Am 17.04.2013 01:13, schrieb Tom Stellard:
From: Tom Stellard
The LLVM C API is considered stable and should never change, so it
is much more desirable to use than the LLVM C++ API, which is constantly in
flux.
Looks good on first glance, but I would separate the initialization and
finding t
On Wed, Apr 17, 2013 at 07:25:12AM +1000, Dave Airlie wrote:
> >> Those are just ideas. I'm open to discussion.
> >
> > The driver is disabled by default and needs to be enabled via
> > --with-gallium-drivers=i965.
>
> I think a warning + maybe something like
> --with-gallium-drivers=i965g-unoffi
- Original Message -
> On Wed, Apr 17, 2013 at 8:09 AM, Marek Olšák wrote:
> > On Tue, Apr 16, 2013 at 9:58 PM, Chia-I Wu wrote:
> >>
> >> On Wed, Apr 17, 2013 at 12:58 AM, Matt Turner wrote:
> >>>
> >>> I think everything Marek said was correct. If you could extend Gallium
> >>> to co
Removes the need of API_DEFINES.
Cc: Brian Paul
---
Brian, do you have somewhere an updated branch of your remove-mfeatures work?
I'd like to rebase my remove API_DEFINES patch.
src/gallium/targets/egl-static/Makefile.am | 12 +++-
1 file changed, 11 insertions(+), 1 deletion(-)
dif
From: José Fonseca
Details on docs/llvmpipe.html
---
bin/perf-annotate-jit | 240
configure.ac |7 -
docs/llvmpipe.html | 40 ++--
src/gallium/auxiliary/gallivm/lp_bld_debug.cpp
Series looks good to me.
Reviewed-by: Jose Fonseca
- Original Message -
> The specification says that the geometry shader should exit if the
> number of emitted vertices is bigger or equal to max_output_vertices and
> we can't do that because we're running in the SoA mode, which means t
https://bugs.freedesktop.org/show_bug.cgi?id=63404
--- Comment #2 from Kristian Høgsberg ---
Joe, if you're up for compiling a new kernel, here are two patches you can try
http://lists.freedesktop.org/archives/dri-devel/2013-April/037101.html
against Linus git master.
--
You are receiving thi
https://bugs.freedesktop.org/show_bug.cgi?id=63404
--- Comment #3 from Joe Konno ---
I'm up for it. I'll give it a go later today.
(In reply to comment #2)
> Joe, if you're up for compiling a new kernel, here are two patches you can
> try
>
> http://lists.freedesktop.org/archives/dri-devel/2013
On 04/17/2013 12:07 AM, Tapani Pälli wrote:
patch fixes a crash that happens if glTexSubImage2D is called with a
negative xoffset.
Signed-off-by: Tapani Pälli
---
src/mesa/main/teximage.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/mesa/main/teximage.c b/src/mesa/m
On 04/17/2013 06:34 AM, jfons...@vmware.com wrote:
From: José Fonseca
Details on docs/llvmpipe.html
---
bin/perf-annotate-jit | 240
configure.ac |7 -
docs/llvmpipe.html |
On 04/17/2013 03:36 AM, Andreas Boll wrote:
Removes the need of API_DEFINES.
Cc: Brian Paul
---
Brian, do you have somewhere an updated branch of your remove-mfeatures work?
I'd like to rebase my remove API_DEFINES patch.
I'm trickling out the patches from that branch. I'll push a few more
https://bugs.freedesktop.org/show_bug.cgi?id=63645
Priority: medium
Bug ID: 63645
Assignee: mesa-dev@lists.freedesktop.org
Summary: g3dvl should be disabled when not building gallium
drivers
Severity: normal
Classificatio
- Original Message -
> On 04/17/2013 06:34 AM, jfons...@vmware.com wrote:
> > From: José Fonseca
> >
> > Details on docs/llvmpipe.html
> > ---
> > bin/perf-annotate-jit | 240
> >
> > configure.ac |
On 04/17/2013 02:31 AM, Jose Fonseca wrote:
- Original Message -
On Wed, Apr 17, 2013 at 8:09 AM, Marek Olšák wrote:
On Tue, Apr 16, 2013 at 9:58 PM, Chia-I Wu wrote:
On Wed, Apr 17, 2013 at 12:58 AM, Matt Turner wrote:
I think everything Marek said was correct. If you could ex
On 04/16/2013 01:58 PM, Chia-I Wu wrote:
If you could agree with me, then, technically, we need to make the new
driver less confusing. The driver is disabled by default, of course.
