pan/decode: Remove last_size

Fixes ../src/panfrost/pandecode/decode.c: In function ‘pandecode_jc’:
../src/panfrost/pandecode/decode.c:2859:14: warning: variable ‘last_size’ set but not used [-Wunused-but-set-variable]
 2859 |         bool last_size;

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3543>

panfrost: Don't use implicit mali_exception_status enum

Fixes ../src/panfrost/pandecode/public.h:53:33: warning: ‘enum mali_exception_access’ declared inside parameter list will not be visible outside of this definition or declaration

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3543>

radv: enable ACO support for GFX6

CTS should pass, as well as Crucible and the few number of Piglit tests.

List of game benchmarks tested:
- Dawn of War 3
- Serious Sam 2017
- Shadow of The Tomb Raider
- The Talos Principle
- Thrones of Britannia
- Total Warhammer 2
- Total War: Three Kingdoms

Note that F12017 hangs with or without ACO on GFX6 at the moment.

My whole pipelinedb (~30 games) doesn't trigger any compiler crashes.

Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2401
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3533>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3533>

aco: copy the literal offset of SMEM instructions to a temporary

GFX6 only supports up to 8-bit for the literal offset, so make sure
it's copied to a temporary SGPR before emitting a SMEM instruction.
The optimizer will propagate the literal offset if possible anyways.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3533>

aco: fix a hazard with v_interp_* and v_{read,readfirst}lane_* on GFX6

It's required to insert 1 wait state if the dst VGPR of any v_interp_*
is followed by a read with v_readfirstlane or v_readlane to fix GPU
hangs on GFX6. Note that v_writelane_* is apparently not affected.
This hazard isn't documented anywhere but AMD confirmed it.

This fixes a GPU hang with the texturemipmapgen Sascha demo on GFX6.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3533>

aco: fix a hardware bug for MRTZ exports on GFX6

GFX6 (except OLAND and HAINAN) has a bug that it only looks at
the X writemask component.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3533>

turnip: Implement vkCmdCopyQueryPoolResults for occlusion queries

Use CP_COND_EXEC and CP_COND_WRITE to conditionally copy the results
of a query to a buffer based off the query's availability.

Fixes: #2238
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3279>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3279>

turnip: Implement vkCmdResetQueryPool

Clears the available bit for each requested query on the GPU.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3279>

turnip: Implement vkGetQueryPoolResults for occlusion queries

Implements fetching the results of a query pool with the
VK_QUERY_RESULT_WAIT_BIT, VK_QUERY_RESULT_WITH_AVAILABILITY_BIT,
and VK_QUERY_RESULT_PARTIAL_BIT flags.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3279>

turnip: Update query availability on render pass end

Unlike on an immidiate-mode renderer, Turnip only renders tiles on
vkCmdEndRenderPass. As such, we need to track all queries that were
active in a given render pass and defer setting the available bit
on those queries until after all tiles have rendered.

This commit adds a draw_epilogue_cs to tu_cmd_buffer that is
executed as an IB at the end of tu_CmdEndRenderPass. We then emit
packets to this command stream that update the availability bit of a
given query in tu_CmdEndQuery.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3279>

turnip: Implement vkCmdEndQuery for occlusion queries

Mostly a translation of freedreno's implementation of glEndQuery for
GL_SAMPLES_PASSED query objects with a slight modification to set the
availability bit of the query bo (slot->available) if the query was
not ended inside a render pass.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3279>

turnip: Implement vkCmdBeginQuery for occlusion queries

Mostly a translation of freedreno's implementation of glBeginQuery for
GL_SAMPLES_PASSED query objects with special logic for handling tiled
render passes.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3279>

turnip: Implement vkCreateQueryPool for occlusion queries

General structure is inspired by anv's implementation in genX_query.c.
We define a packed struct that tracks sample count at the beginning of
the query and at the end; the result of the occlusion query is then
slot->end - slot->begin.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3279>

turnip: Update tu_query_pool with turnip-specific fields

tu_query_pool was forked from radv_query_pool, but we will need a
different set of fields to implement queries in turnip.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3279>

anv: Allow HiZ in read-only depth layouts

This improves the performance of Aztec Ruins by 5% on ICL.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2605>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2605>

anv: Add a usage parameter to anv_layout_to_aux_usage

Most places we actually know the usage and can provide it.  There are
two exceptions to this:

 1. We pass 0 into get_blorp_surf_for_anv_image when we use
    ANV_IMAGE_LAYOUT_EXPLICIT_AUX because anv_layout_to_aux_usage is
    never actually called so it doesn't matter.

