Re: [Mesa-dev] [PATCH v2] glsl: fix consumer_stage restriction to separate shader objects

On Tue, 2016-01-19 at 07:20 +0100, Samuel Iglesias Gonsálvez wrote: > Commit 781d278 did not restrict consumer_stage only to separate > shader > objects, which is when we don't know if the consumer stage would be a > fragment shader added later. In normal programs, when > consumer_stage == -1, it

Re: [Mesa-dev] [PATCH 17/17] st/mesa: expose ARB_compute_shader when compute is supported

On Mon, Jan 25, 2016 at 11:00 PM, Michel Dänzer wrote: > On 26.01.2016 01:17, Nicolai Hähnle wrote: >> On 25.01.2016 10:04, Ilia Mirkin wrote: >>> Yeah so this isn't going to fly. r600/radeonsi both expose >>> PIPE_CAP_COMPUTE, for clover, but won't support these compute

Re: [Mesa-dev] [PATCH v4 03/11] glsl: keep track of ssbo variable being accessed, add access params

Reviewed-by: Iago Toral Quiroga On Sun, 2016-01-24 at 13:59 -0500, Ilia Mirkin wrote: > Currently any access params (coherent/volatile/restrict) are being lost > when lowering to the ssbo load/store intrinsics. Keep track of the > variable being used, and bake its access

Re: [Mesa-dev] [PATCH 17/17] st/mesa: expose ARB_compute_shader when compute is supported

On 26.01.2016 01:17, Nicolai Hähnle wrote: > On 25.01.2016 10:04, Ilia Mirkin wrote: >> Yeah so this isn't going to fly. r600/radeonsi both expose >> PIPE_CAP_COMPUTE, for clover, but won't support these compute shaders >> quite yet. Not sure what the solution is here... maybe just add >> another

[Mesa-dev] Request for MESA support for OpenGL GL_INTEL_* exts supported by Windows driver (2 of them expose new graphics DX12 features)

Hi, with GFXBench 4.0 released last week for Linux I made a quick diff of extensions support in latest Mesa Intel driver (11.2git) vs Windows OpenGL driver on Skylake and apart of complete GL 4.4 support on Windows which parity will be reached eventually on Mesa I assume, I find roughly only few

[Mesa-dev] [Bug 27512] Illegal instruction _mesa_x86_64_transform_points4_general

https://bugs.freedesktop.org/show_bug.cgi?id=27512 --- Comment #16 from Michael Harder --- I've been able to reinstall and get it working with the patch again. Not sure what I was doing wrong before. Do I need to do anything to move this along? -- You are receiving this

Re: [Mesa-dev] [PATCH 04/17] gallium: disable compute shaders for meta ops

On Mon, Jan 25, 2016 at 6:35 PM, Nicolai Hähnle wrote: > On 24.01.2016 16:09, Samuel Pitoiset wrote: >> >> Loosely based on tessellation shaders. > > > Do we actually need this? The graphics pipeline and the compute pipeline are > separate; draw commands should be unaffected

Re: [Mesa-dev] [PATCH 4/6] nv50/ir: optimize shl(shr(a, c), c) to and(a, ~((1 << c) - 1))

Reviewed-by: Ilia Mirkin I'll run it on my collection on both nvc0 and nv50 to make sure this doesn't subtly explode somehow, but looks great! I've been meaning to do this myself but never got around to it. On Mon, Jan 25, 2016 at 9:57 AM, Karol Herbst

Re: [Mesa-dev] [PATCH 1/2] winsys/amdgpu: Handle RADEON_FLAG_NO_CPU_ACCESS

This series is, Reviewed-by: Edward O'Callaghan Good job working out where this issue was. On 2016-01-26 18:40, Michel Dänzer wrote: From: Michel Dänzer Failing to do this was resulting in the kernel driver unnecessarily leaving open

Re: [Mesa-dev] [PATCH 2/6] nv50/ir: swap sources in PostRaConstantFolding when src0 is imm

On Mon, Jan 25, 2016 at 9:57 AM, Karol Herbst wrote: > From: Karol Herbst > > helps some shaders in multiple games > > total instructions in shared programs : 1922267 -> 1922121 (-0.01%) > total gprs used in shared programs: 251878 -> 251878

[Mesa-dev] [PATCH] glsl: remove old FINISHME

This should have been removed long ago. --- src/glsl/linker.cpp | 2 -- 1 file changed, 2 deletions(-) diff --git a/src/glsl/linker.cpp b/src/glsl/linker.cpp index 4e63698..7925709 100644 --- a/src/glsl/linker.cpp +++ b/src/glsl/linker.cpp @@ -4679,8 +4679,6 @@ link_shaders(struct gl_context

Re: [Mesa-dev] [PATCH 6/6] nv50/ir: run DCE backwards

Reviewed-by: Ilia Mirkin On Mon, Jan 25, 2016 at 9:57 AM, Karol Herbst wrote: > reduces Pass rerun by around 40% > > Signed-off-by: Karol Herbst > --- > src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp | 6 +++--- >

