[Mesa-dev] [Bug 103526] Mesa 13.0.6/13.0.4 llvmpipe software rasterizer on PowerPC. Unexpected format PIPE_FORMAT_A8B8G8R8_UNORM in st_new_renderbuffer_fb

https://bugs.freedesktop.org/show_bug.cgi?id=103526 Ben Crocker changed: What|Removed |Added Status|NEW |NEEDINFO -- You

[Mesa-dev] [Bug 104818] mesa fails to build on ia64

https://bugs.freedesktop.org/show_bug.cgi?id=104818 Jason Duerstock changed: What|Removed |Added CC|

[Mesa-dev] [Bug 104818] mesa fails to build on ia64

https://bugs.freedesktop.org/show_bug.cgi?id=104818 Bug ID: 104818 Summary: mesa fails to build on ia64 Product: Mesa Version: unspecified Hardware: Other OS: All Status: NEW Severity: normal

[Mesa-dev] [RFC] [PATCH 2/2] anv: enable VK_EXT_shader_viewport_index_layer

--- This should not be merged until we have proper tests. I've submitted the first test for this to VK-GL-CTS. src/intel/vulkan/anv_extensions.py | 1 + src/intel/vulkan/anv_pipeline.c| 1 + 2 files changed, 2 insertions(+) diff --git a/src/intel/vulkan/anv_extensions.py

[Mesa-dev] [PATCH 1/2] spirv: Add SpvCapabilityShaderViewportIndexLayerEXT

--- In the last chunk of the diff, should accepting stage of vertex shader should be conditioned on shader_viewport_index_layer? src/compiler/shader_info.h | 1 + src/compiler/spirv/spirv_to_nir.c | 4 src/compiler/spirv/vtn_variables.c | 2 +- 3 files changed, 6 insertions(+), 1

Re: [Mesa-dev] [PATCH shaderdb 1/2] run: fall back to v3.1 core ctx if v3.2 fails

On Sat, Jan 27, 2018 at 4:08 PM, Rob Clark wrote: > On Sat, Jan 27, 2018 at 7:02 PM, Matt Turner wrote: >> On Sat, Jan 27, 2018 at 10:06 AM, Rob Clark wrote: >>> Signed-off-by: Rob Clark >>> --- >>> run.c | 12

Re: [Mesa-dev] [PATCH shaderdb 1/2] run: fall back to v3.1 core ctx if v3.2 fails

On Sat, Jan 27, 2018 at 7:08 PM, Matt Turner wrote: > On Sat, Jan 27, 2018 at 4:02 PM, Matt Turner wrote: >> On Sat, Jan 27, 2018 at 10:06 AM, Rob Clark wrote: >>> Signed-off-by: Rob Clark >>> --- >>> run.c | 12

Re: [Mesa-dev] [PATCH shaderdb 1/2] run: fall back to v3.1 core ctx if v3.2 fails

On Sat, Jan 27, 2018 at 4:02 PM, Matt Turner wrote: > On Sat, Jan 27, 2018 at 10:06 AM, Rob Clark wrote: >> Signed-off-by: Rob Clark >> --- >> run.c | 12 >> 1 file changed, 12 insertions(+) >> >> diff --git a/run.c

Re: [Mesa-dev] [PATCH shaderdb 1/2] run: fall back to v3.1 core ctx if v3.2 fails

On Sat, Jan 27, 2018 at 7:02 PM, Matt Turner wrote: > On Sat, Jan 27, 2018 at 10:06 AM, Rob Clark wrote: >> Signed-off-by: Rob Clark >> --- >> run.c | 12 >> 1 file changed, 12 insertions(+) >> >> diff --git a/run.c

Re: [Mesa-dev] [PATCH shaderdb 1/2] run: fall back to v3.1 core ctx if v3.2 fails

