AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
190,947 Commits 1 Branch 0 Tags
738e202dceaa960be4e1379724289a481981cbeb
Commit Graph

9447 Commits

Author SHA1 Message Date
Timothy Arceri 0ac0fbc19e glsl: make glsl_to_nir() more generic 
Here we move anything that expects the IR to have already been linked
so that in a future patch we can use glsl_to_nir() to convert IR that
has only been compiled.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29761>
 2024-06-20 00:56:10 +00:00
Faith Ekstrand 22f1b04a99 nir/format_convert: Assert that UNORM formats are <= 16 bits 
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28793>
 2024-06-19 01:56:23 +00:00
Faith Ekstrand 5f5f4474f6 nir: Add a format unpack helper and tests 
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28793>
 2024-06-19 01:56:22 +00:00
Faith Ekstrand faf4c2edfe nir: Add a format pack helper and tests 
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28793>
 2024-06-19 01:56:22 +00:00
Faith Ekstrand 1ffb0c5af4 nir: Support 0 and 32 bits in some format conversion helpers 
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28793>
 2024-06-19 01:56:22 +00:00
Faith Ekstrand 34161d3fda nir: Move most of nir_format_convert to a C file 
There's no good reason for this to be header-only besides laziness on my
part when I first wrote a few "small" helpers.  Some of those are pretty
good sized and don't need to be inlined.

Keeping the original copyright since this is just moving code.

Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28793>
 2024-06-19 01:56:22 +00:00
Faith Ekstrand 9d3b144018 nir: Add a nir_intrinsic_use for unit tests 
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28793>
 2024-06-19 01:56:22 +00:00
Faith Ekstrand 5b9ac9a68f nir/format_convert: Use fmin/fmax to clamp R9G9B9E5 data 
As long as drivers implement an fmin/fmax that do the right thing with
NaN, there's no reason for the integer comparison.

Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28793>
 2024-06-19 01:56:22 +00:00
Faith Ekstrand 86aad90e2a nir/format_convert: Smash NaN to 0 in pack_r9g9b9e5() 
I have no idea why I flipped the order of these to checks vs. the C
code when I wrote the NIR helper.  We need to deal with NaN first or
else the fmin will smash NaN to MAX_RGB9E5 and it won't get handled as
NaN.

Fixes: 9981709d8f ("nir/format_convert: Add a function to pack RGB9_E5 formats")
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28793>
 2024-06-19 01:56:22 +00:00
Adam Jackson 10d21d4100 mesa: Enable EXT_shadow_samplers for GLES2 
I thought this was just the funny GLES spelling of the extn name, but
there's also some ESSL bits you need to add. Most of which you could
probably yoink from the old Unity glsl-optimizer (which itself yoinked
most of the GLSL compiler from Mesa):

https://github.com/aras-p/glsl-optimizer/commit/94a9b2959b885dba4e355aa344adb4d472b00f18

Signed-off-by: Adam Jackson <ajax@redhat.com>
Co-authored-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>

Acked-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6691>
 2024-06-18 14:40:33 +00:00
Timothy Arceri ef21df917f glsl: remove do_function_inlining() 
This no longer has any users. nir based inlining should be used for any
new code.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29519>
 2024-06-18 12:34:52 +00:00
Timothy Arceri f1ef6517e8 glsl: remove Par-linking from the standalone linker 
lima was the last user of this feature so lets remove it. This will
allow us to drop more soon to be unused glsl ir code once full nir
linker support lands.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29519>
 2024-06-18 12:34:52 +00:00
Timothy Arceri 7df492923a glsl: drop dump-builder support from standalone compiler 
The support is incomplete and largely untested, but more importantly
glsl ir is depreciated at this point. This feature was added to support
building additional passes but that shouldn't ever be needed from here
on.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29469>
 2024-06-18 08:12:45 +00:00
Marek Olšák 75777f1dc8 nir: add a NIR option flag nir_io_prefer_scalar_fs_inputs 
It's a NIR option because passing flags from radeonsi to the GLSL linker is
complicated.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29406>
 2024-06-17 23:48:35 +00:00
Marek Olšák 3622092614 glsl/linker: vectorize lowered IO 
Since we scalarize all IO for nir_opt_varyings, we should re-vectorize it.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29406>
 2024-06-17 23:48:35 +00:00
Marek Olšák 2514999c9c nir: add nir_opt_vectorize_io, vectorizing lowered IO 
Since nir_opt_varyings requires scalar IO and thus all drivers have to
scalarize it, this gives the option to re-vectorize IO after that.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29406>
 2024-06-17 23:48:35 +00:00
Marek Olšák 0058989357 nir/lower_io_to_scalar: don't create output stores that have no effect 
This fixes NIR validation errors that happen with certain shaders.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29406>
 2024-06-17 23:48:35 +00:00
Marek Olšák 756b4f907e nir/lower_io_to_scalar: add new_component temporary variable 
The next commit will use it. No change in behavior.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29406>
 2024-06-17 23:48:35 +00:00
Alyssa Rosenzweig ae3af4c73a nir: document restriction on load_smem_amd constantness 
This came up while reviewing
https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29398 ... Possibly
this intrinsic should be renamed to load_smem_constant_amd for consistency with
load_global_constant. But if we're not going to convey constantness in the
intrinsic name, let's at least document the restriction, because NIR's optimizer
relies on it.

