zink: optimize shader recalc

now we only have to loop over the changed shaders that exist for the
program, and we can avoid a lot of hashing operations by reusing stored
hash values where possible

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11962>

zink: set dedicated allocation when needed

Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11958>

zink: hook up VK_KHR_dedicated_allocation

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11958>

zink: check for dedicated allocation requirements during image alloc

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11958>

freedreno/ir3: Reduce choose_instr_dec() and _inc() overhead.

If you didn't have a freed+ready instruction, you'd redo the live_effect
and check_instr() logic multiple times per instr.  Replace the multiple
loops in each function with a ranking that I think is more readable,
reducing the overhead in the process.

debugoptimized dEQP-GLES31.functional.ubo.random.all_per_block_buffers.20
runtime goes from ~3.5s -> ~3.0s on my lazor.  No shader-db change.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11855>

features: VK_EXT_color_write_enable for lavapipe

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11961>

lavapipe: implement VK_EXT_color_write_enable

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11961>

lavapipe: add a padding member to rendering_state

I always forget where to steal bits from otherwise

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11961>

gallium: Remove dead pb_malloc_buffer_create function prototype

This was deleted by commit ec686a66dbf2180631564a3877884b1e4d0c8ea4.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11922>

wl/shm: don't fetch formats if not requested.

This aligns the code with the drm format code

Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Simon Ser <contact@emersion.fr>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11945>

v3d/driconf: Expose non-MSAA texture limits for mutter and gnome-shell

To enable dual 4k displays on mutter or gnome-shell under X11 we need to
expose the non-MSAA texture limit as we did for Xorg at 60a64f028d75ad6
("v3d: Use driconf to expose non-MSAA texture limits for Xorg.")

https://gitlab.gnome.org/GNOME/mutter/-/issues/1874

Cc: 21.1 21.2 mesa-stable
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11926>

nv50/ir/nir: fix smem size for GL

Originally I fixed the case where the nir itself has a shared mem size of
0, but the frontend (e.g. clover) set it to some other value.

But st/mesa sets the shared mem size on the state object as well and we
end up actually doubling the value in the driver as we set smemSize to the
value from the state object before calling into the compiler.

So just max the value instead.

Fixes the compute_shader.shared-max CTS test.

Fixes: dc667b1f192 ("nv50/ir/nir: fix smem size")
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11047>

v3dv: implement VK_EXT_physical_device_drm

Reviewed-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9320>

nouveau: Support nir_intrinsic_*_atomic_fadd

This fixes failures in piglit's nv_shader_atomic_float tests with
NV50_PROG_USE_NIR=1

Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9866>

softpipe/aniso: move DDQ calculation to after scaling.

I've read the papers on EWA filters and it seems like the calculate
DDQ = 2 * A after the scaling of A happens. This seems to make
things less blurry and more like real aniso.

Fixes: 2135aba8 ("softpipe: Constify variables")
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11917>

pan/bi: Shrink live array to 8-bits

We only actually use 4-bits, so we could shrink again. But this by
itself means 1/2 the memory usage for liveness analysis and 1/2 the
copying/alloc/free.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11936>

pan/bi: Clean up liveness freeing

This is ralloc'd, kind of pointless.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11936>

pan/bi: Clean up useless casts

Left over from removing inheritance with sed instead of coccinelle.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11936>

pan/bi: Copy block bi_block

Gets rid of the silly inheritance everywhere, which has caused _far_
more problems in practice than it has fixed. It was an idea I tried
before the pandemic. It didn't work. I'm finally cleaning it up.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11936>

pan/bi: Copy back bi_foreach_successor

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11936>

pan/bi: Copy back add_successor

Trying to get back independent block types.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11936>

pan/bi: Copy liveness routines back

We'll diverge shortly.

Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11936>

aco: Fix how p_elect interacts with optimizations.

Since p_elect doesn't have any operands, ACO's value numbering and/or
the pre-RA optimizer could currently recognize two p_elect instructions
in two different blocks as the same.

This patch adds exec as an operand to p_elect in order to achieve
correct behavior.

