AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
147,167 Commits 1 Branch 0 Tags
5aa39253cbb700876569b2e400ae8eb66dfc2507
Commit Graph

662 Commits

Author SHA1 Message Date
Timur Kristóf 5aa39253cb nir: Rename nir_get_io_vertex_index_src and include per-primitive I/O. 
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13466>
 2021-11-16 07:46:55 +00:00
Daniel Schürmann ab21183b5d aco: implement D16 texture loads 
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13592>
 2021-11-15 18:28:20 +00:00
Timur Kristóf d80c7f3406 aco: Fix how p_is_helper interacts with optimizations. 
p_is_helper doesn't have any operands, so ACO's value numbering and/or
the pre-RA optimizer could incorrectly recognize two such instructions
as the same.

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

Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5570
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13577>
 2021-11-13 16:32:02 +01:00
Rhys Perry 11b533cb19 aco: optimize load_local_invocation_index with single-wave workgroups 
fossil-db (Sienna Cichlid):
Totals from 668 (0.52% of 128647) affected shaders:
CodeSize: 2201912 -> 2193336 (-0.39%)
Instrs: 403124 -> 402325 (-0.20%)
Latency: 4510940 -> 4510214 (-0.02%); split: -0.02%, +0.00%
InvThroughput: 681057 -> 679453 (-0.24%); split: -0.24%, +0.00%
VClause: 6470 -> 6467 (-0.05%)
SClause: 12759 -> 12755 (-0.03%)
Copies: 26348 -> 26218 (-0.49%); split: -0.50%, +0.00%
PreSGPRs: 26140 -> 26101 (-0.15%)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel-schuermann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13757>
 2021-11-12 18:59:51 +00:00
Bas Nieuwenhuizen 5db098c98b aco: Remove useless sub-expr. 
../src/amd/compiler/aco_instruction_selection.cpp:11915:83: error: expression result unused [-Werror,-Wunused-value]
      bld.vop2(aco_opcode::v_lshrrev_b32, fetch_index_def, div_info, instance_id).instr;

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13164>
 2021-11-09 20:51:14 +00:00
Samuel Pitoiset dc74285d32 aco: only load streamout buffers if streamout is enabled 
The streamout_config SGPR is used to determine if streamout is enabled.

This fixes a GPU hang with various transform feedback tests:
 - dEQP-GLES3.functional.transform_feedback.*
 - KHR-GL46.transform_feedback.api_errors_test
 - KHR-GL46.draw_indirect.basic-draw*-xfbPaused
 - KHR-GL46.geometry_shader.api.draw_calls_while_tf_is_paused

Cc: 21.3 mesa-stable
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/13514>
 2021-10-25 13:43:10 +00:00
Samuel Pitoiset 8ec6824335 radv,aco: decouple shader_info/options from radv_shader_args 
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13287>
 2021-10-22 07:10:40 +00:00
Samuel Pitoiset 996e81fb70 aco: fix loading 64-bit inputs with fragment shaders 
Fixes a bunch of 64-bit IO tests with piglit and Zink.

Cc: 21.3 mesa-stable
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/13454>
 2021-10-21 12:50:55 +02:00
Rhys Perry cd3f0683cd aco: simplify emit_stream_output() 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13438>
 2021-10-20 15:00:23 +01:00
Samuel Pitoiset 572a902566 aco: fix emitting stream outputs when the first component isn't zero 
Fixes a bunch of XFB piglit tests with Zink.

Cc: 21.3 mesa-stable
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/13437>
 2021-10-19 15:19:42 +00:00
Samuel Pitoiset e3cbb0eb6a aco: fix invalid IR generated for b2f64 when the dest is a VGPR 
Fixes few 64-bit piglit tests with Zink.

Cc: 21.3 mesa-stable
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/13435>
 2021-10-19 14:47:32 +00:00
Jason Ekstrand 956199e870 nir: s/nir_var_mem_image/nir_var_image/g 
We typically use nir_var_mem_* for stuff that has an explicit byte-based
memory layout.  Images are opaque.

Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13386>
 2021-10-16 03:47:10 +00:00
Jason Ekstrand 8ab40f517f aco: Split var_mem_image barrier handling from global/ssbo 
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4743>
 2021-10-15 14:58:56 +00:00
Jason Ekstrand 94b9f25883 aco: Add support for nir_var_mem_image 
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4743>
 2021-10-15 14:58:56 +00:00
Rhys Perry d45958f82e aco: implement VS input loads with prologs 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11717>
 2021-10-13 05:13:10 +00:00
Rhys Perry f4ea2d7887 aco: implement aco_compile_vs_prolog 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11717>
 2021-10-13 05:13:10 +00:00
Rhys Perry 6bfacb60ee radv: add and use radv_vs_input_alpha_adjust 
Unlike ac_fetch_format, this enum can be packed into two bits.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11717>
 2021-10-13 05:13:10 +00:00
Samuel Pitoiset 61a3e75242 radv,aco: remove nir_intrinsic_load_layer_id 
This was never used because the layer ID isn't a system value.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13243>
 2021-10-12 09:09:15 +02:00
Samuel Pitoiset d993b59941 aco: constify radv_shader_{info,args} 
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/13192>
 2021-10-08 14:14:50 +02:00
Samuel Pitoiset 64030e79c5 radv,aco: remap PS inputs when declaring shader arguments 
LLVM seems to require all inputs for PS.

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/13192>
 2021-10-08 14:14:09 +02:00
Samuel Pitoiset da50534ada aco: prevent using undeclared shader arguments for PS 
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/13192>
 2021-10-08 14:14:09 +02:00
Samuel Pitoiset a6298b1bc9 radv: remove unnecessary radv_shader_info:num_inline_push_consts 
This can be determined directly from the user SGPR loc.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13149>
 2021-10-08 11:37:19 +00:00
Samuel Pitoiset b172d68e69 radv: remove unnecessary radv_shader_info:base_inline_push_consts 
No need to duplicate it.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13149>
 2021-10-08 11:37:19 +00:00
Samuel Pitoiset 113ce21528 radv: determine if a shader uses indirect descriptors from the SGPR loc 
If the SGPR loc is declared, the shader needs indirect descriptor sets.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13149>
 2021-10-08 11:37:19 +00:00
Samuel Pitoiset 333472f0de aco: fix load_barycentric_at_{offset,sample} 
The linear case was missing.

Cc: 21.2 mesa-stable
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/13257>
 2021-10-08 08:51:24 +00:00
Rhys Perry 1fb63367a8 radv,aco: don't include FMASK in the storage descriptor 
We perform a FMASK expand when transitioning to GENERAL or TRANSFER_DST
layout, so storage images always have an identity FMASK.

radeonsi doesn't appear to expand the FMASK for read-only storage images,
so the sample index adjustment is still needed there.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12214>
 2021-10-07 15:36:39 +00:00
Rhys Perry 1e9f72ffbe radv,aco: use lower_to_fragment_fetch 
This simplifies ACO and will let us optimize the FMASK fetch (for example,
move it out of loops).

fossil-db (Sienna Cichlid):
Totals from 955 (0.64% of 150170) affected shaders:
CodeSize: 4722016 -> 4722952 (+0.02%); split: -0.02%, +0.04%
Instrs: 875619 -> 875760 (+0.02%); split: -0.02%, +0.04%
Latency: 14069089 -> 14071699 (+0.02%); split: -0.02%, +0.04%
InvThroughput: 2321419 -> 2321218 (-0.01%); split: -0.02%, +0.01%
VClause: 23080 -> 23081 (+0.00%)
SClause: 32426 -> 32019 (-1.26%); split: -1.88%, +0.62%
Copies: 42787 -> 42777 (-0.02%); split: -0.19%, +0.16%
Branches: 17900 -> 17902 (+0.01%); split: -0.04%, +0.06%
PreSGPRs: 43229 -> 41002 (-5.15%); split: -5.16%, +0.00%

