mesa/src/intel/compiler/meson.build
Caio Oliveira dba6451ce8 intel/cmat: Add pass to lower cooperative matrix to subgroup operations
This is just the skeleton of the implementation. Future commits will
fill it all in.

v2: Move to src/intel/compiler

v3 (idr): Use vecN instead of array[N] for slice type.

v4 (idr): Refactor lower_cooperative_matrix_load and
lower_cooperative_matrix_store into a single function.

v5 (idr): Remove old, verbose debug logging. Assert that entry is not
NULL in get_coop_type_for_slice. Use nir_component_mask(...) instead of
0xffff. s/cooperative_matrix/cmat/. All suggested by Caio.

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>

I put both R-b on this because, at this point, we've each done equal
parts authoring and reviewing.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25994>
2023-12-29 20:24:15 -08:00

222 lines
6.5 KiB
Meson

# Copyright © 2017 Intel Corporation
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
libintel_compiler_files = files(
'brw_cfg.cpp',
'brw_cfg.h',
'brw_clip.h',
'brw_clip_line.c',
'brw_clip_point.c',
'brw_clip_tri.c',
'brw_clip_unfilled.c',
'brw_clip_util.c',
'brw_compile_clip.c',
'brw_compile_ff_gs.c',
'brw_compile_sf.c',
'brw_compiler.c',
'brw_compiler.h',
'brw_dead_control_flow.cpp',
'brw_dead_control_flow.h',
'brw_debug_recompile.c',
'brw_disasm.c',
'brw_disasm_info.c',
'brw_disasm_info.h',
'brw_eu.c',
'brw_eu_compact.c',
'brw_eu_defines.h',
'brw_eu_emit.c',
'brw_eu.h',
'brw_eu_util.c',
'brw_eu_validate.c',
'brw_fs_bank_conflicts.cpp',
'brw_fs_builder.h',
'brw_fs_cmod_propagation.cpp',
'brw_fs_combine_constants.cpp',
'brw_fs_copy_propagation.cpp',
'brw_fs.cpp',
'brw_fs_cse.cpp',
'brw_fs_dead_code_eliminate.cpp',
'brw_fs_generator.cpp',
'brw_fs.h',
'brw_fs_live_variables.cpp',
'brw_fs_live_variables.h',
'brw_fs_lower_pack.cpp',
'brw_fs_lower_regioning.cpp',
'brw_fs_nir.cpp',
'brw_fs_reg_allocate.cpp',
'brw_fs_register_coalesce.cpp',
'brw_fs_saturate_propagation.cpp',
'brw_fs_scoreboard.cpp',
'brw_fs_sel_peephole.cpp',
'brw_fs_thread_payload.cpp',
'brw_fs_validate.cpp',
'brw_fs_visitor.cpp',
'brw_inst.h',
'brw_interpolation_map.c',
'brw_ir.h',
'brw_ir_allocator.h',
'brw_ir_analysis.h',
'brw_ir_fs.h',
'brw_ir_performance.h',
'brw_ir_performance.cpp',
'brw_ir_vec4.h',
'brw_isa_info.h',
'brw_lower_logical_sends.cpp',
'brw_mesh.cpp',
'brw_nir.h',
'brw_nir.c',
'brw_nir_analyze_boolean_resolves.c',
'brw_nir_analyze_ubo_ranges.c',
'brw_nir_attribute_workarounds.c',
'brw_nir_blockify_uniform_loads.c',
'brw_nir_clamp_per_vertex_loads.c',
'brw_nir_lower_conversions.c',
'brw_nir_lower_cooperative_matrix.c',
'brw_nir_lower_cs_intrinsics.c',
'brw_nir_lower_alpha_to_coverage.c',
'brw_nir_lower_intersection_shader.c',
'brw_nir_lower_non_uniform_resource_intel.c',
'brw_nir_lower_non_uniform_barycentric_at_sample.c',
'brw_nir_lower_ray_queries.c',
'brw_nir_lower_rt_intrinsics.c',
'brw_nir_lower_shader_calls.c',
'brw_nir_lower_shading_rate_output.c',
'brw_nir_lower_sparse.c',
'brw_nir_lower_storage_image.c',
