Marek Olšák [Fri, 19 Mar 2021 22:39:23 +0000 (18:39 -0400)]
radeonsi: remove unused SI_CP_DMA_SKIP_* definitions
The existing uses had no effect.
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9795>
Marek Olšák [Sat, 20 Mar 2021 02:17:38 +0000 (22:17 -0400)]
radeonsi: rename internal compute sync flags
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9795>
Marek Olšák [Fri, 19 Mar 2021 22:48:04 +0000 (18:48 -0400)]
radeonsi: never set DISABLE_WR_CONFIRM for CP DMA clears and copies
Only prefetches set it. Unsynchronized clears and copies shouldn't set it
because syncing later wouldn't wait for the writes.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9795>
Marek Olšák [Fri, 19 Mar 2021 21:54:08 +0000 (17:54 -0400)]
radeonsi: don't insert start/stop pipeline stat events if it has no effect
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9795>
Marek Olšák [Fri, 19 Mar 2021 21:41:59 +0000 (17:41 -0400)]
radeonsi: set the clear/copy cache policy based on L2 cache size
This matches the intent.
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9795>
Marek Olšák [Sun, 28 Mar 2021 12:48:19 +0000 (08:48 -0400)]
radeonsi: disable sparse buffers on gfx7-8
Cc: 20.3 21.0 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9795>
Marek Olšák [Sun, 28 Mar 2021 11:46:25 +0000 (07:46 -0400)]
ac/llvm: don't set unsupported xnack options to fix LLVM crashes on gfx6-8
LLVM prints an error if xnack is unsupported and it uses a global stream
object that is not thread-safe. Since Mesa uses multiple threads to compile
shaders, there is a small chance that it will crash.
Just don't set any xnack options to use LLVM defaults.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4439
Cc: 20.3 21.0 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9795>
Samuel Pitoiset [Wed, 31 Mar 2021 08:08:27 +0000 (10:08 +0200)]
radv: use COLOR_ATTACHMENT_OPTIMAL for fast clear/hw resolve operations
This should be equivalent without needed to force enable FMASK for
some specific internal pipelines.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9940>
Samuel Pitoiset [Wed, 31 Mar 2021 07:08:26 +0000 (09:08 +0200)]
radv: do not force enable FMASK during MSAA blits
This is no longer needed since FMASK is also compressed for
transfer dst operations.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9940>
Samuel Pitoiset [Thu, 1 Apr 2021 16:38:38 +0000 (18:38 +0200)]
radv: cleanup FMASK expand transitions
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9990>
Samuel Pitoiset [Thu, 1 Apr 2021 15:29:22 +0000 (17:29 +0200)]
radv: compress FMASK for all layouts except GENERAL
The COMPRESSION bit is FMASK and this is much faster! Should
speedup transfer dst operations with MSAA images considerably.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9990>
Eric Anholt [Thu, 1 Apr 2021 17:44:13 +0000 (10:44 -0700)]
ci/freedreno: Switch to the trimmed glxgears trace.
The old one had a ton of frames and took ~5 minutes on a306.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9957>
Eric Anholt [Thu, 1 Apr 2021 17:25:00 +0000 (10:25 -0700)]
freedreno/a6xx: Fix alpha tests.
Apparently I inverted the sense of this flag back when we didn't have
piglit testing. Fixes terrible rendering in minetest, HL2, CS:Source, and
CS.
Fixes:
0369dd907778 ("freedreno/a6xx: Add ARB_depth_clamp and separate clamp support.")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9957>
Eric Anholt [Thu, 1 Apr 2021 17:18:01 +0000 (10:18 -0700)]
freedreno/a5xx: Fix alpha test vs early Z bugs.
Just like with discards, we have to disable early Z writes when alpha test
is enabled.
Fixes rendering on HL2, CS: Source, counter-strike, and minetest.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9957>
Eric Anholt [Wed, 31 Mar 2021 19:19:16 +0000 (12:19 -0700)]
ci/freedreno: Add trace testing on a3xx, a5xx.
Having compared rendering between a6xx and these, I found several bugs.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9957>
Eric Anholt [Wed, 31 Mar 2021 19:32:07 +0000 (12:32 -0700)]
ci/freedreno: Rename a306-test and a530-test to drop "arm64" from the name.
We don't have an armhf variant, and probably won't. Now matches a630.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9957>
Eric Anholt [Wed, 31 Mar 2021 18:42:56 +0000 (11:42 -0700)]
ci/freedreno: Add more new traces for a630 (minetest, TDM, pioneer, glyphy).
