AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
78ef51aa040892f1ab4aca7bd45c82f6d3bfcffb
mesa/src/intel/compiler
T
History
Ian Romanick 20cce95ce5 brw/opt: Don't call brw_opt_copy_propagation before brw_lower_load_reg 
On a 36c/72t Xeon system, performance of replaying
hogwarts_legacy.dx12vk-ultra.foz was improved 1.3% +/- 0.77% (n=10).

I picked MTL for the fossil-db results because it was the most negative.

shader-db:

All Intel platforms had fairly similar results. (Lunar Lake)
total instructions in shared programs: 16964217 -> 16964216 (<.01%)
instructions in affected programs: 51777 -> 51776 (<.01%)
helped: 20 / HURT: 27

total cycles in shared programs: 892934916 -> 893041912 (0.01%)
cycles in affected programs: 51245298 -> 51352294 (0.21%)
helped: 96 /HURT: 78

fossil-db:

All Intel platforms had similar results. (Meteor Lake shown)
Totals:
Instrs: 233678547 -> 233678944 (+0.00%); split: -0.00%, +0.00%
Cycle count: 24398049850 -> 24400490877 (+0.01%); split: -0.01%, +0.02%
Max live registers: 42145052 -> 42145038 (-0.00%); split: -0.00%, +0.00%

Totals from 1141 (0.14% of 805934) affected shaders:
Instrs: 1546001 -> 1546398 (+0.03%); split: -0.01%, +0.03%
Cycle count: 1201746062 -> 1204187089 (+0.20%); split: -0.14%, +0.34%
Max live registers: 84247 -> 84233 (-0.02%); split: -0.03%, +0.01%

Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31497>
2025-04-04 06:45:02 +00:00
..
elk
intel/compiler: fix lingering i965 references
2025-04-03 03:17:25 +00:00
tests
brw: Rename shared function enums for clarity
2025-02-27 08:49:24 +00:00
brw_analysis_def.cpp
brw: Add basic infrastructure for load_reg pseudo op
2025-04-04 06:45:02 +00:00
brw_analysis_liveness.cpp
brw: Use brw_ip_ranges in liveness analysis
2025-03-29 00:25:51 +00:00
brw_analysis_performance.cpp
intel/compiler: Add support for MSAA typed load/store messages
2025-03-07 23:06:14 +00:00
brw_analysis.cpp
brw: Add analysis for block IP ranges
2025-03-29 00:25:50 +00:00
brw_analysis.h
brw: Use brw_ip_ranges in liveness analysis
2025-03-29 00:25:51 +00:00
brw_asm_internal.h
brw: Rework label tracking in assembler
2025-03-06 17:06:20 -08:00
brw_asm_tool.c
intel/compiler: fix lingering i965 references
2025-04-03 03:17:25 +00:00
brw_asm.c
brw: Rework label tracking in assembler
2025-03-06 17:06:20 -08:00
brw_asm.h
brw: Fix size in assembler when compacting
2025-03-03 20:43:56 +00:00
brw_builder.cpp
intel/brw: Rename fs_visitor to brw_shader
2025-02-11 09:13:28 +00:00
brw_builder.h
brw: Add basic infrastructure for load_reg pseudo op
2025-04-04 06:45:02 +00:00
brw_cfg.cpp
brw: Remove adjust_block_ips and brw_inst::remove() with defer
2025-03-29 00:25:51 +00:00
brw_cfg.h
brw: Remove adjust_block_ips and brw_inst::remove() with defer
2025-03-29 00:25:51 +00:00
brw_compile_bs.cpp
brw: factor out base prog_data setting
