broadcom/compiler: try harder to merge thread switch earlier

We have been stopping as soon as we find a conflict but that doesn't
mean we can't merge it in an earlier slot, so keep going. Going by
shader-db, this sometimes allows us to merge the final thrsw a bit
earlier and avoid emitting NOP instructions at the program end to
make up for its delay slots. I have not observed cases where this
helps with regular thrsw though, but it doesn't hurt to try with
those too.

total instructions in shared programs: 11526876 -> 11526354 (<.01%)
instructions in affected programs: 10760 -> 10238 (-4.85%)
helped: 236
HURT: 0
Instructions are helped.

total max-temps in shared programs: 2231705 -> 2231677 (<.01%)
max-temps in affected programs: 276 -> 248 (-10.14%)
helped: 27
HURT: 0
Max-temps are helped.

total inst-and-stalls in shared programs: 11545177 -> 11544655 (<.01%)
inst-and-stalls in affected programs: 10777 -> 10255 (-4.84%)
helped: 236
HURT: 0
Inst-and-stalls are helped.

total nops in shared programs: 321624 -> 321152 (-0.15%)
nops in affected programs: 751 -> 279 (-62.85%)
helped: 236
HURT: 0
Nops are helped.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22679>

radv/sqtt: Add RGP Markers for Mesh Shaders

This change updates the sqtt layer to add mesh shader specific RGP
instrumentation logic. This should allow RGP to correctly identify GPU
work derived from vkCmdDrawMeshTasksEXT, vkCmdDrawMeshTasksIndirectEXT,
and vkCmdDrawMeshTasksIndirectCountEXT API calls.

This change also updates the mesa-to-RGP shader stage translation logic
to handle the mesh & task stages.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21917>

ac/sqtt: Add RGP Definitions for Mesh Shaders

This commit adds RGP marker definitions for the following mesh shader
functions:

- vkCmdDrawMeshTasksEXT
- vkCmdDrawMeshTasksIndirectCountEXT
- vkCmdDrawMeshTasksIndirectEXT

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21917>

ac/spm: rename ac_spm_trace_data to ac_spm

This is shorter and it's the main struct that controls SPM, while
ac_spm_trace contains the generated data only.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22707>

ac/spm: introduce ac_spm_trace and ac_spm_get_trace()

For more code isolation.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22707>

radv: only enable extendedDynamicState3ConservativeRasterizationMode on GFX9+

If extendedDynamicState3ConservativeRasterizationMode is enabled,
VK_EXT_conservative_rasterization is mandatory.

Fixes recent updates in dEQP-VK.info.device_mandatory_features.

Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22681>

nir: clean up white-space in deref-printing

Without this, we get some trailing spaces in some cases, and some
double space in some other cases. Let's clean that up a bit.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Acked-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22550>

radv/video: start adding gfx11 vcn decoder

On gfx11 the vcn decoder moved into the vcn encoder ring,
now known as the unified vcn ring.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21980>

radv/video: fix dpb surface programming

This just programs the dpb fields directly from the image surface.

Makes things happier on navi33

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21980>

radv/video: add a video addr gfx mode

This will be different on gfx11+ so add a settable field for it

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21980>

radv/winsys: handle encoder queue padding/submits.

The encoder queue doesn't like padding at all, don't pad in the sysmem
paths for encoder queue.

The encoder queue will be using on gfx11 for decoding.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21980>

radv: set a video decode ip block in physical device.

This makes it easier to add navi 3x

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21980>

radv/video: add missing gfx family

This is compatible with rdna2 but was just missing.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21980>

radeonsi/ac: move some vcn defines to common

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Acked-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21980>

zink: enable EXT_shader_object for generic precompiles

this should match the functionality of GPL, but it should also (theoretically)
have significantly less CPU overhead, so I've enabled this to be the new
default when available

currently I'm not changing any of the requirements for shader object enablement,
so this is probably only be usable on desktops

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: make zink_shader_spirv_compile public

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: use zink_shader_object for pipeline compiles from zink_gfx_program

this matches up the rest of the codebase using zink_shader_object

zink_gfx_program::objects is left in place for shader binding so that
the entire array can always be bound in one call