It can be enabled with --with-gallium-drivers=i965 now. Would it help
if I change it to something like
--enable-g
Am 17.04.2013 14:34, schrieb jfons...@vmware.com:
> From: José Fonseca
>
> Details on docs/llvmpipe.html
> ---
> bin/perf-annotate-jit | 240
>
> configure.ac |7 -
> docs/llvmpipe.html
---
docs/relnotes/9.2.html | 11 ---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/docs/relnotes/9.2.html b/docs/relnotes/9.2.html
index ec9cc47..0dcc960 100644
--- a/docs/relnotes/9.2.html
+++ b/docs/relnotes/9.2.html
@@ -44,8 +44,13 @@ Note: some of the new features are
On Wed, Apr 17, 2013 at 8:27 AM, Andreas Boll
wrote:
> ---
> docs/relnotes/9.2.html | 11 ---
> 1 file changed, 8 insertions(+), 3 deletions(-)
>
> diff --git a/docs/relnotes/9.2.html b/docs/relnotes/9.2.html
> index ec9cc47..0dcc960 100644
> --- a/docs/relnotes/9.2.html
> +++ b/docs/re
On 04/16/2013 05:09 PM, Marek Olšák wrote:
On Tue, Apr 16, 2013 at 9:58 PM, Chia-I Wu wrote:
On Wed, Apr 17, 2013 at 12:58 AM, Matt Turner wrote:
I think everything Marek said was correct. If you could extend Gallium
to consume GLSL IR it might actually be an interesting project.
Yes, it
On 04/16/2013 11:07 PM, Tapani Pälli wrote:
patch fixes a crash that happens if glTexSubImage2D is called with a
negative xoffset.
Signed-off-by: Tapani Pälli
---
src/mesa/main/teximage.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/mesa/main/teximage.c b/src/mesa/
https://bugs.freedesktop.org/show_bug.cgi?id=63650
Priority: medium
Bug ID: 63650
Assignee: mesa-dev@lists.freedesktop.org
Summary: gallium/opencl opencl-examples test failures
Severity: minor
Classification: Unclassified
OS:
Brian Paul writes:
> Use alternate code in intel, r200, radeon drivers.
> v2: use conditional operator instead of bit shifting
> ---
> src/mesa/drivers/dri/i915/intel_tris.c |4 +++-
> src/mesa/drivers/dri/r200/r200_state.c |5 +++--
> src/mesa/drivers/dri/r200/r200_swtcl.c |
Other than a whitespace comment, this series is:
Reviewed-by: Eric Anholt
pgp7WGJWJkC9K.pgp
Description: PGP signature
___
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=63650
--- Comment #1 from Tom Stellard ---
Can you post the config.log file from your mesa build?
--
You are receiving this mail because:
You are the assignee for the bug.
___
mesa-dev mailing list
mesa-dev
https://bugs.freedesktop.org/show_bug.cgi?id=63650
--- Comment #2 from Tom Stellard ---
What commits of Mesa and LLVM are you using?
You may want to try updating your mesa and llvm to the latest versions, if you
haven't already.
--
You are receiving this mail because:
You are the assignee for
Marek Olšák writes:
> The next patch makes sure _NEW_BUFFER_OBJECT is not needlessly set
> for this code.
This seems like a pretty dubious optimization -- on UMA systems you're
increasing the memory bandwidth usage in the "&& data" case, and only
trying to eliminate update_array_object_max_eleme
Marek Olšák writes:
> We don't want to set the flag for Gallium.
>
> I think only swrast needs the flag to be set for occlusion queries.
>
> Reviewed-by: Brian Paul
>
> v2: fix stats_wm updates in i965
Reviewed-by: Eric Anholt
pgp2KgQ82ApWZ.pgp
Description: PGP signature
Marek Olšák writes:
> _NEW_TRANSFORM_FEEDBACK is not used by core Mesa, so it can be removed.
> Instead, an new private flag is added to i965 to serve the same purpose.
>
> If you're new to this:
> diff --git a/src/mesa/drivers/dri/i965/brw_state_upload.c
> b/src/mesa/drivers/dri/i965/brw_state
https://bugs.freedesktop.org/show_bug.cgi?id=63650
--- Comment #3 from Iaroslav ---
I build with OBS and do not have all logs.
2-3 weeks ago, all works.
