[Mesa-dev] Mesa i965 instruction scheduler latency values

Hi, Mesa i965 instruction scheduler latency values look weird. For example: -- void schedule_node::set_latency_gen7(bool is_haswell) { switch (inst->opcode) { ... case SHADER_OPCODE_UNTYPED_SURFACE_READ: case SHADER_OPCODE_UNTYPED_SURFACE_WRITE: case SHADER_OPCODE

Re: [Mesa-dev] [PATCH 1/1] i965: Do not overwrite optimizer dumps

Hi, On 11/25/2015 08:52 PM, Jason Ekstrand wrote: On Wed, Nov 25, 2015 at 10:49 AM, Kenneth Graunke wrote: [...] It almost seems like you want a hash of the program key - and maybe a file containing a printed copy of the key so you can identify which program is which. Alternatively, one coul

Re: [Mesa-dev] [PATCH shader-db 1/3] split-to-files: deal with minimum versions, other shader types

sets (that re-use same program numbers) from the same application. RFC patches for my changes are attached. - Eero >From c52f02ff664af269fa5268627624fe94c647ad37 Mon Sep 17 00:00:00 2001 From: Eero Tamminen Date: Wed, 9 Dec 2015 16:29:43 +0200 Subject: [PATCH 1/2] Rewrite

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

Hi, On 12/14/2015 10:44 AM, Herminio Hernandez, Jr. wrote: I am new to this list. I have been trying to see if I can fix or at least pin point an issue with Radeon r300 driver failing on PowerPC systems. This has been a problem for a while and I would like to help to get this fixed. I have do

Re: [Mesa-dev] [PATCH 6/7] nir: Teach nir_opt_algebraic about adding and subtracting the same thing

Hi, On 12/17/2015 01:52 AM, Matt Turner wrote: On Tue, Dec 15, 2015 at 1:16 AM, Eduardo Lima Mitev wrote: On 12/15/2015 09:28 AM, Kristian Høgsberg Kristensen wrote: This optimizes a + b - b to just a. Modest shader-db results (BDW): total instructions in shared programs: 7842452 -> 78418

Re: [Mesa-dev] [PATCH 2/3] i965: Use the upload BO for push constants on Gen9+.

Hi, On 13.05.2017 22:26, Kenneth Graunke wrote: Shaders can use quite a bit of uniform data. Better to put it in the upload buffers, like we do for client vertex data, rather than the batch buffer state area, which is primarly used for indirect state. This should free up batch space, allowing

Re: [Mesa-dev] [PATCH] i965: Don't try to resolve CCS with MESA_FORMAT_NONE.

Hi, On 04.06.2017 22:26, Kenneth Graunke wrote: Since commit e9f5004d5ea38a21019e0affcd63d353b98d499a, KDE's Plasma Shell started crashing when intel_disable_rb_aux_buffer tried to dereference irb->mt, and there was no miptree. This happened during intel_texture_view_requires_resolve, coming fr

Re: [Mesa-dev] [PATCH] i965: Don't try to resolve CCS with MESA_FORMAT_NONE.

Hi, On 06.06.2017 02:58, Kenneth Graunke wrote: On Monday, June 5, 2017 2:03:45 AM PDT Tapani Pälli wrote: FWIW this change fixes also regression on Android wallpaper since that commit. I'm planning on dropping this patch, as it seems that commit 708664159e18487b6676fd5b4c33f52003f81d9e Auth

Re: [Mesa-dev] i965: disable_throttling

Hi, On 09.06.2017 09:21, Marathe, Yogesh wrote: I’m looking forward to set this brw->disable_throttling to true, I’m actually observing adverse effect on performance benchmarks after I force set that to true. What’s the expectation from disable_throttling here? By default Mesa throttles its bu

Re: [Mesa-dev] i965: disable_throttling

Hi, On 12.06.2017 06:58, Marathe, Yogesh wrote: -Original Message- On 09.06.2017 09:21, Marathe, Yogesh wrote: I’m looking forward to set this brw->disable_throttling to true, I’m actually observing adverse effect on performance benchmarks after I force set that to true. What’s the expe

Re: [Mesa-dev] [PATCH 00/31] Mesa: Moving from _NEW flags to NewDriverState

Hi, Do you have a branch I could check with our automation? On 13.06.2017 02:34, Marek Olšák wrote: Edmondo on IRC reported that this series improves Civilization 5 performance and the improvement is not just tiny. A potentially interesting synthetic test-case for this could be "Driver2" tes

Re: [Mesa-dev] [PATCH 00/31] Mesa: Moving from _NEW flags to NewDriverState

- Eero Marek On Tue, Jun 13, 2017 at 9:58 AM, Eero Tamminen wrote: Hi, Do you have a branch I could check with our automation? On 13.06.2017 02:34, Marek Olšák wrote: Edmondo on IRC reported that this series improves Civilization 5 performance and the improvement is not just tiny. A po

