zink: generate instance creation code with a python script

Acked-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7800>

zink: factor out GET_PROC_ADDR and friends to zink_screen.h

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7800>

zink: allow Extension/Version to be shared across files

Acked-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7800>

mesa: check for extension instead of desktop GL

If we ever decide to allow this extension for GLES, this will make life
easier.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8028>

mesa: do not allow es2-extension enums for es1

Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8028>

.gitlab-ci: verify that Get-Content worked

This will hopefully prevent future breakages similar to what the
previous commit fixed.

Reviewed-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8052>

gitlab-ci: copy piglit expected results to artifacts

I somehow missed this while reworking the CI bits to reduce bandwidth
usage. This emits an error on stdout, but doesn't actually fail the
tests.

Fixes: 195a001d736 ("gitlab-ci: do not clone git-repo for test-job")
Reviewed-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8052>

android: fix build failure with libbacktrace

because 848e7b94 commit cause.it  modify u_debug_stack_android.cpp
location from src/gallium/auxiliary/util to src/util but Android.mk
not modify

Fixes: 848e7b94 ("Move stack debug functions to src/util")
Signed-off-by: cheyang <cheyang@bytedance.com>
Acked-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7851>

v3d: extend the list of formats supported by the TFU unit

TFU can convert from these new formats, but can not autogenerate
mipmaps from them.

Hence we need to set what is the purpose to know if the formats are
supported or not.

v1:
 - Use the same and shorter variable name (Alejandro)

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8050>

panfrost: Fix several depth/stencil format mappings

Stencil should always be loaded in component Y, even when the buffer
only contains stencil values. Patch the blit shaders accordingly, and
fix some format definitions.

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7922>

gallium/util: Fix depth/stencil blit shaders

When loading the depth, we want to store component X of the texel fetch
result into position.Z which can't be expressed without an extra MOV
unless the backend replicates the depth.

Stencil is always expected in the Y component, but some TGSI shaders
assume it will also be available in X, which only works if the backend
replicates the stencil value.

Let's fix those shaders so backend drivers are not forced to replicate
the depth/stencil values.

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7922>

panfrost: Reload depth/stencil when they are read

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7922>

panfrost: Force ->s_writeback_base to ->zs_writeback_base for Z24S8 buffers

Apparently this field is used even if the stencil is part of the ZS
buffer. Fixes a bunch of page faults on Bifrost.

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7922>

panfost: Fix depth/stencil writeback on Bifrost v7

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7922>

panfrost: Allow 2DMS arrays

The midgard compiler has been fixed to handle texture operations on
2DMS arrays, and Bifrost already had this case handled properly. Relax
the assert.

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8042>

pan/mdg: Fix texture handling for 2DMS arrays

While at it, move the coordinate building logic to a dedicated helper
and make it generalize it to handle 1D arrays too.

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8042>

vulkan: Enable VK_KHR_performance_query on android

The tests of this extension seem to be added to Android 11 CTS,
since they can be found on android-11-x branches.

Signed-off-by: Hyunjun Ko <zzoon@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7602>

intel/blorp: Drop support for STC_CCS resolves

There are no users of this feature.

Reverts the following commits:

* 87c57b8dae954c7de6a499506a4161f69076ce67 (effectively)
* 53d472df24d4bc05359120df0615321318922d6d
* 9ab0e92cff8bb8336cebd9dc68d02d7451ad78e1

Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8021>

iris: Stop using blorp_hiz_stencil_op

Delete this function call for two reasons:

1. Calling blorp_hiz_stencil_op within iris_resolve_color seems out of
   place.
2. AFAICT, iris stopped doing STC_CCS resolves with commit
   5b82d8ce8bc681905a8902902fb89151ca32d6c9.

Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8021>

CI: add lavapipe vulkan testing

Acked-by: Michel Dänzer <mdaenzer@redhat.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7952>

st/mesa: Replace mesa_to_tgsi() with prog_to_nir() and nir_to_tgsi().

This introduces NIR optimizations to ARB programs in all gallium drivers,
while deleting code.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7927>

ci: Make sure that osmesa stays warnings-clean in release builds.

Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/1243>

mesa: Retire classic OSMesa.