These are all recent traces that have been added.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9957>
Mike Blumenkrantz [Mon, 22 Mar 2021 20:13:36 +0000 (16:13 -0400)]
zink: emit ImageCubeArray cap when accessing arrayed cube dimension images
required by spec
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9987>
Mike Blumenkrantz [Tue, 9 Mar 2021 14:33:54 +0000 (09:33 -0500)]
zink: use max_rt to determine number of blend state attachments
this is more consistent with how gallium applies the states
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9988>
Erik Faye-Lund [Tue, 30 Mar 2021 11:56:29 +0000 (13:56 +0200)]
bin/gen_release_notes.py: more robust rST escaping
This copies code from the xml2rst to escape rST strings. Hopefully this
will be more robust than what we've done so far.
I really wish docutils would have utils for this directly, seems kinda
essential.
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9917>
Erik Faye-Lund [Tue, 30 Mar 2021 10:46:49 +0000 (12:46 +0200)]
ci: turn sphinx-build warnings into errors
This should make it harder to accidentally introduce formatting issues
into the docs.
Acked-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9914>
Erik Faye-Lund [Thu, 1 Apr 2021 08:28:27 +0000 (10:28 +0200)]
docs: fix release notes for 20.3.5
This is incorrect escaping of the RST.
Fixes:
930d2a92a32 ("docs: add release notes for 20.3.5")
Acked-by: Dylan Baker <dylan@pnwbakers.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9914>
Pierre-Eric Pelloux-Prayer [Tue, 9 Mar 2021 19:50:39 +0000 (20:50 +0100)]
vbo/dlist: use DrawGallium(Complex)
We can build the needed structs during list compilation and then
use DrawGallium (if one draw or a single primitive mode is used) or DrawGalliumComplex
otherwise.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9533>
Pierre-Eric Pelloux-Prayer [Wed, 10 Mar 2021 10:20:37 +0000 (11:20 +0100)]
mesa: make _mesa_HashTable InDeleteAll debug only
It's only used in asserts so we can wrap it inside
This makes the struct 32 bytes instead of 40 in release
builds.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9533>
Pierre-Eric Pelloux-Prayer [Wed, 10 Mar 2021 11:04:39 +0000 (12:04 +0100)]
mesa/hash: switch to simple_mtx
simple_mtx are faster and smaller (4 bytes instead of 8) so switch to using them.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9533>
Pierre-Eric Pelloux-Prayer [Wed, 10 Mar 2021 10:59:24 +0000 (11:59 +0100)]
mesa/hash: make the mtx non-recursive
Now that no one is recursively locking it we can switch to
a plain mutex.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9533>
Pierre-Eric Pelloux-Prayer [Thu, 11 Mar 2021 19:40:11 +0000 (20:40 +0100)]
mesa: remove 2 recursive lock usages of _mesa_HashTable
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9533>
Pierre-Eric Pelloux-Prayer [Tue, 20 Oct 2020 19:08:37 +0000 (21:08 +0200)]
dlist: remove ListExt feature
This is only used by vbo_save_api so let's simplify the code and add a normal
opcode for this one.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9533>
Dave Airlie [Thu, 1 Apr 2021 04:25:08 +0000 (14:25 +1000)]
zink/ci: update results after lavapipe clear fixes
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9971>
Dave Airlie [Thu, 1 Apr 2021 03:52:58 +0000 (13:52 +1000)]
lavapipe: fix only clearing depth or stencil paths.
This fixes the ds clears path to clear only depth or stencil
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Fixes:
b38879f8c5f ("vallium: initial import of the vulkan frontend")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9971>
Ian Romanick [Mon, 29 Mar 2021 20:17:10 +0000 (13:17 -0700)]
mesa: Clean up _mesa_layout_parameters after previous commit
After the previous change, PASS 1 can be trivially pulled out of the
loop.