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12214>
 2021-10-07 15:36:39 +00:00
Rhys Perry cfb816b2a5 aco: use correct dim for FMASK fetches 
I think it somehow worked fine previously, but this is more correct.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12214>
 2021-10-07 15:36:39 +00:00
Rhys Perry bf0cc05227 aco: return 0x76543210 for NULL FMASK fetch 
This can replace several v_cndmask_b32 with a single v_cndmask_b32, and
will be useful when we lower sample index adjustment in NIR.

fossil-db (Sienna Cichlid):
Totals from 955 (0.64% of 150170) affected shaders:
VGPRs: 53232 -> 53208 (-0.05%)
CodeSize: 4712548 -> 4722016 (+0.20%); split: -0.02%, +0.23%
MaxWaves: 19052 -> 19056 (+0.02%)
Instrs: 875891 -> 875619 (-0.03%); split: -0.04%, +0.00%
Latency: 14070164 -> 14069089 (-0.01%); split: -0.02%, +0.01%
InvThroughput: 2322982 -> 2321419 (-0.07%); split: -0.08%, +0.01%
VClause: 23070 -> 23080 (+0.04%); split: -0.00%, +0.05%
SClause: 32463 -> 32426 (-0.11%); split: -0.12%, +0.01%
Copies: 42840 -> 42787 (-0.12%); split: -0.19%, +0.07%
Branches: 17907 -> 17900 (-0.04%); split: -0.06%, +0.02%
PreSGPRs: 43585 -> 43229 (-0.82%)
PreVGPRs: 47676 -> 47625 (-0.11%)

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12214>
 2021-10-07 15:36:39 +00:00
Rhys Perry 225fe37c14 nir: add _amd suffix to fragment_mask_fetch and fragment_fetch texops 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12214>
 2021-10-07 15:36:39 +00:00
Timur Kristóf 6ca66808b5 aco: Fix determining whether any culling is enabled. 
Use 0xB instead of 0x00FFFFFF - this allows to jump over the culling
code when no actual culling is enabled but the ngg_cull_face_is_ccw
flag is set.

Fixes: 182d9b1e60
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13129>
 2021-10-01 15:46:12 +00:00
Timur Kristóf c13a8d20f7 aco: Fix small primitive precision. 
This is a mistake. It should use ngg_culling_settings
instead of ngg_gs_state.

Fixes: 182d9b1e60
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13129>
 2021-10-01 15:46:12 +00:00
Tony Wasserka 0812d440c7 aco: Use std::vector for the underlying container of std::stack 
By default, std::stack uses std::deque to allocate its elements, which has
poor cache efficiency. std::vector makes appending elements more expensive
(due to potential reallocations), but in the changed contexts the element
count should always be low anyway.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11925>
 2021-10-01 09:39:13 +00:00
Timur Kristóf 5c35040da1 aco: Don't write m0 register for LDS instructions on GFX9+. 
Fossil DB stats on Sienna Cichlid:

Totals from 2691 (2.09% of 128647) affected shaders:
VGPRs: 124392 -> 124376 (-0.01%)
CodeSize: 8192352 -> 8174620 (-0.22%); split: -0.22%, +0.00%
MaxWaves: 61516 -> 61524 (+0.01%)
Instrs: 1519774 -> 1514958 (-0.32%); split: -0.32%, +0.00%
Latency: 14767555 -> 14766145 (-0.01%); split: -0.01%, +0.00%
InvThroughput: 3394282 -> 3394173 (-0.00%); split: -0.01%, +0.00%
VClause: 31985 -> 32002 (+0.05%); split: -0.02%, +0.07%
SClause: 47581 -> 47539 (-0.09%); split: -0.14%, +0.05%
Copies: 127533 -> 122709 (-3.78%); split: -3.80%, +0.02%
Branches: 39395 -> 39390 (-0.01%)
PreSGPRs: 84389 -> 82702 (-2.00%)
PreVGPRs: 87520 -> 87519 (-0.00%)

