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

device_select_layer: fix inverted strcmp in device_select_find_dri_prime_tag_default (v1)

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7498
Fixes: 9bc5b2d169d ("vulkan: add initial device selection layer. (v6.1)")

Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19101>

svga: use util_unreference_framebuffer_state to unref fb state

Fixes: f5bde99cbdd ("gallium: plumb resolve attachments through from frontends -> pipe_framebuffer_state")

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22673>

llvmpipe: use util_unreference_framebuffer_state to unref fb state

Fixes: f5bde99cbdd ("gallium: plumb resolve attachments through from frontends -> pipe_framebuffer_state")

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22673>

vc4: use util_unreference_framebuffer_state to unref fb state

Fixes: f5bde99cbdd ("gallium: plumb resolve attachments through from frontends -> pipe_framebuffer_state")

Reviewed-by: Eric Engestrom <eric@igalia.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22673>

v3d: use util_unreference_framebuffer_state to unref fb state

Fixes: f5bde99cbdd ("gallium: plumb resolve attachments through from frontends -> pipe_framebuffer_state")

Reviewed-by: Eric Engestrom <eric@igalia.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22673>

softpipe: use util_unreference_framebuffer_state to unref fb state

Fixes: f5bde99cbdd ("gallium: plumb resolve attachments through from frontends -> pipe_framebuffer_state")

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22673>

iris: use util_unreference_framebuffer_state to unref fb state

Fixes: f5bde99cbdd ("gallium: plumb resolve attachments through from frontends -> pipe_framebuffer_state")

Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22673>

i915: use util_unreference_framebuffer_state to unref fb state

Fixes: f5bde99cbdd ("gallium: plumb resolve attachments through from frontends -> pipe_framebuffer_state")

Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22673>

i915: use util_copy_framebuffer_state to set fb state

Fixes: f5bde99cbdd ("gallium: plumb resolve attachments through from frontends -> pipe_framebuffer_state")

Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22673>

cso: unbind fb state when unbinding the context

this avoids leaking/ballooning references

cc: mesa-stable

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22673>

ci: carry venus-protocol 1.0 release patches in virglrenderer

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22547>

docs: update Virtio-GPU Venus driver page

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22547>

meson/ci: promote virtio-experimental to virtio

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22547>

venus: adopt venus protocol release

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22547>

venus: stop query experimental features

Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22547>

ci/zink+anv: Test piglit quick_gl pre-merge, dropping a few KHR-GL46 tests.

piglit provides way more important coverage than these bits of GL46, and
has been regressing regularly. Thanks to dropping the long KHR-GL46
tests, this is now completing faster than it was before, and things will
get even better with
https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/804.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22811>

ci/zink: Update more xfails for tgl piglit.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22811>

Revert "ci/zink: Try to update TGL results for new MSAA behavior."

This reverts commit 74a8f118a2595479a94a8b7097680c5111d7273c.

This should have been in !22753.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22811>

venus: use pipelineCacheUUID for shader cache id

Use the host driver pipelineCacheUUID as the shader cache id. This is
just used as an identifier for fossilize replay since venus utilizes
the host side shader cache anyways.

Signed-off-by: Juston Li <justonli@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22634>

zink: sync queries at the end of cmdbufs

with tc, query results can be fetched async, and it's impossible to
sync tc in this scenario. to avoid needing to sync when a sync is not
possible, sync ahead of time in all cases

Fixes: 7c96e989755 ("zink: always start/stop/resume queries inside renderpasses")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22827>

zink: null some descriptor buffer pointers during destruction

this shouldn't actually be needed? and yet...

Fixes: 13c6ad0038a ("zink: use a single descriptor buffer for all non-bindless types")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22827>

zink: set depth dynamic state values unconditionally

spec requires the state values to be set if the dynamic state is
enabled regardless of whether they will take effect

cc: mesa-stable

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22827>

nouveau/nir: image_samples/size don't have coordinates

Without this, it treats the src[1] as a coordinate (it's actually LOD)
and may try to read more than one component. I don't think this usually
hurts anything as the coordinate should get ignored later but it can
result in OOB memory reads while translating NIR.

Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22834>

zink: disable a630 traces

I can't even run a fair number of these locally without a gpu hang,
and there are no validation errors, so just disable the whole thing
until at least #8874 can be resolved

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22831>

nv50/codegen: Set lower_uniforms_to_ubo

Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22815>

panfrost: Check blend enabled state in pan_allow_forward_pixel_to_kill()

The shader can write to a specific RT, but the blend descriptor gets
to decide if the RT is really updated. We need to take that into
account when initializing the rt_written local variable in
pan_allow_forward_pixel_to_kill() otherwise we might get inconsistent
results.

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Suggested-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22465>

ci/zink: Disable Freedoom trace on ANV

Apparently it's flaky.

Signed-off-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22819>

radv/ci: document a series of recent regressions

No time to bisect all the regressions, so I am just including the
GOOD/BAD commits if anyone is interested to bisect.

One thing is for sure, !22753 is responsible for at least tens of
these issues.

Signed-off-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22820>

wsi/wayland: Do not assert that all present IDs have been waited on.

EXT_swapchain_maintenance1 allows destruction when the swapchain fence
is complete. That may signal before we have dispatched all present
waits. Just discard those presentation feedbacks.

Signed-off-by: Hans-Kristian Arntzen <post@arntzen-software.no>
Reviewed-by: Joshua Ashton <joshua@froggi.es>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22682>

wsi/wayland: Simplify wait logic for present wait.

CLOCK_REALTIME is generally problematic due to NTP.
Use normal MONOTONIC waits for condition variable,
and remove the timedlock. The lock is never held in a blocking fashion,
so there is little need for a timed lock here.

Signed-off-by: Hans-Kristian Arntzen <post@arntzen-software.no>
Reviewed-by: Joshua Ashton <joshua@froggi.es>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22682>

wsi/wayland: Support VK_KHR_present_wait

Use the wp_presentation extension to detect when a given presentation
has taken effect.

Since this protocol is not guaranteed to be supported,
it must be enabled through driconf (vk_khr_present_wait=true) for the time being.

Signed-off-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Hans-Kristian Arntzen <post@arntzen-software.no>
Co-authored-by: Hans-Kristian Arntzen <post@arntzen-software.no>
Reviewed-by: Joshua Ashton <joshua@froggi.es>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22682>

wayland: generalize wayland-protocols code generation

Make it easy to add a new protocol to the list without duplicating
logic.

Signed-off-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Eric Engestrom <eric@igalia.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22682>

radv: Add driconf to always drain waves before writing timestamps

UE4's Vulkan backend uses vkCmdWriteTimestamp with TOP_OF_PIPE
to measure how long a workload took in the GPU Benchmark. This is wrong
and writes the timestamp before the workload is actually finished,
making it seem like the GPU is much faster than it actually is.
This caused subsequent benchmark passes to contain way too big workloads,
which caused soft hangs on slower GPUs.

Fixes GPU hangs with Splitgate during automatic settings configuration.

Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22823>

radv: Always call si_emit_cache_flush before writing timestamps

In case of barriers with TOP_OF_PIPE as dst stage, writing timestamps
immediately would be invalid.

Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22823>

radv/meta: rename dest to dst

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22794>

radv: replace radv_get_levelCount() by vk_image_subresource_level_count()

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22794>

radv: replace radv_get_layerCount by vk_image_subresource_layer_count()

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22794>

radv: fix fast-clearing images with VK_REMAINING_{ARRAY_LAYERS,MIP_LEVELS}

Found by inspection.

Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22794>

radv: fix invalid type for usage in radv_get_buffer_memory_requirements()

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22794>

broadcom/compiler: move buffer loads to lower register pressure

If we are trying to lower register pressure this can make a big
difference in some cases. To avoid adding even more strategies,
merge this with disabling ubo load sorting, since they are basically
trying to do the same.

total instructions in shared programs: 12848024 -> 12844510 (-0.03%)
instructions in affected programs: 236537 -> 233023 (-1.49%)
helped: 195
HURT: 87
Instructions are helped.

total uniforms in shared programs: 3815601 -> 3814932 (-0.02%)
uniforms in affected programs: 31773 -> 31104 (-2.11%)
helped: 67
HURT: 115
Inconclusive result (value mean confidence interval includes 0).

total max-temps in shared programs: 2210803 -> 2210622 (<.01%)
max-temps in affected programs: 9362 -> 9181 (-1.93%)
helped: 114
HURT: 34
Max-temps are helped.

total spills in shared programs: 2556 -> 2330 (-8.84%)
spills in affected programs: 1391 -> 1165 (-16.25%)
helped: 39
HURT: 9

total fills in shared programs: 3840 -> 3317 (-13.62%)
fills in affected programs: 2379 -> 1856 (-21.98%)
helped: 39
HURT: 23

total sfu-stalls in shared programs: 21965 -> 21978 (0.06%)
sfu-stalls in affected programs: 2618 -> 2631 (0.50%)
helped: 45
HURT: 81
Inconclusive result (value mean confidence interval includes 0).

total inst-and-stalls in shared programs: 12869989 -> 12866488 (-0.03%)
inst-and-stalls in affected programs: 238771 -> 235270 (-1.47%)
helped: 193
HURT: 87
Inst-and-stalls are helped.

total nops in shared programs: 303501 -> 303274 (-0.07%)
nops in affected programs: 4159 -> 3932 (-5.46%)
helped: 87
HURT: 105
Inconclusive result (value mean confidence interval includes 0).

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22824>

docs: fix edit-links

When making a custom-theme, I accidentally hard-coded the edit-URL to
point to the index-file in the mesa3d.org repo instead of pointing to
the correct file in our docs. This fixes that, so the "Edit this
page"-links in the footer works the same way as the old "Edit on
GitLab"-links did.

Fixes: 7da0482636a ("docs: add custom html theme")
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22791>

turnip: fix buffer markers using wrong addresses

Fixes failures in dEQP-VK.api.buffer_marker.*

Signed-off-by: Amber Amber <amber@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22792>

radv: Move all the dirty flags from TES binding to TCS binding.

With merged shaders we might not have an explicit TES.

Fixes: 879ddf97202 ("radv: rework binding shaders to cmdbuf by introducing new helpers")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8939
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22784>

radv: use lower_ballot_bit_count_to_mbcnt_amd

Foz-DB Navi21:
Totals from 62 (0.05% of 134864) affected shaders:
VGPRs: 2464 -> 2440 (-0.97%)
CodeSize: 332408 -> 324276 (-2.45%)
MaxWaves: 1690 -> 1692 (+0.12%)
Instrs: 62356 -> 60828 (-2.45%)
Latency: 595723 -> 592554 (-0.53%)
InvThroughput: 126106 -> 124241 (-1.48%)
SClause: 2163 -> 2162 (-0.05%)
Copies: 6392 -> 6226 (-2.60%); split: -2.94%, +0.34%
Branches: 2295 -> 2298 (+0.13%)
PreSGPRs: 2390 -> 2389 (-0.04%)
PreVGPRs: 2139 -> 2117 (-1.03%); split: -1.08%, +0.05%

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22783>

nir: lower ballot_bit_count_exclusive/inclusive to mbcnt_amd

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22783>

intel/compiler: make uses_pos_offset a tri-state

This value depends on the per-sample value which can be unknown at
compile time with graphics pipeline libraries. So we need to have this
dynamic has well and pick the right value when generating the
3DSTATE_PS/3DSTATE_WM packet.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: d8dfd153c5 ("intel/fs: Make per-sample and coarse dispatch tri-state")
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22728>

mesa/main: drop unused variable

Fixes: c5b3d488f9be ("mesa/main: make ffvertex output nir")

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22812>

broadcom/compiler: don't allocate undef to rf0

rf0 is affected by restrictions in some scenarios so we rather use
a register that does not cause conflicts for scheduling.

total instructions in shared programs: 12850958 -> 12848024 (-0.02%)
instructions in affected programs: 331974 -> 329040 (-0.88%)
helped: 2559
HURT: 201
Instructions are helped.