With PASS 1 removed, the loop can be unrolled, and a lot of code can be
deleted (from the unrolls). This saves a couple lines of code, and it
makes the function a little easier to follow.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9867>
Ian Romanick [Mon, 29 Mar 2021 19:32:57 +0000 (12:32 -0700)]
mesa: Add anything dynamically indexed before any non-dynamically indexed
Things that are not dynamically indexed must be added last. This is
necessary so that values that are both statically indexed (or used
directly) and dynamically indexed will only be added once. With the
above change, if the constant 47 is used as a literal in an instruction
and in an array that is dynamically indexed, it will be added to
`Parameters` twice. On (really old) GPUs that store constants and other
parameters in the same storage, this can cause some valid programs to
exceed the storage limits. I don't know about R300 or NV30, but R200
was limited to something like 256 vec4s. This applies to constants,
state parameters, and local parameters (the assembly shader version of
uniforms).
The problem this causes here is that the final parameter layout created
in `_mesa_layout_parameters` may have more parameters than the input
layout. The fundamental assumption of that routine (and documented as
an assumption of `copy_indirect_accessed_array`) is that the input size
and the output size will be the same.
The affected shader had something like below. This is a common pattern
for ARB assembly shaders generated by NVIDIA's cgc compiler. As far as
I can tell, the majory of applications that use ARB assembly shaders
either use cgc or use some sort of DX9 crosscompiler... that generates
similar patterns.
PARAM c[141] = { program.local[0..133],
{ 255, 0.1, 3, 1 },
{ 0.5, 2, 0.
15915491, 0.25 },
{ 0, 0.5, 1, -1 },
{ 24.980801, -24.980801, -60.145809, 60.145809 },
{ 85.453789, -85.453789, -64.939346, 64.939346 },
{ 19.73921, -19.73921, -9, 0.75 },
{ -999999 } };
The shader contains instructions like
MUL R0.x, R0, c[135].y;
and
DP4 R2.z, c[A0.x + 6], R1;
Starting with
b9bff76b630, the constants at the end of `c` would get
added to `Parameters` twice. The first time they are added due to
instructions that directly access the array (e.g., the `c[135].y`
above). The second time is because they are part of an array that is
dynamically indexed. As a result, the final layout of Parameters
(calculated by `_mesa_layout_parameters`) is 7 elements larger than the
input layout.
Since
bcc61a01d49 fixed the allocation size of `ParameterValues`,
`copy_indirect_accessed_array` will now write past the end of the array.
The eventually results in a crash in `free`. Thankfully Valgrind was
able to help find the real source of the problem.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Fixes:
b9bff76b630 ("mesa: put constants before state vars for ARB programs")
Closes: #4505
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9867>
Adam Jackson [Thu, 18 Mar 2021 22:02:18 +0000 (18:02 -0400)]
mesa/st: Check for successful framebuffer allocation in st_api_make_current
Ran into this while trying to rework fbconfig setup, due to a bug I
ended up trying to allocate a PIPE_FORMAT_NONE framebuffer, which failed
like you'd hope, but which we weren't converting into an error in
st_api_make_current. Instead we'd treat it like binding no drawable to
the context, which is really not what was asked for, so let's go ahead
and make this an error.
Reviewed-by: Eric Faye-Lund <kusmabite@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9956>
Adam Jackson [Wed, 24 Mar 2021 18:20:11 +0000 (14:20 -0400)]
mesa/st: Remove unused ST_ATTACHMENT_SAMPLE
Not sure what this was supposed to do, but whatever it did, it doesn't.
Reviewed-by: Eric Faye-Lund <kusmabite@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9956>
Danylo Piliaiev [Tue, 23 Mar 2021 16:39:32 +0000 (18:39 +0200)]
turnip: enable VK_KHR_16bit_storage on A650
A650 can use the same SSBO descriptor for both 32-bit and 16-bit access,
which makes it easy to enable this extension.
Passes tests that run under:
dEQP-VK.spirv_assembly.instruction.*.16bit_storage.*
Rebased and modified commit from Jonathan Marek.
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9840>
Jonathan Marek [Tue, 23 Mar 2021 13:39:26 +0000 (15:39 +0200)]
turnip: enable VK_KHR_shader_float16_int8
ir3 supports 16-bit floats, so we can enable this.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9840>
Danylo Piliaiev [Thu, 25 Mar 2021 12:53:20 +0000 (14:53 +0200)]
turnip: enable infinities for f16 math and document the register
When float16 is enabled this will allow to pass a number of
float16 tests.
When A6XX_SP_FLOAT_CNTL_F16_NO_INF is set - all operations which
generate +-infinity generate +-MAX_HALF_FLOAT.
