Re: [Mesa-dev] [PATCH v2 1/3] st/glsl_to_tgsi: move nir detection earlier - bisected
On 8 February 2018 at 04:30, Timothy Arceriwrote: > > > On 07/02/18 19:17, Dieter Nützel wrote: >> >> Am 06.02.2018 00:23, schrieb Timothy Arceri: >>> >>> On 05/02/18 15:04, Dieter Nützel wrote: Am 02.02.2018 10:24, schrieb Timothy Arceri: > > On 02/02/18 19:26, Dieter Nützel wrote: >> >> Hello Tim, >> >> _this_ version brake UH, UV, mpv, blender 2.79 (some test files not >> all). >> Must be something with the cache file(s). > > > The cache currently needs to be deleted when switching between nir and > tgsi. I'm not sure it I should try to avoid this or not ... I guess it > will probably save some bug reports so I'll try send a follow up > patch. Hi Tim, it is NOT your fault. I tracked it down to Marek's commit commit be973ed21f6e456ebd753f26a99151d9ea6e765c >>> >>> >>> This should fix things for now: >>> >>> https://patchwork.freedesktop.org/patch/202759/ >> >> >> Apart that it landed already: >> >> Tested-by: Dieter Nützel >> >> But I get some severe hangs with current git code on Polaris 20. >> Steam (Linux only, NOT Wine), UH and UV hang full system. >> I could remotely log in but nothing in the logs. >> Have to bisect, again... > > > I'm seeing deadlocks in piglit caused by the following patch, could be what > you are seeing. > > commit 6a651b6b77b68db71a027c826abccc843ace88ef (HEAD) > Author: Tapani Pälli > Date: Mon Jan 22 11:55:06 2018 +0200 > > disk cache: initialize cache path and index only when used > > This patch makes disk_cache initialize path and index lazily so > that we can utilize disk_cache without a path using callback > functionality introduced by next patch. > > v2: unmap mmap and destroy queue only if index_mmap exists > Just a fly-by comment. While the issue may be elsewhere, an orthogonal approach is to allow selecting the old/new behaviour. Say by adding a "bool delayed/immediate" to disk_cache_create()? Off the top of my head - the old behaviour might be even good for the Vulkan drivers. You don't want app. developers having a wtf moment as they see the odd stall where there it shouldn't be ;-) -Emil ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH v2 1/3] st/glsl_to_tgsi: move nir detection earlier - bisected
On 09/02/18 11:53, Marek Olšák wrote: On Fri, Feb 9, 2018 at 1:45 AM, Timothy Arceriwrote: On 09/02/18 10:49, Marek Olšák wrote: Does this fix the cache/no cache conflict? diff --git a/src/gallium/drivers/radeonsi/si_pipe.c b/src/gallium/drivers/radeonsi/si_pipe.c index 26835d6..97f11ea 100644 --- a/src/gallium/drivers/radeonsi/si_pipe.c +++ b/src/gallium/drivers/radeonsi/si_pipe.c @@ -642,7 +642,8 @@ static void si_disk_cache_create(struct si_screen *sscreen) sscreen->debug_flags & (DBG(FS_CORRECT_DERIVS_AFTER_KILL) | DBG(SI_SCHED) | -DBG(UNSAFE_MATH)); +DBG(UNSAFE_MATH) | +DBG(NIR)); sscreen->disk_shader_cache = disk_cache_create(si_get_family_name(sscreen), Marek That should do it thanks. Feel free to add my r-b and push. Well, I'm not sure. Would you explain to me why this fixes the hangs? If it's a conflict between TGSI and NIR caching in st/mesa, the driver shouldn't contain workarounds for it. Even when NIR is enabled, TGSI shaders can still come from non-GLSL sources. I don't think this fixes any hangs just crashes e.g. when you run a shader on the tgsi backend then try to run the same shader on the nir backend. In this case the nir cache will try to load the tgsi cache from disk as it produces the same key without this flag. Thanks, Marek ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH v2 1/3] st/glsl_to_tgsi: move nir detection earlier - bisected