Re: [Mesa-dev] [PATCH 00/31] Mesa: Moving from _NEW flags to NewDriverState

Hi, On 13.06.2017 16:23, Marek Olšák wrote: On Tue, Jun 13, 2017 at 3:24 PM, Eero Tamminen wrote: On 13.06.2017 12:56, Marek Olšák wrote: Everything is here: git://people.freedesktop.org/~mareko/mesa state-optz Didn't see any changes in benchmark suites that had tests which are part

Re: [Mesa-dev] [PATCH 00/31] Mesa: Moving from _NEW flags to NewDriverState

Eero Regards //Ernst 2017-06-13 16:50 GMT+02:00 Eero Tamminen : Hi, On 13.06.2017 16:23, Marek Olšák wrote: On Tue, Jun 13, 2017 at 3:24 PM, Eero Tamminen wrote: On 13.06.2017 12:56, Marek Olšák wrote: Everything is here: git://people.freedesktop.org/~mareko/mesa state-optz Didn'

Re: [Mesa-dev] [PATCH 10/21] i965: Only add the wpos state reference if we lowered something

Hi Jason, This commit dropped SynMark2 v7 OglShMapPcf performance by 1-2% on GT2 versions of SKL, KBL and CFL. It may have improved it by same amount on SKL GT4e. Effects on other machines were within normal variation. - Eero On 30.09.2017 00:25, Jason Ekstrand wrote: Otherwise, i

Re: [Mesa-dev] Meson's default build type

Hi, On 04.11.2017 03:28, Christian Schmidbauer wrote: On Thu, Nov 2, 2017 at 11:45 AM, Andres Rodriguez wrote: [...] In the autotools system we have today, we have --enable-debug, which adds -g and -O0 if some -g* and -O* are not already in CFLAGS. It also adds -DDEBUG which turns on lots of

Re: [Mesa-dev] [PATCH] i965 : Performance Improvement

Hi, On 14.07.2017 09:38, Marathe, Yogesh wrote: [...] The only reason I could see this helping is if check_state() wasn't inlined, but a release build with -O2 definitely inlines both check_and_emit_atom() and check_state(). Are you using GCC? What are your CFLAGS? -O2? I hope you're not try

Re: [Mesa-dev] [RFC PATCH v1 28/30] RFC: anv: Enable VK_EXT_image_drm_format_modifier

Hi, On 07.11.2017 16:48, Chad Versace wrote: The draft spec lives at http://kiwitree.net/~chadv/vulkan/#1.0-VK_EXT_image_drm_format_modifier. This extension spec has a complete, working API. I'm happy with the API, and believe it's good enough. But the spec language has some outstanding issues

Re: [Mesa-dev] [PATCH 3/3] i965/gen6-7/sol: Bump primitive counter BO size.

Hi, Tested-By: Eero Tamminen On 18.11.2017 00:28, Francisco Jerez wrote: Improves performance of SynMark2 OglGSCloth by a further 9.65%±0.59% due to the reduction in overwraps of the primitive count buffer that lead to a CPU stall on previous rendering. Cummulative performance improvement

Re: [Mesa-dev] GPU (and system) monitoring

Hi, On 21.11.2017 02:39, Gordon Haverland wrote: On Mon, 20 Nov 2017 13:25:26 +0100 Nicolai Hähnle wrote: ... umr does that for the amdgpu kernel module I downloaded the source tree, compiled umr, mounted the debugfs to use umr, and ran umr --top switching to sensor mode, you could see

Re: [Mesa-dev] [PATCH 00/30] i965: Add support for I915_FORMAT_MOD_Y_TILED_CCS

Hi, I run this on few GEN9 machines and SynMark v7 DeferredAA test improved ~3%, so it seems multisample fast-clears do help. :-) - Eero On 17.06.2017 01:41, Jason Ekstrand wrote: This series is a rework of Ben's series to enable the CCS format modifier. It started as an attempt to

Re: [Mesa-dev] [PATCH 3/3] glsl: relax loop unrolling restrictions

Hi, On 21.06.2017 13:12, Timothy Arceri wrote: The main reason these restriction exist is because glsl the loop unrolling pass is super slow with large loops. be5f27a84d0d fixed things so that expression trees were counted against the limit, however it left the limit as max_iterations * 5 which

Re: [Mesa-dev] [PATCH shader-db] Drop Orbital Explorer shader.

Hi, On 22.06.2017 23:14, Chad Versace wrote: On Thu 22 Jun 2017, Chad Versace wrote: On Thu 22 Jun 2017, Kenneth Graunke wrote: The author eventually emailed me and said that he considers it a "finished experiment" and said the rendering method (geometry shader based approach) is inefficient,

Re: [Mesa-dev] [PATCH shader-db] Drop Orbital Explorer shader.

Hi, On 27.06.2017 20:13, Chad Versace wrote: On Mon 26 Jun 2017, Eero Tamminen wrote: On 22.06.2017 23:14, Chad Versace wrote: On Thu 22 Jun 2017, Chad Versace wrote: On Thu 22 Jun 2017, Kenneth Graunke wrote: The author eventually emailed me and said that he considers it a "fin

Re: [Mesa-dev] [PATCH 1/2] intel/fs: Implement GRF bank conflict mitigation pass.

