I know this code has confused others, and it confused me 3 years later,
too.
---
src/glsl/opt_dead_code_local.cpp | 28 ++--
1 file changed, 14 insertions(+), 14 deletions(-)
diff --git a/src/glsl/opt_dead_code_local.cpp b/src/glsl/opt_dead_code_local.cpp
index e7d46ed..70
While we wish our optimization passes could identify all the cases where
we can coalesce our variables, we miss out on a lot of opportunities.
total instructions in shared programs: 1673849 -> 1673166 (-0.04%)
instructions in affected programs: 299521 -> 298838 (-0.23%)
GAINED:
This logic is borrowed from the radeon code. The transfer logic will
only get called for PIPE_BUFFER resources, so it shouldn't be necessary
to worry about them becoming render targets.
Signed-off-by: Ilia Mirkin
---
This was re-tested by someone on both nv50 and nvc0. The performance gain
remai
nouveau_fence_wait has the expectation that an external entity is
holding onto the fence being waited on, not that it is merely held onto
by the current pointer. Fixes a use-after-free in nouveau_fence_wait
when used on the screen's current fence.
Bugzilla: https://bugs.freedesktop.org/show_bug.cg
sizeof(scissor) returns the size of the full array rather than a single
element. Fix it to consider just the one element.
Fixes: 0705fa35cdaf15ec969c28dc85e88b8be1149a3b
Signed-off-by: Ilia Mirkin
---
Noticed this when playing with valgrind for unrelated reasons. Seems fairly
obvious.
src/mesa
On Thu, Mar 6, 2014 at 7:06 AM, Brian Paul wrote:
> ---
> src/mapi/mapi.c | 10 +-
> src/mapi/stub.c |6 +++---
> src/mapi/u_current.c |6 +++---
> src/mapi/u_execmem.c |6 +++---
> src/mapi/u_thread.h | 10 --
> 5 files changed, 14 insertions(+), 24 del
On Thu, Mar 6, 2014 at 7:06 AM, Brian Paul wrote:
> A user would have no idea what "_glthread_" is. This removes the
> last remaining instance of the _glthread_ string in Mesa.
Reviewed-by: Chia-I Wu
> ---
> src/mapi/u_thread.h |6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
On Wed, Mar 5, 2014 at 10:48 PM, Brian Paul wrote:
> To make the functions more understandable.
Both patches are
Reviewed-by: Chia-I Wu
> ---
> src/mapi/glapi/glapi.c |2 +-
> src/mapi/mapi_glapi.c |2 +-
> src/mapi/u_current.c | 32
> src/mapi/u_c
https://bugs.freedesktop.org/show_bug.cgi?id=75226
Chia-I Wu changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
On Mit, 2014-03-05 at 21:35 -0500, Tom Stellard wrote:
> These are sometimes created by the shrink to boolean optimization in the
> globalopt pass.
Reviewed-by: Michel Dänzer
--
Earthling Michel Dänzer| http://www.amd.com
Libre software enthusiast |
On Mit, 2014-03-05 at 21:25 -0500, Tom Stellard wrote:
> And move its definition into r600_pipe_common.h; This struct is a just
> a container for shader code and has nothing to do with LLVM.
Well, it has something to do with LLVM in that the drivers only use it
with LLVM. :)
> diff --git a/src/
These are sometimes created by the shrink to boolean optimization in the
globalopt pass.
---
lib/Target/R600/SIISelLowering.cpp | 6 ++
test/CodeGen/R600/store.ll | 23 ---
2 files changed, 26 insertions(+), 3 deletions(-)
diff --git a/lib/Target/R600/SIISelLoweri
---
src/gallium/drivers/radeon/radeon_llvm_util.c | 35 --
.../state_trackers/clover/core/compiler.hpp| 3 +-
src/gallium/state_trackers/clover/core/device.cpp | 6 +++
src/gallium/state_trackers/clover/core/device.hpp | 1 +
src/gallium/state_trackers/clover/core/pro
---
src/gallium/drivers/radeon/Makefile.sources | 1 +
src/gallium/drivers/radeon/radeon_elf_util.c | 90 +++
src/gallium/drivers/radeon/radeon_elf_util.h | 39
src/gallium/drivers/radeon/radeon_llvm_emit.c | 53 +---
4 files changed, 132 inser
And move its definition into r600_pipe_common.h; This struct is a just
a container for shader code and has nothing to do with LLVM.
