nir/algebraic: i2f(f2i()) -> trunc()

total instructions in shared programs: 12840968 -> 12840784 (<.01%)
instructions in affected programs: 17886 -> 17702 (-1.03%)
helped: 77
HURT: 0

total cycles in shared programs: 302508917 -> 302505592 (<.01%)
cycles in affected programs: 249964 -> 246639 (-1.33%)
helped: 70
HURT: 7

Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/948>

i965: Reuse the new core glsl_count_dword_slots().

The only difference I could see was treating interfaces like structs.
Maintain that case.

Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3297>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3297>

mesa/st: Move the dword slot counting function to glsl_types as well.

To implement NIR-to-TGSI, we need to be able to get the size of the
uniform variable for the TGSI declaration, not just the
.driver_location.  With its location in mesa/st, drivers couldn't link
to it from nir-to-tgsi.

This feels like a common enough function to want, so let's share it in
the core compiler.

Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3297>

mesa/prog: Reuse count_vec4_slots() from ir_to_mesa.

Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3297>

mesa/st: Move the vec4 type size function into core GLSL types.

The only bit that gallium varied on was handling of bindless.  We can
retain previous behavior for count_attribute_slots() by passing in
"true" (though I suspect this is just giving a silly answer to a silly
question), and delete our recursive function from mesa/st.

Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3297>

mesa/st: Deduplicate the NIR uniform lowering code.

Just a little refactor as I go looking at the type size functions.

Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3297>

radeonsi: move PS LLVM code into si_shader_llvm_ps.c

This is an attempt to clean up si_shader.c.

v2: don't move code that is not specific to LLVM

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com> (v1)

radeonsi: remove always constant ballot_mask_bits from si_llvm_context_init

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

radeonsi: fold si_create_function into si_llvm_create_func

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

radeonsi: rename si_shader_create -> si_create_shader_variant for clarity

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

radeonsi: rename si_compile_tgsi_main -> si_build_main_function

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

radeonsi: clean up si_shader_info

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

radeonsi: merge si_tessctrl_info into si_shader_info

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

radeonsi: fork tgsi_shader_info and tgsi_tessctrl_info

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

radeonsi: rename si_shader_info -> si_shader_binary_info

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

radeonsi: remove TGSI from comments

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

radeonsi: rename DBG_NO_TGSI -> DBG_NO_NIR

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

radeonsi: don't adjust depth and stencil PS output locations

this was for compatibility with TGSI

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

nir: Add missing nir_var_mem_global to various passes

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3322>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3322>

spirv: Handle PhysicalStorageBuffer in memory barriers

PhysicalStorageBuffer is lowered to nir_var_mem_global, and
SPIR-V 1.5rev1 in section "3.25. Memory Semantics <id>" says

    UniformMemory

    Apply the memory-ordering constraints to StorageBuffer,
    PhysicalStorageBuffer, or Uniform Storage Class memory.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3322>

spirv: Drop EXT for PhysicalStorageBuffer symbols

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3322>

aco: Flip s_cbranch / s_cselect to optimize out an s_not if possible.

When possible, get rid of an s_not when all it does is invert the SCC,
and its successor s_cbranch / s_cselect can be inverted instead.

Also modify some parts of instruction_selection to take advantage of
this feature.

Example:
s2: %3900,  s1: %3899:scc = s_andn2_b64 %0:exec, %406
s2: %3902 = s_cselect_b64 -1, 0, %3900:scc
s2: %407,  s1: %3903:scc = s_not_b64 %3902
s2: %3906,  s1: %3905:scc = s_and_b64 %407, %0:exec
p_cbranch_z %3905:scc
Can now be optimized to:
s2: %3900,  s1: %3899:scc = s_andn2_b64 %0:exec, %406
p_cbranch_nz %3900:scc

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>

aco: Optimize out s_and with exec, when used on uniform bitwise values.

