Re: [Mesa-dev] [PATCH] st/va: use provided sizes and coords for vlVaGetImage

On Tue, Oct 9, 2018 at 4:32 PM wrote: > > From: Boyuan Zhang > > vlVaGetImage should respect the width, height, and coordinates x and y that > passed in. Therefore, pipe_box should be created with the passed in values > instead of surface width/height. > > v2: add input size check, return error

[Mesa-dev] [Bug 108310] Failure to build with LLVM7

https://bugs.freedesktop.org/show_bug.cgi?id=108310 --- Comment #1 from Thiago Macieira --- Patch used by Clear Linux found in https://bugs.freedesktop.org/attachment.cgi?id=141969 -- You are receiving this mail because: You are the QA Contact for the bug. You are the assignee for the

[Mesa-dev] [Bug 108308] Segfault when using LLVM7 -

https://bugs.freedesktop.org/show_bug.cgi?id=108308 --- Comment #5 from Thiago Macieira --- (In reply to Alok Hota from comment #2) > Hi Thiago, > > Thanks for the details and patch on this. Can you describe what compile > error you're receiving with LLVM 7? And what configure flags you're

[Mesa-dev] [Bug 108310] Failure to build with LLVM7

https://bugs.freedesktop.org/show_bug.cgi?id=108310 Bug ID: 108310 Summary: Failure to build with LLVM7 Product: Mesa Version: git Hardware: Other OS: All Status: NEW Severity: normal Priority:

[Mesa-dev] [Bug 108308] Segfault when using LLVM7 -

https://bugs.freedesktop.org/show_bug.cgi?id=108308 --- Comment #4 from Thiago Macieira --- Created attachment 141969 --> https://bugs.freedesktop.org/attachment.cgi?id=141969=edit Actual patch to make compile with LLVM 7 -- You are receiving this mail because: You are the QA Contact for the

[Mesa-dev] [Bug 108308] Segfault when using LLVM7 -

https://bugs.freedesktop.org/show_bug.cgi?id=108308 --- Comment #3 from Thiago Macieira --- (In reply to Alok Hota from comment #2) > Hi Thiago, > > Thanks for the details and patch on this. Can you describe what compile > error you're receiving with LLVM 7? And what configure flags you're

Re: [Mesa-dev] [PATCH] glsl: ignore trailing whitespace when define redefined

It took me a second to grok what was happening in the test case... it looked like both definitions of EIGHT were identical. :) Reviewed-by: Ian Romanick On 10/09/2018 05:03 PM, Timothy Arceri wrote: > The Nvidia/AMD binary drivers allow this, as does GCC. > > This fixes shader compilation

Re: [Mesa-dev] [PATCH v2] intel: Don't propagate conditional modifiers if a UD source is negated

Works for me. Reviewed-by: Ian Romanick On 10/09/2018 05:19 PM, Jason Ekstrand wrote: > This fixes a bug uncovered by my NIR integer division by constant > optimization series. > > Fixes: 19f9cb72c8b "i965/fs: Add pass to propagate conditional..." > Fixes: 627f94b72e0 "i965/vec4: adding

[Mesa-dev] [Bug 108308] Segfault when using LLVM7 -

https://bugs.freedesktop.org/show_bug.cgi?id=108308 --- Comment #2 from Alok Hota --- Hi Thiago, Thanks for the details and patch on this. Can you describe what compile error you're receiving with LLVM 7? And what configure flags you're using? We have an upcoming patch to support LLVM 7, but it

[Mesa-dev] [Bug 108308] Segfault when using LLVM7 -

https://bugs.freedesktop.org/show_bug.cgi?id=108308 --- Comment #1 from Thiago Macieira --- If it helps: (gdb) p *pJitMgr $28 = {mContext = { = {pImpl = 0xcd3af0}, }, mBuilder = { = {CurDbgLocation = {Loc = { Ref = {MD = 0x0}}}, BB = 0x0, InsertPt = { >> = { = {}, }, NodePtr =

[Mesa-dev] [Bug 108308] Segfault when using LLVM7 -

https://bugs.freedesktop.org/show_bug.cgi?id=108308 Bug ID: 108308 Summary: Segfault when using LLVM7 - Product: Mesa Version: git Hardware: Other OS: All Status: NEW Severity: normal Priority:

Re: [Mesa-dev] [PATCH] radv: reduce max{Uniform, Storage}BufferRange values

On Tue, Oct 9, 2018 at 3:40 PM Samuel Pitoiset wrote: > > 2^32-1 is too high. > > This fixes the following crucible tests: > stress.limits.buffer-update.range.uniform > stress.limits.buffer-update.range.storage This is just an allocation test, it passes for me on Vega so make this GPU specific?

