d2d94b62f2
aco: initialize scratch base registers on GFX9-GFX10.3
...
fossil-db (navi21):
Totals from 1142 (0.70% of 162293) affected shaders:
Instrs: 271636 -> 271974 (+0.12%)
CodeSize: 1532020 -> 1533792 (+0.12%)
Latency: 7484066 -> 7485698 (+0.02%)
InvThroughput: 4048824 -> 4049579 (+0.02%)
SClause: 4171 -> 4212 (+0.98%)
PreSGPRs: 11203 -> 12276 (+9.58%)
fossil-db (vega10):
Totals from 3327 (2.06% of 161355) affected shaders:
Instrs: 257413 -> 257601 (+0.07%)
CodeSize: 1424244 -> 1425372 (+0.08%)
Latency: 8598402 -> 8600466 (+0.02%)
InvThroughput: 7906335 -> 7908234 (+0.02%)
SClause: 4932 -> 4973 (+0.83%)
PreSGPRs: 22010 -> 25405 (+15.42%)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17079 >
2022-07-08 14:49:03 +00:00
39800f0fa3
amd: change chip_class naming to "enum amd_gfx_level gfx_level"
...
This aligns the naming with PAL.
Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Pierre-Eric Pellou-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16469 >
2022-05-13 14:56:22 -04:00
a2701bfdb8
aco: move info pointer to a copy.
...
This is just setup to move this to a different struct later.
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16342 >
2022-05-11 19:07:11 +00:00
8d8c59b4cd
aco: split num_waves adjustment into separate function
...
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16039 >
2022-04-29 15:39:10 +00:00
6220046ad1
aco: remove 'max_waves' and use 'num_waves' to adjust for LDS and workgroup size
...
Totals from 21 (0.02% of 134913) affected shaders: (GFX10.3)
VGPRs: 1024 -> 1176 (+14.84%)
CodeSize: 127824 -> 127664 (-0.13%); split: -0.17%, +0.04%
MaxWaves: 416 -> 378 (-9.13%)
Instrs: 22521 -> 22502 (-0.08%); split: -0.17%, +0.09%
Latency: 146386 -> 143154 (-2.21%); split: -2.21%, +0.00%
InvThroughput: 28379 -> 28944 (+1.99%); split: -0.23%, +2.22%
VClause: 575 -> 579 (+0.70%); split: -0.87%, +1.57%
SClause: 692 -> 645 (-6.79%)
Copies: 780 -> 747 (-4.23%); split: -4.74%, +0.51%
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16039 >
2022-04-29 15:39:10 +00:00
b10c4d7dee
aco: make program->needs_vcc independent of VCC hints
...
Totals from 5 (0.00% of 135048) affected shaders: (GFX9)
SGPRs: 208 -> 160 (-23.08%)
CodeSize: 2700 -> 2692 (-0.30%)
Instrs: 533 -> 531 (-0.38%)
Latency: 41688 -> 41680 (-0.02%)
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15408 >
2022-04-13 21:52:43 +00:00
44fb9ba84a
aco/ra: only use VCC if program->needs_vcc == true
...
A future commit will make VCC register assignment independent
from register hints. Up to GFX9, VCC can alternatively be used
as regular SGPR, so prevent overlap.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15408 >
2022-04-13 21:52:43 +00:00
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
589ccf3d77
aco: Consider maximum number of workgroups per CU/WGP on Navi.
...
No Fossil DB changes.
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/12517 >
2021-08-27 16:41:08 +00:00
c8698199a1
aco: Consider LDS usage by PS inputs in MaxWaves calculation.
...
Before PS waves are launched, PS inputs are moved from PC to LDS
and the corresponding part of the PC is deallocated.
Each PS input occupies 3 * vec4 (3 * 16 = 48 bytes) of LDS space.
See Figure 10.3 in the GCN3 ISA manual.
These limit occupancy the same way as other stages' LDS usage does.
Note that PS can request additional LDS space via EXTRA_LDS_SIZE,
so that also must be taken into account here.
No Fossil DB changes.
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/12517 >
2021-08-27 16:41:08 +00:00
5a536eca9c
aco: calculate correct register demand for branch instructions
...