Previously all booleans needed an s_and with exec when they were turned
into a scalar condition. However, this is not needed for uniform booleans.

v2 by Daniel Schürmann:
- Make the code more readable
v3 by Timur Kristóf:
- Fix regressions, make it work in wave32 mode

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>

aco: Don't skip combine_instruction when definitions[1] is used.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>

aco: Allow optimizing vote_all and nir_op_iand.

By adding an extra instruction, we can replace the operands of
the s_cselect_b64, which allows it to get picked up by the
optimizer when it looks for uniform booleans.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>

aco: Implement 64-bit constant propagation.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>

panfrost: Fix linear depth textures

As pointed out by Boris, what we were calling PAN_LINEAR depth textures
was in fact u-interleaved tiled (!), but we never noticed since we
flipped the flag used for sampling, leading to all sorts of fun bugs
when attempting to directly acess depth textures from the CPU. Which
begs the question -- if what we called LINEAR was tiled, how do we
actually render linear depth textures? It turns out the flags for AFBC
form a mali_block_format 2-bit code just like their render-target
counterparts, so we can render to any of the above.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reported-by: Boris Brezillon <boris.brezillon@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3393>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3393>

vulkan/wsi: Add a driconf option to force WSI to advertise BGRA8_UNORM first

The Aztec Ruins benchmark just grabs the first format in the list and
SRGB causes it to render washed out.  With this workaround, it renders
the same as OpenGL.

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3350>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3350>

intel/fs: Only use SLM fence in compute shaders

Fixes: b390ff35170 ("intel/fs: Add support for SLM fence in Gen11")
Fixes: e142061399c ("intel/fs: Implement scoped_memory_barrier")
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>

radeonsi: actually enable VBOs in user SGPRs

Fixes: 363b4027fcb - radeonsi: put up to 5 VBO descriptors into user SGPRs
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>

radeonsi: fix assertion and other failures in si_emit_graphics_shader_pointers

The assertion was failing.

Fixes: 363b4027fcb - radeonsi: put up to 5 VBO descriptors into user SGPRs
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>

nir/algebraic: a & ~(a >> 31) -> imax(a, 0)

Found in some Doom shaders

Totals from affected shaders:
SGPRS: 30056 -> 30064 (0.03 %)
VGPRS: 28024 -> 28024 (0.00 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 4278648 -> 4270852 (-0.18 %) bytes
Max Waves: 1476 -> 1476 (0.00 %)
Instructions: 835287 -> 833338 (-0.23 %)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3089>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3089>

etnaviv: Fix assert when try to accumulate an invalid fd

Check if it is a valid fd before merging it to the context's fd.

Signed-off-by: Marco Felsch <m.felsch@pengutronix.de>
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3381>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3381>

pan/midgard: Fix midgard_compile.h includes

We now use enum mali_format which is defined in panfrost-job.h

Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3243>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3243>

anv: only use VkSamplerCreateInfo::compareOp if enabled

The spec says nothing about the validity of the compareOp field when
compareEnable is false.

v2: use vulkan enum to pick default value (Caio)

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: <mesa-stable@lists.freedesktop.org>
Closes: https://gitlab.freedesktop.org/mesa/mesa/issues/2350
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3387>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3387>

nir/sink,nir/move: move/sink nir_op_mov

Can uncover opportunities to move other instructions. This can increase
register usage, but that doesn't seem to actually happen.

This optimizes a pattern of a load_per_vertex_input followed by several
moves and then a store_output in a different block.

v2: add nir_move_copies to make it optional

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net> (v1)
Acked-by: Rob Clark <robdclark@chromium.org>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2420>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2420>

nir/sink,nir/move: move/sink load_per_vertex_input

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Acked-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2420>

gitlab-ci: Consolidate container and build stages for LAVA

Use the normal build job to also prepare the artifacts for LAVA jobs.

For that, the build container needs to also build the test suites,
kernel, ramdisk, etc.

Then the build job will place the just-built Mesa in the ramdisk and the
test job can generate a LAVA job and point to those artifacts.

Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3295>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3295>

aco: add integer min/max to can_swap_operands

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2883>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2883>

aco: improve readfirstlane after uniform LDS loads

Totals from affected shaders:
SGPRS: 976 -> 968 (-0.82 %)
VGPRS: 580 -> 584 (0.69 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 106032 -> 103076 (-2.79 %) bytes
Max Waves: 237 -> 237 (0.00 %)
Instructions: 19452 -> 18740 (-3.66 %)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2883>

aco: replace extract_vector with copies

Helps a small number of small shaders with situations like this:
a = p_create_vector ...
b = p_extract_vector a, 3
and copy propagation can't be done

Totals from affected shaders:
SGPRS: 14304 -> 14416 (0.78 %)
VGPRS: 8716 -> 6592 (-24.37 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 184664 -> 176888 (-4.21 %) bytes
Max Waves: 6260 -> 6260 (0.00 %)
Instructions: 35561 -> 33617 (-5.47 %)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2883>

aco: allow input modifiers on v_cndmask_b32

Totals from affected shaders:
SGPRS: 594099 -> 594019 (-0.01 %)
VGPRS: 441016 -> 441124 (0.02 %)
Spilled SGPRs: 101 -> 101 (0.00 %)
Spilled VGPRs: 18 -> 18 (0.00 %)
Code Size: 30266652 -> 30125256 (-0.47 %) bytes
Max Waves: 67044 -> 67057 (0.02 %)
Instructions: 5753097 -> 5726607 (-0.46 %)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2883>

aco: don't move literal to reg when making an instruction VOP3 on GFX10

pipeline-db (Navi):
Totals from affected shaders:
SGPRS: 163398 -> 163398 (0.00 %)
VGPRS: 143820 -> 143820 (0.00 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 13065744 -> 13044308 (-0.16 %) bytes
Max Waves: 18921 -> 18921 (0.00 %)
Instructions: 2514644 -> 2509285 (-0.21 %)

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/2883>

aco: add min(-max(), ) and max(-min(), ) optimization

No pipeline-db changes.

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/2883>

aco: improve clamp optimization

Not sure why it checked the use count, it doesn't apply the constants.

pipeline-db (Navi):
Totals from affected shaders:
SGPRS: 269409 -> 269745 (0.12 %)
VGPRS: 238120 -> 238132 (0.01 %)
Spilled SGPRs: 305 -> 305 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 22908584 -> 22904672 (-0.02 %) bytes
Max Waves: 20217 -> 20217 (0.00 %)
Instructions: 4275312 -> 4263869 (-0.27 %)

pipeline-db (Vega):
Totals from affected shaders:
SGPRS: 155409 -> 155233 (-0.11 %)
VGPRS: 153072 -> 153072 (0.00 %)
Spilled SGPRs: 269 -> 269 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 14650824 -> 14650396 (-0.00 %) bytes
Max Waves: 9609 -> 9609 (0.00 %)
Instructions: 2762802 -> 2755517 (-0.26 %)

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/2883>

aco: fix clamp optimization

We can't do the optimization if there are neg/abs in-between.

No pipeline-db changes.

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/2883>

aco: improve creation of v_madmk_f32/v_madak_f32

Using needs_vop3 check was flawed because it would only combine the
literal if the first operand is the literal. If the second or third
operand is the literal, then needs_vop3 will be true and the literal will
not be combined.

pipeline-db (Navi):
Totals from affected shaders:
SGPRS: 782051 -> 782051 (0.00 %)
VGPRS: 630048 -> 630048 (0.00 %)
Spilled SGPRs: 195 -> 195 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 54743740 -> 54585548 (-0.29 %) bytes
Max Waves: 67340 -> 67340 (0.00 %)
Instructions: 10182030 -> 10182030 (0.00 %)

pipeline-db (Vega):
Totals from affected shaders:
SGPRS: 701990 -> 699590 (-0.34 %)
VGPRS: 566632 -> 566784 (0.03 %)
Spilled SGPRs: 218 -> 218 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 49173564 -> 49007856 (-0.34 %) bytes
Max Waves: 59650 -> 59612 (-0.06 %)
Instructions: 9315135 -> 9293330 (-0.23 %)

