mesa/st: drop ssbo, image and sampler driver state flags bits

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14261>

mesa: drop unused transform feedback state driver flags

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14261>

mesa/st: drop new uniform driver state bit

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14261>

mesa/st: drop new tess state driver bit

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14261>

mesa/st: drop poly stipple driver state bit

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14261>

mesa/st: drop new depth/stencil state bits

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14261>

mesa/st: drop NewBlend driver state flags

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14261>

mesa/st: remove the viewport driver state flags

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14261>

mesa/st: drop the rasterizer driver flags

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14261>

mesa/st: drop the new array driver state bit

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14261>

mesa/st: merge NewDepthClamp state flag

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14261>

mesa: drop optional tex/tnl maintains mode.

These are always going to be on with gallium

v2: drop call, tidy up switch (kwg)

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14261>

mesa/st: move default enabled extensions into mesa.

This just moves a bunch of true assignments into the core

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14261>

mesa/draw: drop the multi draw with indices fallback.

Gallium drivers don't need this.

v2: drop some more code

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14261>

mesa: remove StripTextureBorder option.

Always make this true.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14261>

ci: debian-clang: build more drivers

Add gallium drivers: i915 + asahi
Add vulkan drivers: swrast + panfrost

These can now compile with the current no-error list

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14289>

panvk: cast negative value to unint8_t

The index is a uint8_t but can be assigned a negative 1
value in panvk_pipeline_builder_parse_color_blend()
The comparison to ~0 thus makes sense but clang will complain:

"result of comparison of constant -1 with expression of type
'const uint8_t' (aka 'const unsigned char') is always true
[-Wtautological-constant-out-of-range-compare]"

Fix this by casting to a uint8_t before comparison.

Fixes a warning with clang

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14289>

panvk: use FALLTHROUGH to stop a warning

Fixes a warning with clang

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14289>

i915g: avoid left shifting a negative number

Fixes a warning with clang

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14289>

lavapipe: fix string-plus-int warning

Fixes a warning with clang

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14289>

pan/bi: Don't call useless NIR passes

Cargo culted from the Midgard compiler.

nir_move_vec_src_uses_to_dest is intended for vec4 backends, which does
not apply to Bifrost. nir_lower_locals_to_regs runs much earlier in the
compiler and is a no-op here.

total instructions in shared programs: 107252 -> 107242 (<.01%)
instructions in affected programs: 2403 -> 2393 (-0.42%)
helped: 10
HURT: 0
helped stats (abs) min: 1.0 max: 1.0 x̄: 1.00 x̃: 1
helped stats (rel) min: 0.33% max: 0.57% x̄: 0.43% x̃: 0.42%
95% mean confidence interval for instructions value: -1.00 -1.00
95% mean confidence interval for instructions %-change: -0.49% -0.37%
Instructions are helped.

total tuples in shared programs: 89664 -> 89664 (0.00%)
tuples in affected programs: 333 -> 333 (0.00%)
helped: 1
HURT: 1
helped stats (abs) min: 1.0 max: 1.0 x̄: 1.00 x̃: 1
helped stats (rel) min: 0.52% max: 0.52% x̄: 0.52% x̃: 0.52%
HURT stats (abs)   min: 1.0 max: 1.0 x̄: 1.00 x̃: 1
HURT stats (rel)   min: 0.70% max: 0.70% x̄: 0.70% x̃: 0.70%

total cycles in shared programs: 8103.88 -> 8103.79 (<.01%)
cycles in affected programs: 29.42 -> 29.33 (-0.28%)
helped: 3
HURT: 1
helped stats (abs) min: 0.041665999999999315 max: 0.04166700000000034 x̄:
0.04 x̃: 0
helped stats (rel) min: 0.49% max: 0.55% x̄: 0.53% x̃: 0.54%
HURT stats (abs)   min: 0.04166700000000034 max: 0.04166700000000034 x̄:
0.04 x̃: 0
HURT stats (rel)   min: 0.74% max: 0.74% x̄: 0.74% x̃: 0.74%
95% mean confidence interval for cycles value: -0.09 0.05
95% mean confidence interval for cycles %-change: -1.22% 0.80%
Inconclusive result (value mean confidence interval includes 0).