On Fri, Feb 9, 2018 at 1:45 AM, Timothy Arceriwrote: > On 09/02/18 10:49, Marek Olšák wrote: >> >> Does this fix the cache/no cache conflict? >> >> diff --git a/src/gallium/drivers/radeonsi/si_pipe.c >> b/src/gallium/drivers/radeonsi/si_pipe.c >> index 26835d6..97f11ea 100644 >> --- a/src/gallium/drivers/radeonsi/si_pipe.c >> +++ b/src/gallium/drivers/radeonsi/si_pipe.c >> @@ -642,7 +642,8 @@ static void si_disk_cache_create(struct si_screen >> *sscreen) >> sscreen->debug_flags & >> (DBG(FS_CORRECT_DERIVS_AFTER_KILL) | >> DBG(SI_SCHED) | >> -DBG(UNSAFE_MATH)); >> +DBG(UNSAFE_MATH) | >> +DBG(NIR)); >> >> sscreen->disk_shader_cache = >> >> disk_cache_create(si_get_family_name(sscreen), >> >> >> Marek > > > That should do it thanks. Feel free to add my r-b and push. Well, I'm not sure. Would you explain to me why this fixes the hangs? If it's a conflict between TGSI and NIR caching in st/mesa, the driver shouldn't contain workarounds for it. Even when NIR is enabled, TGSI shaders can still come from non-GLSL sources. Thanks, Marek ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH v2 1/3] st/glsl_to_tgsi: move nir detection earlier - bisected
On 09/02/18 10:49, Marek Olšák wrote: Does this fix the cache/no cache conflict? diff --git a/src/gallium/drivers/radeonsi/si_pipe.c b/src/gallium/drivers/radeonsi/si_pipe.c index 26835d6..97f11ea 100644 --- a/src/gallium/drivers/radeonsi/si_pipe.c +++ b/src/gallium/drivers/radeonsi/si_pipe.c @@ -642,7 +642,8 @@ static void si_disk_cache_create(struct si_screen *sscreen) sscreen->debug_flags & (DBG(FS_CORRECT_DERIVS_AFTER_KILL) | DBG(SI_SCHED) | -DBG(UNSAFE_MATH)); +DBG(UNSAFE_MATH) | +DBG(NIR)); sscreen->disk_shader_cache = disk_cache_create(si_get_family_name(sscreen), Marek That should do it thanks. Feel free to add my r-b and push. Reviewed-by: Timothy ArceriOn Thu, Feb 8, 2018 at 11:51 PM, Dieter Nützel wrote: Am 08.02.2018 05:30, schrieb Timothy Arceri: On 07/02/18 19:17, Dieter Nützel wrote: Am 06.02.2018 00:23, schrieb Timothy Arceri: On 05/02/18 15:04, Dieter Nützel wrote: Am 02.02.2018 10:24, schrieb Timothy Arceri: On 02/02/18 19:26, Dieter Nützel wrote: Hello Tim, _this_ version brake UH, UV, mpv, blender 2.79 (some test files not all). Must be something with the cache file(s). The cache currently needs to be deleted when switching between nir and tgsi. I'm not sure it I should try to avoid this or not ... I guess it will probably save some bug reports so I'll try send a follow up patch. Hi Tim, it is NOT your fault. I tracked it down to Marek's commit commit be973ed21f6e456ebd753f26a99151d9ea6e765c This should fix things for now: https://patchwork.freedesktop.org/patch/202759/ Apart that it landed already: Tested-by: Dieter Nützel But I get some severe hangs with current git code on Polaris 20. Steam (Linux only, NOT Wine), UH and UV hang full system. I could remotely log in but nothing in the logs. Have to bisect, again... I'm seeing deadlocks in piglit caused by the following patch, could be what you are seeing. commit 6a651b6b77b68db71a027c826abccc843ace88ef (HEAD) Author: Tapani Pälli Date: Mon Jan 22 11:55:06 2018 +0200 disk cache: initialize cache path and index only when used This patch makes disk_cache initialize path and index lazily so that we can utilize disk_cache without a path using callback functionality introduced by next patch. v2: unmap mmap and destroy queue only if index_mmap exists Signed-off-by: Tapani Pälli Reviewed-by: Jordan Justen Reviewed-by: Emil Velikov Ah, thanks. Will verify this, too. But now, my only remaining NIR sigfaults are your 'announced' (;-)) NIR/without NIR cache fighting. Even _with_ Tapani's commit. [ 504.814523] si_shader:2[10216]: segfault at 7f1959a377cc ip 839230e0 sp 97aa8f3e error 4 in libc-2.26.so[7f19865b8000+1b1000] My above reported deadlocks (hangs) was (must be) a LLVM git (7.0.0) bug. With LLVM git from yesterday all are gone. Steam Linux, DiRT Rally and F1 2017 and UH/UV all solved. The later 'faster than ever'. Xeon X3470, 4/8 c/t, 2,93 GHz, 24 GB, Polaris 20 (RX 580), NIR (!): HD (1920x1080) UH: ~91 fps (!!!) UV: ~86 fps GREAT work! Greetings, Dieter ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH v2 1/3] st/glsl_to_tgsi: move nir detection earlier - bisected
