fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2025-12-23 02:30:12 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 1
mesa/src/intel/compiler
History
Francisco Jerez 8eb4f2092a intel/fs: Add support for copy-propagating a block of multiple FIXED_GRFs. 
In cases where a LOAD_PAYLOAD instruction copies a single block of
sequential GRF registers into the destination (see
is_identity_payload()), splitting the block copy into a number of ACP
entries (one for each LOAD_PAYLOAD source) is undesirable, because
that prevents copy propagation into any instructions which read
multiple components at once with the same source (the barycentric
source of the LINTERP instruction is going to be the overwhelmingly
most common example).

Technically it would also be possible to do this for VGRF sources, but
there is little benefit from that since register coalesce already
covers many of those cases -- There is no way for a block of
FIXED_GRFs to be coalesced into a VGRF though.

This prevents the following shader-db regressions (including SIMD32
programs) in combination with the interpolation rework part of this
series.  On SKL:

   total instructions in shared programs: 18595160 -> 18828562 (1.26%)
   instructions in affected programs: 13374946 -> 13608348 (1.75%)
   helped: 7
   HURT: 108977

   total spills in shared programs: 9116 -> 9106 (-0.11%)
   spills in affected programs: 404 -> 394 (-2.48%)
   helped: 7
   HURT: 9

   total fills in shared programs: 8994 -> 9176 (2.02%)
   fills in affected programs: 898 -> 1080 (20.27%)
   helped: 7
   HURT: 9

   LOST:   469
   GAINED: 220

On SNB:

   total instructions in shared programs: 13996898 -> 14096222 (0.71%)
   instructions in affected programs: 8088546 -> 8187870 (1.23%)
   helped: 2
   HURT: 66520

   total spills in shared programs: 2985 -> 2961 (-0.80%)
   spills in affected programs: 632 -> 608 (-3.80%)
   helped: 2
   HURT: 0

   total fills in shared programs: 3144 -> 3128 (-0.51%)
   fills in affected programs: 1515 -> 1499 (-1.06%)
   helped: 2
   HURT: 0