2025-02-22 08:30:22 +00:00
brw_compile_cs.cpp
brw: Embed at_end() inside brw_builder(brw_shader *) constructor
2025-03-06 23:33:38 +00:00
brw_compile_fs.cpp
nir, compiler: Rename needs_quad_helper_invocations
2025-03-14 21:55:57 +00:00
brw_compile_gs.cpp
brw: Embed at_end() inside brw_builder(brw_shader *) constructor
2025-03-06 23:33:38 +00:00
brw_compile_mesh.cpp
brw: Embed at_end() inside brw_builder(brw_shader *) constructor
2025-03-06 23:33:38 +00:00
brw_compile_tcs.cpp
brw: Embed at_end() inside brw_builder(brw_shader *) constructor
2025-03-06 23:33:38 +00:00
brw_compile_tes.cpp
brw: factor out base prog_data setting
2025-02-22 08:30:22 +00:00
brw_compile_vs.cpp
brw: factor out base prog_data setting
2025-02-22 08:30:22 +00:00
brw_compiler.c
intel/brw, lp: enable lower_pack_64_4x16
2025-02-26 20:43:39 +00:00
brw_compiler.h
brw: store source_hash in prog_data
2025-02-22 08:30:22 +00:00
brw_debug_recompile.c
intel/brw: Simplify @file annotations
2024-07-22 22:48:03 +00:00
brw_device_sha1_gen_c.py
intel/compiler: drop unused ray-tracing fields from cache hash
2024-03-22 00:01:28 +00:00
brw_disasm_info.cpp
brw: Remove extra interface in brw_cfg types
2025-03-10 19:23:17 +00:00
brw_disasm_info.h
intel/brw: Rename fs_inst to brw_inst
2025-01-31 00:57:21 +00:00
brw_disasm_tool.c
intel/brw: Remove Gfx8- code from disassembler
2024-02-28 05:45:38 +00:00
brw_disasm.c
brw: Fix disassembler trying to decode 3src_hstride in Gfx9
2025-03-26 18:12:46 +00:00
brw_disasm.h
intel/brw: Rename brw_inst to brw_eu_inst
2024-12-30 17:16:15 +00:00
brw_eu_compact.c
brw: Add support for GOTO/JOIN in the assembler
2025-03-06 17:06:20 -08:00
brw_eu_defines.h
brw: Add basic infrastructure for load_reg pseudo op
2025-04-04 06:45:02 +00:00
brw_eu_emit.c
brw: Add BRW_TYPE_BF for bfloat16
2025-03-25 05:23:37 +00:00
brw_eu_inst.h
brw: Add BRW_TYPE_BF for bfloat16
2025-03-25 05:23:37 +00:00
brw_eu_validate.c
brw: Expand EU validation for DPAS
2025-03-25 07:38:08 +00:00
brw_eu.c
brw: Add support for GOTO/JOIN in the assembler
2025-03-06 17:06:20 -08:00
brw_eu.h
brw: Add assembler support for DPAS
2025-03-25 01:40:02 +00:00
brw_from_nir.cpp
brw: move texture offset packing to NIR
2025-03-29 02:15:18 +00:00
brw_generator.cpp
brw: Add assembler support for DPAS
2025-03-25 01:40:02 +00:00
brw_generator.h
brw: factor out base prog_data setting
2025-02-22 08:30:22 +00:00
brw_gram.y
brw: Add EU assembler support for bfloat16
2025-03-25 05:23:37 +00:00
brw_inst.cpp
brw: Add basic infrastructure for load_reg pseudo op
2025-04-04 06:45:02 +00:00
brw_inst.h
brw: Remove adjust_block_ips and brw_inst::remove() with defer
2025-03-29 00:25:51 +00:00
brw_isa_info.h
intel/compiler: Use #pragma once instead of header guards
2024-12-11 19:47:44 +00:00
brw_kernel.c
treewide: Switch to nir_progress
2025-02-26 15:19:53 +00:00
brw_kernel.h