Hi, Thanks for finally having this handled in Mesa! This patch series, live intervals and "Don't let undefined values prevent copy propagation" commits help performance in following tests: * GfxBench CarChase (2% by live intervals) * GfxBench AztecRuins & Manhattan 3.0 (very marginally) * GfxB

Re: [Mesa-dev] [PATCH 0/2] i965: scratch space fixes

Hi, Tested-by: Eero Tamminen On 12.12.2017 12:04, kevin.rogo...@intel.com wrote: From: Kevin Rogovin This patch series fixes 2 issues for scratch space on compute shaders for GEN. Together with the ASTC5x5 fixes posted before, carchase on GLES works on my SKL GT4. These patches fix GPU

Re: [Mesa-dev] [PATCH 1/2] intel/fs: Implement GRF bank conflict mitigation pass.

Hi, On 11.12.2017 12:28, Eero Tamminen wrote: Thanks for finally having this handled in Mesa! This patch series, live intervals and "Don't let undefined values prevent copy propagation" commits help performance in following tests: * GfxBench CarChase (2% by live intervals)

Re: [Mesa-dev] [PATCH 0/2] i965: scratch space fixes (v2)

Hi, Tested-by: Eero Tamminen Fixes GPU hangs I'm seeing also with the *GL* version of CarChase on KBL GT3e (when using Ubuntu 16.04 kernel). - Eero On 12.12.2017 14:17, kevin.rogo...@intel.com wrote: From: Kevin Rogovin This patch series offers a readability improvemen

Re: [Mesa-dev] [PATCH 3/3] Revert "i965: Disable regular fast-clears (CCS_D) on gen9+"

in the render cache, it should     be safe to re-enable CCS_D for sRGB. * s/How taht/Now that/ * Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=104163 * Tested-by: Eero Tamminen - Eero On Wed, Dec 13, 2017 at 5:52 PM, Jason Ekstrand <mailto:ja...@jlekstrand.net>>

Re: [Mesa-dev] [PATCH] intel/fs: Optimize and simplify the copy propagation dataflow logic.

Hi, On 20.12.2017 16:29, Eero Tamminen wrote: I got unexpected results with this, when testing it on BXT & SKL GT2. While performance in GpuTest Volplosion and GfxBench v4 Tessellation test improved slightly, performance in SynMark v7 CSDof and GpuTest v0.7 Piano dropped clearly. P

Re: [Mesa-dev] [PATCH] intel/fs: Optimize and simplify the copy propagation dataflow logic.

Hi, I got unexpected results with this, when testing it on BXT & SKL GT2. While performance in GpuTest Volplosion and GfxBench v4 Tessellation test improved slightly, performance in SynMark v7 CSDof and GpuTest v0.7 Piano dropped clearly. Piano dropped only on SKL, but there the drop was >10

[Mesa-dev] [PATCH] spirv: avoid infinite loop / freeze in vtn_cfg_walk_blocks()

Fixes: 9702fac68e (spirv: consider bitsize when handling OpSwitch cases) Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104359 --- src/compiler/spirv/vtn_cfg.c | 14 +++--- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/src/compiler/spirv/vtn_cfg.c b/src/compiler/sp

Re: [Mesa-dev] [PATCHv2] intel/fs: Optimize and simplify the copy propagation dataflow logic.

Hi, I tested this on HSW GT2, BXT & SKL GT3e, and didn't see any significant regressions this time. I'll try it also on a machine with smaller variance than those (now that it became free), and send a note if that does show something. - Eero On 20.12.2017 21:27, Francisco Jerez wro

Re: [Mesa-dev] [PATCH] spirv: avoid infinite loop / freeze in vtn_cfg_walk_blocks()

t disables testing on master. My fix is rather obvious, just moving few lines, to make sure loop is incremented every round, like it was before the bad commit. - Eero Eero Tamminen writes: Fixes: 9702fac68e (spirv: consider bitsize when handling OpSwitch cases) Bugzi

Re: [Mesa-dev] [PATCH] spirv: avoid infinite loop / freeze in vtn_cfg_walk_blocks()

Hi, On 22.12.2017 11:42, Eero Tamminen wrote: On 21.12.2017 22:19, Mark Janes wrote: This patch doesn't apply to master as formatted. It was against master.  Could you try the attached patch instead? Argh, I meant the patch attached to this mail. Sorry again. - Eero I ass

Re: [Mesa-dev] [PATCH] spirv: Loosten the validation for load/store type matching

with them, it works fine. Tested-by: Eero Tamminen On Mon, Jan 1, 2018 at 8:09 PM, Jason Ekstrand <mailto:ja...@jlekstrand.net>> wrote: Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104424 <https://bugs.freedesktop.org/show_bug.cgi?id=104424> ---

Re: [Mesa-dev] [PATCH 0/6] i965: Misc sRGB and CCS fixes

CarChase & SynMark Deferred KBL GT3e improvements: ~2% in GfxBench Manhattan 3.1 & CarChase and SynMark Deferred By accident I had left Aztec Ruins out, but I assume ~1% drop in that is also fixed. Tested-by: Eero Tamminen - Eero [1] SynMark DrvState drops few percent be

