AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
154,669 Commits 1 Branch 0 Tags
7a088d3583db84fd73256bc3ff6c1660800812bf
Commit Graph

154669 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Erik Faye-Lund
7a088d3583 editorconfig: remove scons-ism 
Fixes: 6e6cd7d93c ("scons: Remove.")
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16786>
 2022-06-01 09:08:13 +00:00
Erik Faye-Lund
8901df12ab dzn: add debug option to redirect stdout/stderr 
For applications that doesn't have a terminal, it's useful to get output
like nir_log etc output somewhere.

This outputs these to stderr.txt and stdout.txt in the current user's
home directory, typically in C:\Users\my-user\.

Acked-by: Boris Brezillon <boris.brezillon@collabora.com>
Acked-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16640>
 2022-06-01 08:54:22 +00:00
Erik Faye-Lund
9f3a5b4313 dzn: add a debug-flag to wait for the debugger 
It's geneunely useful to wait for the debugger to attach in some cases.
So let's add a debug-flag and staill untik the debugger has attached.

Acked-by: Boris Brezillon <boris.brezillon@collabora.com>
Acked-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16640>
 2022-06-01 08:54:21 +00:00
Erik Faye-Lund
19044cf1de tests/graw: rename shaders from .sh to .txt 
The vertex and fragment shaders are named foo.sh, while the geometry
shaders are named foo.txt. Since .sh can easily be confused with shell
scripts, let's rename the vertex and fragment shaders to .txt to match
the geometry shaders.

These tests aren't hooked up to run automatically, so there's no need to
update references to the file names.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16791>
 2022-06-01 08:23:46 +00:00
Samuel Pitoiset
b968b99c45 radv: reduce radv_discard_rectangle_info::count to 8-bit 
DiscardRectangleCount must be less than or equal to
maxDiscardRectangles which is limited to 4.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16778>
 2022-06-01 07:59:27 +00:00
Samuel Pitoiset
b62f1cdee8 radv: reduce radv_rasterization_info::polygon_mode to 8-bit 
It's large enough.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16778>
 2022-06-01 07:59:27 +00:00
Samuel Pitoiset
38b6f2e0ce radv: reduce radv_input_assembly_info::primitive_topology to 8-bit 
It's large enough.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16778>
 2022-06-01 07:59:27 +00:00
Samuel Pitoiset
2f176f86c2 radv: reduce radv_tessellation_info::patch_control_points to 8-bit 
The value must be less than or equal to maxTessellationPatchSize which
is limited to 32 with RADV.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16778>
 2022-06-01 07:59:27 +00:00
Samuel Pitoiset
c869a5d550 radv: ignore DYNAMIC_STENCIL_OP if stencil test isn't enabled 
Like compare/write masks and reference.

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/16783>
 2022-06-01 07:34:28 +00:00
Iago Toral Quiroga
18985e8030 v3dv: use the global RCL EZ disable if we don't have any EZ draws in the job 
Until now we would only disable EZ globally if we had a depth or stencil
load operation or if we had no draw calls at all, but even if we have draw
calls if all of them disable EZ we should also us the global disable.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16794>
 2022-06-01 08:11:04 +02:00
Iago Toral Quiroga
0f65838933 v3dv: don't be too aggressive disabling early Z 
When we have a draw call that is incompatible with EZ we should only
disable EZ for the remaining of the job in the case that both of the
following conditions are met:

1. The cause for the incompatibility is an incompatible depth test
   direction.

2. The pipeline does Z writes.

Otherwise it is enough to disable EZ temporarily only for draw calls
with the incompatible pipeline.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16794>
 2022-06-01 08:10:57 +02:00
Martin Roukala (né Peres)
2f7f4060ee radv/ci: update the list of flakes 
This commit has been created by aggregating all the flakes found in the
last 2 weeks, excluding the ones seen once.

v2 (Martin Roukala):
 - stop trying to hardcode all the flakes, use a wildcard

Signed-off-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org>
Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16777>
 2022-06-01 05:26:12 +00:00
Lionel Landwerlin
5078b4fff1 nir/divergence: handle load_ray_num_dss_rt_stacks_intel 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16797>
 2022-06-01 04:58:50 +00:00
Lionel Landwerlin
d3c1b0ac28 nir/divergence: handle load_scratch_base_ptr 
v2: divergent (Jason)

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16797>
 2022-06-01 04:58:50 +00:00
Vinson Lee
6dfaedd607 d3d12: Remove unnecessary NULL check. 
Fix defect reported by Coverity Scan.

Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking pD3D12VideoBuffer suggests that it may
be null, but it has already been dereferenced on all paths leading to
the check.

Fixes: d8206f6286 ("d3d12: Add video decode implementation of pipe_video_codec")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16714>
 2022-06-01 04:46:27 +00:00
Connor Abbott
2e9c53549b tu: Expose VK_KHR_synchronization2 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16374>
 2022-06-01 04:06:46 +00:00
Connor Abbott
59259a0167 tu: Convert to sync2 entrypoints 
Use the common Vulkan fallbacks to implement the now-deprecated original
entrypoints.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16374>
 2022-06-01 04:06:46 +00:00
Timothy Arceri
abe4536c51 ci: uprev piglit 2022-05-31 
Also document additional piglit failures and passes.

Multiple changes, mostly notable:
  - few new tests
  - fixed test for upcoming mesa MR

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16785>
 2022-06-01 03:14:29 +00:00
Mike Blumenkrantz
d67a3ecc9a st/texture: allow compute blits for (some) non-finalized textures 
this yields sizable perf improvements in some cases

KHR-GL46.copy_image.functional timing (zink+anv-icl):

before
MESA_LOADER_DRIVER_OVERRIDE=zink ./glcts -n   74.77s user 76.44s system 33% cpu 7:32.38 total

after
MESA_LOADER_DRIVER_OVERRIDE=zink ./glcts -n   69.95s user 68.84s system 33% cpu 6:51.54 total

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16545>
 2022-06-01 02:50:32 +00:00
Mike Blumenkrantz
debb016d96 st/texture: use base teximage for compute blits 
this matches behavior of st_texture_image_map

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16545>
 2022-06-01 02:50:32 +00:00
Mike Blumenkrantz
9f7029169b zink: make swapchain kill message more descriptive 
cc: mesa-stable

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16776>
 2022-06-01 02:34:59 +00:00
Mike Blumenkrantz
964a651e6f zink: add a #define for descriptor compaction 
make those 2s a little less random

fixes #6556

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16755>
 2022-06-01 02:02:31 +00:00
Mike Blumenkrantz
517894e819 zink: flag all cached descriptors as needing update on program change 
programs have different usages of descriptors, so forcing a recalc on program
change ensures that the right hash values are always set, especially for compact
sets where there's more descriptors going into each hash value

this can (ideally) be optimized later to check for matching interfaces between old
program and new program to avoid recalc if both programs have identical descriptor
usage for a given set

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16755>
 2022-06-01 02:02:31 +00:00
Mike Blumenkrantz
9d34cab4a8 zink: fix change flagging for compact descriptor cache 
using pool existence alone is broken since the pools are compacted,
so instead flag based on whether usage exists

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16755>
 2022-06-01 02:02:31 +00:00
Mike Blumenkrantz
ddcd9c730c zink: store the real (non-compact) binding usage for programs 
this is useful for quickly determining whether a program has a certain
type of descriptor usage

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16755>
 2022-06-01 02:02:31 +00:00
Mike Blumenkrantz
41fce7ae57 zink: drop requirement for 6 descriptorsets to use bindless 
compact mode takes care of this

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16755>
 2022-06-01 02:02:31 +00:00
Mike Blumenkrantz
3304e5b5fb zink: only infer modifiers if winsys handle has a stride 
if no stride is passed, the image creation will fail, so avoid this case

fixes:
spec@ext_external_objects@vk-image-overwrite

Fixes: d79c716331 ("zink: create images with modifiers any time there is an import handle")