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/2883>

aco: take advantage of GFX10's constant bus limit and VOP3 literals

pipeline-db (Navi):
Totals from affected shaders:
SGPRS: 2397159 -> 2392494 (-0.19 %)
VGPRS: 1756036 -> 1753920 (-0.12 %)
Spilled SGPRs: 461 -> 470 (1.95 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 110287304 -> 109946304 (-0.31 %) bytes
Max Waves: 318341 -> 318475 (0.04 %)
Instructions: 21019327 -> 20533618 (-2.31 %)

pipeline-db (Vega):
Totals from affected shaders:
SGPRS: 0 -> 0 (0.00 %)
VGPRS: 0 -> 0 (0.00 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 0 -> 0 (0.00 %) bytes
Max Waves: 0 -> 0 (0.00 %)
Instructions: 0 -> 0 (0.00 %)

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/2883>

aco: allow an extra SGPR with multiple uses to be applied to VOP3

This is in a separate patch from the apply_sgprs() rewrite so that the
rewrite can be more easily tested.

pipeline-db (Navi):
Totals from affected shaders:
SGPRS: 3056 -> 3056 (0.00 %)
VGPRS: 1632 -> 1632 (0.00 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 156468 -> 156304 (-0.10 %) bytes
Max Waves: 288 -> 288 (0.00 %)
Instructions: 29510 -> 29469 (-0.14 %)

pipeline-db (Vega):
Totals from affected shaders:
SGPRS: 2984 -> 2984 (0.00 %)
VGPRS: 1616 -> 1616 (0.00 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 156132 -> 155968 (-0.11 %) bytes
Max Waves: 289 -> 289 (0.00 %)
Instructions: 29426 -> 29385 (-0.14 %)

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/2883>

aco: allow applying two sgprs to an instruction

We could create VALU instructions which read two sgprs, but only if isel
created an instruction which already read one of them.

This change is in a separate patch from the apply_sgprs() rewrite so that
it can be tested if the rewrite affected anything.

pipeline-db (Navi):
Totals from affected shaders:
SGPRS: 216 -> 216 (0.00 %)
VGPRS: 64 -> 64 (0.00 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 1756 -> 1708 (-2.73 %) bytes
Max Waves: 120 -> 120 (0.00 %)
Instructions: 312 -> 300 (-3.85 %)

pipeline-db (Vega):
Totals from affected shaders:
SGPRS: 216 -> 216 (0.00 %)
VGPRS: 64 -> 64 (0.00 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 1784 -> 1736 (-2.69 %) bytes
Max Waves: 120 -> 120 (0.00 %)
Instructions: 319 -> 307 (-3.76 %)

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/2883>

aco: follow through temporary when merging tests into constant comparisons

This can happen with v_mov_b32(s_mov_b32(literal))

pipeline-db (Navi):
Totals from affected shaders:
SGPRS: 632 -> 632 (0.00 %)
VGPRS: 492 -> 492 (0.00 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 77488 -> 76928 (-0.72 %) bytes
Max Waves: 67 -> 67 (0.00 %)
Instructions: 14426 -> 14332 (-0.65 %)

pipeline-db (Vega):
Totals from affected shaders:
SGPRS: 632 -> 632 (0.00 %)
VGPRS: 492 -> 492 (0.00 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 77512 -> 76952 (-0.72 %) bytes
Max Waves: 67 -> 67 (0.00 %)
Instructions: 14432 -> 14338 (-0.65 %)

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/2883>

aco: be more careful with literals in combine_salu_{n2,lshl_add}

No pipeline-db changes.

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/2883>

aco: add check_vop3_operands()

This will be useful when taking advantage of GFX10 features.

No pipeline-db changes.

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/2883>

aco: rewrite apply_sgprs()

This will make it easier to apply two different sgprs (for GFX10) or apply
the same sgpr twice (just remove the break).