Fixes some tests from:
dEQP-VK.spirv_assembly.instruction.*.float16.*
dEQP-VK.spirv_assembly.instruction.*.float_controls.fp16.*
E.g.:
dEQP-VK.spirv_assembly.instruction.graphics.float16.arithmetic_1.sinh_vert
dEQP-VK.spirv_assembly.instruction.compute.float16.arithmetic_4.length
dEQP-VK.spirv_assembly.instruction.compute.float_controls.fp16.input_args.log_denorm_flush_to_zero_nostorage
dEQP-VK.spirv_assembly.instruction.compute.float_controls.fp16.input_args.log2_denorm_flush_to_zero_nostorage
dEQP-VK.spirv_assembly.instruction.compute.float_controls.fp16.input_args.inv_sqrt_denorm_flush_to_zero_nostorage
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9840>
Danylo Piliaiev [Wed, 24 Mar 2021 15:24:30 +0000 (17:24 +0200)]
ir3: convert shift amount to 16b for 16b shifts
NIR has shifts defined as:
opcode("*shr", 0, tuint, [0, 0], [tuint, tuint32], False, ...
However, in ir3 we have to ensure that both operators of shift
instruction have the same bitness.
Let's hope that in future the additional COV for constants would
be optimized away.
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9840>
Jonathan Marek [Tue, 23 Mar 2021 13:20:41 +0000 (15:20 +0200)]
turnip: implement VK_KHR_shader_float_controls
This matches the blob and doesn't require actually implementing controls
since the supported modes are just what the HW does.
Passes tests under:
dEQP-VK.spirv_assembly.*.float_controls.*
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9840>
Danylo Piliaiev [Wed, 31 Mar 2021 19:58:00 +0000 (22:58 +0300)]
ir3: nir_op_f2f16 should round to even
cat1 instructions round to zero by default.
When fp16 is enabled this will fix:
dEQP-VK.spirv_assembly.instruction.graphics.float_controls.fp16.input_args.rounding_rte_conv_from_fp32_nostorage_frag
dEQP-VK.spirv_assembly.instruction.graphics.float_controls.fp16.input_args.rounding_rte_conv_from_fp32_nostorage_vert
dEQP-VK.spirv_assembly.instruction.compute.float_controls.fp16.input_args.rounding_rte_conv_from_fp32_nostorage
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9840>
Keith Packard [Sun, 4 Feb 2018 01:07:15 +0000 (17:07 -0800)]
wsi/x11: Fix type of target_msc argument to x11_present_to_x11_dri3
Broken out of VK_GOOGLE_display_timing patch
Cc: stable
Co-author: Jakob Bornecrantz <jakob@collabora.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Jakob Bornecrantz <jakob@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9939>
Andrii Simiklit [Tue, 30 Mar 2021 15:42:58 +0000 (18:42 +0300)]
gitlab-ci: remove fixed tests
These tests are fixed by MR2333
Signed-off-by: Andrii Simiklit <andrii.simiklit@globallogic.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2333>
Andrii Simiklit [Wed, 24 Feb 2021 12:30:25 +0000 (14:30 +0200)]
glsl/linker: Fix xfb stride alignment for buffers containing 64bit types
Per OpenGL 4.6 spec:
"If no xfb_stride qualifier is specified for a
binding point, the stride is derived by identifying the variable associated with the
binding point having the largest offset, and then adding the offset and the size of
the variable, in basic machine units. If any variable associated with the binding
point contains double-precision floating-point components, the derived stride is
aligned to the next multiple of eight basic machine units. If a binding point has no
xfb_stride qualifier and no associated output variables, its stride is zero."
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Signed-off-by: Andrii Simiklit <andrii.simiklit@globallogic.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2333>
Danylo Piliaiev [Thu, 10 Oct 2019 16:11:00 +0000 (19:11 +0300)]
glsl/linker: Fix xfb with explicit locations and 64bit types
1) Per GL_ARB_enhanced_layouts if explicit location is set for varying,
each struct member, array element and matrix row will take
separate location. With GL_ARB_gpu_shader_fp64/GL_ARB_gpu_shader_int64
they may take two locations.
Examples:
| layout(location=0) dvec3[2] a; | layout(location=4) vec2[4] b; |
| | |
| 32b 32b 32b 32b | 32b 32b 32b 32b |
| 0 X X Y Y | 4 X Y 0 0 |
| 1 Z Z 0 0 | 5 X Y 0 0 |
| 2 X X Y Y | 6 X Y 0 0 |
| 3 Z Z 0 0 | 7 X Y 0 0 |
Previously it wasn't taken into account.