I had some problem with the libclc corrected in this way
./configure.py --prefix=/usr --with-llvm-config=/usr/bin/llvm-config
for 32/64 bit i
https://bugs.freedesktop.org/show_bug.cgi?id=63650
--- Comment #4 from Iaroslav ---
Created attachment 78147
--> https://bugs.freedesktop.org/attachment.cgi?id=78147&action=edit
strace log
--
You are receiving this mail because:
You are the assignee for the bug.
__
On 16 April 2013 16:21, Eric Anholt wrote:
> The upside is less CPU overhead in fiddling with GL error handling, the
> ability to use the constant color write message in most cases, and no GLSL
> clear shaders appearing in MESA_GLSL=dump output. The downside is more
> batch flushing and a total
We were always treating the vertex index as a scalar but when the
shader is using indirect addressing it will be a vector of indices
for each channel. This was causing some nasty crashes insides
LLVM.
Signed-off-by: Zack Rusin
---
src/gallium/auxiliary/draw/draw_llvm.c | 34 ++
---
src/mesa/drivers/dri/r200/r200_swtcl.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/mesa/drivers/dri/r200/r200_swtcl.c
b/src/mesa/drivers/dri/r200/r200_swtcl.c
index 274e338..0f1e013 100644
--- a/src/mesa/drivers/dri/r200/r200_swtcl.c
+++ b/src/mesa/drivers/dri/r200
Am 17.04.2013 21:17, schrieb Zack Rusin:
> We were always treating the vertex index as a scalar but when the
> shader is using indirect addressing it will be a vector of indices
> for each channel. This was causing some nasty crashes insides
> LLVM.
>
> Signed-off-by: Zack Rusin
> ---
> src/gall
On Wed, Apr 17, 2013 at 1:40 PM, Laurent Carlier wrote:
> ---
> src/mesa/drivers/dri/r200/r200_swtcl.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/mesa/drivers/dri/r200/r200_swtcl.c
> b/src/mesa/drivers/dri/r200/r200_swtcl.c
> index 274e338..0f1e013 100644
> --- a
This list appears in the fixed bugs section of the release notes.
NOTE: This is a candidate for the stable branches.
---
bin/bugzilla_mesa.sh | 43 +++
1 file changed, 43 insertions(+)
create mode 100755 bin/bugzilla_mesa.sh
diff --git a/bin/bugzilla_me
---
docs/devinfo.html |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/docs/devinfo.html b/docs/devinfo.html
index 7176824..f5eb4d4 100644
--- a/docs/devinfo.html
+++ b/docs/devinfo.html
@@ -206,8 +206,8 @@ branch is relevant.
Create a docs/relnotes/x.y.z.html file.
The following patch set makes the necessary changes to support geometry shaders
in the nv50 driver. There are no piglit tests yet for geometry shader corner
cases yet, so these changes were tested with all of the GS demos in mesa/demos
and several corner case tests, using Paul Berry's "gs" branch
From: Christoph Bumiller
v2 (Bryan Cain ): Fix emission of PFETCH instructions
using the SHL form.
---
src/gallium/drivers/nv50/codegen/nv50_ir.h |1 +
.../drivers/nv50/codegen/nv50_ir_emit_nv50.cpp | 62 ++--
src/gallium/drivers/nv50/codegen/nv50_ir_print.cpp |
Instead of emitting an SHL 4 io an address register on the TGSI ARL and UARL
instructions, emit the shift when the loaded address is actually used. This
is necessary because input vertex and attribute indices in geometry shaders on
nv50 need to be shifted left by 2 instead of 4.
---
.../drivers/n
Layer output probably doesn't work yet, but other than that everything seems
to be working.
---
.../drivers/nv50/codegen/nv50_ir_emit_nv50.cpp | 25 +++-
src/gallium/drivers/nv50/nv50_program.c| 17 +
src/gallium/drivers/nv50/nv50_shader_state.c
Hi,
I put a 3Ghz Core2 Q35 box i found in the office to good use (so much
nicer than a pineview atom).
http://people.freedesktop.org/~airlied/piglit/i915c/
is a full run with i915c forced to advertise GL2.0 using the stub
occlusion query hack in drirc, and i915g advertising GL2.1, and built
agai
On 04/17/2013 01:59 PM, Dave Airlie wrote:
Hi,
I put a 3Ghz Core2 Q35 box i found in the office to good use (so much
nicer than a pineview atom).
http://people.freedesktop.org/~airlied/piglit/i915c/
is a full run with i915c forced to advertise GL2.0 using the stub
occlusion query hack in drirc
On 04/16/2013 04:21 PM, Eric Anholt wrote:
Improves GLB2.7 performance 0.13% +/- 0.09% (n=104/105, outliers removed).
