anv: Add GPU breakpoint before/after specific draw call

This change allow us to insert the MI_SEMAPHORE_WAIT before/after
specific draw call. With GTX tool, we can always update the memory
address to unblock spinning wait.

v2:
- Make sure draw_call_count is thread-safe (Lionel)
- Add static inline helper (Lionel)

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24308>

intel: Add env variable to add break point on/before draw

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24308>

ci/panfrost: t760-gles is nightly job, test also GLES 3 and 3.1

We don't care about how long it takes since it's nightly.

Acked-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24546>

CI: Re-enable G52 Vulkan testing

This gives us coverage back on panvk on Bifrost. There are a lot of
fails since it was last tested though.

[daniels: Updated with new expectations.]

Acked-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24546>

ci/panfrost: re-enable t760 and t860 traces as a nightly job

Two crashing jobs documented in: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9473

Acked-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24546>

ci: add perfetto into mesa git-cache

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8821

Fixes: 8aff2281276c ("ci: Enable building the testing drivers with perfetto.")
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23611>

zink: fix setting VkShaderCreateInfoEXT::nextStage

nextStage has some restrictions depending on the current stage.

Fixes: cd6625c6eb9 ("zink: use EXT_shader_object to (re)implement separate shaders")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24558>

nir/print: always group variables by type when printing

Acked-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23752>

nir/print: print location names for (some) tess slots

these should be fine to print

Acked-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23752>

Gallium: Fix shared memory segment leak

Commit abe6d750e58d371624de75f4bad365c61e0196c1 caused shared memory
segments to be leaked. We need to mark shared memory segments for
deletion upon construction.

Fixes: abe6d750e5 xlib: fix glXDestroyContext in Gallium frontends
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9425
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24556>

radv: Expose VK_EXT_external_memory_acquire_unmodified.

No-op, since we generally don't do anything to revalidate images.

In general on external/foreign queues we prepare on "export" that
it might be used on more queues, but we do pretty much nothing on
"import".

Gitlab: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9348
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24524>

ci: build nvk

Signed-off-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24542>

meson: Export winsys function symbols for target va

Export winsys function symbols of target va, even if the user
links with '-Bsymbolic-functions'.

It refers target vdpau, which commit is:
8c136b53b79e90b9e8f30f891b8bef112fee375d
("fix vdpau interop when using -Bsymbolic-functions in ldflags")

Signed-off-by: Feng Jiang <jiangfeng@kylinos.cn>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23177>

vc4/ci: update expected results

Signed-off-by: Juan A. Suarez Romero <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24550>

radv: use vk_query

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24416>

radv: use common vkCmdBegin/EndQuery wrappers

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24416>

radv: use vk_sampler

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24416>

radv: use vk_buffer_view

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24416>

vulkan: add init/finish helpers for vk_buffer_view

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24416>

anv: fake non intel vendorID for Death Stranding

The assumption is the same issue that is plaguing Cyberpunk 2077 is
also at play here. That is the XeSS library is looking for the Windows
driver binary and not finding them in the Wine/Proton distribution.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24237>

radv: update cmdbuf scratch size info when shaders are bound

This will automatically update the scratch size info for shader object.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24502>

radv: update the number of scratch waves for RT prolog at bind time

The compute scratch size is computed later because the RT stack size
can be dynamic, but the number of waves shouldn't change.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24502>

radv: add a helper to get the maximum number of scratch waves per shader

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24502>

radv: use the RT prolog scratch size directly for tracing rays

It should be the same as the pipeline scratch size value.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24502>

intel/genxml: Add filter_engines() to GenXml class

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24547>

intel/genxml: Add GenXml class into intel_genxml module

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24547>

intel/genxml: Convert gen_pack_header to use ElementTree

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24547>

intel/genxml: Convert gen_bits_header to use ElementTree

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24547>

intel/genxml: Split some genxml sorting code into a intel_genxml module

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24547>

intel/genxml: Align "Texture Coordinate Mode" naming

Some older gens used '_' (underscore) while newer gens used a ' '
(space).