2) Captured double-precision variables should be aligned to
8 bytes per GL_ARB_gpu_shader_fp64:
"If any variable captured in transform feedback has double-precision
components, the practical requirements for defined behavior are:
...
(c) each double-precision variable captured must be aligned to a
multiple of eight bytes relative to the beginning of a vertex."
v2: fix `output_size` calculations
( Andrii Simiklit <andrii.simiklit@globallogic.com> )
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/1667
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Signed-off-by: Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2333>
Danylo Piliaiev [Thu, 10 Oct 2019 14:59:35 +0000 (17:59 +0300)]
glsl/linker: Fix attempts to split up 64bit varyings between slots
When packing varyings when there is only 32bit of space
left in slot 64bit type is attempted to be divided between
current and next slot. However there is neither code for
splitting the 64bit type nor for assembling it back.
Instead we add 32bit padding.
The above happens only in structs because their
members aren't sorted by packing order.
Example of the issue:
struct S {
vec3 a;
double d;
};
out flat S s;
Before, the packing went as:
slot 32b 32b 32b 32b
0 a.x a.y a.z d
1 d 0 0 0
After:
slot 32b 32b 32b 32b
0 a.x a.y a.z 0
1 d d 0 0
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Signed-off-by: Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/2333>
Michel Dänzer [Tue, 30 Mar 2021 10:54:15 +0000 (12:54 +0200)]
ci: Merge ARM testing docker images to a single arm_test one
The merged image contains kernels & rootfs for both arm64 & armhf
baremetal test jobs, and is smaller than either arm{64,hf}_test image
before.
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9955>
Michel Dänzer [Mon, 29 Mar 2021 13:09:06 +0000 (15:09 +0200)]
ci: Build ARM baremetal rootfs in native container
Doing so in an x86 container via qemu was slow, and started failing
recently after updating to a newer qemu version.
This also results in smaller arm*_test* docker images, since we need to
install fewer Debian packages in them.
As a bonus, this turns some piglit tests from fail to pass (Or maybe
they'll turn out to be flakes? They've passed at least 3 times in a
row).
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9955>
Michel Dänzer [Wed, 31 Mar 2021 07:32:20 +0000 (09:32 +0200)]
ci: Remove INCLUDE_PIGLIT
It was enabled by all callers.
Also remove the unused $CI_FAIRY_PACKAGES and $VK_CTS_PACKAGES
references.
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9955>
Eric Anholt [Mon, 29 Mar 2021 10:50:05 +0000 (12:50 +0200)]
ci: Drop the custom db820c kernel/dtb from the kernel+rootfs.
We use the CI-built kernel+rootfs these days. I haven't bumped image tags
because the files are definitely unused, and I'm rebuilding it all in the
next commit.
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9955>
Jesse Natalie [Wed, 31 Mar 2021 20:50:07 +0000 (13:50 -0700)]
wgl, d3d12: Add a d3d12-specific test for swapchain leaks
Reviewed-By: Bill Kristiansen <billkris@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9959>
Jesse Natalie [Wed, 31 Mar 2021 21:12:50 +0000 (14:12 -0700)]
d3d12: Clean up swapchains on framebuffer destruction
Reviewed-By: Bill Kristiansen <billkris@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9959>
Jesse Natalie [Wed, 31 Mar 2021 22:55:00 +0000 (15:55 -0700)]
d3d12: Add a constant for num_buffers
Reviewed-By: Bill Kristiansen <billkris@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9959>
Jesse Natalie [Wed, 31 Mar 2021 21:12:29 +0000 (14:12 -0700)]
wgl: Add a context to framebuffer destruction
If the window is destroyed on a thread that has a currently-bound
context, use that context for destroying the framebuffer. This
ensures that the winsys can wait for in-flight work before
destroying any resources.
If the window did have a context bound beforehand but it was unbound,
we should've already done a glFinish. If the window is destroyed from
an unrelated thread... well, we're screwed, but that's the best we can do.