intel: rework CL pre-compile
2025-01-25 03:28:07 +00:00
brw_lex.l
brw: Add EU assembler support for bfloat16
2025-03-25 05:23:37 +00:00
brw_load_reg.cpp
brw: Add passes to generate and lower load_reg
2025-04-04 06:45:02 +00:00
brw_lower_dpas.cpp
brw: Simplify brw_builder "insert before inst" constructor
2025-03-06 23:33:38 +00:00
brw_lower_integer_multiplication.cpp
brw: Remove bblock_t parameters from various passes
2025-03-06 23:33:38 +00:00
brw_lower_logical_sends.cpp
brw: move texture offset packing to NIR
2025-03-29 02:15:18 +00:00
brw_lower_pack.cpp
brw: Simplify brw_builder "insert before inst" constructor
2025-03-06 23:33:38 +00:00
brw_lower_regioning.cpp
brw: Remove bblock_t parameters from various passes
2025-03-06 23:33:38 +00:00
brw_lower_scoreboard.cpp
brw: Use brw_ip_ranges in passes
2025-03-29 00:25:50 +00:00
brw_lower_simd_width.cpp
brw: Embed at_end() inside brw_builder(brw_shader *) constructor
2025-03-06 23:33:38 +00:00
brw_lower_subgroup_ops.cpp
brw: Remove bblock_t parameters from various passes
2025-03-06 23:33:38 +00:00
brw_lower.cpp
brw: Add missing dependency classes to various passes
2025-03-12 22:44:10 +00:00
brw_nir_analyze_ubo_ranges.c
intel/compiler: take reg_unit size into account with ubo ranges
2025-01-07 21:38:06 +00:00
brw_nir_lower_alpha_to_coverage.c
treewide: Switch to nir_progress
2025-02-26 15:19:53 +00:00
brw_nir_lower_cooperative_matrix.c
intel/brw/xe2+: Allow vec16 for cooperative matrix
2024-06-25 14:17:47 -07:00
brw_nir_lower_cs_intrinsics.c
treewide: Switch to nir_progress
2025-02-26 15:19:53 +00:00
brw_nir_lower_fsign.py
intel/brw: Use range analysis to optimize fsign
2024-05-14 01:28:21 +00:00
brw_nir_lower_intersection_shader.c
treewide: Switch to nir_progress
2025-02-26 15:19:53 +00:00
brw_nir_lower_ray_queries.c
treewide: Switch to nir_progress
2025-02-26 15:19:53 +00:00
brw_nir_lower_rt_intrinsics.c
treewide: Switch to nir_progress
2025-02-26 15:19:53 +00:00
brw_nir_lower_sample_index_in_coord.c
intel/compiler: Lower sample index into coord for MSRT messages
2025-03-07 23:06:14 +00:00
brw_nir_lower_shader_calls.c
treewide: Switch to nir_progress
2025-02-26 15:19:53 +00:00
brw_nir_lower_storage_image.c
brw: add support for 64bit storage images load/store
2025-02-23 15:16:50 +00:00
brw_nir_lower_texel_address.c
intel/compiler: Use correct enum type
2025-03-13 20:11:10 +00:00
brw_nir_lower_texture.c
brw: move texture offset packing to NIR
2025-03-29 02:15:18 +00:00
brw_nir_opt_fsat.c
treewide: Switch to nir_progress
2025-02-26 15:19:53 +00:00
brw_nir_rt_builder.h
intel/compiler: Drop primitive leaf desc load code
2025-02-12 05:23:05 +00:00
brw_nir_rt.c
treewide: Switch to nir_progress
2025-02-26 15:19:53 +00:00
brw_nir_rt.h
brw: enable A64 pulling of push constants
2025-02-05 09:56:04 +00:00
brw_nir_trig_workarounds.py
…
brw_nir.c