Re: [Mesa-dev] [RFC] NIR serialization

Hi, On 13.09.2017 19:08, Jason Ekstrand wrote: On Wed, Sep 13, 2017 at 8:42 AM, Nicolai Hähnle <mailto:nhaeh...@gmail.com>> wrote: On 13.09.2017 11:54, Eero Tamminen wrote: On 12.09.2017 09:55, Jordan Justen wrote: On 2017-09-11 21:44:32, Timothy Arc

Re: [Mesa-dev] [PATCH] intel/compiler/gen9: Pixel shader header only workaround

Hi, Ran several tens of rounds on 2 different 18 EU BXT devices with the patch -> no hangs. Earlier that (internal) test-case always hung within 1-3 rounds. Tested-by: Eero Tamminen On 25.09.2017 16:39, Topi Pohjolainen wrote: Fixes intermittent GPU hangs on Broxton with an Intel inter

Re: [Mesa-dev] [PATCH] glsl: Trigger precision mismatch error only if both symbols are used

Hi, On 26.09.2017 11:33, Tomasz Figa wrote: An update on my testing inline. On Tue, Sep 26, 2017 at 7:13 AM, Tomasz Figa wrote: On Tue, Sep 26, 2017 at 6:21 AM, Tomasz Figa wrote: Hi Ian, On Tue, Sep 26, 2017 at 5:59 AM, Ian Romanick wrote: On 09/25/2017 02:53 AM, Tomasz Figa wrote: Com

Re: [Mesa-dev] [PATCH v2 7/7] i965: make use of nir linking

Hi, On 26.09.2017 12:04, Timothy Arceri wrote: On 26/09/17 17:50, Kenneth Graunke wrote: [...] 2. It would be great to use this in anv as well, especially given that     it has no GLSL IR linking to do this sort of stuff.  Plus, the SPIR-V     might be compiled independently...and when you bui

Re: [Mesa-dev] [PATCH] intel/compiler/gen9: Pixel shader header only workaround

Hi, Is there some reason/problem why this isn't in upstream repo yet? If not, could somebody push it? (Topi is currently on vacation) - Eero On 25.09.2017 17:41, Eero Tamminen wrote: Ran several tens of rounds on 2 different 18 EU BXT devices with the patch -> no hangs.

Re: [Mesa-dev] [PATCH] mesa: Use a 565 format for GL_RGB and GL_UNSIGNED_SHORT_5_6_5 textures.

e any performance changes either. Tested-by: Eero Tamminen Reviewed-by: Eric Anholt ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev

Re: [Mesa-dev] [PATCH v2 05/12] meson: de-tabularize meson_options.txt

Hi, On 05.10.2017 20:12, Dylan Baker wrote: This ends up being unworkable as more options get added, and with description wrapped onto a new line it doesn't improve readability anyway. Signed-off-by: Dylan Baker --- meson_options.txt | 12 ++-- 1 file changed, 6 insertions(+), 6 del

Re: [Mesa-dev] [PATCH] i965: Fix src0 vs src1 typo

Hi, On 03.10.2017 08:20, Matt Turner wrote: A typo caused us to copy src0's reg file to src1 rather than reading src1's as intended. This caused us to fail to compact instructions like mov(8) g4<1>D0D { align1 1Q }; because src1 was set to immediate rather than architect

Re: [Mesa-dev] [PATCH] mesa: Use a 565 format for GL_RGB and GL_UNSIGNED_SHORT_5_6_5 textures.

Hi, On 05.10.2017 18:15, Eero Tamminen wrote: On 05.10.2017 01:16, Eric Anholt wrote: Kenneth Graunke writes: Found while trying to optimize an application. Not observed to help performance on i965, but should at least reduce the memory usage of such textures a bit. I run benchmarks on

Re: [Mesa-dev] [PATCH] configure.ac: set -O0 on --enable-debug

Hi, On 16.10.2017 02:35, Jason Ekstrand wrote: I'm not sure what I think of this patch.  On the one hand, I'm constantly setting CFLAGS="-O0 -ggdb3" when I configure and it seems like a nicer default. I use a script for configuring Mesa with suitable options (as CFLAGS isn't not the only thi

Re: [Mesa-dev] [PATCH] configure.ac: set -O0 on --enable-debug

Hi, On 16.10.2017 12:40, Eero Tamminen wrote: On 16.10.2017 02:35, Jason Ekstrand wrote: I'm not sure what I think of this patch.  On the one hand, I'm constantly setting CFLAGS="-O0 -ggdb3" when I configure and it seems like a nicer default. I use a script for c

Re: [Mesa-dev] [PATCH] nir: fix loop iteration count calculation for floats

Hi, Tested-by: Eero Tamminen May also impact other programs having float indexed loops: - Invisible Inc - Talos Principle - Steam Big Picture - Serious Sam 3 - Eero On 04.01.2017 02:59, Timothy Arceri wrote: Fixes performance regression in SynMark PSPom caused by loops with float