On Sat, Jan 27, 2018 at 10:06 AM, Rob Clark wrote: > Signed-off-by: Rob Clark > --- > run.c | 12 > 1 file changed, 12 insertions(+) > > diff --git a/run.c b/run.c > index 69fe23d..20a0a6a 100644 > --- a/run.c > +++ b/run.c > @@ -520,6

[Mesa-dev] [Bug 42540] glxgears segfaults on POWER / Xvnc

https://bugs.freedesktop.org/show_bug.cgi?id=42540 Ben Crocker changed: What|Removed |Added Status|NEW |NEEDINFO -- You are

[Mesa-dev] [Bug 42540] glxgears segfaults on POWER / Xvnc

https://bugs.freedesktop.org/show_bug.cgi?id=42540 --- Comment #3 from Ben Crocker --- This is (or was) a bug in the LLVM pipe; it has nothing to do, per se, with Xvnc. The stack trace indicates a call to the runtime-compiled vertex shader program. A lot of development has

Re: [Mesa-dev] [PATCH v2 00/22] Introducing SPIR-V support to clover

Pierre Moreau writes: > On 2018-01-24 — 09:19, Tomeu Vizoso wrote: >> On 01/24/2018 12:03 AM, Karol Herbst wrote: >> > On Tue, Jan 23, 2018 at 11:46 PM, Francisco Jerez >> > wrote: >> > > Pierre Moreau writes: >> > > >> > >

[Mesa-dev] [PATCH 5/8] mesa: Remove unused bit in ffvertex_prog state_key.

From: Mathias Fröhlich Remove set but not read field from the state key used for hashing fixed function vertex shaders. Signed-off-by: Mathias Fröhlich --- src/mesa/main/ffvertex_prog.c | 6 -- 1 file changed, 6 deletions(-) diff --git

[Mesa-dev] [PATCH 4/8] mesa: texgen_enabled is only 1 bit.

From: Mathias Fröhlich For the state key for hashing fixed function vertex shaders, the texgen_enabled field requires only a single bit. Signed-off-by: Mathias Fröhlich --- src/mesa/main/ffvertex_prog.c | 2 +- 1 file changed, 1

[Mesa-dev] [PATCH 3/8] mesa: Encode fog modes in a 2 bit field.

From: Mathias Fröhlich For the state key for hashing fixed function vertex shaders, encode the different fog modes, including if fog is generally enabled or not, into a 2 bit field. Signed-off-by: Mathias Fröhlich ---

[Mesa-dev] [PATCH 7/8] mesa: Reduce ffvertex_prog state_key to 36 bytes.

From: Mathias Fröhlich Using lower alignment restrictions for the state key fields finally yields to a smaller hasing state key. Signed-off-by: Mathias Fröhlich --- src/mesa/main/ffvertex_prog.c | 29 +++-- 1 file

[Mesa-dev] [PATCH 0/8] Clean up fixed function vertex shader key

From: Mathias Fröhlich Hi all, The following series applies some optimizations to fixed function vertex shader hash key generation. Most is targeted to get a smaller hash key to get a smaller cache footprint and a shorter final key to compare. Two of the changes avoid

[Mesa-dev] [PATCH 8/8] mesa: Care for differences in fog mode only if fog is consumed.

From: Mathias Fröhlich In creating fixed function vertex shader hash keys do only care for producing the varying output if fog is enabled and the varing is consumed in the fragment stage. Signed-off-by: Mathias Fröhlich ---

[Mesa-dev] [PATCH 1/8] mesa: Get the point size array state from varying_vp_inputs.

From: Mathias Fröhlich For the state key for hashing fixed function vertex shaders, The varying_vp_inputs bitmask already contains the point size array enabled information. Signed-off-by: Mathias Fröhlich --- src/mesa/main/ffvertex_prog.c |

[Mesa-dev] [PATCH 2/8] mesa: Move seperate_specular into the lighting section.

