-
fa4796af
by Eric Engestrom at 2024-01-10T21:14:17+00:00
docs: add sha256sum for 23.3.3
-
15a237d7
by Eric Engestrom at 2024-01-12T18:00:41+00:00
.pick_status.json: Update to 68f5277887aae1cdc202f45ecd44df2c3c59ba7d
-
3dcea0be
by Pierre-Eric Pelloux-Prayer at 2024-01-12T18:00:52+00:00
ac/surface: don't oversize surf_size
Yet another iteration on the same YUV surfaces.
The change from 87ecfdfbf0a has 2 odd things:
* it's using MAX2(original value, new value) but the point of updating
surf_slice_size / surf_size is to make it correct relative to the new
value of surf_pitch
* it's multiplying surf_pitch (= number of elements per row) by height (ok)
by surf->bpe (= number of bytes per element) by surf->blk_w (= number of
"horizontal" pixels in an element) so the end unit doesn't make sense.
Fix this by computing a reasonnable value based on unit: the surf_slice_size
is the number of elements per row (surf_pitch) x number of bytes per element
(bpe) x number of rows.
This makes the expected size correct and thus fixes users of eglCreateImageKHR,
like the issue #6131.
I tested a bunch of gst pipelines and ffmpeg scripts on various files I have
and didn't notice any issues (on gfx10.3 and gfx9).
Fixes: 87ecfdfbf0a ("ac/surface: adapt surf_size when modifying surf_pitch")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6131
Acked-by: Marek Olšák <marek.olsak@amd.com>
Tested-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26693>
(cherry picked from commit 115b61e51f619df0b8d920b8ee572b56e7be575f)
-
7979ca07
by Pierre-Eric Pelloux-Prayer at 2024-01-12T18:00:55+00:00
radeonsi: compute epitch when modifying surf_pitch
In the linear case with no mipmaps addrlib sets epitch to surf_pitch - 1
so lets do the same thing here.
The change in si_descriptors.c looks like it's papering over a bug but I
couldn't find any other changes that wouldn't break at least one use case.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10375
Fixes: 115b61e51f6 ("ac/surface: don't oversize surf_size")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26947>
(cherry picked from commit 4e76c4ecb462ec6bc1b114b93161c7c732f3a65b)
-
6468166a
by Max R at 2024-01-12T18:03:07+00:00
d3d10umd: Fix compilation
Fixes: 4eb4c9bba ("d3d10umd: use cso_context to set vertex buffers and elements")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10054
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26370>
(cherry picked from commit 75fa621d34967773c0b9a40d29ba9b24d1c4bb6b)
-
761ef45a
by Matt Turner at 2024-01-12T18:03:09+00:00
symbols-check: Add _GLOBAL_OFFSET_TABLE_
This is exported on hppa/parisc.
See also: https://gitlab.freedesktop.org/glvnd/libglvnd/-/merge_requests/291
Cc: mesa-stable
Bug: https://bugs.gentoo.org/908079
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26978>
(cherry picked from commit 0ab7ea56b2558d30400a7462a05014e758c9c9c1)
-
ebfe4254
by Matt Turner at 2024-01-12T18:03:10+00:00
nir: Fix cast
We were wrongly telling `nir_const_value_as_uint()` that `iter` had
`bit_size` bits, but in one case it is explicitly i64. This works on
little endian platforms, but caused the nir_loop_unroll_test.fadd{,_rev}
tests to fail on big endian platforms.
Bug: https://bugs.gentoo.org/921297
Fixes: 268ad47c111 ("nir/loop_analyze: Handle bit sizes correctly in calculate_iterations")
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26964>
(cherry picked from commit 5997cf7587ce56aedac9114c0db9b250f1b54461)
-
62a8493a
by Samuel Pitoiset at 2024-01-12T18:03:11+00:00
radv: do not issue SQTT marker with DISPATCH_MESH_INDIRECT_MULTI
According to PAL, only DISPATCH_TASKMESH_GFX is supposed to emit a
SQTT marker as part of the packet, probably because there is also
a packet emitted on ACE for executing task shaders.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10401
Fixes: 312103e0ffb ("radv: set THREAD_TRACE_MARKER_ENABLE for mesh/task draws")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26936>
(cherry picked from commit 3f655bc47c388d9fb37d4569be7bd79208820a6a)
-
4b10fa97
by Mike Blumenkrantz at 2024-01-12T18:03:12+00:00
lavapipe: fix devenv icd filename
fixes #10408
cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26985>
(cherry picked from commit 465e26dd9806728f4689c56097d80d6c04f610f6)
-
5361f9c7
by Pavel Ondračka at 2024-01-12T18:06:30+00:00
r300: fix reusing of color varying slots for generic ones
This was broken when I added texcoord support, the problem is that we
failed to properly count the number of used fs inputs and thus we failed
to make the proper decision when to reuse the color varying slot
Also fix the error messages, they were incorrect after the rewrite as
well. This fixes a bunch of piglits.
Fixes: d4b8e8a48144f4b899d48c271558f0dc613632cb
Signed-off-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Reviewed-by: Filip Gawin <filip.gawin@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27003>
(cherry picked from commit 53c17d85abad095c9d381785d29531e8a4532ffc)
-
d85271f3
by Timur Kristóf at 2024-01-12T18:21:23+00:00
radv: Correctly select SDMA support for PRIME blit.
Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10317
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27015>
(cherry picked from commit 436b89e8388810dc456c366b8ab0ee03f8fb1357)
-
572e3a1b
by Lionel Landwerlin at 2024-01-12T18:21:25+00:00
anv: fix disabled Wa_14017076903/18022508906
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: d0669f3ede ("intel/dev: switch defect identifiers to use lineage numbers")
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27008>
(cherry picked from commit 695b4a299250ee7ff5da5f7eda70c6b245dd0f78)
-
35b321e9
by Mike Blumenkrantz at 2024-01-12T18:44:27+00:00
zink: always force flushes when originating from api frontend
flags=0 is used for e.g., glFenceSync, which apps use to insert sync points
to determine when all prior work has completed. eliding these flushes into no-ops
is fine for all scenarios except when the last op was a present, in which
case the no-op (previous) fence will not sync as expected for the present and
graphical artifacts will result
in the future, this may be changed back to the previous behavior if/when presentation
gains timeline semaphore capabilities by providing the last timeline id
as a fence instead of the last batch
fixes #10386
cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26935>
(cherry picked from commit 03f049f49730b8b1268f2975fc34ac2b1f9a4ef9)
-
5aab597e
by Mike Blumenkrantz at 2024-01-12T18:44:50+00:00
zink: ignore tc buffer replacement info
without tc tracking cross-context buffer usage, this is intensely broken
Fixes: 96cf4531e11 ("Revert "gallium/u_threaded: buffer sharedness tracking"")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26959>
(cherry picked from commit 0444d057ae6b0765cefa760845bed488396f9fde)
-
92f39d1c
by Mike Blumenkrantz at 2024-01-12T18:44:53+00:00
zink: fix buffer rebind early-out check
this was accidentally inverted; the rebind attempt is over if the
number of enacted rebinds >= the expected rebinds
Fixes: c32bcb9a8c2 ("zink: improve handling of buffer rebinds using tc info")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26959>
(cherry picked from commit 6e4d901a2eeb3ede16fc423c6347fca2442ea811)
-
62161b74
by Alessandro Astone at 2024-01-12T18:44:53+00:00
zink: Fix resizable BAR detection logic
This was broken in two ways:
* When looking for the MAX biggest_ram it was actually comparing
a candidate against biggest_vis_ram
* mem_props.memoryTypes[] should be accessed with the memory type
index as found in heap_map
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10341
Cc: 23.3 <mesa-stable>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26979>
(cherry picked from commit a077c14f150f1c4f670dce381ac2eb548f1a4ac2)
-
151cd31e
by Mike Blumenkrantz at 2024-01-12T18:44:54+00:00
zink: fix separate shader patch variable location adjustment
in spirv, these start at location 0, not location 32
fixes #10414
Fixes: d9942442f20 ("zink: handle patch variable locations for separate shaders better")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26981>
(cherry picked from commit 565ee4fafc9490a4ee7eec117123444843376d1c)
-
0c942965
by Eric Engestrom at 2024-01-15T09:43:40+00:00
.pick_status.json: Update to 4fe5f06d400a7310ffc280761c27b036aec86646
-
c24841d1
by Matt Turner at 2024-01-15T09:43:50+00:00
util: Add DETECT_ARCH_HPPA macro
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26991>
(cherry picked from commit 0540c9de447730e5efe73a0c5a1a5b6c1e902722)
-
685cc5f6
by Matt Turner at 2024-01-15T09:43:54+00:00
util/tests: Disable half-float NaN test on hppa/old-mips
Bug: https://bugs.gentoo.org/908079
Fixes: 067023dce2c ("util: Add some unit tests of the half-float conversions.")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26991>
(cherry picked from commit 5b7c73390247caa847c56c442298107a1e568a6d)
-
4c37d02f
by Jesse Natalie at 2024-01-15T22:18:42+00:00
mesa: Consider mesa format in addition to internal format for mip/cube completeness
Prior to 06b526de, the mesa format was used for these completeness checks.
That was to address the case where a *different* internal format selected
the *same* mesa format, and the texture shouldn't be considered compatible.
But this didn't address the case where the *same* internal format selected
a *different* mesa format, e.g. because the type passed to the TexImage
API was different.
An old WGL demo app called TexFilter.exe tries to redefine a mipped RGBA16
texture as RGBA8. This incorrect logic caused Mesa to try to copy the RGBA16
data from the smaller mips into the newly created RGBA8 data, because it
thought that the texture was still mip-complete, despite the format changing.
Cc: mesa-stable
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27023>
(cherry picked from commit 4cb9c77e8e08507b5c181a480259e42b43dd647e)
-
5d7b3812
by Lionel Landwerlin at 2024-01-15T22:19:04+00:00
anv: hide vendor ID for The Finals
XeSS workaround.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10436
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27057>
(cherry picked from commit a34a113059f55947cc08624897999f7f066f000a)
-
2f0a118a
by Tapani Pälli at 2024-01-15T22:20:07+00:00
anv: check for wa 16013994831 in emit_so_memcpy_end
We are toggling preemption on/off during streamout, this is also
happening on gfx12 platforms, not just dg2.
Cc: mesa-stable
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27002>
(cherry picked from commit 36f428f1de78d6bd2c0aa6719da06cd5233a8c7f)
-
8f69ee06
by Eric Engestrom at 2024-01-16T18:42:33+00:00
.pick_status.json: Update to ff84aef116f9d0d13440fd13edf2ac0b69a8c132
-
fc920acf
by Lucas Stach at 2024-01-16T18:42:45+00:00
etnaviv: disable 64bpp render/sampler formats
Vivante hardware handles 64bpp render targets and samplers in a odd way
by splitting the buffer and using a pair of texture samplers or a pair
of MRT outputs to access those resources. This isn't implemented in the
driver right now, so we should not advertise support for those formats.
CC: mesa-stable
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <cgmeiner@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26982>
(cherry picked from commit e481c1269c36efae6fad9e3c60af9c66cc8bbf74)
-
a34a657b
by Tatsuyuki Ishi at 2024-01-16T18:42:46+00:00
radv: never set DISABLE_WR_CONFIRM for CP DMA clears and copies
This mirrors the changes in 69ff9c16bbb ("radeonsi: never set
DISABLE_WR_CONFIRM for CP DMA clears and copies").
Cc: mesa-stable
Suggested-by: Vitaliy Triang3l Kuzmin <triang3l@yandex.ru>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27053>
(cherry picked from commit 43fb43ba2cfe673d5b6693bfe93d0331f86817ed)
-
3cc00515
by Karol Herbst at 2024-01-16T18:42:47+00:00
rusticl/kernel: run opt/lower_memcpy later to fix a crash
nir_opt_memcpy requires explicit types to function properly. So run them
after lowering vars to explicit types.