[Mesa-dev] [PATCH 1/2] winsys/amdgpu: Handle RADEON_FLAG_NO_CPU_ACCESS

From: Michel Dänzer Failing to do this was resulting in the kernel driver unnecessarily leaving open the possibility of CPU access to tiled BOs. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93862 (This change shouldn't be backported to stable branches, because

[Mesa-dev] [PATCH 2/2] winsys/amdgpu: Process RADEON_FLAG_* independently from RADEON_DOMAIN_*

From: Michel Dänzer In particular, AMDGPU_GEM_CREATE_CPU_GTT_USWC can affect even BOs created in VRAM if they get evicted to GTT. In general there's no need to restrict any of the flags to any particular domains. Signed-off-by: Michel Dänzer ---

Re: [Mesa-dev] [PATCH 3/6] nv50/ir: optimize neg(add(bool, 1)) to bool for OP_SET and OP_SLCT

On Mon, Jan 25, 2016 at 9:57 AM, Karol Herbst wrote: > From: Karol Herbst > > helps shaders in saints row IV, bioshock infinite and shadow warrior > > total instructions in shared programs : 1922121 -> 192 (-0.57%) > total gprs used in shared

Re: [Mesa-dev] [PATCH] glsl: remove old FINISHME

Reviewed-by: Chris Forbes On Tue, Jan 26, 2016 at 6:22 PM, Timothy Arceri < timothy.arc...@collabora.com> wrote: > This should have been removed long ago. > --- > src/glsl/linker.cpp | 2 -- > 1 file changed, 2 deletions(-) > > diff --git a/src/glsl/linker.cpp

Re: [Mesa-dev] [PATCH 15/17] st/mesa: add a state tracker for compute

How about st/mesa: add compute program dispatch callbacks or something along those lines. I don't think "state tracker" is the right term to use. On Sun, Jan 24, 2016 at 4:09 PM, Samuel Pitoiset wrote: > This state tracker implements DispatchCompute() and

Re: [Mesa-dev] [PATCH] appveyor: Bump shallow clone depth.

Am 24.01.2016 um 19:42 schrieb Jose Fonseca: > To prevent build failures when a large patch series is committed, like > happened in >

Re: [Mesa-dev] [PATCH 1/2] split-to-files.py: fix parsing when #version is not at the beginning

Hi, On 23.01.2016 15:24, Ilia Mirkin wrote: On Sat, Jan 23, 2016 at 8:05 AM, Marek Olšák wrote: On Sat, Jan 23, 2016 at 1:28 PM, Ilia Mirkin wrote: H... Because you can have // asdf #version 150? Unfortunately this would also catch // #version

Re: [Mesa-dev] [PATCH v3 2/10] glsl: keep track of ssbo variable being accessed, add access params

On Mon, Jan 25, 2016 at 2:38 AM, Iago Toral wrote: > On Sun, 2016-01-24 at 11:46 -0500, Ilia Mirkin wrote: >> On Thu, Jan 21, 2016 at 3:27 AM, Iago Toral wrote: >> >> *offset = new(mem_ctx) ir_constant(0u); >> >> *row_major =

Re: [Mesa-dev] [PATCH 17/17] st/mesa: expose ARB_compute_shader when compute is supported

Yeah so this isn't going to fly. r600/radeonsi both expose PIPE_CAP_COMPUTE, for clover, but won't support these compute shaders quite yet. Not sure what the solution is here... maybe just add another cap? PIPE_CAP_GLSL_COMPUTE or something? Or require a backend to also set PIPE_CAP_GLSL_VERSION

Re: [Mesa-dev] [PATCH v2] glsl: fix consumer_stage restriction to separate shader objects

This patch is still unreviewed. Sam On Tue, 2016-01-19 at 07:20 +0100, Samuel Iglesias Gonsálvez wrote: > Commit 781d278 did not restrict consumer_stage only to separate > shader > objects, which is when we don't know if the consumer stage would be a > fragment shader added later. In normal

Re: [Mesa-dev] [PATCH v2] glsl: fix consumer_stage restriction to separate shader objects

On 25 January 2016 11:35:04 pm AEDT, "Samuel Iglesias Gonsálvez" wrote: >This patch is still unreviewed. I'll take another look tomorrow unless someone gets to it first. I've looked at it but didn't really look hard enough to understand the problem the previous patch

Re: [Mesa-dev] [PATCH 09/17] st/mesa: add a second pipeline for compute

On Sun, Jan 24, 2016 at 4:09 PM, Samuel Pitoiset wrote: > Compute needs a new and different validation path. > > Signed-off-by: Samuel Pitoiset > --- > src/gallium/include/state_tracker/st_api.h | 8 + > src/mesa/state_tracker/st_atom.c

