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

aco: Use NIR IO semantics for tess factor IO locations.

Previously we relied on looping over the NIR output variables
to remember the driver location of the tess factors, now use
the new NIR IO semantics instead.

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/6689>

anv: implement VK_KHR_copy_commands2

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6801>

vulkan: bump headers/registry to 1.2.154

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6801>

nir/lower_samplers: Clamp out-of-bounds access to array of samplers

Section 5.11 (Out-of-Bounds Accesses) of the GLSL 4.60 spec says:

"In the subsections described above for array, vector, matrix and
structure accesses, any out-of-bounds access produced undefined
behavior.... Out-of-bounds reads return undefined values, which
include values from other variables of the active program or zero."

Robustness extensions suggest to return zero on out-of-bounds
accesses, however it's not applicable to the arrays of samplers,
so just clamp the index.

Otherwise instr->sampler_index or instr->texture_index would be out
of bounds, and they are used as an index to arrays of driver state.

E.g. this fixes such dereference:
if (options->lower_tex_packing[tex->sampler_index] !=
in nir_lower_tex.c

CC: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6428>

nir/large_constants: Eliminate out-of-bounds writes to large constants

Out-of-bounds writes could be eliminated per spec:

Section 5.11 (Out-of-Bounds Accesses) of the GLSL 4.60 spec says:

"In the subsections described above for array, vector, matrix and
structure accesses, any out-of-bounds access produced undefined
behavior.... Out-of-bounds writes may be discarded or overwrite
other variables of the active program."

Fixes: 1235850522cd5e7b07701f7065996430ca1514b6
Signed-off-by: Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6428>

nir/lower_io: Eliminate oob writes and return zero for oob reads

Out-of-bounds writes could be eliminated per spec:

Section 5.11 (Out-of-Bounds Accesses) of the GLSL 4.60 spec says:

"In the subsections described above for array, vector, matrix and
  structure accesses, any out-of-bounds access produced undefined
  behavior....
  Out-of-bounds writes may be discarded or overwrite
  other variables of the active program.
  Out-of-bounds reads return undefined values, which
  include values from other variables of the active program or zero."

GL_KHR_robustness and GL_ARB_robustness encourage us to return zero
for reads.

Otherwise get_io_offset would return out-of-bound offset which may
result in out-of-bound loading/storing of inputs/outputs,
that could cause issues in drivers down the line.

E.g. this fixes such dereference:
int vue_slot = vue_map->varying_to_slot[intrin->const_index[0]];
in brw_nir.c

CC: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6428>

radeon: rename RADEON_TRANSFER_* -> RADEON_MAP_*

Acked-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5749>

gallium: rename transfer flags -> map flags in comments

Acked-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5749>

gallium: rename pipe_transfer_usage -> pipe_map_flags

Acked-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5749>

gallium: rename PIPE_TRANSFER_* -> PIPE_MAP_*

Acked-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5749>

radeonsi: add 16-bit ALU vectorization

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6622>

radeonsi: don't scalarize 16-bit vec2 ALU opcodes

prerequisite for enabling vectorization, so that we don't scalarize
and vectorize at the same time, causing an infinite loop.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6622>

radeonsi: vectorize IO for better ALU vectorization

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6622>

radeonsi: implement 16-bit FS color outputs

This removes type conversions from 16 bits to 32 bits in the main function
and then back to 16 bits in the epilog.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6622>

radeonsi: set flags for FP16 in shaders

v2: remove the enablement code

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> (v1)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6622>

radeonsi: don't lower pack for better 16-bit vectorization

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6622>

ac/nir: implement nir_op_[un]pack_[64/32]_*

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6622>

glsl: Initialize ir_constant member const_elements in all constructors.

Fix defects reported by Coverity Scan.

Uninitialized pointer field (UNINIT_CTOR)
uninit_member: Non-static class member const_elements is not
initialized in this constructor nor in any functions that it calls.

Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6770>

pan/bi: Fix typo.

Fix defect reported by Coverity Scan.

Same on both sides (CONSTANT_EXPRESSION_RESULT)
pointless_expression: The expression src0_u8 || src0_u8 does not
accomplish anything because it evaluates to either of its
identical operands, src0_u8.

Fixes: 2ff53879f214 ("pan/bi: Use new packing")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6780>

st/mesa: Deal with empty textures/buffers in semaphore wait/signal.

The actual texture might not have been created yet.

Gitlab: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3257
CC: mesa-stable
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6788>

turnip/kgsl: Associate fences with submits.

