Re: [Mesa-dev] [PATCH] i965: Gut the separate OpenGL ES extension enabling.

2012-05-15 Thread Charles, Daniel
On Tue, May 15, 2012 at 3:49 PM, Kenneth Graunke wrote: > We should just set the bits of functionality that we support; the > GL/ES1/ES2 flags in extensions.c will take care of advertising the > appropriate extensions for the current API. > > This enables the GL_EXT_texture_compression_dxt1 extens

[Mesa-dev] [Bug 49538] [softpipe] piglit fs-texelFetchOffset-2D regression

2012-05-15 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=49538 Vinson Lee changed: What|Removed |Added CC||jfons...@vmware.com -- Configure bugmail:

[Mesa-dev] [Bug 49539] [softpipe] piglit vs-attrib-ivec4-implied regression

2012-05-15 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=49539 --- Comment #1 from Vinson Lee 2012-05-15 22:43:39 UTC --- vs-attrib-ivec4-precision, vs-attrib-uvec4-implied, and vs-attrib-uvec4-precision have also regressed on softpipe. -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=

Re: [Mesa-dev] [PATCH] llvmpipe: Implement TXQ.

2012-05-15 Thread Roland Scheidegger
Wouldn't it be easier (and more efficient) if you could just reuse the sampler code which does mipmap minification, i.e. get the texture sizes from bld.int_size and do a lp_build_minify (followed by extract shuffles)? Also, is it on purpose that the non-explicit lod path doesn't handle all cases (u

[Mesa-dev] [Bug 49766] [BUG] a typo in the help string of "r600-llvm-compiler" in configure.ac

2012-05-15 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=49766 Tom Stellard changed: What|Removed |Added Status|NEW |RESOLVED Resolution|

[Mesa-dev] [Bug 49768] [BUG][FIX] incorrect command format in the help string of "r600-llvm-compiler" in "configure.ac"

2012-05-15 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=49768 Tom Stellard changed: What|Removed |Added Status|NEW |RESOLVED Resolution|

[Mesa-dev] [PATCH] mesa: Remove the OES_draw_texture extension from ES2.

2012-05-15 Thread Kenneth Graunke
This extension appears to be written against ES 1.0. In ES 2.0, you really want to be using FBOs instead. --- src/mesa/main/extensions.c |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/mesa/main/extensions.c b/src/mesa/main/extensions.c index a843a40..cd76eeb 100644 ---

[Mesa-dev] [PATCH] i965: Gut the separate OpenGL ES extension enabling.

2012-05-15 Thread Kenneth Graunke
We should just set the bits of functionality that we support; the GL/ES1/ES2 flags in extensions.c will take care of advertising the appropriate extensions for the current API. This enables the GL_EXT_texture_compression_dxt1 extension on ES1/ES2 when libtxc_dxtn is installed or the force_s3tc dri

Re: [Mesa-dev] [PATCH 2/3] lp_bld_arit: Fix lp_build_sgn.

2012-05-15 Thread Jose Fonseca
- Original Message - > On Wed, May 09, 2012 at 10:01:37AM -0700, Jose Fonseca wrote: > > Something else is not right here, as > > > > lp_build_const_vec(bld->gallivm, type, -1.0) > > > > and > > > > lp_build_const_int_vec(bld->gallivm, type, -1) > > > > should produce the same outpu

[Mesa-dev] [PATCH] llvmpipe: Color slot interpolation can be flat or perspective, not linear.

2012-05-15 Thread Olivier Galibert
Fixes a bunch of glsl 1.10 interpolation piglit tests. Signed-off-by: Olivier Galibert --- src/gallium/drivers/llvmpipe/lp_bld_interp.h |2 +- src/gallium/drivers/llvmpipe/lp_state_fs.c|2 +- src/gallium/drivers/llvmpipe/lp_state_setup.c |2 +- 3 files changed, 3 insertions(+),

[Mesa-dev] [PATCH] llvmpipe: Implement TXQ.

2012-05-15 Thread Olivier Galibert
Piglits test for fragment shaders pass, vertex shaders fail. The actual failure seems to be in the interpolators, and not the textureSize query. Signed-off-by: Olivier Galibert --- src/gallium/auxiliary/draw/draw_llvm_sample.c | 23 + src/gallium/auxiliary/gallivm/lp_bld_sample.h

Re: [Mesa-dev] [PATCH v2 5/5] i965/gen6: Initial implementation of MSAA.

