fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2025-12-21 11:30:11 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 1
165299 commits 64 branches 914 tags 1 GiB
Commit graph

287 commits

Author SHA1 Message Date
Rhys Perry
254b178d5b aco: disallow SGPRS/constants with interpolation instructions 
https://reviews.llvm.org/D137575

The VINTRP format cannot encode anything except VGPRs.

Reading VINTERPInstructions.td, looks like it's the same for GFX11.

No fossil-db changes.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20251>
 2023-01-10 16:01:38 +00:00
Timur Kristóf
db5c3f170f aco: Emulate Wave64 bpermute on GFX11. 
Similar to emit_gfx10_wave64_bpermute, but uses the new
v_permlane64_b32 instruction to swap data between wave halves.

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/20293>
 2022-12-14 13:54:04 +00:00
Timur Kristóf
640e801651 aco: Split opcodes for GFX6 and GFX10 emulated bpermute. 
Different sequences are emitted for these, so it makes sense to
have different opcodes too.

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/20293>
 2022-12-14 13:54:04 +00:00
Timur Kristóf
614348f28b aco: Don't accept constants on p_bpermute. 
The sequence emitted for this pseudo instruction is not ready
to handle constants or literals at all.

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/20293>
 2022-12-14 13:54:04 +00:00
Rhys Perry
381de3c809 aco: more carefully apply constant offsets into scratch accesses 
Death stranding does scratch_arr[80-idx]. This doesn't seem to work if we
try to combine the subtraction into the access.

fossil-db (navi21):
Totals from 52 (0.04% of 135636) affected shaders:
Instrs: 78560 -> 79036 (+0.61%)
CodeSize: 427940 -> 431188 (+0.76%)
Latency: 1313809 -> 1318142 (+0.33%)
InvThroughput: 292833 -> 293842 (+0.34%)
VClause: 2361 -> 2555 (+8.22%); split: -0.51%, +8.73%
Copies: 8767 -> 8746 (-0.24%); split: -0.35%, +0.11%

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Fixes: 0e783d687a ("aco: use scratch_* for scratch load/store on GFX9+")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7735
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20117>
 2022-12-06 15:23:38 +00:00
Rhys Perry
917cfd587c aco: use v_minmax/v_maxmin opcodes 
fossil-db (gfx1100):
Totals from 29868 (22.12% of 135032) affected shaders:
MaxWaves: 741336 -> 741344 (+0.00%)
Instrs: 34624902 -> 34539766 (-0.25%); split: -0.25%, +0.00%
CodeSize: 187196804 -> 187192100 (-0.00%); split: -0.01%, +0.01%
VGPRs: 1816860 -> 1816788 (-0.00%); split: -0.01%, +0.01%
Latency: 502597202 -> 502245627 (-0.07%); split: -0.08%, +0.01%
InvThroughput: 84813176 -> 84586122 (-0.27%); split: -0.28%, +0.01%
VClause: 633826 -> 633749 (-0.01%); split: -0.02%, +0.01%
SClause: 1317738 -> 1317047 (-0.05%); split: -0.06%, +0.01%
Copies: 2130610 -> 2130954 (+0.02%); split: -0.03%, +0.05%
Branches: 766093 -> 765969 (-0.02%); split: -0.02%, +0.00%
PreSGPRs: 1630250 -> 1630034 (-0.01%); split: -0.02%, +0.00%
PreVGPRs: 1590777 -> 1590664 (-0.01%); split: -0.01%, +0.00%

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/19933>
 2022-12-01 21:43:28 +00:00
Rhys Perry
dfbc8e0192 aco: change order in combine_minmax() 
Prepare for future optimizations.

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/19933>
 2022-12-01 21:43:28 +00:00
Georg Lehmann
7aa94efe82 aco: Combine constant bit test to s_bitcmp. 
Foz-DB Navi21:
Totals from 73988 (54.84% of 134913) affected shaders:
VGPRs: 2959768 -> 2959752 (-0.00%)
SpillSGPRs: 10250 -> 10697 (+4.36%); split: -0.64%, +5.00%
SpillVGPRs: 2326 -> 2291 (-1.50%); split: -2.24%, +0.73%
CodeSize: 261339476 -> 261045912 (-0.11%); split: -0.12%, +0.00%
Scratch: 239616 -> 238592 (-0.43%)
Instrs: 49214044 -> 49188242 (-0.05%); split: -0.06%, +0.00%
Latency: 413214139 -> 413296229 (+0.02%); split: -0.03%, +0.05%
InvThroughput: 71741622 -> 71786300 (+0.06%); split: -0.07%, +0.13%
VClause: 856838 -> 856973 (+0.02%); split: -0.01%, +0.02%
SClause: 1504502 -> 1504567 (+0.00%); split: -0.01%, +0.02%
Copies: 4058433 -> 4060424 (+0.05%); split: -0.03%, +0.08%
Branches: 1502953 -> 1502945 (-0.00%); split: -0.00%, +0.00%
PreSGPRs: 3081927 -> 3081531 (-0.01%); split: -0.02%, +0.01%
PreVGPRs: 2513990 -> 2513992 (+0.00%)

The vast majority of instruction count regressions are caused by parallel-rdp.

Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18870>
 2022-11-28 18:43:53 +00:00
Georg Lehmann
73be938c48 aco: Combine bit test to s_bitcmp. 
Foz-DB Navi21:
Totals from 6396 (4.74% of 134913) affected shaders:
VGPRs: 483280 -> 483152 (-0.03%); split: -0.03%, +0.01%
SpillSGPRs: 8119 -> 7941 (-2.19%)
CodeSize: 63377880 -> 63268556 (-0.17%); split: -0.20%, +0.03%
MaxWaves: 86778 -> 86810 (+0.04%)
Instrs: 11745621 -> 11725857 (-0.17%); split: -0.20%, +0.03%
Latency: 162400148 -> 162282230 (-0.07%); split: -0.08%, +0.01%
InvThroughput: 29179429 -> 29133173 (-0.16%); split: -0.16%, +0.00%
VClause: 208032 -> 208100 (+0.03%); split: -0.01%, +0.05%
SClause: 431390 -> 430849 (-0.13%); split: -0.24%, +0.11%
Copies: 896222 -> 893285 (-0.33%); split: -0.62%, +0.30%
Branches: 349806 -> 348770 (-0.30%); split: -0.90%, +0.60%
PreSGPRs: 618908 -> 613773 (-0.83%); split: -0.83%, +0.00%
PreVGPRs: 482901 -> 482893 (-0.00%)

Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18870>
 2022-11-28 18:43:53 +00:00
Georg Lehmann
853d2cb6f1 aco: Combine s_abs and s_sub/s_add to s_absdiff. 
Totals from 2 (0.00% of 134913) affected shaders:
CodeSize: 1344 -> 1336 (-0.60%)
Instrs: 277 -> 275 (-0.72%)

Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18870>
 2022-11-28 18:43:53 +00:00
Georg Lehmann
7e1d77fd90 aco: Ignore instructions with exec operands in follow_operand. 
No Foz-DB changes.

Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18870>
 2022-11-28 18:43:53 +00:00
Georg Lehmann
65a3328b4c aco/optimizer: Cleanup ctx.uses handling for patterns which use follow_operand(..., true). 
No Foz-DB changes.

Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18870>
 2022-11-28 18:43:53 +00:00
Yonggang Luo
cdbe1ad570 aco: Fixes -Werror,-Wbitwise-instead-of-logical for clang-15 in aco_optimizer.cpp 
error message:
error: use of bitwise '|' with boolean operands [-Werror,-Wbitwise-instead-of-logical]

Signed-off-by: Yonggang Luo <luoyonggang@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19969>
 2022-11-24 02:56:03 +00:00
Samuel Pitoiset
bb90d29660 aco: add p_dual_src_export_gfx11 for dual source blending on GFX11 
Dual source blending must be in strict WQM mode.

Cc: 22.3 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/19643>
 2022-11-16 18:35:10 +00:00
Georg Lehmann
6d02054047 aco: Combine v_cvt_u32_f32 with insert to v_cvt_pk_u8_f32. 
No Foz-DB difference on Navi21.

Foz-DB GFX11:
Totals from 746 (0.55% of 134913) affected shaders:
CodeSize: 8430248 -> 8416128 (-0.17%); split: -0.17%, +0.00%
Instrs: 1617202 -> 1614707 (-0.15%)
Latency: 13943398 -> 13934161 (-0.07%); split: -0.07%, +0.00%
InvThroughput: 2601620 -> 2596624 (-0.19%); split: -0.20%, +0.01%
Copies: 114346 -> 114334 (-0.01%); split: -0.01%, +0.00%
PreVGPRs: 48314 -> 48312 (-0.00%)

Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18492>
 2022-11-16 16:49:04 +00:00
Rhys Perry
6113ee650a aco/gfx11: fix FS input loads in quad-divergent control flow 
This is not ideal and it would be great to somehow make it better some
day.

fossil-db (gfx1100):
Totals from 5208 (3.86% of 135032) affected shaders:
MaxWaves: 127058 -> 126962 (-0.08%); split: +0.01%, -0.09%
Instrs: 3983440 -> 4072736 (+2.24%); split: -0.00%, +2.24%
CodeSize: 21872468 -> 22230852 (+1.64%); split: -0.00%, +1.64%
VGPRs: 206688 -> 206984 (+0.14%); split: -0.05%, +0.20%
Latency: 37447383 -> 37491197 (+0.12%); split: -0.05%, +0.17%
InvThroughput: 6421955 -> 6422348 (+0.01%); split: -0.03%, +0.03%
VClause: 71579 -> 71545 (-0.05%); split: -0.09%, +0.04%
SClause: 148289 -> 147146 (-0.77%); split: -0.84%, +0.07%
Copies: 259011 -> 258084 (-0.36%); split: -0.61%, +0.25%
Branches: 101366 -> 101314 (-0.05%); split: -0.10%, +0.05%
PreSGPRs: 223482 -> 223460 (-0.01%); split: -0.21%, +0.20%
PreVGPRs: 184448 -> 184744 (+0.16%)