The classic OSMesa renders directly into user memory using
src/mesa/swrast, while gallium OSMesa renders using softpipe or llvmpipe
and copies out at glFlush() time.  This would make gallium look like a
worse choice for OSMesa, except that swrast is:

1) Painfully slow to render compared to llvmpipe
2) Incorrect at derivatives
3) Limited to GL 2.1 instead of GL 4.6

In my survey of OSMesa users, debian was the remaining holdout with
classic OSMesa in use on hurd and some rare non-LLVM-supported
architectures (sh4, alpha, etc.).  As of today, they've switched to
softpipe-based gallium OSMesa for them.

To prevent people from running the wrong OSMesa (to the extent that
running OSMesa can ever be the right thing), delete the classic
version.

Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Closes: #320
Closes: #877
Closes: #2297
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/1243>

docs: Fix the documentation of the OSMesa path.

Typo in gallium, along with the fact that gallium's osmesa installs to the
same directory as classic's.

Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/1243>

osmesa/test: Clear the stencil bits in the depth test.

Since the depth buffer starts out as a malloc, and we weren't clearing it,
you could get undefined values in your top 8 bits.  This should fix
intermittent failures of the depth test.

(Sadly, valgrind wasn't catching this, presumably because the 32-bit value
there *is* written, just some bits are left undef)

Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/1243>

aco: Skip TCS s_barrier when VS outputs are not stored in the LDS.

When VS outputs are known to be never stored in LDS, there is no
reason for HS waves to wait for all LS waves to complete. So, the
s_barrier between the LS and HS can be safely skipped.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7727>

util+treewide: container_of() cleanup

Replace mesa's slightly different container_of() with one more aligned
to the linux kernel's version which takes a type as the 2nd param.  This
avoids warnings like:

  freedreno_context.c:396:44: warning: variable 'batch' is uninitialized when used within its own initialization [-Wuninitialized]

At the same time, we can add additional build-time type-checking asserts

Signed-off-by: Rob Clark <robdclark@chromium.org>
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7941>

util: Allow STATIC_ASSERT() everywhere

Remove -Werror=vla from c_msvc_compat_args so we can use STATIC_ASSERT()
in core code.  We have a CI job for this.

(And arguably we could probably just drop c_msvc_compat_args entirely.)

Signed-off-by: Rob Clark <robdclark@chromium.org>
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7941>

util: Promote __builtin_types_compatible_p compat

Signed-off-by: Rob Clark <robdclark@chromium.org>
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7941>

swr: fix crashes caused by incorrectly reporting SSBO support

Reviewed-by: Krzysztof Raszkowski <krzysztof.raszkowski@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8036>

ci: Only run the sanity job if there's a MR

As it will fail right away if there isn't, and that prevents the CI to
run on people's branches.

$ ci-fairy check-merge-request --require-allow-collaboration --junit-xml=check-merge-request.xml
ERROR: No open merge request against mesa/mesa with sha 9f6aba4be0ceb59d35732651cc44f676e93b70b5

Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Fixes: d4151f2e ("ci: Run sanity job only in pre-merge pipelines")
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8032>

zink: remove useless import in zink_device_info.py

Acked-by: Adam Jackson <ajax@redhat.com>
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7756>

zink: move blend_operation_advanced conditions to zink_device_info.py

Acked-by: Adam Jackson <ajax@redhat.com>
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7756>

zink: decouple features and enabling conditions in zink_device_info.py

replace `have_feature` with `conditions` and add `features=True`
to places where `have_feature` was used.

Acked-by: Adam Jackson <ajax@redhat.com>
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7756>

zink: define and use <%guard> helper in zink_device_info

this is an attempt to improve readability of the template code, with
the elimination of excessive whitespace as a bonus.

Acked-by: Adam Jackson <ajax@redhat.com>
Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7756>

ci: Move deploy stage to the end of the pipeline

This should make more important jobs visible without scrolling on
pipeline pages.

The deploy stage jobs only depend on the sanity job or none at all, so
this has no impact on when the former can run.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7926>

ci: Run sanity job only in pre-merge pipelines

It's more of a nuisance than useful for forked branches.