total max-temps in shared programs: 2210893 -> 2210803 (<.01%)
max-temps in affected programs: 1486 -> 1396 (-6.06%)
helped: 96
HURT: 7
Max-temps are helped.

total sfu-stalls in shared programs: 21975 -> 21965 (-0.05%)
sfu-stalls in affected programs: 32 -> 22 (-31.25%)
helped: 16
HURT: 6
Sfu-stalls are helped.

total inst-and-stalls in shared programs: 12872933 -> 12869989 (-0.02%)
inst-and-stalls in affected programs: 332036 -> 329092 (-0.89%)
helped: 2560
HURT: 189
Inst-and-stalls are helped.

total nops in shared programs: 305911 -> 303501 (-0.79%)
nops in affected programs: 11215 -> 8805 (-21.49%)
helped: 2131
HURT: 3
Nops are helped.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22797>

r600: fix refcnt imbalance related to atomic_buffer_state

Indeed, these references are not freed.

For instance, this issue is triggered on an evergreen card with
"piglit/bin/shader_runner tests/spec/arb_shader_atomic_counter_ops/execution/all_touch_test.shader_test -auto -fbo"
while setting GALLIUM_REFCNT_LOG=refcnt.log.

Fixes: 06993e4ee350 ("r600: add support for hw atomic counters. (v3)")
Signed-off-by: Patrick Lerda <patrick9876@free.fr>
Reviewed-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22798>

vulkan: rename vk_physical_device_features.py to vk_physical_device_features_gen.py

In-line with other files that generate source code

Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22333>

vulkan: put TEMPLATE_H before TEMPLATE_C

Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22333>

vulkan: put interesting code before boring code

Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22333>

vulkan: inline vk_get_physical_device_features into vk_common_GetPhysicalDeviceFeatures2

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22333>

v3dv: use common vkGetPhysicalDeviceFeatures2

Reviewed-by: Eric Engestrom <eric@igalia.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22333>

panvk: use common vkGetPhysicalDeviceFeatures2

Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22333>

anv: move get_features after get_device_extensions (ugly diff)

Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22333>

anv: switch to using the common vkGetPhysicalDeviceFeatures2

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22333>

vulkan: introduce supported_features parameter to vk_physical_device_init

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22333>

vulkan: add common implementation of vkGetPhysicalDeviceFeatures2

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22333>

vulkan: Actually increment the count of objects in GetPipelineCacheData.

Otherwise a load of the cache won't load any of the objects in it!

Fixes: 591da9877900 ("vulkan: Add a common VkPipelineCache implementation")
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22700>

vulkan: Handle alignment failure in the pipeline cache.

On malloc failure, or more likely a passed in pDataSize < pipeline cache
data size, we should return incomplete. Otherwise,
vk_pipeline_cache_object_serialize will assertion fail about the start not
being aligned.

Closes: #8868
Fixes: 591da9877900 ("vulkan: Add a common VkPipelineCache implementation")
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22700>

dzn: Ensure bindless is factored into pipeline/nir hash

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22810>

dzn: Ensure subgroup size control is factored into pipeline hash

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22810>

dzn: Use the nir hash as an input to the dxil hash

Instead of re-computing all the same bits we already computed,
just use the nir hash. Helps prevent missing a bit between the two.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22810>

dzn: Ensure sample-rate shading is factored into nir hash

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22810>

dzn: Hook up subgroup size to compute shader compilation

Previously this was only in the graphics path... where it does nothing,
since D3D only supports wave size control for compute. Whoops.

Fixes: db083070 ("dzn: Implement subgroup size control extension")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22810>

crocus: fix refcnt imbalance related to crocus_create_surface()

Indeed, the objects are not freed when the function returns NULL.

"psurf->texture = tex;" is redundant with
"pipe_resource_reference(&psurf->texture, tex);".

For instance, this issue is triggered with
"piglit/bin/ext_texture_array-compressed teximage pbo -fbo -auto"
while setting GALLIUM_REFCNT_LOG=refcnt.log.