[Mesa-dev] [PATCH v2] intel: Don't propagate conditional modifiers if a UD source is negated

This fixes a bug uncovered by my NIR integer division by constant optimization series. Fixes: 19f9cb72c8b "i965/fs: Add pass to propagate conditional..." Fixes: 627f94b72e0 "i965/vec4: adding vec4_cmod_propagation..." Cc: Matt Turner --- src/intel/compiler/brw_fs.cpp| 19 +++

Re: [Mesa-dev] [PATCH] intel: Don't propagate conditional modifiers if a UD source is negated

On Tue, Oct 9, 2018 at 6:55 PM Jason Ekstrand wrote: > On Tue, Oct 9, 2018 at 6:45 PM Jason Ekstrand > wrote: > >> On Tue, Oct 9, 2018 at 6:35 PM Ian Romanick wrote: >> >>> On 10/09/2018 10:03 AM, Jason Ekstrand wrote: >>> > On Tue, Oct 9, 2018 at 11:24 AM Ian Romanick >> >

[Mesa-dev] [PATCH] glsl: ignore trailing whitespace when define redefined

The Nvidia/AMD binary drivers allow this, as does GCC. This fixes shader compilation issues in the latest update of No Mans Sky. Cc: mesa-sta...@lists.freedesktop.org --- src/compiler/glsl/glcpp/glcpp-parse.y | 14 ++ .../glsl/glcpp/tests/122-redefine-whitespace.c |

Re: [Mesa-dev] [PATCH 3/9] nir/opt_peephole_select: Don't peephole_select expensive math instructions

On 10/10/18 10:31 am, Ian Romanick wrote: On 10/09/2018 03:42 PM, Timothy Arceri wrote: Hi Ian, You might be interested in doing something like this [1]. It needs some tidy up (and likely further testing) but my initial results showed a fairly large number of select can be converted to ifs

Re: [Mesa-dev] [PATCH] intel: Don't propagate conditional modifiers if a UD source is negated

On Tue, Oct 9, 2018 at 6:45 PM Jason Ekstrand wrote: > On Tue, Oct 9, 2018 at 6:35 PM Ian Romanick wrote: > >> On 10/09/2018 10:03 AM, Jason Ekstrand wrote: >> > On Tue, Oct 9, 2018 at 11:24 AM Ian Romanick > > > wrote: >> > >> > On 10/09/2018 09:00 AM, Jason

Re: [Mesa-dev] [PATCH] intel: Don't propagate conditional modifiers if a UD source is negated

On Tue, Oct 9, 2018 at 6:35 PM Ian Romanick wrote: > On 10/09/2018 10:03 AM, Jason Ekstrand wrote: > > On Tue, Oct 9, 2018 at 11:24 AM Ian Romanick > > wrote: > > > > On 10/09/2018 09:00 AM, Jason Ekstrand wrote: > > > This fixes a bug uncovered by my NIR

Re: [Mesa-dev] [PATCH] intel: Don't propagate conditional modifiers if a UD source is negated

On 10/09/2018 10:03 AM, Jason Ekstrand wrote: > On Tue, Oct 9, 2018 at 11:24 AM Ian Romanick > wrote: > > On 10/09/2018 09:00 AM, Jason Ekstrand wrote: > > This fixes a bug uncovered by my NIR integer division by constant > > optimization series. > >

Re: [Mesa-dev] [PATCH 3/9] nir/opt_peephole_select: Don't peephole_select expensive math instructions

On 10/09/2018 03:42 PM, Timothy Arceri wrote: > Hi Ian, > > You might be interested in doing something like this [1]. It needs some > tidy up (and likely further testing) but my initial results showed a > fairly large number of select can be converted to ifs just based on > having a ubo load. >

Re: [Mesa-dev] [PATCH] dri: Include kernel release in vendor info

On 10/09/2018 06:24 AM, Chris Wilson wrote: > The userspace driver does not exist in isolation and occasionally > depends on kernel uapi, and so it is useful in bug reports to include > that information. (radeonsi, r600 and radv already include utsname) > > References:

Re: [Mesa-dev] [PATCH 3/9] nir/opt_peephole_select: Don't peephole_select expensive math instructions

Hi Ian, You might be interested in doing something like this [1]. It needs some tidy up (and likely further testing) but my initial results showed a fairly large number of select can be converted to ifs just based on having a ubo load. [1]