 2. We pass 0 into anv_layout_to_aux_usage in transition_color_buffer.
    However, the coming commits which will begin using the usage
    parameter only care about depth.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2605>

anv: Use isl_aux_state for HiZ resolves

Rather than looking at the aux usage, we look at the isl_aux_state which
provides us with more detailed information.  This commit adds a couple
helpers to isl which let us quickly determine if we have valid depth/hiz
on the initial layout and if we need valid depth/hiz for the final
layout.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2605>

anv: Add a layout_to_aux_state helper

This new helper maps VkImageLayout enums to isl_aux_state enums which
are the hardware's concept of image layouts.  We can then use the aux
state to get the fast clear type and the aux usage.  This should yield
no functional change in driver behavior.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2605>

anv: Use TRANSFER_SRC_OPTIMAL for depth/stencil MSAA resolves

As of 52ad1712ed62, TRANSFER_SRC_OPTIMAL and SHADER_READ_ONLY_OPTIMAL
are now identical for depth buffers so there's no reason why we need to
use the "wrong" layout.  Technically, according to Vulkan, blits and
MSAA resolves are transfer ops so we should use the transfer layout now
that we can.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2605>

intel/blorp: resize src and dst surfaces separately

When copying to an RGB surface, we treat it as an R only one of three
times the width, which may end up being larger than the maximum size
supported by the hardware and so it hits the shrink path. This forced
both source and destination surfaces to be shrunk, even though it's not
necessary for the former, and may even hit some assertions in some
cases, such as the surface being compressed.

Fixes several tests under dEQP-VK.api.copy_and_blit.core.image_to_image.dimensions.*

Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3422>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3422>

aco: combine MRTZ (depth, stencil, sample mask) exports

Instead of emitting up to 3 for each different components (depth,
stencil and sample mask). This is needed to fix a hw bug on GFX6.

Totals from affected shaders:
SGPRS: 34728 -> 35056 (0.94 %)
VGPRS: 26440 -> 26476 (0.14 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size: 1346088 -> 1344180 (-0.14 %) bytes
LDS: 0 -> 0 (0.00 %) blocks
Max Waves: 3922 -> 3915 (-0.18 %)
Wait states: 0 -> 0 (0.00 %)

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3538>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3538>

aco/gfx10: Fix VcmpxExecWARHazard mitigation.

The SOPP instruction shouldn't have a definition, and its block
should be set to -1 in order to prevent it from being recognized
as a branch.
Also fix a typo in the readme.

Fixes: d6dfce02d074d615a3b88a3fccd8ee8c7e13c010
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3552>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3552>

aco: Transform uniform bitwise instructions to 32-bit if possible.

This allows removing superfluous s_cselect instructions
that come from turning booleans into 64-bit vector condition.

v2 by Daniel Schürmann:
- Make the code massively simpler
v3 by Timur Kristóf:
- Fix regressions, make it work in wave32 mode
- Eliminate extra moves by not always using the SCC definition
- Use s_absdiff_i32 for uniform XOR
- Skip the transformation for uncommon or invalid instructions

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3450>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3450>

etnaviv: update Android build files

etnaviv no longer builds on Android, fix this.

Signed-off-by: Martin Fuzzey <martin.fuzzey@flowbird.group>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3447>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3447>

aco: use nir_move_copies

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>

radv/aco: use ACO for GS copy shaders

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>

aco: implement GS copy shaders

v5: rebase on float_controls changes
v7: rebase after shader args MR and load/store vectorizer MR

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>

aco: remove needs_instance_id

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>

aco: explicitly mark end blocks for exports

For GS copy shaders, whether we want to do exports is conditional. By
explicitly marking the end blocks, we can mark an IF's then branch as an
export block and ensure that's where the assembler inserts null exports.

v6: only fixup exports in the end block, like before
v8: simplify some code

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>

radv/aco: allow ACO for GS

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>

aco: implement GS on GFX7-8

GS is the same on GFX6, but GFX6 isn't fully supported yet.

v4: fix regclass
v7: rebase after shader args MR

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>

radv/aco,aco: implement GS on GFX9+

v2: implement GFX10
v3: rebase
v7: rebase after shader args MR
v8: fix gs_vtx_offset usage on GFX9/GFX10
v8: use unreachable() instead of printing intrinsic
v8: rename output_state to ge_output_state
v8: fix formatting around nir_foreach_variable()
v8: rename some helpers in the scheduler
v8: rename p_memory_barrier_all to p_memory_barrier_common
v8: fix assertion comparing ctx.stage against vertex_geometry_gs

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>

aco: improve support for s_sendmsg

In particular, the messages needed for GS.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>

radv: move gs copy shader creation before other variants

ACO lowers output derefs which breaks the shader_info pass used by gs copy
shader creation.

v3: rebase

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2421>

aco: Make a better guess at which instructions need the VCC hint.