Fixes: e66f54e5c83fd545e1a4062e683b584a35dacc00
Closes: #5080
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/11943>

glsl: replace some C++ code with C

This replaces some new/delete uses with malloc/free.

This is more consistent with most of the other glsl IR code but
more importantly it allows the game "Battle Block Theater" to
start working on some mesa drivers. The game overrides new and
ends up throwing an assert and crashing when it sees this
function calling new [0].

Note: The game still crashes with radeonsi due to similar conflicts
with LLVM.

CC: mesa-stable
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11907>

mesa: fix incorrect comment in draw_gallium_multimode

Fixes: 4566383ae4b ("gallium: move pipe_draw_info::index_bias to pipe_draw_start_count_bias")

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11939>

mesa: remove unused drawid_offset parameter from DrawGalliumMultiMode

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11939>

st/mesa: inline st_setup_arrays on MSVC too by adding a wrapper

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11939>

util/foz: stop crashing on destroy if prepare hasn't been called

Fixes: eca6bb9540d ("util/fossilize_db: add basic fossilize db util to read/write shader caches")

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11931>

anv: fix feature/property/sizes reported for fragment shading rate

Fixes: 790ff1ceac8 ("anv: fixup physical device properties of fragment shading rate")

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11897>

util/tc: make clear calls async

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11864>

radeonsi: document a missing synchronization for bindless textures

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11900>

radeonsi: set scratch_state dirty only if ctx->scratch_buffer allocated

if ctx->scratch_buffer is NULL, then no need to update SPI_TMPRING_SIZE
size register.

Signed-off-by: Yogesh mohan marimuthu <yogesh.mohanmarimuthu@amd.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11900>

radeonsi: remove redundant setting scratch_state atom dirty

Whenever scratch buffer is allocated, current spi_tmpring_size and
previous spi_tmpring_size cannot be same and hence scratch_state will be
set dirty as part of "if (spi_tmpring_size != sctx->spi_tmpring_size)".
Removing redundant dirty bit sat while allocating scratch buffer.

Signed-off-by: Yogesh mohan marimuthu <yogesh.mohanmarimuthu@amd.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11900>

radeonsi: don't expose no-attachment MSAA 16x on all 1 RB chips due to issues

Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11900>

lavapipe: don't read line stipple info in pipeline creation if stipple is disabled

otherwise these values may be (harmlessly) garbage

Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11933>

gallium/ttn: Add a debug flag for dumping the shaders.

Just like for ntt, it's nice to be able to dump the shader that assertion
failed on you.

Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11915>

nir_to_tgsi: Fix image declarations.

We failed to translate the target type, which virgl needs for translation.
Also the read_only flag is for consts, shader inputs, and uniforms.  The
access flag gives you the readonly qualifier.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11916>

anv: Stop reference counting semaphores

The only reason we had to refcount semaphores was for the ancient
sync_file semaphores which we used for pre-syncobj kernels.  Now that we
assume syncobj and that code is gone, we don't need reference counting
anymore either.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9777>

anv: Drop unused sync_file and BO semaphore code

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9777>

anv: Assume syncobj support

Sync object for i915 support has been in upstream Linux since 4.14 which
is 3.5 years old at this point and, as far as we can tell, it also
exists in all the ChromeOS kernels.  Assuming it allows us to drop some
of our more gnarly synchronization fall-back paths.

At the time of merge, ChromeOS was on the following kernels:

 - kernel 3.18: SKL
 - kernel 4.4: BYT, KBL, APL
 - Kernel 4.14: BDW, GLK

All of the pre-4.14 kernels have had syncobj support back-ported.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9777>

intel/eu: Start validating LSC message descriptors

This is certainly not a full validation but it at least gets the
framework in place and validates one hard-to-find restriction.

Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11657>

intel/compiler: Enable has_iadd3 option on XeHP

shader-db result is inconclusive but doesn't harm to include it for
reference.