Re: [Mesa-dev] [PATCH 0/8] i965: gl_TessLevel rescrambling in NIR

Hi, On 04.01.2017 13:07, Kenneth Graunke wrote: This series reworks i965's handling of gl_TessLevelInner/Outer[] arrays. Instead of using lower_tess_levels to turn them into vec4/vec2s, we pass them through to NIR and make them compact arrays (where array indexing translates to enhanced layouts

Re: [Mesa-dev] [PATCH 0/8] i965: gl_TessLevel rescrambling in NIR

Hi, On 05.01.2017 01:55, Kenneth Graunke wrote: On Wednesday, January 4, 2017 3:16:41 PM PST Eero Tamminen wrote: Are there yet other use-cases for Vulkan tessellation besides Sacha Willems' three tests here: https://github.com/SaschaWillems/Vulkan ? Does it matter? It'

Re: [Mesa-dev] [PATCH 11/11] anv: Enable tessellation shaders.

Hi, On 09.01.2017 07:26, Kenneth Graunke wrote: Signed-off-by: Kenneth Graunke --- src/intel/vulkan/anv_device.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) I'm not clear whether we want to do this yet. On Skylake, the status is: 99.3% of the CTS passing, Sascha demos work. Teste

Re: [Mesa-dev] [PATCH 11/11] anv: Enable tessellation shaders.

Hi, On 11.01.2017 20:07, Lionel Landwerlin wrote: On 11/01/17 17:02, Eero Tamminen wrote: On 09.01.2017 07:26, Kenneth Graunke wrote: Signed-off-by: Kenneth Graunke --- src/intel/vulkan/anv_device.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) I'm not clear whether we want

Re: [Mesa-dev] [PATCH] i965: Unbind deleted shaders from brw_context, fixing malloc heisenbug.

Hi, On 27.01.2017 00:08, Kenneth Graunke wrote: On Thursday, January 26, 2017 12:00:14 PM PST Eric Anholt wrote: Kenneth Graunke writes: Any reason you're not just using _mesa_reference_program()? That might be a better plan. Conceptually, I thought of this more as a "weak reference" - we w

Re: [Mesa-dev] very strange intermittent frame-dropping

Hi, On 04.04.2017 09:23, Luke Kenneth Casson Leighton wrote: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=848895 there's a really strange and comprehensively cross-application intermittent bug that's only occurring on opengl-based applications, that's been introduced some time in the past

Re: [Mesa-dev] [PATCH] anv/query: Busy-wait for available query entries

tu default) powersave [1] or performance governor with Mesa master, the improvement is 10-25%. Even with everything set to Ultra, there's small improvement. Tested-by: Eero Tamminen - Eero [1] With powersave scheduler, without this patch kernel doesn't even request GPU to run

Re: [Mesa-dev] very strange intermittent frame-dropping

Hi, On 04.04.2017 14:10, Luke Kenneth Casson Leighton wrote: On Tue, Apr 4, 2017 at 11:04 AM, Eero Tamminen wrote: anyway, apologies: nobody whose applications are affected by this really has a clue where the *actual* bug is so i am escalating it down the chain of library dependencies, making

Re: [Mesa-dev] Enabling verbose modes in mesa and its libraries

Hi, On 05.04.2017 13:11, J Karanje, Kedar wrote: Could someone please send a link to a comprehensive document on enabling the various DEBUG levels within mesa, I’ve already tried : https://www.mesa3d.org/debugging.html , however this just mentions MESA_DEBUG (which only dumps errors) ,

Re: [Mesa-dev] very strange intermittent frame-dropping

Hi, On 06.04.2017 00:36, Luke Kenneth Casson Leighton wrote: eero, apologies, the mesa-dev list's traffic is very high, i'll switch over to digest, so please do cc me each time, to mitigate that. OK. there's a whole stack of people reporting this occurring, running the full range of window

Re: [Mesa-dev] [PATCH 00/53] intel/fs: SIMD32 support for fragment shaders

ave a good enough heuristic to start turning it on by default. The objective of this series is to just to get the compiler infrastructure landed so that it stops bit-rotting in Curro's branch. Tested v3 on BXT & SKL. Everything seems to work otherwise fine. Tested-by Eero Tamminen

Re: [Mesa-dev] [PATCH 00/53] intel/fs: SIMD32 support for fragment shaders

Hi, On 29.05.2018 18:58, Eero Tamminen wrote: On 25.05.2018 00:55, Jason Ekstrand wrote: This patch series adds back-end compiler support for SIMD32 fragment shaders.  Support is added and everything works but it's currently hidden behind INTEL_DEBUG=do32.  We know that it improves perfor

Re: [Mesa-dev] [PATCH 00/53] intel/fs: SIMD32 support for fragment shaders

Hi, On 29.05.2018 18:58, Eero Tamminen wrote: On 25.05.2018 00:55, Jason Ekstrand wrote: This patch series adds back-end compiler support for SIMD32 fragment shaders.  Support is added and everything works but it's currently hidden behind INTEL_DEBUG=do32.  We know that it improves perfor