Cc: mesa-stable
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27068>
(cherry picked from commit f8966598940ad46fb1ff2cbd9c23013289ef0736)
-
cd711b42
by Patrick Lerda at 2024-01-16T18:42:47+00:00
glsl/nir: fix gl_nir_cross_validate_outputs_to_inputs() memory leak
For instance, this issue is triggered with
vs-to-fs-overlap.shader_test -auto -fbo:
Direct leak of 24 byte(s) in 1 object(s) allocated from:
#0 0x7fe64f58e9a7 in calloc (/usr/lib64/libasan.so.6+0xb19a7)
#1 0x7fe642ca2839 in _mesa_symbol_table_ctor ../src/mesa/program/symbol_table.c:286
#2 0x7fe642ff003d in gl_nir_cross_validate_outputs_to_inputs ../src/compiler/glsl/gl_nir_link_varyings.c:728
#3 0x7fe642d7c7d8 in gl_nir_link_glsl ../src/compiler/glsl/gl_nir_linker.c:1357
#4 0x7fe642be6931 in st_link_glsl_to_nir ../src/mesa/state_tracker/st_glsl_to_nir.cpp:562
#5 0x7fe642be6931 in st_link_shader ../src/mesa/state_tracker/st_glsl_to_nir.cpp:944
#6 0x7fe642acab55 in link_program ../src/mesa/main/shaderapi.c:1336
#7 0x7fe642acab55 in link_program_error ../src/mesa/main/shaderapi.c:1447
#8 0x7fe6424aa389 in _mesa_unmarshal_LinkProgram src/mapi/glapi/gen/marshal_generated2.c:1911
#9 0x7fe641fd912b in glthread_unmarshal_batch ../src/mesa/main/glthread.c:139
#10 0x7fe641f48d48 in util_queue_thread_func ../src/util/u_queue.c:309
#11 0x7fe641fa442a in impl_thrd_routine ../src/c11/impl/threads_posix.c:67
Fixes: 7d1948e9b5d9 ("glsl: implement cross_validate_outputs_to_inputs() in nir linker")
Signed-off-by: Patrick Lerda <patrick9876@free.fr>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27071>
(cherry picked from commit bacace8634346f853547f51a0ea6ff8082a8dcb8)
-
077c8540
by David Rosca at 2024-01-16T18:42:48+00:00
radeonsi/vcn: Fix H264 slice header when encoding I frames
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27029>
(cherry picked from commit 865abfde6393689d17d33fb907cc39d63ef02df4)
-
c12de84e
by Eric Engestrom at 2024-01-17T22:18:21+00:00
.pick_status.json: Update to 6e4bb8253ed36f911a0a45dfecf89c237a8cd362
-
0946a4bf
by Yiwei Zhang at 2024-01-17T22:18:25+00:00
vulkan/wsi/wayland: fix returns and avoid leaks for failed swapchain
Cc: mesa-stable
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Tested-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27080>
(cherry picked from commit dc5725ee29727a3272ecf30141a249cf4f91f1fc)
-
46bafee3
by Yonggang Luo at 2024-01-17T22:18:27+00:00
compiler/spirv: The spirv shader is binary, should write in binary mode
Fixes: 53265c8798e ("spirv: Add a mechanism for dumping failing shaders")
Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26775>
(cherry picked from commit fd118188285d0f3b2d963b0a4807f3f4cfd0efbb)
-
13ceea55
by Sviatoslav Peleshko at 2024-01-17T22:18:28+00:00
nir: Use alu source components count in nir_alu_srcs_negative_equal
When we use source from ALU instruction directly, the default swizzle array
should be populated with the same amount of components as the src has.
Otherwise, if we use nir_ssa_alu_instr_src_components, it can return
the destination components count that is lower than component index
actually used in that source. This can lead to false equality
between 0 (uninitialized) and 0 (.x) in swizzle comparison below.
Fixes: c6ee46a7 ("nir: Add nir_alu_srcs_negative_equal")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8704
Signed-off-by: Sviatoslav Peleshko <sviatoslav.peleshko@globallogic.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22655>
(cherry picked from commit 6b0bfdfa9e83aba5316821296fb9cf68fdc958f8)
-
6fb9f61f
by Boris Brezillon at 2024-01-17T22:18:29+00:00
panvk: Fix tracing
pandecode_next_frame() take a decode context. Passing NULL leads to a
NULL deref.
Fixes: 56be9a55be03 ("pan/decode: handle more than one panfrost_device")
Cc: mesa-stable
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Constantine Shablya <constantine.shablya@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27107>
(cherry picked from commit 35a02560c8932d4b1841772b016b8e672e409eda)
-
ae7d41bd
by Boris Brezillon at 2024-01-17T22:18:30+00:00
panvk: Fix access to unitialized panvk_pipeline_layout::num_sets field
Commit 73eecffabdd3 ("panvk: Use the vk_pipeline_layout base struct")
reworked the panvk logic to use vk_pipeline_layout, which contains the
number of descriptor set layout referenced by a pipeline layout, thus
deprecating panvk_pipeline_layout::num_sets.
Make panvk_fill_non_vs_attribs() use vk_pipeline_layout::set_count
instead of panvk_pipeline_layout::num_sets and kill the latter so we
can't introduce new users.
Fixes: 73eecffabdd3 ("panvk: Use the vk_pipeline_layout base struct")
Cc: mesa-stable
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Constantine Shablya <constantine.shablya@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27107>
(cherry picked from commit b18bfed2c5fc9c8ad93b08bae8b9e65380a1b772)
-
dd6e3a1a
by Friedrich Vock at 2024-01-17T22:25:49+00:00
radv/rt: Add workaround to make leaves always active
DOOM Eternal builds acceleration structures with inactive primitives and
tries to make them active in later AS updates. This is disallowed by the
spec and triggers a GPU hang. Fix the hang by working around the bug.
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27034>
(cherry picked from commit a9831caa144f9944fec936608faf03d253e9bb7d)
-
01b374ec
by Eric Engestrom at 2024-01-17T23:28:12+00:00
ci/deqp: ensure that in `default` builds, wayland + x11 + xcb are all built
If someone were to remove the libraries that are needed for these,
`default` would simply not enable these tests, and the only thing we
could notice is that test jobs would suddenly take less time to run.
Instead, let's have a check to make sure dEQP's cmake has detected
everything and enabled these platforms.
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27041>
(cherry picked from commit 27a1b4e4f314832c164380ea332c096fe394c8f0)
-
a062b043
by David Heidelberg at 2024-01-17T23:29:18+00:00
ci/deqp: uprev deqp-runner for Linux too to 0.18.0
Previous commit upreved deqp only for the Android
Fixes: 1ff4687e8660 ("ci: uprev deqp-runner from 0.16.1 to 0.18.0")
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
[Eric]
- rename the deqp-runner version to DEQP_RUNNER_VERSION instead of DEQP_VERSION
- update image tags
- fix expectations lists
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27062>
(cherry picked from commit 4ff77f08e476e37bf387178b5151093296491016)
-
22c416e1
by Eric Engestrom at 2024-01-18T13:07:49+00:00
.pick_status.json: Update to d2b08f9437f692f6ff4be2512967973f18796cb2
-
ed75400a
by Ryan Neph at 2024-01-18T13:07:51+00:00
venus: fix shmem leak on vn_ring_destroy
Missed shmem unref when moving ring internals out of vn_instance.c.