This means the test-docs job can no longer have a direct dependency on
sanity for forked branches, so split it up into two jobs: one for
pre-merge pipelines, one for forked branches.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7926>

egl/wayland: remove libwayland < 1.18 workaround

Require libwayland 1.18 and remove the workaround for
WL_SHM_FORMAT_{A,X}BGR16161616F.

Signed-off-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7771>

zink: use _mesa_pointer_set_create for simplicity

Acked-by: Adam Jackson <ajax@redhat.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7985>

zink: fail if set failed to create

Fixes: a03d17ede77 ("zink: refcount zink_gfx_program objects")

Acked-by: Adam Jackson <ajax@redhat.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7985>

gallium/u_threaded: set has_user_indices = false for merged draws

Fixes an assert when running this piglit test:
   arb_shader_draw_parameters-drawid-single-draw

Fixes: 351ba767afe ("gallium/u_threaded: set has_user_indices = false in the driver thread")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8004>

radeonsi: fix si_get_draw_start_count count value

Fixes: 0ce68852c1a ("radeonsi: implement multi_draw but supporting only 1 draw")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3932
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8004>

ci: .lava-test:amd64 template needs arm_build

It uses the arm_build image, but didn't depend on the job which ensures
it exists. So jobs using the template could run before the arm_build job
had finished, and fail if the image didn't exist.

Fixes: 6c8b921572a5 "ci: Build kernels and rootfs for x86 devices"
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3632
Reviewed-by: Andres Gomez <agomez@igalia.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8003>

docs: Adapt to FDO_DISTRIBUTION_TAG → MESA_IMAGE_TAG rename

Fixes: 0781d9825b31 "ci: Append $MESA_TEMPLATES_COMMIT to image tags"
Reviewed-by: Andres Gomez <agomez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7957>

ci: Adapt armhf_test job to MESA_TEMPLATES_COMMIT related changes

I missed this before somehow.

Fixes: 0781d9825b31 "ci: Append $MESA_TEMPLATES_COMMIT to image tags"
Reviewed-by: Andres Gomez <agomez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7957>

ci: Add .use-base-image template

And use it in jobs for images using another Mesa image as their base.

Should fix the build of images which don't use another Mesa image as
their base (by no longer setting the FDO_BASE_IMAGE variable).

Fixes: 0781d9825b31 "ci: Append $MESA_TEMPLATES_COMMIT to image tags"
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Andres Gomez <agomez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7957>

ci: Move BASE_TAG expansion to FDO_BASE_IMAGE assignment

This fixes the build of images which use another Mesa image as the base.

v2:
* Move $CI_REGISTRY_IMAGE expansion into FDO_BASE_IMAGE assignment as
  well (Dave Airlie)

Fixes: 0781d9825b31 "ci: Append $MESA_TEMPLATES_COMMIT to image tags"
Reported-by: Dave Airlie <airlied@redhat.com>
Acked-by: Eric Anholt <eric@anholt.net> # v1
Reviewed-by: Andres Gomez <agomez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7957>

anv: fix calculation of buffer size in case dynamic size is used

VK spec got clarification about the pSizes parameter.

Fixes set of new tests:
   dEQP-VK.pipeline.extended_dynamic_state*with_offset*

v2: move offset subtract to be part of size calculation (Jason)

CC: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3871
Fixes: b9a05447a19 ("anv: dynamic vertex input binding stride and size support")
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/7439>

lavapipe: implement VK_EXT_vertex_attribute_divisor (v2)

This is more or less just compile-tested, but this seems about right to
me. I see the extension being supported when running on top of Zink,
which makes me happy enough for now ;)

v2: fixed up to copy the structs on pipeline create [airlied]
gallium doesn't support the 0 divisor case yet.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7930>

lavapipe: interpret inputRate as an enum-value

This isn't an integer, it's an enum, We need to support
VK_EXT_vertex_attribute_divisor if we want to support using values other
than 0 and 1 here.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7930>

lavapipe: don't copy pNext

We have to ignore unknown structs, so don't init the deep copy pNext
with the src.

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7978>

lavapipe: split out pipeline struct duplication to a macro.

This just pulls a common pattern into a macro.

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7978>

lavapipe: use ralloc for pipeline copies.