---
src/gallium/drivers/r600/r600_llvm.c | 4 ++--
src/gallium/drivers/radeon/Makefile.am| 1 +
src/gallium/drivers/radeon/r600_pipe_common.h | 16
On Wed, Mar 5, 2014 at 3:42 PM, Brian Paul wrote:
> With a non-debug build, gcc has two complaints:
> 1. 'found' var not used. Silence with '(void) found;'
> 2. 'id' not initialized. It's assigned by the UniformHash->get()
>call, actually. But init it to zero to silence gcc.
> ---
> src/gl
On 03/05/2014 04:17 PM, Eric Anholt wrote:
> We've had several problems now with FinishRenderTexture not getting called
> enough, and we're ready to just give up on it ever doing what we need. In
> particular, an upcoming Steam title had rendering bugs that could be fixed
> by always_flush_cache=t
On 03/05/2014 04:17 PM, Eric Anholt wrote:
> Now you can call the blit functions without worrying about flushing either
> too much or too little.
> ---
> src/mesa/drivers/dri/i965/intel_blit.c | 8 +---
> src/mesa/drivers/dri/i965/intel_buffer_objects.c | 20
>
We've had several problems now with FinishRenderTexture not getting called
enough, and we're ready to just give up on it ever doing what we need. In
particular, an upcoming Steam title had rendering bugs that could be fixed
by always_flush_cache=true.
Instead of hoping Mesa core can figure out wh
Now you can call the blit functions without worrying about flushing either
too much or too little.
---
src/mesa/drivers/dri/i965/intel_blit.c | 8 +---
src/mesa/drivers/dri/i965/intel_buffer_objects.c | 20
2 files changed, 5 insertions(+), 23 deletions(-)
diff
With a non-debug build, gcc has two complaints:
1. 'found' var not used. Silence with '(void) found;'
2. 'id' not initialized. It's assigned by the UniformHash->get()
call, actually. But init it to zero to silence gcc.
---
src/glsl/link_uniforms.cpp |3 ++-
1 file changed, 2 insertions(+
---
src/mapi/mapi.c | 10 +-
src/mapi/stub.c |6 +++---
src/mapi/u_current.c |6 +++---
src/mapi/u_execmem.c |6 +++---
src/mapi/u_thread.h | 10 --
5 files changed, 14 insertions(+), 24 deletions(-)
diff --git a/src/mapi/mapi.c b/src/mapi/mapi.c
index 5
A user would have no idea what "_glthread_" is. This removes the
last remaining instance of the _glthread_ string in Mesa.
---
src/mapi/u_thread.h |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/src/mapi/u_thread.h b/src/mapi/u_thread.h
index 0fc9392..78f2269 100644
-
On Wed, Mar 5, 2014 at 8:02 AM, Brian Paul wrote:
> This was overlooked in 7cc9df4b8a1c. Spotted by Chia-I Wu.
> ---
> src/mesa/main/format_unpack.c |8
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/src/mesa/main/format_unpack.c b/src/mesa/main/format_unpack.c
>
https://bugs.freedesktop.org/show_bug.cgi?id=75814
Priority: medium
Bug ID: 75814
Assignee: mesa-dev@lists.freedesktop.org
Summary: Heap-buffer-overflow WRITE in memcpy_texture
Severity: normal
Classification: Unclassified
OS
Reviewed-by: Marek Olšák
Marek
On Wed, Mar 5, 2014 at 12:20 AM, Aaron Watry wrote:
> Fix a leaked vertex shader in u_blitter.c
>
> Signed-off-by: Aaron Watry
>
> CC: "10.1"
> ---
> src/gallium/auxiliary/util/u_blitter.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/src/gallium/a
Ian Romanick writes:
> This is the last set of core API changes for
> ARB_separate_shader_objects. The next set of patches will just be in
> the linker and compiler.
Assuming my clarification restatement of the comment about ctx->Shader
wasn't misinformed, all the uncommented patches (or patche
Ian Romanick writes:
> From: Gregory Hainaut
>
> Implementation note:
> I don't use context for ralloc (don't know how).
>
> The check on PROGRAM_SEPARABLE flags is also done when the pipeline
> isn't bound. It doesn't make any sense in a DSA style API.