Marek, great work. This one solve it. I've tested both ways: NO_NIR -> NIR : OK NIR -> NO_NIR : OK (deleted cache files before both tests) Tested-by: Dieter NützelLand this stuff...;-) Dieter Am 09.02.2018 00:49, schrieb Marek Olšák: Does this fix the cache/no cache conflict? diff --git a/src/gallium/drivers/radeonsi/si_pipe.c b/src/gallium/drivers/radeonsi/si_pipe.c index 26835d6..97f11ea 100644 --- a/src/gallium/drivers/radeonsi/si_pipe.c +++ b/src/gallium/drivers/radeonsi/si_pipe.c @@ -642,7 +642,8 @@ static void si_disk_cache_create(struct si_screen *sscreen) sscreen->debug_flags & (DBG(FS_CORRECT_DERIVS_AFTER_KILL) | DBG(SI_SCHED) | -DBG(UNSAFE_MATH)); +DBG(UNSAFE_MATH) | +DBG(NIR)); sscreen->disk_shader_cache = disk_cache_create(si_get_family_name(sscreen), Marek On Thu, Feb 8, 2018 at 11:51 PM, Dieter Nützel wrote: Am 08.02.2018 05:30, schrieb Timothy Arceri: On 07/02/18 19:17, Dieter Nützel wrote: Am 06.02.2018 00:23, schrieb Timothy Arceri: On 05/02/18 15:04, Dieter Nützel wrote: Am 02.02.2018 10:24, schrieb Timothy Arceri: On 02/02/18 19:26, Dieter Nützel wrote: Hello Tim, _this_ version brake UH, UV, mpv, blender 2.79 (some test files not all). Must be something with the cache file(s). The cache currently needs to be deleted when switching between nir and tgsi. I'm not sure it I should try to avoid this or not ... I guess it will probably save some bug reports so I'll try send a follow up patch. Hi Tim, it is NOT your fault. I tracked it down to Marek's commit commit be973ed21f6e456ebd753f26a99151d9ea6e765c This should fix things for now: https://patchwork.freedesktop.org/patch/202759/ Apart that it landed already: Tested-by: Dieter Nützel But I get some severe hangs with current git code on Polaris 20. Steam (Linux only, NOT Wine), UH and UV hang full system. I could remotely log in but nothing in the logs. Have to bisect, again... I'm seeing deadlocks in piglit caused by the following patch, could be what you are seeing. commit 6a651b6b77b68db71a027c826abccc843ace88ef (HEAD) Author: Tapani Pälli Date: Mon Jan 22 11:55:06 2018 +0200 disk cache: initialize cache path and index only when used This patch makes disk_cache initialize path and index lazily so that we can utilize disk_cache without a path using callback functionality introduced by next patch. v2: unmap mmap and destroy queue only if index_mmap exists Signed-off-by: Tapani Pälli Reviewed-by: Jordan Justen Reviewed-by: Emil Velikov Ah, thanks. Will verify this, too. But now, my only remaining NIR sigfaults are your 'announced' (;-)) NIR/without NIR cache fighting. Even _with_ Tapani's commit. [ 504.814523] si_shader:2[10216]: segfault at 7f1959a377cc ip 839230e0 sp 97aa8f3e error 4 in libc-2.26.so[7f19865b8000+1b1000] My above reported deadlocks (hangs) was (must be) a LLVM git (7.0.0) bug. With LLVM git from yesterday all are gone. Steam Linux, DiRT Rally and F1 2017 and UH/UV all solved. The later 'faster than ever'. Xeon X3470, 4/8 c/t, 2,93 GHz, 24 GB, Polaris 20 (RX 580), NIR (!): HD (1920x1080) UH: ~91 fps (!!!) UV: ~86 fps GREAT work! Greetings, Dieter ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH v2 1/3] st/glsl_to_tgsi: move nir detection earlier - bisected
