Mark Collins [Tue, 3 Oct 2023 18:22:49 +0000 (18:22 +0000)]
tu: Support higher descriptor set count for A7XX
Allows for the descriptor set count to vary at runtime depending on
the specific GPU to allow for 7 usable descriptor sets on A7XX with
one reserved for dynamic offsets.
Passing VK-CTS: dEQP-VK.binding_model.*
Signed-off-by: Mark Collins <mark@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25534>
Mark Collins [Tue, 3 Oct 2023 15:58:56 +0000 (15:58 +0000)]
freedreno/common: Add max_sets property to A6xxGPUInfo
A7XX increases the maximum amount of descriptor sets from 5 to 8,
the amount of descriptor sets was added as a property of the
A6xxGPUInfo to accommodate it changing across GPUs.
Signed-off-by: Mark Collins <mark@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25534>
Lionel Landwerlin [Sat, 7 Oct 2023 13:13:35 +0000 (16:13 +0300)]
anv: fix index buffer size programming
This is a merge issue due to 2 MRs touching the same code.
Fixes a few maintence5 tests like : dEQP-VK.robustness.bind_index_buffer2.offset_0.draw_indexed.oo_size
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes:
50f6903bd9 ("anv: add new low level emission & dirty state tracking")
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25599>
Marek Olšák [Fri, 7 Apr 2023 03:34:47 +0000 (23:34 -0400)]
nir: handle nir_var_mem_ubo in nir_clone_uniform_variable
for UBOs
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25394>
Marek Olšák [Wed, 5 Apr 2023 10:40:59 +0000 (06:40 -0400)]
nir: expose reusable linking helpers for cloning uniform loads
for the new varying optimizer
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25394>
Marek Olšák [Tue, 26 Sep 2023 04:10:44 +0000 (00:10 -0400)]
nir: gather dual slot input information
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25394>
Marek Olšák [Tue, 26 Sep 2023 04:09:49 +0000 (00:09 -0400)]
nir: take dual slot input info into account when computing IO driver locations
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25394>
Marek Olšák [Mon, 13 Mar 2023 04:18:47 +0000 (00:18 -0400)]
nir: add dual-slot input information into load_input intrinsics
This is necessary to allow optimizing VS inputs after nir_lower_io, which
is currently impossible because the loss of dual-slot information in NIR
would break VS inputs. With this, driver locations can be recomputed by
calling nir_recompute_io_bases. It's a prerequisite for optimizing varyings
with lowered IO.
When this is used, we will be able to eliminate unused dual-slot VS inputs
as well as unused low and high halves of dual-slot VS inputs for the first
time, which can happen due to optimizations of varyings. Without this,
st/mesa binds vertex buffers for dual-slot inputs that are fully or
partially unused in the shader.
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25394>
Marek Olšák [Thu, 31 Aug 2023 14:54:39 +0000 (10:54 -0400)]
nir: recompute IO bases after DCE in nir_lower_io_passes
otherwise the IO bases can be incorrect due to non-DCE'd input loads
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25394>
Marek Olšák [Fri, 1 Sep 2023 22:56:10 +0000 (18:56 -0400)]
nir: sort variables by location in nir_lower_io_passes to work around a bug
I don't know why this is necessary, but it unblocks the work on varying
optimizations.
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25394>
Gert Wollny [Fri, 6 Oct 2023 07:31:24 +0000 (09:31 +0200)]
virgl: Use common clear_texture if host doesn't support the feature
v2: Fix include (osy)
Fixes:
a1eabeff (gallium: remove PIPE_CAP_CLEAR_TEXTURE)
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9944
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25580>
Vignesh Raman [Fri, 6 Oct 2023 05:14:35 +0000 (10:44 +0530)]
ci: Uprev crosvm
Recent commit in linux kernel 6.6 rc3 broke booting
in crosvm. Latest crosvm contains a fix for this issue.
So bump the crosvm version to latest.
https://issuetracker.google.com/issues/
303128596
https://chromium-review.googlesource.com/c/crosvm/crosvm/+/4906858
Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25582>
David Heidelberg [Sat, 7 Oct 2023 08:17:57 +0000 (10:17 +0200)]
ci/lima: farm is down, disable for now
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25596>
Martin Roukala (né Peres) [Fri, 6 Oct 2023 18:20:11 +0000 (21:20 +0300)]
ci/b2c: switch containers to a back-up ahead of valve-infra renaming
We are about to rename mupuf/valve-infra into gfx-ci/ci-tron.
While most resources will transparently be redirected, gitlab does
not allow us to keep our containers during the migration.
To work around that, I uploaded the current containers to Eric's fork
of valve-infra. Let's use these containers until the migration is over!
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25593>
Faith Ekstrand [Fri, 6 Oct 2023 23:46:07 +0000 (18:46 -0500)]
nvk: Invalidate SKED caches at the top of command buffers
This is the cache that caches QMDs. We need to invalidate it or else we
can end up with cache collisions and end up running the wrong shader.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25595>
Faith Ekstrand [Thu, 5 Oct 2023 17:34:35 +0000 (12:34 -0500)]
nvk: Invalidate sampler/texture header caches in BeginCommandBuffer()
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25595>
Faith Ekstrand [Fri, 6 Oct 2023 23:45:32 +0000 (18:45 -0500)]
nvk: Add a nvk_cmd_buffer_compute_cls() helper
We have one of these for 3D and it's more ergonamic.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25595>
Faith Ekstrand [Fri, 6 Oct 2023 23:42:34 +0000 (18:42 -0500)]
nouveau: Generate headers for Maxwell B compute
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25595>
Ian Romanick [Thu, 5 Oct 2023 00:41:08 +0000 (17:41 -0700)]
intel/fs: Add DP4A to get_lowered_simd_width
While working on cooperative matrix support, I noticed some invalid
DP4A instructions being generated.
dp4a(32) g33<1>UD g21<8,8,1>UD g1.0<0,1,0>UD g9<1,1,1>UD
This violates the constraint that the destination or a source can only
access two consecutive GRFs.
I'm a little surprised that validation didn't catch this. Perhaps
because it's a 3 source instruction? Either way, it seems like a bigger
project to fix that.
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Fixes:
0f809dbf404 ("intel/compiler: Basic support for DP4A instruction")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25554>
Caio Oliveira [Mon, 18 Sep 2023 21:55:58 +0000 (14:55 -0700)]
util: Add function print information about a ralloc tree
In release mode print the ralloc tree header pointers. In debug mode,
will look at the child allocations recursively checking for the canary
values. In most cases this should work and give us extra information
about the type of the subtree (GC, Linear) and the allocation sizes.
For example, calling it at the end of spirv_to_nir() will output the
following tree with a summary at the end (lines elided to focus on the
general structure of the output)
```
0xca8d00 (472 bytes)
0xdf2c60 (32 bytes)
0xdf2c00 (32 bytes)
0xdf2ba0 (32 bytes)
0xdf2b40 (32 bytes)
(...)
0xcde8b0 (64 bytes)
0xcde760 (72 bytes)
0xd2de40 (168 bytes)
0xcce660 (64 bytes)
0xcce510 (72 bytes)
0xcce5a0 (120 bytes)
0xcce490 (64 bytes)
(...)
0xcbc450 (456 bytes)
0xdf55c0 (160 bytes)
0xdf5730 (72 bytes)
0xdf57c0 (80 bytes)
0xdf5530 (72 bytes)
0xdf56a0 (80 bytes)
(...)
0xcbe840 (128 bytes)
0xcc4310 (4 bytes)
0xcbc660 (536 bytes) (gc context)
0xde6b40 (32576 bytes) (gc slab or large block)
0xddb160 (32704 bytes) (gc slab or large block)
0xdc8d50 (32704 bytes) (gc slab or large block)
(...)
0xcde9a0 (32704 bytes) (gc slab or large block)
0xcd6720 (32704 bytes) (gc slab or large block)
0xcce6e0 (32768 bytes) (gc slab or large block)
0xcbc330 (72 bytes)
0xd680a0 (208 bytes)
0xca9010 (78560 bytes)
0xca8f20 (176 bytes)
==== RALLOC INFO ptr=0xca8d30 info=0xca8d00
ralloc allocations = 4714
- linear = 0
- gc = 23
- other = 4691
content bytes = 1055139
ralloc metadata bytes = 226272
linear metadata bytes = 0
====
```
There's a flag to pass so only the summary at the end is printed.
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25482>
Caio Oliveira [Fri, 29 Sep 2023 22:34:02 +0000 (15:34 -0700)]
util: Add a canary to identify gc_ctx in debug mode
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25482>
Caio Oliveira [Mon, 18 Sep 2023 21:24:00 +0000 (14:24 -0700)]
util: Add size to ralloc_header in debug mode
For 64-bit environments this just fills a padding. Will enable
dumping more information when debugging ralloc trees.
Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25482>
Caio Oliveira [Sat, 16 Sep 2023 17:19:22 +0000 (10:19 -0700)]
rusticl: Add Rust bindings for inline glsl_types functions
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25470>
Caio Oliveira [Fri, 29 Sep 2023 07:36:13 +0000 (00:36 -0700)]
compiler/types: Flip wrapping of basic "get type" functions
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25470>
Caio Oliveira [Thu, 7 Sep 2023 23:49:47 +0000 (16:49 -0700)]
compiler/types: Flip wrapping of convenience accessors for vector types
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25470>
Caio Oliveira [Sat, 2 Sep 2023 06:49:26 +0000 (23:49 -0700)]
compiler/types: Flip wrapping of various type identification checks
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25470>
Caio Oliveira [Sat, 2 Sep 2023 05:35:26 +0000 (22:35 -0700)]
compiler/types: Flip wrapping of base_type checks
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25470>
Caio Oliveira [Fri, 29 Sep 2023 18:09:55 +0000 (11:09 -0700)]
compiler/types: Move C declarations into glsl_types.h
This ensures they'll be visible for the C++ inline implementations.
Reordered the functions to better organize them: queries, getters,
transformers, layout functions.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25470>
Caio Oliveira [Fri, 1 Sep 2023 23:26:02 +0000 (16:26 -0700)]
compiler/types: Move the C++ inline functions in glsl_type out of the struct body
Just move code, will make easier to flip these to be wrappers to the C code. Keep
those in a separate header file to reduce cluttering glsl_types.h.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25470>
Yiwei Zhang [Fri, 6 Oct 2023 06:41:54 +0000 (23:41 -0700)]
venus: fix re-export of imported classic 3d resources
When the guest driver is Virgl while Xwayland is on Zink, Virgl can
request virtgpu classic 3d resource allocations for swapchain images.
Zink will import when the image is shared with xserver and will export
for fd info of all 2d images later to be forwarded.
Cc: mesa-stable
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25579>
Samuel Pitoiset [Tue, 19 Sep 2023 13:47:34 +0000 (15:47 +0200)]
radv: fix re-emitting streamout descriptors for NGG streamout
In a scenario like:
CmdBindTransformFeedbackBuffers()
BeginTransformFeedback()
CmdDraw() --> streamout descriptors emitted
EndTransformFeedback() --> streamout descriptors emitted as 0 (disabled)
CmdDraw()
BeginTransformFeedback()
CmdDraw() --> streamout descriptor not re-emitted
EndTransformFeedback()
Fix this by re-emitting streamout descriptors when streamout is
enabled/disabled because a buffer size of 0 acts like a disable bit.
This fixes dEQP-VK.transform_feedback.simple.backward_dependency_indirect*
on NAVI31.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25583>
antonino [Fri, 6 Oct 2023 10:42:20 +0000 (12:42 +0200)]
zink: store bindless var when creating it to avoid creating it again
When bindless variables are created they get used directly and never
stored anywhere so if another bindless instruction is encountered
duplicate variables are created.
Fixes:
fe2ba184d8c ("zink: use descriptor indices in compiler")
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25591>
Danylo Piliaiev [Fri, 6 Oct 2023 08:26:26 +0000 (10:26 +0200)]
tu/kgsl: Fix field order in kgsl_command_object init
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9942
Fixes:
ec268fa5b666a49adafc431dbc05b73bfd74526e
("tu/kgsl: Support u_trace and perfetto")
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25581>
Ian Douglas Scott [Mon, 28 Aug 2023 23:18:01 +0000 (16:18 -0700)]
egl/wayland: Don't segfault if `create_wl_buffer` returns `NULL`
Normally, this shouldn't fail, but it has various cases where it returns
`NULL`. Without this change, it would result in a segfault when
`wl_buffer_add_listener` is called.
This instead makes `EGLSwapBuffers` return a `EGL_BAD_ALLOC` error.
The other place `create_wl_buffer` is called already checks the return
value, and the Vulkan WSI code doesn't seem to have an issue like this.
Signed-off-by: Ian Douglas Scott <ian@iandouglasscott.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24915>
Eric Engestrom [Tue, 3 Oct 2023 09:17:08 +0000 (10:17 +0100)]
ci: skip dEQP-VK.api.driver_properties.conformance_version for everyone
This test checks the driver's reported conformance version against the
version of the CTS we're running. This check fails every few months
and everyone has to go and bump the number in every driver.
Running this check only makes sense while preparing a conformance
submission, so skip it in the regular CI.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25519>
Eric Engestrom [Thu, 5 Oct 2023 11:52:40 +0000 (12:52 +0100)]
docs/ci: rewrite the "farm maintenance ^ other change" rule to mean what we actually meant
Sometimes updates to the farm config are needed before re-enabling it,
or updating test expectations with new failures that happened while the
farm was down, etc.
These need to be allowed, as the alternative is to update the
config/expectations/etc. blindly in one MR, merge it, and then merge
another MR with the farm re-enablement.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25560>
Kai Wasserbäch [Fri, 6 Oct 2023 09:48:30 +0000 (11:48 +0200)]
fix: clover: LLVM 18: s/CodeGenOpt::/CodeGenOptLevel::/
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9833
Reference: https://github.com/llvm/llvm-project/commit/
0a1aa6cda2758b0926a95f87d39ffefb1cb90200
Signed-off-by: Kai Wasserbäch <kai@dev.carbon-project.org>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25586>
Kai Wasserbäch [Fri, 6 Oct 2023 09:32:59 +0000 (11:32 +0200)]
fix: clover: LLVM 18 renamed/moved CGFT_*, update compat layer
LLVm renamed and moved the CGFT_* stuff, we need to update the clover
compat header to follow suit.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9833
Reference: https://github.com/llvm/llvm-project/commit/
0a1aa6cda2758b0926a95f87d39ffefb1cb90200
Signed-off-by: Kai Wasserbäch <kai@dev.carbon-project.org>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25586>
Samuel Pitoiset [Thu, 31 Aug 2023 08:35:38 +0000 (10:35 +0200)]
radv: fix gang submissions with chaining
Gang submissions are mostly only be used for task shaders to both
submit GFX and ACE command buffers in the same submission. Though,
the gang leader (the last submitted CS) IP type should match the
queue IP type to determine which fence to signal.
But if chaining is enabled, it's possible that all GFX cmdbufs are
chained to the first GFX cmdbuf, which means the last CS is ACE and
not GFX.
Fix this by resetting chaining for GFX when a cmdbuffer has GFX+ACE.
Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9724
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24966>
Samuel Pitoiset [Thu, 5 Oct 2023 10:23:58 +0000 (12:23 +0200)]
radv: fill the scratch BO in radv_fill_shader_rings()
Cleanup.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25529>
Samuel Pitoiset [Thu, 5 Oct 2023 10:21:49 +0000 (12:21 +0200)]
radv: always write the sample positions when a new descriptor BO is created
This was completely broken, for example in the following scenario:
- submits something which needs sample positions (this creates a new
descriptor BO with sample positions)
- submits something which needs the tess rings (this creates a new
descriptor BO with tess rings but without sample positions, ie.
add_sample_positions would be FALSE)
- submits something which needs sample positions again (this won't
create a new descriptor BO because it incorrectly remembered that
sample positions were set)
Fix this by always writing the sample positions.
This should fix the following flakes:
- dEQP-VK.fragment_shading_barycentric.*.weights.pipeline_topology_dynamic.msaa_interpolate_at_sample.*
- dEQP-VK.pipeline.fast_linked_library.multisample_interpolation.sample_interpolate_at_distinct_values.*
- dEQP-VK.pipeline.fast_linked_library.multisample_interpolation.sample_interpolation_consistency.*
- dEQP-VK.draw.renderpass.linear_interpolation.*
- dEQP-VK.draw.dynamic_rendering.primary_cmd_buff.linear_interpolation.*
These flakes were extremely hard to reproduce!
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25529>
Lionel Landwerlin [Wed, 10 Aug 2022 11:10:10 +0000 (11:10 +0000)]
intel/ds: track acceleration RT commands
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25570>
Lionel Landwerlin [Thu, 5 Oct 2023 14:54:35 +0000 (17:54 +0300)]
anv: use buffer pools for BVH build buffers
Private memory for BVH builds doesn't need to be mapped on the host,
it's purely for use by the GPU. So it can be put into a different
buffer pool that can put into VRAM only buffers.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25570>
Lionel Landwerlin [Thu, 5 Oct 2023 15:03:41 +0000 (18:03 +0300)]
anv: move bo_pool allocation flags to init caller
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25570>
Lionel Landwerlin [Thu, 5 Oct 2023 14:09:54 +0000 (17:09 +0300)]
anv: reduce working temporary memory for BVH builds
Part of the memory allocated (private) is a temporary working buffer
for the GRL kernels. Once the build operation is done, the buffer
becomes unused.
Rather than allocate a new buffer each time, reuse the current last
allocated one if its size fits the next build operation.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25570>
Eric Engestrom [Wed, 4 Oct 2023 12:15:07 +0000 (13:15 +0100)]
ci: drop unused ephemeral packages in alpine image
There's nothing between installing these packages and removing them that
uses them, so let's just drop them.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25546>
Caio Oliveira [Tue, 3 Oct 2023 18:12:47 +0000 (11:12 -0700)]
intel/fs: Tweak default case of fs_inst::size_read()
In the default case, there's a special case with a few conditions.
Prefer the cheapest conditions first, so we can take advantage of
short-circuiting.
Effect is a small but still significant reduce in shader compilation
times, as can be seen by:
- Fossil replay for Rise of the Tomb Raider
```
Difference at 95.0% confidence
-0.433333 +/- 0.028609
-1.42556% +/- 0.0941163%
(Student's t, pooled s = 0.0337886)
```
- Fossil replay for Batman Arkham City
```
Difference at 95.0% confidence
-8.84 +/- 0.146083
-1.65932% +/- 0.0274207%
(Student's t, pooled s = 0.125423)
```
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25549>
Mike Blumenkrantz [Thu, 5 Oct 2023 16:56:10 +0000 (12:56 -0400)]
zink: don't check submit count for unflushed usage
unflushed usage is unflushed regardless of the submit count and is
critical for detecting multi-context synchronization
fixes Wolfenstein: The New Order load screen deadlock
Fixes:
db12b881c7f ("zink: track/check submit info on resource batch usage")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25572>
Mike Blumenkrantz [Wed, 13 Sep 2023 13:22:55 +0000 (09:22 -0400)]
egl/wayland: return sooner from swrast_update_buffers() if zink
Fixes:
0f50cc03ef0 ("egl/wayland: don't block in swrast when updating buffers for zink")
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24700>
Mike Blumenkrantz [Tue, 15 Aug 2023 16:21:24 +0000 (12:21 -0400)]
egl/wayland: don't block in swrast when updating buffers for zink
this is broken, let vulkan wsi handle buffer management
Fixes:
74451ed3f08 ("egl/wayland: wait for compositor to release shm buffers")
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24700>
Mike Blumenkrantz [Tue, 15 Aug 2023 16:07:36 +0000 (12:07 -0400)]
Revert "egl/wayland: Add image loader extension for swrast"
This reverts commit
45b9b0ba32c6a6eed388ff080f7eaf5a765fe970.
Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24700>
Emma Anholt [Thu, 5 Oct 2023 20:15:30 +0000 (13:15 -0700)]
ci/hasvk: Add a bunch of new CTS border color fails.
pretty sure this is from new coverage since the CTS uprev..
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25574>
Emma Anholt [Thu, 5 Oct 2023 20:13:09 +0000 (13:13 -0700)]
ci/crocus: Add known piglit flakes
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25574>
Emma Anholt [Thu, 5 Oct 2023 20:10:41 +0000 (13:10 -0700)]
ci/etnaviv: return gl-1.4-tex1d-2dborder as a known flake
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25574>
Simon Zeni [Thu, 5 Oct 2023 14:09:55 +0000 (10:09 -0400)]
nouveau/winsys: use mmap instead of mmap64 in nouveau_bo
The function `mmap64` is part of the large file extension and should not be
called directly. Instead `mmap` should be use and let the system use the
correct interface.
Signed-off-by: Simon Zeni <simon.zeni@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25569>
Chia-I Wu [Wed, 13 Sep 2023 23:21:27 +0000 (16:21 -0700)]
radv: hard code format features for emulated formats
The format features are known. No need to rely on
radv_is_sampler_format_supported which will not work for ASTC.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25071>
Chia-I Wu [Wed, 13 Sep 2023 23:23:45 +0000 (16:23 -0700)]
radv: simplify view format override for emulated formats
Override the view format to the format of plane 1 when the view format
is also emulated. There is no functional change.
v2: check iview->vk.format directly (@yogeshmohan)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25071>
Chia-I Wu [Wed, 13 Sep 2023 22:54:00 +0000 (15:54 -0700)]
radv: add radv_is_format_emulated
This is a cleanup with no functional change.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25071>
Chia-I Wu [Tue, 26 Sep 2023 23:48:00 +0000 (16:48 -0700)]
vulkan/runtime, radv: remove 1D support from ETC2 emulation
The nir code deos not support 1D. There is also no point in supporting
1D ETC2 images.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25071>
Chia-I Wu [Tue, 26 Sep 2023 23:45:56 +0000 (16:45 -0700)]
vulkan/runtime: fix a harmless typo for ETC2 emulation
Init input_img_3d correctly.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25071>
Chia-I Wu [Tue, 5 Sep 2023 22:20:31 +0000 (15:20 -0700)]
vulkan/runtime: fix image type check for ETC2 emulation
There was a typo causing the wrong push constant to be loaded.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25071>
Chia-I Wu [Tue, 5 Sep 2023 21:08:32 +0000 (14:08 -0700)]
radv: use vk_tecompress_etc2 from the runtime
There are some minor differences
- fix incorrectly use of device->meta_state.resolve_compute.p_layout
- when on_demand is true, the creation of ds and pipeline layouts are
also deferred
- unlike radv_meta_get_view_type, vk_texcompress_etc2_image_view_type
returns 1d/2d array image views
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25071>
Chia-I Wu [Thu, 31 Aug 2023 23:33:37 +0000 (16:33 -0700)]
vulkan/runtime: add a helper for ETC2 emulation
This is based on radv's ETC2 emulation. There is no real change to the
generated NIR shader.
v2: rename vk_texcompress_etc2_image_format to vk_texcompress_etc2_emulation_format
update the comments
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25071>
Danylo Piliaiev [Thu, 5 Oct 2023 13:11:35 +0000 (15:11 +0200)]
tu: Fix VK_FORMAT_A8_UNORM_KHR using UBWC when !has_8bpp_ubwc
Fixes hangs in Anno 1800 with DXVK 2.3
Fixes:
302907e347612c3d748c88aed90fd6ab60a638f5
("tu: Expose VK_KHR_maintenance5")
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25567>
Konstantin Seurer [Fri, 22 Sep 2023 11:21:30 +0000 (13:21 +0200)]
mesa: Fix glBegin/End when LINE_LOOP is not supported
Emits the first vertex inside glEnd.
cc: mesa-stable
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25346>
Georg Lehmann [Thu, 14 Sep 2023 11:25:07 +0000 (13:25 +0200)]
aco/gfx11: support vinterp as fma_mix
Totals from 718 (0.94% of 76572) affected shaders:
Instrs: 657897 -> 654219 (-0.56%)
CodeSize: 3471668 -> 3457352 (-0.41%); split: -0.41%, +0.00%
VGPRs: 34200 -> 34164 (-0.11%)
Latency:
11687698 ->
11677030 (-0.09%); split: -0.10%, +0.00%
InvThroughput: 1455371 -> 1451537 (-0.26%); split: -0.26%, +0.00%
VClause: 7598 -> 7600 (+0.03%)
SClause: 18293 -> 18241 (-0.28%); split: -0.44%, +0.15%
Copies: 34641 -> 34644 (+0.01%); split: -0.05%, +0.06%
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25220>
Georg Lehmann [Thu, 14 Sep 2023 11:15:39 +0000 (13:15 +0200)]
aco: support v_fma_f32_dpp as fma_mix
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25220>
Georg Lehmann [Thu, 14 Sep 2023 11:01:09 +0000 (13:01 +0200)]
aco/gfx11: apply clamp/omod to vinterp
Totals from 2504 (3.27% of 76572) affected shaders:
MaxWaves: 74098 -> 74106 (+0.01%)
Instrs: 1829278 -> 1823427 (-0.32%); split: -0.32%, +0.00%
CodeSize: 9775908 -> 9759308 (-0.17%); split: -0.18%, +0.01%
Latency:
13494107 ->
13485390 (-0.06%); split: -0.10%, +0.04%
InvThroughput: 2052428 -> 2048724 (-0.18%); split: -0.18%, +0.00%
VClause: 26637 -> 26640 (+0.01%); split: -0.04%, +0.05%
SClause: 62027 -> 61988 (-0.06%); split: -0.14%, +0.08%
Copies: 73776 -> 73815 (+0.05%); split: -0.07%, +0.12%
PreVGPRs: 84403 -> 84397 (-0.01%)
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25220>
Marek Olšák [Thu, 5 Oct 2023 08:52:29 +0000 (04:52 -0400)]
Revert "ac/gpu_info: replace ib_alignment with per-IP IB base and size alignments"
This reverts commit
b6f435888b718506e8b806eae2a17d6ecf027dfc.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25558>
Marek Olšák [Thu, 5 Oct 2023 08:52:24 +0000 (04:52 -0400)]
Revert "ac/gpu_info: pad IBs according to ib_size_alignment"
This reverts commit
4f660f99378382871f0c29e0d4b4d6a044dfe715.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25558>
Marek Olšák [Thu, 5 Oct 2023 08:52:16 +0000 (04:52 -0400)]
Revert "winsys/amdgpu: pad gfx and compute IBs with a single NOP packet"
This reverts commit
43e72850697e42cdf6d794407a5523a5e481bd41.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25558>
Marek Olšák [Thu, 5 Oct 2023 08:54:06 +0000 (04:54 -0400)]
Revert "radv: fix alignment of DGC command buffers"
This reverts commit
b01e87423433f6b39d244daee4b59f53baa4a02e.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25558>
Marek Olšák [Thu, 5 Oct 2023 08:53:28 +0000 (04:53 -0400)]
Revert "radv/amdgpu: fix alignment of command buffers"
This reverts commit
4bc58c9f11f0f701be5446ee5e7e8ef9494e3c1e.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25558>
Marek Olšák [Thu, 5 Oct 2023 08:49:34 +0000 (04:49 -0400)]
Revert "ac/gpu_info: override ib_size_alignment for VCN_DEC and JPEG"
This reverts commit
867a995ce759a725b58d3fa23e5e319035b7e3f9.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25558>
David Heidelberg [Fri, 22 Sep 2023 03:56:17 +0000 (09:26 +0530)]
ci/bare-metal: drop unused imports, sort, use SPDX license
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25285>
David Heidelberg [Tue, 19 Sep 2023 06:16:36 +0000 (11:46 +0530)]
ci/bare-metal: correct workaround for R8152 issue while retrieving TFTP data
1. Move block used for detecting R8152 problems to the bootloader
phase where it belongs. Also remove requirement to 100 failures and just
retry immediatelly.
2. Consider job failed after 10 errors, not 100. From the logs on
cheza-14, ~ 30 errors is enough to fail.
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25285>
Erik Faye-Lund [Mon, 21 Aug 2023 10:04:01 +0000 (12:04 +0200)]
docs/panfrost: use math-role to denote powers of two
We do this elsewhere in the article, so let's be consistent here.
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24921>
Erik Faye-Lund [Mon, 21 Aug 2023 09:39:31 +0000 (11:39 +0200)]
docs/panfrost: use code-blocks with wrapping for long blocks
This makes it a lot easier to read the code-blocks, because we don't
*always* need to scroll.
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24921>
Erik Faye-Lund [Mon, 21 Aug 2023 08:50:32 +0000 (10:50 +0200)]
docs/panfrost: link to lima
In the age of the internet, we can use hyperlinking to content instead
of just telling users about something! Let's do that here as well!
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24921>
Samuel Pitoiset [Thu, 5 Oct 2023 08:40:34 +0000 (10:40 +0200)]
radv/ci: cleanup list of expected failures for NAVI10/NAVI21/VEGA10
These are flakes, not failures. Remove them from the list of expected
failures to avoid confusion.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25557>
Samuel Pitoiset [Thu, 5 Oct 2023 08:38:07 +0000 (10:38 +0200)]
radv/ci: remove no longer existing test for VANGOGH
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25557>
Samuel Pitoiset [Thu, 5 Oct 2023 06:53:04 +0000 (08:53 +0200)]
radv: fix synchronization with emulated GS primitives/invocations queries
Move emitting the EOP even which writes the availability bit after the
GDS copy to ensure it's available.
This should fix all GS primitives/invocations flakes in CI.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25457>
Samuel Pitoiset [Thu, 5 Oct 2023 07:41:16 +0000 (09:41 +0200)]
radv/ci: exclude dEQP-VK.texture.explicit_lod.2d.sizes.128x128_* for all jobs
These tests usually take more than 20s to finish which is slow.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25556>
Samuel Pitoiset [Thu, 5 Oct 2023 07:39:16 +0000 (09:39 +0200)]
radv/ci: remove duplicate skipped tests for RAVEN/STONEY
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25556>
Georg Lehmann [Wed, 4 Oct 2023 09:51:21 +0000 (11:51 +0200)]
aco/gfx11: optimize dual source export
We can combine dpp with the v_cndmask_b32.
Foz-DB Navi31:
Totals from 222 (0.28% of 79330) affected shaders:
Instrs: 564392 -> 563373 (-0.18%); split: -0.19%, +0.01%
CodeSize: 2867040 -> 2864728 (-0.08%); split: -0.09%, +0.01%
Latency: 4278957 -> 4275199 (-0.09%); split: -0.09%, +0.00%
InvThroughput: 586636 -> 585824 (-0.14%); split: -0.14%, +0.00%
SClause: 20210 -> 20211 (+0.00%); split: -0.02%, +0.02%
Copies: 39763 -> 39778 (+0.04%); split: -0.13%, +0.17%
PreVGPRs: 13924 -> 13922 (-0.01%)
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25541>
Ganesh Belgur Ramachandra [Thu, 14 Sep 2023 09:55:37 +0000 (04:55 -0500)]
radeonsi: "clear_12bytes_buffer" shader in nir
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25221>
Ganesh Belgur Ramachandra [Thu, 14 Sep 2023 12:33:35 +0000 (07:33 -0500)]
radeonsi: "clear_render_target_1d_array" shader in nir
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25221>
Ganesh Belgur Ramachandra [Thu, 14 Sep 2023 07:02:52 +0000 (02:02 -0500)]
radeonsi: "clear_render_target" shader in nir
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25221>
Marek Olšák [Wed, 4 Oct 2023 09:17:11 +0000 (05:17 -0400)]
ac/surface: don't require exact pitch for gfx6-8 tiled imports
It was reported that it broke Stoney. Something probably uses a suboptimal
pitch, like minigbm.