   LOST:   0
   GAINED: 4

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2020-01-17 13:21:41 -08:00
..
brw_cfg.cpp intel/ir: Represent physical edge of unconditional CONTINUE instruction. 2019-10-11 12:24:16 -07:00
brw_cfg.h intel/ir: Represent physical and logical subsets of the CFG. 2019-10-11 12:24:16 -07:00
brw_clip.h
brw_clip_line.c
brw_clip_point.c
brw_clip_tri.c i965: Don't emit MOVs with undefined registers for Gen4 point clipping. 2018-02-28 15:03:51 -08:00
brw_clip_unfilled.c
brw_clip_util.c
brw_compile_clip.c intel/common: move gen_debug to intel/dev 2019-04-10 13:15:33 -07:00
brw_compile_sf.c intel/common: move gen_debug to intel/dev 2019-04-10 13:15:33 -07:00
brw_compiler.c iris: Set nir_shader_compiler_options::unify_interfaces. 2020-01-03 00:41:50 +00:00
brw_compiler.h intel/compiler: Add a flag to avoid compacting push constants 2019-11-18 18:35:14 +00:00
brw_dead_control_flow.cpp
brw_dead_control_flow.h
brw_debug_recompile.c intel/compiler: Add a "base class" for program keys 2019-07-10 19:35:55 +00:00
brw_disasm.c intel/disasm: Fix decoding of src0 of SENDS 2020-01-08 14:14:16 +00:00
brw_disasm_info.c intel/common: move gen_debug to intel/dev 2019-04-10 13:15:33 -07:00
brw_disasm_info.h
brw_eu.cpp intel/compiler: replace 0 pointer with NULL 2019-12-13 20:16:20 +00:00
brw_eu.h intel/fs: Add DWord scattered read/write opcodes 2019-11-11 17:17:02 +00:00
brw_eu_compact.c intel/compiler: Fix 'comparison is always true' warning 2019-12-04 15:34:03 -08:00
brw_eu_defines.h intel/fs: Add DWord scattered read/write opcodes 2019-11-11 17:17:02 +00:00
brw_eu_emit.c intel/compiler: Don't change hstride if not needed 2019-11-18 14:19:41 -08:00
brw_eu_util.c
brw_eu_validate.c intel/compiler: Don't left-shift by >= the number of bits of the type 2019-10-24 16:16:49 +02:00
brw_fs.cpp intel/fs: Rework fs_inst::is_copy_payload() into multiple classification helpers. 2020-01-17 13:21:19 -08:00
brw_fs.h intel/fs: Make implied_mrf_writes() an fs_inst method. 2020-01-10 11:02:30 -08:00
brw_fs_bank_conflicts.cpp intel/fs: Make implied_mrf_writes() an fs_inst method. 2020-01-10 11:02:30 -08:00
brw_fs_builder.h intel/compiler: remove the operand restriction for src1 on GLK 2019-11-05 00:08:34 +00:00
brw_fs_cmod_propagation.cpp intel/fs: Allow cmod propagation across reads and writes of different flags 2019-06-05 17:03:45 -07:00
brw_fs_combine_constants.cpp intel/compiler: Don't move immediate in register 2019-10-21 20:32:43 -07:00
brw_fs_copy_propagation.cpp intel/fs: Add support for copy-propagating a block of multiple FIXED_GRFs. 2020-01-17 13:21:41 -08:00
brw_fs_cse.cpp intel/fs: Rework fs_inst::is_copy_payload() into multiple classification helpers. 2020-01-17 13:21:19 -08:00
brw_fs_dead_code_eliminate.cpp intel/fs: Properly stride NULL replacement regs in DCE 2019-07-17 18:44:35 +00:00
brw_fs_generator.cpp intel/fs: Implement the new load/store_scratch intrinsics 2019-11-11 17:17:02 +00:00
brw_fs_live_variables.cpp intel/compiler: Fix C++ one definition rule violations 2019-10-28 12:02:40 +02:00
brw_fs_live_variables.h intel/compiler: Fix C++ one definition rule violations 2019-10-28 12:02:40 +02:00
brw_fs_lower_pack.cpp
brw_fs_lower_regioning.cpp intel/fs: Add an UNDEF instruction to avoid excess live ranges 2019-06-04 14:27:30 -05:00
brw_fs_nir.cpp intel/compiler: Fix illegal mutation in get_nir_image_intrinsic_image 2020-01-15 19:25:35 +00:00
brw_fs_reg_allocate.cpp intel/fs: Make implied_mrf_writes() an fs_inst method. 2020-01-10 11:02:30 -08:00
