeon_elf_util.c | 53 +--
src/gallium/drivers/radeon/radeon_elf_util.h | 7
3 files changed, 65 insertions(+)
v2:
- Only emit write SPI_TMPRING_SIZE once per packet.
- Use context global scratch buffers, one for each shader type.
---
src/gallium/drivers/radeon/radeon_elf_util.c| 10 ++-
src/gallium/drivers/radeon/radeon_elf_util.h| 5 +-
src/gallium/drivers/radeonsi/si_compute.c | 46
This was inadvertently disabled by
761e36b4caab4e8e09a4c2b1409a825902fc7d2c.
---
src/gallium/drivers/radeon/radeon_llvm_emit.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/src/gallium/drivers/radeon/radeon_llvm_emit.c
b/src/gallium/drivers/radeon/radeon_llvm_emit.c
index
---
src/gallium/drivers/radeon/radeon_elf_util.c | 6 ++
src/gallium/drivers/radeon/radeon_elf_util.h | 5 +
src/gallium/drivers/radeonsi/si_compute.c| 4 +---
src/gallium/drivers/radeonsi/si_shader.c | 4 +---
4 files changed, 13 insertions(+), 6 deletions(-)
diff --git a/src/gal
The GLSL compiler optimizes the pattern
result = [!]gl_FrontFacing ? x : y;
where x and y are ±0.0 and ±1.0 into
result = [-]b2f([!]gl_FrontFacing);
We can do these expressions in two instructions.
total instructions in shared programs: 5928518 -> 5927775 (-0.01%)
instructions in affecte
On Thu, Jan 15, 2015 at 4:08 PM, Brian Paul wrote:
>
> I have a question on behalf of a coworker. He's using Ubuntu 14.04 (Mesa
> 10.1.3, kernel 3.13.0) on a Haswell system. But glxinfo does not list
> support for GL_ARB_transform_feedback[23]. It does list
> ARB_conservative_depth.
>
> Accordi
On Thu, Jan 15 2015, Matt Turner wrote:
> I wouldn't put the spaces after and before the [ ] (there's an
> occurrence of this in the previous patch as well, that gets removed in
> this one).
OK. I'll fix that.
> Both are:
>
> Reviewed-by: Matt Turner
Thanks!
-Carl
pgpPQ3Jww2upP.pgp
Descripti
Reviewed-by: Connor Abbott
On Thu, Jan 15, 2015 at 7:39 PM, Brian Paul wrote:
> Silences a compiler warning.
> ---
> src/glsl/nir/nir_validate.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/glsl/nir/nir_validate.c b/src/glsl/nir/nir_validate.c
> index 77bc5af..22
Silences a compiler warning.
---
src/glsl/nir/nir_validate.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/glsl/nir/nir_validate.c b/src/glsl/nir/nir_validate.c
index 77bc5af..228dce2 100644
--- a/src/glsl/nir/nir_validate.c
+++ b/src/glsl/nir/nir_validate.c
@@ -381,7 +38
On Thu, Jan 8, 2015 at 10:59 PM, Matt Turner wrote:
> total instructions in shared programs: 5886767 -> 5881642 (-0.09%)
> instructions in affected programs: 532199 -> 527074 (-0.96%)
> GAINED:9
> LOST: 6
> ---
On top of Ian's i
I believe this is expected. See
http://people.freedesktop.org/~imirkin/glxinfo/glxinfo.html -- it's
missing tf2+/indirect on Haswell.
On Thu, Jan 15, 2015 at 7:08 PM, Brian Paul wrote:
>
> I have a question on behalf of a coworker. He's using Ubuntu 14.04 (Mesa
> 10.1.3, kernel 3.13.0) on a Hasw
Btw, apparently nir_validate.c has the same problem on line 391.
On Thu, Jan 15, 2015 at 6:31 PM, Connor Abbott wrote:
> Reviewed-by: Connor Abbott
>
> On Thu, Jan 15, 2015 at 6:16 PM, Brian Paul wrote:
>> v2: use proper argument
>> ---
>> src/glsl/nir/nir.c | 2 +-
>> 1 file changed, 1 insert
I have a question on behalf of a coworker. He's using Ubuntu 14.04
(Mesa 10.1.3, kernel 3.13.0) on a Haswell system. But glxinfo does not
list support for GL_ARB_transform_feedback[23]. It does list
ARB_conservative_depth.