Fossil DB stats on Sienna Cichlid with NGGC on:

Totals from 60930 (47.36% of 128647) affected shaders:
VGPRs: 2180712 -> 2180696 (-0.00%)
CodeSize: 169122736 -> 167474304 (-0.97%); split: -0.97%, +0.00%
MaxWaves: 1703698 -> 1703706 (+0.00%)
Instrs: 32301234 -> 31888743 (-1.28%); split: -1.28%, +0.00%
Latency: 152526083 -> 152367301 (-0.10%); split: -0.10%, +0.00%
InvThroughput: 25090218 -> 25089812 (-0.00%); split: -0.00%, +0.00%
VClause: 577302 -> 577319 (+0.00%); split: -0.00%, +0.00%
SClause: 801614 -> 801572 (-0.01%); split: -0.01%, +0.00%
Copies: 3399700 -> 2987201 (-12.13%); split: -12.13%, +0.00%
Branches: 1262859 -> 1262854 (-0.00%)
PreSGPRs: 2175752 -> 2141331 (-1.58%)
PreVGPRs: 1785088 -> 1785087 (-0.00%)

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11224>
 2021-09-29 16:00:19 +02:00
Daniel Schürmann 40a93e271c aco: clang-format 
No changes, just formatting.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13087>
 2021-09-28 19:48:00 +00:00
Timur Kristóf 9478901824 aco: Implement integer conversions using p_extract. 
Fossil DB stats on Sienna Cichlid:

Totals from 563 (0.44% of 128647) affected shaders:
SpillSGPRs: 1381 -> 1382 (+0.07%)
SpillVGPRs: 1606 -> 1552 (-3.36%)
CodeSize: 2474724 -> 2446612 (-1.14%); split: -1.15%, +0.02%
Scratch: 181248 -> 180224 (-0.56%)
Instrs: 440973 -> 435091 (-1.33%); split: -1.35%, +0.01%
Latency: 9123609 -> 8517830 (-6.64%); split: -6.66%, +0.02%
InvThroughput: 3685256 -> 3383293 (-8.19%); split: -8.22%, +0.02%
VClause: 8425 -> 8372 (-0.63%)
Copies: 66553 -> 66681 (+0.19%); split: -0.49%, +0.68%
Branches: 13824 -> 13825 (+0.01%); split: -0.01%, +0.01%
PreSGPRs: 21816 -> 21824 (+0.04%)

Fossil DB stats on Sienna Cichlid with NGGC on:

Totals from 58802 (45.71% of 128647) affected shaders:
SpillSGPRs: 6541 -> 6542 (+0.02%)
SpillVGPRs: 1606 -> 1552 (-3.36%)
CodeSize: 162976608 -> 162244340 (-0.45%); split: -0.45%, +0.00%
Scratch: 181248 -> 180224 (-0.56%)
Instrs: 31163521 -> 31098078 (-0.21%); split: -0.21%, +0.00%
Latency: 146893569 -> 144920070 (-1.34%); split: -1.34%, +0.00%
InvThroughput: 25384324 -> 25035940 (-1.37%); split: -1.38%, +0.00%
VClause: 552310 -> 552257 (-0.01%)
Copies: 3356856 -> 3356984 (+0.00%); split: -0.01%, +0.01%
Branches: 1237314 -> 1237315 (+0.00%); split: -0.00%, +0.00%
PreSGPRs: 2185339 -> 2185347 (+0.00%)

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11560>
 2021-09-28 17:59:27 +00:00
Samuel Pitoiset deede6b03d radv: pass the pipeline key to the backend compilers 
It exactly matches the shader keys now. Everything was copied from
the pipeline key to the shader keys.