Re: [Mesa-dev] [PATCH 00/17] st/mesa: add compute shaders support

Assuming you incorporate the feedback I've given on the various patches, everything except the last one are Reviewed-by: Ilia Mirkin . You should wait for more feedback before resending the series though. On Sun, Jan 24, 2016 at 4:09 PM, Samuel Pitoiset

Re: [Mesa-dev] [PATCH 07/17] gallium: add indirect compute parameters to pipe_grid_info

On Sun, Jan 24, 2016 at 4:09 PM, Samuel Pitoiset wrote: > Like indirect draw, we need to store a resource and an offset that > needs to be 4 byte aligned. When indirect is used, the size of the > grid (in blocks) is stored with three 32-bit integers. > > Signed-off-by:

[Mesa-dev] [Bug 93820] Change from Mesa 10 to 11 made remote opengl stop working

https://bugs.freedesktop.org/show_bug.cgi?id=93820 --- Comment #2 from Christoph Weiss --- The X server version changed from 1.16.4 to 1.17.4. I can try running it with 1.16.4 if necessary (requires a bit of work since gentoo won't let me compile it with Mesa 11 for some

Re: [Mesa-dev] [PATCH 1/2] split-to-files.py: fix parsing when #version is not at the beginning

On Mon, Jan 25, 2016 at 10:35 AM, Eero Tamminen wrote: > Hi, > > > On 23.01.2016 15:24, Ilia Mirkin wrote: >> >> On Sat, Jan 23, 2016 at 8:05 AM, Marek Olšák wrote: >>> >>> On Sat, Jan 23, 2016 at 1:28 PM, Ilia Mirkin >>> wrote:

Re: [Mesa-dev] [PATCH 13/17] st/mesa: add mappings for compute shader sysvals

On Sun, Jan 24, 2016 at 4:09 PM, Samuel Pitoiset wrote: > LOCAL_INVOCATION_ID, WORK_GROUP_ID and NUM_WORK_GROUPS are respectively > mapped to THREAD_ID, BLOCK_ID and GRID_SIZE. > > Signed-off-by: Samuel Pitoiset > --- >

Re: [Mesa-dev] [PATCH 11/17] st/mesa: add conversion for compute shaders

On Sun, Jan 24, 2016 at 4:09 PM, Samuel Pitoiset wrote: > The size of shared variables needs to be stored in gl_compute_program > in order to set up pipe_compute_state::req_local_mem. According to the > spec, there are no predefined inputs nor any fixed-function

Re: [Mesa-dev] [PATCH v4 04/11] mesa: add PROGRAM_IMMEDIATE, PROGRAM_BUFFER

Reviewed-by: Marek Olšák Marek On Sun, Jan 24, 2016 at 7:59 PM, Ilia Mirkin wrote: > This makes PROGRAM_IMMEDIATE a first-class gl_register_file type, and > adds PROGRAM_BUFFER to the list. These are used purely inside > glsl_to_tgsi conversion. > >

Re: [Mesa-dev] [PATCH] trace: fix a segfault when tracing indirect draw calls

On Sun, Jan 24, 2016 at 4:47 PM, Samuel Pitoiset wrote: > Like other resources, the indirect draw buffer must be unwrapped. > > Signed-off-by: Samuel Pitoiset > --- > src/gallium/drivers/trace/tr_context.c | 15 ++- > 1 file

Re: [Mesa-dev] gallium r300 driver for PowerPC

On 24.01.2016 20:56, Herminio Hernandez, Jr. wrote: So I believe I have all the debugging symbols installed. From what I am seeing in gdb and valgrind I am still thinking the issue is in the glx branch. For gdb I ran it twice and stopped it during it attempt to load the r300 driver and in it

Re: [Mesa-dev] gallium r300 driver for PowerPC

Someone suggested that I should kill the program at runtime to see what the issue was. I did the same thing with valgrind and saw some similar out puts. See below It is just a sample I can send more output. I wanted to compare the result I got from gdb with what I was seeing with Valgrind.

Re: [Mesa-dev] [PATCH 1/2] glsl: don't attempt to link empty program

On 01/25/2016 03:56 AM, Timothy Arceri wrote: > Previously an empty program would go through the entire > link_shaders() function and we would have to be careful > not to cause a segfault. > > We also now set link_status to false by throwing an error, > it was previously set to true. > > From

[Mesa-dev] [PATCH] i965/bxt: Fix conservative wm thread counts.