No pipeline-db changes.

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/2883>

aco: rewrite literal combining

Should make taking advantage of GFX10's increased constant bus limit and
VOP3 literals easier.

No pipeline-db changes

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/2883>

aco: improve can_use_VOP3()

No pipeline-db changes

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/2883>

aco: combine two sgprs into a VALU if they're the same

This was supposed to be done before but it wasn't done correctly and
everywhere.

pipeline-db (Navi):
Totals from affected shaders:
SGPRS: 784680 -> 786128 (0.18 %)
VGPRS: 574012 -> 573892 (-0.02 %)
Spilled SGPRs: 461 -> 461 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 45477088 -> 45478172 (0.00 %) bytes
Max Waves: 81294 -> 81277 (-0.02 %)
Instructions: 8657970 -> 8622483 (-0.41 %)

pipeline-db (Vega):
Totals from affected shaders:
SGPRS: 780664 -> 782072 (0.18 %)
VGPRS: 573880 -> 573760 (-0.02 %)
Spilled SGPRs: 629 -> 629 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 45445244 -> 45448340 (0.01 %) bytes
Max Waves: 81178 -> 81161 (-0.02 %)
Instructions: 8649902 -> 8614918 (-0.40 %)

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/2883>

aco: apply literals to split mads

Removing the return is also needed to apply literals to mads (which can be
done on GFX10).

pipeline-db (Navi):
Totals from affected shaders:
SGPRS: 368787 -> 367555 (-0.33 %)
VGPRS: 312436 -> 312448 (0.00 %)
Spilled SGPRs: 461 -> 461 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 26113388 -> 26098260 (-0.06 %) bytes
Max Waves: 35982 -> 35982 (0.00 %)
Instructions: 5038670 -> 5028941 (-0.19 %)

pipeline-db (Vega):
Totals from affected shaders:
SGPRS: 369843 -> 368659 (-0.32 %)
VGPRS: 317224 -> 317196 (-0.01 %)
Spilled SGPRs: 629 -> 629 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 26310540 -> 26295156 (-0.06 %) bytes
Max Waves: 36324 -> 36326 (0.01 %)
Instructions: 5073957 -> 5064164 (-0.19 %)

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/2883>

aco: update IR validator

GFX10 increased the constant bus limit and allowed literals on VOP3

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/2883>

nir/lower_gs_intrinsics: add option for per-stream counts

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2422>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2422>

nir/divergence: handle load_primitive_id in GS

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2323>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2323>

mesa/st: use float literals

This removes a warning on MSVC.

Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Brian Paul <brianp@vmware.com>

gallium: fix a warning

On some platforms (like Win64), unsigned long is 32-bit, so the first
cast doesn't do anything, and the compiler complains about an implicit
cast to a smaller type. So let's cast to an uintptr_t instead first,
as that's large enough on all platforms.

Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Brian Paul <brianp@vmware.com>

st/wgl: eliminate implicit cast warning

I get warnings on MSVC for these implicit casts. Let's use explicit
casts instead.

Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Brian Paul <brianp@vmware.com>

util: initialize float-array with float-literals

We currently initialize this float-array with double-literals. Some
compilers generate warnings for this, so let's switch these to
float-literals instead.

Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Brian Paul <brianp@vmware.com>

anv: Implement Gen12 workaround for non pipelined state

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3365>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3365>

iris: Implement Gen12 workaround for non pipelined state

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3365>

lima: add new findings to texture descriptor

Lower 8 bits of unknown_1_3 seems to be min_lod,
rest of 4 bits + miplevels are max_lod and min_mipfilter seems to be
lod bias. All are in fixed format with 4 bit integer and 4 bit fraction,
lod_bias also has sign bit.

Blob also seems to do some magic with lod_bias if min filter is nearest --
it adds 0.5 to lod_bias in this case. Same story when all filters are
nearest and mipmapping is enabled, but in this case it subtracts 1/16
from lod_bias.