Re: [Mesa-dev] [PATCH 00/53] intel/fs: SIMD32 support for fragment shaders

Hi, On 30.05.2018 17:30, Jason Ekstrand wrote: On May 30, 2018 06:45:29 Eero Tamminen wrote: On 29.05.2018 18:58, Eero Tamminen wrote: On 25.05.2018 00:55, Jason Ekstrand wrote: This patch series adds back-end compiler support for SIMD32 fragment shaders.  Support is added and everything

Re: [Mesa-dev] [PATCH 00/16] Move the Mesa Website to Sphinx

Hi, On 07.06.2018 12:01, Erik Faye-Lund wrote: Just as a fun toy, I decided to give an animated SVG "variation" of this a go myself: https://codepen.io/kusma/pen/vrXppL The actual SVG can be found here: https://gitlab.freedesktop.org/snippets/492 The gears were generated by this python scrip

Re: [Mesa-dev] Mesa logo animation (was: Move the Mesa Website to Sphinx)

Hi, On 08.06.2018 15:06, robdcl...@gmail.com wrote: On Fri, Jun 8, 2018 at 3:02 AM, Jordan Justen wrote: On Thu, Jun 7, 2018 at 2:56 AM Eero Tamminen wrote: ... For Mesa, WebGL could be more fitting implementation than SVG though... https://github.com/gears3d/gears3d.github.io/blob

Re: [Mesa-dev] [PATCH 6/7] mesa/util: add allow_glsl_relaxed_es driconfig override

Hi, On 08.06.2018 15:19, Timothy Arceri wrote: This relaxes a number of ES shader restrictions allowing shaders to follow more desktop GLSL like rules. This initial implementation relaxes the following: - allows linking ES shaders with desktop shaders - allows mismatching precision qualifi

Re: [Mesa-dev] [PATCH] anv/allocator: Don't srink either end of the block pool

o not shrink in either direction. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105374 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106147 Verified that "texture3d" demo works with this patch, crashes without it. Tested-by: Eero Tamminen - Eero Cc: mesa-sta

Re: [Mesa-dev] [PATCH] mesa/main: Rework the shader capture naming convention

Hi, On 28.04.2018 17:39, Benedikt Schemmer wrote: Change from a purely number.shader_test naming scheme to sha_number.shader_test because especially games often use the same number for entirely different shaders based on graphics settings etc. and then already captured shaders get overwritten.

Re: [Mesa-dev] [PATCH] dri3: Only update number of back buffers in loader_dri3_get_buffers

patch, same compiz process works fine for the whole run. -> looks good to me. Tested-by: Eero Tamminen - Eero src/loader/loader_dri3_helper.c | 19 +++ 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/src/loader/loader_dri3_helper.c b/src/loader/loade

Re: [Mesa-dev] [PATCH 0/4] improve buffer cache and reuse

Hi, On 02.05.2018 02:25, James Xiong wrote: From: "Xiong, James" With the current implementation, brw_bufmgr may round up a request size to the next bucket size, result in 25% more memory allocated in the worst senario. For example: Request sizeActual size 32KB+1Byte 40KB . 8MB+1Byte

Re: [Mesa-dev] [PATCH v3] i965: Fix ETC2/EAC GetCompressed* functions on Gen7 GPUs

Hi, On 02.05.2018 20:19, Matt Turner wrote: On Wed, May 2, 2018 at 9:13 AM, Eleni Maria Stea wrote: Gen 7 GPUs store the compressed EAC/ETC2 images in other non-compressed formats that can render. When GetCompressed* functions are called, the pixels are returned in the non-compressed format th

Re: [Mesa-dev] [PATCH 0/4] improve buffer cache and reuse

Hi, On 02.05.2018 21:19, James Xiong wrote: On Wed, 2 May 2018 14:18:21 +0300 Eero Tamminen wrote: [...] You're missing information on: * On which plaform you did the testing (affects variance) * how many test rounds you ran, and * what is your variance > I ran these tests o

Re: [Mesa-dev] [PATCH v3] i965: Fix ETC2/EAC GetCompressed* functions on Gen7 GPUs

Hi, On 04.05.2018 14:11, Eleni Maria Stea wrote: Thanks for your feedback, On Thu, 3 May 2018 13:30:38 +0300 Eero Tamminen wrote: On 02.05.2018 20:19, Matt Turner wrote: On Wed, May 2, 2018 at 9:13 AM, Eleni Maria Stea wrote: Gen 7 GPUs store the compressed EAC/ETC2 images in other non

Re: [Mesa-dev] [PATCH v3] i965: Fix ETC2/EAC GetCompressed* functions on Gen7 GPUs

Hi, On 04.05.2018 22:06, Eleni Maria Stea wrote: On Fri, 4 May 2018 18:29:55 +0300 Eero Tamminen wrote: You mean returning CAVEAT_SUPPORT in params for compressed formats which are transparently converted to uncompressed data? Well, that would be the best solution I think, if it's pos

Re: [Mesa-dev] [PATCH 0/4] improve buffer cache and reuse

Hi, On 05.05.2018 03:56, James Xiong wrote: From: "Xiong, James" With the current implementation, brw_bufmgr may round up a request size to the next bucket size, result in 25% more memory allocated in the worst senario. For example: Request sizeActual size 32KB+1Byte 40KB . 8MB+1Byte

Re: [Mesa-dev] gcc bug / crash in ast_type_qualifier::validate_in_qualifier()?