When setting the conservative thread counts, I halved everything. That isn't correct for the wm, which has nothing to do with actual thread counts. I suck. BXT only has 1 slice, and there is some ambiguity about subslices, so just reserve the max possible for now. It looks like this might fix:

Re: [Mesa-dev] [PATCH 17/17] st/mesa: expose ARB_compute_shader when compute is supported

On 25.01.2016 10:04, Ilia Mirkin wrote: Yeah so this isn't going to fly. r600/radeonsi both expose PIPE_CAP_COMPUTE, for clover, but won't support these compute shaders quite yet. Not sure what the solution is here... maybe just add another cap? PIPE_CAP_GLSL_COMPUTE or something? Or require a

Re: [Mesa-dev] gallium r300 driver for PowerPC

On 25.01.2016 10:58, Herminio Hernandez Jr. wrote: Someone suggested that I should kill the program at runtime to see what the issue was. I did the same thing with valgrind and saw some similar out puts. See below It is just a sample I can send more output. I wanted to compare the result I

Re: [Mesa-dev] [PATCH 3/3] si-report.py: process Max Waves

For the series: Reviewed-by: Nicolai Hähnle On 23.01.2016 11:07, Marek Olšák wrote: From: Marek Olšák --- si-report.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/si-report.py b/si-report.py index 12e16c3..c7fe1b5

[Mesa-dev] [PATCH 2/2] gallium/ddebug: add 'verbose' option

From: Nicolai Hähnle This currently just writes out the name of dump files, which can be useful to easily correlate those files with other log outputs (driver debug output, apitrace calls, etc.) --- src/gallium/drivers/ddebug/dd_draw.c| 3 ++-

[Mesa-dev] [Bug 89599] symbol 'x86_64_entry_start' is already defined when building with LLVM/clang

https://bugs.freedesktop.org/show_bug.cgi?id=89599 --- Comment #7 from Armin K --- (In reply to Matt Turner from comment #6) > What do I have to do to reproduce this? Mesa builds with clang (3.7.0) > fine for me. I've built with clang occasionally for a few years and > have

[Mesa-dev] [PATCH 1/2] gallium/ddebug: make 'noflush' also affect 'always' mode

From: Nicolai Hähnle This changes the default behavior of 'always' mode to be consistent with hang detection mode. I have used this to more easily compare dumped command streams using diff. --- src/gallium/drivers/ddebug/dd_draw.c | 3 +++

Re: [Mesa-dev] [PATCH] radeonsi: fix shader precompilation for shader-db

On 23.01.2016 11:04, Marek Olšák wrote: From: Marek Olšák The addition of spi_shader_col_format killed all color outputs in precompiled shaders. Reviewed-by: Nicolai Hähnle --- src/gallium/drivers/radeonsi/si_state_shaders.c | 43

[Mesa-dev] [Bug 89599] symbol 'x86_64_entry_start' is already defined when building with LLVM/clang

https://bugs.freedesktop.org/show_bug.cgi?id=89599 --- Comment #6 from Matt Turner --- What do I have to do to reproduce this? Mesa builds with clang (3.7.0) fine for me. I've built with clang occasionally for a few years and have never encountered this problem. -- You are

Re: [Mesa-dev] [PATCH 17/17] st/mesa: expose ARB_compute_shader when compute is supported

On Mon, Jan 25, 2016 at 11:17 AM, Nicolai Hähnle wrote: > On 25.01.2016 10:04, Ilia Mirkin wrote: >> >> Yeah so this isn't going to fly. r600/radeonsi both expose >> PIPE_CAP_COMPUTE, for clover, but won't support these compute shaders >> quite yet. Not sure what the solution

[Mesa-dev] [Bug 93813] Incorrect viewport range when GL_CLIP_ORIGIN is GL_UPPER_LEFT

https://bugs.freedesktop.org/show_bug.cgi?id=93813 --- Comment #4 from Mathias Fröhlich --- Hi, I can't remember why I thought there must be a minus. Today this looks rather wrong to me ... So I think that the proposed change to mesa is a good one. Also I have

[Mesa-dev] [Bug 77449] Tracker bug for all bugs related to Steam titles

https://bugs.freedesktop.org/show_bug.cgi?id=77449 Bug 77449 depends on bug 92944, which changed state. Bug 92944 Summary: [Fiji/LLVM/RadeonSI] CS:GO segfaults in llvm https://bugs.freedesktop.org/show_bug.cgi?id=92944 What|Removed |Added

[Mesa-dev] [PATCH] radeonsi: use llvm.amdgcn.s.barrier instead of llvm.AMDGPU.barrier.local

From: Nicolai Hähnle The new name for the intrinsic was introduced in LLVM r258558. --- src/gallium/drivers/radeonsi/si_shader.c | 7 ++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/src/gallium/drivers/radeonsi/si_shader.c

Re: [Mesa-dev] [PATCH] i965/bxt: Fix conservative wm thread counts.

Tested-by: Mark Janes Ben Widawsky writes: > When setting the conservative thread counts, I halved everything. That isn't > correct for the wm, which has nothing to do with actual thread counts. I suck. > > BXT only has 1 slice, and there is

[Mesa-dev] [PATCH V2 1/2] glsl: don't attempt to link empty program

Previously an empty program would go through the entire link_shaders() function and we would have to be careful not to cause a segfault. In core profile also now set link_status to false by generating an error, it was previously set to true. From Section 7.3 (PROGRAM OBJECTS) of the OpenGL 4.5

Re: [Mesa-dev] [PATCH] i965/bxt: Fix conservative wm thread counts.