There is still some work to completely remove radv_shader_variant_key.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13032>
 2021-09-27 11:57:25 +02:00
Bas Nieuwenhuizen 8ca54b4d38 radv: Support nir_intrinsic_load_global_constant. 
SPIR-V parsing can result in some direct constant usage for shader
records. Lower this early to a global based intrinsic so that it
doesn't interfere with the later 32-bit offset based constants
for scratch usage.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12592>
 2021-09-21 01:53:39 +00:00
Bas Nieuwenhuizen c299968988 aco: Add support for ray launch size. 
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12592>
 2021-09-21 01:53:39 +00:00
Bas Nieuwenhuizen 817553c052 aco: Implement call scope. 
Since we do no repacking yet, just use invocation.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12592>
 2021-09-21 01:53:39 +00:00
Bas Nieuwenhuizen b6be96a2bd radv: Modify load_sbt_amd intrinsic to get the descriptor. 
That way we can get the address to the entry, which is needed for
some nir builtins because extra data in the entry can be used as
shader input.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12592>
 2021-09-21 01:53:39 +00:00
Timur Kristóf 966cff9cfa aco/isel: Fix emit_vop2_instruction to apply 16/24-bit flags properly. 
Previously it used a builder function but didn't use the return
value from that function, so the flags were not applied.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12786>
 2021-09-20 12:39:03 +02:00
Rhys Perry 2a7fa132be aco: implement udot_4x8/sdot_4x8/udot_2x16/sdot_2x16 opcodes 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12617>
 2021-09-03 13:21:28 +00:00
Rhys Perry e0d232c2fc aco: implement nir_op_pack_32_4x8 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12617>
 2021-09-03 13:21:28 +00:00
Rhys Perry 4dd420f76d radv,aco: implement iadd_sat 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12617>
 2021-09-03 13:21:28 +00:00
Daniel Schürmann 0988f7b9ba aco: remove explicit dst_preserve flag 
Instead, we can rely on the fact that subdword definitions
must preserve the unused bits while dword definitions either
pad or sign-extend.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12640>
 2021-09-02 20:39:17 +02:00
Daniel Schürmann 9e3ff06c38 aco: rewrite SDWA selector 
This commit introduces a new struct SubdwordSel
in order to ease and clean up the usage of SDWA
selections. This includes removing the distinction
between register-allocated and fixed SDWA selections.
Instead, SDWA selections can now also access the high
bits of subdword variables. Alignment and sizes are
validated accordingly. Size, offset and sign_extend
can be evaluated via helper methods.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12640>
 2021-09-02 20:39:17 +02:00
Rhys Perry 9df9fe7dfa aco: include utility in isel 
For std::exchange().

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Fixes: c1d11bb92c ("aco: Add loop creation helpers.")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5301
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12614>
 2021-08-30 14:28:00 +00:00
Timur Kristóf cfb0d931f2 aco: Emit zero for the derivatives of uniforms. 
Observed in a shader from Resident Evil Village.
This also helps prevent emitting invalid IR.

Cc: mesa-stable
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12599>
 2021-08-27 20:34:22 +00:00
Daniel Schürmann 23d5865f42 aco: refactor nir_op_imul selection 
Previously, the optimization to use v_mul_lo_u16 for
32bit multiplications was done in instruction_selection.
This was moved to the optimizer to ease some case distinctions.

The mixed results are due to increased use of SDWA.

Totals from 2616 (1.74% of 150170) affected shaders: (GFX10.3)
VGPRs: 143888 -> 143872 (-0.01%); split: -0.02%, +0.01%
CodeSize: 5604032 -> 5604080 (+0.00%); split: -0.01%, +0.01%
Instrs: 1086798 -> 1083915 (-0.27%); split: -0.27%, +0.01%
Latency: 8215793 -> 8213023 (-0.03%); split: -0.10%, +0.07%
InvThroughput: 20765157 -> 20773766 (+0.04%); split: -0.02%, +0.06%
VClause: 35256 -> 35260 (+0.01%); split: -0.02%, +0.03%
SClause: 29021 -> 29024 (+0.01%); split: -0.00%, +0.01%
Copies: 74163 -> 74306 (+0.19%); split: -0.05%, +0.24%

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11678>
 2021-08-27 19:57:59 +00:00