This fixes all the I was seeing in the multiview tests.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4479>

turnip: Add kgsl backend

Lacking a bit around fences and wsi integration, but there's enough
here to actually drive the GPU.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4479>

intel/compiler: fixup Gen12 workaround for array sizes

We didn't handle the case of NULL images/textures for which we should
return 0.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 397ff2976ba281 ("intel: Implement Gen12 workaround for array textures of size 1")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3522
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6729>

wsi/x11: Create a present queue for VK_PRESENT_MODE_FIFO_RELAXED_KHR

This activates the target_msc computation that we'd otherwise be using
if we were just FIFO; otherwise we'd always set target_msc = 0 and it'd
act like IMMEDIATE.

Fixes: 77d25627933 "wsi/x11: Hook up VK_PRESENT_MODE_FIFO_RELAXED_KHR"
Fixes: mesa/mesa#3512
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6804>

gallium/auxiliary/vl: Include src region in scale_y calculation

Signed-off-by: Thong Thai <thong.thai@amd.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3521
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6736>

frontends/va: Enabled packed headers for H.264 encoder

Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4184>

radeon: Pass HEVC encode crop parameters to the encoder

Signed-off-by: Thong Thai <thong.thai@amd.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2351
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4184>

gallium: Parse packed HEVC SPS encode header for crop parameters

The crop / conformance window parameters are set by ffmpeg but they only
seem to be made available in packed headers. This commit copies the H265
header parsing code from st/omx (planning in the future to move this
code to a common place to be shared by the different state trackers) in
order to grab the crop parameters

Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4184>

r600/sfn: Fix typo in comment

Thanks Stanislav Jakúbek for ponting it out.

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6787>

r600/sfn: Fix component count for fdph

The second parameter has four components, and this is the one that would
be copied when splitting constants.

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6787>

r600/sfn: make number of source components a local variable

There is no need to keep that value as a member and it irritates
Coverity.

Fixes: 688680decce9
r600/nir: fetch sources and split uniforms before emittting alu instructions

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6787>

gallium/auxiliary/vl: Odd Dimensions are failing

Issue: While running the tast for odd resolutions there are green lines
observed in the dumped image. The resolution is 321x241, the extra 1 pixel
data is missing in the image. The reason for this is in the post
processing when we adjust the size of height and width we are
dividing it by 2.

Fix: To resolve this issue we first need to align it to 2 and then divide
by 2 to get the required values. Once we do this we will have proper data
in the dumped image and missing pixel data will be available.

Signed-off-by: Krunal Patel <krunalkumarmukeshkumar.patel@amd.corp-partner.google.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6795>

radv: do not lower UBO/SSBO access to offsets

Use nir_lower_explicit_io instead of lowering to offsets. Extra
(useless) additions are removed by lowering load_vulkan_descriptor
to vec2(src.x, 0).

fossils-db (Navi):
Totals from 18236 (13.21% of 138013) affected shaders:
SGPRs: 1172766 -> 1168278 (-0.38%); split: -0.89%, +0.50%
VGPRs: 940156 -> 952232 (+1.28%); split: -0.08%, +1.37%
SpillSGPRs: 30286 -> 31109 (+2.72%); split: -0.78%, +3.50%
SpillVGPRs: 1893 -> 1909 (+0.85%)
CodeSize: 87910396 -> 88113592 (+0.23%); split: -0.35%, +0.58%
Scratch: 819200 -> 823296 (+0.50%)
MaxWaves: 205535 -> 202102 (-1.67%); split: +0.05%, -1.72%
Instrs: 17052527 -> 17113484 (+0.36%); split: -0.32%, +0.67%
Cycles: 670794876 -> 669084540 (-0.25%); split: -0.38%, +0.13%
VMEM: 5274728 -> 5388556 (+2.16%); split: +3.10%, -0.94%
SMEM: 1196146 -> 1165850 (-2.53%); split: +2.06%, -4.60%
VClause: 381463 -> 399217 (+4.65%); split: -1.08%, +5.73%
SClause: 666216 -> 631135 (-5.27%); split: -5.44%, +0.18%
Copies: 1292720 -> 1289318 (-0.26%); split: -1.28%, +1.01%
Branches: 467336 -> 473028 (+1.22%); split: -0.67%, +1.89%
PreSGPRs: 766459 -> 772175 (+0.75%); split: -0.53%, +1.28%
PreVGPRs: 819746 -> 825327 (+0.68%); split: -0.05%, +0.73%

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6202>

radv: add a helper for loading meta descriptors

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6202>

ci: Don't create test-docs job if the pages one exists in the pipeline

It's redundant in that case.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6566>

ci: Prevent pages job from running in pre-merge pipelines

Since pre-merge pipelines run in the "mesa" namespace now, the rule
would have created the job if the source branch was named "master" (and
the job might have updated the public website).

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6566>

ci: Always use CI_PROJECT_NAMESPACE instead of CI_PROJECT_PATH

And always check it first.