2012-05-15 Thread Kenneth Graunke
On 05/14/2012 06:24 PM, Paul Berry wrote: [snip] I'm not going to try to change your mind, but in the spirit of clarifying the method that was behind my madness: I did not have the clarity of mind during our discussion today to point out that in creating the brw_blorp_params class, I was in fact

Re: [Mesa-dev] [PATCH 0/4] add i965 primitive restart hardware support where possible

2012-05-15 Thread Kenneth Graunke
On 05/13/2012 07:06 AM, Jordan Justen wrote: Note: This changeset depends on my other changeset: "move software primitive restart into VBO module" v1: * Enable NV_primitive_restart for all hardware. (Software primitive restart is used where necessary) * If hardware supports cut in

Re: [Mesa-dev] bugfix series: mesa core, glsl, one i965.

2012-05-15 Thread Ian Romanick
On 05/14/2012 05:36 PM, Eric Anholt wrote: I took a little tour of the Intel oglconform failures, and found some easy fixes. Enjoy. The series is Reviewed-by: Ian Romanick ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freede

Re: [Mesa-dev] [PATCH v2 0/6] move software primitive restart into VBO module

2012-05-15 Thread Kenneth Graunke
On 05/13/2012 07:00 AM, Jordan Justen wrote: v2: * change non-gallium drivers to not enable NV_primitive_restart by default. (this matches the current mesa behavior.) * gallium drivers will continue to use software primitivie restart where needed and will always declare the NV_primiti

Re: [Mesa-dev] bugfix series: mesa core, glsl, one i965.

2012-05-15 Thread Kenneth Graunke
On 05/14/2012 05:36 PM, Eric Anholt wrote: I took a little tour of the Intel oglconform failures, and found some easy fixes. Enjoy. ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev For t

Re: [Mesa-dev] [PATCH] i965/blorp: Move exec() out of brw_blorp_params.

2012-05-15 Thread Kenneth Graunke
On 05/15/2012 07:31 AM, Paul Berry wrote: No functional change. This patch replaces the brw_blorp_params::exec() method with a global function brw_blorp_exec() that performs the operation described by the params data structure. Reviewed-by: Kenneth Graunke ___

Re: [Mesa-dev] [PATCH 3/9] glsl: Fix indentation of switch code.

2012-05-15 Thread Kenneth Graunke
On 05/14/2012 05:36 PM, Eric Anholt wrote: I managed to completely trash it in 22d81f15. --- src/glsl/ast_to_hir.cpp | 545 +++ 1 file changed, 263 insertions(+), 282 deletions(-) diff --git a/src/glsl/ast_to_hir.cpp b/src/glsl/ast_to_hir.cpp index

[Mesa-dev] [PATCH] i965/blorp: Move exec() out of brw_blorp_params.

2012-05-15 Thread Paul Berry
No functional change. This patch replaces the brw_blorp_params::exec() method with a global function brw_blorp_exec() that performs the operation described by the params data structure. --- src/mesa/drivers/dri/i965/brw_blorp.cpp |6 +++--- src/mesa/drivers/dri/i965/brw_blorp.h|

Re: [Mesa-dev] [PATCH 0/9] radeon/llvm fixes

2012-05-15 Thread Tom Stellard
On Tue, May 15, 2012 at 01:41:25PM +0400, Vadim Girlin wrote: > With these patches piglit results are finally equal with non-llvm backend. > Nice! > radeon/llvm: add generated files to .gitignore > radeon/llvm: add names for AMDGPU* passes > radeon/llvm: fix BUILD_VECTOR lowering for replic

Re: [Mesa-dev] [PATCH 1/9] mesa: Throw error on glGetActiveUniform inside Begin/End.

2012-05-15 Thread Brian Paul
On 05/14/2012 06:36 PM, Eric Anholt wrote: Fixes piglit GL_ARB_shader_objeccts/getactiveuniform-beginend. --- src/mesa/main/uniform_query.cpp |2 ++ 1 file changed, 2 insertions(+) diff --git a/src/mesa/main/uniform_query.cpp b/src/mesa/main/uniform_query.cpp index 08d330a..f5d998f 100644

[Mesa-dev] [PATCH 9/9] radeon/llvm: add support for texture offsets, fix TEX_LD

2012-05-15 Thread Vadim Girlin
Signed-off-by: Vadim Girlin --- src/gallium/drivers/radeon/AMDGPUIntrinsics.td |2 +- src/gallium/drivers/radeon/R600CodeEmitter.cpp | 15 src/gallium/drivers/radeon/R600Instructions.td |7 ++-- .../drivers/radeon/radeon_setup_tgsi_llvm.c| 37 +