[Mesa-dev] [PATCH 1/3 v2] glsl: prevent qualifiers modification of predeclared variables

From: Ian Romanick Section 3.7 (Identifiers) of the GLSL spec says: However, as noted in the specification, there are some cases where previously declared variables can be redeclared to change or add some property, and predeclared "gl_" names are allowed to be redeclared in a

[Mesa-dev] [PATCH 3/3] glsl: Refactor type checking for redeclarations

From: Ian Romanick Signed-off-by: Ian Romanick --- src/compiler/glsl/ast_to_hir.cpp | 19 +++ 1 file changed, 7 insertions(+), 12 deletions(-) diff --git a/src/compiler/glsl/ast_to_hir.cpp b/src/compiler/glsl/ast_to_hir.cpp index 6f6e7efe9aa..6b0d4d874c9 100644 ---

[Mesa-dev] [PATCH 2/3] glsl: Omit redundant qualifier checks on redeclarations

From: Ian Romanick Signed-off-by: Ian Romanick --- src/compiler/glsl/ast_to_hir.cpp | 9 +++-- 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/src/compiler/glsl/ast_to_hir.cpp b/src/compiler/glsl/ast_to_hir.cpp index 2e4c9ef6776..6f6e7efe9aa 100644 ---

Re: [Mesa-dev] [PATCH] glsl: prevent qualifiers modification of predeclared variables

On 10/09/2018 06:52 AM, andrey simiklit wrote: > Hello, > > Please find my comments below: > > On Sat, Oct 6, 2018 at 1:07 AM Ian Romanick > wrote: > > On 10/05/2018 03:02 PM, Ian Romanick wrote: > > On 10/04/2018 07:08 AM, asimiklit.w...@gmail.com >

[Mesa-dev] [Bug 106465] No test for Image Load/Store on format-incompatible texture buffer

https://bugs.freedesktop.org/show_bug.cgi?id=106465 Nanley Chery changed: What|Removed |Added Resolution|--- |FIXED Status|NEW

[Mesa-dev] [Bug 106465] No test for Image Load/Store on format-incompatible texture buffer

https://bugs.freedesktop.org/show_bug.cgi?id=106465 --- Comment #5 from Nanley Chery --- (In reply to Andrés Gómez García from comment #4) > (In reply to Andrés Gómez García from comment #3) > > Curro, is this fixed by? > > [...] > > My bad, I just realized that the report is about a missing

Re: [Mesa-dev] [PATCH 07/12] nir: Allow [iu]mul_high on non-32-bit types

On Tue, Oct 9, 2018 at 3:30 PM Ian Romanick wrote: > On 10/09/2018 10:17 AM, Jason Ekstrand wrote: > > On Tue, Oct 9, 2018 at 11:32 AM Ian Romanick > > wrote: > > > > On 10/08/2018 02:14 PM, Jason Ekstrand wrote: > > > On Mon, Oct 8, 2018 at 3:46 PM Ian

[Mesa-dev] [PATCH] st/va: use provided sizes and coords for vlVaGetImage

From: Boyuan Zhang vlVaGetImage should respect the width, height, and coordinates x and y that passed in. Therefore, pipe_box should be created with the passed in values instead of surface width/height. v2: add input size check, return error when size out of bounds v3: fix the size check for

Re: [Mesa-dev] [PATCH 07/12] nir: Allow [iu]mul_high on non-32-bit types

On 10/09/2018 10:17 AM, Jason Ekstrand wrote: > On Tue, Oct 9, 2018 at 11:32 AM Ian Romanick > wrote: > > On 10/08/2018 02:14 PM, Jason Ekstrand wrote: > > On Mon, Oct 8, 2018 at 3:46 PM Ian Romanick > >

Re: [Mesa-dev] [PATCH] st/va: use provided sizes and coords for vlVaGetImage

On 2018-10-09 04:09 PM, Ilia Mirkin wrote: On Tue, Oct 9, 2018 at 4:03 PM wrote: From: Boyuan Zhang vlVaGetImage should respect the width, height, and coordinates x and y that passed in. Therefore, pipe_box should be created with the passed in values instead of surface width/height. v2:

Re: [Mesa-dev] [PATCH] st/va: use provided sizes and coords for vlVaGetImage

On Tue, Oct 9, 2018 at 4:03 PM wrote: > > From: Boyuan Zhang > > vlVaGetImage should respect the width, height, and coordinates x and y that > passed in. Therefore, pipe_box should be created with the passed in values > instead of surface width/height. > > v2: add input size check, return error