More importantly, once color glClear()s are done through blorp in the next
commit, this reduces regression in GLES3 conformance tests that rely on
queueing up many glClear()s and
On 04/16/2013 05:21 PM, Eric Anholt wrote:
The next commit introduces what is apparently our first one, which tripped
over this in glReadPixels.
---
src/mesa/drivers/dri/intel/intel_blit.c | 21 ++---
1 file changed, 14 insertions(+), 7 deletions(-)
Commit message here shou
On 04/16/2013 04:21 PM, Eric Anholt wrote:
Improves GLB2.7 performance 0.13% +/- 0.09% (n=104/105, outliers removed).
More importantly, once color glClear()s are done through blorp in the next
commit, this reduces regression in GLES3 conformance tests that rely on
queueing up many glClear()s and
On 04/16/2013 05:21 PM, Eric Anholt wrote:
This should already be handled by _mesa_base_tex_format() calls in
TexImage*.
---
NOTE!
I've replaced the actual patch full of unindentation of code described by:
src/mesa/main/texformat.c | 1184 -
1 file
On 04/17/2013 02:53 PM, Kenneth Graunke wrote:
On 04/16/2013 05:21 PM, Eric Anholt wrote:
The next commit introduces what is apparently our first one, which
tripped
over this in glReadPixels.
---
src/mesa/drivers/dri/intel/intel_blit.c | 21 ++---
1 file changed, 14 insertio
On 04/17/2013 01:02 PM, Andreas Boll wrote:
This list appears in the fixed bugs section of the release notes.
NOTE: This is a candidate for the stable branches.
That's really similar to the one I came up with for the 9.1.1 release
(attached). I think you want -n on sort, though. I also had
On 04/16/2013 06:08 PM, Dave Airlie wrote:
From: Dave Airlie
For some reason I made this happen under indirect rendering,
I think we might have a leak, valgrind gave out, so I said I'd
fix the basic problem.
NOTE: This is a candidate for stable branches.
Signed-off-by: Dave Airlie
Review
we were using the wrong vars, reporting incorrect stream output
statistics.
Signed-off-by: Zack Rusin
---
src/gallium/drivers/softpipe/sp_query.c | 20 +++-
1 file changed, 15 insertions(+), 5 deletions(-)
diff --git a/src/gallium/drivers/softpipe/sp_query.c
b/src/gallium/dri
we were ignoring leading/provoking vertex settings which was
breaking decomposition of some strips.
Signed-off-by: Zack Rusin
---
src/gallium/auxiliary/draw/draw_so_emit_tmp.h |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/gallium/auxiliary/draw/draw_so_emit_tmp.h
b/
On 12 April 2013 01:26, Christian König wrote:
> From: Christian König
>
> Since we now have UVD support we should enable them by default.
>
> Signed-off-by: Christian König
>
Those of us who don't build Gallium now need to add
"--disable-gallium-g3dvl" to our configure parameters, or the buil
This series does two things: first, it plumbs ShaderCompilerOptions into
do_common_optimization(), since optimization & lowering passes may want
to consult that.
Secondly, it adds an optimization pass that converts (matrix * vector)
operations to (vector * matrixTranspose) for certain built-in mat
This code is copied from _mesa_init_shader_state().
Signed-off-by: Kenneth Graunke
---
src/glsl/standalone_scaffolding.cpp | 12
1 file changed, 12 insertions(+)
diff --git a/src/glsl/standalone_scaffolding.cpp
b/src/glsl/standalone_scaffolding.cpp
index 0c1f52f..3d9e1bf 100644
--
We can't include shaderobj.h from the standalone utilities, so we
unfortunately have to copy this function.
Signed-off-by: Kenneth Graunke
---
src/glsl/standalone_scaffolding.h | 17 +
1 file changed, 17 insertions(+)
diff --git a/src/glsl/standalone_scaffolding.h
b/src/glsl/st
This flag essentially tells the compiler whether it prefers
dot products or multiply/adds for matrix operations. As such,
ShaderCompilerOptions seems like the right place for it.
This also lets us specify it on a per-stage basis. This patch makes all
existing users set the flag for the Vertex Sh
do_common_optimization may need to make choices about whether to emit
certain kinds of instructions. gl_context::ShaderCompilerOptions
contains exactly that information, so it makes sense to pass it in.
Rather than passing the whole array, pass the structure for the stage
that's currently being w
Doing matrix multiplies with DP4s is fewer instructions than MUL/ADD,
especially since we don't support MAD in the vertex shader.