total arith in shared programs: 3376.42 -> 3376.33 (<.01%)
arith in affected programs: 29.42 -> 29.33 (-0.28%)
helped: 3
HURT: 1
helped stats (abs) min: 0.041665999999999315 max: 0.04166700000000034 x̄:
0.04 x̃: 0
helped stats (rel) min: 0.49% max: 0.55% x̄: 0.53% x̃: 0.54%
HURT stats (abs)   min: 0.04166700000000034 max: 0.04166700000000034 x̄:
0.04 x̃: 0
HURT stats (rel)   min: 0.74% max: 0.74% x̄: 0.74% x̃: 0.74%
95% mean confidence interval for arith value: -0.09 0.05
95% mean confidence interval for arith %-change: -1.22% 0.80%
Inconclusive result (value mean confidence interval includes 0).

total quadwords in shared programs: 79681 -> 79681 (0.00%)
quadwords in affected programs: 283 -> 283 (0.00%)
helped: 1
HURT: 1
helped stats (abs) min: 1.0 max: 1.0 x̄: 1.00 x̃: 1
helped stats (rel) min: 0.62% max: 0.62% x̄: 0.62% x̃: 0.62%
HURT stats (abs)   min: 1.0 max: 1.0 x̄: 1.00 x̃: 1
HURT stats (rel)   min: 0.82% max: 0.82% x̄: 0.82% x̃: 0.82%

total threads in shared programs: 2226 -> 2227 (0.04%)
threads in affected programs: 1 -> 2 (100.00%)
helped: 1
HURT: 0

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14293>

glcpp: Disable the valgrind tests.

We have the glcpp unit tests covered with ASan and MSan in CI, no need to
make everyone doing a "meson test" suffer through valgrind slowly churning
through these.

Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14236>

ci: Enable a build with MSan.

This will catch uninitialized data usage (such as 37855fd59d18 ("glcpp:
Fully initialize struct gl_context")) much faster than valgrind does.

Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14236>

isaspec: Sort field names to generate deterministic output.

Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Acked-by: Rob Clark <robclark@freedesktop.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14259>

microsoft/compiler: dxil_nir_opt_alu_deref_srcs: return progress

dxil_nir_opt_alu_deref_srcs will always return false because
the progress variable is declared both for the function and also
inside the loop.

Spotted by a unused-but-set-variable warning from clang

Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14290>

pan/va: Generalize LD_VAR_IMM_* to support flat varyings

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14292>

pan/va: Add .signed bit to right shift instructions

This makes the RSHIFT_* family of instructions act like ARSHIFT.* on
Bifrost.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14292>

pan/va: Rename LEA_ATTR to LEA_VARY

This more accurately reflects the function of the instruction. Unlike
earlier Malis, there don't seem to be attribute records for varyings.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14292>

pan/va: Remove extra LD_VAR_IMM_F32 source

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14292>

intel/compiler: drop unused decleration

Acked-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14202>

intel/compiler: don't lower swizzles in backend.

These are lowered by crocus in the frontend, the key
entries are still used.

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/14202>

intel/compiler: remove gfx6 gather wa from backend.

Crocus lowers this in the frontend, they key member is still used
but reset prior to backend.

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/14202>

mesa/dd: drop unused InvalidateBufferSubData entry.

I already removed the users of this, but forgot the entry.

Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14286>

mesa: drop unused _mesa_new_program.

This isn't used since classic removal.

Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14286>

mesa: drop unused new renderbuffer code.

This isn't used anywhere currently

Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14286>

mesa/st: use has_stencil_export instead of querying screen cap.

We already query at context setup, just use that value.

Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14286>

mesa: drop texformat code this isn't used.

v2: dropped comment (Tapani)

Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14286>

spirv: Allow VRS with mesh shaders.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14291>

draw: drop unused function

Introduced in 381e9fe6. Never used.

