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

nir/algebraic: Simplify min and max of b2i

This prevents ~400 shader-db regresssions and a handful of fossil-db
regressions after i2b is always lowered.

All Ivy Bridge and newer Intel platforms had similar results. (Ice Lake shown)
total cycles in shared programs: 855301494 -> 855302118 (<.01%)
cycles in affected programs: 52787 -> 53411 (1.18%)
helped: 4 / HURT: 5

All Intel platforms had similar results. (Ice Lake shown)
Instructions in all programs: 141206055 -> 141205979 (-0.0%)
Instructions helped: 14

Cycles in all programs: 9133376616 -> 9133387327 (+0.0%)
Cycles helped: 13
Cycles hurt: 3

Acked-by: Jesse Natalie <jenatali@microsoft.com>
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tested-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15121>

nir/algebraic: Reassociate some iand to eliminate an operation

No shader-db changes on any Intel platform.

All of the helped shaders were presumably regressed by 4676b3d3dd9 (nir:
Use nir_test_mask instead of i2b(iand)).

v2: Add some comments explaining why specific replacements are used. In
the umin pattern, only markup the first usage of 'b' in the source
pattern.

Tiger Lake, Ice Lake, and Skylake had similar results. (Ice Lake shown)
Instructions in all programs: 141384970 -> 141200966 (-0.1%)
Instructions helped: 45842

Cycles in all programs: 9133648977 -> 9133282672 (-0.0%)
Cycles helped: 26812
Cycles hurt: 6025

Gained: 23
Lost: 135

Acked-by: Jesse Natalie <jenatali@microsoft.com>
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tested-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15121>

nir/algebraic: Remove redundant i2b(b2i(x)) patterns

A loop below already adds all the permutations... including the 1-bit
version that isn't included in this group.

No shader-db or fossil-db changes on any Intel platform.

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Acked-by: Jesse Natalie <jenatali@microsoft.com>
Tested-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15121>

nir/algebraic: Remove redundant i2b(-x) pattern

The exact same pattern appears later (around line 1323).

No shader-db or fossil-db changes on any Intel platform.

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Acked-by: Jesse Natalie <jenatali@microsoft.com>
Tested-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15121>

nir/algebraic: Catch some kinds of copy-and-paste bugs in algebraic patterns

A later commit adds a pattern

   (('umin', ('iand', a, '#b(is_pos_power_of_two)'),
             ('iand', c, '#b(is_pos_power_of_two)')),
    ('iand', ('iand', a, b), ('iand', c, b))),

When I originally made that pattern, I copied and pasted the search to
the replacement as

  (('umin', ('iand', a, '#b(is_pos_power_of_two)'),
            ('iand', c, '#b(is_pos_power_of_two)')),
   ('iand', ('iand', a, '#b(is_pos_power_of_two)'),
            ('iand', c, '#b(is_pos_power_of_two)'))),

The caused the variables in the replacement to be marked is_constant,
and that resulted in an assertion failure deep inside nir_search.

    src/compiler/nir/nir_search.c:530: construct_value: Assertion `!var->is_constant' failed.

These extra validation rules catch this kind of error at compile time
rather than at run time.

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Acked-by: Jesse Natalie <jenatali@microsoft.com>
Tested-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15121>

gallium/pp: typedef and use pp_st_invalidate_state_func to avoid cast

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20042>

gallium/hud: typedef and use hud_st_invalidate_state_func to avoid cast

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20042>

intel: Allow CCS_E on R11G11B10_FLOAT for TGL+

We now support blorp_copy with this format.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jianxun Zhang <jianxun.zhang@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19937>

intel/isl: Bump format_info entries from 100 to 110

The new format support is only tested on Ice Lake and onward. Makes the
next patch clearer.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jianxun Zhang <jianxun.zhang@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19937>

intel/isl: Lessen CCS_E-compatibility checks for TGL+

Tiger Lake and onward allow drivers to specify a compression format
independently from the surface format. So, even if the surface format
changes, hardware is still able to determine how to access the CCS.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jianxun Zhang <jianxun.zhang@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19937>

intel: Hook up RENDER_SURFACE_STATE::DecompressInL3

The sampler's decompressor seems to lack support for some types of
format re-interpretation. Use the more capable decompressor for these
cases. This will be needed to avoid regressing piglit's
arb_texture_view-rendering-formats in later commits.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jianxun Zhang <jianxun.zhang@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19937>

v3dv: skip some invalid tests

Cc: mesa-stable
Signed-off-by: Eric Engestrom <eric@igalia.com>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20305>

asahi: unbind samplers and fix sampler_count if state is NULL

When states is NULL, unbind samplers (to avoid dangling pointers) and
set sampler_count to the highest non-null samplers[] entry instead of
setting it to 0.