Fixes: d1e29b75578 ("venus: move ring shmem into vn_ring")
Signed-off-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27125>
(cherry picked from commit 6e4bb8253ed36f911a0a45dfecf89c237a8cd362)
-
2ff92193
by Faith Ekstrand at 2024-01-18T13:07:54+00:00
nvk: Unref shaders on pipeline free
Fixes: d6a1e29ccdc1 ("nvk: pipeline shader cache")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27130>
(cherry picked from commit be0f04f5bd3155c05af14b2c8ffee887838af807)
-
74ee323c
by Dave Airlie at 2024-01-18T13:07:56+00:00
radv/video: refactor sq start/end code to avoid decode hangs.
The extra cmd buffer layer was done wrong, need to emit the
sq start and ends around every reset/decode packet.
Fixes dEQP-VK.video.decode.h264_i on navi3x
Fixes: d8f3060bd915 ("radv/video: start adding gfx11 vcn decoder")
Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25932>
(cherry picked from commit d32f2ee7b632b87c9ea8de66aa41423bf36f8268)
-
085612fc
by Dave Airlie at 2024-01-18T13:07:57+00:00
radv: don't submit empty command buffers on encoder ring.
the vcn enc/unified rings don't do nop packets, and hang with 0 sized
cmd buffers. This just stops submitting 0 sized cmd buffers to the hw.
Fixes hangs with dEQP-VK.video.decode.h264_i on navi3x
Cc: mesa-stable
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25932>
(cherry picked from commit f33683e4dad9e1dfb7dcd4f86bb86ef3e3954315)
-
edc8f709
by Eric Engestrom at 2024-01-23T13:18:32+00:00
.pick_status.json: Update to d0a3bac163ca803eda03feb3afea80e516568caf
-
bc9a92aa
by Eric Engestrom at 2024-01-23T13:18:34+00:00
.pick_status.json: Update to d0a3bac163ca803eda03feb3afea80e516568caf
-
ee3ab325
by Dave Airlie at 2024-01-23T13:18:48+00:00
radv/video: refactor sq start/end code to avoid decode hangs.
The extra cmd buffer layer was done wrong, need to emit the
sq start and ends around every reset/decode packet.
Fixes dEQP-VK.video.decode.h264_i on navi3x
Fixes: d8f3060bd915 ("radv/video: start adding gfx11 vcn decoder")
Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25932>
(cherry picked from commit d32f2ee7b632b87c9ea8de66aa41423bf36f8268)
-
9b6bce4b
by Dave Airlie at 2024-01-23T13:18:49+00:00
radv: don't submit empty command buffers on encoder ring.
the vcn enc/unified rings don't do nop packets, and hang with 0 sized
cmd buffers. This just stops submitting 0 sized cmd buffers to the hw.
Fixes hangs with dEQP-VK.video.decode.h264_i on navi3x
Cc: mesa-stable
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25932>
(cherry picked from commit f33683e4dad9e1dfb7dcd4f86bb86ef3e3954315)
-
e0790e87
by Hans-Kristian Arntzen at 2024-01-23T13:18:50+00:00
wsi/x11: Add workaround for Detroit Become Human.
Game needs strict image count to not crash in non-vsync mode.
Signed-off-by: Hans-Kristian Arntzen <post@arntzen-software.no>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27038>
(cherry picked from commit efc0131d5bf42c7671b6ebcda61de06c9b954b11)
-
ba54cfa0
by Hans-Kristian Arntzen at 2024-01-23T13:19:02+00:00
wsi/x11: Add workaround for Detroit Become Human.
Game needs strict image count to not crash in non-vsync mode.
Signed-off-by: Hans-Kristian Arntzen <post@arntzen-software.no>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27038>
(cherry picked from commit efc0131d5bf42c7671b6ebcda61de06c9b954b11)
-
219cd6dc
by Ian Romanick at 2024-01-23T13:19:03+00:00
intel/compiler: Disable DPAS instructions on MTL
Reviewed-by: Mark Janes <markjanes@swizzler.org>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 3756f605586 ("intel/fs: DPAS lowering")
Closes: #10376
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26993>
(cherry picked from commit 951e08fc18a32f8a5ee9faa39cab69f8f0767e24)
-
057493ce
by Ian Romanick at 2024-01-23T13:19:04+00:00
intel/compiler: Track lower_dpas flag in brw_get_compiler_config_value
This user-settable flag affects compiler output, so it should be tracked
in the cache hash.
Fixes: 3756f605586 ("intel/fs: DPAS lowering")
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Suggested-by: Lionel Landwerlin
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26993>
(cherry picked from commit 6f237a23c771e3dc74adc1cc0ab5cbc3e3b03be8)
-
b4285304
by Ian Romanick at 2024-01-23T13:19:05+00:00
intel/compiler: Track mue_compaction and mue_header_packing flags in brw_get_compiler_config_value
v2: Use u_foreach_bit64. Suggested by Lionel.
Fixes: 48885c7fe34 ("intel/compiler: load debug mesh compaction options once")
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26993>
(cherry picked from commit 7481d61a5d5a8bef71c855182f1d958c4e6c4f0f)
-
69cac7ae
by Tapani Pälli at 2024-01-23T13:19:06+00:00
iris: expand pre-hiz data cache flush to gfx >= 125
Cc: mesa-stable
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27132>
(cherry picked from commit 93706d5c2fb8cd47fa444fd8598b5cb190cf74a7)
-
cfa818f1
by Tapani Pälli at 2024-01-23T13:19:07+00:00
anv: expand pre-hiz data cache flush to gfx >= 125
Cc: mesa-stable
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27132>
(cherry picked from commit 02d7f5e4ff1a93d149778d8b40f327cccbb412ee)
-
484a051a
by Konstantin Seurer at 2024-01-23T13:19:58+00:00
ac/llvm: Enable helper invocations for quad OPs
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9239
cc: mesa-stable
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27110>
(cherry picked from commit 220c91208037a499ff7a553f263d20e5844094a4)
-
6d1dae87
by Konstantin Seurer at 2024-01-23T13:19:59+00:00
lavapipe: Fix DGC vertex buffer handling
Fixes: 976dd26 ("lavapipe: NV_device_generated_commands")
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27019>
(cherry picked from commit 6d88c1bb6ce7b64188ff1a19238a40d9218fd021)
-
83b5a3a3
by Konstantin Seurer at 2024-01-23T13:20:00+00:00
lavapipe: Mark vertex elements dirty if the stride changed
Fixes: 7672545 ("gallium: move vertex stride to CSO")
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27019>
(cherry picked from commit cc94ff081c9b431dbb1242a04bd4efe0feaf5dde)
-
364835c5
by Konstantin Seurer at 2024-01-23T13:20:01+00:00
lavapipe: Report the correct preprocess buffer size
There can be multiple sequences.