From: Mathias Fröhlich For the state key for hashing fixed function vertex shaders, the information is only evaluated if lighting is generally switched on. Signed-off-by: Mathias Fröhlich --- src/mesa/main/ffvertex_prog.c | 6 +++--- 1 file

[Mesa-dev] [PATCH 6/8] mesa: Remove unused ffvertex_prog texunit_really_enabled.

From: Mathias Fröhlich Remove set but not read field from the state key used for hashing fixed function vertex shaders. Signed-off-by: Mathias Fröhlich --- src/mesa/main/ffvertex_prog.c | 4 1 file changed, 4 deletions(-) diff --git

[Mesa-dev] [PATCH 3/3] radeonsi: remove DBG_PRECOMPILE

From: Marek Olšák it's useless and shader-db stats only report the main shader part. --- src/gallium/drivers/radeon/r600_pipe_common.h | 1 - src/gallium/drivers/radeonsi/si_pipe.c | 1 - src/gallium/drivers/radeonsi/si_state_shaders.c | 49

Re: [Mesa-dev] [PATCH v2 00/22] Introducing SPIR-V support to clover

On 2018-01-24 — 09:19, Tomeu Vizoso wrote: > On 01/24/2018 12:03 AM, Karol Herbst wrote: > > On Tue, Jan 23, 2018 at 11:46 PM, Francisco Jerez > > wrote: > > > Pierre Moreau writes: > > > > > > > On 2018-01-23 — 14:02, Francisco Jerez wrote: > > >

[Mesa-dev] [PATCH 2/3] radeonsi: print shader-db stats for main parts, not final binaries

From: Marek Olšák This is needed to get shader-db stats for LS,HS,ES,GS stages on gfx9. --- src/gallium/drivers/radeonsi/si_shader.c| 33 +++-- src/gallium/drivers/radeonsi/si_shader.h| 2 ++

[Mesa-dev] [PATCH 1/3] radeonsi: move max_simd_waves computation into a separate function

From: Marek Olšák --- src/gallium/drivers/radeonsi/si_shader.c | 34 +--- src/gallium/drivers/radeonsi/si_shader.h | 1 + 2 files changed, 23 insertions(+), 12 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_shader.c

[Mesa-dev] [PATCH shaderdb 2/2] run: add support for GLES shaders

Signed-off-by: Rob Clark --- So, I think this is correct, but somehow with GLES I don't seem to get the shaders pre-compiled, so I don't get any output stats. So I suspect some fix is needed on the mesa side of things. Hints welcome, if anyone has some idea. It would be

[Mesa-dev] [PATCH shaderdb 1/2] run: fall back to v3.1 core ctx if v3.2 fails

Signed-off-by: Rob Clark --- run.c | 12 1 file changed, 12 insertions(+) diff --git a/run.c b/run.c index 69fe23d..20a0a6a 100644 --- a/run.c +++ b/run.c @@ -520,6 +520,18 @@ main(int argc, char **argv) }; EGLContext core_ctx =

Re: [Mesa-dev] [PATCH v2 03/22] clover/api: Fail if trying to build a non-executable binary

On 2018-01-23 — 15:09, Francisco Jerez wrote: > Pierre Moreau writes: > > > On 2018-01-23 — 14:03, Francisco Jerez wrote: > >> Pierre Moreau writes: > >> > >> > From the OpenCL 1.2 Specification, Section 5.6.2 (about clBuildProgram: > >> > > >> >>

[Mesa-dev] [PATCH] i965: Move down genX_upload_sbe in profiles.

From: Mathias Fröhlich Hi all, Avoid looping over all VARYING_SLOT_MAX urb_setup array entries from genX_upload_sbe. Prepare an array indirection to the active entries of urb_setup already in the compile step. On upload only walk the active arrays. The change moves

[Mesa-dev] [PATCH] mesa: Remove unused gl_fog_attrib::_Scale.