This is ported from a similar fix in panfrost:
https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20285

Signed-off-by: Iago López Galeiras <iaguis@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20306>

radv/rt: Handle no-null shader flags

If those flags are set, we can assume that idx is not 0.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20286>

nir: validate that store_buffer_amd doesn't use a non-trivial writemask

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19422>

ac/llvm: implement ACCESS_USE_FORMAT_AMD as buffer_load/store_format

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19422>

ac/llvm: unify load_buffer_amd and store_buffer_amd code

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19422>

nir: add ACCESS_USES_FORMAT_AMD for typed buffer opcodes

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19422>

nir: replace IS_SWIZZLED flag with ACCESS_IS_SWIZZLED_AMD

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19422>

nir: remove redundant SLC_AMD in favor of ACCESS_STREAM_CACHE_POLICY

ACCESS_STREAM_CACHE_POLICY was added to map to SLC for AMD.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19422>

nir: add nir_texop_sampler_descriptor_amd

We'll use it to query the min/mag filter in the shader.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19422>

anv: move functions around to plan for generated draws

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20295>

genxml: add gen12/12.5 CS prefetch disable bits

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20295>

anv: move draw command helpers to their own file

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20295>

anv: move total_batch_size to anv_batch

We'll want 2 batches :

   * the main one

   * another to contain dispatch commands to generate stuff in the
     main batch

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20295>

anv: add helper to get current address of a batch

And allow the function to get the very first address in the batch.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20295>

anv: add a utility function to ensure command buffer space

Ensuring the space is contiguous.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20295>

anv: document utrace field

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20295>

anv: Use extended parameters on Gen11+

Gen11 added a nifty feature where we have three custom system-generated
values called extended parameters that we can set to any 32-bit values
we want. These work just like vertex and instance ID and are controlled
in the pipeline by the 3DSTATE_SGVS_2 packet. They are provided to the
draw call either by extra DWORDs on the end of 3DSTATE_PRIMITIVE or by
storing values to more state registers.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20295>

intel/blorp: Emit a dummy 3DSTATE_VF_SGVS_2

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20295>

intel/genxml: Add 3DPRIMITIVE_EXTENDED opcodes on Gen11+

On Gen11 and above, the 3DPRIMITIVE command takes an optional additional
three DWORDs of data as "extended parameters".  These extended
parameters only exist in the packet if "Extended Parameters Present" is
set.  Because our packing code doesn't handle variable-length commands
well, this commit adds a second version of the command which isn't real
but is just a copy of 3DPRIMITIVE with the additional dwords where the
"Extended Parameters Present" defaults to true and "DWord Length" is
adjusted by 3 as needed.  The 3DPRIMITIVE command is then the gen4-9
version which still works fine but doesn't have the new parameters.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20295>

intel/genxml: Make some fields Booleans

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20295>

anv/utrace: only record the emitted pipecontrol flags

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20295>

intel/ds: trace a couple of more pipe control flags

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20295>

cso: fix a regression from draw_vbo rework

Fixes: c9b13a93385 - cso: remove cso_draw_vbo from all draws, call the driver or u_vbuf directly

Reviewed-by: Brian Paul <brianp@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20288>

panfrost: fix sampler_count and unbind samplers in bind_sampler_states

1. Old approach did not support unbind (set to NULL) samplers because
it only copied memory if sampler is not empty. New approach checks
if sampler is empty - it will set NULL.

2. Old approach just set sampler_count to 0 if sampler is empty.
That's wrong and we need to find highest non-null samplers[] entry.
It was done in new approach.

3. Gallium dosc says:
```
NOTE: at this time, start is always zero ...
This may change in the future.
```
It's better to take into consideration start parameter in new approach.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20285>

anv: disable Wa_1806565034 when robustImageAccess is enabled

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5711
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7859
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20280>

intel/fs: make Wa_1806565034 conditional to non robust access

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20280>

isl: make Wa_1806565034 conditional to non robust access

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20280>

frontends/va: fix crash during grayscale rendering

crash is observed as format is assumed to be yuv and buffer plane
order is undefined for monochrome. luma only format is not to be
considered yuv format. It has to be rendered with rgb=rrr.

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Reviewed-by: Thong Thai <thong.thai@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20165>

radeonsi/vcn: fix regression in yuv422 jpeg decode

- yuv422 decode was blocked in 12acee17fad5, enable it back.
- nv12 yuv422 and grayscale decode is supported on all versions.
- JPEG2 and higher versions supports 444p decode.
- add l8_unorm to supported formats, can be used for grayscale.