On Monday, January 25, 2016 12:34:15 PM PST Ben Widawsky wrote: > When setting the conservative thread counts, I halved everything. That isn't > correct for the wm, which has nothing to do with actual thread counts. I suck. > > BXT only has 1 slice, and there is some ambiguity about subslices,

[Mesa-dev] [Bug 89599] symbol 'x86_64_entry_start' is already defined when building with LLVM/clang

https://bugs.freedesktop.org/show_bug.cgi?id=89599 --- Comment #8 from Tomasz Paweł Gajc --- I've build mesa-11.1.1 without a patch from comment #2. make[4]: Entering directory '/builddir/build/BUILD/mesa-11.1.1/src/mapi' /bin/sh ../../libtool --tag=CC --mode=link

[Mesa-dev] [PATCH 07/12] nir: Add opcodes to extract bytes or words.

The uint versions zero extend while the int versions sign extend. --- src/glsl/nir/nir.h| 3 +++ src/glsl/nir/nir_opcodes.py | 9 + src/glsl/nir/nir_opt_algebraic.py | 16 3 files changed, 28 insertions(+) diff --git a/src/glsl/nir/nir.h

[Mesa-dev] [PATCH 08/12] i965/fs: Implement support for extract_word.

The vec4 backend will lower it. --- src/mesa/drivers/dri/i965/brw_defines.h| 12 src/mesa/drivers/dri/i965/brw_fs_cse.cpp | 2 ++ src/mesa/drivers/dri/i965/brw_fs_generator.cpp | 22 ++ src/mesa/drivers/dri/i965/brw_fs_nir.cpp | 16

[Mesa-dev] [PATCH 06/12] glsl: Remove 2x16 half-precision pack/unpack opcodes.

i965/fs was the only consumer, and we're now doing the lowering in NIR. --- src/glsl/ir.cpp| 9 -- src/glsl/ir.h | 19 src/glsl/ir_optimization.h | 15 ++- src/glsl/ir_validate.cpp

[Mesa-dev] [PATCH 02/12] i965: Move brw_compiler_create() to new brw_compiler.c.

A future patch will want to use designated initalizers, which aren't available in C++, but this is C. --- src/mesa/drivers/dri/i965/Makefile.sources | 1 + src/mesa/drivers/dri/i965/brw_compiler.c | 157 + src/mesa/drivers/dri/i965/brw_compiler.h | 3 +

[Mesa-dev] [PATCH 04/12] nir: Add lowering of nir_op_unpack_half_2x16.

--- src/glsl/nir/nir.h | 3 +++ src/glsl/nir/nir_lower_alu_to_scalar.c | 30 ++ 2 files changed, 29 insertions(+), 4 deletions(-) diff --git a/src/glsl/nir/nir.h b/src/glsl/nir/nir.h index e2bd2bf..1113030 100644 --- a/src/glsl/nir/nir.h +++

[Mesa-dev] [PATCH 12/12] i965/gen7+: Use NIR for lowering of pack/unpack opcodes.

--- src/mesa/drivers/dri/i965/brw_compiler.c | 15 + .../dri/i965/brw_fs_channel_expressions.cpp| 8 +++ src/mesa/drivers/dri/i965/brw_link.cpp | 25 ++ 3 files changed, 29 insertions(+), 19 deletions(-) diff --git

[Mesa-dev] [PATCH 09/12] nir: Add lowering support for packing opcodes.

--- src/glsl/nir/nir.h | 4 src/glsl/nir/nir_lower_alu_to_scalar.c | 32 src/glsl/nir/nir_opcodes.py| 10 ++ src/glsl/nir/nir_opt_algebraic.py | 20 4 files changed, 66 insertions(+) diff

[Mesa-dev] [PATCH 10/12] nir: Add lowering support for unpacking opcodes.

--- src/glsl/nir/nir.h| 4 src/glsl/nir/nir_opt_algebraic.py | 28 2 files changed, 32 insertions(+) diff --git a/src/glsl/nir/nir.h b/src/glsl/nir/nir.h index bbd5b1a..3b90b51 100644 --- a/src/glsl/nir/nir.h +++ b/src/glsl/nir/nir.h @@ -1474,6

[Mesa-dev] [PATCH 01/12] nir: Make argument order of unop_convert match binop_convert.

Strangely the return and parameter types were reversed. --- src/glsl/nir/nir_opcodes.py | 20 ++-- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/src/glsl/nir/nir_opcodes.py b/src/glsl/nir/nir_opcodes.py index e79810c..a8bbe1a 100644 ---

[Mesa-dev] [PATCH 00/12] i965+nir: Do pack/unpack lowering in NIR

This series adds NIR support for lowering pack/unpack opcodes and switches i965 over to using that. The only place we still use the GLSL IR lower_packing_builtins.cpp is for the half-precision functions on Sandybridge. As a result, I get to remove the ir_unop_unpack_half_2x16_split_x/y opcodes

[Mesa-dev] [PATCH 05/12] i965/fs: Switch from GLSL IR to NIR for un/packHalf2x16 lowering.