Hi, On 08.05.2018 06:45, Matt Turner wrote: On Mon, May 7, 2018 at 8:02 PM, Brian Paul wrote: I don't know when this started happening (I'll try bisecting tomorrow) but we're seeing a crash in ast_type_qualifier::validate_in_qualifier() in -O3 builds with gcc 5.4.0 on Ubuntu 16.04. Specifica

Re: [Mesa-dev] gcc bug / crash in ast_type_qualifier::validate_in_qualifier()?

Hi, On 08.05.2018 20:33, Francisco Jerez wrote: Kenneth Graunke writes: On Tuesday, May 8, 2018 1:23:32 AM PDT Eero Tamminen wrote: On 08.05.2018 06:45, Matt Turner wrote: On Mon, May 7, 2018 at 8:02 PM, Brian Paul wrote: This probably prevents use of xmm instructions, but I haven&#

Re: [Mesa-dev] [PATCH 09/22] intel/compiler: implement 16-bit multiply-add

Hi, On 17.05.2018 11:46, Iago Toral Quiroga wrote: The PRM for MAD states that F, DF and HF are supported, however, then it requires that the instruction includes a 2-bit mask specifying the types of each operand like this: > 00: 32-bit float 01: 32-bit signed integer 10: 32-bit unsigned integ

Re: [Mesa-dev] [PATCH 09/22] intel/compiler: implement 16-bit multiply-add

Hi, On 21.05.2018 10:42, Iago Toral wrote: On Fri, 2018-05-18 at 12:08 +0300, Eero Tamminen wrote: On 17.05.2018 14:25, Eero Tamminen wrote: On 17.05.2018 11:46, Iago Toral Quiroga wrote: The PRM for MAD states that F, DF and HF are supported, however, then it requires that the instruction

Re: [Mesa-dev] Mesa documentation (was: Gitlab migration)

Hi, On 23.05.2018 22:34, Jason Ekstrand wrote: [...]> One of the motivations for doing this now is that there has been some desire to move the mesa website away from raw HTML and over to a platform such as sphinx.  If we're going to do that, we need a system for building the website whenever s

Re: [Mesa-dev] [PATCH 0/2] Implement INTEL_fragment_shader_ordering

Hi, On 28.08.2018 19:15, Rogovin, Kevin wrote: On the questions of tests: If people want, I can adapt the test in piglit for ARB_fragment_shader_interlock to this INTEL one. In general, I have an app/library that uses the extension and testing of that does definitely work on i965 (which sho

Re: [Mesa-dev] [PATCH 3/3] i965/screen: Allow modifiers on sRGB formats

r comment) fixes both SynMark & GfxBench regressions. Tested-By: Eero Tamminen - Eero Rejecting such formats caused us to not allow CCS in some cases where we should have been allowing it. There's some question of whether or not we really should be using SRGB "fou

Re: [Mesa-dev] [PATCH] Bounding box avx2 intrinsic algorithm for openGL/GLES

Hi, On 31.08.2018 08:25, kedar.j.kara...@intel.com wrote: From: "J Karanje, Kedar" The feature is enabled by default during make however we need to add the following to drirc to enable the feature at runtime. vbo: Main algorithm & code to check for MVP & vertex position location Build Files:

Re: [Mesa-dev] [PATCH] glsl: fixer lexer for unreachable defines

Hi, Seems a lot of replicated code to deal with regression from few line line patch, but at least it fixed the failing GfxBench ALU2 test. :-) Tested-By: Eero Tamminen - Eero On 01.09.2018 16:57, Timothy Arceri wrote: If we have something like: #ifdef NOT_DEFINED #define

Re: [Mesa-dev] [PATCH 3/4] drirc: Initial blacklist for adaptive sync

Hi, On 11.09.2018 19:24, Nicholas Kazlauskas wrote: Applications that don't present at a predictable rate (ie. not games) shouldn't have adapative sync enabled. This list covers some of the common desktop compositors and some web browsers. Why videos (in video players like vlc & mpv) wouldn't

Re: [Mesa-dev] [PATCH] util/u_queue: don't inherit thread affinity from parent thread

Hi, On 10/3/18 2:00 AM, Marek Olšák wrote: On Tue, Oct 2, 2018 at 6:36 PM Rob Clark wrote: On Tue, Oct 2, 2018 at 6:30 PM Marek Olšák wrote: From: Marek Olšák [...] Just curious (and maybe I missed some previous discussion), would this override taskset? Asking because when benchmarking o

Re: [Mesa-dev] [PATCH] anv: Use separate MOCS settings for external BOs

s and/or display and those which are entirely internal. That, together with an anv_mocs_for_bo helper lets us choose the right MOCS settings for each BO use. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99507 Cc: mesa-sta...@lists.freedesktop.org Tested-By: Eero Tamminen This fixes the