Reviewed-By: Bill Kristiansen <billkris@microsoft.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9959>
Jesse Natalie [Wed, 31 Mar 2021 20:49:31 +0000 (13:49 -0700)]
wgl: Add unit test infrastructure for OpenGL32.dll on Windows
Reviewed-By: Bill Kristiansen <billkris@microsoft.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9959>
Jesse Natalie [Wed, 31 Mar 2021 20:47:19 +0000 (13:47 -0700)]
d3d12: Use CreateDXGIFactory2 and use the debug flag when appropriate
This enables leak detection and other debug messages for DXGI objects
Reviewed-By: Bill Kristiansen <billkris@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9959>
Mike Blumenkrantz [Sat, 5 Dec 2020 13:46:01 +0000 (08:46 -0500)]
zink: zink_push_constant -> zink_gfx_push_constant
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9979>
Samuel Pitoiset [Tue, 30 Mar 2021 11:06:33 +0000 (13:06 +0200)]
radv: use a sampled image descriptor for reads for the MSAA color decompress
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9915>
Mike Blumenkrantz [Wed, 31 Mar 2021 19:35:11 +0000 (15:35 -0400)]
zink: use VkSubresourceLayout::depthPitch as layer_stride when mapping 3D imgs
this is not only more correct according to vk spec, it avoids having a 0-sized
layer_stride, which totally breaks the transfer map
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9969>
Mike Blumenkrantz [Wed, 31 Mar 2021 19:31:30 +0000 (15:31 -0400)]
zink: be more explicit about blit layer/depth usage
same as for resource copying, ensure that it's very clear we're abiding
by vk spec for array vs 3d usage
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9969>
Jason Ekstrand [Wed, 31 Mar 2021 22:45:42 +0000 (17:45 -0500)]
anv: Implement VK_EXT_conservative_rasterization
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4480
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9960>
Mike Blumenkrantz [Wed, 10 Feb 2021 23:40:55 +0000 (18:40 -0500)]
zink: enforce device lost status
many apps don't request device-lost notification, so just calling the reset
callback isn't enough; once the device has been lost, no more cmdbufs should
be submitted and the queue should not be waited on
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9963>
Mike Blumenkrantz [Sat, 23 Jan 2021 23:09:11 +0000 (18:09 -0500)]
zink: optimize batch states for timeline use
finishing a timeline wait guarantees that a given fence has completed,
meaning the accompanying batch static is implicitly available for use
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9963>
Mike Blumenkrantz [Tue, 29 Dec 2020 18:28:51 +0000 (13:28 -0500)]
zink: add timeline semaphore fastpath for checking/triggering batch completion
we can avoid locking to access batch states in these cases by just using a semaphore
to fast-forward the gpu to the batch we need completed
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9963>
Mike Blumenkrantz [Fri, 12 Feb 2021 19:15:27 +0000 (14:15 -0500)]
zink: hook up timeline semaphore signalling during batch submission
just basic parts, no waiting on semaphores yet
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9963>
Mike Blumenkrantz [Tue, 16 Mar 2021 19:21:16 +0000 (15:21 -0400)]
zink: handle expired deferred fences more reasonably
now that there's some tracking for the last-finished batch id, this can
be used to detect when an application holds onto a sync object for way too long,
to the point that the sync object has expired so far into the past that we
no longer have any record of it existing
fixes things like unigine superposition
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9963>
Mike Blumenkrantz [Tue, 29 Dec 2020 17:10:08 +0000 (12:10 -0500)]
zink: track last completed batch id to optimize checking states
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9963>
Mike Blumenkrantz [Tue, 12 Jan 2021 22:05:42 +0000 (17:05 -0500)]
zink: add batch tracking id for program struct
this reduces overhead of batch tracking
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9967>
Mike Blumenkrantz [Thu, 28 Jan 2021 15:15:45 +0000 (10:15 -0500)]
zink: add set_context_param hook
for handling PIPE_CONTEXT_PARAM_PIN_THREADS_TO_L3_CACHE
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9970>
Mike Blumenkrantz [Fri, 18 Dec 2020 17:06:12 +0000 (12:06 -0500)]
zink: handle PIPE_MAP_DONTBLOCK for buffer read maps
this is mostly used by queries
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9962>
Rhys Perry [Wed, 6 Jan 2021 15:24:26 +0000 (15:24 +0000)]
nir/gather_info: implement partial masking of struct and compact I/O
fossil-db (Sienna):
Totals from 138 (0.10% of 138791) affected shaders:
CodeSize: 504060 -> 482136 (-4.35%)
Instrs: 97318 -> 94518 (-2.88%)
Cycles: 389272 -> 378072 (-2.88%)
VMEM: 14397 -> 14614 (+1.51%); split: +1.76%, -0.25%
SMEM: 9088 -> 9024 (-0.70%)
VClause: 2915 -> 2430 (-16.64%)
SClause: 1790 -> 1791 (+0.06%)
PreVGPRs: 5013 -> 4998 (-0.30%)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8364>
Rhys Perry [Thu, 4 Feb 2021 14:10:59 +0000 (14:10 +0000)]
lavapipe: fix initialization of pipe_stream_output with unwritten outputs
nir_assign_io_var_locations() does not use outputs_written when assigning
driver locations. Use driver_location to avoid incorrectly guessing what
locations it assigned.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8364>
Tony Wasserka [Tue, 30 Mar 2021 15:17:27 +0000 (17:17 +0200)]
gitlab: rename RADV bug report template
Acked-by: Daniel Schürmann <daniel@schuermann.dev>
Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9923>
Erik Faye-Lund [Wed, 3 Feb 2021 12:06:51 +0000 (13:06 +0100)]
docs: no-op'd -> disabled
"Disabled" is a common English word, that perfectly covers this usage.