Fixes a compile warning with clang

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14252>

gallium/tgsi_exec: drop unused function

Introduced in 9ca6cf0f and last usage
dropped in 31369987

Fixes a compile warning with clang

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14252>

gallium/u_threaded: drop unused function

tc_drop_sampler_view_reference is unused. It was introduced
in 340703e0 and its last usage was dropped in bb89cf4bf39

Fixes a compile warning with clang

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14252>

glx: remove a set but not used variable

total_sent was never used. It was introduced in
fdb07636f2e6324c5250cd5ee97778b7f5933bea

Fixes a warning in clang

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14252>

r600: remove a set but not used variable

grid_size was never used. Not even when introduced
in 6a829a1b724ca0d960decee217d260b4de8a5463

Fixes a warning on clang

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14252>

r300: remove a set but not used variable

The use of phase_refmask was removed 12 years ago
in b7cf887ca74561469c144f1d12227e1bcf277e7e

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14252>

i915g: fix implicit-fallthrough warning

Fixes a warning on clang.

Uses FALLTHROUGH like the surrounding code.

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14252>

lavapipe: fix implicit-fallthrough warning

Fixes a warning on clang

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14252>

intel/compiler: disable workaround not applicable to gfx >= 11

There's nothing in bspec that would suggest this is still needed.
It only affected gfx 9 and 10.

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14280>

etnaviv: Use mesa_log*

Makes it consistent with the DRM bits

Signed-off-by: Guido Günther <guido.gunther@puri.sm>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Acked-by: Lucas Stach <l.stach@pengutronix.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10546>

entaviv/drm: Use same log format as gallium bits

We prefix with __func__:__LINE__ there.

Signed-off-by: Guido Günther <guido.gunther@puri.sm>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Acked-by: Lucas Stach <l.stach@pengutronix.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10546>

etnaviv/drm: Use mesa_log* for debugging

This makes sure errors, warnings and info messages don't get
compiled out in non debug builds.

Signed-off-by: Guido Günther <agx@sigxcpu.org>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Acked-by: Lucas Stach <l.stach@pengutronix.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10546>

etnaviv/drm: Print gpu model at debug verbosity

Otherwise we print it at every application start.

Signed-off-by: Guido Günther <guido.gunther@puri.sm>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Acked-by: Lucas Stach <l.stach@pengutronix.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10546>

etnaviv/drm: Add some bo debug output

This makes it simpler to trace BO usage.

Signed-off-by: Guido Günther <agx@sigxcpu.org>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Acked-by: Lucas Stach <l.stach@pengutronix.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10546>

etnaviv/drm: Use etna_mesa_debug for debugging messages

We use the variable from gallium but fall back to a weak symbol
in case there's usage out of galllium in the future.

Signed-off-by: Guido Günther <agx@sigxcpu.org>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Acked-by: Lucas Stach <l.stach@pengutronix.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10546>

glsl: fix invariant qualifer usage and matching rule for GLSL 4.20

I noticed that GLSL version referenced here was wrong, version 4.20 is
first spec that does not allow invariant keyword for inputs.

v2: fix all comments (Timothy Arceri)

Fixes: f9f462936ad ("glsl: Fix invariant matching in GLSL 4.30 and GLSL ES 1.00.")
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14241>

ci: debian-clang: drop -Wno-error for self-assign

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14272>

gallivm: avoid a self-assign warning

Fixes a clang warning

Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14272>

panfrost: Avoid double unlock.

Fix defect reported by Coverity Scan.

Double unlock (LOCK)
double_unlock: pthread_mutex_unlock unlocks dev->indirect_draw_shaders.lock while it is unlocked.

Fixes: 2e6d94c198e ("panfrost: Add helpers to support indirect draws")
Suggested-by: Alyssa Rosenzweig <alyssa@collabora.com>
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14262>

ir3: Make shift operand 64-bit.

Fix defect reported by Coverity Scan.

Unintentional integer overflow (OVERFLOW_BEFORE_WIDEN)
overflow_before_widen: Potentially overflowing expression 2 << W
with type int (32 bits, signed) is evaluated using 32-bit
arithmetic, and then used in a context that expects an expression
of type uint64_t (64 bits, unsigned).

Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Acked-by: Rob Clark <robclark@freedesktop.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14258>

aco/optimizer_postRA: Fix applying VCC to branches.

Fixes: a93092d0edc92eea8e8e96709ad9857f05c45cef
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14281>

aco/optimizer_postRA: Fix combining DPP into VALU.