>
> Maybe we could replace _mesa_validate
Ian Romanick writes:
> From: Gregory Hainaut
>
> To avoid NULL pointer check a default pipeline object is installed in
> _Shader when no program is current
> diff --git a/src/mesa/main/mtypes.h b/src/mesa/main/mtypes.h
> index d05649c..8a03afd 100644
> --- a/src/mesa/main/mtypes.h
> +++ b/src/
Ian Romanick writes:
> From: Gregory Hainaut
>
> Now arb_separate_shader_object-GetProgramPipelineiv should pass.
>
> V3 (idr):
> * Change spec references to core OpenGL versions instead of issues in
> the extension spec.
> * Split out from previous uber patch.
>
> v4 (idr): Use _mesa_has_geom
Ian Romanick writes:
> From: Gregory Hainaut
>
> To avoid NULL pointer check a default pipeline object is installed in
> _Shader when no program is current
>
> The spec say that UseProgram/UseShaderProgramEXT/ActiveProgramEXT got an
> higher priority over the pipeline object. When default progra
On Wed, Mar 5, 2014 at 5:37 AM, Ilia Mirkin wrote:
> Signed-off-by: Ilia Mirkin
> ---
>
> Noticed by inspection, untested. Would be interesting to see if this fixes
> anything.
I noticed that the nvc0 bit is bogus -- mt->ms_x/y aren't used to
determine ms there (it does log2(samples) directly).
Hello all!
While debugging an application crash, we found a problem in mesa i915 buffer
allocation, and we think we at least have a workaround for this.
In more detail, the vertex buffers used in src/mesa/drivers/dri/i915/*.c, which
are pointed to by intel->prim.vb, usually have a size of INTEL_V
On 03/05/2014 09:02 AM, Brian Paul wrote:
On 03/04/2014 08:17 PM, Chia-I Wu wrote:
On Wed, Mar 5, 2014 at 12:10 AM, Brian Paul wrote:
Spotted by Chia-I Wu.
unpack_ubyte_ARGB_REV() needs the same care too. With that fixed,
this patch is
Reviewed-by: Chia-I Wu
I missed that function in
On 03/04/2014 08:17 PM, Chia-I Wu wrote:
On Wed, Mar 5, 2014 at 12:10 AM, Brian Paul wrote:
Spotted by Chia-I Wu.
unpack_ubyte_ARGB_REV() needs the same care too. With that fixed,
this patch is
Reviewed-by: Chia-I Wu
I missed that function in the original patch. Follow-up patch poste
This was overlooked in 7cc9df4b8a1c. Spotted by Chia-I Wu.
---
src/mesa/main/format_unpack.c |8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/src/mesa/main/format_unpack.c b/src/mesa/main/format_unpack.c
index d8bf57d..f9c42e7 100644
--- a/src/mesa/main/format_unpack
- Original Message -
> It was really only used in the radeon driver for a debug printf.
> And evidently, libGL.so referenced it just to work around some sort
> of linker issue.
>
> This patch removes the two calls to the function and the function
> itself.
>
> Fixes undefined _glthread_
https://bugs.freedesktop.org/show_bug.cgi?id=70410
--- Comment #20 from Laurent carlier ---
Patch increase size a lot. Linking failure is fixed when enable-shared-llvm
option isn't enabled. Something is weird somewhere
--
You are receiving this mail because:
You are the assignee for the bug.
__
Put "table" in the names to make things more understandable.
---
src/mapi/entry.c |4 ++--
src/mapi/glapi/glapi.c |2 +-
src/mapi/mapi.c|2 +-
src/mapi/mapi_glapi.c |2 +-
src/mapi/u_current.c |6 +++---
src/mapi/u_current.h | 10 +-
6 files changed
To make the functions more understandable.
---
src/mapi/glapi/glapi.c |2 +-
src/mapi/mapi_glapi.c |2 +-
src/mapi/u_current.c | 32
src/mapi/u_current.h | 20 ++--
4 files changed, 28 insertions(+), 28 deletions(-)
diff --git a/sr
https://bugs.freedesktop.org/show_bug.cgi?id=75797
Priority: medium
Bug ID: 75797
Assignee: mesa-dev@lists.freedesktop.org
Summary: EGL application crashes with BadDrawable at
SwapBuffers
Severity: normal
Classification:
When building out of tree, the file ends up dangling which
may result in a binary with the old git sha.
Signed-off-by: Emil Velikov
---
src/mesa/Makefile.am | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/mesa/Makefile.am b/src/mesa/Makefile.am
index ffe6599..16ac148 1006
Thanks. I'm going to push this shortly.