So let's use that instead of this non-dictionary word.
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9947>
Erik Faye-Lund [Wed, 3 Feb 2021 13:22:06 +0000 (14:22 +0100)]
docs: spell out freedesktop.org
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9947>
Erik Faye-Lund [Wed, 3 Feb 2021 12:14:34 +0000 (13:14 +0100)]
docs: spell out environment
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9947>
Erik Faye-Lund [Wed, 3 Feb 2021 11:14:47 +0000 (12:14 +0100)]
docs: spell out development
The "-devel"-suffix is only helpful to RedHat users. Debian based
distros use "-dev" instead. But let's get out of the distro-specific
business, and instead just spell out "development", as this applies
regardless of naming-scheme.
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9947>
Erik Faye-Lund [Mon, 28 Sep 2020 13:05:15 +0000 (15:05 +0200)]
docs: spell out full name of gitlab instance
While we're at it, quote it so it's clear that it's something special.
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9947>
Erik Faye-Lund [Fri, 19 Mar 2021 12:46:26 +0000 (13:46 +0100)]
docs: fix rst-quoting issues in release-notes
The first issue is benign, but the other two create rogue, invalid
links.
Fixes:
ca79b041cbf ("docs: add release notes for 21.0.0")
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9710>
Erik Faye-Lund [Fri, 19 Mar 2021 12:44:38 +0000 (13:44 +0100)]
docs: fix invalid rst
We need a single empty line between the code-block state and the text
in the block, otherwise the rST is invalid and the entire block will be
dropped, as is currently the case on the website.
While we're at it, remove some needless colons from these code-blocks as
well. They're not needed, and we usually don't have these in the docs.
Fixes:
a2a8c6a36c1 ("docs: Add some documentation of game GL buffer object mapping behavior.")
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9710>
Lionel Landwerlin [Tue, 23 Mar 2021 14:37:40 +0000 (16:37 +0200)]
intel/fs: implement another copy propagation restriction
We are missing an additional restriction on CHV & upcoming Xe-Hp.
v2: Quote BSW PRMs (Curro)
Check source is not a scalar (Curro)
Fix comment (Marcin)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9929>
Tapani Pälli [Tue, 30 Mar 2021 13:57:49 +0000 (16:57 +0300)]
iris: clamp PointWidth in 3DSTATE_SF like i965 does
Values match how MinimumPointWidth, MaximumPointWidth is setup. This
fixes assert hit in debug build when packing the struct with too large
value for genxml.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9942>
Vinson Lee [Mon, 29 Mar 2021 03:34:41 +0000 (20:34 -0700)]
Remove leftover dead code.
Fix defect reported by Coverity Scan.
Logically dead code (DEADCODE)
dead_error_line: Execution cannot reach this statement: return;.
Fixes:
bdf93f4e3b3 ("v3dv/cmd_buffer: return early for draw commands if there is nothing to draw")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9890>
Timothy Arceri [Mon, 22 Mar 2021 06:43:10 +0000 (17:43 +1100)]
glsl: enforce restrictions on builtin functions moved to compat
Section 8.9.4 (Compatibility Profile Texture Functions) of the
GLSL 4.20 spec outlines a number of builtin texture functions that
have been moved to compatibility shaders.
This change enforces those restrictions. Note we don't worry about
enforcing restrictions on the EXT_gpu_shader4 extensions of these
functions because EXT_gpu_shader4 should only be enabled for compat
already.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9887>
Bas Nieuwenhuizen [Wed, 20 Jan 2021 20:26:30 +0000 (21:26 +0100)]
zink: Only set the needed number of scissors.