brw_fs_register_coalesce.cpp intel/fs: Rework fs_inst::is_copy_payload() into multiple classification helpers. 2020-01-17 13:21:19 -08:00
brw_fs_saturate_propagation.cpp Revert "intel/compiler: split is_partial_write() into two variants" 2019-04-25 09:19:10 +02:00
brw_fs_scoreboard.cpp intel/fs: Make implied_mrf_writes() an fs_inst method. 2020-01-10 11:02:30 -08:00
brw_fs_sel_peephole.cpp Revert "intel/compiler: split is_partial_write() into two variants" 2019-04-25 09:19:10 +02:00
brw_fs_validate.cpp intel: disable FS IR validation in release mode. 2018-10-15 18:10:27 -07:00
brw_fs_visitor.cpp intel/fs: Check for NULL key in fs_visitor constructor 2019-10-24 16:20:04 +02:00
brw_inst.h intel/compiler: Add instruction compaction support on Gen12 2019-10-30 11:11:50 -07:00
brw_interpolation_map.c intel/compiler: Silence unused parameter warning in brw_interpolation_map.c 2019-03-06 08:35:36 -08:00
brw_ir_allocator.h intel/ir: Don't allow allocating zero registers 2018-12-11 21:26:23 -06:00
brw_ir_fs.h intel/fs: Rework fs_inst::is_copy_payload() into multiple classification helpers. 2020-01-17 13:21:19 -08:00
brw_ir_vec4.h intel: Don't propagate conditional modifiers if a UD source is negated 2018-10-10 13:13:12 -05:00
brw_nir.c anv,nir: Lower quad_broadcast with dynamic index in NIR 2020-01-15 08:34:57 -06:00
brw_nir.h i965: Reuse the new core glsl_count_dword_slots(). 2020-01-14 23:55:00 +00:00
brw_nir_analyze_boolean_resolves.c intel/fs: Mark source 0 of bcsel as needing Boolean resolve 2019-06-11 12:12:07 -07:00
brw_nir_analyze_ubo_ranges.c nir: Add explicit signs to image min/max intrinsics 2019-08-21 17:19:55 +00:00
brw_nir_attribute_workarounds.c nir/builder: Remove the use_fmov parameter from nir_swizzle 2019-05-24 08:38:11 -05:00
brw_nir_lower_alpha_to_coverage.c nir: Add alpha_to_coverage lowering pass 2019-10-21 11:27:29 -07:00
brw_nir_lower_conversions.c intel/compiler: add a NIR pass to lower conversions 2019-04-18 11:05:18 +02:00
brw_nir_lower_cs_intrinsics.c intel/nir: Stop adding redundant barriers 2020-01-13 17:23:47 +00:00
brw_nir_lower_image_load_store.c nir: move data.image.access to data.access 2019-11-19 18:20:05 -05:00
brw_nir_lower_mem_access_bit_sizes.c intel/fs: Implement the new load/store_scratch intrinsics 2019-11-11 17:17:02 +00:00
brw_nir_opt_peephole_ffma.c util: rename list_empty() to list_is_empty() 2019-10-28 11:24:38 +00:00
brw_nir_tcs_workarounds.c util: use C99 declaration in the for-loop set_foreach() macro 2018-10-25 12:43:18 +01:00
brw_nir_trig_workarounds.py intel/nir: do not apply the fsin and fcos trig workarounds for consts 2019-09-17 23:39:18 +03:00
brw_packed_float.c intel/compiler: Cast to target type before shifting left 2019-10-24 16:19:23 +02:00
brw_predicated_break.cpp intel/ir: Represent physical and logical subsets of the CFG. 2019-10-11 12:24:16 -07:00
brw_reg.h intel/compiler: Expand size of the 'nr' field 2019-01-09 16:42:41 -08:00
brw_reg_type.c intel/compiler: Remove unreachable() from brw_reg_type.c 2019-10-30 11:11:50 -07:00
brw_reg_type.h intel/compiler: add a brw_reg_type_is_integer helper 2019-04-18 11:05:18 +02:00
brw_schedule_instructions.cpp intel/fs: Make implied_mrf_writes() an fs_inst method. 2020-01-10 11:02:30 -08:00
brw_shader.cpp intel/fs: Add DWord scattered read/write opcodes 2019-11-11 17:17:02 +00:00
brw_shader.h intel/nir: Take a nir_tex_instr and src index in brw_texture_offset 2019-04-14 22:25:56 +02:00
brw_vec4.cpp intel/compiler: Add a flag to avoid compacting push constants 2019-11-18 18:35:14 +00:00
brw_vec4.h intel/compiler: Fill a compiler statistics struct 2019-08-12 22:56:07 +00:00