Shader-db result on XeHPG:

total instructions in shared programs: 1397405 -> 1397315 (<.01%)
instructions in affected programs: 88252 -> 88162 (-0.10%)
helped: 20
HURT: 7
helped stats (abs) min: 1 max: 18 x̄: 7.20 x̃: 7
helped stats (rel) min: 0.03% max: 2.20% x̄: 0.37% x̃: 0.23%
HURT stats (abs)   min: 4 max: 23 x̄: 7.71 x̃: 4
HURT stats (rel)   min: 0.10% max: 0.68% x̄: 0.22% x̃: 0.11%
95% mean confidence interval for instructions value: -6.81 0.14
95% mean confidence interval for instructions %-change: -0.42% -0.02%
Inconclusive result (value mean confidence interval includes 0).

total cycles in shared programs: 119924219 -> 119931868 (<.01%)
cycles in affected programs: 45029193 -> 45036842 (0.02%)
helped: 11
HURT: 16
helped stats (abs) min: 15 max: 5490 x̄: 1655.73 x̃: 140
helped stats (rel) min: <.01% max: 0.35% x̄: 0.11% x̃: <.01%
HURT stats (abs)   min: 1 max: 2944 x̄: 1616.38 x̃: 1743
HURT stats (rel)   min: <.01% max: 0.17% x̄: 0.09% x̃: 0.10%
95% mean confidence interval for cycles value: -606.11 1172.70
95% mean confidence interval for cycles %-change: -0.04% 0.07%
Inconclusive result (value mean confidence interval includes 0).

v2:
- Include shader-db result (Jason)

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11596>

nir: Add optimizations for iadd3

This patch also adds has_iadd3 bit to give more control if backend
supports ternary add instruction or not.

v2:
- Add patterns in late optimization (Connor Abbott)

Suggested-by: Alyssa/Jason
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11596>

intel/compiler: Allow ternary add to promote source to immediate

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11596>

intel/compiler: Make decision based on source type instead of opcode

This patch restructure code a little bit to check if source can be
represented as immediate operand. This is a foundation for next patch
which add checks for integer operand as well.

Suggested-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11596>

intel/compiler: Add support for ternary add instruction on XeHP

v2:
- Re-arragne opcode in correct order (Matt Turner)
- Move ADD3 case closer to LRP (Jason)

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11596>

nir: Add new opcode for ternary addition

v2:
- Make it 2src commutative (Connor Abbott)

Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11596>

radv: Use pre-computed viewport transform for NGG culling state.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11839>

radv: remove unused variable from radv_emit_viewport

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11839>

radv: pre-calculate viewport transforms

this requires more storage in the viewport struct, but it avoids
the need to repeatedly calculate the same transform if e.g., a meta
operation occurs, which can save about 5% cpu in some cases

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11839>

radv: merge si_write_viewport into radv_emit_viewport

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11839>

freedreno: fix wrong tile aligment for 3 CCU gpu

Fixes: 78c8a8af80d5b0ccd68dd85ff7ecbb699be31f7d "freedreno: Generate device-info tables at build time"
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5060

Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11928>

aco: Remove s_and with exec when all lanes are active.

This helps NGG GS and culling shaders.
No Fossil DB changes without NGG culling.

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/11458>

aco: Allow elect to take advantage of knowing when all lanes are active.

Implement elect using a pseudo-op which is lowered during the
insert_exec_mask pass. This makes it possible to emit a more
optimal sequence when the exec mask is constant.

Fossil DB results on Sienna Cichlid:
Totals from 211 (0.16% of 128647) affected shaders:
CodeSize: 2254356 -> 2240468 (-0.62%); split: -0.62%, +0.00%
Instrs: 438471 -> 434996 (-0.79%); split: -0.80%, +0.01%
Latency: 2717082 -> 2709400 (-0.28%); split: -0.28%, +0.00%
InvThroughput: 566987 -> 566342 (-0.11%); split: -0.11%, +0.00%
Copies: 40058 -> 40162 (+0.26%)
Branches: 31209 -> 31211 (+0.01%)
PreSGPRs: 9927 -> 10125 (+1.99%)

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/11458>

aco: Swap s_and operand order for ballot.

This allows our optimizer to recognize this and eliminate it when
it can prove that the s_and with exec is unneeded.