[Mesa-dev] [PATCH] st/va: use provided sizes and coords for vlVaGetImage

From: Boyuan Zhang vlVaGetImage should respect the width, height, and coordinates x and y that passed in. Therefore, pipe_box should be created with the passed in values instead of surface width/height. v2: add input size check, return error when size out of bounds Signed-off-by: Boyuan Zhang

Re: [Mesa-dev] [PATCH] st/va: use provided sizes and coords for getimage

On Tue, Oct 9, 2018 at 3:09 PM Boyuan Zhang wrote: > > Thanks for the explanation ilia. > > I'm curious too here that if it's legal for player to not respect the > image size when calling vlVaGetImage. If player already know the size of > image is 100x100, then why should it still call

Re: [Mesa-dev] [PATCH] st/va: use provided sizes and coords for getimage

Thanks for the explanation ilia. I'm curious too here that if it's legal for player to not respect the image size when calling vlVaGetImage. If player already know the size of image is 100x100, then why should it still call vlVaGetImage with width/height=600? I mean when VA-API player calls

[Mesa-dev] [Bug 35268] initial-exec TLS model breaks dlopen'ed libGL

https://bugs.freedesktop.org/show_bug.cgi?id=35268 --- Comment #22 from Rich Felker --- Ping. Looking at this again, I see the original proposed patch introduced a lot of extra asm to save/restore registers that might be clobbered by __tls_get_addr. That could all go away if tlsdesc support is

Re: [Mesa-dev] [PATCH 3/9] nir/opt_peephole_select: Don't peephole_select expensive math instructions

Drive-by comment: At some point, maybe we just want to replace all these booleans with a function pointer that takes an instruction and returns a bool. That way the driver can define "expensive". Do with that what you will. On October 9, 2018 11:50:45 Ian Romanick wrote: On 10/08/2018

Re: [Mesa-dev] [PATCH] st/va: use provided sizes and coords for getimage

On Tue, Oct 9, 2018 at 1:59 PM Boyuan Zhang wrote: > > Hi ilia, > > I saw the function > u_box_clip_2d(https://cgit.freedesktop.org/mesa/mesa/tree/src/gallium/auxiliary/util/u_box.h#n74). > > But I still don't quite understand why we need to do this? Or say, what > will happen if we don't do this

Re: [Mesa-dev] [PATCH] svga: change svga_destroy_shader_variant() to return void

Reviewed-by: Neha Bhende Regards, Neha From: Brian Paul Sent: Tuesday, October 9, 2018 8:06:26 AM To: mesa-dev@lists.freedesktop.org Cc: Charmaine Lee; Neha Bhende Subject: [PATCH] svga: change svga_destroy_shader_variant() to return void

Re: [Mesa-dev] [PATCH] st/va: use provided sizes and coords for getimage

Hi ilia, I saw the function u_box_clip_2d(https://cgit.freedesktop.org/mesa/mesa/tree/src/gallium/auxiliary/util/u_box.h#n74). But I still don't quite understand why we need to do this? Or say, what will happen if we don't do this box clipping here? Can you provide more information about

[Mesa-dev] [Bug 108275] Breaking out of loop creates broken code on RADV

https://bugs.freedesktop.org/show_bug.cgi?id=108275 --- Comment #8 from mais...@archlinux.us --- So, apparently not *everything* is falsely culled, just most of it, depending on view angle. -- You are receiving this mail because: You are the QA Contact for the bug. You are the assignee for the

[Mesa-dev] [Bug 108275] Breaking out of loop creates broken code on RADV

https://bugs.freedesktop.org/show_bug.cgi?id=108275 --- Comment #7 from mais...@archlinux.us --- Broken capture: https://drive.google.com/file/d/1qdUNsJqLE0kzBKz_h7kPqb6ivbVLEH9P/view?usp=sharing Working capture (without spirv-opt on the offending compute shader):

Re: [Mesa-dev] [Mesa-stable] [PATCH] radv: switch on EOP when primitive restart is enabled with triangle strips

This will decrease geometry performance 2x. The correct workaround is to set PARTIAL_VS_WAVE_ON for all strip primitive types with restart indices. Marek On Tue, Oct 9, 2018 at 8:15 AM Samuel Pitoiset wrote: > > Otherwise, Yakuza hangs the GPU with DXVK. We don't know if > linetrip and