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: add a union to zink_gfx_pipeline_cache_entry for gpl

just code motion for now

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: avoid accessing zink_gfx_program::modules during pipeline compile

this allows a different array of modules to be passed in

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: allow zink_shader_module to be either a shobj or a mod using a bool

this allows shader modules to more accurately represent the contents of
the zink_shader_object for which they are containers

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: store spirv onto zink_shader_object structs

this makes it convenient for reuse

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: use zink_destroy_shader_module() for compute to deduplicate code

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: use zink_shader_object for zink_shader_module

this streamlines a bit more code, though nothing changes functionally

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: delete redundant conditional

this is checked literally 2 lines above

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: deduplicate separable program replacement handling

no functional changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: use local screen var in zink_gfx_program_update_optimal()

no functional changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: add zink_program::uses_shobj for managing shader object binds

this is going to be more flexible than just checking for separable+EXT

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: explicitly block sample shading in the GPL precompile path

this can't be precompiled, so don't bother trying

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: rename 'separate' param in shader compilation to 'can_shobj'

this will better reflect the usage of the param

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: add ZINK_DEBUG=noshobj to disable EXT_shader_object

for debugging/testing

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: don't run update_so_info if shader has no outputs

this probably means all outputs were eliminated by the linker,
but in any case it's not going to do anything

fixes spec@glsl-1.50@execution@interface-blocks-api-access-members

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

zink: add a workaround for a nir_assign_io_var_locations bug

drivers that use nir_assign_io_var_locations() with EXT_shader_object all
have the same bug with a shader interface that looks like this:

shader output block:
* PSIZ
* VAR0
* VAR8

shader input block:
* VAR0
* VAR8

in this case, output driver locations will be assigned like:
* PSIZ=0
* VAR0=1
* VAR8=2

and input driver locations will be:
* VAR0=0
* VAR8=1

which breaks the shaders even though this is a totally legitimate thing
to do

thus, a second set of shaders have to be created without PSIZ to work around
the bug since I've already spent 18+ hours trying to fix it and have only succeeded
in breaking every driver that uses it

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22725>

anv: Improve image/view usage bits verification

This change makes usage bits verification closer to the Vulkan spec.
i.e. VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT does not always require all formats
to support all the requested usage bits.
Also, VK_IMAGE_CREATE_EXTENDED_USAGE_BIT, when combined with
VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT can relax the requirements for the
usage supported by the original image format.

v2: Removed strict verification of the format_list_info formats usage
    per chadversary's suggestion. Other minor style/comments tweaks.

v3: Added checking of all compatible formats when
    VK_IMAGE_CREATE_MUTABLE_FORMAT_BIT and VK_IMAGE_CREATE_EXTENDED_USAGE_BIT
    are specified, but no list of possible formats was given.

v4: Add VK_IMAGE_CREATE_BLOCK_TEXEL_VIEW_COMPATIBLE_BIT handling.

Cc: 22.2 <mesa-stable>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6031
Signed-off-by: Sviatoslav Peleshko <sviatoslav.peleshko@globallogic.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17182>

anv: Handle UNDEFINED format in image format list

It's not invalid to have this value in the list, but the only case it
is actually valid as format in the creation of an image or image view
is with Android Hardware Buffers which have their format specified
externally.

So we can just ignore all entries with VK_FORMAT_UNDEFINED.

Cc: 22.2 <mesa-stable>
Signed-off-by: Sviatoslav Peleshko <sviatoslav.peleshko@globallogic.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17182>

isl: Check all channels in isl_formats_have_same_bits_per_channel

Cc: 22.2 <mesa-stable>
Signed-off-by: Sviatoslav Peleshko <sviatoslav.peleshko@globallogic.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17182>

r600: remove TGSI code path

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7780
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7342

Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21521>

glx: fix typos

Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22663>

egl: fix typos

Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22663>

docs: use custom html theme

This uses the new custom HTML theme for Sphinx, so our documentation
looks similar to our website.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8399>

docs: remove support for old sphinx-versions

We're not using them, so no point in keeping support around.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8399>

docs: translate admonitions into bootstrap alerts

Sphinx and Bootstraps disagree on what these are called. Let's try to
bridge that gap, by rewriting things a bit.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8399>

docs: add bootstrap extension

To get Sphinx and Bootstrap to work well together, we need to massage
the output from Sphinx a bit. This adds an extension to do such changes,
based on work from here:

https://github.com/pydata/pydata-sphinx-theme

...However, because we don't ship as an external theme, we can't just do
things as a part of __init__.py, so instead we register an extension
that does the heavy lifting for us.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8399>

docs: add custom html theme

This custom theme will make the docs look more like the main website,
which makes it slightly less confusing to navigate around when crossing
between the two.