brw/nir: Lower fsign again after last call to brw_nir_optimize
2025-04-02 01:59:49 +00:00
brw_nir.h
brw: move texture offset packing to NIR
2025-03-29 02:15:18 +00:00
brw_opt_address_reg_load.cpp
brw: Remove adjust_block_ips and brw_inst::remove() with defer
2025-03-29 00:25:51 +00:00
brw_opt_algebraic.cpp
brw/algebraic: Constant folding for BROADCAST and SHUFFLE
2025-04-04 06:45:02 +00:00
brw_opt_bank_conflicts.cpp
intel/brw: Rename fs_visitor to brw_shader
2025-02-11 09:13:28 +00:00
brw_opt_cmod_propagation.cpp
brw: Remove adjust_block_ips and brw_inst::remove() with defer
2025-03-29 00:25:51 +00:00
brw_opt_combine_constants.cpp
brw: Use brw_inst::block in Combine Constants
2025-03-06 23:33:38 +00:00
brw_opt_copy_propagation.cpp
brw/opt: Don't call brw_opt_copy_propagation before brw_lower_load_reg
2025-04-04 06:45:02 +00:00
brw_opt_cse.cpp
brw: Remove adjust_block_ips and brw_inst::remove() with defer
2025-03-29 00:25:51 +00:00
brw_opt_dead_code_eliminate.cpp
brw: Remove adjust_block_ips and brw_inst::remove() with defer
2025-03-29 00:25:51 +00:00
brw_opt_register_coalesce.cpp
brw/coalesce: Prepare brw_opt_register_coalesce for load_reg
2025-04-04 06:45:02 +00:00
brw_opt_saturate_propagation.cpp
brw/sat: Eliminate non-defs saturate propagation
2025-04-04 06:45:02 +00:00
brw_opt_txf_combiner.cpp
brw: Simplify brw_builder "insert before inst" constructor
2025-03-06 23:33:38 +00:00
brw_opt_virtual_grfs.cpp
brw: Simplify brw_builder "insert before inst" constructor
2025-03-06 23:33:38 +00:00
brw_opt.cpp
brw/opt: Don't call brw_opt_copy_propagation before brw_lower_load_reg
2025-04-04 06:45:02 +00:00
brw_packed_float.c
…
brw_prim.h
intel/compiler: Use #pragma once instead of header guards
2024-12-11 19:47:44 +00:00
brw_print.cpp
brw: Add basic infrastructure for load_reg pseudo op
2025-04-04 06:45:02 +00:00
brw_private.h
intel/brw: Rename a few remaining functions to remove fs prefix
2025-02-11 09:13:28 +00:00
brw_reg_allocate.cpp
brw: Use brw_ip_ranges in scheduling / regalloc
2025-03-29 00:25:51 +00:00
brw_reg_type.c
brw: Add BRW_TYPE_BF for bfloat16
2025-03-25 05:23:37 +00:00
brw_reg_type.h
brw: Add BRW_TYPE_BF for bfloat16
2025-03-25 05:23:37 +00:00
brw_reg.cpp
intel/compiler: Use unreachable instead of assert(!"...")
2025-03-13 20:11:10 +00:00
brw_reg.h
brw: Add BRW_TYPE_BF for bfloat16
2025-03-25 05:23:37 +00:00
brw_rt.h
intel/compiler: Use #pragma once instead of header guards
2024-12-11 19:47:44 +00:00
brw_schedule_instructions.cpp
brw: Use brw_ip_ranges in scheduling / regalloc
2025-03-29 00:25:51 +00:00
brw_shader.cpp
brw: Use brw_ip_ranges in scheduling / regalloc
2025-03-29 00:25:51 +00:00
brw_shader.h
brw: Add passes to generate and lower load_reg
2025-04-04 06:45:02 +00:00
brw_simd_selection.cpp
intel/brw/xe3+: Optimize CS/TASK/MESH compile time optimistically assuming SIMD32.