Since copies for the successor's linear phis are inserted before the
branch, we should consider the definitions and operands of the successor's
linear phis.
Fixes a Detroit: Become Human spilling failure with GCM+GVN.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12035 >
2021-08-05 12:01:58 +00:00
71aab9607d
aco/live_var_analysis: change worklist to a single integer
...
Reduces overall compile times by ~0.45%.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11879 >
2021-07-14 18:10:56 +02:00
1e2639026f
aco: Format.
...
Manually adjusted some comments for more intuitive line breaks.
Reviewed-by: Tony Wasserka <tony.wasserka@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11258 >
2021-07-12 21:27:31 +00:00
3f9e986d33
aco: add missing Licenses and remove Authors from files
...
Reviewed-by: Tony Wasserka <tony.wasserka@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11271 >
2021-07-12 12:09:31 +00:00
776ba40115
aco: add and use Program::progress
...
This is used when printing the program and to avoid updating register
demand during post-RA liveness analysis.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10315 >
2021-04-21 11:09:33 +00:00
655ba1e3a9
aco: don't update register demand during RA validation
...
It isn't intended to be accurate after RA, so num_waves can become zero,
breaking the sgpr_limit calculation.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10315 >
2021-04-21 11:09:33 +00:00
d75c73e6a6
aco: fix kill flags on phi operands
...
Fossil-db changes are likely due to how the CSSA pass works.
Totals from 1782 (1.31% of 136546) affected shaders (Raven):
CodeSize: 25333292 -> 25294020 (-0.16%); split: -0.16%, +0.00%
Instrs: 4916059 -> 4908218 (-0.16%); split: -0.16%, +0.00%
Latency: 282860167 -> 282707176 (-0.05%); split: -0.08%, +0.03%
InvThroughput: 136487564 -> 136394958 (-0.07%); split: -0.12%, +0.05%
VClause: 74791 -> 74795 (+0.01%)
Copies: 542115 -> 534280 (-1.45%); split: -1.48%, +0.04%
Branches: 168977 -> 168966 (-0.01%); split: -0.01%, +0.01%
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9196 >
2021-04-13 18:40:57 +00:00
3d4c13f3b8
aco: add DeviceInfo
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8761 >
2021-02-15 13:44:22 +00:00
b759557cac
aco: consider that GFX10.3 allocates LDS in 1024 byte blocks
...
fossil-db (GFX10.3):
Totals from 3 (0.00% of 139391) affected shaders:
VMEM: 513 -> 511 (-0.39%)
SMEM: 94 -> 92 (-2.13%)
VClause: 31 -> 30 (-3.23%)
fossil-db (GFX10.3, wave32):
Totals from 4 (0.00% of 139391) affected shaders:
VClause: 82 -> 81 (-1.22%)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8761 >
2021-02-15 13:35:38 +00:00
f520f4c299
aco: add Program::wgp_mode
...
Instead of assuming WGP mode on GFX10+ in different places, add a member
to Program that can be used instead.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8761 >
2021-02-15 13:35:14 +00:00
592d64611c
aco: fix waves calculation for wave32
...
fossil-db (GFX10.3, wave32):
Totals from 176 (0.13% of 139391) affected shaders:
SGPRs: 16648 -> 16640 (-0.05%)
VGPRs: 18920 -> 19076 (+0.82%); split: -0.30%, +1.12%
CodeSize: 2354172 -> 2354288 (+0.00%); split: -0.01%, +0.01%
MaxWaves: 1618 -> 1627 (+0.56%); split: +0.68%, -0.12%
Instrs: 435756 -> 435761 (+0.00%); split: -0.02%, +0.02%
Cycles: 8858360 -> 8869960 (+0.13%); split: -0.01%, +0.14%
VMEM: 55899 -> 57220 (+2.36%); split: +2.53%, -0.17%
SMEM: 10323 -> 10374 (+0.49%); split: +0.73%, -0.23%
VClause: 8307 -> 8290 (-0.20%); split: -0.24%, +0.04%
SClause: 16573 -> 16577 (+0.02%); split: -0.01%, +0.03%
Copies: 24641 -> 24652 (+0.04%); split: -0.24%, +0.28%
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8761 >
2021-02-15 13:35:13 +00:00
5d7b3bf1a7
aco: handle non-temp phi definitions and operands
...