The icon link.svg is from Feathericons:
https://feathericons.com/?query=link

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8399>

docs: sort extensions

It's easier to keep this sorted if we have each element on a separate
line. Let's do that for long-term ease of maintainability.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8399>

frontend/nine: Fix num_textures count

util_last_bit already included the +1

Fixes: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11865?commit_id=9205a45da50ebc6159169ba42b02d8475b09bfd7
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8894

cc: mesa-stable

Signed-off-by: Axel Davy <davyaxel0@gmail.com>
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22645>

zink: reapply VkImageViewUsageCreateInfo when rebinding a surface

attachment bits must always be removed if they are not supported

fixes (ANV):
KHR-GL46.texture_view.view_classes

cc: mesa-stable

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22713>

zink: break out VkImageViewUsageCreateInfo applying for reuse

no functional changes

cc: mesa-stable

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22713>

crocus: fix refcnt imbalance related to framebuffer

Indeed, the hardcoded framebuffer cleanup doesn't handle "resolve".

For instance, this issue is triggered with "piglit/bin/glx-copy-sub-buffer -samples=2 -auto"
while setting GALLIUM_REFCNT_LOG=refcnt.log.

Fixes: f5bde99cbdd2 ("gallium: plumb resolve attachments through from frontends -> pipe_framebuffer_state")
Signed-off-by: Patrick Lerda <patrick9876@free.fr>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22554>

rusticl/program: use if let to get rid of an unwrap in build

Signed-off-by: Karol Herbst <git@karolherbst.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22434>

rusticl/program: rework dynamic Program state

We had a lot of implicit locks going on even though there was strictly no
need in doing so. This makes the compilation APIs more atomic while also
providing a cleaner interface.

Not in the mood of splitting it up without deadlocking in the middle. So
it's one big commit sadly.

Signed-off-by: Karol Herbst <git@karolherbst.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22434>

rusticl: Move nir compilation to Program

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22434>

zink: prune some validation errors from ci

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22715>

pan/bi: Don't set has_fsub

All this option controls is fusing fadd(x, fneg(y)) into a single instruction,
but our modifier propagation can already do that without the special cases.

shader-db changes are mostly noise.

total instructions in shared programs: 2698039 -> 2697812 (<.01%)
instructions in affected programs: 156742 -> 156515 (-0.14%)
helped: 280
HURT: 109
helped stats (abs) min: 1.0 max: 12.0 x̄: 1.69 x̃: 1
helped stats (rel) min: 0.08% max: 10.71% x̄: 0.65% x̃: 0.32%
HURT stats (abs)   min: 1.0 max: 8.0 x̄: 2.26 x̃: 2
HURT stats (rel)   min: 0.15% max: 5.77% x̄: 2.08% x̃: 2.22%
95% mean confidence interval for instructions value: -0.80 -0.36
95% mean confidence interval for instructions %-change: -0.05% 0.28%
Inconclusive result (%-change mean confidence interval includes 0).

total cycles in shared programs: 143679.94 -> 143676.33 (<.01%)
cycles in affected programs: 617.98 -> 614.38 (-0.58%)
helped: 148
HURT: 3
helped stats (abs) min: 0.015625 max: 0.1875 x̄: 0.02 x̃: 0
helped stats (rel) min: 0.17% max: 3.33% x̄: 0.71% x̃: 0.57%
HURT stats (abs)   min: 0.015625 max: 0.046875 x̄: 0.03 x̃: 0
HURT stats (rel)   min: 0.66% max: 2.27% x̄: 1.46% x̃: 1.45%
95% mean confidence interval for cycles value: -0.03 -0.02
95% mean confidence interval for cycles %-change: -0.77% -0.56%
Cycles are helped.