Cleanup for consistency.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6566>

ci: Add "is for Marge Bot" YAML anchor

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6566>

ci: Add "is pre-merge pipeline" YAML anchor

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6566>

ci: Add "is post-merge pipeline" YAML anchor

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6566>

ci: Add "is forked branch" YAML anchor

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6566>

ci: Add "is forked branch or pre-merge pipeline" YAML anchor

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6566>

ci: Add "is post-merge pipeline, not for Marge Bot" YAML anchor

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6566>

ci: Add "is pre-merge pipeline for Marge Bot" YAML anchor

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6566>

ci: Add "is master branch of main project" YAML anchor

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6566>

ci: Add "is scheduled pipeline" YAML anchor

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6566>

radv: fix transform feedback crashes if pCounterBufferOffsets is NULL

From the Vulkan 1.2.154 spec:
"If pCounterBufferOffsets is NULL, then it is assumed the
offsets are zero."

Fix new CTS
dEQP-VK.transform_feedback.simple.backward_dependency_no_offset_array.

CC: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6798>

aco: use bit vectors for liveness sets

This seems to be much faster than hash sets. When compiling pipelines from
5 games, live_var_analysis takes about a third the time it used to and
fossilize-replay is ~1.77% faster.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Tony Wasserka <tony.wasserka@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6733>

aco: keep track of temporaries' regclasses in the Program

A future change will switch the liveness sets to bit vectors, which don't
contain regclass information.

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/6733>

radeonsi: change vendor name to AMD

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6754>

radeonsi: reduce PIPE_SHADER_CAP_MAX_CONST_BUFFER_SIZE value

The new value (64 MB) is a compromise:
- the old value was very large (max_alloc & 0xffffffff) and caused https://gitlab.freedesktop.org/mesa/mesa/-/issues/3301
- amdgpu-pro allows 512MB, nvidia 64MB

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6754>

radeonsi: move GL vendor workaround to drirc

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6754>

gallium: add PIPE_CAP_MAX_TEXTURE_MB

Allows driver to override the default value (1024) from mesa.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6754>

radv,aco: fix reading primitive ID in FS after TES

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3530
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6760>

panfrost: decode: Misc formatting improvements

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/6797>

panfrost: Avoid copying job descriptors around when we can

Job descriptors are written section by section and are never modified
after them been emitted. Let's avoid copying things around by allocating
descriptors upfront and letting the scoreboard logic only write the
header section.

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/6797>

panfrost: XML-ify the compute job descriptor

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/6797>

panfrost: Rename the Blend dither disable flag

It's actually encoding the "Round to FB precision" information.

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/6797>

panfrost: XML-ify the fragment job descriptor

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/6797>

panfrost: XML-ify the write value job descriptor

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/6797>

panfrost: XML-ify the job header descriptor

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/6797>

panfrost: XML-ify the multi-target framebuffer descriptors

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/6797>

panfrost: XML-ify the bifrost tiler descriptors

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/6797>

panfrost: XML-ify the single target framebuffer descriptor

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/6797>

panfrost: XML-ify the midgard tiler descriptor

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/6797>

panfrost: Add Tiled linear mode to the Block Format enum

Add it for completeness.

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/6797>

panfrost: Clarify what TILED mode is

We have 2 tile modes: tiled linear and tiled U-interleaved. Let's rename
the existing value to clarify that.

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/6797>

panfrost: XML-ify the local storage descriptor

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/6797>

panfrost: decode: Use pan_{unpack,print}() when applicable

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/6797>

panfrost: decode: Add a macro to dump unpacked descriptors

And use it where it makes sense.

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/6797>

panfrost: decode: Rework the DUMP_{CL,ADDR}() macros

We don't need to pass the indentation level explicitly if we use
pandecode_log() instead of fprintf(). This helps keeping a consistent
indentation across descriptors.

While at it, let's rework those macros to allow creating titles from
a format string.

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/6797>

panfrost: decode: Make the indentation consistent with auto-generated print helpers

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/6797>

panfrost: gen_pack: Allow enum/define values expressed in hexadecimal

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/6797>

panfrost: gen_pack: Add a log2 modifier

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Suggested-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6797>

panfrost: gen_pack: Add an align() modifier

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/6797>

panfrost: gen_pack: Allow empty structs

This is useful if we want to declare padding sections which can be
packed (filled with zeros) and unpacked (checked for non zero entries).

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/6797>

panfrost: gen_pack: Add the aggregate concept

Panfrost descriptors are big and are usually built from a combination of
sub-descriptors. On top of that, layout of sub-descriptors might vary
depending on the architecture version. Since unions are not really an
option (too complex), here is a thin abstraction layer allowing us to
manipulate aggregates in their packed format. Each aggregate is formed
of one or more sections that are meant to be packed/unpacked/printed
separately. Section overlapping is allowed to facilitate handling of
descriptor variants.

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/6797>

panfrost: gen_pack: Move the group get_length() logic to its own method

So we can re-use it elsewhere.