Re: [Mesa-dev] [PATCH 07/12] nir: Allow [iu]mul_high on non-32-bit types

On Tue, Oct 9, 2018 at 11:32 AM Ian Romanick wrote: > On 10/08/2018 02:14 PM, Jason Ekstrand wrote: > > On Mon, Oct 8, 2018 at 3:46 PM Ian Romanick > > wrote: > > > > On 10/05/2018 09:10 PM, Jason Ekstrand wrote: > > > --- > > >

Re: [Mesa-dev] [PATCH] intel: Don't propagate conditional modifiers if a UD source is negated

On Tue, Oct 9, 2018 at 11:24 AM Ian Romanick wrote: > On 10/09/2018 09:00 AM, Jason Ekstrand wrote: > > This fixes a bug uncovered by my NIR integer division by constant > > optimization series. > > > > Fixes: 19f9cb72c8b "i965/fs: Add pass to propagate conditional..." > > Fixes: 627f94b72e0

Re: [Mesa-dev] [PATCH] svga: change svga_destroy_shader_variant() to return void

Reviewed-by: Charmaine Lee From: Brian Paul Sent: Tuesday, October 9, 2018 8:06:26 AM To: mesa-dev@lists.freedesktop.org Cc: Charmaine Lee; Neha Bhende Subject: [PATCH] svga: change svga_destroy_shader_variant() to return void

Re: [Mesa-dev] [PATCH 3/9] nir/opt_peephole_select: Don't peephole_select expensive math instructions

On 10/08/2018 01:34 PM, Thomas Helland wrote: > Den tor. 30. aug. 2018 kl. 07:37 skrev Ian Romanick : >> >> From: Ian Romanick >> >> On some GPUs, especially older Intel GPUs, some math instructions are >> very expensive. On those architectures, don't reduce flow control to a >> csel if one of

Re: [Mesa-dev] [PATCH v2 1/2] spirv/nir: handle memory access qualifiers for SSBO loads/stores

On Tue, Oct 9, 2018 at 10:05 AM Samuel Pitoiset wrote: > v2: change how the access qualifiers are accumulated > > Signed-off-by: Samuel Pitoiset > --- > src/compiler/nir/nir_intrinsics.py | 4 +-- > src/compiler/spirv/spirv_to_nir.c | 14 +++-- > src/compiler/spirv/vtn_private.h | 6

Re: [Mesa-dev] [PATCH 07/12] nir: Allow [iu]mul_high on non-32-bit types

On 10/08/2018 02:14 PM, Jason Ekstrand wrote: > On Mon, Oct 8, 2018 at 3:46 PM Ian Romanick > wrote: > > On 10/05/2018 09:10 PM, Jason Ekstrand wrote: > > --- > >  src/compiler/nir/nir_constant_expressions.py |  1 + > > 

Re: [Mesa-dev] [PATCH] nvc0: fix blitting red to srgb8_alpha

yeah, no. Everything should be handled. Got a little bit confused with the luminance check. On Tue, Oct 9, 2018 at 4:07 PM Ilia Mirkin wrote: > > What other single component formats are there that aren't covered by the > earlier ifs? > > On Tue, Oct 9, 2018, 07:49 Karol Herbst wrote: >> >>

Re: [Mesa-dev] [PATCH] intel: Don't propagate conditional modifiers if a UD source is negated

On 10/09/2018 09:00 AM, Jason Ekstrand wrote: > This fixes a bug uncovered by my NIR integer division by constant > optimization series. > > Fixes: 19f9cb72c8b "i965/fs: Add pass to propagate conditional..." > Fixes: 627f94b72e0 "i965/vec4: adding vec4_cmod_propagation..." > Cc: Matt Turner >

Re: [Mesa-dev] [PATCH] glsl/linker: Check the subroutine associated functions names

Hi Ian, Thanks for the explanation. I've just sent a patch for review with the compile time check https://patchwork.freedesktop.org/patch/255542/ Regards, Vadym пт, 5 окт. 2018 г. в 16:55, Ian Romanick : > On 10/03/2018 11:52 PM, Iago Toral wrote: > > On Wed, 2018-10-03 at 16:24 +0300, Vadim

[Mesa-dev] [PATCH] glsl: Check the subroutine associated functions names

Adding compile time check for subroutine functions with the same names. Similar check for intrastage linking was already landed in commit 5f0567a4f60. From Section 6.1.2 (Subroutines) of the GLSL 4.00 specification "A program will fail to compile or link if any shader or stage contains

[Mesa-dev] [PATCH] intel: Don't propagate conditional modifiers if a UD source is negated