Does this fix the cache/no cache conflict? diff --git a/src/gallium/drivers/radeonsi/si_pipe.c b/src/gallium/drivers/radeonsi/si_pipe.c index 26835d6..97f11ea 100644 --- a/src/gallium/drivers/radeonsi/si_pipe.c +++ b/src/gallium/drivers/radeonsi/si_pipe.c @@ -642,7 +642,8 @@ static void si_disk_cache_create(struct si_screen *sscreen) sscreen->debug_flags & (DBG(FS_CORRECT_DERIVS_AFTER_KILL) | DBG(SI_SCHED) | -DBG(UNSAFE_MATH)); +DBG(UNSAFE_MATH) | +DBG(NIR)); sscreen->disk_shader_cache = disk_cache_create(si_get_family_name(sscreen), Marek On Thu, Feb 8, 2018 at 11:51 PM, Dieter Nützelwrote: > Am 08.02.2018 05:30, schrieb Timothy Arceri: >> >> On 07/02/18 19:17, Dieter Nützel wrote: >>> >>> Am 06.02.2018 00:23, schrieb Timothy Arceri: On 05/02/18 15:04, Dieter Nützel wrote: > > Am 02.02.2018 10:24, schrieb Timothy Arceri: >> >> On 02/02/18 19:26, Dieter Nützel wrote: >>> >>> Hello Tim, >>> >>> _this_ version brake UH, UV, mpv, blender 2.79 (some test files not >>> all). >>> Must be something with the cache file(s). >> >> >> The cache currently needs to be deleted when switching between nir and >> tgsi. I'm not sure it I should try to avoid this or not ... I guess it >> will probably save some bug reports so I'll try send a follow up >> patch. > > > Hi Tim, > > it is NOT your fault. > I tracked it down to Marek's commit commit > be973ed21f6e456ebd753f26a99151d9ea6e765c This should fix things for now: https://patchwork.freedesktop.org/patch/202759/ >>> >>> >>> Apart that it landed already: >>> >>> Tested-by: Dieter Nützel >>> >>> But I get some severe hangs with current git code on Polaris 20. >>> Steam (Linux only, NOT Wine), UH and UV hang full system. >>> I could remotely log in but nothing in the logs. >>> Have to bisect, again... >> >> >> I'm seeing deadlocks in piglit caused by the following patch, could be >> what you are seeing. >> >> commit 6a651b6b77b68db71a027c826abccc843ace88ef (HEAD) >> Author: Tapani Pälli >> Date: Mon Jan 22 11:55:06 2018 +0200 >> >> disk cache: initialize cache path and index only when used >> >> This patch makes disk_cache initialize path and index lazily so >> that we can utilize disk_cache without a path using callback >> functionality introduced by next patch. >> >> v2: unmap mmap and destroy queue only if index_mmap exists >> >> Signed-off-by: Tapani Pälli >> Reviewed-by: Jordan Justen >> Reviewed-by: Emil Velikov > > > Ah, thanks. Will verify this, too. > But now, my only remaining NIR sigfaults are your 'announced' (;-)) > NIR/without NIR cache fighting. Even _with_ Tapani's commit. > > [ 504.814523] si_shader:2[10216]: segfault at 7f1959a377cc ip > 839230e0 sp 97aa8f3e error 4 in > libc-2.26.so[7f19865b8000+1b1000] > > My above reported deadlocks (hangs) was (must be) a LLVM git (7.0.0) bug. > With LLVM git from yesterday all are gone. > > Steam Linux, DiRT Rally and F1 2017 and > UH/UV all solved. The later 'faster than ever'. > > Xeon X3470, 4/8 c/t, 2,93 GHz, 24 GB, Polaris 20 (RX 580), NIR (!): > HD (1920x1080) > UH: ~91 fps (!!!) > UV: ~86 fps > > GREAT work! > > Greetings, > Dieter ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH v2 1/3] st/glsl_to_tgsi: move nir detection earlier - bisected