This will be necessary as we make exec non-temp.
No fossil-db changes.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8870 >
2021-02-12 22:41:31 +00:00
44a76ba16d
aco: use VCC as regular SGPR pair on GFX10
...
There is no need to reserve it for special purposes, only.
Totals from 139391 (100.00% of 139391) affected shaders (Navi10):
VGPRs: 4738296 -> 4738156 (-0.00%); split: -0.01%, +0.00%
SpillSGPRs: 16188 -> 14968 (-7.54%); split: -7.60%, +0.06%
CodeSize: 294204472 -> 294118048 (-0.03%); split: -0.04%, +0.01%
MaxWaves: 2119584 -> 2119619 (+0.00%); split: +0.00%, -0.00%
Instrs: 56075079 -> 56056235 (-0.03%); split: -0.05%, +0.01%
Cycles: 1757781564 -> 1755354032 (-0.14%); split: -0.16%, +0.02%
VMEM: 52995887 -> 52996319 (+0.00%); split: +0.07%, -0.07%
SMEM: 9005338 -> 9004858 (-0.01%); split: +0.16%, -0.17%
VClause: 1178436 -> 1178331 (-0.01%); split: -0.02%, +0.01%
SClause: 2403649 -> 2404542 (+0.04%); split: -0.14%, +0.18%
Copies: 3447073 -> 3432417 (-0.43%); split: -0.66%, +0.23%
Branches: 1166542 -> 1166422 (-0.01%); split: -0.11%, +0.10%
PreSGPRs: 4229322 -> 4235538 (+0.15%)
PreVGPRs: 3817111 -> 3817040 (-0.00%)
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8921 >
2021-02-12 19:00:18 +00:00
947bf0bd67
aco: don't decrease the vgpr_limit when encountering bpermute
...
Instead we recalculate vgpr_limit on demand, depending on
the number of needed shared VGPRs.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8921 >
2021-02-12 19:00:18 +00:00
b98a4d4dd7
aco: refactor GPR limit calculation
...
This patch delays the calculation of GPR limits in order to
precisely incorporate extra registers (VCC etc.) and shared VGPRs.
Additionally, the allocation granularity is used to set the config.
This has some effect on the reported SGPR stats.
Totals (Navi10):
SGPRs: 6971787 -> 17753642 (+154.65%)
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8921 >
2021-02-12 19:00:18 +00:00
eaf681724e
aco: change gpr_alloc_granule to full alignment
...
This also switches the alloc_granule of Tonga and Iceland
to 96, so that the calculation is consistent.
Also changes the granularity for RDNA to 16 to keep
better stats with the upcoming patch.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8921 >
2021-02-12 19:00:18 +00:00
d1f93261b1
aco: always set exec_live=false
...
Register demand calculation for exec masks doesn't always match
get_live_changes() and get_temp_registers(). For now, just set
exec_live=false.
fossil-db (GFX10.3):
Totals from 108230 (77.64% of 139391) affected shaders:
SGPRs: 5759658 -> 5756818 (-0.05%); split: -0.08%, +0.03%
VGPRs: 4061104 -> 4061248 (+0.00%); split: -0.00%, +0.01%
SpillSGPRs: 14114 -> 15198 (+7.68%); split: -0.10%, +7.78%
CodeSize: 266548396 -> 266603288 (+0.02%); split: -0.01%, +0.03%
MaxWaves: 1390885 -> 1390855 (-0.00%); split: +0.00%, -0.00%
Instrs: 50983353 -> 50992972 (+0.02%); split: -0.02%, +0.04%
Cycles: 1733042048 -> 1735443264 (+0.14%); split: -0.02%, +0.16%
VMEM: 41933625 -> 41914722 (-0.05%); split: +0.04%, -0.09%
SMEM: 7197675 -> 7197789 (+0.00%); split: +0.16%, -0.16%
VClause: 1050885 -> 1050978 (+0.01%); split: -0.02%, +0.03%
SClause: 2074913 -> 2071844 (-0.15%); split: -0.23%, +0.08%
Copies: 3181464 -> 3188125 (+0.21%); split: -0.38%, +0.59%
Branches: 1127526 -> 1127716 (+0.02%); split: -0.10%, +0.12%
PreSGPRs: 3376687 -> 3586076 (+6.20%); split: -0.00%, +6.20%
PreVGPRs: 3339740 -> 3339811 (+0.00%)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8807 >
2021-02-05 17:03:17 +00:00
489aa8c7cb
aco: fix num_waves on GFX10+
...