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16722>
 2022-06-01 01:48:26 +00:00
Jason Ekstrand
2a22885a45 st,nir: Use nir_shader::xfb_info in nir_lower_io_passes 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16750>
 2022-05-31 23:09:30 +00:00
Jason Ekstrand
16b0719441 glsl/nir: Stash the xfb_info in the nir_shader when linking XFB 
This pass is used for shaders coming in from SPIR-V.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16750>
 2022-05-31 23:09:30 +00:00
Jason Ekstrand
36d8a2f1d7 glsl/nir: Stop leaking varyings_info 
Fixes: 34b3b92bbe ("nir/xfb: move varyings info out of nir_xfb_info")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16750>
 2022-05-31 23:09:30 +00:00
Jason Ekstrand
7c5dc0b11a glsl/nir: Populate nir_shader::xfb_info after linking varyings 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16750>
 2022-05-31 23:09:30 +00:00
Jason Ekstrand
64cc35d2ac nir: Drop nir_shader_get_xfb_info 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16750>
 2022-05-31 23:09:30 +00:00
Jason Ekstrand
7003ff2e3a radv: Use nir_gather_xfb_info 
Instead of gathering XFB info over and over again every time we even
want to know if the shader uses XFB, gather it once when we compile the
shader and refer to the copy in the NIR.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16750>
 2022-05-31 23:09:30 +00:00
Jason Ekstrand
c8e191e369 turnip: Use nir_gather_xfb_info 
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Acked-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16750>
 2022-05-31 23:09:30 +00:00
Jason Ekstrand
0eab41559f lavapipe: Use nir_shader_gather_xfb_info 
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16750>
 2022-05-31 23:09:30 +00:00
Jason Ekstrand
e5ff2c2242 anv: Use nir_shader_gather_xfb_info 
Now that the resulting xfb_info is stashed on the shader, we can put
this with all the other NIR stuff and only fetch it out at the last
minute when we upload the kernel.

Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16750>
 2022-05-31 23:09:30 +00:00
Jason Ekstrand
23b55dcff4 nir: Add a nir_xfb_info to nir_shader 
We want to be able to carry this along with the shader instead of always
having to re-generate it from scratch.  A new nir_gather_xfb_info()
helper is also added which, instead of returning it, adds it to the
shader.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16750>
 2022-05-31 23:09:30 +00:00
Jason Ekstrand
3e04432b3a nir: Rename nir_gather_xfb_info to nir_shader_get_xfb_info 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16750>
 2022-05-31 23:09:30 +00:00
Jesse Natalie
f812cc0fe6 nir: Consider PNTC to be a varying 
Fixes: 3528dcdf ("nir: add nir_io_semantics::no_varying, no_sysval_output, and helpers")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6091
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16761>
 2022-05-31 20:51:22 +00:00
Mike Blumenkrantz
1c019ee1ba st/pbo_compute: fix z coords for compute pbos 
without manually taking the value from the global_id vec, this will
end up being offset.y again, which breaks z-indexing

Fixes: e7b9561959 ("gallium: implement compute pbo download")

Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16728>
 2022-05-31 20:05:49 +00:00
Mike Blumenkrantz
15f37d8d6f st/pbo_compute: make compute download conditional in shader slightly more readable 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16728>
 2022-05-31 20:05:49 +00:00
Mike Blumenkrantz
70fb3a4700 st/pbo_compute: use different calc for non-3d compute buffer sizing 
this avoids looking at irrelevant 3d pixelstore params like
GL_PACK_IMAGE_HEIGHT when they don't apply, which will cause the storage
buffer to be incorrectly sized and break the operation

Fixes: e7b9561959 ("gallium: implement compute pbo download")

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16728>
 2022-05-31 20:05:49 +00:00
Mike Blumenkrantz
5b02facc8c st/pbo_compute: do pbo readback directly to the buffer object if it exists 
no point in copying more than needed

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16802>
 2022-05-31 19:23:11 +00:00
Mike Blumenkrantz
4b5bb21318 zink: drop largePoints requirement 
this is not required by any version of GL, so don't pretend it's needed

cc: mesa-stable

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16758>
 2022-05-31 19:10:42 +00:00
Mike Blumenkrantz
63a9e0a986 zink: drop wideLines requirement 
this is not a requirement for any version of GL, so don't pretend it's
a requirement

cc: mesa-stable

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16758>
 2022-05-31 19:10:42 +00:00
Mike Blumenkrantz
9bb9511edc zink: remove first_frame stalling 
this is fixed by kopper

cc: mesa-stable

Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16796>
 2022-05-31 18:57:44 +00:00
Jesse Natalie
7f81333f71 CI: Re-enable Windows builds 
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16756>
 2022-05-31 18:32:59 +00:00
Jesse Natalie
cb2cc69ebb dzn: Update fails/flakes 
Acked-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16756>
 2022-05-31 18:32:59 +00:00
Jesse Natalie
b819f80d87 CI/windows: Move --fraction for deqp-runner to run command line 
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16756>
 2022-05-31 18:32:59 +00:00
Jesse Natalie
ad96c3079f CI/windows: Disable LLVM CMake developer warnings to shrink log size 
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16756>
 2022-05-31 18:32:59 +00:00