Fossil DB changes on Sienna Cichlid:
Totals from 1969 (1.53% of 128647) affected shaders:
CodeSize: 9468228 -> 9469348 (+0.01%); split: -0.00%, +0.01%
Instrs: 1773566 -> 1773581 (+0.00%); split: -0.01%, +0.01%
Latency: 19504042 -> 19503385 (-0.00%); split: -0.00%, +0.00%
InvThroughput: 3617406 -> 3617333 (-0.00%)
Copies: 108998 -> 110592 (+1.46%)

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/11458>

intel/perf: Use intel_i915_query_flags instead of hand-rolling it

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11770>

intel/dev: Use intel_i915_query_alloc in query_topology

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11770>

iris: Use intel_i915_query for meminfo

Error handling with DRM_IOCTL_I915_QUERY is tricky and we got it wrong
in one of the two calls here.  Use the common helper instead.  This also
fixes a theoretical bug where calloc() fails.  While we're here, inline
iris_bufmgr_update_meminfo because we're not really benefiting from
having it separate anymore.

Fixes: e60114b2ae11 "iris/bufmgr: Query memory region info."
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11770>

anv: Use intel_i915_query_alloc for memory regions

Error handling with DRM_IOCTL_I915_QUERY is tricky and we got it wrong
in one of the two calls here.  Use the common helper instead.  This also
fixes a theoretical bug where calloc() fails.  While we're here, inline
anv_track_meminfo because we're not really benefiting from having it
separate anymore.

Fixes: 65e8d72bc106 "anv: Query memory region info"
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11770>

intel: Pull anv_i915_query into common code

We also add a helper which contains the standard query+alloc+query
pattern used by anv_gem_get_engine_info().  The caller is required to
free the pointer.

These are declared static inline not because we care about the
performance of these helpers but because we're going to use them in the
intel_device_info code and we don't want a link dependency.

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11770>

anv: Handle errors properly in anv_i915_query

DRM_IOCTL_I915_QUERY is a multi-query.  The most egregious errors are
returned via the usual ioctl error mechanism but there are also
per-query errors that are indicated by item.length < 0.  We need to
handle those as well.  While we're at it, scrape errno so we can return
a proper integer error.

Fixes: c0d07c838a9f "anv: Support i915 query (DRM_IOCTL_I915_QUERY)..."
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11770>

radv: Remove num_viewports from radv_skip_ngg_culling.

NGG culling is not compiled into shaders that can use multiple
viewports, so it's not necessary to check it here.

Fixes: 9a95f5487f5ab83fa44bea12afa30cf1a25fc9db
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/11910>

radv: Don't compile NGG culling into shaders that write viewport index.

We don't support NGG culling with multiple viewports yet.

Fixes: f30e4351de2f562955435a04cf75dd641639d31c
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/11910>

iris: Re-emit MEDIA_VFE_STATE for variable group size shaders

It implicitly contains the number of threads via the CURBE allocation
size field.

Fixes: 33c61eb2f105 "iris: Implement ARB_compute_variable_group_size"
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10640>

v3dv: don't emit frame setup more than once for multilayered framebuffers

We only need to do this once per job, not once per layer.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11923>

v3dv: don't overallocate tile state for meta TLB operations

Generally, all meta operations that are implemented via the TLB and
only involve tile load/store operations only need one layer worth
of tile state memory and a single frame setup pointing the binner
to that memory.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11923>

v3dv: allow limiting amount of tile state allocated

With multilayered framebuffer we want to allocate enough tile state for
all layers involved, so te binner can handle layered rendering where
a geometry shader is used to redirect primitives to specific layers by
writing to gl_Layer.

However, we may also have layered framebuffers in cases where layered
rendering won't be used. Typically this will happen for meta copy/clear
operations, where we setup multilayered framebuffers but then we just
load and/or store the tile buffer without ever rendering a primitive,
let alone use a geometry shader to do layered rendering. In these cases
we can reduce the amount of tile state allocated to a sigle layer.