According to the driver code (from master, 10.1.3 lacks ARB_textur
https://bugs.freedesktop.org/show_bug.cgi?id=88478
Bug ID: 88478
Summary: #error " has been replaced by "
Product: Mesa
Version: git
Hardware: x86-64 (AMD64)
OS: FreeBSD
Status: NEW
Keywords: regression
Reviewed-by: Connor Abbott
On Thu, Jan 15, 2015 at 6:16 PM, Brian Paul wrote:
> v2: use proper argument
> ---
> src/glsl/nir/nir.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/glsl/nir/nir.c b/src/glsl/nir/nir.c
> index f112db8..81dec1c 100644
> --- a/src/glsl/ni
On Fri, Dec 19, 2014 at 6:43 PM, Carl Worth wrote:
> We don't actually have the code for the shader cache just yet, but
> this configure machinery puts everything in place so that the shader
> cache can be optionally compiled in.
>
> Specifically, if the user passes no option (neither
> --disable-
v2: use proper argument
---
src/glsl/nir/nir.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/glsl/nir/nir.c b/src/glsl/nir/nir.c
index f112db8..81dec1c 100644
--- a/src/glsl/nir/nir.c
+++ b/src/glsl/nir/nir.c
@@ -1541,7 +1541,7 @@ static bool
visit_tex_src(nir_tex_instr
On 01/15/2015 10:05 PM, Iago Toral wrote:
> Hi,
>
> We have 16 deqp tests that fail, at least on i965, because of
> insufficient precision of the mod GLSL function.
>
> Mesa lowers mod(x,y) to y * fract(x,y) so there can be some precision
> lost due to fract operation. Since the result is multipl
On Thursday, January 15, 2015 01:06:24 PM Jason Ekstrand wrote:
> On Thu, Jan 15, 2015 at 1:01 PM, Eric Anholt wrote:
>
> > Jason Ekstrand writes:
> >
> > > ---
> > > src/glsl/nir/nir.h| 2 +-
> > > src/mesa/Makefile.sources | 2 -
> > > src/mesa/drivers/dri/
On Thu, Jan 15, 2015 at 5:28 PM, Brian Paul wrote:
> Warning seen with gcc 4.8.2
> ---
> src/glsl/nir/nir.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/glsl/nir/nir.c b/src/glsl/nir/nir.c
> index f112db8..6d6f910 100644
> --- a/src/glsl/nir/nir.c
> +++ b/src/glsl/
Reviewed-by: Matt Turner
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
To fix MSVC build.
---
src/mesa/main/renderbuffer.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/mesa/main/renderbuffer.c b/src/mesa/main/renderbuffer.c
index cc97ab2..98f3c13 100644
--- a/src/mesa/main/renderbuffer.c
+++ b/src/mesa/main/renderbuffer.c
@@ -38,6 +38,8 @
Warning seen with gcc 4.8.2
---
src/glsl/nir/nir.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/glsl/nir/nir.c b/src/glsl/nir/nir.c
index f112db8..6d6f910 100644
--- a/src/glsl/nir/nir.c
+++ b/src/glsl/nir/nir.c
@@ -1541,7 +1541,7 @@ static bool
visit_tex_src(nir_tex_in
Series is
Reviewed-by: Connor Abbott
As you know, I have a branch that generalizes this and adds a worklist
for SSA definitions as well. But I think we won't want the SSA-def
worklist for DCE, since just like with phi-node placement we only ever
push something onto the worklist once, and we use
On Thu, Jan 15, 2015 at 07:25:56PM +0100, Niels Ole Salscheider wrote:
> Since 8e7df519bd8556591794b2de08a833a67e34d526, we initialise all targets in
> clover. This fixes bug 85189.