total fma in shared programs: 22102.44 -> 22098.33 (-0.02%)
fma in affected programs: 395.98 -> 391.88 (-1.04%)
helped: 123
HURT: 3
helped stats (abs) min: 0.015625 max: 0.1875 x̄: 0.03 x̃: 0
helped stats (rel) min: 0.43% max: 18.75% x̄: 1.81% x̃: 0.89%
HURT stats (abs)   min: 0.015625 max: 0.046875 x̄: 0.03 x̃: 0
HURT stats (rel)   min: 0.66% max: 4.17% x̄: 2.09% x̃: 1.45%
95% mean confidence interval for fma value: -0.04 -0.03
95% mean confidence interval for fma %-change: -2.17% -1.27%
Fma are helped.

total cvt in shared programs: 14735.23 -> 14735.80 (<.01%)
cvt in affected programs: 969.64 -> 970.20 (0.06%)
helped: 179
HURT: 121
helped stats (abs) min: 0.015625 max: 0.09375 x̄: 0.02 x̃: 0
helped stats (rel) min: 0.17% max: 46.15% x̄: 1.13% x̃: 0.31%
HURT stats (abs)   min: 0.015625 max: 0.125 x̄: 0.03 x̃: 0
HURT stats (rel)   min: 0.42% max: 28.57% x̄: 6.62% x̃: 6.25%
95% mean confidence interval for cvt value: -0.00 0.01
95% mean confidence interval for cvt %-change: 1.32% 2.68%
Inconclusive result (value mean confidence interval includes 0).

total quadwords in shared programs: 1460952 -> 1460680 (-0.02%)
quadwords in affected programs: 10216 -> 9944 (-2.66%)
helped: 39
HURT: 5
helped stats (abs) min: 8.0 max: 8.0 x̄: 8.00 x̃: 8
helped stats (rel) min: 0.97% max: 11.11% x̄: 4.53% x̃: 3.45%
HURT stats (abs)   min: 8.0 max: 8.0 x̄: 8.00 x̃: 8
HURT stats (rel)   min: 12.50% max: 25.00% x̄: 15.71% x̃: 14.29%
95% mean confidence interval for quadwords value: -7.74 -4.62
95% mean confidence interval for quadwords %-change: -4.43% -0.04%
Quadwords are helped.

total threads in shared programs: 53538 -> 53537 (<.01%)
threads in affected programs: 2 -> 1 (-50.00%)
helped: 0
HURT: 1

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22687>

docs: add missing ACO_DEBUG=force-waitdeps

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22688>

docs: rename ACO_DEBUG=noscheduling to ACO_DEBUG=nosched

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22688>

egl: Clear EGL_WINDOW_BIT for non-double-buffered EGLConfigs

EGL windows are not required to support single-buffered rendering,
and it's awful, so let's not.

Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22685>

radv: reserve command buffer index for SQTT

These indexes are used to match command buffers with queue events.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22670>

ac/sqtt: add a helper to get cmdbuf IDs per queue

These will be used by RADV to implement queue event timings.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22670>

ac/sqtt: add rgp_sqtt_marker_cb_id definition

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22670>

aco: don't move exec reads around exec writes

Fixes flickering and blocky plants in Jedi: Fallen Order.

Also fixes flickering squares in The Last of Us Part 1.

fossil-db (navi21):
Totals from 92 (0.07% of 135636) affected shaders:
Instrs: 35324 -> 35354 (+0.08%); split: -0.03%, +0.11%
CodeSize: 189568 -> 189668 (+0.05%); split: -0.03%, +0.08%
Latency: 345305 -> 346529 (+0.35%); split: -0.02%, +0.37%
InvThroughput: 78632 -> 78625 (-0.01%)
SClause: 1955 -> 1972 (+0.87%); split: -0.61%, +1.48%
Copies: 1311 -> 1304 (-0.53%); split: -0.69%, +0.15%

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8883
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8878
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22696>

bump VVL to 1.3.248

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22672>

intel/tools: add ability to dump out raw kernels data

Useful for debug.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22703>

r300: add CI list of known rv370 dEQP failures

We don't have CI, but its still convenient for local testing.