Fixes: 976dd26 ("lavapipe: NV_device_generated_commands")
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27019>
(cherry picked from commit 024f144165f7b53e78bdd18f8b5afe11bf4e36e1)
-
0bdce868
by Ian Romanick at 2024-01-23T13:20:51+00:00
intel/compiler: Track mue_compaction and mue_header_packing flags in brw_get_compiler_config_value
v2: Use u_foreach_bit64. Suggested by Lionel.
Fixes: 48885c7fe34 ("intel/compiler: load debug mesh compaction options once")
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26993>
(cherry picked from commit 7481d61a5d5a8bef71c855182f1d958c4e6c4f0f)
-
0392e4bf
by Friedrich Vock at 2024-01-23T13:21:07+00:00
radv: Fix shader replay allocation condition
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26891>
(cherry picked from commit 43bdfebbff2c5557355df88816e737ef19618ed2)
-
e1d20b69
by Daniel Schürmann at 2024-01-23T13:21:08+00:00
aco: give spiller more room to assign spilled SGPRs to VGPRs
On chordal graphs, a greedy coloring can be done in a way that never uses
more colors than are required for the largest clique. However, since we
have vector values and force phi resources into the same spill slots, the
interference graphs are not chordal, and thus, this assumption doesn't hold.
Use twice as many spill slots as upper bound.
Totals from 10 (0.01% of 79242) affected shaders: (GFX11)
MaxWaves: 52 -> 54 (+3.85%)
Instrs: 271386 -> 271779 (+0.14%)
CodeSize: 1362544 -> 1365432 (+0.21%)
VGPRs: 2536 -> 2532 (-0.16%)
SpillVGPRs: 778 -> 818 (+5.14%)
Scratch: 73472 -> 76800 (+4.53%)
Latency: 3331718 -> 3328798 (-0.09%); split: -0.14%, +0.05%
InvThroughput: 1665860 -> 1643350 (-1.35%); split: -1.40%, +0.05%
VClause: 3292 -> 3329 (+1.12%); split: -0.06%, +1.18%
Copies: 46082 -> 46257 (+0.38%)
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27011>
(cherry picked from commit e3098bb23284b598ec6d5030ceaa1c3d5bd9d428)
-
ebd56d7a
by Georg Lehmann at 2024-01-23T13:21:08+00:00
aco: stop scheduling at p_logical_end
No Foz-DB changes, but this fixes some issues when the spiller inserts
scratch loads after p_logical_end for p_return.
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27119>
(cherry picked from commit 74fc2e287fc79b9451919b21957bc5d0ef186a5a)
-
21d22653
by Samuel Pitoiset at 2024-01-23T13:21:09+00:00
radv: fix indirect dispatches on the compute queue on GFX7
GFX7 CP requires the indirect dispatch VA to be aligned to 32-bytes.
This fixes dEQP-VK.api.command_buffers.many_indirect_disps_on_secondary,
but it's unexpected that it uncovered this bug.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27148>
(cherry picked from commit 5c03cdbd02a69884ce759e0cbd0cf76dc212e2d3)
-
812bcc29
by Samuel Pitoiset at 2024-01-23T13:21:10+00:00
radv: fix indirect draws with NULL index buffer on GFX10
GFX10 has a hw bug and it can't handle 0-sized index buffer. The
non-indirect draw path was fine but not the indirect path where RADV
emits the index buffer.
This fixes flakes with dEQP-VK.*maintenance6* on NAVI14, and possibly
GPU hangs if there is an indirect draw with a valid index buffer right
before because it would re-use the same index buffer.
Fixes: db9816fd666 ("radv: add support for NULL index buffer")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27142>
(cherry picked from commit 783e3c096fe34b06a251b3355330feac3a015e4e)
-
2e4623bd
by Eric R. Smith at 2024-01-23T13:21:14+00:00
panfrost: fix panfrost drm-shim
The panfrost driver now makes an ioctl to retrieve some new memory
parameters, and DRM_PANFROST_PARAM_MEM_FEATURES is required (does not
default in the caller). This caused drm-shim to stop working. This
patch adds some defaults to get drm-shim working again.
Signed-off-by: Eric R. Smith <eric.smith@collabora.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Fixes: 91fe8a0d2859 ("panfrost: Back panfrost_device with pan_kmod_dev object")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27162>
(cherry picked from commit a50b2f8f258eb71984a3d63ca031b8051c380344)
-
8039f6a5
by Sil Vilerino at 2024-01-23T13:21:25+00:00
d3d12: Implement cap for PIPE_VIDEO_CAP_ENC_INTRA_REFRESH
Fixes: c81967fa89d ("d3d12: Implement Intra Refresh for H264, HEVC, AV1")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27201>
(cherry picked from commit a3c91624f487ee7676dfce0f8ed4e4c87ce2c0a2)
-
95167b21
by Yiwei Zhang at 2024-01-23T13:21:27+00:00
venus: fix pipeline layout lifetime
Should check the count instead of random ptr addr.
Fixes: 19f2b9d0bbd ("venus: extend VkPipelineLayout lifetime for ...")
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27054>
(cherry picked from commit b551b6e48a4b69e0b1b6eb36acfbebe359025c24)
-
252a87e7
by Yiwei Zhang at 2024-01-23T13:21:28+00:00
venus: fix pipeline derivatives
This was unexpected dropped in the initial GPL impl.
Fixes: a65ac274acf ("venus: Do pipeline fixes for VK_EXT_graphics_pipeline_library")
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27054>
(cherry picked from commit f713b17a16d7899ef78f85b91c14a46b9f91b3c8)
-
bfa31de5
by Yiwei Zhang at 2024-01-23T13:21:29+00:00
venus: fix to respect the final pipeline layout
This fixes VUID-vkCmdDraw-None-08600 violation when running gpl cts:
dEQP-VK...graphics_library.misc.bind_null_descriptor_set.*, where the
final pipeline layout is falsely dropped, leading to incompatible with
the pipeline layout of the bound descriptor set.
Fixes: a65ac274acf ("venus: Do pipeline fixes for VK_EXT_graphics_pipeline_library")
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27054>
(cherry picked from commit 80a5df16fe81d39fd771d5805b141c4eff623886)
-
296e6d31
by Tapani Pälli at 2024-01-23T13:21:39+00:00
iris: expand pre-hiz data cache flush to gfx >= 125
Cc: mesa-stable
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27132>
(cherry picked from commit 93706d5c2fb8cd47fa444fd8598b5cb190cf74a7)
-
f948ccf7
by Tapani Pälli at 2024-01-23T13:21:40+00:00
anv: expand pre-hiz data cache flush to gfx >= 125
Cc: mesa-stable
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27132>
(cherry picked from commit 02d7f5e4ff1a93d149778d8b40f327cccbb412ee)
-
13b0648c
by Konstantin Seurer at 2024-01-23T13:21:41+00:00
ac/llvm: Enable helper invocations for quad OPs
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9239
cc: mesa-stable
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27110>
(cherry picked from commit 220c91208037a499ff7a553f263d20e5844094a4)
-
596cfa44
by Konstantin Seurer at 2024-01-23T13:21:42+00:00
lavapipe: Fix DGC vertex buffer handling
Fixes: 976dd26 ("lavapipe: NV_device_generated_commands")
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27019>
(cherry picked from commit 6d88c1bb6ce7b64188ff1a19238a40d9218fd021)
-
b3918fe5
by Konstantin Seurer at 2024-01-23T13:21:43+00:00
lavapipe: Mark vertex elements dirty if the stride changed
Fixes: 7672545 ("gallium: move vertex stride to CSO")
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27019>
(cherry picked from commit cc94ff081c9b431dbb1242a04bd4efe0feaf5dde)
-
a26d09b4
by Konstantin Seurer at 2024-01-23T13:21:44+00:00
lavapipe: Report the correct preprocess buffer size
There can be multiple sequences.
Fixes: 976dd26 ("lavapipe: NV_device_generated_commands")
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27019>
(cherry picked from commit 024f144165f7b53e78bdd18f8b5afe11bf4e36e1)
-
5d602c5e
by Friedrich Vock at 2024-01-23T13:21:45+00:00
radv: Fix shader replay allocation condition
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26891>
(cherry picked from commit 43bdfebbff2c5557355df88816e737ef19618ed2)
-
faa72750
by Daniel Schürmann at 2024-01-23T13:21:46+00:00
aco: give spiller more room to assign spilled SGPRs to VGPRs
On chordal graphs, a greedy coloring can be done in a way that never uses
more colors than are required for the largest clique. However, since we
have vector values and force phi resources into the same spill slots, the
interference graphs are not chordal, and thus, this assumption doesn't hold.
Use twice as many spill slots as upper bound.
Totals from 10 (0.01% of 79242) affected shaders: (GFX11)
MaxWaves: 52 -> 54 (+3.85%)
Instrs: 271386 -> 271779 (+0.14%)
CodeSize: 1362544 -> 1365432 (+0.21%)
VGPRs: 2536 -> 2532 (-0.16%)
SpillVGPRs: 778 -> 818 (+5.14%)
Scratch: 73472 -> 76800 (+4.53%)
Latency: 3331718 -> 3328798 (-0.09%); split: -0.14%, +0.05%
InvThroughput: 1665860 -> 1643350 (-1.35%); split: -1.40%, +0.05%
VClause: 3292 -> 3329 (+1.12%); split: -0.06%, +1.18%
Copies: 46082 -> 46257 (+0.38%)
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27011>
(cherry picked from commit e3098bb23284b598ec6d5030ceaa1c3d5bd9d428)
-
c529e4af
by Georg Lehmann at 2024-01-23T13:21:48+00:00
aco: stop scheduling at p_logical_end
No Foz-DB changes, but this fixes some issues when the spiller inserts
scratch loads after p_logical_end for p_return.
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27119>
(cherry picked from commit 74fc2e287fc79b9451919b21957bc5d0ef186a5a)
-
535cef74
by Samuel Pitoiset at 2024-01-23T13:21:49+00:00
radv: fix indirect dispatches on the compute queue on GFX7
GFX7 CP requires the indirect dispatch VA to be aligned to 32-bytes.
This fixes dEQP-VK.api.command_buffers.many_indirect_disps_on_secondary,
but it's unexpected that it uncovered this bug.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27148>
(cherry picked from commit 5c03cdbd02a69884ce759e0cbd0cf76dc212e2d3)
-
e99d28b4
by Lionel Landwerlin at 2024-01-23T19:49:06+00:00
anv: fix pipeline executable properties with graphics libraries
We're missing the ISA code in renderdoc. You can reproduce with the
Sascha Willems graphics pipeline demo.
The change is large here because we have to fix a confusion between
anv_shader_bin & anv_pipeline_executable. anv_pipeline_executable is
there as a representation for the user and multiple
anv_pipeline_executable can point to a single anv_shader_bin.
In this change we split the anv_shader_bin related logic that was
added in anv_pipeline_add_executable*() and move it to a new
anv_pipeline_account_shader() function.
When importing RT libraries, we add all the anv_pipeline_executable
from the libraries.
When importing Gfx libraries, we add the anv_pipeline_executable only
if not doing link time optimization.
anv_shader_bin related properties are added whenever we're importing a
shader from a library, compiling or finding in the cache.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 3d49cdb71e ("anv: implement VK_EXT_graphics_pipeline_library")
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26594>
(cherry picked from commit 58c9f817cbed2fc5263c86b514730e54a2d9cbe4)
-
72d36448
by Lionel Landwerlin at 2024-01-23T19:49:07+00:00
anv: implement undocumented tile cache flush requirements
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27169>
(cherry picked from commit ba87656079a7fb745c06e78641d2fa6ac4112b82)
-
7af4d666
by Tapani Pälli at 2024-01-23T19:49:09+00:00
iris: replace constant cache invalidate with hdc flush
This implements Wa_14010840176.
Cc: mesa-stable
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21364>
(cherry picked from commit 231ede4f0ca92b33da8b16e7921af9877379c7d1)
-
725541c3
by Yiwei Zhang at 2024-01-23T19:49:26+00:00
venus: fix pipeline layout lifetime
Should check the count instead of random ptr addr.
Fixes: 19f2b9d0bbd ("venus: extend VkPipelineLayout lifetime for ...")
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27054>
(cherry picked from commit b551b6e48a4b69e0b1b6eb36acfbebe359025c24)
-
287e146d
by Yiwei Zhang at 2024-01-23T20:14:57+00:00
venus: fix pipeline derivatives
This was unexpected dropped in the initial GPL impl.
Fixes: a65ac274acf ("venus: Do pipeline fixes for VK_EXT_graphics_pipeline_library")
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27054>
(cherry picked from commit f713b17a16d7899ef78f85b91c14a46b9f91b3c8)
-
1f89910f
by Yiwei Zhang at 2024-01-23T20:14:57+00:00
venus: fix to respect the final pipeline layout
This fixes VUID-vkCmdDraw-None-08600 violation when running gpl cts:
dEQP-VK...graphics_library.misc.bind_null_descriptor_set.*, where the
final pipeline layout is falsely dropped, leading to incompatible with
the pipeline layout of the bound descriptor set.