There are half the SIMDs per CU and physical_vgprs should be 512 instead
of 256.
fossil-db (GFX10.3):
Totals from 3622 (2.60% of 139391) affected shaders:
VGPRs: 298192 -> 289732 (-2.84%); split: -3.43%, +0.59%
CodeSize: 29443432 -> 29458388 (+0.05%); split: -0.00%, +0.06%
MaxWaves: 21703 -> 23395 (+7.80%); split: +7.84%, -0.05%
Instrs: 5677920 -> 5681438 (+0.06%); split: -0.01%, +0.07%
Cycles: 280715524 -> 280895676 (+0.06%); split: -0.00%, +0.07%
VMEM: 981142 -> 981894 (+0.08%); split: +0.18%, -0.10%
SMEM: 243315 -> 243454 (+0.06%); split: +0.07%, -0.02%
VClause: 88991 -> 89767 (+0.87%); split: -0.02%, +0.89%
SClause: 200660 -> 200659 (-0.00%); split: -0.00%, +0.00%
Copies: 430729 -> 434160 (+0.80%); split: -0.07%, +0.86%
Branches: 158004 -> 158021 (+0.01%); split: -0.01%, +0.02%
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8523 >
2021-01-20 16:46:54 +00:00
97afb2a0a9
aco: remove unused radv_shader.h includes
...
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/7061 >
2020-10-14 15:09:34 +00:00
408195ec53
aco: remove useless occurences of radv_nir_compiler_options
...
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/7061 >
2020-10-14 15:09:34 +00:00
d2c18b7bf3
aco: use bit vectors for liveness sets
...
This seems to be much faster than hash sets. When compiling pipelines from
5 games, live_var_analysis takes about a third the time it used to and
fossilize-replay is ~1.77% faster.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Tony Wasserka <tony.wasserka@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6733 >
2020-09-21 13:47:28 +00:00
c2b1978aa4
aco: rework the way various compilation/validation errors are reported
...
The upcoming change will allow to report all ACO errors (or warnings)
directly to the app via VK_EXT_debug_report. This is similar to what
we already do for reporting various SPIRV->NIR errors.
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/6318 >
2020-08-20 08:15:06 +02:00
37988b5b8e
aco: fix max_waves_per_simd on Polaris, VegaM and GFX10.3
...
fossil-db (Polaris):
Totals from 20263 (14.75% of 137414) affected shaders:
SGPRs: 871407 -> 871679 (+0.03%); split: -0.00%, +0.03%
VGPRs: 513828 -> 550028 (+7.05%); split: -1.68%, +8.72%
CodeSize: 18869680 -> 18828148 (-0.22%); split: -0.23%, +0.01%
MaxWaves: 162012 -> 162030 (+0.01%); split: +0.01%, -0.00%
Instrs: 3629172 -> 3618817 (-0.29%); split: -0.30%, +0.02%
Cycles: 15682244 -> 15638244 (-0.28%); split: -0.30%, +0.02%
VMEM: 10675942 -> 10673344 (-0.02%); split: +0.18%, -0.21%
SMEM: 1209717 -> 1206088 (-0.30%); split: +0.03%, -0.33%
VClause: 81780 -> 81227 (-0.68%); split: -0.73%, +0.06%
SClause: 231724 -> 231561 (-0.07%); split: -0.07%, +0.00%
Copies: 187126 -> 180831 (-3.36%); split: -3.62%, +0.26%
Branches: 26841 -> 26837 (-0.01%); split: -0.03%, +0.01%
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5546 >
2020-08-04 20:39:33 +01:00
2ae27b96ef
aco: change live_out variables to std::unordered_set
...