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/6797>

panfrost: gen_pack: Add pan_{unpack,print}() helpers

And use them in pandecode.

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/6797>

panfrost: gen_pack: Fix __gen_unpack_uint()

The mask should be a 64-bit value and we should promote cl bytes to u64
before shifting them.

Fixes: 75cc5b8c2922 ("panfrost: Adopt gen_pack_header.py via v3d")
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/6797>

panfrost: gen_pack: Minor formatting improvement

Escape the \ in pan_pack() so the end result is actually a multi-line
macro.

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/6797>

Revert "panfrost: Drop implicit blend pooling"

This reverts commit dc7fbe114bac338e7075efdb325053ef41dad91c.

Fixes INSTR_INVALID_PC faults with the SuperTuxKart advanced rendering
pipeline, which occurred when blend shader BOs were allocated far apart.

Fixes: dc7fbe114ba ("panfrost: Drop implicit blend pooling")
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6785>

Revert "radv: emit {CB,DB}_RMI_L2_CACHE_CONTROL at framebuffer time"

This reverts commit d6bc0f26c918a67ebf85595d06628e8092e271d0.

These registers are now constant.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6726>

Revert "radv: set BIG_PAGE to improve performance on GFX10.3"

This reverts commit f4d861696dfb11dc2b6242a683a13238981f705f.

Turns out we cannot use BIG_PAGE with GTT and we can't tell
when a buffer is spilled to GTT.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6726>

radv,radeonsi: Disable compression on interop depth images

If we want to use HTILE correctly we need to communicate extra stuff
like clear colors. (Unlike DCC there is no HTILE FCE)

CC: mesa-stable
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6617>

ac/surface: Fix depth import on GFX6-GFX8.

Lets just do depth interop imports by convention between radv and
radeonsi for now. The only thing using this should be Vulkan interop
anyway.

CC: mesa-stable
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6617>

intel/fs/swsb: SCHEDULING_FENCE only emits SYNC_NOP

It's not really unordered in the sense that it can still stall on
ordered things and we don't need a SYNC_NOP for that because it is a
SYNC_NOP. However, it also doesn't count when computing instruction
distances.

Fixes: 18e72ee2108 "intel/fs: Add FS_OPCODE_SCHEDULING_FENCE"
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6781>

nir: Add a pass for lowering CL-style image ops to texture ops

In CL 1.2, images are required to be either read-only or write-only.  We
can always translate the read-only image ops to texture ops.  In CL 2.0
(and an extension), the ability is added to have read-write images but
sampling (with a sampler) is only allowed on read-only images.  As long
as we only lower read-only images to texture ops, everything should stay
consistent.

Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6578>

spirv: Plumb access qualifiers through from image types

In SPIR-V, the access qualifiers for an image are provided on the image
type.  Assuming no one swaps the types around on us (I think that should
be illegal), this means we can reliably fetch the access qualifiers from
the type itself.  The ops which don't really have an easy-to-fetch type
are the atomics because they use OpImageTexelPointer.  However, those
are only allowed on read/write images and that's the default.

Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6578>

spirv: Access qualifiers are not a bitfield

They're an actual enum. My bad.

Fixes: de36b5b805be "nir/vtn: Add support for kernel images to SPIRV-to-NIR"
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6578>

clover: Pull the stride from pipe_transfer for image maps

Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6578>

pan/mdg: Use nir_shader_instructions_pass for fdot2 lowering

Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6718>

pan/mdg: Use nir_shader_instructions_pass for nir_lod_errata

Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6718>

pan/mdg: Return a bool from midgard_nir_lod_errata

To match other NIR passes.

Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6718>

panfrost: Set modifier_constant to true for exported resources

Not actually seen in the wild but could theoretically be a problem for
applications that explicitly import/export resources.

v2: Set it on exporting as well as importing (Daniel Stone)

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6717>

lima: fix vertex shader uniform buffer size

In some cases when switching shader programs, mesa does not switch the
currently set pipe_constant_buffer, which keeps pointing to the one
previously set.
If the two shader programs have a different number of uniforms, the size
of the constant buffer may be different and this needs to be considered
while generating the next draw command.
This patch fixes the uniform buffer creation in the lima vertex shader
command to avoid an out of bounds memcpy due to a previously set
pipe_constant_buffer.

Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6701>

lima: allocate new bo for stream draw

In stream draws, the resource bo might be in use in a previous draw.
Allocate a new one for the resource to avoid overwriting data in use.

Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6503>

radeonsi: Move display dcc dirty tracking to framebuffer emission.

To improve performance.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6783>

radeonsi: Put retile map in separate buffers.

The retile maps are a software mechanism and hence very suceptible
to change. As such I'd like to avoid making it part of the cross
driver ABI.

Ideally we'd just use the cached tile info + a shader to avoid these
buffers altogether.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6783>