(I didn't inspect every call site, but it looks like load_smem_amd is just used
for descriptor loads so there's no bug to fix.)

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29743>
 2024-06-17 21:17:09 +00:00
Alyssa Rosenzweig 15257b65c6 treewide: use nir_metadata_control_flow 
Via Coccinelle patch:

    @@
    @@

    -nir_metadata_block_index | nir_metadata_dominance
    +nir_metadata_control_flow

...plus some manual fixups for call sites missed by coccinelle.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Acked-by: Karol Herbst <kherbst@redhat.com>
Acked-by: Juan A. Suarez Romero <jasuarez@igalia.com> [broadcom]
Acked-by: Vasily Khoruzhick <anarsoul@gmail.com> [lima]
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29745>
 2024-06-17 16:28:14 -04:00
Alyssa Rosenzweig 90b6dba772 nir: add nir_metadata_control_flow 
Most passes want to preserve this specific combination of metadata, so let's add
an alias for the combination. The alias communicates that the control flow graph
is preserved, rather than a particular statement about e.g. dominance
preservation.

You don't need to understand dominance to write a simple
nir_shader_instructions_pass. And since you were going to cargo cult the
metadata anyway, this way you'll cargo cult a version you're more likely to
understand.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Acked-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29745>
 2024-06-17 16:28:11 -04:00
Daniel Schürmann cfa5beeeab spirv: workaround for tests assuming that OpKill terminates invocations or loops 
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27617>
 2024-06-17 19:37:16 +00:00
Daniel Schürmann 7af16e9f1e nir/shader_info: remove uses_demote 
This flag is mostly redundant with uses_discard and was only
introduced to implement demote with LLVM when it didn't have
that intrinsic.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27617>
 2024-06-17 19:37:16 +00:00
Daniel Schürmann 9b1a748b5e nir: remove nir_intrinsic_discard 
The semantics of discard differ between GLSL and HLSL and
their various implementations. Subsequently, numerous application
bugs occurred and SPV_EXT_demote_to_helper_invocation was written
in order to clarify the behavior. In NIR, we now have 3 different
intrinsics for 2 things, and while demote and terminate have clear
semantics, discard still doesn't and can mean either of the two.

This patch entirely removes nir_intrinsic_discard and
nir_intrinsic_discard_if and replaces all occurences either with
nir_intrinsic_terminate{_if} or nir_intrinsic_demote{_if} in the
case that the NIR option 'discard_is_demote' is being set.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27617>
 2024-06-17 19:37:16 +00:00
Daniel Schürmann 073e69c7dc nir/opt_peephole_select: handle nir_terminate{_if} 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27617>
 2024-06-17 19:37:15 +00:00
Daniel Schürmann f3d8bd18dd nir: introduce discard_is_demote compiler option 
This new option indicates that the driver emits the same
code for nir_intrinsic_discard and nir_intrinsic_demote.
Otherwise, it is assumed that discard is implemented as
terminate.

spirv_to_nir uses this option in order to directly emit
nir_demote in case of OpKill.