>
> Signed-off-by: Niels Ole Salscheider
Reviewed-by: Tom Stellard
> ---
> configure.ac | 2 +-
> 1 file changed,
On Thu, Jan 15, 2015 at 1:01 PM, Eric Anholt wrote:
> Jason Ekstrand writes:
>
> > ---
> > src/glsl/nir/nir.h| 2 +-
> > src/mesa/Makefile.sources | 2 -
> > src/mesa/drivers/dri/i965/intel_fbo.c | 2 +-
> > src/mesa/main/set.c | 346
> ---
Jason Ekstrand writes:
> ---
> src/glsl/nir/nir.h| 2 +-
> src/mesa/Makefile.sources | 2 -
> src/mesa/drivers/dri/i965/intel_fbo.c | 2 +-
> src/mesa/main/set.c | 346
> --
> src/mesa/main/set.h
Jason Ekstrand writes:
> I just pushed NIR to master. I wanted to write a couple of notes about the
> interaction between that code and MSVC and what the current plan is. Right
> now, NIR won't build with MSVC; at least not MSVC 2010 (or is it 11?) as it
> uses quite a bit of C99. It should be
On Thu, Jan 15, 2015 at 11:33:22AM -0800, Kenneth Graunke wrote:
> On Thursday, January 15, 2015 08:56:23 AM Ben Widawsky wrote:
> > On Thu, Jan 15, 2015 at 01:41:14AM -0800, Kenneth Graunke wrote:
> > > There's some debate about whether we should use Meta or BLORP,
> > > but either should run circ
On Thu, 15 Jan 2015 15:32:59 +0100, Roland Scheidegger
wrote:
Am 15.01.2015 um 10:05 schrieb Iago Toral:
Hi,
We have 16 deqp tests that fail, at least on i965, because of
insufficient precision of the mod GLSL function.
Mesa lowers mod(x,y) to y * fract(x,y) so there can be some precision
On 15.01.2015 19:35, Jan Vesely wrote:
On Thu, 2015-01-15 at 19:26 +0100, Tobias Klausmann wrote:
I: Program returns random data in a function
E: Mesa no-return-in-nonvoid-function ../../src/mesa/main/format_utils.c:180
E: Mesa no-return-in-nonvoid-function ../../src/mesa/main/glformats.c:2714
On Thu, Jan 08 2015, Jose Fonseca wrote:
> Note that Windows build is only supported with SCons. Never with
> autobuild.
OK. That's good for me to learn. I've requested that the folks doing our
automated build testing here will also start testing scons builds, so
hopefully on our end we can avoid
On Thursday, January 15, 2015 08:56:23 AM Ben Widawsky wrote:
> On Thu, Jan 15, 2015 at 01:41:14AM -0800, Kenneth Graunke wrote:
> > There's some debate about whether we should use Meta or BLORP,
> > but either should run circles around the BLT engine.
> >
> > Improves performance in "copypixrate
On Thursday, January 15, 2015 07:09:40 AM Jason Ekstrand wrote:
> On Jan 15, 2015 1:49 AM, "Kenneth Graunke" wrote:
> >
> > On Thursday, January 15, 2015 01:41:14 AM Kenneth Graunke wrote:
> > > There's some debate about whether we should use Meta or BLORP,
> > > but either should run circles arou
From: Ian Romanick
Some shaders end up with code that looks something like:
if (some_condition)
result = 1.0;
else
result = 0.0;
This pass converts those if-statements to
result = float(some_condition);
total instructions in shared programs: 5934568 -> 5921818 (-0
Yes, the bit shifting can be a crude approximation.
llvmpipe did that everywhere but we had to fix it in a few places, e.g.:
http://cgit.freedesktop.org/mesa/mesa/commit/?id=1569b3e536da9337a28a16d0cc6ed07043bf094b
The multiplication is unavoidable (*), but one can avoid the division of
(2*n
On Thu, 2015-01-15 at 14:47 +, Jose Fonseca wrote:
> On 14/01/15 20:53, Jan Vesely wrote:
> > these two patches fix cca 90 warnings when compiling src/mesa.
> > The idea is to reduce warning bloat so that the remaining stuff
> > can be looked into to check whether the warning hints at bigger is
v2: s/unsigned int/unsigned/ in prog_optimize.c
Signed-off-by: Jan Vesely
---
src/mesa/main/bufferobj.c | 18 +-
src/mesa/main/buffers.c | 2 +-
src/mesa/main/clear.c | 2 +-
src/mesa/main/dlist.c | 2 +-
https://bugs.freedesktop.org/show_bug.cgi?id=84019
--- Comment #3 from Jan Vesely ---
I have added mesa git access to the request
--
You are receiving this mail because:
You are the assignee for the bug.
___
mesa-dev mailing list
mesa-dev@lists.freede
https://bugs.freedesktop.org/show_bug.cgi?id=84019
Jan Vesely changed:
What|Removed |Added
Summary|Request commit access to|Request commit access to
On Thu, 2015-01-15 at 19:26 +0100, Tobias Klausmann wrote:
> I: Program returns random data in a function
> E: Mesa no-return-in-nonvoid-function ../../src/mesa/main/format_utils.c:180
> E: Mesa no-return-in-nonvoid-function ../../src/mesa/main/glformats.c:2714
Are these errors generated by a comp
Since 8e7df519bd8556591794b2de08a833a67e34d526, we initialise all targets in
clover. This fixes bug 85189.