Previously, bool_to_vector_condition would always set the VCC hint
on its result. This commit improves it by having the optimizer set
the VCC hint only when the result really needs to be in the VCC.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3451>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3451>

gallium/swr: implementation of tessellation shaders compilation

TCS and TES shaders compilation mechanisms in SWR and state
management implementation.

Reviewed-by: Krzysztof Raszkowski <krzysztof.raszkowski@intel.com>
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Acked-by: Roland Scheidegger <sroland@vmware.com>
Acked-by: Dave Airlie <airlied@redhat.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3484>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3484>

radv: Allow DCC & TC-compat HTILE with VK_IMAGE_CREATE_EXTENDED_USAGE_BIT.

I misunderstood the flag when initially disabling. But this flag
only does something with mutable formats. If we have DCC and
mutable formats, the formats are close enough that the allowed
usage flags are not meaningfully different nor used during
allocation.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3424>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3424>

radv: Expose VK_KHR_swapchain_mutable_format.

Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2354
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3425>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3425>

freedreno: Document CP_INDIRECT_BUFFER_CHAIN

This will let us use batch chaining instead of growing batches on a5xx
and a6xx.

Reviewed-by: Rob Clark <robdclark@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3537>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3537>

freedreno: Document CP_UNK_A6XX_55

Reviewed-by: Rob Clark <robdclark@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3537>

freedreno: Document CP_COND_REG_EXEC more

The vulkan blob uses the RENDER_MODE mode to condition a blit on the
render mode in traces of a dEQP triangle test.

Reviewed-by: Rob Clark <robdclark@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3182>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3182>

ac/llvm: fix missing casts in ac_build_readlane()

Because ac_build_optimization_barrier() overwrites the original
src_type, we have to keep track of it before emitting that barrier.
Otherwise, wrong conversions are expected for pointers or small
bitsizes.

By doing this, we no longer need to do the cast dance in
ac_build_readlane_no_opt_barrier(), it was just necessary for
ac_build_optimization_barrier().

This fixes a bunch of crashes with subgroups related tests when
RADV_DEBUG=checkir is enabled, and it also fixes a compiler crash
with The Surge 2.

Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2395
Fixes: 0f45d4dc2b1 ("ac: add ac_build_readlane without optimization barrier")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3535>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3535>

anv/apply_pipeline_layout: Initialize the nir_builder before use

Fixes: #2410
Fixes: 3c754900b5f "nir: don't emit ishl in _nir_mul_imm() if backend doesn't support bitops"
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3548>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3548>

meson: Prefer 'iris' by default over 'i965'.

This changes the default driver for Intel Gen8-11 hardware to be
the newer 'iris' driver rather than the older 'i965' driver.  To
continue using i965, pass -Dprefer-iris=false when building.

Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3540>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3540>

drisw: Cache the depth of the X drawable

This is not always ->rgbBits, because there are cases where that could
be 32 but we're (legally) bound to a depth-24 pixmap. The important
thing to have match here is the actual server-side notion of depth.  You
can look this up (at modest expense) from the xlib visual info if the
fbconfig has a visual. But it might not, so if not, fetch it (at
slightly greater expense) from XGetGeometry. Do this at GLX drawable
creation so you don't have to do it on the SwapBuffers path.

Apparently this fixes glx/glx-swap-singlebuffer, which is unintentional
but quite pleasant.

Fixes: mesa/mesa#2291
Fixes: 90d58286 ("drisw: Fix and simplify drawable setup")
Reviewed-by: Eric Anholt <eric@anholt.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3305>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3305>

turnip: Convert the rest of tu_cmd_buffer.c over to the new pack macros.

There are only a couple of hard cases left using pkt4, where the register
number to write is computed.

Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3455>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3455>

turnip: Convert renderpass setup to the new register packing macros.

This gets a lot of the hard code converted over to the new macros,
resulting in (I feel) much more readable code with
LESS_SHOUTING_ABOUT_THE_REG().  I decided to consistently put the reg on
its own line, so that all the register names line up.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3455>

turnip: Port krh's packing macros from freedreno to tu.

This introduces some minor unpacking of the temporary fd_reg_pair structs
to code that previously was packing a whole register field.

In the pack wrapper in tu_cs.h, I added some explanatory docs, dropped the
relocs handling since we don't need it, and removed the extra regs[] in
the __ONE_REG() macro (which was causing gcc's optimizer to fall on its
face in my release build).

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3455>

freedreno: Fix OUT_REG() on address regs without a .bo supplied.