Not observed to improve performance in any fixed function applications,
but is useful for the next patch.
I've left this unset for the fragment shader because the scal
This pass flips (matrix * vector) operations to (vector *
matrixTranspose) for certain built-in matrices (currently
gl_ModelViewProjectionMatrix and gl_TextureMatrix).
This is equivalent, but results in dot products rather than multiplies
and adds. On some hardware, this is more efficient.
This
On 18/04/13 01:21, Paul Berry wrote:
> On 12 April 2013 01:26, Christian König wrote:
>
>> From: Christian König
>>
>> Since we now have UVD support we should enable them by default.
>>
>> Signed-off-by: Christian König
>>
>
> Those of us who don't build Gallium now need to add
> "--disable-ga
Ian Romanick writes:
> On 04/16/2013 04:21 PM, Eric Anholt wrote:
>> Improves GLB2.7 performance 0.13% +/- 0.09% (n=104/105, outliers removed).
>> More importantly, once color glClear()s are done through blorp in the next
>> commit, this reduces regression in GLES3 conformance tests that rely on
Kenneth Graunke writes:
> On 04/16/2013 05:21 PM, Eric Anholt wrote:
>> The next commit introduces what is apparently our first one, which tripped
>> over this in glReadPixels.
>> ---
>> src/mesa/drivers/dri/intel/intel_blit.c | 21 ++---
>> 1 file changed, 14 insertions(+),
On 18/04/13 01:55, Emil Velikov wrote:
> On 18/04/13 01:21, Paul Berry wrote:
>> On 12 April 2013 01:26, Christian König wrote:
>>
>>> From: Christian König
>>>
>>> Since we now have UVD support we should enable them by default.
>>>
>>> Signed-off-by: Christian König
>>>
>>
>> Those of us who do
Kenneth Graunke writes:
> On 04/17/2013 01:59 PM, Dave Airlie wrote:
>> Hi,
>>
>> I put a 3Ghz Core2 Q35 box i found in the office to good use (so much
>> nicer than a pineview atom).
>>
>> http://people.freedesktop.org/~airlied/piglit/i915c/
>>
>> is a full run with i915c forced to advertise GL2
On Wed, Apr 17, 2013 at 6:07 PM, Emil Velikov wrote:
> On 18/04/13 01:55, Emil Velikov wrote:
>> On 18/04/13 01:21, Paul Berry wrote:
>>> On 12 April 2013 01:26, Christian König wrote:
>>>
From: Christian König
Since we now have UVD support we should enable them by default.
>
On Wed, Apr 17, 2013 at 5:30 PM, Kenneth Graunke wrote:
> We can't include shaderobj.h from the standalone utilities, so we
> unfortunately have to copy this function.
This is happening a lot. It'd be nice if src/glsl was just built as
part of the core.
Oh no, I've suggested a project for myself
pre_clip_pos is a float[4] we just use (*float)[4] to be able to
jump within the array of vertex_headers with it. So if the idx
happened to be anything but 0, we'd actually read from some garbage
in memory. Always use 0 to correctly offset withing float[4].
Signed-off-by: Zack Rusin
---
src/gall
On 04/17/2013 05:12 PM, Brian Paul wrote:
On 04/17/2013 12:07 AM, Tapani Pälli wrote:
patch fixes a crash that happens if glTexSubImage2D is called with a
negative xoffset.
Signed-off-by: Tapani Pälli
---
src/mesa/main/teximage.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --
On Wed, Apr 17, 2013 at 10:26 PM, Brian Paul wrote:
> On 04/17/2013 02:31 AM, Jose Fonseca wrote:
[snipped]
>>> How hard would it be to make GLSL IR an alternative IR for pipe
>>> drivers? On the driver side, it seems I only need to add
>>> PIPE_SHADER_IR_GLSL for the preferred IR and change the
patch fixes a crash that happens if glTexSubImage2D is called with a
negative xoffset.
NOTE: This is a candidate for stable branches.
Signed-off-by: Tapani Pälli
---
v2: corrected xoffset cast to happen before sign + added more checks
src/mesa/main/teximage.c | 10 +-
1 file changed,
Hi list,
Per the discussion, i965g is confusing and misleading. Instead of
preventing the confusions via
--with-gallium-drivers=i965g-unofficial or
--with-gallium-drivers=experimental-i965g,
which kind of makes i965g a second-class citizen, I think it is better
to rename it. Then whenever
72 matches
Mail list logo