Fixes 134 dEQP tests in dEQP-GLES2.functional.texture.*

Reviewed-by: Qiang Yu <yuq825@gmail.com>
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3359>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3359>

intel: Use similar brand strings to the Windows drivers

This updates our product name strings to match the ones reported
by the Windows driver, which is typically the marketing name.

We retain a platform abbreviation and GT level in parenthesis so that
we're able to distinguish similar parts more easily, helping us better
understand at a glance which GPU a bug reporter has.

Acked-by: Matt Turner <mattst88@gmail.com>
Acked-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3371>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3371>

iris: Simplify iris_get_renderer_string()

We use gen_get_device_name() instead of PCI ID list munging.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3371>

i965: Simplify brw_get_renderer_string()

This stops using driGetRendererString() in favor of a simple snprintf().
This should have the same functionality on 64-bit systems, but drops
a "x86/MMX/SSE2" suffix on 32-bit systems.  (People shouldn't be using
the GL_RENDERER string to check for CPU features...)

We also use gen_get_device_name() instead of PCI ID list munging.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3371>

Revert "nir: assert that nir_lower_tex runs after lowering derefs"

This reverts commit 4cda61f11e922fb5914ae73d22cc0c495abf0377 for now,
as it appears to break i965 CI (32,000+ failures).  Rob and I suspect
we need to do the equivalent of 1c6a2efa06e9bb5914f4557118930fc61065a467
on i965 - we are doing nir_lower_tex and brw_nir_lower_resources in the
wrong order and that's likely triggering this condition.  Once we fix
that, we should put this patch back.

zink: fixup initialization of operand_mask / num_extra_operands

This doesn't change behavior, but makes the code a bit easier to read.
Both values are zero, but I somehow swapped the logical meaning of them
when initializing.

mesa: Fix detection of invalidating both depth and stencil.

Fixes an extra 1024x1024x4 MSAA Z/S store on WebGL fishtank on cheza.

Reported-by: Dave Airlie <airlied@redhat.com>
Fixes: db2ae5112106 ("mesa: Skip partial InvalidateFramebuffer of packed depth/stencil.")
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3370>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3370>

mesa/st: lower samplers before nir_lower_tex

Fixes incorrect lowering of YUV samplers when there are non-yuv
samplers.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3368>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3368>

nir: assert that nir_lower_tex runs after lowering derefs

It isn't going to do the right thing, because texture_index/
sampler_index defaults to zero.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3368>

i965: support EXT_EGL_image_storage

i965 can support this.

Reviewed-by: Tapani Pälli <tapani.palli@intel.com>

i965: refactor intel_image_target_texture_2d

intel_image_target_texture_tex_storage can reuse much of this
code.

Reviewed-by: Tapani Pälli <tapani.palli@intel.com>

i965: track if image is created by a dmabuf

Will be used by EXT_EGL_image_storage later.

Reviewed-by: Tapani Pälli <tapani.palli@intel.com>

dri_util: add driImageFormatToSizedInternalGLFormat function

This is needed to implement the EXT_EGL_image_storage spec:

"If <target> is GL_TEXTURE_2D, then the resultant texture must have a
sized internal format which is colorspace and size compatible with the
dma-buf.  If the GL is unable to determine such a format, the error
INVALID_OPERATION is generated."

Reviewed-by: Tapani Pälli <tapani.palli@intel.com>

glapi / teximage: implement EGLImageTargetTexStorageEXT

Check various parts of the EXT_EGL_image_storage spec, and add a
new vfunc for drivers implementing it.

Reviewed-by: Tapani Pälli <tapani.palli@intel.com>

teximage: split out helper from EGLImageTargetTexture2DOES

The major differences between EXT_EGL_image_storage and
EGLImageTargetTexture2DOES are:

(1) The texture target is made immutable
(2) EXT_EGL_image_storage supports non-2D targets.

We can reuse EGLImageTargetTexture2D and FreeTextureImageBuffer
for (1) pretty easily.  For (2), let's just not support the
complicated targets.  Let's reuse aspects of the
EGLImageTargetTexture2DOES implementation.