This fixes a bug uncovered by my NIR integer division by constant optimization series. Fixes: 19f9cb72c8b "i965/fs: Add pass to propagate conditional..." Fixes: 627f94b72e0 "i965/vec4: adding vec4_cmod_propagation..." Cc: Matt Turner --- .../compiler/brw_fs_cmod_propagation.cpp | 25

[Mesa-dev] [Bug 108305] dEQP-VK.wsi.xlib.swapchain.render.basic hangs

https://bugs.freedesktop.org/show_bug.cgi?id=108305 Jason Ekstrand changed: What|Removed |Added Assignee|mesa-dev@lists.freedesktop. |fdo-b...@engestrom.ch

[Mesa-dev] [Bug 108305] dEQP-VK.wsi.xlib.swapchain.render.basic hangs

https://bugs.freedesktop.org/show_bug.cgi?id=108305 Bug ID: 108305 Summary: dEQP-VK.wsi.xlib.swapchain.render.basic hangs Product: Mesa Version: unspecified Hardware: Other OS: All Status: NEW Severity:

[Mesa-dev] [PATCH] svga: change svga_destroy_shader_variant() to return void

svga_destroy_shader_variant() itself flushes and retries the command if there's a failure. So no need for the callers to do it. Other callers of the function were already ignoring the return value. This also fixes a corner-case double-free reported by Coverity (and reported by Dave Airlie).

[Mesa-dev] [PATCH v2 1/2] spirv/nir: handle memory access qualifiers for SSBO loads/stores

v2: change how the access qualifiers are accumulated Signed-off-by: Samuel Pitoiset --- src/compiler/nir/nir_intrinsics.py | 4 +-- src/compiler/spirv/spirv_to_nir.c | 14 +++-- src/compiler/spirv/vtn_private.h | 6 src/compiler/spirv/vtn_variables.c | 48

Re: [Mesa-dev] [PATCH mesa] st/dri: remove leftover local variable

On Tue, 9 Oct 2018 at 14:34, Eric Engestrom wrote: > > Left over from the cleanup in 6ccc435e7ad92bb0ba77d "pipe-loader: move dup(fd) > within pipe_loader_drm_probe_fd" > Reviewed-by: Emil Velikov -Emil ___ mesa-dev mailing list

Re: [Mesa-dev] [PATCH] nvc0: fix blitting red to srgb8_alpha

What other single component formats are there that aren't covered by the earlier ifs? On Tue, Oct 9, 2018, 07:49 Karol Herbst wrote: > just asking, because your commit message more or less hinted towards > that being about r -> srgb conversions, but it makes sense that any > single component

Re: [Mesa-dev] [PATCH] glsl: prevent qualifiers modification of predeclared variables

Hello, Please find my comments below: On Sat, Oct 6, 2018 at 1:07 AM Ian Romanick wrote: > On 10/05/2018 03:02 PM, Ian Romanick wrote: > > On 10/04/2018 07:08 AM, asimiklit.w...@gmail.com wrote: > >> From: Andrii Simiklit > >> > >> GLSL 3.7 (Identifiers): > >> "However, as noted in the

[Mesa-dev] [PATCH] radv: reduce max{Uniform, Storage}BufferRange values

2^32-1 is too high. This fixes the following crucible tests: stress.limits.buffer-update.range.uniform stress.limits.buffer-update.range.storage Signed-off-by: Samuel Pitoiset --- src/amd/vulkan/radv_device.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git

[Mesa-dev] [PATCH mesa] st/dri: remove leftover local variable

Left over from the cleanup in 6ccc435e7ad92bb0ba77d "pipe-loader: move dup(fd) within pipe_loader_drm_probe_fd" Signed-off-by: Eric Engestrom --- src/gallium/state_trackers/dri/dri2.c | 1 - 1 file changed, 1 deletion(-) diff --git a/src/gallium/state_trackers/dri/dri2.c

Re: [Mesa-dev] [PATCH 1/3] meson: only build clapi tests when OpenGL is being built

On Friday, 2018-10-05 11:05:10 -0700, Dylan Baker wrote: > Otherwise building just vulkan (among other things) will build these > tests, pull in a bunch of stuff they shouldn't, and potentially fail to > compile. Series looks good to me: Reviewed-by: Eric Engestrom > --- >

Re: [Mesa-dev] [PATCH mesa] radv: add missing meson c++ visibility arguments

On Monday, 2018-10-08 08:45:18 -0700, Dylan Baker wrote: > Reviewed-by: Dylan Baker > > You could add radv_flags to future safety, but it's not used currently so this > should fix things. Good point, thanks. Added with your r-b and pushed :) > > Quoting Eric Engestrom (2018-10-08 08:25:58) >