Signed-off-by: Niels Ole Salscheider
---
configure.ac | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index c72fe92..1761c32 100644
--- a/con
I: Program returns random data in a function
E: Mesa no-return-in-nonvoid-function ../../src/mesa/main/format_utils.c:180
E: Mesa no-return-in-nonvoid-function ../../src/mesa/main/glformats.c:2714
---
src/mesa/main/format_utils.c | 1 +
src/mesa/main/glformats.c| 1 +
2 files changed, 2 insert
The series is
Reviewed-by: Matt Turner
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
https://bugs.freedesktop.org/show_bug.cgi?id=88467
Jason Ekstrand changed:
What|Removed |Added
Status|NEW |ASSIGNED
Assignee|mesa-dev@li
We already have search_pre_hashed. This makes the APIs match better.
---
src/glsl/nir/nir_lower_locals_to_regs.c | 2 +-
src/mesa/main/hash.c| 2 +-
src/util/hash_table.c | 4 ++--
src/util/hash_table.h | 4 ++--
4 files changed, 6 insertion
---
src/glsl/nir/nir.h| 2 +-
src/mesa/Makefile.sources | 2 -
src/mesa/drivers/dri/i965/intel_fbo.c | 2 +-
src/mesa/main/set.c | 346 --
src/mesa/main/set.h | 94 -
src/mesa/main/sh
Previously, the set API required the user to do all of the hashing of keys
as it passed them in. Since the hashing function is intrinsically tied to
the comparison function, it makes sense for the hash set to know about
it. Also, it makes for a somewhat clumsy API as the user is constantly
callin
https://bugs.freedesktop.org/show_bug.cgi?id=88467
Bug ID: 88467
Summary: nir.c:140: error: ‘nir_src’ has no member named ‘ssa’
Product: Mesa
Version: git
Hardware: x86-64 (AMD64)
OS: All
Status: NEW
Keywo
https://bugs.freedesktop.org/show_bug.cgi?id=84019
Jan Vesely changed:
What|Removed |Added
Assignee|sitewranglers@lists.freedes |mesa-dev@lists.freedesktop.
On Thu, Jan 15, 2015 at 01:41:14AM -0800, Kenneth Graunke wrote:
> There's some debate about whether we should use Meta or BLORP,
> but either should run circles around the BLT engine.
>
> Improves performance in "copypixrate -blit -back" (from Mesa demos)
> by 232.037% +/- 3.15795% (n=10) on Broa
Jason Ekstrand writes:
> This looks fine to me. We should probably also do this for snorm formats.
> I don't care if that's part of this or in a separate patch.
> --Jason
The snorm formats are a bit more fiddly because the hardware doesn't
quite seem to be doing what I'd expect. For example, wh
On 15/01/15 15:39, Jason Ekstrand wrote:
I just pushed NIR to master. I wanted to write a couple of notes about
the interaction between that code and MSVC and what the current plan
is. Right now, NIR won't build with MSVC; at least not MSVC 2010 (or is
it 11?) as it uses quite a bit of C99. It
A worklist is a common concept in optimizations. This adds a structure
that we can reuse for many different types of optimizations.
---
src/glsl/Makefile.sources | 2 +
src/glsl/nir/nir_worklist.c | 144
src/glsl/nir/nir_worklist.h | 91 ++
This is a rework of the liveness algorithm using a worklist as suggested by
Connor. Doing so reduces the number of times we walk over the instructions
because we don't have to do an entire pointless walk over the instructions
just to figure out it's time to stop. Also, the stuff after the last lo
I just pushed NIR to master. I wanted to write a couple of notes about the
interaction between that code and MSVC and what the current plan is. Right
now, NIR won't build with MSVC; at least not MSVC 2010 (or is it 11?) as it
uses quite a bit of C99. It should be mostly MSVC 2013 safe except for
On Jan 15, 2015 12:27 AM, "Eero Tamminen" wrote:
>
> Hi,
>
> On 01/15/2015 03:04 AM, Matt Turner wrote:
>>
>> On Wed, Jan 14, 2015 at 1:52 PM, Matt Turner wrote:
>>>
>>> On Wed, Jan 14, 2015 at 1:29 PM, Matt Turner wrote:
glsl: Optimize certain if-statements to just casts from the cond
On Jan 15, 2015 1:49 AM, "Kenneth Graunke" wrote:
>
> On Thursday, January 15, 2015 01:41:14 AM Kenneth Graunke wrote:
> > There's some debate about whether we should use Meta or BLORP,
> > but either should run circles around the BLT engine.