$ sed -i 's/Texture_Coordinate_Mode/Texture\ Coordinate\ Mode/' \
      src/intel/genxml/*.xml

The naming needs to be aligned for importing enums later on.

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24547>

llvmpipe: fix UAF in lp_scene_is_resource_referenced.

reworked slightly by airlied

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24414>

nvk: Fix assert

Fix defect reported by Coverity Scan.

Assign instead of compare (PW.ASSIGN_WHERE_COMPARE_MEANT)
assign_where_compare_meant: use of "=" where "==" may have been intended

Fixes: e41031d8ffd ("nvk: Enable multiplane images and image views")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24530>

nouveau/mme: fix OOB inside tu104 simulator

Signed-off-by: Karol Herbst <git@karolherbst.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24544>

nouveau/mme: fix OOB access inside while_ine builder test

Signed-off-by: Karol Herbst <git@karolherbst.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24544>

anv: Override vendorID for Hogwarts Legacy

This is another game that makes use of XeSS but works when we fake
the vendorID.

With this temporary hack it works on i915 and Xe KMDs.

Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24506>

glsl: fix spirv sso validation

The api validation calls will segfault without this as it will
try to fallback to string matching names which are NULL. This
would be incorrect behaviour even if the names weren't NULL so
here we correctly set the explicit location flag.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9483
Fixes: ffdb44d3a0a2 ("nir/linker: Add inputs/outputs to the program resource list")

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24529>

zink: turn off threaded cpu access if not visible.

This turns off the threaded cpu access it the resource isn't visible.

Fixes a bunch of crashes with current nvk.

Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24548>

nvk: Remove plane sources from tex instructions

The plane source is entirely handled by lower_tex() so there's no need
to keep it around.  Codegen currently just ignores these but NAK will
assert if it sees an unknown source type.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24549>

nir/linking_helpers: force type matching in does_varying_match

this otherwise breaks when i/o is scalarized in the producer but not
the consumer

cc: mesa-stable

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24458>

nir/lower_io_to_scalar: fix 64bit io splitting

this was creating broken 64bit loads/stores using 32bit component
size

cc: mesa-stable

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24458>

anv/video: copy from correct H264 scaling lists

Vulkan defines the scaling lists according to the H264 ITU spec, which
only defines ScalingList8x8[0] and ScalingList8x8[1] for
non-444 formats.

Reviewed-by: Lynne <dev@lynne.ee>
Reviewed-by: Hyunjun Ko <zzoon@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24474>

radv/video: copy from correct H264 scaling lists

Vulkan defines the scaling lists according to the H264 ITU spec, which
only defines ScalingList8x8[0] and ScalingList8x8[1] for
non-444 formats. Since RADV only supports 420, just directly use those.

Reviewed-by: Lynne <dev@lynne.ee>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24413>

nvk: align sampler allocation counts with nvidia.

Fill out the missing property as well.

found running some zink piglit runs.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24532>

nir/gl: Move glsl_type::sampler_target() into a helper in its one caller

The new version also doesn't need to worry about arrays of textures
because the caller already takes care of that.

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24503>

tu: Add virtgpu support

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23533>

tu/drm: Split out helper for iova alloc

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23533>

tu/drm: Add missing error path cleanup

Fixes: bea6290ca090 ("turnip: device global bo list")
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23533>

tu/drm: Factor out shared helpers

Factor out a few things that we can re-use between virtio and msm
backends.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23533>

tu: close submitqueues before device_finish()