Reviewed-by: Tapani Pälli <tapani.palli@intel.com>

anv: Memset array properties

This is probably better than possibly leaving those bytes uninitialized
even if the app will theoretically not use them.

Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3369>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3369>

anv: Don't over-advertise descriptor indexing features

We should only advertise sub-features if we advertise the extension.

Fixes: 6e230d7607f "anv: Implement VK_EXT_descriptor_indexing"
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Ivan Briano <ivan.briano@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3369>

intel/blorp: Fill out all the dwords of MI_ATOMIC

This makes us valgrind clean again.

Fixes: 9175c7058efb "intel/blorp: Make blorp update the clear color..."
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3366>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3366>

gitlab-ci: Upgrade kernel for LAVA jobs to v5.5-rc5

Some fixes got in that should prevent hangs in lima jobs.

Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3363>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3363>

aco: fix unconditional demote_to_helper

This patch fixes an out-of-bounds access on p_exit_early
and binds the exec register to the correct operand.

Fixes: 2ea9e59e8d976ec77800d2a20645087b96d1e241 ('aco: move s_andn2_b64 instructions out of the p_discard_if')
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3347>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3347>

radeonsi: don't enable VBOs in user SGPRs if compute-based culling can be used

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>

radeonsi: put up to 5 VBO descriptors into user SGPRs

gfx6-8: 1 VBO descriptor in user SGPRs
gfx9-10: 5 VBO descriptors in user SGPRs

We no longer pull up to 5 VBO descriptors from GTT when SDMA is disabled.

Totals from affected shaders:
SGPRS: 1110528 -> 1170528 (5.40 %)
VGPRS: 952896 -> 951936 (-0.10 %)
Spilled SGPRs: 83 -> 61 (-26.51 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size: 23766296 -> 22843920 (-3.88 %) bytes
LDS: 0 -> 0 (0.00 %) blocks
Max Waves: 179344 -> 179344 (0.00 %)
Wait states: 0 -> 0 (0.00 %)

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>

ac,radeonsi: increase the maximum number of shader args and return values

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>

radeonsi: simplify si_set_vertex_buffers

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>

radeonsi: don't allow draw calls with uninitialized VS inputs

These always hang, because vertex buffer descriptors are not set up.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>

radeonsi: add si_context::num_vertex_elements

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>

radeonsi: rename desc_list_byte_size -> vb_desc_list_alloc_size

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>

anv: set stencil layout for input attachments

If an input attachment has a stencil format, we need to set this.

v2: Fish out VkAttachmentReferenceStencilLayoutKHR from
    VkAttachmentReference2KHR::pNext (Jason)

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reported-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Fixes: c1c346f16673 ("anv: implement VK_KHR_separate_depth_stencil_layouts")
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2891>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/2891>

anv: Drop an unused variable

nir/lower_atomics_to_ssbo: Also lower barriers

This is more correct for a pass which is supposed to completely lower
away atomic counters.  It also lets us stop supporting atomic counter
barriers in most of the drivers.

Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3307>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3307>

nir: Rename nir_intrinsic_barrier to control_barrier

This is a more explicit name now that we don't want it to be doing any
memory barrier stuff for us.

Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3307>

intel/nir: Stop adding redundant barriers

Now that both GLSL and SPIR-V are adding shared and tcs_patch barriers
(as appropreate) prior to the nir_intrinsic_barrier, we don't need to do
it ourselves in the back-end.  This reverts commit
26e950a5de01564e3b5f2148ae994454ae5205fe.

Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3307>

nir/glsl: Emit memory barriers as part of barrier()

The GLSL barrier() intrinsic does an implicit shared memory barrier in
compute shaders and an implicit TCS patch output barrier in tessellation
control shaders.  We'd like NIR's barrier intrinsic to just be a control
flow barrier and not have memory implications.  To satisfy this, we need
to add an extra memory barrier in front of each nir_intrinsic_barrier.

Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3307>