Fixes: 4ac47ad1cd7976d7effbbfae37fa69e26a288ad2
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14281>

ci: clean up debian-clang no-error list

I see no warnings for these on a local build

Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14269>

anv: Simplify assertions related to graphics stages

In all three cases, COMPUTE was on the table but with an invalid
value (zero).  Drop it from the tables and the extra assertion, so if
a COMPUTE is passed it will just fail the ARRAY_SIZE assertion.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14274>

aco/ra: fix get_reg_for_operand() in case of stride mismatches

We have to clear the register file from the previous operand
as otherwise, there might be no space left.

Totals from 5 (0.00% of 134572) affected shaders: (GFX10.3)
CodeSize: 21144 -> 21000 (-0.68%); split: -0.72%, +0.04%
Instrs: 3738 -> 3720 (-0.48%); split: -0.51%, +0.03%
Latency: 517229 -> 516319 (-0.18%); split: -0.18%, +0.00%
InvThroughput: 49068 -> 48902 (-0.34%); split: -0.38%, +0.04%
Copies: 501 -> 483 (-3.59%); split: -3.79%, +0.20%

Cc: mesa-stable
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14279>

d3d12: Fix NV12 resource importing

Fixes: a6db8054 ("d3d12: Handle opening planar resources")
Reviewed-by: Sil Vilerino <sivileri@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14276>

nir/opt_algebraic: lower fneg_hi/lo to fmul

This pattern, found in the FSR upscaling shader,
helps the vectorization efforts by keeping the
chain of vectorized instructions intact.
Radeon can optimize it to per-component fneg modifiers.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13688>

aco/optimizer: propagate and fold inline constants on VOP3P instructions

This patch aims to propagate and fold constants on VOP3P instructions
by using omod selection and the fneg modifier.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13688>

aco: change fneg for VOP3P to use fmul with +1.0

This will be useful to be able to also apply
fneg_lo and fneg_hi.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13688>

aco/optimizer: fix fneg modifier propagation on VOP3P

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13688>

anv: Refactor dirty masking in cmd_buffer_flush_state

Instead of masking the dirty variable itself, use an appropriate mask
in the users of dirty.  This will avoid extra tracking when dealing
with Task/Mesh later.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14275>

anv/blorp: Split blorp_exec into a render and compute

And set the relevant push_constants_dirty for each case.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14275>

v3dv: add Android support

Acknowledgements to android-rpi team and lineage-rpi maintainer (KonstaT)
for creating/testing initial vulkan support. Their experience was used as
a baseline for this work.

Most of the code is a copy of turnip and anv.
Improved by cleaning dEQP failures:

 - Improved gralloc support (use allocation time stride, size, modifier).
 - Fixed some dEQP crashes due to memory allocation issues.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Signed-off-by: Roman Stratiienko <r.stratiienko@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14016>

r300/vs: Fix flow control processing just after an endloop.

We tried to step over the instruction we just generated, except we didn't
always just generate one.  In the sequence_vertex tests, that meant we
skipped processing the next BGNLOOP and then underflowed our stack.

Cc: mesa-stable
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14271>

r300/vs: Reuse rc_match_bgnloop().

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14271>

r300/vs: Allocate temps we see a use as a source, too.

This is a quick hack for a bunch of the fail in #5766.

Cc: mesa-stable
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14271>

ci/r300: Add another xfail on the main branch.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14271>

intel/fs: Add physical fall-through CFG edge for unconditional BREAK instruction.

This adds a missing CFG edge that represents a possible physical
control flow path the EU might take under some conditions which isn't
part of the logical CFG of the program.  This possibility shouldn't
have led to problems on platforms prior to Gfx12, since the missing
control flow edge cannot possibly influence liveness intervals.
However on Gfx12+ it becomes the compiler's responsibility to resolve
data dependencies across instructions, and the missing physical
control flow paths may lead to a WaR data hazard currently not visible
to the software scoreboard pass, which could lead to data corruption.