Fixes: f3630548f1da ("crocus: initial gallium driver for Intel gfx 4-7")
Signed-off-by: Patrick Lerda <patrick9876@free.fr>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22799>

virgl: check a debug option again at context creation

Android apps commonly use HWUI (a GLES-based UI framework provided by
the system), that generally performs eglInitialize() before the app can
do the same for its custom rendering needs.

If an app is going to set VIRGL_DEBUG to enable case-by-case driver
behaviors (e.g. experimental shader_sync option), it should be checked
again during context creation.

Signed-off-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22744>

dzn: Fix UBO descriptors pointing to the end of the buffer

UBO descriptors can re-use the default descriptor when their upper
bound is 64KiB, not the end of the buffer.

Fixes dEQP-VK.memory.pipeline_barrier.host_write_uniform_buffer.1048576

Fixes: d34ac0a70b7 ("dzn: Re-design custom buffer descriptors")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22805>

Revert "zink: don't create separate shader dsls if there are no bindings"

This reverts commit 83a1b74240df3afcdb0e723c5bf6514b3321b411.

this is creating the pipeline layout, which is always required, not the dsl

Fixes: 83a1b74240d ("zink: don't create separate shader dsls if there are no bindings")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22801>

zink: only unset a generated tcs if the bound tcs is the generated one

this could otherwise unbind a non-generated tcs if the tes had at some
point generated a tcs

cc: mesa-stable

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22801>

radv/rt: Fix and improve VkPipelineCreationFeedback

Due to a copy-paste error, we asserted pipelineStageCreationFeedbackCount == 1
and wrote the stage feedback of the combined shader into the feedback of the first
stage. This is fixed.
Instead, we now write the precompilation feedback for each stage. This not ideal,
but definitely an improvement.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22100>

radv/rt: remove merged VkPipelineShaderStageCreateInfo

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22100>

radv/rt: refactor compute_rt_stack_size() to use radv_ray_tracing_stage information

instead of pStages.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22100>

radv/rt: use precompiled stages to create RT shader

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22100>

radv/rt: retain parsed NIR shaders in radv_ray_tracing_lib_pipeline

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22100>

radv/rt: introduce struct radv_ray_tracing_stage

The purpose of this struct is to bundle per-stage information
for ray-tracing pipelines.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22100>

radv/rt: expose radv_parse_rt_stage()

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22100>

radv/pipeline_cache: add NIR caching capabilities

This patch adds functions to radv_pipeline_cache in order to cache
serialized NIR shaders as opaque cache objects.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22100>

vulkan/pipeline_cache: expose vk_raw_data_cache_object

together with vk_raw_data_cache_object_ops and vk_raw_data_cache_object_create().

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22100>

nvc0: do not randomly emit fences.

We track fences in a global list and have a per context "current" fence
which we randomly attach things to. If we take such a fence and emit it
without also creating a new fence for future tasks we can get out of sync
leading to random failures.

Some of our queries could trigger such cases and even though this issues
appears to be triggered by the MT rework, I'm convinced that this was only
made more visible by those fixes and we had this bug lurking for quite a
while.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7429
Fixes: df0a4d02f26 ("nvc0: make state handling race free")
Signed-off-by: Karol Herbst <git@karolherbst.de>
Acked-by: M Henning <drawoc@darkrefraction.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22722>

ci: Uprev crosvm and virglrenderer

Update crosvm, virglrenderer and bindgen-cli to their latest version on time.

Signed-off-by: Corentin Noël <corentin.noel@collabora.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22712>

nir/lower_non_uniform_access: add get_ssbo_size handling

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22793>

rusticl/nir: finish blob after serializing

Signed-off-by: Karol Herbst <git@karolherbst.de>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22776>

clc: free kernel args in clc_free_kernels_info

Signed-off-by: Karol Herbst <git@karolherbst.de>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22776>

radv: add padding to radv_shader_binary_legacy

This fixes an alignment mismatch in radv_shader_serialize()

Fixes: 577d76f60f788c3a8d43b4b388bb5a04e4e64994 ('radv: add radv_shader_serialize() and radv_shader_deserialize() functions')
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22735>

vulkan/pipeline_cache: use vk_pipeline_cache_insert_object() to replace raw data objects

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22735>

vulkan/pipeline_cache: replace raw data objects on cache insertion of real objects

It might happen that a raw data object (from pipeline cache creation)
was never looked up, and thus never deserialized, before it gets
inserted again into the cache. In this case, the deserialized object
got replaced by the raw data object.
Instead, replace the raw data object with the real object in the cache.