Fixes: a65ac274acf ("venus: Do pipeline fixes for VK_EXT_graphics_pipeline_library")
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27054>
(cherry picked from commit 80a5df16fe81d39fd771d5805b141c4eff623886)
-
2a4f8de5
by Caio Oliveira at 2024-01-23T20:21:13+00:00
intel/compiler: Fix rebuilding the CFG in fs_combine_constants
When building the CFG the instructions are taken of the list in
fs_visitor and added to the lists inside each block. The single
"exec_node" in the instruction is used for those memberships.
In the case the pass rebuilt the CFG, it had no instructions, so
calculate_cfg() had nothing to work with. For now fix the bug by
pulling all the instructions back to the original list.
We can do better here, but punting until upcoming work on
CFG itself.
Issue found in an unpublished CTS test. Small reproduction in our
unit tests now enabled.
Fixes: 65237f8bbca ("intel/fs: Don't add MOV instructions to DO blocks in combine constants")
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27131>
(cherry picked from commit 4dbf9181cd53baad71a2dba7b3a9198c57ba1941)
-
ee57c9df
by Karol Herbst at 2024-01-23T20:34:30+00:00
nir: rework and fix rotate lowering
No driver supports urol/uror on all bit sizes. Intel gen11+ only for 16
and 32 bit, Nvidia GV100+ only for 32 bit. Etnaviv can support it on 8,
16 and 32 bit.
Also turn the `lower` into a `has` option as only two drivers actually
support `uror` and `urol` at this momemt.
Fixes crashes with CL integer_rotate on iris and nouveau since we emit
urol for `rotate`.
v2: always lower 64 bit
Fixes: fe0965afa6b ("spirv: Don't use libclc for rotate")
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by (Intel and nir): Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Yonggang Luo <luoyonggang@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27090>
(cherry picked from commit f2b7c4ce29b36556968ccf0480393180455d498a)
-
e2178ddc
by Eric Engestrom at 2024-01-23T20:34:30+00:00
.pick_status.json: Update to 90939e93f6657e1334a9c5edd05e80344b17ff66
-
725af5b5
by Karol Herbst at 2024-01-23T20:34:31+00:00
nak/opt_out: fix comparison in try_combine_outs
clippy complained it was comparing the same thing
Fixes: 5b355ff25a7 ("nak: Fix opt_out")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27216>
(cherry picked from commit 0a414ecdf5b55bb5a1717693dbf0fbaba9867792)
-
b85673b0
by Rhys Perry at 2024-01-23T20:34:31+00:00
radv: do nir_shader_gather_info after radv_nir_lower_rt_abi
Fixes compilation of a Doom Eternal shader with
PIPELINE_CREATE_DISABLE_OPTIMIZATION_BIT.
ac_nir_lower_resinfo() was not happening because it is predicated on
uses_resource_info_query and no later optimization updated it.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27195>
(cherry picked from commit 90939e93f6657e1334a9c5edd05e80344b17ff66)
-
08229beb
by Karol Herbst at 2024-01-23T20:52:01+00:00
nir: rework and fix rotate lowering
No driver supports urol/uror on all bit sizes. Intel gen11+ only for 16
and 32 bit, Nvidia GV100+ only for 32 bit. Etnaviv can support it on 8,
16 and 32 bit.
Also turn the `lower` into a `has` option as only two drivers actually
support `uror` and `urol` at this momemt.
Fixes crashes with CL integer_rotate on iris and nouveau since we emit
urol for `rotate`.
v2: always lower 64 bit
Fixes: fe0965afa6b ("spirv: Don't use libclc for rotate")
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by (Intel and nir): Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Acked-by: Yonggang Luo <luoyonggang@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27090>
(cherry picked from commit f2b7c4ce29b36556968ccf0480393180455d498a)
-
7e322ae5
by Lionel Landwerlin at 2024-01-23T21:13:46+00:00
anv: fix pipeline executable properties with graphics libraries
We're missing the ISA code in renderdoc. You can reproduce with the
Sascha Willems graphics pipeline demo.
The change is large here because we have to fix a confusion between
anv_shader_bin & anv_pipeline_executable. anv_pipeline_executable is
there as a representation for the user and multiple
anv_pipeline_executable can point to a single anv_shader_bin.
In this change we split the anv_shader_bin related logic that was
added in anv_pipeline_add_executable*() and move it to a new
anv_pipeline_account_shader() function.
When importing RT libraries, we add all the anv_pipeline_executable
from the libraries.
When importing Gfx libraries, we add the anv_pipeline_executable only
if not doing link time optimization.
anv_shader_bin related properties are added whenever we're importing a
shader from a library, compiling or finding in the cache.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 3d49cdb71e ("anv: implement VK_EXT_graphics_pipeline_library")
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26594>
(cherry picked from commit 58c9f817cbed2fc5263c86b514730e54a2d9cbe4)
-
b6045c5f
by Lionel Landwerlin at 2024-01-23T21:14:02+00:00
anv: implement undocumented tile cache flush requirements
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27169>
(cherry picked from commit ba87656079a7fb745c06e78641d2fa6ac4112b82)
-
b0be9a58
by Tapani Pälli at 2024-01-23T21:23:45+00:00
iris: replace constant cache invalidate with hdc flush
This implements Wa_14010840176.
Cc: mesa-stable
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21364>
(cherry picked from commit 231ede4f0ca92b33da8b16e7921af9877379c7d1)
-
2519220a
by Eric Engestrom at 2024-01-24T14:21:43+00:00
.pick_status.json: Update to eca4f0f632b1e3e6e24bd12ee5f00522eb7d0fdb
-
4410947e
by Eric Engestrom at 2024-01-24T14:22:16+00:00
.pick_status.json: Update to eca4f0f632b1e3e6e24bd12ee5f00522eb7d0fdb
-
10596139
by Rhys Perry at 2024-01-24T14:22:19+00:00
nir/lower_non_uniform: set non_uniform=false when lowering is not needed
Fixes RADV compilation of a Doom Eternal pipeline with
PIPELINE_CREATE_DISABLE_OPTIMIZATION_BIT, because
nir_opt_non_uniform_access was skipped and later passes don't expect
non-uniform access.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: b1619109ca91 ("nir/lower_non_uniform: remove non_uniform flags after lowering")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27192>
(cherry picked from commit 015b0d678f8027008a5ae4cc3bb066a859839c8a)
-
73dcdc7a
by Rhys Perry at 2024-01-24T14:22:20+00:00
nir/lower_shader_calls: remove CF before nir_opt_if
Otherwise, opt_if_simplification() can attempt to insert an inot after a
jump.