Signed-off-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22705>

r300: fix unconditional KIL on R300/R400

  0: KIL -none.1111

Negate is not allowed for texturing opcodes, so the incorrect swizzle
was detected, however later optimization, where we try to rewrite incorrect
swizzles from constant (immediate) registers by adding a new ones with
correct order was interfering and not handling this correctly, so we
ended with

CONST[0] = {    -1.0000    -1.0000    -1.0000    -1.0000 }
  0: KIL const[0].xyz-w;

Even if it would get the swizzle right, texturing opcodes can't read from
constant registers, so just skip it and let this be handled by a later
part which inserts an extra mov instead.

Signed-off-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Reviewed-by: Filip Gawin <filip@gawin.net>
Fixes: a8e1e5b5c2aeb7c2fb4eff2203a026090f0853b9
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22704>

radv/ci: document another vkcts flake on vega10

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/22702>

docs: correct spelling of "frame"

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22286>

docs: correct spelling of "tagged"

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22286>

docs: correct spelling of "source"

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22286>

docs: toplevel -> top-level

This is how we spell it elsewhere.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22286>

docs: vlan -> VLAN

VLAN is an abbreviation, so let's spell it in all-caps for clarity.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22286>

docs: use correct tick for "doesn't"

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22286>

docs: perfetto -> Perfetto

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22286>

docs: Anv -> ANV

We're usually spelling ANV in allcaps, so let's do that here as well.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22286>

docs: cma -> CMA

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22286>

docs: did't -> didn't

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22286>

docs: codepath -> code-path

This is consistent with how we spell this elsewhere in the docs.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22286>

docs: backfacing -> back-facing

This is consistent with how the OpenGL spec spells it.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22286>

docs: statechanges -> state changes

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22286>

docs: renderpass -> render pass

The Vulkan spec spells render pass in two words, so let's do the same.

Reviewed-by: Eric Engestrom <eric@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22286>

glsl: remove ir_state_slot::swizzle

Same story as with the NIR counterpart in the previous commit.

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22620>

nir: remove nir_state_slot::swizzle

This is only ever written to, never read from. Let's just get rid of it!

This also saves us a few needless includes.

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22620>

zink: use EXT_shader_object to implement generic separate shader precompile

this adds precompile for all separate shader stages (+tcs,tes,geom)
using separate shaders, which should eliminate stuttering for games
using it (e.g., Tomb Raider)

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>

zink: fix longstanding TODO for generated tcs

with dynamic pcp this doesn't matter, and this should only be reached
in async mode if dynamic pcp is available

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>

zink: handle all stages in fixup_io_locations()

this makes the handling a bit more complex, as both input and output
need to be handled for most stages, and also the per-component handling
gets trickier

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>

zink: move separate shader creation to shader CSO creation

this is a more logical place for it and also enables u_blitter
shaders to be fast-linked

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>

zink: use a more standardized loop for initing separate shader program descriptors

this should be identical to previous behavior

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>

zink: assign separate shader prog stages from ctx->shader_stages

this is functionally equivalent given the checks above which already
restrict which stages can be passed

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>

zink: move some shader CSO functions around

no functional changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>

zink: switch to a regular loop to wait on precompile shader fences

even if these aren't done yet, it'll still be faster to wait than
to start compiling new pipelines now

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>

zink: streamline separate shader descriptor update

no functional changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>

zink: simplify separate shader prog init a little

no functional changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>

zink: use intermediate variable for separate shader db resize check

no functional changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>

zink: use intermediate variable for separate shader descriptor update loop

no functional changes

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22671>

aco,radv: remove unused aco compile options

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>

aco,ac/llvm,radv,radeonsi: handle ps bc optimization in nir for radv

The side effect is removing the aco/llvm backend bc optimization code
and linear/persp_centroid variable.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>

ac/nir/ps: remove used nir_variable if created

RADV won't do this, so remove them at last.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>

radv: implement nir_load_barycentric_optimize_amd

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>

ac/llvm: remove output variable declaration for radv ps

radv ps does not support epilog when llvm, so outputs will always
be lowered to exports in nir.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>

aco,radv: lower outputs to exports when nir for monolithic ps

Remove the compiler backend code for outputs to exports.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>

aco: support nir_export_amd with ps targets

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22199>