Am 08.02.2018 05:30, schrieb Timothy Arceri: On 07/02/18 19:17, Dieter Nützel wrote: Am 06.02.2018 00:23, schrieb Timothy Arceri: On 05/02/18 15:04, Dieter Nützel wrote: Am 02.02.2018 10:24, schrieb Timothy Arceri: On 02/02/18 19:26, Dieter Nützel wrote: Hello Tim, _this_ version brake UH, UV, mpv, blender 2.79 (some test files not all). Must be something with the cache file(s). The cache currently needs to be deleted when switching between nir and tgsi. I'm not sure it I should try to avoid this or not ... I guess it will probably save some bug reports so I'll try send a follow up patch. Hi Tim, it is NOT your fault. I tracked it down to Marek's commit commit be973ed21f6e456ebd753f26a99151d9ea6e765c This should fix things for now: https://patchwork.freedesktop.org/patch/202759/ Apart that it landed already: Tested-by: Dieter NützelBut I get some severe hangs with current git code on Polaris 20. Steam (Linux only, NOT Wine), UH and UV hang full system. I could remotely log in but nothing in the logs. Have to bisect, again... I'm seeing deadlocks in piglit caused by the following patch, could be what you are seeing. commit 6a651b6b77b68db71a027c826abccc843ace88ef (HEAD) Author: Tapani Pälli Date: Mon Jan 22 11:55:06 2018 +0200 disk cache: initialize cache path and index only when used This patch makes disk_cache initialize path and index lazily so that we can utilize disk_cache without a path using callback functionality introduced by next patch. v2: unmap mmap and destroy queue only if index_mmap exists Signed-off-by: Tapani Pälli Reviewed-by: Jordan Justen Reviewed-by: Emil Velikov Ah, thanks. Will verify this, too. But now, my only remaining NIR sigfaults are your 'announced' (;-)) NIR/without NIR cache fighting. Even _with_ Tapani's commit. [ 504.814523] si_shader:2[10216]: segfault at 7f1959a377cc ip 839230e0 sp 97aa8f3e error 4 in libc-2.26.so[7f19865b8000+1b1000] My above reported deadlocks (hangs) was (must be) a LLVM git (7.0.0) bug. With LLVM git from yesterday all are gone. Steam Linux, DiRT Rally and F1 2017 and UH/UV all solved. The later 'faster than ever'. Xeon X3470, 4/8 c/t, 2,93 GHz, 24 GB, Polaris 20 (RX 580), NIR (!): HD (1920x1080) UH: ~91 fps (!!!) UV: ~86 fps GREAT work! Greetings, Dieter ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH v2 1/3] st/glsl_to_tgsi: move nir detection earlier - bisected
On 07/02/18 19:17, Dieter Nützel wrote: Am 06.02.2018 00:23, schrieb Timothy Arceri: On 05/02/18 15:04, Dieter Nützel wrote: Am 02.02.2018 10:24, schrieb Timothy Arceri: On 02/02/18 19:26, Dieter Nützel wrote: Hello Tim, _this_ version brake UH, UV, mpv, blender 2.79 (some test files not all). Must be something with the cache file(s). The cache currently needs to be deleted when switching between nir and tgsi. I'm not sure it I should try to avoid this or not ... I guess it will probably save some bug reports so I'll try send a follow up patch. Hi Tim, it is NOT your fault. I tracked it down to Marek's commit commit be973ed21f6e456ebd753f26a99151d9ea6e765c This should fix things for now: https://patchwork.freedesktop.org/patch/202759/ Apart that it landed already: Tested-by: Dieter NützelBut I get some severe hangs with current git code on Polaris 20. Steam (Linux only, NOT Wine), UH and UV hang full system. I could remotely log in but nothing in the logs. Have to bisect, again... I'm seeing deadlocks in piglit caused by the following patch, could be what you are seeing. commit 6a651b6b77b68db71a027c826abccc843ace88ef (HEAD) Author: Tapani Pälli Date: Mon Jan 22 11:55:06 2018 +0200 disk cache: initialize cache path and index only when used This patch makes disk_cache initialize path and index lazily so that we can utilize disk_cache without a path using callback functionality introduced by next patch. v2: unmap mmap and destroy queue only if index_mmap exists Signed-off-by: Tapani Pälli Reviewed-by: Jordan Justen Reviewed-by: Emil Velikov Sadly, I've a bad cold. Thanks, Dieter ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH v2 1/3] st/glsl_to_tgsi: move nir detection earlier - bisected