Otherwise we would have already closed the device.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23533>

freedreno/drm/virtio: Use global_faults

Rather than doing a synchronous round trip to the host to query fault
count, use the shmem->global_faults field if available.  The kernel is
already only reporting faults associated with the same drm_file (which
in this case maps 1:1 to guest process), so the only thing we really
miss is, if the app is using multiple contexts, _which_ context the
fault was in.  But vulkan doesn't even use that, which sounds like a
good enough argument that per-submitqueue granularity isn't really
needed.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23533>

freedreno: Update virtgpu proto

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23533>

drm-uapi: Update virtgpu header

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23533>

tu/knl: Remove some random const'ness

This gets in the way with virtio_gpu backend.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23533>

tu/msm: staticify a couple things

Stop leaking some private implementation details as global symbols.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23533>

freedreno/drm/virtio: Remove unused header

A remnant from a very early version, prior to refactoring out
fd_ringbuffer_sp to share more code btwn msm and virtgpu backends.
Somehow this never got removed.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23533>

freedreno: move virtgpu msm_proto.h to common

src/freedreno/virtgpu seems a bit overkill for a single header.. but
maybe there are some other bits we could share?

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23533>

r600/sfn: take number of dest values into account

With the write mask removed we have to run the loop correctly.

Fixes: b870988b77f70acd69feb1d4e52343b7c3d4e328
    r600/sfn: Stop referencing legacy functionality

Related: https://gitlab.freedesktop.org/mesa/mesa/-/issues/9490
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24533>

Force vk vendor for spider-man remastered

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24194>

zink: unset line stipple ds3 state flags when stipple not available

Fixes: 7b4c1b3a426 ("zink: track and apply ds3 states only on change")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24478>

zink: use VK_WHOLE_SIZE when binding null db buffer descriptors

using range=0 is illegal

cc: mesa-stable

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24478>

ci: bump VVL to 1.3.261

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24478>

intel: Sync xe_drm.h and rename engine to exec_queue

Sync with commit f16c04291100 ("drm/xe: Rename engine to exec_queue").
With that Iris and ANV had some major renames that were done manually
as there is to many "engine" in non-related code.

Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24476>

egl: fixup _eglFilterConfigArray() params and drop _eglFallbackMatch() wrapper

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24292>

egl: make _eglFilterConfigArray static

Signed-off-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24292>

panfrost: add PAN_MESA_DEBUG=yuv for debugging yuv sampler

Signed-off-by: Italo Nicola <italonicola@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21109>

panfrost: mandate proper alignment requirement depending format and arch

v7+ mandates row_stride alignment to be equal to surface alignment.
AFBC surface alignment is always 16-bytes.
NV12/NV21 formats on v7+ have 16-byte surface alignment.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21109>

panfrost: advertise support for YUYV and variants

Signed-off-by: Italo Nicola <italonicola@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21109>

panfrost: prepare the driver to support YUYV and variants

Signed-off-by: Italo Nicola <italonicola@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21109>

panfrost: prepare pan_image_view for multiplanar formats

Signed-off-by: Italo Nicola <italonicola@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21109>

pan/decode: decode Multiplanar Surface descriptors

Signed-off-by: Italo Nicola <italonicola@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21109>

panfrost: refactor (un)packing of surface descriptors

Use the genxml helpers to pack/unpack midgard and bifrost surface
descriptors. Also changed how we describe midgard's descriptors, to make
it more straightforward. We currently only use SURFACE_WITH_STRIDE for
midgard, but pandecode should still be able to decode other surface
descriptor types.

This commit shouldn't change panfrost/pandecode behavior.

Consequenctly, this refactor also prepares pandecode to handle the
SURFACE_YUV descriptor for bifrost in the following patch.

Signed-off-by: Italo Nicola <italonicola@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21109>

panfrost/genxml: add Multiplanar Surface descriptor

Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21109>

panfrost/genxml: fix Surface With Stride descriptor alignment

Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21109>

pvr: do not claim support for ASTC texture compression

We don't support any ASTC formats yet, but the textureCompressionASTC_LDR
feature was incorrectly set to true. Fix this by setting it to false and
don't advertise ASTC support for texture compression.