Fixes: 12acee17fad5 (frontends/va: reallocate surface for yuv400/yuv444 picture)

v2: indent the switch case correctly (Thong Thai)

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Reviewed-by: Thong Thai <thong.thai@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20165>

radv: do not remove the PointSize built-in for polygon mode as points

Determine if it can be removed when generating the graphics pipeline
key because of dynamic states.

Fixes new CTS dEQP-VK.rasterization.polygon_as_large_points.* and also
spec@!opengl 1.1@polygon-offset with Zink/RADV.

Cc: 22.3 mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20073>

intel/compiler/mesh: extract emit_urb_direct_vec4_write

No functional changes.

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20292>

intel/compiler: optimize away local_inv_index and local_inv_id if workgroup size is 1

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20292>

intel/compiler: split lower_cs_intrinsics_convert_block

No functional changes.

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20292>

intel/compiler/mesh: extract shared code for offset adjustment

No functional changes.

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20292>

anv: simplify EXT_mesh_shader lowering

No functional changes.

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20292>

anv,intel/compiler/mesh: drop lowering of gl_Primitive*IndicesEXT

Until U888X index format lands this change shouldn't have any impact on performance.

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20292>

intel: fix the wrong function name

pipe not create_vertex_elements fuction

Signed-off-by: cheyang <cheyang@bytedance.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20291>

anv: fixup descriptor copies

I did not properly understood that we cannot access the views written
to the descriptor sets because they might have been destroyed after
the write operation and the copy operation is allowed to copy what is
invalid data. The shader just can't access it.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 03e1e19246 ("anv: Refactor descriptor copy")
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20222>

r600: Lower all bitfield ops on pre-EG

Fixes: 4dff3ff005b47befd3e4a903b08d5b4bdbef6ae3
nir/opt_algebraic: Optimize open coded bfm.

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20283>

r600/sfn: Use variable length DOT on Evergreen and Cayman

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20283>

r600/sfn: prepare alu split for use of variable length dot

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20283>

r600/sfn: check used channels when evaluating allowed mask

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7878

v2: Fix ws in comment (iorn10)

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20283>

ci/test-vk: uprev vkd3d

This commit uprevs vkd3d, which flags
test_primitive_restart_list_topology_stream_output as being a flake on
RADV / NAVI21.

Suggested-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Signed-off-by: Martin Roukala (né Peres) <martin.roukala@mupuf.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20278>

ac/nir: add ac_nir_lower_legacy_gs

For legacy (non-NGG) GS to lower outputs to memory stores and add
shader query when required.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20158>

ac/nir: move ngg_gs_shader_query to a common function

To be shared by NGG GS and legacy GS. Legacy GS need this when
GFX10 which mix use NGG and legacy GS. For example when streamout
is enabled, it uses legacy GS, otherwise uses NGG GS. So legacy
GS also need to update query emulation which is a sum of NGG and
legacy GS results.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20158>

nir,ac/llvm,radv: add stream id index to nir_load_ring_gsvs_amd

For used by legacy GS to store output to different ring according
to stream id.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20158>

ac/llvm,radv,radeonsi: pass instruction to intrinsic_load abi

For simple intrinsic which also need other fields to translate
to LLVM like stream_id.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20158>

nir: add nir_load_ring_gs2vs_offset_amd

Used by legacy GS output lowering.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20158>

ac/nir,radeonsi: replace llvm ngg gs invocation query with nir

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20156>

nir,radv,radeonsi: add nir_atomic_add_gs_invocation_count_amd

For shader query emulation.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20156>

freedreno/registers: Add some docs for PM4 events.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19733>

doc/freedreno: Add a bunch of docs of the hardware and drivers.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19733>

freedreno: Clarify that CACHE_FLUSH is pre-5xx.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19733>

ci/docs: Always expose the built docs.

When testing my docs changes outside of an MR, there was no way to look at
the result.

Acked-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19733>

intel: Fix crashes for importing drm buffer

image_aspect_to_binding() converts aspect to index by subrracting
VK_IMAGE_ASPECT_MEMORY_PLANE_0_BIT_EXT, however these enum values
are bitfields, not consecutive numbers, so comparing and subtracting
them won't work.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20269>

hasvk: pipelineStageCreationFeedbackCount is allowed to be 0

Fixes: 6601e5d6fc6 ("anv: implement VK_EXT_pipeline_creation_feedback")

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20216>

vdpau: allow building vdpau st with virgl only

Fixes: 6b5aecb1955 ("virgl: add support for hardware video acceleration")

Signed-off-by: Jan Palus <jpalus@fastmail.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20108>

glsl: fix function inlining for images

Here we skip replacing parameters with their actual values for
images as glsl_to_nir() expects them to be copied to temps first.
Tree grafting has a similiar rule to avoid this happening also.