We have to store every struct that is in the pipeline for processing
later, use ralloc to make freeing this mess easier.

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7978>

iris: Drop res variable in resolve_sampler_views

Instead of storing isv->res in a local variable, just use it directly to
increase consistency with similar field accesses.

Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7996>

iris: Fix resource ptr in resolve_sampler_views

Use the sampler view's iris_resource instead of its pipe_resource. For
stencil views of a depth-stencil resource, this enables the stencil
resource to cause a depth cache flush when needed.

Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3287
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7996>

softpipe: Fix swizzled texture gather of int textures.

We need to pick 1u vs 1.0f based on the type of the texture, just like for
normal samples.  Move the decision up to the create_sampler_view, and use
that value from both sampler paths.

Cc: mesa-stable
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8012>

radeonsi: use a C++ template to decrease draw_vbo overhead by 13 %

With GALLIUM_THREAD=0 to disable draw merging.

Before:
   1, DrawElements ( 1 VBO| 0 UBO|  0    ) w/ no state change,                 8736

After:
   1, DrawElements ( 1 VBO| 0 UBO|  0    ) w/ no state change,                10059

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7807>

radeonsi: rename si_state_draw.c to .cpp

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7807>

radeonsi: resolve a tricky C++ failure with goto jumping over initializations

C++ doesn't allow jumping over variable initializations, so we have to use
a macro.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7807>

radeonsi: fix future C++ compile failures and warnings

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7807>

radeonsi: allow including a few files from C++

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7807>

amd/llvm: fix C++ compile failures

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7807>

gallium/util: allow including a few files in C++

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7807>

compiler: fix glsl_types.h compile failures when including as C++ in drivers

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7807>

ac/llvm: handle no_(un)signed_wrap NIR flags

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7939>

ac: unify shader arguments that are duplicated

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7939>

radeonsi: move si_llvm_compiler_shader and deps into si_shader_llvm.c

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7939>

radeonsi: move si_build_main_function into si_shader_llvm.c

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7939>

radeonsi: move si_create_function into si_shader_llvm.c

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7939>

radeonsi: split ac_shader_args initialization from LLVM code

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7939>

ac: add shader return values into ac_shader_args

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7939>

ac: correct ac_shader_args types, remove sgpr_count

sgpr_count is unused. The size of the others is too small.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7939>

radeonsi: fix a crash in si_fence_server_sync

Fixes: 1f31a216640 - radeonsi: remove SDMA support

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8010>

aco: rename s_subb_u32 operands to borrow

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8007>

aco: fix various s_subb_u32 operands to SCC

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8007>

v3d: use job's nr_cbufs field

Instead of computing the number of color buffers in the start binning
function, just use the job's nr_cbufs, which already contains that
value.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8006>

panfrost: Fix provoking vertex selection for lines

For line primitives, the provoking vertex selection is done through the
DRAW.flat_shading_vertex field and PRIMITIVE.first_provoking_vertex must
be set to true.

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7923>

panfrost: Preload SampleID when reloading multisample FBs

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7923>

panfrost: Take the number of samples into account in blend shaders

Midgard has to split the writeout instruction if the number of bits per
pixel exceeds 128. We thus need to take the number of samples into
account when creating blend shaders.

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7984>

pan/mdg: Add support for multi sample iteration writeout

Some MSAA+fmt combination require writeout to be split. Right now, it
only impacts blend shaders since we only support MSAA 4x, and the only
formats that could exceed the 128bit/pixel limit in MSAA 4x are
not supported by the fixed-function blend unit. We thus rely on the
blend shader to split things properly. Things will change once we add
MSAA 8x/16x to the mix, since even the blendable formats will exceed
the 128b/pixel limit in that case.

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7984>

panfrost: Fix fencing

Commit 64d6f56ad26f ("panfrost: Allocate syncobjs in panfrost_flush")
aimed at optimizing the fencing logic but it looks it also broke the
fence-based synchronization in subtle ways.

Indeed, now that the fence only waits on a single syncobj, we're not
guaranteed that all jobs queued in panfrost_flush_all_batches() will
be done when the fence is signaled, because jobs at the top level
(those stored in the batches hashmap) have not inter-dependencies.