Fixes dEQP-VK.api.info.format_properties.compressed_formats

Fixes: 8991e646 ("pvr: Add a Vulkan driver for Imagination Technologies PowerVR Rogue GPUs")
Signed-off-by: Luigi Santivetti <luigi.santivetti@imgtec.com>
Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24448>

freedreno: add ci flake

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24539>

virgl: add ci flake

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24539>

panfrost/vk: Use correct sampler dimensions for MSAA

2D-MS samplers are SAMPLER_DIM_MS, not SAMPLER_DIM_2D. There is no MSAA
for non-2D textures.

Signed-off-by: Daniel Stone <daniels@collabora.com>
Closes: mesa/mesa#9474
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24494>

anv: enable INTEL_DEBUG=nofc

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24510>

v3dv: drop cpu path for buffer to image copies

We really want to avoid cpu jobs as much as possible. Also,
the texel buffer path we have for this should be able to
handle most cases.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24466>

radv: initialize result when pipeline cache creation fails

Signed-off-by: Chris Spencer <spencercw@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24521>

zink: use fprintf instead of printf to align the requirements warnings

This just seems to have been overlooked and nvk triggers it right now.

Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24527>

zink: force image barriers after dmabuf import

if the image will be used, then it has to be on the right queue

cc: mesa-stable

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24492>

zink: better handle separate shader dsl creation when no bindings exist

this otherwise underflows the array and provides a (probably huge) garbage
value for the binding id, which then causes the driver to massively overallocate
both the layout and set/pool/buffer

the main result of this is that on radv any simple test that should be near-instant
takes 2-3 seconds to execute, which somehow nobody noticed

Fixes: e3b746e3a31 ("zink: use GPL to handle (simple) separate shader objects")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24501>

docs: add two nvk exts to features.txt

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24523>

intel/decoder: Make intel_spec_load_filename() have separate dir and name strings

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20889>

intel/decoder: Add intel_spec_load_common()

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20889>

intel/decoder: Fix xml filename when verx10 % 10 is not 0

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20889>

r600/sfn: Take source uses into account when switching channels

When we switch the channels by re-creating vec4 values we have to
take into account that the source values may be used in an ALU op,
and with that we have to take read-port limitations into account.

Fixes: 18a8d148d862d4a0f5b83f4385f00cf8a4a71b6d
    r600/sfn: Cleanup copy-prop into vec4 source values

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24519>

nvk: Call nir_lower_int64

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24518>

nvk: Use more consistent device variable names

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24518>

ci/freedreno: document another flakes on Adreno 530

Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24511>

nv50/ir: convert system values to gl_system_value

Signed-off-by: Karol Herbst <git@karolherbst.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24447>

nv50/ir: use own info struct for sys vals

This makes it more obvious what's actually needed by the driver and allows
us to drop some code in the nir to nvir step.

Signed-off-by: Karol Herbst <git@karolherbst.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24447>

nvk: Enable the new UAPI

We also add a meson option to disable it and re-enable the legacy API.
This code probably won't last long but it gives developers the option.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>

nvk: Print an error message if VM_BIND support is missing

This should prevent a lot of bug reports.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>

nouveau/winsys: Allow nouveau_ws_device_new() without VM_BIND

Silently fall back to non-VM_BIND.  The client can check for the
has_vm_bind flag in the device to detect whether VM_BIND is actually
available or not.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>

nvk: add new internal bo allocation flag.

This is to allow the kernel to use a single resv object, this might
need more work.

Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>

nvk: Use an empty EXEC for the empty submit case

The new UAPI allows for an EXEC with zero pushes just fine so we no
longer need the no-op push just for synchronization.  This lets us drop
the whole empty push we allocate per-queue as well.

Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>

nvk: enable sparse residency buffer on maxwell+

Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>

nvk: enable a bunch of external fence/semaphore bits

Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24326>