[Mesa-dev] [PATCH] dri: Include kernel release in vendor info

The userspace driver does not exist in isolation and occasionally depends on kernel uapi, and so it is useful in bug reports to include that information. (radeonsi, r600 and radv already include utsname) References: https://bugs.freedesktop.org/show_bug.cgi?id=108282 ---

Re: [Mesa-dev] [PATCH] radv: switch on EOP when primitive restart is enabled with triangle strips

Reviewed-by: Bas Nieuwenhuizen On Tue, Oct 9, 2018 at 2:15 PM Samuel Pitoiset wrote: > > Otherwise, Yakuza hangs the GPU with DXVK. We don't know if > linetrip and pointlist are affected, so my point is to do that > only for triangle strips. > > Cc: mesa-sta...@lists.freedesktop.org >

[Mesa-dev] [Bug 108105] [DXVK] Dauntless Helmets rendering incorrectly on Vega, works in AMDVLK

https://bugs.freedesktop.org/show_bug.cgi?id=108105 --- Comment #11 from Samuel Pitoiset --- What version of renderdoc did you use? If that doesn't work with rdc, can you try recording an apitrace? -- You are receiving this mail because: You are the QA Contact for the bug. You are the assignee

[Mesa-dev] [PATCH] radv: switch on EOP when primitive restart is enabled with triangle strips

Otherwise, Yakuza hangs the GPU with DXVK. We don't know if linetrip and pointlist are affected, so my point is to do that only for triangle strips. Cc: mesa-sta...@lists.freedesktop.org Signed-off-by: Samuel Pitoiset --- src/amd/vulkan/radv_pipeline.c | 3 +-- 1 file changed, 1 insertion(+), 2

[Mesa-dev] [Bug 108275] Breaking out of loop creates broken code on RADV

https://bugs.freedesktop.org/show_bug.cgi?id=108275 --- Comment #6 from Samuel Pitoiset --- Yes, please record a renderdoc capture. -- You are receiving this mail because: You are the QA Contact for the bug. You are the assignee for the bug.___

Re: [Mesa-dev] [PATCH] etnaviv: Make sure rs alignment checks match

Hi, On Mon, Sep 17, 2018 at 04:06:57PM +0200, Guido Günther wrote: > etna_resource_alloc and etna_resource_from_handle currently use different > checks. > This leads to > >etna_resource_from_handle:492: target=2, > format=PIPE_FORMAT_B8G8R8X8_UNORM, 1080x1920x1, array_size=1, last_level=0,

Re: [Mesa-dev] [PATCH] nvc0: fix blitting red to srgb8_alpha

just asking, because your commit message more or less hinted towards that being about r -> srgb conversions, but it makes sense that any single component blits wouldn't work. Reviewed-by: Karol Herbst On Tue, Oct 9, 2018 at 1:03 PM Ilia Mirkin wrote: > > Ones that blit to srgb, yes. Couldn't

Re: [Mesa-dev] [PATCH] nvc0: fix blitting red to srgb8_alpha

Ones that blit to srgb, yes. Couldn't get those to work with the 2d engine. On Tue, Oct 9, 2018, 04:46 Karol Herbst wrote: > but this movs all single color blits to the 3d blitter, right? > On Sun, Oct 7, 2018 at 11:50 PM Ilia Mirkin wrote: > > > > For some reason the 2d engine can't handle

Re: [Mesa-dev] [PATCH v2 2/2] radv: implement clear operations for R32G32B32

On Tue, Oct 9, 2018 at 12:46 PM Samuel Pitoiset wrote: > > > > On 10/9/18 12:38 PM, Bas Nieuwenhuizen wrote: > > On Tue, Oct 9, 2018 at 12:25 PM Samuel Pitoiset > > wrote: > >> > >> This fixes crashes for some CTS: > >> dEQP-VK.api.copy_and_blit.core.blit_image.all_formats.color.*.linear_*_* >

Re: [Mesa-dev] [PATCH v2 2/2] radv: implement clear operations for R32G32B32

On 10/9/18 12:38 PM, Bas Nieuwenhuizen wrote: On Tue, Oct 9, 2018 at 12:25 PM Samuel Pitoiset wrote: This fixes crashes for some CTS: dEQP-VK.api.copy_and_blit.core.blit_image.all_formats.color.*.linear_*_* dEQP-VK.api.copy_and_blit.core.blit_image.all_formats.color.*.*_linear_* v2: fix