Commit 9e397956b092 ("panfrost: signal syncobj if nothing is going to
be flushed") made this even more apparent by signaling the fence right
away if nothing was left to be drawn in the current context, thus
ignoring any of the batches left to flushed in the ->batches map.

If we want to keep relying the existing kernel APIs there's clearly no
ideal solution here. We can either go back to the original fencing
mechanism where each fence contained an array of syncobjs to be tested
or serialize jobs that have no explicit dependencies so we know the last
submitted job will also be the last one to return. The orginal approach
has proven to add quite a significant overhead (caused by the amount of
ioctls and the time spent in kernel space to gather dma fences attached
to those syncobjs and test them). So let's go for the simple solution
where we have a single syncobj bound to the context which we update to
point to the last job out_sync every time we submit a top-level job.

This approach implies reworking the way we create fences since we
need to capture the syncobj state at the time the fence is created.
Unfortunately, there's not SYNCOBJ_CLONE ioctl, which forces us to
export/create/import a fence so we have a new object that's not
subject to changes done to the context syncobj.

If we want to further optimize the logic, we should probably explore
some of those options:

1/ Adding array based SYNCOBJ ioctls (SYNCOBJ_{CREATE,DESTROY,CLONE}_ARRAY)
   so we can mitigate the cost of ioctls when we need to manipulate
   arrays of syncobjs
2/ Support synchronization jobs. That is, jobs that have a NULL job chain
   but an array of sync_in and a sync_out to allow creating
   synchronization points
3/ Add syncobj aggregators so we only have to wait on one syncobj from
   userspace. The syncobj aggregator would wait for all sub syncobjs to
   be signaled before signaling the top-level one.

Fixes: 64d6f56ad26f ("panfrost: Allocate syncobjs in panfrost_flush")
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7831>

panfrost: Make sure we always add a reader -> write dependency when needed

We shouldn't reset the ->writer field when a reader comes in because we
want subsequent readers to have a dependency on the writer too. Let's
add a new field encoding the last access type and use it to replace the
writer != NULL test.

Reported-by: Roman Elshin
Fixes: c6ebff3ecdde ("panfrost: Remove panfrost_bo_access type")
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7831>

st/mesa: enable compute shader derivatives in SPIR-V

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6799>

ac,radeonsi: implement GL_NV_compute_shader_derivatives

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6799>

winsys/amdgpu: use VRAM for command buffers if all VRAM is visible

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7951>

radeonsi: map PIPE_USAGE_STREAM to VRAM if all VRAM is visible

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7951>

radeonsi: unify uploaders and upload to VRAM if all VRAM is visible

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7951>

radeonsi: only use staging for linear textures when all VRAM is not visible

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7951>

radeons: only force staging uploads for VRAM when all VRAM is not visible

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7951>

ac: add radeon_info::all_vram_visible for Smart Access Memory

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7951>

radeonsi: initialize ctx and gfx_cs first, then allocators

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7951>

compiler: update gl_access_qualifier comments

Clarify ACCESS_NON_{READABLE,WRITEABLE} and update ACCESS_CAN_REORDER.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6483>

nir/opt_access: infer writeonly

This isn't always done for GL because it could cause ImageAccess to be
incorrect: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3278

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6483>

nir/opt_access: consider global stores

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6483>

nir/opt_access: handle variable pointers

We might not be able to determine what variable a instruction accesses.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6483>

nir/opt_access: add basic Vulkan support

This involves determining the variables referenced by intrinsics, setting and
using the access qualifier correctly and considering that images and buffers
can alias.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6483>

nir/opt_access: rename can_reorder() and set ACCESS_NON_WRITEABLE in it

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6483>

nir/opt_access: don't check restrict in can_reorder()

ACCESS_NON_WRITEABLE means that the memory is read-only, not the variable,
so we don't have to check for aliasing.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6483>

nir/opt_access: check restrict before marking a variable as readonly

ACCESS_NON_WRITEABLE means that the memory is read-only, not the variable,
so we have to check for aliasing first.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6483>

nir/opt_access: ignore barriers and coherent qualifier

It shouldn't matter whether an access/variable is coherent or not, just
that it's not written. The coherent qualifier doesn't mean anything with
read-only memory.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6483>