Worse, the possibility for this path to be taken by the EU increases
on Gfx12+ due to a hardware bug affecting EU fusion -- However the
same physical path can be potentially taken on earlier platforms as
well, so this patch extends the CFG on all platforms for consistency,
even though the lack of this edge shouldn't lead to any functional
issues on platforms earlier than Gfx12.  There are no shader-db
changes on earlier platforms, so there seems to be no disadvantage
from using the same CFG representation as on later platforms.

This issue has ben reported on TGL with the following conformance
test, thanks to Ian for bringing the FULSIM dependency check warning
to my attention:

   dEQP-VK.graphicsfuzz.spv-stable-pillars-volatile-nontemporal-store

Fixes: 4d1959e69328cf ("intel/cfg: Represent divergent control flow paths caused by non-uniform loop execution.")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4940
Reported-by: Tapani Pälli <tapani.palli@intel.com>
Reported-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14248>

glsl: Retire unused modes for lower_64bit_integer_instructions.

Unused since 424ac809bfac ("i965: Do int64 lowering in NIR")

Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14249>

glsl: Remove comment about non-existing DFREXP_TO_ARITH

Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14249>

glsl: Remove dead prototype for old do_discard_simplification().

Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14249>

glsl: Delete the optimize_redundant_jumps pass.

Nothing here that NIR doesn't do.  No effect on shader-db of hsw or
softpipe.

Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14249>

glsl: Delete the vectorization opt pass.

Nothing uses it, and i965 was the last thing to.  Even if I enable it for
softpipe or crocus, it quickly causes NIR validation failures in shader-db
from swizzles outside the bounds of vectors.  Retire it in favor of
nir_opt_vectorize().

Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14249>

freedreno/ir3: Dump const state with shader disasm

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14231>

freedreno/computerator: Mark shader bo for dumping

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14231>

freedreno/computerator: Fix @buf header

Order is important in the grammar, the more specific match needs to go
first.

Fixes: ba1c989348d ("freedreno/computerator: pass iova of buffer to const register")
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14231>

freedreno/ir3: Handle instr->address when cloning

Without this, a cloned instruction that takes full regs will trigger an
ir3_validate assert.  This can happen, for ex, if an instruction that
writes p0.x and has a relative src gets cloned in ir3_sched.

Fixes an assert in Genshin Impact with a debug build.

Fixes: 9af795d9b98 ("ir3: Make ir3_instruction::address a normal register")
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14231>

pan/bi: Specialize shaders for IDVS

We need to compile multiple variants and report them together in a
common shader info. To do so, we split off per-variant shader infos and
combine at the end.

glmark2 is very happy: https://people.collabora.com/~alyssa/idvs-g52.txt
Highlights include -bshading up 41% fps and -bbump:bump-render=high-poly
up 62% faster

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14154>

pan/bi: Add helper to decide if IDVS should be used

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14154>

pan/bi: Use position shader ST_CVT path

We need to use a preload instead of the LEA_ATTR. Not sure why.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14154>

pan/bi: Split out varying store paths

This means we don't need to special case IDVS quite so hard.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14154>

pan/bi: Remove the "wrong" stores in IDVS variants

Position shaders should only write gl_Position (and gl_PointSize on
Valhall), varying shaders should only write varyings.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14154>

pan/bi: Add IDVS mode to bi_context

Various parts of the compiler switch behaviour based on IDVS variant.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14154>

pan/bi: Allow UBO pushing to run multiple times

For IDVS.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14154>

pan/bi: Extract bi_finalize_nir

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14154>

panfrost: Add panfrost_compile_inputs->no_idvs option

panvk will want IDVS support eventually, but not right now. Allow the
driver to opt out of IDVS in the mean time.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14154>

panfrost: Align instance size for IDVS

Hardware requirement. Failing to do this raises a DATA_INVALID_FAULT.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14154>

panfrost: Skip rasterizer discard draws without side effects

Minor optimization, but more importantly fixes an interaction of IDVS
with rasterizer discard.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14154>

panfrost: Extract panfrost_batch_skip_rasterization

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14154>

panfrost: Emit IDVS jobs

When trying to draw with an IDVS capable shader.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14154>

panfrost: Extract panfrost_draw_emit_vertex_section

To be shared with IDVS.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14154>

panfrost: Set secondary_* fields for IDVS

Easy now that we've split everything out nicely.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14154>