RADV GFX11:
Totals from 3965 (4.99% of 79439) affected shaders:
MaxWaves: 119418 -> 119424 (+0.01%); split: +0.03%, -0.03%
Instrs: 1608753 -> 1620830 (+0.75%); split: -0.18%, +0.93%
CodeSize: 8759152 -> 8785152 (+0.30%); split: -0.18%, +0.48%
VGPRs: 152292 -> 149232 (-2.01%); split: -2.37%, +0.36%
Latency: 9162314 -> 10033923 (+9.51%); split: -0.46%, +9.97%
InvThroughput: 1491656 -> 1493408 (+0.12%); split: -0.10%, +0.22%
VClause: 21424 -> 21452 (+0.13%); split: -0.31%, +0.44%
SClause: 53598 -> 55871 (+4.24%); split: -2.15%, +6.39%
Copies: 90553 -> 90462 (-0.10%); split: -2.91%, +2.81%
Branches: 16283 -> 16311 (+0.17%)
PreSGPRs: 113993 -> 113254 (-0.65%); split: -1.84%, +1.19%
PreVGPRs: 110951 -> 108914 (-1.84%); split: -2.08%, +0.24%
VALU: 963192 -> 963167 (-0.00%); split: -0.01%, +0.01%
SALU: 87926 -> 90795 (+3.26%); split: -2.92%, +6.18%
VMEM: 25937 -> 25936 (-0.00%)
SMEM: 110012 -> 109799 (-0.19%); split: -0.20%, +0.01%

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27617>
 2024-06-17 19:37:15 +00:00
Daniel Schürmann e0ab1ed14e spirv: make gl_HelperInvocation volatile if demote is being used 
Non-volatile gl_HelperInvocation after demote is undefined.
In order to avoid application bugs, make it volatile if we use demote.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Faith Ekstrand <faith.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27617>
 2024-06-17 19:37:15 +00:00
Alyssa Rosenzweig 574c5c70de nir/lower_robust_access: handle MSAA images 
We need to check the sample too. fixes on Honeykrisp with MSAA storage images:

dEQP-VK.robustness.robustness2.bind.notemplate.r32i.dontunroll.nonvolatile.storage_image.fmt_qual.img.samples_4.2d_array.comp

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29741>
 2024-06-17 15:28:15 +00:00
Karol Herbst 358e09f9ff nir: add global_atomic_2x32 variants to nir_get_io_offset_src_number 
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29711>
 2024-06-17 10:07:56 +00:00
Karol Herbst d2d966a3c2 nir_lower_mem_access_bit_sizes: support unaligned store_scratch 
This can be trivially be added as it doesn't even need atomics.

Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29711>
 2024-06-17 10:07:56 +00:00
Faith Ekstrand 7e3d157bee nak,nir: Drop r2ur_nv in favor of as_uniform 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29737>
 2024-06-15 06:14:27 +00:00
Job Noorman 0c1bb92690 nir/opt_offsets: add load/store_ssbo_ir3 
Signed-off-by: Job Noorman <jnoorman@igalia.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28664>
 2024-06-14 17:12:59 +00:00
Job Noorman 609a56d170 nir/opt_offsets: add option to allow offset wrapping 
On some ISAs (e.g., ir3) the offset calculation wraps the same way as
normal unsigned addition so potentially wrapping operations do not have
to be ignored.

Signed-off-by: Job Noorman <jnoorman@igalia.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28664>
 2024-06-14 17:12:59 +00:00
Job Noorman 518c93768b nir/opt_offsets: add callback for max base offset 
To support cases where different instructions may be used for the same
storage type. For example, to load from an SSBO on ir3, either ldib (max
offset 127) or isam.v (max offset 255) can be used.

Signed-off-by: Job Noorman <jnoorman@igalia.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28664>
 2024-06-14 17:12:59 +00:00
Job Noorman d3f8de791d ir3: lower SSBO access imm offsets 
Add the BASE index to the load/store_ssbo_ir3 intrinsic to store an
immediate offset. This offset is encoded in the corresponding fields of
isam.v/ldib.b/stib.b.

One extra optimization is implemented: whenever the regular offset is
also a constant, the total offset (regular plus immediate) is aligned
down to a multiple of the max immediate offset and this is used as the
regular offset while the immediate is set to the remainder. This ensures
that the register used for the regular offset can often be reused among
multiple contiguous accesses.