From: Mathias Fröhlich Hi all, The patch removes a variable that is only written to. Please review. best Mathias Signed-off-by: Mathias Fröhlich --- src/mesa/main/fog.c| 16 src/mesa/main/mtypes.h | 1 - 2 files

Re: [Mesa-dev] [PATCH 11/11] vbo: overhaul display list vertex array setup/binding code

Hi Brian, That one will not work as is. Display lists are shared objects across contexts. Means past that change the const struct gl_vertex_array *inputs[VBO_ATTRIB_MAX]; may be concurrently written to from different threads filling in current value pointers from the context where the list

Re: [Mesa-dev] [PATCH 01/11] vbo: s/cl/array/ in vbo_context.c

Hi Brian, for Patch #1 through #9 they are Reviewed-by: Mathias Fröhlich For #10 and #11 I have some comments. best Mathias On Thursday, 25 January 2018 00:20:25 CET Brian Paul wrote: > I think 'cl' used to mean client array. > --- > src/mesa/vbo/vbo_context.c |

Re: [Mesa-dev] [PATCH 10/11] vbo: optimize loops in bind_vertex_list(), vbo_exec_bind_arrays()

Hi Brian, The change should not change the current behavior. Nevertheless the current behavior as well as past your change the code has a corner case that will not work correctly. I do have a hand full of unpublished changes here that will fix that potential problem and achieve something

Re: [Mesa-dev] [PATCH v2] meson: generate translations for driconf

Hi Dylan, Am Donnerstag, 25. Januar 2018, 20:32:23 CET schrieb Dylan Baker: > Currently meson implements the same logic as SCons for translations, > namely it doesn't do them. This patch changes meson to use logic more > like autotools, and generate translations. To do this we have to go > behind

[Mesa-dev] [PATCH] radv: Signal fence correctly after sparse binding.

It did not signal syncobjs in the fence, and also signalled to early if there was work on the queue already, as we have to wait till that work is done. Fixes: d27aaae4d2 "radv: Add external fence support." --- src/amd/vulkan/radv_device.c | 46 ++-- 1 file

Re: [Mesa-dev] [PATCH 3/3] mesa: skip validation of legality of size/type queries for format queries

Am 27.01.2018 um 09:52 schrieb Alejandro Piñeiro: > On 27/01/18 01:59, srol...@vmware.com wrote: >> From: Roland Scheidegger >> >> The size/type query is always legal (if we made it that far). >> This causes a difference for GL_TEXTURE_BUFFER - the reason is that these >>

Re: [Mesa-dev] [PATCH] ac/nir: call glsl_get_sampler_dim() only once, where possible (v2)

Ping. This patch has a T-b by Dieter Nützel (including v2), any R-b and commit would be much appreciated. (Timothy indicated on Wednesday he would try to commit this (), but I guess it must have slipped

Re: [Mesa-dev] [PATCH 3/3] mesa: skip validation of legality of size/type queries for format queries

On 27/01/18 01:59, srol...@vmware.com wrote: > From: Roland Scheidegger > > The size/type query is always legal (if we made it that far). > This causes a difference for GL_TEXTURE_BUFFER - the reason is that these > parameters are valid only with GetTexLevelParameter() if gl

Re: [Mesa-dev] [PATCH 2/3] mesa: restrict formats being supported by target type for formatquery

On 27/01/18 01:59, srol...@vmware.com wrote: > From: Roland Scheidegger > > The code just considered all formats as being supported if they were either > a valid fbo or texture format. > This was quite awkward since then the query would return "supported" for > e.g. GL_RGB9E5

Re: [Mesa-dev] [PATCH 1/3] mesa: remove misleading gles checks for formatquery

On 27/01/18 01:59, srol...@vmware.com wrote: > From: Roland Scheidegger > > Testing for gles there is just confusing - this is about target being > supported, if it was valid at all was already determined earlier > (in _legal_parameters). It didn't make sense at all in any