Marek
On Wed, Mar 5, 2014 at 9:05 AM, Niels Ole Salscheider
wrote:
> Using DMA for reads is much faster.
>
> Signed-off-by: Niels Ole Salscheider
> ---
> src/gallium/drivers/radeon/r600_buffer_common.c | 74
> +++--
> 1 file changed,
On 05/03/14 07:55, Christian König wrote:
> Am 04.03.2014 22:12, schrieb Emil Velikov:
>> All three (xvmc, vdpau and omx) do not have an alternative loading
>> similar to the dri modules. Thus one needs to explicitly install
>> them in order to use/test them.
>
> It's not mainline yet, but at leas
https://bugs.freedesktop.org/show_bug.cgi?id=75212
Eero Tamminen changed:
What|Removed |Added
Status|RESOLVED|VERIFIED
--- Comment #4 from Eero Tammin
On Wed, Mar 5, 2014 at 12:42 PM, Emil Velikov wrote:
> On 05/03/14 10:13, Erik Faye-Lund wrote:
>> On Tue, Mar 4, 2014 at 10:12 PM, Emil Velikov
>> wrote:
>>> There is little point in echoing everything that the script does
>>> to stdout. Wrap it in AM_V_GEN so that a reasonable message is
>>> p
On 05/03/14 10:13, Erik Faye-Lund wrote:
> On Tue, Mar 4, 2014 at 10:12 PM, Emil Velikov
> wrote:
>> There is little point in echoing everything that the script does
>> to stdout. Wrap it in AM_V_GEN so that a reasonable message is
>> printed as a indication of it's invocation.
>>
>> Signed-off-b
https://bugs.freedesktop.org/show_bug.cgi?id=71547
--- Comment #6 from David "okias" Heidelberger ---
Patch [1] tested on amd64, with both SSE4.1 enabled and -mno-sse4.1.
Both passed correctly without additional tweaking.
Thank you Matt
Tested-by: David Heidelberger
[1] http://lists.freedesk
Same here. I'll try to make clean in the LLVM tree.
Marek
On Wed, Mar 5, 2014 at 3:52 AM, Michel Dänzer wrote:
> On Mit, 2014-03-05 at 00:43 +0100, Marek Olšák wrote:
>> I'm not an automake expect, but this might need to be done for gallivm
>> too. It currently fails to compile with latest LLVM.
On 05/03/14 10:11, Erik Faye-Lund wrote:
> On Tue, Mar 4, 2014 at 10:12 PM, Emil Velikov
> wrote:
>> There is little gain in printing whenever a folder is created.
>>
>> Signed-off-by: Emil Velikov
>> ---
>> src/gallium/auxiliary/Makefile.am | 8
>> src/glsl/Makefile.am |
Signed-off-by: Ilia Mirkin
---
Noticed by inspection, untested. Would be interesting to see if this fixes
anything.
src/gallium/drivers/nouveau/nv50/nv50_miptree.c | 4 ++--
src/gallium/drivers/nouveau/nvc0/nvc0_miptree.c | 4 ++--
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/
From: Gregory Hainaut
V2 (idr):
* Keep the behavior of other info logs in Mesa: and empty info log
reports a GL_INFO_LOG_LENGTH of zero.
* Use a NULL pointer to denote an empty info log.
* Split out from previous uber patch.
Reviewed-by: Ian Romanick
---
src/mesa/main/mtypes.h | 2 ++
From: Gregory Hainaut
Test become green in piglit:
The updated ext_transform_feedback-api-errors:useprogstage_noactive
useprogstage_active bind_pipeline
arb_separate_shader_object-GetProgramPipelineiv
arb_separate_shader_object-IsProgramPipeline
For the moment I reuse Driver.UseProgram but I g
From: Gregory Hainaut
Basically a sed but shaderapi.c and get.c.
get.c => GL_CURRENT_PROGAM always refer to the "old" UseProgram behavior
shaderapi.c => the old api stil update the Shader object directly
V2: formatting improvement
V3 (idr):
* Rebase fixes after a block of code was moved from ir
From: Gregory Hainaut
save and restore _Shader/Pipeline binding point. Rational we don't want any
conflict when the program will be unattached.