2025-01-29 23:39:32 +00:00
brw_spirv.c
nir/peephole_select: add options struct
2025-02-20 21:59:16 +00:00
brw_thread_payload.cpp
brw: Embed at_end() inside brw_builder(brw_shader *) constructor
2025-03-06 23:33:38 +00:00
brw_thread_payload.h
intel/brw: Rename fs_visitor to brw_shader
2025-02-11 09:13:28 +00:00
brw_validate.cpp
brw: Add basic infrastructure for load_reg pseudo op
2025-04-04 06:45:02 +00:00
brw_vue_map.c
intel/brw: Simplify @file annotations
2024-07-22 22:48:03 +00:00
brw_workaround.cpp
brw: Simplify brw_builder "insert before inst" constructor
2025-03-06 23:33:38 +00:00
intel_clc.c
intel_clc: remove NIR output support
2025-02-01 07:54:37 +00:00
intel_gfx_ver_enum.h
intel/compiler: Use #pragma once instead of header guards
2024-12-11 19:47:44 +00:00
intel_nir_blockify_uniform_loads.c
intel: switch to nir_metadata_divergence
2025-02-13 10:08:43 +00:00
intel_nir_clamp_image_1d_2d_array_sizes.c
treewide: use nir_metadata_control_flow
2024-06-17 16:28:14 -04:00
intel_nir_clamp_per_vertex_loads.c
treewide: use nir_metadata_control_flow
2024-06-17 16:28:14 -04:00
intel_nir_lower_conversions.c
intel/nir: Don't needlessly split u2f16 for nir_type_uint32
2024-07-11 02:37:05 -07:00
intel_nir_lower_non_uniform_barycentric_at_sample.c
intel: switch to nir_metadata_divergence
2025-02-13 10:08:43 +00:00
intel_nir_lower_non_uniform_resource_intel.c
treewide: use nir_metadata_control_flow
2024-06-17 16:28:14 -04:00
intel_nir_lower_printf.c
nir: drop printf_base_identifier
2025-02-05 20:33:15 +00:00
intel_nir_lower_shading_rate_output.c
treewide: use nir_metadata_control_flow
2024-06-17 16:28:14 -04:00
intel_nir_lower_sparse.c
treewide: use nir_metadata_control_flow
2024-06-17 16:28:14 -04:00
intel_nir_opt_peephole_ffma.c
treewide: use nir_metadata_control_flow
2024-06-17 16:28:14 -04:00
intel_nir_opt_peephole_imul32x16.c
treewide: use nir_metadata_control_flow
2024-06-17 16:28:14 -04:00
intel_nir_tcs_workarounds.c
treewide: Switch to nir_progress
2025-02-26 15:19:53 +00:00
intel_nir.c
intel/compiler: Rename the passes and files related to intel_nir.h
2024-02-16 22:35:05 +00:00
intel_nir.h
intel: move lower_texture to brw
2025-03-29 02:15:18 +00:00
intel_shader_enums.h
anv/brw: move INTEL_MSAA_* flag computation to the compiler
2025-03-05 17:20:12 +00:00
meson.build
brw: Add passes to generate and lower load_reg
2025-04-04 06:45:02 +00:00
test_eu_compact.cpp
intel/brw: Rename brw_inst_* helpers to brw_eu_inst_*
2024-12-30 17:16:15 +00:00
test_eu_validate.cpp
brw: Expand EU validation for DPAS
2025-03-25 07:38:08 +00:00
test_helpers.cpp
brw: Simplify the test code for brw passes
2025-03-13 17:43:17 +00:00
test_helpers.h
brw: Simplify the test code for brw passes
2025-03-13 17:43:17 +00:00
test_insert_load_reg.cpp
brw: Add passes to generate and lower load_reg
2025-04-04 06:45:02 +00:00
test_lower_scoreboard.cpp
brw: Use block's num_instructions in scoreboard tests
2025-03-29 00:25:50 +00:00
test_opt_cmod_propagation.cpp
brw: Simplify the test code for brw passes
2025-03-13 17:43:17 +00:00
test_opt_combine_constants.cpp
brw: Simplify the test code for brw passes
2025-03-13 17:43:17 +00:00
test_opt_copy_propagation.cpp
brw: Simplify the test code for brw passes
2025-03-13 17:43:17 +00:00
test_opt_cse.cpp
brw: Simplify the test code for brw passes
2025-03-13 17:43:17 +00:00
test_opt_saturate_propagation.cpp
brw/sat: Eliminate non-defs saturate propagation
2025-04-04 06:45:02 +00:00
test_simd_selection.cpp
intel/brw/xe3+: Optimize CS/TASK/MESH compile time optimistically assuming SIMD32.
2025-01-29 23:39:32 +00:00
test_vf_float_conversions.cpp
…