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9968>
Bas Nieuwenhuizen [Wed, 20 Jan 2021 20:20:35 +0000 (21:20 +0100)]
zink: Remove initialization of some arrays
We're only using the same number of elements we're writing to.
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9968>
Mike Blumenkrantz [Tue, 30 Mar 2021 21:28:39 +0000 (17:28 -0400)]
zink: ci updates
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9935>
Mike Blumenkrantz [Mon, 22 Mar 2021 15:01:30 +0000 (11:01 -0400)]
zink: implement threaded context
this includes:
* async buffer mapping/replacement
* async queue submission
* async (threaded) gallium flush handling
the main churn here is from handling async gallium flushes, which involves
creating multiple gallium fences (zink_tc_fence) for each zink fence (zink_fence).
a tc fence may begin waiting for completion at any time, even before the zink_fence
has had its cmdbuf(s) submitted, so handling this type of desync ends up needing
almost a complete rewrite of the existing queue architecture
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9935>
Charmaine Lee [Tue, 30 Mar 2021 06:01:41 +0000 (23:01 -0700)]
gallivm: increase size of texture target enum bitfield
Need to bump up the size of texture target bitfield for MSVC.
Fixes:
0ce7c4a7c981 ("gallivm: Use the proper enum for the texture target bitfield.")
Reviewed-by: Neha Bhende <bhenden@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9928>
Eric Anholt [Tue, 30 Mar 2021 20:45:37 +0000 (13:45 -0700)]
ci/freedreno: Demote a630-asan to a manual test for now.
It's flaky in producing Missing results. I've got an uprev that should
avoid the issue (and possibly a followon actual fix), but it's blocked on
being able to rebuild the arm containers.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9932>
Erik Faye-Lund [Wed, 3 Feb 2021 11:45:17 +0000 (12:45 +0100)]
docs: fix incorrect possessive form
When a word ends in an 's', the possessive form ends with a trailing
apostrophe instead.
Acked-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9946>
Erik Faye-Lund [Wed, 3 Feb 2021 12:13:25 +0000 (13:13 +0100)]
docs: dfsm -> DFSM
This abberation is spelled in all-caps other places in this document.
Acked-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9946>
Erik Faye-Lund [Wed, 3 Feb 2021 11:11:10 +0000 (12:11 +0100)]
docs: lex / yacc -> Lex / Yacc
These are names, so use proper nouns for them.
Acked-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9946>
Erik Faye-Lund [Wed, 3 Feb 2021 10:59:34 +0000 (11:59 +0100)]
docs: lod -> LOD
While we're at it, fix a few cases of incorrect usage of apostrophes.
Acked-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9946>
Erik Faye-Lund [Wed, 28 Oct 2020 14:07:05 +0000 (15:07 +0100)]
docs: nops -> NOPs
Acked-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9946>
Erik Faye-Lund [Wed, 3 Feb 2021 12:50:20 +0000 (13:50 +0100)]
docs: Xorg -> X.Org
Acked-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9946>
Erik Faye-Lund [Wed, 3 Feb 2021 10:55:51 +0000 (11:55 +0100)]
docs: opencl -> OpenCL
Acked-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9946>
Erik Faye-Lund [Wed, 28 Oct 2020 10:53:30 +0000 (11:53 +0100)]
docs: optimisation -> optimization
We generally prefer US English in the docs over UK English.
Acked-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8572>
Samuel Pitoiset [Tue, 30 Mar 2021 15:55:10 +0000 (17:55 +0200)]
radv: only set WRITE_COMPRESS_ENABLE for storage image descriptors
This will allow us to reduce the number of situations where the
compiler workaround is needed on GFX10.3.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9924>
Danylo Piliaiev [Wed, 24 Mar 2021 12:09:56 +0000 (14:09 +0200)]
ir3/isa: account for randomly set by blob lowest bit of ibo atomics
As far as I could see - blob randomly sets the lowest bit of atomic.b.*
instructions.
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9841>
Adam Jackson [Tue, 30 Mar 2021 14:58:05 +0000 (10:58 -0400)]
ci: Bump the llvmpipe test timeout to 240 seconds
lp_test_blend occasionally exceeds this on s390x, but we don't really
want to lose what little (hah) big-endian coverage we have.
Related: mesa/mesa#3437
Acked-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9921>