V2: formatting improvement
V3 (idr):
* Build fix. The original patch added calls to _mesa_use_shader_program
with 4 parameters, but the fourth param
From: Gregory Hainaut
To avoid NULL pointer check a default pipeline object is installed in
_Shader when no program is current
The spec say that UseProgram/UseShaderProgramEXT/ActiveProgramEXT got an
higher priority over the pipeline object. When default program is
uninstall, the pipeline is use
From: Gregory Hainaut
Reviewed-by: Ian Romanick
---
src/mesa/main/get.c | 9 +
src/mesa/main/get_hash_params.py | 3 +++
2 files changed, 12 insertions(+)
diff --git a/src/mesa/main/get.c b/src/mesa/main/get.c
index 6e061e4..8a509c4 100644
--- a/src/mesa/main/get.c
+++ b/s
From: Gregory Hainaut
This is much like _mesa_sampler_uniforms_are_valid, but it operates
across an entire pipeline object.
This function differs from _mesa_sampler_uniforms_are_valid in that it
directly creates the gl_pipeline_object::InfoLog instead of writing to
some temporary buffer.
This w
From: Gregory Hainaut
Now arb_separate_shader_object-GetProgramPipelineiv should pass.
V3 (idr):
* Change spec references to core OpenGL versions instead of issues in
the extension spec.
* Split out from previous uber patch.
v4 (idr): Use _mesa_has_geometry_shaders in _mesa_UseProgramStages t
From: Gregory Hainaut
Implementation note:
I don't use context for ralloc (don't know how).
The check on PROGRAM_SEPARABLE flags is also done when the pipeline
isn't bound. It doesn't make any sense in a DSA style API.
Maybe we could replace _mesa_validate_program by
_mesa_validate_program_pip
From: Gregory Hainaut
Extend use_shader_program to support a different target. Allow to reuse the
function to update the pipeline state. Note I bypass the flush when target
isn't current. Maybe it would be better to create a new UseProgramStages
driver function
This was originally included in an
From: Ian Romanick
Signed-off-by: Ian Romanick
---
src/glsl/linker.cpp | 15 ---
1 file changed, 4 insertions(+), 11 deletions(-)
diff --git a/src/glsl/linker.cpp b/src/glsl/linker.cpp
index f6b2661..3bf2789 100644
--- a/src/glsl/linker.cpp
+++ b/src/glsl/linker.cpp
@@ -2303,17 +23
From: Ian Romanick
Remove extra "any" and re-word-wrap the comment.
Signed-off-by: Ian Romanick
---
src/glsl/lower_packed_varyings.cpp | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/glsl/lower_packed_varyings.cpp
b/src/glsl/lower_packed_varyings.cpp
index c23d180.
This is the last set of core API changes for
ARB_separate_shader_objects. The next set of patches will just be in
the linker and compiler.
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
On Tue, Mar 4, 2014 at 10:12 PM, Emil Velikov wrote:
> There is little point in echoing everything that the script does
> to stdout. Wrap it in AM_V_GEN so that a reasonable message is
> printed as a indication of it's invocation.
>
> Signed-off-by: Emil Velikov
> ---
> install-lib-links.mk | 2
On Tue, Mar 4, 2014 at 10:12 PM, Emil Velikov wrote:
> There is little gain in printing whenever a folder is created.
>
> Signed-off-by: Emil Velikov
> ---
> src/gallium/auxiliary/Makefile.am | 8
> src/glsl/Makefile.am | 4 ++--
> src/mesa/Makefile.am | 4 ++--
On Thu, Feb 6, 2014 at 5:09 PM, Erik Faye-Lund wrote:
> In the specification text of NV_vertex_program1_1, the upper
> limit of the RCC instruction is written as 1.884467e+19 in
> scientific notation, but as 0x5F80 in binary. But the binary
> version translates to 1.84467e+19 rather than 1.884
On Tuesday 04 March 2014, 23:43:01, Marek Olšák wrote:
> You check for streamout and CP DMA support, but you don't use
> resource_copy_region if DMA is not supported. The CP DMA and
> streamout-based buffer copying is only used by resource_copy_region.
Oh, right. I initially used resource_copy_reg
Using DMA for reads is much faster.
Signed-off-by: Niels Ole Salscheider
---
src/gallium/drivers/radeon/r600_buffer_common.c | 74 +++--
1 file changed, 56 insertions(+), 18 deletions(-)
diff --git a/src/gallium/drivers/radeon/r600_buffer_common.c
b/src/gallium/drivers/rade
69 matches
Mail list logo