Improves performance of live_var_analysis for larger shaders
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4130 >
2020-04-09 15:08:57 +00:00
b66f474121
aco: improve speed of live_var_analysis
...
by merging live_sgprs and live_vgprs sets.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4130 >
2020-04-09 15:08:57 +00:00
2f424c83e0
aco: only break SMEM clauses if XNACK is enabled (mostly APUs)
...
According to LLVM, it seems only required for APUs like RAVEN, but
we still ensure that SMEM stores are in their own clause.
pipeline-db (VEGA10):
Totals from affected shaders:
SGPRS: 1775364 -> 1775364 (0.00 %)
VGPRS: 1287176 -> 1287176 (0.00 %)
Spilled SGPRs: 725 -> 725 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 65386620 -> 65107460 (-0.43 %) bytes
Max Waves: 287099 -> 287099 (0.00 %)
pipeline-db (POLARIS10):
Totals from affected shaders:
SGPRS: 1797743 -> 1797743 (0.00 %)
VGPRS: 1271108 -> 1271108 (0.00 %)
Spilled SGPRs: 730 -> 730 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 64046244 -> 63782324 (-0.41 %) bytes
Max Waves: 254875 -> 254875 (0.00 %)
This only affects GFX6-GFX9 chips because the compiler uses a
different pass for GFX10.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4349 >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4349 >
2020-04-01 17:50:31 +00:00
0f35b3795d
aco: Fix workgroup size calculation.
...
Clear the workgroup size for all supported shader stages.
Also, unify the workgroup size calculation accross various places.
As a result, insert_waitcnt can use the proper workgroup size
which means that some waits can be dropped from tessellation
shaders. Also, in cases where the previous calculation was wrong,
we now insert s_barrier instructions.
Totals from affected shaders (GFX10):
Code Size: 340116 -> 338484 (-0.48 %) bytes
Fixes: a8d15ab6dahttps://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4165 >
2020-03-30 13:09:08 +00:00
1872759f55
aco: add a late kill flag
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3914 >
2020-03-16 16:09:02 +00:00
c51348bd9b
aco: move some register demand helpers into aco_live_var_analysis.cpp
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3914 >
2020-03-16 16:09:02 +00:00
b088a4b113
aco: only reserve sgprs for vcc if it's used
...
pipeline-db (Vega):
Totals:
SGPRS: 5186302 -> 5075616 (-2.13 %)
VGPRS: 3704580 -> 3704580 (0.00 %)
Spilled SGPRs: 144859 -> 144859 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Scratch size: 4124 -> 4124 (0.00 %) dwords per thread
Code Size: 247315944 -> 247315944 (0.00 %) bytes
LDS: 1311 -> 1311 (0.00 %) blocks
Max Waves: 674560 -> 674562 (0.00 %)
Totals from affected shaders:
SGPRS: 536992 -> 426306 (-20.61 %)
VGPRS: 356404 -> 356404 (0.00 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size: 8498748 -> 8498748 (0.00 %) bytes
LDS: 8 -> 8 (0.00 %) blocks
Max Waves: 113832 -> 113834 (0.00 %)
There are some small code size changes in a few RotTR shaders and a small
increase in max_waves in two Detroit: Become Human shaders.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3906 >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3906 >
2020-03-05 20:18:34 +00:00
4e83e05e62
aco: error when block has no logical preds but VGPRs are live at the start
...
This would have caught the liveness error fixed in the previous commit.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3257 >
2020-01-29 18:02:27 +00:00
3f96a1ed86
aco: fix operand kill flags when a temporary is used more than once
...
Helps create v_mac_f32 from v_mad_f32(b, a, b)
Totals from affected shaders:
SGPRS: 35824 -> 35824 (0.00 %)
VGPRS: 33460 -> 33456 (-0.01 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size: 2187264 -> 2180976 (-0.29 %) bytes
LDS: 127 -> 127 (0.00 %) blocks
Max Waves: 3802 -> 3802 (0.00 %)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3486 >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3486 >
2020-01-22 15:55:00 +00:00
b5c9688516
aco: limit register usage for large work groups
...
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
2020-01-10 12:10:37 +00:00
afe1a8ff5b
aco: fix vgpr alloc granule with wave32
...