Re: [Mesa-dev] [PATCH] meson: Bump version to 0.46 for python module

Hi, On 21.11.2018 20.13, Dylan Baker wrote: From: Arfrever Frehtes Taifersar Arahesis Meson has two modules for finding python, the python3 module and the python module. Python3 is older, and has some corner cases, python is newer, has no known corner cases and can detect python2. Things have

Re: [Mesa-dev] Lets talk about autotools

Hi, On 27.11.2018 19.05, Matt Turner wrote: On Tue, Nov 27, 2018 at 1:13 AM Timo Aaltonen wrote: On 17.11.2018 6.04, Dylan Baker wrote: Quoting Dylan Baker (2018-09-17 09:44:07) I feel like for !windows meson is in good enough shape at this point that we can start having the discussion about

Re: [Mesa-dev] [PATCH 5/6] nir: Fix holes in nir_instr

Hi, On 28.11.2018 2.38, Ian Romanick wrote: From: Ian Romanick Found using pahole. While Pahole does static analysis for binaries, Valgrind has tool that can do also _run-time_ analysis of structures and other heap allocations utilization: http://valgrind.org/docs/manual/dh-manual.ht

Re: [Mesa-dev] [PATCH 4/4] nir: detect more induction variables

Hi, On 28.11.2018 5.25, Timothy Arceri wrote: This adds allows loop analysis to detect inductions varibales that are incremented in both branches of an if rather than in a main loop block. For example: [...]> Unfortunatly GCM could move the addition out of the if for us (making this patch unre

Re: [Mesa-dev] Let's talk about -DDEBUG

Hi, On 13.12.2018 12.19, Eric Engestrom wrote: On Wednesday, 2018-12-12 19:45:06 -0500, Marek Olšák wrote: On Wed, Dec 12, 2018 at 7:35 PM Rob Clark wrote: On Wed, Dec 12, 2018 at 7:14 PM Dylan Baker wrote: Quoting Rob Clark (2018-12-12 15:52:47) ... I guess I should have covered that: a

Re: [Mesa-dev] Suggestions for improving meson-based builds

Hi, On 17.12.2018 2.33, Jason Ekstrand wrote: On Sun, Dec 16, 2018 at 1:49 PM Ilia Mirkin > wrote: [...]> 2. Summary at the end. This is not strictly an autotools-specific issue, but rather how mesa has used it. That summary at the end is very useful to

Re: [Mesa-dev] last call for autotools

Hi, On 17.12.2018 8.08, Marek Olšák wrote: [...] I think one of the serious usability issues is that environment variables such as CFLAGS, CXXFLAGS, LDFLAGS, and PKG_CONFIG_PATH are not saved by meson for future reconfigures. I don't know what Meson is supposed to do, but to me that would be

Re: [Mesa-dev] [PATCH] autotools: Deprecate the use of autotools

Hi, On 20.12.2018 4.40, Stuart Young wrote: Could this be reduced this from an error to a warning, with the command-line option suppressing the warning? Perhaps as well as producing the warning, the build could sleep for say 30 seconds after producing the warning message. This would be notice

Re: [Mesa-dev] NIR documentation (was: NIR constant problem for GPU which doesn't have native integer support)

Hi, On 7.1.2019 12.42, Erik Faye-Lund wrote: On Fri, 2019-01-04 at 09:40 -0600, Jason Ekstrand wrote: [...] Yeah... Patches welcome? There have been many attempts by Connor and myself to better document NIR. They all end up in /dev/null due to EBIGGERFIRES. :-( That said, if you ever want

Re: [Mesa-dev] Chromium - Application-level nouveau blacklist

Hi, On 8.1.2019 8.56, Stéphane Marchesin wrote: Yes I think the Chrome-side is very simple here: because there isn't time or means for in-depth investigation, if a driver crashes too much, it gets blacklisted. The situation is not unique, the GPU blacklist file is 1700 lines: https://chromium.go

Re: [Mesa-dev] [PATCH 4/4] i965: Implement dual-patch tessellation evaluation shaders.

Hi, Could you also give dual & single shaders different names when dumping them, e.g. something like the attached patch? - Eero On 23.02.2018 10:36, Kenneth Graunke wrote: Normally, SIMD8 tessellation evaluation shaders operate on a single patch, with each channel operating on a diffe

Re: [Mesa-dev] [PATCH 08/15] glsl: Switch ast_type_qualifier to a 128-bit bitset.

ovide overloads for signed integer types, but it should be harmless since the two candidates have the same semantics, and should go away with a C++11-capable compiler. I think the attached patch should shut the warnings on older compilers. Yes, it fixed the compilation with gcc 5.4, and INTEL_DEBU

Re: [Mesa-dev] [PATCH 5/6] intel/fs: Handle surface opcode sample masks via predication.

GT2. CSDof improved by: * 9% on BYT * 7-8% on BXT J4205, and on SKL GT2 desktop (Variance on HSW was too large in this test, to conclude anything.) Changes in all the other tests were within daily variance. Tested-By: Eero Tamminen - Eero --- src/intel/compiler/brw

<    1   2   3   >