brw_vec4_builder.h intel/compiler: Lower flrp32 on Gen11+ 2018-02-28 11:15:47 -08:00
brw_vec4_cmod_propagation.cpp intel/compiler: use correct swizzle for replacement 2019-02-27 20:06:42 +00:00
brw_vec4_copy_propagation.cpp intel/compiler: Re-prefix non-logical surface opcodes with VEC4 2019-02-28 16:58:20 -06:00
brw_vec4_cse.cpp i965/vec4: Allow CSE on subset VF constant loads 2018-03-08 15:26:26 -08:00
brw_vec4_dead_code_eliminate.cpp i965/vec4/dce: Don't narrow the write mask if the flags are used 2018-12-17 13:47:06 -08:00
brw_vec4_generator.cpp intel/compiler: Report the number of non-spill/fill SEND messages on vec4 too 2019-10-30 21:27:03 -07:00
brw_vec4_gs_nir.cpp intel/vec4: Drop all of the 64-bit varying code 2019-07-31 18:14:09 -05:00
brw_vec4_gs_visitor.cpp intel/compiler: Fill a compiler statistics struct 2019-08-12 22:56:07 +00:00
brw_vec4_gs_visitor.h
brw_vec4_live_variables.cpp intel/compiler: Fix C++ one definition rule violations 2019-10-28 12:02:40 +02:00
brw_vec4_live_variables.h intel/compiler: Fix C++ one definition rule violations 2019-10-28 12:02:40 +02:00
brw_vec4_nir.cpp intel/vec4: Support scoped_memory_barrier 2020-01-13 17:23:46 +00:00
brw_vec4_reg_allocate.cpp intel/compiler: Prevent warnings in the following patch 2019-01-09 16:42:41 -08:00
brw_vec4_surface_builder.cpp intel/compiler: Re-prefix non-logical surface opcodes with VEC4 2019-02-28 16:58:20 -06:00
brw_vec4_surface_builder.h intel/vec4: Drop dead code for handling typed surface messages 2019-02-28 16:58:20 -06:00
brw_vec4_tcs.cpp nir: Rename nir_intrinsic_barrier to control_barrier 2020-01-13 17:23:47 +00:00
brw_vec4_tcs.h
brw_vec4_tes.cpp intel/vec4: Drop all of the 64-bit varying code 2019-07-31 18:14:09 -05:00
brw_vec4_tes.h
brw_vec4_visitor.cpp intel/vec4: Delete vec4_visitor::emit_lrp 2019-07-08 11:30:11 -07:00
brw_vec4_vs.h i965: Use NIR to lower legacy userclipping. 2019-07-24 18:00:13 +00:00
brw_vec4_vs_visitor.cpp i965: Use NIR to lower legacy userclipping. 2019-07-24 18:00:13 +00:00
brw_vue_map.c intel/compiler: silence a warning of using different enum type 2019-06-25 10:09:22 +03:00
brw_wm_iz.cpp intel: Use a system value for gl_FragCoord 2019-07-29 23:30:26 +00:00
gen6_gs_visitor.cpp intel/compiler: Prevent warnings in the following patch 2019-01-09 16:42:41 -08:00
gen6_gs_visitor.h
meson.build nir: Add alpha_to_coverage lowering pass 2019-10-21 11:27:29 -07:00
test_eu_compact.cpp intel/eu: Encode and decode native instruction opcodes from/to IR opcodes. 2019-10-11 12:24:16 -07:00
test_eu_validate.cpp intel/eu/validate/gen12: Add TGL to eu_validate tests. 2019-10-30 14:08:51 -07:00
test_fs_cmod_propagation.cpp intel/fs: Drop the gl_program from fs_visitor 2019-08-25 01:02:52 -05:00
test_fs_copy_propagation.cpp intel/fs: Drop the gl_program from fs_visitor 2019-08-25 01:02:52 -05:00
test_fs_saturate_propagation.cpp intel/fs: Drop the gl_program from fs_visitor 2019-08-25 01:02:52 -05:00
test_fs_scoreboard.cpp intel/fs/gen12: Add tests for scoreboard pass 2019-10-17 10:02:35 -07:00
test_vec4_cmod_propagation.cpp i965/vec4: Silence unused parameter warnings in vec4 compiler tests 2018-12-17 13:47:06 -08:00
test_vec4_copy_propagation.cpp i965/vec4: Silence unused parameter warnings in vec4 compiler tests 2018-12-17 13:47:06 -08:00
test_vec4_dead_code_eliminate.cpp i965/vec4/dce: Don't narrow the write mask if the flags are used 2018-12-17 13:47:06 -08:00
test_vec4_register_coalesce.cpp i965/vec4: Silence unused parameter warnings in vec4 compiler tests 2018-12-17 13:47:06 -08:00
test_vf_float_conversions.cpp