Re: [Mesa-dev] [PATCH v2 2/2] radv: implement clear operations for R32G32B32

On Tue, Oct 9, 2018 at 12:25 PM Samuel Pitoiset wrote: > > This fixes crashes for some CTS: > dEQP-VK.api.copy_and_blit.core.blit_image.all_formats.color.*.linear_*_* > dEQP-VK.api.copy_and_blit.core.blit_image.all_formats.color.*.*_linear_* > > v2: fix stride > > Bugzilla:

[Mesa-dev] [Bug 108275] Breaking out of loop creates broken code on RADV

https://bugs.freedesktop.org/show_bug.cgi?id=108275 --- Comment #5 from mais...@archlinux.us --- To see the result, you would need a repro. I can create a RenderDoc capture if that helps (I run this on an RX470), although "bogus" in this case is actually that nothing is rendered. The shader in

[Mesa-dev] [PATCH v2 2/2] radv: implement clear operations for R32G32B32

This fixes crashes for some CTS: dEQP-VK.api.copy_and_blit.core.blit_image.all_formats.color.*.linear_*_* dEQP-VK.api.copy_and_blit.core.blit_image.all_formats.color.*.*_linear_* v2: fix stride Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108113 Signed-off-by: Samuel Pitoiset ---

Re: [Mesa-dev] [PATCH 2/2] radv: implement clear operations for R32G32B32

On 10/9/18 11:22 AM, Bas Nieuwenhuizen wrote: " On Mon, Oct 8, 2018 at 2:38 PM Samuel Pitoiset wrote: This fixes crashes for some CTS: dEQP-VK.api.copy_and_blit.core.blit_image.all_formats.color.*.linear_*_* dEQP-VK.api.copy_and_blit.core.blit_image.all_formats.color.*.*_linear_* Bugzilla:

[Mesa-dev] [Bug 108263] glGetTexImage with PBO is not accelerated on Gallium

https://bugs.freedesktop.org/show_bug.cgi?id=108263 mirh changed: What|Removed |Added CC||m...@protonmail.ch -- You are receiving this

Re: [Mesa-dev] [PATCH 2/2] radv: implement clear operations for R32G32B32

" On Mon, Oct 8, 2018 at 2:38 PM Samuel Pitoiset wrote: > > This fixes crashes for some CTS: > dEQP-VK.api.copy_and_blit.core.blit_image.all_formats.color.*.linear_*_* > dEQP-VK.api.copy_and_blit.core.blit_image.all_formats.color.*.*_linear_* > > Bugzilla:

Re: [Mesa-dev] [PATCH] nvc0: fix blitting red to srgb8_alpha

but this movs all single color blits to the 3d blitter, right? On Sun, Oct 7, 2018 at 11:50 PM Ilia Mirkin wrote: > > For some reason the 2d engine can't handle this. Red formats get special > treatment there, so perhaps related. > > Fixes dEQP-GLES3 tests of the form: > >

Re: [Mesa-dev] [PATCH] nv50,nvc0: guard against zero-size blits

Reviewed-by: Karol Herbst On Sun, Oct 7, 2018 at 11:50 PM Ilia Mirkin wrote: > > The current state tracker can generate these sometimes. Fixing this is > more involved, and due to some integer math we can generate > divisions-by-zero. > > Signed-off-by: Ilia Mirkin > Cc:

Re: [Mesa-dev] [PATCH v2 3/6] radeonsi:optimizing SET_CONTEXT_REG for shaders VS

On 2018-10-09 12:18 a.m., Marek Olšák wrote: > I didn't see any issue when I was testing Tahiti and Hawaii on amdgpu. > radeon is too unstable for a 16-thread piglit, so I never use it. > > You can try to set reg_saved = 0 in si_begin_new_gfx_cs. That should > prevent issues with CLEAR_STATE.

[Mesa-dev] [Bug 108275] Breaking out of loop creates broken code on RADV

https://bugs.freedesktop.org/show_bug.cgi?id=108275 --- Comment #4 from Samuel Pitoiset --- Just compiled your stuff, how do I see the bogus rendering? -- You are receiving this mail because: You are the QA Contact for the bug. You are the assignee for the

[Mesa-dev] [Bug 108118] AMDGPU sometimes hangs forever when running graphical applications

https://bugs.freedesktop.org/show_bug.cgi?id=108118 Samuel Pitoiset changed: What|Removed |Added Resolution|--- |FIXED Status|NEW