> >
> > Improves performance in "copypixrate -blit -bac
On 14/01/15 20:53, Jan Vesely wrote:
these two patches fix cca 90 warnings when compiling src/mesa.
The idea is to reduce warning bloat so that the remaining stuff
can be looked into to check whether the warning hints at bigger issue
gcc 4.9.2 with Wextra:
from
2 -Warray-bounds
29 -
Am 15.01.2015 um 10:05 schrieb Iago Toral:
> Hi,
>
> We have 16 deqp tests that fail, at least on i965, because of
> insufficient precision of the mod GLSL function.
>
> Mesa lowers mod(x,y) to y * fract(x,y) so there can be some precision
> lost due to fract operation. Since the result is multip
https://bugs.freedesktop.org/show_bug.cgi?id=88275
--- Comment #21 from Eugene ---
Created attachment 112298
--> https://bugs.freedesktop.org/attachment.cgi?id=112298&action=edit
GPU crash dump from /sys/class/drm/card0/error
--
You are receiving this mail because:
You are the assignee for th
https://bugs.freedesktop.org/show_bug.cgi?id=88275
Eugene changed:
What|Removed |Added
Attachment #112071|0 |1
is obsolete|
https://bugs.freedesktop.org/show_bug.cgi?id=88275
Eugene changed:
What|Removed |Added
Attachment #112070|0 |1
is obsolete|
https://bugs.freedesktop.org/show_bug.cgi?id=88275
Eugene changed:
What|Removed |Added
Attachment #112069|0 |1
is obsolete|
https://bugs.freedesktop.org/show_bug.cgi?id=88275
--- Comment #17 from Eugene ---
Greetings. I just updated to latest Mesa git and drm-intel-nightly after which
new backtrace and even GPU crash appeared again. Mouse pointer is late now for
hand movements. Although it says acceleration is turning
We were incorrectly attributing VS time to FS8 on Gen8+, which now use
fs_visitor for vertex shaders.
We don't hit this for geometry shaders yet, but we may as well add
support now - the fix is obvious, and we'll just forget later.
Signed-off-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/br
On Thursday, January 15, 2015 01:41:14 AM Kenneth Graunke wrote:
> There's some debate about whether we should use Meta or BLORP,
> but either should run circles around the BLT engine.
>
> Improves performance in "copypixrate -blit -back" (from Mesa demos)
> by 232.037% +/- 3.15795% (n=10) on Broa
There's some debate about whether we should use Meta or BLORP,
but either should run circles around the BLT engine.
Improves performance in "copypixrate -blit -back" (from Mesa demos)
by 232.037% +/- 3.15795% (n=10) on Broadwell GT3e.
Signed-off-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965
On Thursday, January 15, 2015 10:21:30 AM Tapani wrote:
> On 01/14/2015 05:59 PM, Samuel Iglesias Gonsálvez wrote:
> > Hello,
> >
> > Mesa fails the following dEQP GLES 3 test:
> >
> > dEQP-GLES3.functional.fbo.api.attachment_query_default_fbo
> >
> > This test calls glGetFramebufferAttachmentPa
Hi,
We have 16 deqp tests that fail, at least on i965, because of
insufficient precision of the mod GLSL function.
Mesa lowers mod(x,y) to y * fract(x,y) so there can be some precision
lost due to fract operation. Since the result is multiplied by y the
total precision lost usually grows together
Hi,
On 01/15/2015 03:04 AM, Matt Turner wrote:
On Wed, Jan 14, 2015 at 1:52 PM, Matt Turner wrote:
On Wed, Jan 14, 2015 at 1:29 PM, Matt Turner wrote:
glsl: Optimize certain if-statements to just casts from the condition
Cherry-picked to master, the shader-db results are
total instruction
On 01/14/2015 05:59 PM, Samuel Iglesias Gonsálvez wrote:
Hello,
Mesa fails the following dEQP GLES 3 test:
dEQP-GLES3.functional.fbo.api.attachment_query_default_fbo
This test calls glGetFramebufferAttachmentParameteriv() to get
FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE information from COLOR_ATTACHM
88 matches
Mail list logo