We still need to increase the number of physical vgprs
Totals from affected shaders:
SGPRS: 671976 -> 675288 (0.49 %)
VGPRS: 550112 -> 562596 (2.27 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Code Size: 27621660 -> 27606532 (-0.05 %) bytes
Max Waves: 81083 -> 87833 (8.32 %)
Instructions: 5391560 -> 5389031 (-0.05 %)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
2019-12-21 12:38:42 +01:00
e0bcefc3a0
aco/wave32: Use lane mask regclass for exec/vcc.
...
Currently all usages of exec and vcc are hardcoded to use s2 regclass.
This commit makes it possible to use s1 in wave32 mode and
s2 in wave64 mode.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
2019-12-04 10:36:01 +00:00
78bca0d0ce
aco: improve live variable analysis
...
This patch makes the live variable analysis more precise
w.r.t. killed phi operands and the block's register pressure.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
2019-10-30 19:48:32 +00:00
c52ebbcea4
aco: Introduce vgpr_limit to keep track of available VGPRs.
...
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
2019-10-28 23:52:50 +00:00
fc04a2fc31
aco: take LDS into account when calculating num_waves
...
pipeline-db (Vega):
SGPRS: 344 -> 344 (0.00 %)
VGPRS: 424 -> 524 (23.58 %)
Spilled SGPRs: 84 -> 80 (-4.76 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size: 52812 -> 52484 (-0.62 %) bytes
LDS: 135 -> 135 (0.00 %) blocks
Max Waves: 56 -> 53 (-5.36 %)
v2: consider WGP, rework to be clearer and apply the
"maximum 16 workgroups per CU" limit properly
v2: use "SIMD" instead of "EU"
v2: fix spiller by introducing "Program::max_waves"
v2: rename "lds_size" to "lds_limit"
v3: make max_waves actually independant of register usage
v3: fix issue where max_waves was way too high
v3: use DIV_ROUND_UP(a, b) instead of max(a / b, 1)
v3: rename "workgroups_per_cu" to "workgroups_per_cu_wgp"
v4: fix typo from "workgroups_per_cu" rename
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev> (v3)
2019-10-23 19:11:21 +01:00
08d510010b
aco: increase accuracy of SGPR limits
...
SGPRs are allocated in groups of 16 on GFX8/GFX9. GFX10 allocates a fixed
number of SGPRs and has 106 addressable SGPRs.
pipeline-db (Vega):
SGPRS: 5912 -> 6232 (5.41 %)
VGPRS: 1772 -> 1780 (0.45 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size: 88228 -> 87904 (-0.37 %) bytes
LDS: 0 -> 0 (0.00 %) blocks
Max Waves: 559 -> 571 (2.15 %)
piepline-db (Navi):
SGPRS: 341256 -> 363384 (6.48 %)
VGPRS: 171536 -> 170960 (-0.34 %)
Spilled SGPRs: 832 -> 581 (-30.17 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size: 14207332 -> 14190872 (-0.12 %) bytes
LDS: 33 -> 33 (0.00 %) blocks
Max Waves: 18072 -> 18251 (0.99 %)
v2: unconditionally count vcc as an extra sgpr on GFX10+
v3: pass SGPRs rounded to 8
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
2019-10-23 19:11:21 +01:00
93c8ebfa78
aco: Initial commit of independent AMD compiler
...
ACO (short for AMD Compiler) is a new compiler backend with the goal to replace
LLVM for Radeon hardware for the RADV driver.
ACO currently supports only VS, PS and CS on VI and Vega.
There are some optimizations missing because of unmerged NIR changes
which may decrease performance.
Full commit history can be found at
https://github.com/daniel-schuermann/mesa/commits/backend
Co-authored-by: Daniel Schürmann <daniel@schuermann.dev>
Co-authored-by: Rhys Perry <pendingchaos02@gmail.com>
Co-authored-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Co-authored-by: Connor Abbott <cwabbott0@gmail.com>
Co-authored-by: Michael Schellenberger Costa <mschellenbergercosta@googlemail.com>
Co-authored-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-09-19 12:10:00 +02:00
Rhys Perry
Marek Olšák
Dave Airlie
Daniel Schürmann
Timur Kristóf
Samuel Pitoiset