Fixes: 8d10a6835f32 ("glsl: dont create temps for builtin function inputs")

Tested-by: Martin Roukala <martin.roukala@mupuf.org>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20274>

radv: Improved export conflict bug workaround.

Based on f129db911bd ("radeonsi/gfx11: use a better workaround for the export conflict bug")
which claims better performance.

I couldn't be bothered to do the refactor to check the sample count with
dynamic sample counts, so this is just conservative there.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20273>

mesa: move the _mesa_set_varying_vp_inputs call to where the state changes

This removes the _mesa_update_vao_state() call from all glDraw* functions
where it was executed unconditionally.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: move _DrawVAOEnabledAttribs determination into st_update_array

This can just be computed where it's used.

Now the non-glDraw paths like glRasterPos, glBegin/End, and the slow
display list path have to save and restore _VPModeInputFilter, which
is the only field that's different from the glDraw* path.

_VPModeInputFilter is a bitmask of VP inputs that might have to be bound.
The only difference with glBegin/End and the slow display list path is
that they also add VERT_BIT_MAT_ALL to the bitmask, whereas the glDraw* path
doesn't have that.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: inline _mesa_draw_array_bits & _mesa_draw_current_bits

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: don't AND with VERT_BIT_ALL because it's ~0u

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: simplify VBO state flagging in _mesa_update_vao_state

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: remove VAO::NewVertexBuffers/Elements, set the dirty flags directly

These intermediate dirty flags are unnecessary now.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: move calling _mesa_update_vao_derived_arrays into st_update_array

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: remove the NewVAO flag, set directly what we need to set

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

st/mesa: move st_atom.c contents into st_context.c

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

st/mesa: inline st_validate_state and remove redundant checking in callers

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

st/mesa: remove the switch from st_validate_state by passing state mask directly

Instead of passing the enum that represents which states should be updated
to st_validate_state, pass the state bitmask to st_validate_state directly.
This removes the switch statement over the enums.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

st/mesa: remove st_context::dirty, use gl_context::NewDirtyState instead

We stored the same state in 2 different variables, and then we OR'd them.

This changes the st_validate_state logic slightly, but should be identical.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: remove per_vertex_edgeflags parameter from DrawGalliumVertexState

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: completely rewrite the edge flag state update logic

This removes the edge flag logic from st_validate_state, which was always
executed and added overhead there. Now it's done only when needed.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: don't call _mesa_set_varying_vp_inputs in set_vertex_processing_mode

It's redundant because it's always called in _mesa_update_draw_vao.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: don't call _mesa_update_vao_derived_arrays for dynamic VAOs

This just moves the exit into the caller. There is nothing else in
_mesa_update_vao_derived_arrays that dynamic VAOs care about.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: merge 2 switch statements into one in _mesa_get_derived_vao_masks

_mesa_vao_enable_to_vp_inputs has a switch statement and we call it twice.
Let's inline it.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: remove derived _EffEnabled* fields and compute them at their only use

We were passing the fields to their only use and used them only once,
so let's just compute them there.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: merge _mesa_draw_(user_array,nonzero_divisor)_bits into one helper

preparation for the next change

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: remove unused inline functions from arrayobj.h

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: keep the current VAO also set as _DrawVAO

This removes setting _DrawVAO from all glDraw* functions that draw
primitives because now _DrawVAO is always equal to the current VAO.

glEnd, glRasterPos, and the slow display list path now have to save
_DrawVAO before changing it, and restore it after they are done.

This also removes _EmptyVAO, which becomes unused because we no longer
bind "no VAO" into _DrawVAO.

Rules:
- only glBindVertexArray and non-glDraw* paths should call
_mesa_set_draw_vao, other places shouldn't
- only draws should call _mesa_update_vao_state, other places shouldn't

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

mesa: split _mesa_set_draw_vao into set and update functions

This is the beginning of splitting _mesa_set_draw_vao and moving pieces
of it out of draw functions.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19953>

radv: Handle NULL miss shaders

Fixes reflections in DOOM Eternal.

Fixes: 85580faa ("radv: Add ray traversal loop.")
Closes: #6210
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20270>

radv: Change shaderGroupBaseAlignment to RADV_RT_HANDLE_SIZE

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20270>

radv: Force non uniform texture sampling with NV Remix

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20243>

radv: Add a driconf option to force non uniform texture sampling

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20243>

spirv: Add a debug option to force non uniform texture sampling

Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20243>