'brw_nir_opt_peephole_ffma.c',
'brw_nir_opt_peephole_imul32x16.c',
'brw_nir_rt.h',
'brw_nir_rt.c',
'brw_nir_rt_builder.h',
'brw_nir_tcs_workarounds.c',
'brw_nir_clamp_image_1d_2d_array_sizes.c',
'brw_packed_float.c',
'brw_predicated_break.cpp',
'brw_prim.h',
'brw_private.h',
'brw_reg.h',
'brw_reg_type.c',
'brw_reg_type.h',
'brw_rt.h',
'brw_schedule_instructions.cpp',
'brw_shader.cpp',
'brw_shader.h',
'brw_simd_selection.cpp',
'brw_vec4_builder.h',
'brw_vec4_cmod_propagation.cpp',
'brw_vec4_copy_propagation.cpp',
'brw_vec4.cpp',
'brw_vec4_cse.cpp',
'brw_vec4_dead_code_eliminate.cpp',
'brw_vec4_generator.cpp',
'brw_vec4_gs_visitor.cpp',
'brw_vec4_gs_visitor.h',
'brw_vec4.h',
'brw_vec4_live_variables.cpp',
'brw_vec4_live_variables.h',
'brw_vec4_nir.cpp',
'brw_vec4_gs_nir.cpp',
'brw_vec4_reg_allocate.cpp',
'brw_vec4_surface_builder.cpp',
'brw_vec4_surface_builder.h',
'brw_vec4_tcs.cpp',
'brw_vec4_tcs.h',
'brw_vec4_tes.cpp',
'brw_vec4_tes.h',
'brw_vec4_visitor.cpp',
'brw_vec4_vs_visitor.cpp',
'brw_vec4_vs.h',
'brw_vue_map.c',
'gfx6_gs_visitor.cpp',
'gfx6_gs_visitor.h',
)
brw_nir_trig = custom_target(
'brw_nir_trig_workarounds.c',
input : 'brw_nir_trig_workarounds.py',
output : 'brw_nir_trig_workarounds.c',
command : [
prog_python, '@INPUT@', '-p', dir_compiler_nir,
],
depend_files : nir_algebraic_depends,
capture : true,
)
libintel_compiler = static_library(
'intel_compiler',
[libintel_compiler_files, brw_nir_trig, ir_expression_operation_h],
include_directories : [inc_include, inc_src, inc_intel],
c_args : [no_override_init_args],
gnu_symbol_visibility : 'hidden',
dependencies : [idep_nir_headers, idep_mesautil, idep_intel_dev],
build_by_default : false,
)
# For now this tool is only going to be used by Anv
if get_option('intel-clc') == 'system'
prog_intel_clc = find_program('intel_clc', native : true)
dep_prog_intel_clc = []
elif with_intel_clc
prog_intel_clc = executable(
'intel_clc',
['intel_clc.c', 'brw_kernel.c'],
link_with : [
libintel_compiler, libintel_common,libisl,
],
include_directories : [inc_include, inc_src, inc_intel],
c_args : [pre_args, no_override_init_args],
link_args : [ld_args_build_id],
dependencies : [idep_nir, idep_vtn, idep_mesaclc, idep_mesautil, idep_intel_dev],
native : true,
)
dep_prog_intel_clc = [prog_intel_clc]
endif
if with_tests
test(
'intel_compiler_tests',
executable(
'intel_compiler_tests',
files(
'test_predicated_break.cpp',
'test_eu_compact.cpp',
'test_eu_validate.cpp',
'test_fs_cmod_propagation.cpp',
'test_fs_copy_propagation.cpp',
'test_fs_saturate_propagation.cpp',
'test_fs_scoreboard.cpp',
'test_simd_selection.cpp',
'test_vec4_cmod_propagation.cpp',
'test_vec4_copy_propagation.cpp',
'test_vec4_dead_code_eliminate.cpp',
'test_vec4_register_coalesce.cpp',
'test_vf_float_conversions.cpp',
),
ir_expression_operation_h,
include_directories : [inc_include, inc_src, inc_intel],
link_with : [
libintel_compiler, libintel_common, libisl
],
dependencies : [idep_gtest, idep_nir, idep_mesautil, idep_intel_dev],
),
suite : ['intel'],
protocol : 'gtest',
)
endif