Sometimes you want to zero out an address by supplying a NULL BO, but
without this we would end up only emitting one dword.  Increases size of
fd6_gmem.o by .8%, though it's not clear to me why (no obvious terrible
codegen happening)

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3455>

freedreno: Add some missing a6xx address declarations.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3455>

relnotes: Add GL_INTEL_shader_integer_functions2 and VK_INTEL_shader_integer_functions2

Suggested-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>

lima: use imul for calculations with intrinsic src

It's source is supposed to be int, so we have to use integer
multiplication otherwise we'll get undefined result.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Erico Nunes <nunes.erico@gmail.com>
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3529>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3529>

nir: don't emit ishl in _nir_mul_imm() if backend doesn't support bitops

Otherwise we'll have to lower it later.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Erico Nunes <nunes.erico@gmail.com>
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3529>

pan/decode: Rotate trace files

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3525>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3525>

pan/decode: Dump to a file

The file name is taken from the environment variable
PANDECODE_DUMP_FILE, defaulting to pandecode.dump if it is not set.

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3525>

pan/decode: Support dumping to a file

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3525>

pan/bifrost: Support disassembling to a file

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3525>

pan/midgard: Support disassembling to a file

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3525>

pan/midgard: Fix a memory leak in the disassembler

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3525>

turnip: Fix execution of secondary cmd bufs with nothing in primary.

We want to finish off cmd emission in the primary CS and add its entry to
the IB, but regardless of whether there had been anything in the primary
CS to emit, we still need a reserved CS entry for the loop below.

Fixes crashes in dEQP-VK.binding_model.shader_access.secondary_cmd_buf.*
and many more in dEQP-VK.renderpass*

Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3524>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3524>

panfrost: Drop mysterious zero=0xFFFF field

It doesn't seem to affect any results and it's not at all clear if/why
the blob sometimes(?) sets it? So let's clean this up since this
solution isn't correct anyway.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3513>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3513>

pan/midgard: Fix bundle dynarray leak

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3496>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3496>

radeonsi: separate LLVM compilation from non-LLVM code

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3421>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3421>

radeonsi: change prototypes of si_is_multi_part_shader & si_is_merged_shader

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3421>

radeonsi: make si_compile_shader return bool

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3421>

radeonsi: make si_compile_llvm return bool

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3421>

radeonsi: move more LLVM functions into si_shader_llvm.c

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3421>

radeonsi: fold si_shader_context_set_ir into si_build_main_function

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3421>

radeonsi: move si_nir_build_llvm into si_shader_llvm.c

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3421>

radeonsi: minor cleanup in si_shader_internal.h

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3421>

radeonsi: move si_shader_llvm_build.c content into si_shader_llvm.c

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3421>

radeonsi: move VS shader code into si_shader_llvm_vs.c

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3421>

radeonsi: move non-LLVM code out of si_shader_llvm.c

There was also some redundant code in si_shader_nir.c

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3421>

radeonsi: use ctx->ac. for types and integer constants

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3421>

turnip: simplify tu_physical_device_get_format_properties

Fixes the "bad VkImageTiling" error when tiling is
VK_IMAGE_TILING_DRM_FORMAT_MODIFIER_EXT.

Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3485>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3485>

vulkan/wsi: remove unused image_get_modifier

Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Acked-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3485>

turnip: set linear tiling for scanout images

Fixes: 210e6887 "vulkan/wsi: Use the interface from the real modifiers extension"
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Acked-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3485>

turnip: hook up GetImageDrmFormatModifierPropertiesEXT

Fixes: 210e6887 "vulkan/wsi: Use the interface from the real modifiers extension"
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Acked-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3485>

freedreno/drm: Don't miscalculate timeout

The current code overflows (s * 1000000000) for s >= 5 but that is
e.g. used in msm_bo_cpu_prep.

Signed-off-by: Guido Günther <agx@sigxcpu.org>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3514>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3514>

turnip: Add support for fine derivatives.

This does appear to be the required instruction sequence (dsxpp_1 dst src;
dsxpp_1.p dst src) as dropping either instruction fails the testsuite.

Fixes dEQP-VK.glsl.derivate.*

Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3494>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3494>

freedreno/ir3: Plumb the ir3_shader_variant into legalize.

legalize is computing a lot of state that goes in the variant, let's just
store it directly instead of passing pointers around.  This leaves
max_bary in place, which is doing some surprising work (overwriting the
original total_in in some cases).

Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3494>

util/hash_table: update users to use new optimal integer hash functions

Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3475>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3475>

util/hash_table: added hash functions for integer types

A few hash_table users roll their own integer hash functions which
call _mesa_hash_data to perform the hashing which ultimately calls
into XXH32 with a dynamic key length. When using small keys with a
constant size the hash rate can be greatly improved by inlining
XXH32 and providing it a constant key length, see:
https://fastcompression.blogspot.com/2018/03/xxhash-for-small-keys-impressive-power.html

Additionally, this patch removes calls to _mesa_key_hash_string and
makes them instead call _mesa_has_string directly, matching the new
integer hash functions.

Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3475>

util/hash_table: replace _mesa_hash_data's fnv1a hash function with xxhash

For most key sizes, xxhash outperforms fnv1a's hash rate substantially (bug
2153). In particular, the V3D driver hashes multiple ~200 byte keys as part
of the shader cache lookup which can easily eat up 10-20% of the runtime on
the Raspberry Pi. Swapping over to xxhash drops this to ~1% of the runtime.

Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3475>

util: move fnv1a hash implementation into its own header

Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3475>

util: import xxhash

Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3475>

winsys/amdgpu: Close KMS handles for other DRM file descriptions

When a BO or amdgpu_screen_winsys is destroyed.

Should fix leaking such BOs in other DRM file descriptions.

v2:
* Pass the correct file descriptor to drmIoctl (Pierre-Eric
  Pelloux-Prayer)
* Use _mesa_hash_table_remove
v3:
* Close handles in amdgpu_winsys_unref as well
v4:
* Adapt to amdgpu_winsys::sws_list_lock.

Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2270
Fixes: 11a3679e3aba "winsys/amdgpu: Make KMS handles valid for original
                     DRM file descriptor"

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3202>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3202>

winsys/amdgpu: Re-use amdgpu_screen_winsys when possible

Namely, if os_same_file_description determined that the DRM file
descriptor references the same file description.

v2:
* Adapt to amdgpu_winsys::sws_list_lock.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3202>

util: Add os_same_file_description helper

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3202>

winsys/amdgpu: Only re-export KMS handles for different DRM FDs

When the amdgpu_screen_winsys uses the same FD as the amdgpu_winsys
(which is always the case for the first amdgpu_screen_winsys), we can
just use bo->u.real.kms_handle.

v2:
* Also only create the kms_handles hash table if the
  amdgpu_screen_winsys fd is different from the amdgpu_winsys one.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3202>

winsys/amdgpu: Keep track of retrieved KMS handles using hash tables

The assumption being that KMS handles are only retrieved for relatively
few BOs, so hash tables should be efficient both in terms of performance
and memory consumption.

We use the address of struct amdgpu_winsys_bo as the key and its
kms_handle field (the KMS handle valid for the DRM file descriptor
passed to amdgpu_device_initialize) as the hash value.

v2:
* Add comment above amdgpu_screen_winsys::kms_handles (Pierre-Eric
  Pelloux-Prayer)
v3:
* Protect kms_handles hash table with amdgpu_winsys::sws_list_lock
  mutex.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3202>

winsys/amdgpu: Keep a list of amdgpu_screen_winsyses in amdgpu_winsys

v2:
* Add dedicated mutex for the list.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3202>

aco: implement nir_op_f2i64/nir_op_f2u64 on GFX6

V_TRUNC_F64 and V_FLOOR_F64 needs to be lowered on GFX6.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>

aco: implement 64-bit nir_op_ffloor on GFX6

GFX6 doesn't have V_FLOOR_F64, it needs to be lowered. Loosely based
on the AMDGPU LLVM backend.

Introduce a new function because it will be useful for some other
64-bit operations.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>

aco: implement 64-bit nir_op_fround_even on GFX6

GFX6 doesn't have V_RNDNE_F64, it needs to be lowered. Loosely based
on the AMDGPU LLVM backend.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>

aco: implement 64-bit nir_op_fceil on GFX6

GFX6 doesn't have V_CEIL_F64, it needs to be lowered. Loosely based
on the AMDGPU LLVM backend.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>

aco: implement 64-bit nir_op_ftrunc on GFX6

GFX6 doesn't have V_TRUNC_F64, it needs to be lowered. Loosely based
on the AMDGPU LLVM backend.

Introduce a new function because it will be useful for some other
64-bit operations.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>

aco: implement nir_intrinsic_global_atomic_* on GFX6

GFX6 doesn't have FLAT instructions, use MUBUF instructions instead.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>

aco: implement nir_intrinsic_load_global on GFX6

GFX6 doesn't have FLAT instructions, use MUBUF instructions instead.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>

aco: implement nir_intrinsic_store_global on GFX6

GFX6 doesn't have FLAT instructions, use MUBUF instructions instead.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3477>