--- src/mesa/drivers/dri/i965/brw_compiler.c | 2 ++ src/mesa/drivers/dri/i965/brw_fs_channel_expressions.cpp | 4 src/mesa/drivers/dri/i965/brw_link.cpp | 12 +--- 3 files changed, 7 insertions(+), 11 deletions(-) diff --git

[Mesa-dev] [PATCH 03/12] i965: Make separate nir_options for scalar/vector stages.

We'll want to have different lowering options set for scalar/vector stages. --- src/mesa/drivers/dri/i965/brw_compiler.c | 61 +--- 1 file changed, 33 insertions(+), 28 deletions(-) diff --git a/src/mesa/drivers/dri/i965/brw_compiler.c

[Mesa-dev] [PATCH 11/12] i965/vec4: Implement nir_op_pack_uvec2_to_uint.

And mark nir_op_pack_uvec4_to_uint unreachable, since it's only produced by lowering pack[SU]norm4x8 which the vec4 backend does not need. --- src/mesa/drivers/dri/i965/brw_vec4_nir.cpp | 18 ++ 1 file changed, 18 insertions(+) diff --git

Re: [Mesa-dev] [PATCH 04/17] gallium: disable compute shaders for meta ops

On 24.01.2016 16:09, Samuel Pitoiset wrote: Loosely based on tessellation shaders. Do we actually need this? The graphics pipeline and the compute pipeline are separate; draw commands should be unaffected by the currently set compute shader. Or am I missing something? Cheers, Nicolai

Re: [Mesa-dev] [PATCH v4 02/11] glsl: always initialize image_* fields, copy them on interface init

Reviewed-by: Marek Olšák Marek On Sun, Jan 24, 2016 at 7:59 PM, Ilia Mirkin wrote: > Interfaces can have image properties set in case they are buffer > interfaces. Make sure not to lose this information. > > Signed-off-by: Ilia Mirkin

[Mesa-dev] [PATCH 2/6] nv50/ir: swap sources in PostRaConstantFolding when src0 is imm

From: Karol Herbst helps some shaders in multiple games total instructions in shared programs : 1922267 -> 1922121 (-0.01%) total gprs used in shared programs: 251878 -> 251878 (0.00%) total local used in shared programs : 5673 -> 5673 (0.00%) total bytes used in

[Mesa-dev] Command buffer error in vmwgfx kernel module

When running OpenGL app in Linux VM with 3D acceleration enabled and vmwgfx installed, i get this kind of errors in dmesg: [xx.x] [drm:vmw_cmdbuf_work_func [cmwgfx]] *ERROR* Command buffer error. This causes some OpenGL apps to not run at all. System info: VMWare Workstation Pro 12.1.0

[Mesa-dev] [PATCH 5/6] nv50/ir: add PostRADCE Pass

From: Karol Herbst helps shaders in some games total instructions in shared programs : 1901958 -> 1895185 (-0.36%) total gprs used in shared programs: 251739 -> 251739 (0.00%) total local used in shared programs : 5673 -> 5673 (0.00%) total bytes used in shared

[Mesa-dev] [PATCH 1/6] nv50/ir: enable PostRaConstantFolding for [c0, f0)

From: Karol Herbst helps shaders in multiple games total instructions in shared programs : 1926020 -> 1922267 (-0.19%) total gprs used in shared programs: 251878 -> 251878 (0.00%) total local used in shared programs : 5673 -> 5673 (0.00%) total bytes used in shared

[Mesa-dev] [PATCH 4/6] nv50/ir: optimize shl(shr(a, c), c) to and(a, ~((1 << c) - 1))

From: Karol Herbst helps shaders in multiple games total instructions in shared programs : 192 -> 1901958 (-0.48%) total gprs used in shared programs: 251739 -> 251739 (0.00%) total local used in shared programs : 5673 -> 5673 (0.00%) total bytes used in shared

[Mesa-dev] [PATCH 3/6] nv50/ir: optimize neg(add(bool, 1)) to bool for OP_SET and OP_SLCT

From: Karol Herbst helps shaders in saints row IV, bioshock infinite and shadow warrior total instructions in shared programs : 1922121 -> 192 (-0.57%) total gprs used in shared programs: 251878 -> 251739 (-0.06%) total local used in shared programs : 5673 -> 5673

[Mesa-dev] [PATCH 0/6] nv50/ir: various compiler optimizations

changes in my shader-db: total instructions in shared programs : 1926020 -> 1895185 (-1.60%) total gprs used in shared programs: 251878 -> 251739 (-0.06%) total local used in shared programs : 5673 -> 5673 (0.00%) total bytes used in shared programs : 17659256 -> 17379448 (-1.58%)

[Mesa-dev] [PATCH 6/6] nv50/ir: run DCE backwards

reduces Pass rerun by around 40% Signed-off-by: Karol Herbst --- src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp

Re: [Mesa-dev] [PATCH 1/6] nv50/ir: enable PostRaConstantFolding for [c0, f0)

On Mon, Jan 25, 2016 at 9:57 AM, Karol Herbst wrote: > From: Karol Herbst > > helps shaders in multiple games > > total instructions in shared programs : 1926020 -> 1922267 (-0.19%) > total gprs used in shared programs: 251878 -> 251878 (0.00%) >

Re: [Mesa-dev] [PATCH 3/4] radeonsi: add max waves / SIMD to shader stats (v2)

On Fri, Jan 22, 2016 at 03:18:12PM +0100, Marek Olšák wrote: > From: Marek Olšák > > v2: account for LDS usage in PS > the limit is per SIMD, not per CU > --- > src/gallium/drivers/radeonsi/si_shader.c | 54 > +--- > 1 file changed, 49

[Mesa-dev] [PATCH V3 1/2] glsl: don't attempt to link empty program

Previously an empty program would go through the entire link_shaders() function and we would have to be careful not to cause a segfault. In core profile also now set link_status to false by generating an error, it was previously set to true. From Section 7.3 (PROGRAM OBJECTS) of the OpenGL 4.5

Re: [Mesa-dev] [Mesa-dev, v6, 1/3] clover: separate compile and link stages

Hi, >diff --git a/src/gallium/state_trackers/clover/api/program.cpp >b/src/gallium/state_trackers/clover/api/program.cpp >index 27ca2ef..f8d946e 100644 >--- a/src/gallium/state_trackers/clover/api/program.cpp >+++ b/src/gallium/state_trackers/clover/api/program.cpp >@@ -181,13 +181,20 @@

Re: [Mesa-dev] [PATCH V3 1/2] glsl: don't attempt to link empty program

On 01/25/2016 04:46 PM, Timothy Arceri wrote: > Previously an empty program would go through the entire > link_shaders() function and we would have to be careful > not to cause a segfault. > > In core profile also now set link_status to false by > generating an error, it was previously set to

Re: [Mesa-dev] [PATCH] mapi: Fix build with clang

On 25.1.2016 3:57, Matt Turner wrote: On Sun, Jan 24, 2016 at 2:53 AM, Armin K wrote: From: Tomasz Paweł Gajc Currently, building with clang fails due to: error in backend: symbol 'x86_64_entry_start' is already defined This patch fixes it. Bugzilla:

Re: [Mesa-dev] [PATCH] radeonsi: fix shader precompilation for shader-db

On 24.01.2016 01:04, Marek Olšák wrote: > From: Marek Olšák > > The addition of spi_shader_col_format killed all color outputs > in precompiled shaders. Reviewed-by: Michel Dänzer -- Earthling Michel Dänzer |

Re: [Mesa-dev] [PATCH 1/2] configure.ac: don't require EGL/DRM ang GBM if OpenGL is disabled

On 23.01.2016 02:14, Marek Olšák wrote: > On Fri, Jan 22, 2016 at 6:02 PM, Emil Velikov > wrote: >> On 22 January 2016 at 16:50, Marek Olšák wrote: >>> On Fri, Jan 22, 2016 at 5:32 PM, Emil Velikov >>> wrote: On 22

Re: [Mesa-dev] [RFC libdrm] intel: Add support for softpin

On Wed, Sep 09, 2015 at 04:07:10PM +0200, Michał Winiarski wrote: > Softpin allows userspace to take greater control of GPU virtual address > space and eliminates the need of relocations. It can also be used to > mirror addresses between GPU and CPU (shared virtual memory). > Calls to

Re: [Mesa-dev] [PATCH 5/6] glsl: don't sort varying in separate shader mode

On Mon, 2016-01-25 at 12:47 +0200, Tapani Pälli wrote: > > On 01/25/2016 12:29 PM, Timothy Arceri wrote: > > On Mon, 2016-01-25 at 16:41 +1100, Timothy Arceri wrote: > > > On Mon, 2015-11-30 at 14:31 +0200, Tapani Pälli wrote: > > > > Reviewed-by: Tapani Pälli > > > > >

Re: [Mesa-dev] [PATCH 5/6] glsl: don't sort varying in separate shader mode

On Mon, 2016-01-25 at 16:41 +1100, Timothy Arceri wrote: > On Mon, 2015-11-30 at 14:31 +0200, Tapani Pälli wrote: > > Reviewed-by: Tapani Pälli > > > > On 11/25/2015 11:54 AM, Timothy Arceri wrote: > > > From: Gregory Hainaut > > > > > > This

Re: [Mesa-dev] [PATCH 5/6] glsl: don't sort varying in separate shader mode

On 01/25/2016 12:29 PM, Timothy Arceri wrote: On Mon, 2016-01-25 at 16:41 +1100, Timothy Arceri wrote: On Mon, 2015-11-30 at 14:31 +0200, Tapani Pälli wrote: Reviewed-by: Tapani Pälli On 11/25/2015 11:54 AM, Timothy Arceri wrote: From: Gregory Hainaut