Fixes: 8b13ee75ba9f27ceac6b6180ca05d321caa13612 ('vulkan: Fall back to raw data objects when deserializing if ops == NULL')
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22735>

etnaviv: Add util_blitter_save_so_targets(..) call

We do not support stream outputs yet so be better at faking it.
Fixes the following assertion when running deqp-gles3:
u_blitter.c:635: blitter_check_saved_vertex_states: Assertion `!ctx->has_stream_out || ctx->base.saved_num_so_targets != ~0u' failed

Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Lucas Stach <l.stach@pengutronix.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22754>

pvr: Correctly validate PBE accum format

Fixes: dEQP-VK.pipeline.monolithic.blend.format.e5b9g9r9_ufloat_pack32
  .states
  .color_sas_1mda_add_alpha_1msa_sas_rsub
    -color_1mda_cc_min_alpha_cc_cc_min
    -color_1mcc_1mca_rsub_alpha_1mca_1mda_min
    -color_sc_1mda_sub_alpha_sa_cc_min

Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reported-by: James Glanville <james.glanville@imgtec.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22795>

radv: fix bc optimization with POS_W_FLOAT_ENA(1)

0286CC_PERSP_CENTER_ENA is also set if:
- No barycentrics are used
- POS_W_FLOAT is enabled

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Fixes: a4b60295a79 ("aco,ac/llvm,radv,radeonsi: handle ps bc optimization in nir for radv")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8928
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22782>

pvr: Unmap mapped memory on free

From the Vulkan spec (§11.2.13. Freeing Device Memory):
If a memory object is mapped at the time it is freed, it is implicitly
unmapped.

Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22762>

pvr: Fix unaligned VDMCTRL_PDS_STATE1 data address

For indirect draw commands the data segment's address was being
calculated with an unaligned code size which could potentially
send an unaligned address to the hw causing problems.

Reported-by: James Glanville <james.glanville@imgtec.com>
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/22746>

pvr: Fix image to buffer copies

The code to increment the destination buffer address was wrong. Instead of
incrementing an address on every iteration, the buffer slice size was added
as a static offset to all blits.

Seen in dEQP-VK.draw.renderpass.depth_clamp.d16_unorm, though issues remain.

Signed-off-by: James Glanville <james.glanville@imgtec.com>
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/22747>

pvr: Fix vtxin special var allocation count

When pvr_graphics_pipeline_alloc_vertex_special_vars() was
allocating vtxin regs for special vars, it wasn't returning back
the updated amount of allocated regs so when the PDS was filling
them in, it would try to write out of bounds.

Signed-off-by: James Glanville <james.glanville@imgtec.com>
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/22750>

pvr: Improve support for image clears

Depth/stencil surfaces are not PBE renderable, so a compatible format must
be used instead. The code to calculate this compatible format was not called
when configuring the PBE, and it was missing formats.

Also the code to calculate PBE swizzles was throwing an error in unhandled
cases, rather than using the pre-calculated defaults which was the correct
behaviour.

Fixes blits in dEQP-VK.draw.renderpass.depth_clamp.d32_sfloat

Signed-off-by: James Glanville <james.glanville@imgtec.com>
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/22748>

nir: fix constant-folding of 64-bit fpow

We need to do full pow if 64-bit, and we can do fpow() otherwise. Not
the other way around.

Fixes: 9076c4e289d ("nir: update opcode definitions for different bit sizes")
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22774>

radv: fix dynamic depth clamp enable support

The Vulkan spec says:

"If the depth clamping state is changed dynamically, and the pipeline
was not created with VK_DYNAMIC_STATE_DEPTH_CLIP_ENABLE_EXT enabled,
then depth clipping is enabled when depth clamping is disabled and
vice versa"

Fixes: e48c0fbd8f7 ("radv: add support for dynamic depth clamp enable")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22777>