This patch allows us to specify if we should allocate tile state for all
layers when we start a new frame. We will take advantage of this in
later patches targetting the meta copy/clear code paths.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11923>

v3dv: improve TLB layered image clears

Our meta image clearing code was emitting a job per layer to clear,
but we can emit a single job that packs multiple layer clears for
better performance. For this we just need to make sure we pass
the maximum layer we can emit to v3dv_job_start_frame so we
allocate sufficent tile state memory.

Actually, since this operation is TLB based and doesn't make use of
layered rendering, we could do with allocating less tile state
memory too. We will improve this for this and many others TLB meta
implementations in follow up patches.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11923>

v3dv: remove unused layer field from struct rcl_clear_info

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11923>

radv: fix specifying the stencil layout for separate depth/stencil layouts

The Vulkan spec has been updated few months ago again and pNext is
always honored if present.

Found this with vkd3d-proton which implemented separate depth/stencil
layouts recently.

Cc: 21.2 mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11903>

radv: report APUs as discrete GPUs for Red Dead Redemption 2

On APUs, we fake heaps to simulate a dGPU setup because it seems to
have the maximum compatibility. Though, some applications like RDR2
still only looks at GTT if the driver reports an iGPU which means it
will only use 1/3rd of total memory available.

This is currently behind a drirc option because it might have
implications for other apps but we might want to extend this later
if everything is fine.

Cc: 21.2 mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11891>

radv: fix bounds checking for zero vertex stride on GFX6-7

GFX6 and GFX10+ have similar logic.

This fixes test_zero_vertex_stride from vkd3d-proton on
Pitcairn (GFX6) and on Bonaire (GFX7).

Cc: 21.2 mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11904>

nir: Better document the Boissinot algorithm in nir_from_ssa()

Reviewed-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8815>

intel/dev: Drop a bogus assert

In c24ba6cecbac, I copied over the CHV thread count code from i965 and
ANV into common code.  While at it, I added an assert that I thought
should be true based on the comments.  It turns out to be invalid.  Drop
it.

Fixes: c24ba6cecbac "intel/dev: Handle CHV CS thread weirdness in..."
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5072
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11918>

anv: Unbreak wide lines on HSW/BDW

I knew there was a reason the limit was at ~8 and not rounded to it.

Fixes: 004fcfe6985 ("anv: fix some multisample lines_wide CTS tests")

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11914>

anv: Zero initialize pipeline structs

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11890>

vulkan/util: Add and use vk_multialloc_zalloc variants

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11890>

vulkan/wsi/wayland: Initialise wl_shm pointer in VkImage

We don't explicitly calloc the wsi_wl_image to zero; anything that's
expected to be valid needs to be explicitly initialised.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5063
Fixes: 6b36f35734a9 ("vulkan/wsi/wl: add wl_shm support for lavapipe.")
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Tested-by: Jan Beich <jbeich@freebsd.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11909>

vulkan/wsi/wayland: Cosmetic alignment fix

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11909>

draw: handle resetting draw_id between instances.

If we have a multi draw then the draw_id needs to get reset
between each instance.

Fixes the rest of the VK CTS multidraw tests

Fixes: ce0b1f23c30e ("aux/draw: track increment_draw_id value from draw info")
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11899>

lavapipe: fix indexed multi draw draw_id increment

I cut-n-paste this badly from the non-indexed, and used the wrong
struct.

Fixes a bunch of multi-draw CTS

Fixes: cc8f7dbf2a1a ("lavapipe: fix multi-draw regression in shader parameters test")
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11899>

nir: Validate after deserialization.

It's a particularly relevant place for NIR bugs to occur, and if you make
a mistake in this code it gets caught in your debug build in something
like mesa/st's call to nir_split_var_copies() during finalization, which is
rather misleading.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11860>

docs: mark mesa 21.0 as done

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11326>

anv: fix some multisample lines_wide CTS tests

We can use a better algorithm from ICL and onward by setting a chicken
bit, but prior to that we need to resort to disabling rectangular lines.
Since we don't support strictLines anyway, this shouldn't be a major
issue.