Am 06.02.2018 00:23, schrieb Timothy Arceri: On 05/02/18 15:04, Dieter Nützel wrote: Am 02.02.2018 10:24, schrieb Timothy Arceri: On 02/02/18 19:26, Dieter Nützel wrote: Hello Tim, _this_ version brake UH, UV, mpv, blender 2.79 (some test files not all). Must be something with the cache file(s). The cache currently needs to be deleted when switching between nir and tgsi. I'm not sure it I should try to avoid this or not ... I guess it will probably save some bug reports so I'll try send a follow up patch. Hi Tim, it is NOT your fault. I tracked it down to Marek's commit commit be973ed21f6e456ebd753f26a99151d9ea6e765c This should fix things for now: https://patchwork.freedesktop.org/patch/202759/ Apart that it landed already: Tested-by: Dieter NützelBut I get some severe hangs with current git code on Polaris 20. Steam (Linux only, NOT Wine), UH and UV hang full system. I could remotely log in but nothing in the logs. Have to bisect, again... Sadly, I've a bad cold. Thanks, Dieter ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH v2 1/3] st/glsl_to_tgsi: move nir detection earlier - bisected
On 05/02/18 15:04, Dieter Nützel wrote: Am 02.02.2018 10:24, schrieb Timothy Arceri: On 02/02/18 19:26, Dieter Nützel wrote: Hello Tim, _this_ version brake UH, UV, mpv, blender 2.79 (some test files not all). Must be something with the cache file(s). The cache currently needs to be deleted when switching between nir and tgsi. I'm not sure it I should try to avoid this or not ... I guess it will probably save some bug reports so I'll try send a follow up patch. Hi Tim, it is NOT your fault. I tracked it down to Marek's commit commit be973ed21f6e456ebd753f26a99151d9ea6e765c This should fix things for now: https://patchwork.freedesktop.org/patch/202759/ ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH v2 1/3] st/glsl_to_tgsi: move nir detection earlier - bisected
Am 02.02.2018 10:24, schrieb Timothy Arceri: On 02/02/18 19:26, Dieter Nützel wrote: Hello Tim, _this_ version brake UH, UV, mpv, blender 2.79 (some test files not all). Must be something with the cache file(s). The cache currently needs to be deleted when switching between nir and tgsi. I'm not sure it I should try to avoid this or not ... I guess it will probably save some bug reports so I'll try send a follow up patch. Hi Tim, it is NOT your fault. I tracked it down to Marek's commit commit be973ed21f6e456ebd753f26a99151d9ea6e765c /opt/mesa> git bisect bad be973ed21f6e456ebd753f26a99151d9ea6e765c is the first bad commit commit be973ed21f6e456ebd753f26a99151d9ea6e765c Author: Marek OlšákDate: Tue Jan 30 18:34:25 2018 +0100 radeonsi: load the right number of components for VS inputs and TBOs The supported counts are 1, 2, 4. (3=4) The following snippet loads float, vec2, vec3, and vec4: Before: buffer_load_format_x v9, v4, s[0:3], 0 idxen ; E0002000 8904 buffer_load_format_xyzw v[0:3], v5, s[8:11], 0 idxen ; E00C2000 80020005 s_waitcnt vmcnt(0); BF8C0F70 buffer_load_format_xyzw v[2:5], v6, s[12:15], 0 idxen ; E00C2000 80030206 s_waitcnt vmcnt(0); BF8C0F70 buffer_load_format_xyzw v[5:8], v7, s[4:7], 0 idxen ; E00C2000 80010507 After: buffer_load_format_x v10, v4, s[0:3], 0 idxen ; E0002000 8A04 buffer_load_format_xy v[8:9], v5, s[8:11], 0 idxen; E0042000 80020805 buffer_load_format_xyzw v[0:3], v6, s[12:15], 0 idxen ; E00C2000 80030006 s_waitcnt vmcnt(0); BF8C0F70 buffer_load_format_xyzw v[3:6], v7, s[4:7], 0 idxen ; E00C2000 80010307 Reviewed-by: Samuel Pitoiset :04 04 262b88d9e9f462b32595d6f15eddc0c6be4b997d cf45e1bd87a8a0e12553f6476d51a750e114ea10 M src But can't revert it clean for the time being. Another week, another night,... Cheers, Dieter ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev