review.tizen.org Git - platform/upstream/mesa.git/log

turnip: Correctly store separate stencil in gmem store

- When resolving d32s8 to s8 we stored stencil with a wrong format.
- For unaligned multi-sample store we used wrong gmem offset for stencil.

If unaligined store is forced this change fixes a hang in:
dEQP-VK.renderpass2.depth_stencil_resolve.image_2d_32_32.samples_2.d32_sfloat_s8_uint_separate_layouts.compatibility_depth_zero_stencil_zero_testing_stencil

Fixes: b157a5d0d68ee8a1b4cb862a56b97bd881841413
("tu: Implement non-aligned multisample GMEM STORE_OP_STORE")

Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15532>

radv: enable VK_KHR_pipeline_library

This has been initially implemented for raytracing but
VK_EXT_graphics_pipeline_library requires it.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15737>

zink: use GENERAL layout for mixed zs fb attachments

this interaction requires read-only sampler access from
depth component with writes to the stencil component, which can only
be done in the GENERAL layout

affects:
GTF-GL46.gtf30.GL3Tests.framebuffer_blit.framebuffer_blit_functionality_color_and_stencil_blit

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: update samplerview layouts for zs attachments during renderpass prep

this interaction might require layout changes

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: use store op NONE when necessary for depth usage

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: delete some code in get_layout_for_binding()

should be no functional changes

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: add a ctx param to zink_descriptor_util_image_layout_eval

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: add a renderpass flag for mixed zs layout

this indicates that the layout requires reading from the depth aspect
and writing to the stencil aspect

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: further simplify zs case for zink_descriptor_util_image_layout_eval

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: remove commented code

probably never actually going to do this since it serves no purpose

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: refactor zink_descriptor_util_image_layout_eval

slightly more readable

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: use EXT_image_2d_view_of_3d

fixes #4562

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: unset resource layout+access when doing storage setup

the previous access info is transferred to the staging resource for the copy,
and the new image has no access info

cc: mesa-stable

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: prune shader i/o

to avoid validation spam when variables are declared but never used,
eliminate persisting i/o variables which are never used

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: fix max geometry input component advertising

forgot to divide by 4 somehow

cc: mesa-stable

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: convert all 64bit vertex attribs to 32bit

this applies not only to dvec3/dvec4, but to any type of 64bit input
since that's what gallium gives us

fixes #6211

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: apply fb attachment layout to dummy attachments

this doesn't actually matter but it still triggers validation spam

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: clamp min viewport width to 1

fixes validation spam

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: handle 1bit xor as OpLogicalNotEqual

fixes more validation spam

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: set Geometry capability for fs if geometry inputs are read

super legal.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: always set stencil dynamic states before draw

these are dynamic states set on the pipeline, so they must always
be set on a cmdbuf before draw

fixes some validation spam

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: merge stencil test case for draw-time dynamic state

these cases were identical but using different conditionals

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: only uncommit sparse pages that have been committed

avoid spamming drivers unnecessary with submits since this is expensive

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15716>

zink: fix barrier generation for ssbo descriptors

ssbo binds are always at least readable, so without deeper shader
inspection, never assume that write binds mean no read access

this is different than image descriptors which can explicitly get
write-only access set

cc: mesa-stable

fixes #6231

THANKS TO @daniel-schuermann FOR SOLVING THIS WITH HIS INCREDIBLE TALENT AND WIT

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15697>

Revert "venus: Increase the base sleep of vn_relax"

This reverts commit 737937f45e4231dd596780a856270693566de658.

Testing has revealed sizable performance drops arising out of this change.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15760>

panvk: Enable VK_EXT_debug_report and VK_EXT_debug_utils

They're both implemented in common code as long as you use
vk_command_buffer.

Acked-by: Emma Anholt <emma@anholt.net>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15560>

anv: Enable VK_EXT_debug_utils

It's implemented in common code as long as you use vk_command_buffer.

Acked-by: Emma Anholt <emma@anholt.net>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15560>

lavapipe: Enable VK_EXT_debug_utils

It's implemented in common code as long as you use vk_command_buffer.

Acked-by: Emma Anholt <emma@anholt.net>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15560>

turnip: Enable VK_EXT_debug_utils

It's implemented in common code as long as you use vk_command_buffer.

Acked-by: Emma Anholt <emma@anholt.net>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15560>

v3dv: Enable VK_EXT_debug_utils

It's implemented in common code as long as you use vk_command_buffer.

Acked-by: Emma Anholt <emma@anholt.net>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15560>

lavapipe: Switch to the common sync framework

The common Vulkan sync framework will do most of the queueing for us.
It will even sort out timeline semaphore dependencies and ensure
everything executes in-order. All we have to do is make sure our
vk_sync type implements VK_SYNC_FEATURE_WAIT_PENDING. This lets us get
rid of a big pile of code.

Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15651>

vulkan: Use timespec_add_nsec in vk_sync_timeline

Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15651>

util/timespec: Return overflow from timespec_add_[mn]sec()

To avoid altering any currently existing callers, we continue on with
the calculation regardless of overflow. This also matches the behavior
of GCC's __builtin_add_overflow().

Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15651>

gallivm/nir: Add a short circuit uniform-offset mode for load_ssbo/load_shared.

dEQP-VK.binding_model.buffer_device_address.set3.depth3.basessbo.convertuvec2.nostore.multi.scalar.vert
runtime -24.4002% +/- 1.94375% (n=7). The win (I think) is in LLVM not
having to chew through handling the extra loops on every constant-offset
SSBO load, not in actual rendering time.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14999>

gallivm/nir: Add a short circuit uniform-offset mode for load_global.

If we know the offset is constant, we don't have ask LLVM to loop over the
elements pulling the same value out over and over.

This doesn't seem to have produced a win in the testcase I was looking at,
but it was an easier entrypoint to figuring out how to do scalar memory
access than load_memory, and will probably affect some workload.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14999>

gallivm/nir: Refactor out some repeated code to generate 0 values.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14999>

gallivm/nir: Refactor out some repeated logic for SSBO/shared access.

I needed to be able to get these pointers/limits from another location,
and missing some of the repeated steps was giving me bugs.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14999>

gallivm/nir: Pull some repeated exec_mask computation out of loops.

If the exec mask hasn't changed, don't hassle LLVM to set it up

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14999>

gallivm/nir: Don't do uniform-and-broadcast access on inactive invocations.

In a fragment shader or inside of control flow, invocation 0 might be
inactive, and so our use-first-invocation-and-broadcast optimizations
would be invalid, and the loop logic of an emit_read_invocation would
defeat the point of these optimized paths.

For load_kernel_input, I didn't guard the uniform path with the check
because some CL tests that are currently passing are doing the
load_kernel_input under (presumably) uniform control flow. Instead, I
dropped in a once warning for the next person to be debugging CL.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14999>

util/log: Add support for logging once.

These are not data-race safe (like many other once patterns in Mesa), so
they might not log exactly once, but it should be good enough for not
spamming the console.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14999>

ci/lava: Simplify passthrough of the request to upload results/ to minio.

We already have a way to pass env vars around, just use that instead of
packing/unpacking it on the kernel command line.

Cleans up HW runner job log output some more.

Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15332>

ci/deqp: Move the set +e just before the deqp-runner invocation.

You don't want to proceed to running deqp-runner if you failed at the
vtest or runner options environment setup.

Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15332>

ci/deqp: Add gitlab-ci sections to deqp-runner.sh.

This should help highlight the actual test results, as opposed to the setup
and teardown.

Also tuned the "set -x"es a little bit so we get less surrounding noise in
the echo process.

Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15332>

spirv: Silence "Decoration not allowed on struct members: SpvDecorationRestrict"

VK-GL-CTS causes tons of these due to a bug in glslang, to the point where
it's hard to find actual issues in test logs. Disable the warning for
now, with a link to the issue we're waiting on being resolved.

Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15332>

util/log: Don't print an extra \n if the format string had one.

You're not supposed to include a '\n' in mesa_log*() messages because
android logging will log what you provide on its own line anyway, so each
mesa_log() should be the body of a log line. But also, getting everyone
to consistently not do that is hopeless because we're all so trained by
printf(). So, just detect an existing \n and don't add a new one.

Cleans up deqp-vk debug output a bunch from turnip.

Acked-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15332>

docs: update features for VK_EXT_image_2d_view_of_3d

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15757>

anv: expose VK_EXT_image_2d_view_of_3d

sampling only available on gen9+

Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15754>

lavapipe: expose VK_EXT_image_2d_view_of_3d

Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15754>

vulkan: check 3D image type for VK_IMAGE_CREATE_2D_VIEW_COMPATIBLE_BIT_EXT

Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15754>

tu: Expose VK_KHR_maintenance4

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15488>

tu: Remove tu_pipeline::layout

This makes it more obvious that the layout is never used after creating
the pipeline, which is required by VK_KHR_maintenance4.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15488>

tu: Fill out maxBufferSize

It seems this is really a workaround for silly issues in
GetBufferMemoryRequirements when you ask for a really large buffer. Just
expose the maximum possible size ATM.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15488>

tu: Implement GetDevice*MemoryRequirements()

Based mostly on anv, which is a bit more optimized than radv - we at
allocate the image on the stack.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15488>

ci: do not specify c_std and cpp_std for windows-build

When parts of the tree needs later c and c++ versions, they should ask
for it in the build-system itself, not expect the user to ask for it on
the command-line instead. So let's not paper over things by specifying
them here.

Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15706>

dozen: require c++20 for designated initializers

We do require C++20 still, because designated initializers is part of
that standard. This is almost a revert, but conditionally selecting
between c++latest or c++20 when available, as that's what we really want.

Fixes: 55ca1c8db37 ("vulkan/microsoft: Remove `override_options: ['cpp_std=c++latest']` option for visual studio")
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15706>

nir/tests: do not use designated initializers in c++ code

Designated initializers require C++20, which is a bit easier said than
done to support well across meson versions. Let's avoid using them
for now instead.

Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15706>

aco: do not use designated initializers

Designated initializers are a C++20 feature, but we don't use C++20. In
fact, enabling C++20 for ACO triggers new compiler errors due to some
equality semantics details.

So let's instead stop using designated initializers here.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15706>

frontends/va: fix decode issues introduced by efc change

Fixes: 96025265689 ("frontends/va: add encoder format conversion (EFC) support")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6153
Signed-off-by: Thong Thai <thong.thai@amd.com>
Tested-by: Andrew Falcon <bluestang2006@gmail.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15717>

radv: Remove radv_util.c

Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15725>

vulkan: spec update to 1.3.211

Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15752>

panfrost: Don't crash on panfrost_bo_create() with size==0 invocation

1. Clamp bucket_index from both ends to avoid returning negative index.
2. Return NULL in case BO allocation/fetching failure to prevent invalid
bo mapping.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6247
Signed-off-by: Roman Stratiienko <r.stratiienko@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15748>

radv: only declare dynamic states that are used by internal operations

Initialize some default static PSO states instead.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15729>

radv: use radv_dynamic_state for saving/restoring meta operations

Instead of duplicating every fields.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15729>

radv: save/restore more dynamic states during internal driver operations

This doesn't fix anything known but it could happen in theory.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15729>

radv/ci: update CI lists against CTS 1.3.1.1

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15745>

venus: add VK_EXT_{conditional_rendering,index_type_uint8} extensions

Implements all the necessary code in the device initialization
and extensions functions.

Signed-off-by: Igor Torrente <igor.torrente@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15652>

venus: Update venus-protocol to add two new extensions

These are the changes automatically generated from the venus-protocol
repository.

Update the file to add `VK_EXT_index_type_uint8` and
`VK_EXT_conditional_rendering`

Signed-off-by: Igor Torrente <igor.torrente@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15652>

venus: workaround an ANGLE assumption on FORMAT_IMPLEMENTATION_DEFINED

ANGLE expects VK_FORMAT_UNDEFINED to be returned for such AHB prop
query.

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15721>

ci: bump VK-GL-CTS to 1.3.1.1

Signed-off-by: Omar Akkila <omar.akkila@collabora.com>
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15668>

anv: Advertise two more formats

These both require swizzling so border colors won't work.  However,
they're conveniently in the list of formats for which custom border
colors require you to specify a format in the sampler.  That list
constists of:

    - VK_FORMAT_B4G4R4A4_UNORM_PACK16
    - VK_FORMAT_B5G6R5_UNORM_PACK16
    - VK_FORMAT_B5G5R5A1_UNORM_PACK16

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6226
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15624>

anv: Generalize border color swizzles

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15624>

anv: Disallow blending on swizzled formats

Fixes: c20f78dc5d7e ("anv: Support swizzled formats.")
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15624>

intel/isl: Add a helper for swizzling color values

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15624>

vulkan/queue: Destroy wait temps if they are skipped

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6223
Fixes: 8a11d2a31bf9 ("vulkan: Add a dummy sync type")
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Tested-by: Jakob Bornecrantz <jakob@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15727>

aco: don't use v_mad_mix on GFX9 if 16-bit denormals must be preserved

This probably effectively disables the v_mad_mix optimization on GFX9.

fossil-db (Vega):
Totals from 11545 (7.15% of 161366) affected shaders:
MaxWaves: 43025 -> 42780 (-0.57%); split: +0.06%, -0.63%
Instrs: 18571635 -> 18734201 (+0.88%); split: -0.00%, +0.88%
CodeSize: 96483568 -> 96611012 (+0.13%); split: -0.11%, +0.24%
SGPRs: 1079056 -> 1077616 (-0.13%); split: -0.14%, +0.01%
VGPRs: 819248 -> 821868 (+0.32%); split: -0.04%, +0.36%
SpillSGPRs: 13313 -> 12464 (-6.38%)
Latency: 293804093 -> 295046122 (+0.42%); split: -0.09%, +0.51%
InvThroughput: 110002239 -> 110994978 (+0.90%); split: -0.03%, +0.93%
VClause: 342458 -> 342596 (+0.04%); split: -0.12%, +0.16%
SClause: 648566 -> 648046 (-0.08%); split: -0.12%, +0.04%
Copies: 1728225 -> 1726679 (-0.09%); split: -0.66%, +0.57%
Branches: 552973 -> 552963 (-0.00%); split: -0.02%, +0.02%
PreSGPRs: 862360 -> 856820 (-0.64%); split: -0.69%, +0.05%
PreVGPRs: 773689 -> 776818 (+0.40%); split: -0.02%, +0.42%

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6178
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15718>

iris: Replace unconditional QBO flush with iris_dirty_for_history().

We can now use the same cache tracking mechanism for synchronizing QBO
writes instead of the unconditional PIPE_CONTROL performed currently,
which is unable to invalidate any incoherent caches which may contain
stale data for the buffer object.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15738>

iris: Remove remaining history flushes.

This removes a couple of remaining history flushes which were
open-coded instead of using the iris_flush_and_dirty_for_history()
helper.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15738>

iris: Demote all callers of iris_flush_and_dirty_for_history() to iris_dirty_for_history().

The unconditional flushing performed by
iris_flush_and_dirty_for_history() is now redundant with the memory
barriers introduced previously in this series, which should be in a
better position to determine from which domain the buffer will
actually be used in the future, and whether an additional flush or
invalidation is required or redundant with other PIPE_CONTROL commands
emitted elsewhere.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15738>

aux/trace: dump format in set_shader_images

Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15323>

ci/freedreno: Reduce concurrency when replaying traces on a630

We are running out of memory when replaying traces sometimes, reduce the
number of concurrent retrace processes.

   Mesa: User error: GL_OUT_OF_MEMORY in glReadPixels
   warning: GL_OUT_OF_MEMORY while getting snapshot
   1074335: warning: failed to get snapshot

https://gitlab.freedesktop.org/mesa/mesa/-/jobs/20519522

Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15682>

v3d: fix some leaks in cache

Fix a couple of leaks introduced when adding support for on-disk shader
cache.

Fixes: 4468db20f7f ("v3d: add support for on-disk shader cache")
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15733>

v3dv: fix limits for inline uniform blocks

We don't support 'Update After Bind', however, the limits for this
model also include the ones without it. See the with or without remark
in the spec below:

"maxPerStageDescriptorUpdateAfterBindInlineUniformBlocks is similar to
maxPerStageDescriptorInlineUniformBlocks but counts descriptor bindings
from descriptor sets created with or without the
VK_DESCRIPTOR_SET_LAYOUT_CREATE_UPDATE_AFTER_BIND_POOL_BIT bit set."

Fixes:
dEQP-VK.api.info.vulkan1p2_limits_validation.ext_inline_uniform_block

Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15732>

Revert "ci/panfrost: Disable some jobs due to a lab failure"

Machines are back.

This reverts commit b5fd1fddd9cf24e9250b659130057289d5080746.

Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15730>

radv: save/restore the stencil reference during internal driver operations

I think I should improve this to be more robust.

Gitlab: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6243
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15700>

radv: fix cleaning the image view for CmdCopyImageToBuffer()

Fixes: f07e67272e8 ("radv: fix vk_object_base_init/finish for internal image views")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15707>

pvr: Add stricter type checking in pvr_csb_pack().

Since the packing functions generated by csbgen use a void pointer
for the buffer in which to pack, it's possible to easily write out
of bounds. This commits attempts to reduce the chances by
having the pack macro check that the pointer passed points to an
element sized equally to the state word being packed. Catching
these errors earlier.

As can be seen in this commit, there already was a case of this:
"pds_ctrl". The word size is meant to be 64 bits but the pointer
was pointing to a 32 bit field.

Although it's fine for the word size to be smaller than the
storage pointed to by the pointer, this is not allowed just to
be extra careful.

Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15687>

radv: Replace magic constants with enum values

Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15722>

radv: Advertise ray primitive culling

Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15722>

radv: Fully implement ray primitive culling

Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15722>

turnip: Allow image access on swapped formats.

This is apparently something that gamescope would like to have, and the
CTS's test coverage is happy with it.

Fixes: #6011 (we hope)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15293>

turnip: Disable tiling on 1D images.

If we know the height is 1, then it would be a waste to align each
miplevel to tile height. For non-mipmapped textures, it doesn't save us
memory (since you still align to 4 on the last miplevel), but it should be
better cache locality by not loading those unused lines.

Incidentally, this gets us some more coverage of swap != WZYX cases in CTS
tests, which often use optimal tiling without also testing linear.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15293>

freedreno/a6xx: Set the color_swap field for storage descriptors.

This field does appear to work as expected: with 1D/1DArray turnip storage
images switched to be always linear, it fixes the dEQP-VK.image.*store*
tests using a color swapped format (once we allow color swap).

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15293>

turnip: Add support for VK_KHR_format_feature_flags2.

This reports all of our storage formats as supporting read/write without
format, since we don't have any in-shader format conversions. Similarly,
shadow comparisons were already supported on all the depth formats.

This extension is required for VK 1.3.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15293>

nir_to_tgsi: Add support for nir_intrinsic_image_samples.

Found in 1 piglit test on r600.

Reviewed-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15627>

lima/ci: enable piglit in lima CI

There are spare boards for it and the results appear stable.
Dividing the load in 2 parallel runs results in a run time
of around 10 minutes each.

Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15598>

lima/ci: enable CI again

Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15598>

lima/ci: update deqp results

Unfortunately some regressions sneaked in while CI was down.

Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15598>

turnip: Use correct type for OUTARRAY in FormatProperties2

Fixes: 799a9db24c334713c0ba5995f59b7db5f782e85e
("turnip: Stop using VK_OUTARRAY_MAKE()")

Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15694>

freedreno/registers: update dsi registers to support dsc

Display Stream compression (DSC) compresses the display stream in
host which is later decoded by panel. This requires addition of 3 new
DSI registers to support DSC over DSI.

Signed-off-by: Vinod Koul <vkoul@kernel.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14967>