Closes #2833
Fixes dEQP-VK.rasterization.interpolation_multisample_*_bit.*lines_wide

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11672>

anv: Don't advertise unsupported shader stages

We don't currently support mesh or ray tracing, so don't advertise their
shader stages as supported for subgroups.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11893>

radv: Use 128-sized vertex grouping for NGG shaders.

This matches what RadeonSI also does.
It seems to improve performance especially with NGG culling shaders.

Eg. in Doom Eternal this gives me +5ish fps.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11810>

mesa: (correctly) flush more in _mesa_make_current

The code being removed here requires that the old context have a
window-system drawable bound to the default framebuffer as a condition
of the context-release flush. I can find no justification for
restricting flushes to contexts like this, and GL 3.0 (etc) let you make
a context current without any drawable bound.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11423>

glx/drisw: Nerf PutImage when loaderPrivate == NULL

This means the drawable was already destroyed. This can happen during
diplay teardown, destroying the context will make it current first so it
can flush rendering and destroy textures and such, and if the drawable
is already destroyed then flushing to nowhere would crash.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11423>

asahi: Move assignment after null check.

Fix defect reported by Coverity Scan.

Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking so suggests that it may be null,
but it has already been dereferenced on all paths leading to the
check.

Fixes: dcd2d8ca500 ("asahi: Track more Gallium state")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11898>

Revert "ci:  disable Broadcom CI"

Our POE switch rebooted with factory settings. So devices weren't turning
on/off because of failing SMNP auth.

This reverts commit 87cdcf56455b78b0ca799adccb20c7d23d1b3a51.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11906>

util: document that workaround also fixes Riptale

The games use an idential executable.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11871>

v3dv: refactor meta copy/clear code

This change does the following:

- Moves image clearing code from v3dv_meta_copy.c to v3dv_meta_clear.c
  where it belongs.
- Renames v3dv_meta_copy.h to v3dv_meta_common.h since now it is required
  by the copy and clear code.
- Renames 'struct framebuffer_data' type.
- Merges v3dvx_meta_clear.c and v3dvx_meta_copy.c into v3dvx_meta_common.c.
- Prefixes all "public" meta functions with the 'meta' prefix.

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11902>

v3dv: remove more dead clearing code

We forgot to remove the v3dvx code for the clearing code fallbacks
we removed after we implemented layered clears.

Fixes: 5ba7f64b45 ('v3dv: remove fallback path for vkCmdClearAttachments')

Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11902>

ci:  disable Broadcom CI

The CI is down at the moment, disable it while we figure out
a solution.

Acked-by: Rob Clark <robclark@freedesktop.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11901>

radv: only init the TC-compat ZRANGE metadata for the depth aspect

With separate depth/stencil layouts, if the depth aspect is first
initialized and then cleared, the ZRANGE_PRECISION metadata might
be different than 0. Initializing it again for the stencil aspect
will overwrite the value.

Fixes rendering glitches with Scarlet Nexus on GFX8-9.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5052
Cc: 21.1 21.2 mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11883>

nv50/ir: Initialize Value member id in constructor.

Fix defect reported by Coverity Scan.

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

Suggested-by: Karol Herbst <kherbst@redhat.com>
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10401>

lima: handle fp16 vertex formats

12128fb1351e marked fp16 vertex formats supported, but they aren't actually handled
by lima_pipe_format_to_attrib_type(). Fix it by handling it there.
FP16 seems to be the only missing index which is 0x3.

Fixes: 12128fb1351e ("lima: add natively supported vertex buffer formats")

Cc: 21.1 mesa-stable
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11872>

venus: refactor vn_EndCommandBuffer

Add vn_cmd_submit as a wrapper to vn_instance_ring_submit.  It is also
designed such that it can be called from any of the vn_Cmd* functions.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11896>

lavapipe: fix multi-draw regression in shader parameters test

dEQP-VK.draw.shader_draw_parameters.base_instance.draw,Fail
dEQP-VK.draw.shader_draw_parameters.base_instance.draw_indexed,Fail

Fixes: 6b6dda56b5ff ("lavapipe: increment drawid for multidraws")
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11894>

pan/gen_pack: Use 1U for unpacking log2 to avoid undefined behaviour

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11869>