Fixes RADV compilation of a Cyberpunk 2077 pipeline with
PIPELINE_CREATE_DISABLE_OPTIMIZATION_BIT.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27193>
(cherry picked from commit e465ac25618b3ce2d8666b5015101414f05f9876)
-
d2094c1e
by Boris Brezillon at 2024-01-24T14:22:21+00:00
panfrost: Clamp the render area to the damage region
The render area clamping was lost during the transition to the FB
helpers. Restore the original logic so we can benefit from
EGL_KHR_partial_update on v4, and on v5 when only one damage
rectangle is passed.
Fixes: ff3eada7eb4e ("panfrost: Use the generic preload and FB helpers in the gallium driver")
Reported-by: Sjoerd Simons <sjoerd.simons@collabora.com>
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Tested-by: Sjoerd Simons <sjoerd.simons@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27215>
(cherry picked from commit f6f7715c5824922e867aa739c587eb1718db66c1)
-
466ae8c3
by Friedrich Vock at 2024-01-24T14:22:22+00:00
nir: Make is_trivial_deref_cast public
Cc: mesa-stable
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27197>
(cherry picked from commit 6c845ed548f87fdade9293c83858f2876d3b7cc6)
-
3f1d5726
by Friedrich Vock at 2024-01-24T14:22:23+00:00
nir: Handle casts in nir_opt_copy_prop_vars
Cc: mesa-stable
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27197>
(cherry picked from commit 9f22b95956cb11ccba12dd9f7e4510851fb744a3)
-
d25222c7
by Karol Herbst at 2024-01-24T14:22:24+00:00
rusticl/kernel: check that local size on dispatch doesn't exceed limits
Cc: mesa-stable
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27232>
(cherry picked from commit eca4f0f632b1e3e6e24bd12ee5f00522eb7d0fdb)
-
9a453527
by Rhys Perry at 2024-01-24T14:22:40+00:00
radv: do nir_shader_gather_info after radv_nir_lower_rt_abi
Fixes compilation of a Doom Eternal shader with
PIPELINE_CREATE_DISABLE_OPTIMIZATION_BIT.
ac_nir_lower_resinfo() was not happening because it is predicated on
uses_resource_info_query and no later optimization updated it.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27195>
(cherry picked from commit 90939e93f6657e1334a9c5edd05e80344b17ff66)
-
eb965b58
by Rhys Perry at 2024-01-24T14:23:01+00:00
nir/lower_non_uniform: set non_uniform=false when lowering is not needed
Fixes RADV compilation of a Doom Eternal pipeline with
PIPELINE_CREATE_DISABLE_OPTIMIZATION_BIT, because
nir_opt_non_uniform_access was skipped and later passes don't expect
non-uniform access.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: b1619109ca91 ("nir/lower_non_uniform: remove non_uniform flags after lowering")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27192>
(cherry picked from commit 015b0d678f8027008a5ae4cc3bb066a859839c8a)
-
14277e07
by Rhys Perry at 2024-01-24T14:23:02+00:00
nir/lower_shader_calls: remove CF before nir_opt_if
Otherwise, opt_if_simplification() can attempt to insert an inot after a
jump.
Fixes RADV compilation of a Cyberpunk 2077 pipeline with
PIPELINE_CREATE_DISABLE_OPTIMIZATION_BIT.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27193>
(cherry picked from commit e465ac25618b3ce2d8666b5015101414f05f9876)
-
73e2fce1
by Boris Brezillon at 2024-01-24T14:23:03+00:00
panfrost: Clamp the render area to the damage region
The render area clamping was lost during the transition to the FB
helpers. Restore the original logic so we can benefit from
EGL_KHR_partial_update on v4, and on v5 when only one damage
rectangle is passed.
Fixes: ff3eada7eb4e ("panfrost: Use the generic preload and FB helpers in the gallium driver")
Reported-by: Sjoerd Simons <sjoerd.simons@collabora.com>
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Tested-by: Sjoerd Simons <sjoerd.simons@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27215>
(cherry picked from commit f6f7715c5824922e867aa739c587eb1718db66c1)
-
261ed780
by Friedrich Vock at 2024-01-24T14:23:04+00:00
nir: Make is_trivial_deref_cast public
Cc: mesa-stable
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27197>
(cherry picked from commit 6c845ed548f87fdade9293c83858f2876d3b7cc6)
-
5688a6ea
by Friedrich Vock at 2024-01-24T14:23:04+00:00
nir: Handle casts in nir_opt_copy_prop_vars
Cc: mesa-stable
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27197>
(cherry picked from commit 9f22b95956cb11ccba12dd9f7e4510851fb744a3)
-
39987bd9
by Karol Herbst at 2024-01-24T14:23:05+00:00
rusticl/kernel: check that local size on dispatch doesn't exceed limits
Cc: mesa-stable
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27232>
(cherry picked from commit eca4f0f632b1e3e6e24bd12ee5f00522eb7d0fdb)
-
a0453a14
by Eric Engestrom at 2024-01-24T20:01:03+00:00
docs: add release notes for 23.3.4
-
27405fd5
by Eric Engestrom at 2024-01-24T20:01:24+00:00
VERSION: bump for 23.3.4
-
808f0566
by Eric Engestrom at 2024-01-24T20:01:30+00:00
VERSION: bump for 24.0.0-rc3
-
3be98c19
by Timo Aaltonen at 2024-01-25T15:20:17+02:00
Merge branch 'upstream-unstable' into debian-unstable
-
20dfcc32
by Timo Aaltonen at 2024-01-25T15:20:38+02:00
version bump
-
afb01b38
by Timo Aaltonen at 2024-01-25T15:41:48+02:00
release to sid
-
bc2d798c
by Timo Aaltonen at 2024-01-25T15:45:52+02:00
Merge branch 'upstream-experimental' into debian-experimental
-
fb549b68
by Timo Aaltonen at 2024-01-25T15:46:27+02:00
version bump
-
ef109819
by Timo Aaltonen at 2024-01-25T15:46:36+02:00
Merge tag 'mesa-23.3.4' into debian-experimental
mesa-23.3.4
-
70a0983a
by Timo Aaltonen at 2024-01-25T15:46:40+02:00
Merge branch 'debian-unstable' into debian-experimental
-
ffc91a83
by Timo Aaltonen at 2024-01-25T15:53:40+02:00
patches: Attempt to fix failing autopkgtests by changing zink init failure logging to a warning instead of an error.
-
ea8f7853
by Timo Aaltonen at 2024-01-25T15:53:49+02:00
release to experimental