Re: [Mesa-dev] [PATCH 5/6] glsl: don't sort varying in separate shader mode

On 01/25/2016 01:14 PM, Timothy Arceri wrote: On Mon, 2016-01-25 at 12:47 +0200, Tapani Pälli wrote: On 01/25/2016 12:29 PM, Timothy Arceri wrote: On Mon, 2016-01-25 at 16:41 +1100, Timothy Arceri wrote: On Mon, 2015-11-30 at 14:31 +0200, Tapani Pälli wrote: Reviewed-by: Tapani Pälli

[Mesa-dev] [PATCH 2/2] glsl: clean up and fix bug in varying linking rules

The existing code was very hard to follow and has been the source of at least 3 bugs in the past year. The existing code also has a bug for SSO where if we have a multi-stage SSO for example a tes -> gs program, if we try to use transform feedback with gs the existing code would look for the

[Mesa-dev] [PATCH 1/2] glsl: don't attempt to link empty program

Previously an empty program would go through the entire link_shaders() function and we would have to be careful not to cause a segfault. We also now set link_status to false by throwing an error, it was previously set to true. From Section 7.3 (PROGRAM OBJECTS) of the OpenGL 4.5 spec:

Re: [Mesa-dev] [PATCH v4 01/11] tgsi: add MEMBAR opcode to handle memoryBarrier* GLSL intrinsics

Reviewed-by: Roland Scheidegger Am 24.01.2016 um 19:59 schrieb Ilia Mirkin: > Signed-off-by: Ilia Mirkin > Reviewed-by: Marek Olšák (v1) > > v1 -> v2: add defines for the various bits > --- >

Re: [Mesa-dev] [PATCH 5/6] glsl: don't sort varying in separate shader mode

On Mon, 2016-01-25 at 13:27 +0200, Tapani Pälli wrote: > > On 01/25/2016 01:14 PM, Timothy Arceri wrote: > > On Mon, 2016-01-25 at 12:47 +0200, Tapani Pälli wrote: > > > > > > On 01/25/2016 12:29 PM, Timothy Arceri wrote: > > > > On Mon, 2016-01-25 at 16:41 +1100, Timothy Arceri wrote: > > > > >

Re: [Mesa-dev] [PATCH 1/2] configure.ac: don't require EGL/DRM ang GBM if OpenGL is disabled

On Mon, Jan 25, 2016 at 10:24 AM, Michel Dänzer wrote: > On 23.01.2016 02:14, Marek Olšák wrote: >> On Fri, Jan 22, 2016 at 6:02 PM, Emil Velikov >> wrote: >>> On 22 January 2016 at 16:50, Marek Olšák wrote: On Fri, Jan 22,

Re: [Mesa-dev] [PATCH 5/6] nv50/ir: add PostRADCE Pass

On Mon, Jan 25, 2016 at 9:57 AM, Karol Herbst wrote: > From: Karol Herbst > > helps shaders in some games > > total instructions in shared programs : 1901958 -> 1895185 (-0.36%) > total gprs used in shared programs: 251739 -> 251739 (0.00%) >

Re: [Mesa-dev] [PATCH 1/2] configure.ac: don't require EGL/DRM ang GBM if OpenGL is disabled

On 25.01.2016 21:08, Marek Olšák wrote: > On Mon, Jan 25, 2016 at 10:24 AM, Michel Dänzer wrote: >> On 23.01.2016 02:14, Marek Olšák wrote: >>> On Fri, Jan 22, 2016 at 6:02 PM, Emil Velikov >>> wrote: On 22 January 2016 at 16:50, Marek Olšák

Re: [Mesa-dev] [PATCH V3 1/2] glsl: don't attempt to link empty program

On Mon, 2016-01-25 at 16:58 -0800, Ian Romanick wrote: > On 01/25/2016 04:46 PM, Timothy Arceri wrote: > > Previously an empty program would go through the entire > > link_shaders() function and we would have to be careful > > not to cause a segfault. > > > > In core profile also now set

Re: [Mesa-dev] [PATCH] radeonsi: use llvm.amdgcn.s.barrier instead of llvm.AMDGPU.barrier.local

On 26.01.2016 06:49, Nicolai Hähnle wrote: > From: Nicolai Hähnle > > The new name for the intrinsic was introduced in LLVM r258558. > --- > src/gallium/drivers/radeonsi/si_shader.c | 7 ++- > 1 file changed, 6 insertions(+), 1 deletion(-) > > diff --git

Re: [Mesa-dev] [PATCH] radeonsi: use llvm.amdgcn.s.barrier instead of llvm.AMDGPU.barrier.local

On 26.01.2016 06:49, Nicolai Hähnle wrote: > From: Nicolai Hähnle > > The new name for the intrinsic was introduced in LLVM r258558. > --- > src/gallium/drivers/radeonsi/si_shader.c | 7 ++- > 1 file changed, 6 insertions(+), 1 deletion(-) > > diff --git