Signed-off-by: Job Noorman <jnoorman@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28664>
 2024-06-14 17:12:59 +00:00
Qiang Yu b1d0ecd00d glsl: respect GL_EXT_shader_image_load_formatted when image is embedded in a struct 
Fix compilation failure when image is embedded in struct when
GL_EXT_shader_image_load_formatted is enabled:

  struct GpuPointShadow {
      image2D RayTracedShadowMapImage;
  };

  layout(std140, binding = 2) uniform ShadowsUBO {
      GpuPointShadow PointShadows[1];
  } shadowsUBO;

Compile log:
  error: image not qualified with `writeonly' must have a format layout qualifier

Fixes: 082d180a22 ("mesa, glsl: add support for EXT_shader_image_load_formatted")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29693>
 2024-06-14 02:21:59 +00:00
Faith Ekstrand e05cb967e7 nir: Add nir_foreach_block_in_cf_node_safe() iterators 
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:46 +00:00
Faith Ekstrand b107240474 nir: Add some new _nv intrinsics 
The ldc_nv and ldcx_nv intrinsics correspond to the index and bindless
forms of NVIDIA's LDC instruction, respectively.  ldc_nv is pretty much
load_ubo without some of the unnecessary constant bits while ldcx_nv
takes a 64-bit bindless handle instead of an index.  The other two give
us a little control over register allocation at the NIR level to ensure
that LDCX handles are placed in uniform registers.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:45 +00:00
Faith Ekstrand 290cbf413c nir/print: Improve divergence information 
Reviewed-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29591>
 2024-06-13 20:43:44 +00:00
Timothy Arceri 4c3d1a09de nir: add additional opt_loop_merge() test of deref handling 
Here we test the rematerialization of the deref produces valid nir
when both the deref and array index value are moved to the else branch of
the first terminator during the merge.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29686>
 2024-06-13 15:00:35 +00:00
Timothy Arceri abb51f449d nir: test opt_loop_merge_terminators() skips unhandled loops 
This test makes sure the merge if pass skips loops with trainling phis
as those are not handled by the pass.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29686>
 2024-06-13 15:00:35 +00:00
Timothy Arceri b26ef8f153 nir: correctly track current loop in nir_opt_loop() 
We were not restoring an outer loop as the current loop after we had
finished processing a nested loop.

Fixes: 9995f336e6 ("nir: add merge loop terminators optimisation")

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29686>
 2024-06-13 15:00:35 +00:00
Timothy Arceri 3d2a821198 nir: add test for opt_loop_merge_terminators 
Makes sure we correctly rematerialize derefs moved during the merge.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29686>
 2024-06-13 15:00:35 +00:00
Rhys Perry 92af96e0b3 nir/opt_loop: fix formatting 
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29686>
 2024-06-13 15:00:35 +00:00
Rhys Perry cb51a93c1e nir/opt_loop: rematerialize derefs instead of creating phis 
Fixes NIR validation of hogwarts_legacy/00ac08423ad6e422.

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Fixes: 9995f336e6 ("nir: add merge loop terminators optimisation")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29686>
 2024-06-13 15:00:35 +00:00
Alyssa Rosenzweig f1144aa56f nir/builtin_builder: factor out nir_build_texture_query 
useful for other queries too.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29614>
 2024-06-11 13:10:22 +00:00
Timothy Arceri 9995f336e6 nir: add merge loop terminators optimisation 
Merge two consecutive basic terminators.

Acked-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28998>
 2024-06-11 01:42:23 +00:00
Timothy Arceri e25da8d8d7 nir: support more loop unrolling for logical operators 
Here we support finding loop count when the termination condition
is a logical or.

Acked-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28998>
 2024-06-11 01:42:23 +00:00
Timothy Arceri 987cf4b47d nir: more aggressively remove in loop during partial unroll 
Acked-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28998>
 2024-06-11 01:42:23 +00:00
Timothy Arceri 9702570994 nir: clarify and update loop conditional instruction 
This value is intended to be used to remove out of bounds array
access when unrolling loops so it should contain the comparison
that contains the the induction variable not the overall
condition of the loop terminator. So here we update the instruction
when dealing with iand/ior loop terminator conditions.

Acked-by: Pavel Ondračka <pavel.ondracka@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/28998>
 2024-06-11 01:42:23 +00:00