[Mesa-dev] [PATCH 8/9] radeon/llvm: add SET_GRADIENTS*, fix SAMPLE_G

2012-05-15 Thread Vadim Girlin
Signed-off-by: Vadim Girlin --- src/gallium/drivers/r600/r600_llvm.c | 14 +++--- src/gallium/drivers/radeon/AMDGPUIntrinsics.td |2 +- src/gallium/drivers/radeon/AMDGPUUtil.cpp |2 + src/gallium/drivers/radeon/R600ISelLowering.cpp| 47 ++

[Mesa-dev] [PATCH 7/9] radeon/llvm: increase const regs count

2012-05-15 Thread Vadim Girlin
Signed-off-by: Vadim Girlin --- src/gallium/drivers/radeon/R600GenRegisterInfo.pl |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/gallium/drivers/radeon/R600GenRegisterInfo.pl b/src/gallium/drivers/radeon/R600GenRegisterInfo.pl index 406f3df..9e7cf61 100644 --- a/src/g

[Mesa-dev] [PATCH 6/9] radeon/llvm: use IntrNoMem property for intrinsics

2012-05-15 Thread Vadim Girlin
Signed-off-by: Vadim Girlin --- src/gallium/drivers/r600/r600_llvm.c | 23 ++-- src/gallium/drivers/radeon/AMDGPUIntrinsics.td | 76 +-- src/gallium/drivers/radeon/AMDILFormats.td | 18 +-- .../drivers/radeon/R600IntrinsicsNoOpenCL.td |2 +- src/g

[Mesa-dev] [PATCH 5/9] radeon/llvm: use correct intrinsic for CEIL

2012-05-15 Thread Vadim Girlin
Should be round_posinf instead of round_neginf. Signed-off-by: Vadim Girlin --- src/gallium/drivers/radeon/R600Instructions.td |4 ++-- src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c |2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/src/gallium/drivers/radeo

[Mesa-dev] [PATCH 4/9] radeon/llvm: improve ABS_i32 lowering

2012-05-15 Thread Vadim Girlin
We can save one instruction by lowering it to: SUB_INT tmp, 0, src MAX_INT dst, src, tmp Signed-off-by: Vadim Girlin --- src/gallium/drivers/radeon/R600LowerInstructions.cpp | 18 +- 1 file changed, 5 insertions(+), 13 deletions(-) diff --git a/src/gallium/drivers/radeon/R

[Mesa-dev] [PATCH 3/9] radeon/llvm: fix BUILD_VECTOR lowering for replicated value

2012-05-15 Thread Vadim Girlin
We expect that all elements will be assigned even if they are equal Signed-off-by: Vadim Girlin --- src/gallium/drivers/radeon/AMDILISelLowering.cpp |2 ++ 1 file changed, 2 insertions(+) diff --git a/src/gallium/drivers/radeon/AMDILISelLowering.cpp b/src/gallium/drivers/radeon/AMDILISelLo

[Mesa-dev] [PATCH 2/9] radeon/llvm: add names for AMDGPU* passes

2012-05-15 Thread Vadim Girlin
Signed-off-by: Vadim Girlin --- src/gallium/drivers/radeon/AMDGPUConvertToISA.cpp |3 +++ src/gallium/drivers/radeon/AMDGPULowerInstructions.cpp |2 ++ 2 files changed, 5 insertions(+) diff --git a/src/gallium/drivers/radeon/AMDGPUConvertToISA.cpp b/src/gallium/drivers/radeon/AMDGP

[Mesa-dev] [PATCH 1/9] radeon/llvm: add generated files to .gitignore

2012-05-15 Thread Vadim Girlin
Signed-off-by: Vadim Girlin --- src/gallium/drivers/radeon/.gitignore | 18 ++ 1 file changed, 18 insertions(+) create mode 100644 src/gallium/drivers/radeon/.gitignore diff --git a/src/gallium/drivers/radeon/.gitignore b/src/gallium/drivers/radeon/.gitignore new file mode 10

[Mesa-dev] [PATCH 0/9] radeon/llvm fixes

2012-05-15 Thread Vadim Girlin
With these patches piglit results are finally equal with non-llvm backend. radeon/llvm: add generated files to .gitignore radeon/llvm: add names for AMDGPU* passes radeon/llvm: fix BUILD_VECTOR lowering for replicated value radeon/llvm: improve ABS_i32 lowering radeon/llvm: use correct i