Fixes:
7d066330e0c3d47a0ad - ac/surface: relax custom pitch requirements to any multiple of 256B on gfx10.3+
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25540>
Marek Olšák [Sat, 21 Jan 2023 02:54:57 +0000 (21:54 -0500)]
radeonsi: upload shaders via a staging buffer so as not to map VRAM directly
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25494>
Marek Olšák [Sun, 1 Oct 2023 00:08:23 +0000 (20:08 -0400)]
radeonsi: add another aux context for uploading shaders
When the first auxiliary context is locked and wants to compile and upload
a shader asynchronously, we need to use another auxiliary context
in the compiler thread because the first one is locked at that point.
This adds an array of auxiliary contexts into si_screen and changes how aux
contexts are accessed.
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25494>
Sviatoslav Peleshko [Thu, 28 Sep 2023 17:26:06 +0000 (20:26 +0300)]
intel/eu/validate: Validate "packed word exception" stricter
Fixes:
75b7f5a2 ("i965: Validate "Region Alignment Rules"")
Signed-off-by: Sviatoslav Peleshko <sviatoslav.peleshko@globallogic.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25378>
Sviatoslav Peleshko [Mon, 25 Sep 2023 16:16:50 +0000 (19:16 +0300)]
intel/fs: Fix "packed word exception" condition for register regioning
Fixes:
a6bf5f88 ("i965/fs: Enforce common regioning restrictions by SIMD splitting.")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9432
Signed-off-by: Sviatoslav Peleshko <sviatoslav.peleshko@globallogic.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25378>
Faith Ekstrand [Wed, 4 Oct 2023 20:25:38 +0000 (15:25 -0500)]
nvk: Use align() and align64() instead of ALIGN_POT
For one thing, they actually assert that the alignment factor is a power
of two. More importantly, though, because they're real functions and
not macros, they up-cast the alignment to a full uint32_t or uint64_t
before doing the alignment calculation. If you do ALIGN_POT() on a
64-bit value with a 32-bit alignment, it truncates to 32 bits because
the NOT is done before the up-cast due to C implicit cast rules.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25555>
Faith Ekstrand [Wed, 4 Oct 2023 20:02:20 +0000 (15:02 -0500)]
nvk: Handle zero-sized sparse buffers
In the zero case, we don't want to allocate any VMA.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25555>
Konstantin Seurer [Thu, 21 Sep 2023 17:25:31 +0000 (19:25 +0200)]
zink: Enable edge flags with points
Fixes: 90a8525 ("zink: handle edgeflags")
Reviewed-by: Antonino Maniscalco <antonino.maniscalco@collabora.com>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25335>
Konstantin Seurer [Thu, 21 Sep 2023 17:24:49 +0000 (19:24 +0200)]
nir/passthrough_gs: Support edge flags with points
Fixes: 24535ff ("nir: handle edge flags in nir_create_passthrough_gs")
Reviewed-by: Antonino Maniscalco <antonino.maniscalco@collabora.com>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25335>
Konstantin Seurer [Thu, 21 Sep 2023 17:14:20 +0000 (19:14 +0200)]
zink: Initialize primitive types to an invalid value
The memory is zero initialized which corresponds to MESA_PRIM_POINTS.
Fixes: 659c39f ("zink: rework primitive rasterization type logic")
Reviewed-by: Antonino Maniscalco <antonino.maniscalco@collabora.com>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25335>
Leo Liu [Tue, 3 Oct 2023 02:32:26 +0000 (22:32 -0400)]
ac/gpu_info: override ib_size_alignment for VCN_DEC and JPEG
With the commit
4f660f99 ("ac/gpu_info: pad IBs according to ib_size_alignment"),
we found kernel isn't reporting ib_base/size_alignment correctly, thus causing
VCN_DEC and JPEG functions broken. We will fix the kernel and bump the kernel
version, and now for the older kernel, we need this override.
closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9916
Signed-off-by: Leo Liu <leo.liu@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25511>
Igor Torrente [Thu, 28 Sep 2023 16:26:25 +0000 (12:26 -0400)]
zink: fix for startup crash of weston running on top of zink + venus
Venus asserts if we are transitioning to/from
`VK_IMAGE_LAYOUT_PRESENT_SRC_KHR` if we are not dealing with a wsi
image.
Now we detects this case and avoid transition the layout in this case.
Signed-off-by: Igor Torrente <igor.torrente@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25423>