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/19370>
 2022-11-01 12:42:43 +00:00
Georg Lehmann
2eac571d61 aco: Use opsel for the third operand. 
Foz-DB Navi21:
Totals from 2 (0.00% of 134913) affected shaders:
CodeSize: 7788 -> 7772 (-0.21%)
Instrs: 1305 -> 1303 (-0.15%)
Latency: 7175 -> 7163 (-0.17%)
InvThroughput: 2082 -> 2078 (-0.19%)
Copies: 57 -> 55 (-3.51%)

Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19380>
 2022-10-31 09:54:01 +00:00
Georg Lehmann
361b47b1f0 aco: Implement signed idot instructions on GFX11. 
Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19114>
 2022-10-24 19:07:16 +00:00
Georg Lehmann
616d3908dc aco: Don't use opsel for p_insert. 
This doesn't make sense, opsel preserves the not selected half of the register,
p_insert zeros it.

No Foz-DB changes.

Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>

Fixes: 54292e99c7 ("aco: optimize 32-bit extracts and inserts using SDWA")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19253>
 2022-10-24 18:40:40 +00:00
Timur Kristóf
dd90273aaa aco: Optimize MUBUF 0 offset when idxen is also being used. 
Now that we added an index src to the NIR intrinsic, it can
happen that these generate MUBUF instructions which have both
an index and an offset.

Extend this ACO optimization to the case when idxen is used.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17551>
 2022-10-20 20:00:50 +00:00
Rhys Perry
7cecc81683 aco/gfx11: fix s_waitcnt printing 
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/17710>
 2022-09-30 20:57:02 +00:00
Daniel Schürmann
cd36a29759 aco/optimizer: change inverse_comparison in-place 
This avoids creating a second comparison which is more expensive
than just using the existing s_not.

Totals from 1089 (0.81% of 134913) affected shaders: (GFX10.3)
VGPRs: 50472 -> 50184 (-0.57%)
CodeSize: 4724692 -> 4760824 (+0.76%); split: -0.03%, +0.79%
MaxWaves: 23964 -> 24012 (+0.20%)
Instrs: 859588 -> 859687 (+0.01%); split: -0.11%, +0.12%
Latency: 10674653 -> 10650353 (-0.23%); split: -0.41%, +0.18%
InvThroughput: 1752987 -> 1750238 (-0.16%); split: -0.20%, +0.04%
VClause: 20921 -> 20872 (-0.23%); split: -0.68%, +0.45%
SClause: 31417 -> 31550 (+0.42%)
Copies: 69428 -> 68738 (-0.99%); split: -1.52%, +0.53%
PreSGPRs: 48033 -> 49649 (+3.36%)
PreVGPRs: 44490 -> 43699 (-1.78%)

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18253>
 2022-09-22 12:35:11 +00:00
Timur Kristóf
c8445c1691 aco: Change inverse-comparison optimization to work with s_not 
Some time ago we stopped using s_andn2 with exec for boolean NOT.
The reasoning behind that change was that those booleans will be
always ANDed with exec when necessary.

This inhibited the inverse-comparison optimization in most cases
which is fixed by this patch.

Fossil DB stats on Navi 21:

Totals from 12251 (9.08% of 134913) affected shaders:
VGPRs: 801744 -> 802016 (+0.03%); split: -0.00%, +0.04%
SpillSGPRs: 8863 -> 8893 (+0.34%)
CodeSize: 100593244 -> 100370684 (-0.22%); split: -0.22%, +0.00%
MaxWaves: 204994 -> 204948 (-0.02%); split: +0.00%, -0.02%
Instrs: 18717001 -> 18668965 (-0.26%); split: -0.26%, +0.00%
Latency: 263255046 -> 262874896 (-0.14%); split: -0.16%, +0.02%
InvThroughput: 52760249 -> 52721736 (-0.07%); split: -0.08%, +0.01%
VClause: 329631 -> 329680 (+0.01%); split: -0.03%, +0.04%
SClause: 681563 -> 681435 (-0.02%); split: -0.02%, +0.00%
Copies: 1331612 -> 1372446 (+3.07%); split: -0.03%, +3.10%
Branches: 548325 -> 548301 (-0.00%)
PreSGPRs: 911317 -> 909700 (-0.18%)
PreVGPRs: 766279 -> 767070 (+0.10%)

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18253>
 2022-09-22 12:35:11 +00:00
Daniel Schürmann
cf5f9854bc aco/optimizer: optimize s_and(exec, s_and(x, y)) more aggressively 
It is enough if either of the two operands is respecting the
current exec mask.

Totals from 1680 (1.25% of 134913) affected shaders: (GFX10.3)
CodeSize: 3929436 -> 3922372 (-0.18%); split: -0.18%, +0.00%
Instrs: 730305 -> 728536 (-0.24%); split: -0.24%, +0.00%
Latency: 6839314 -> 6835154 (-0.06%); split: -0.07%, +0.01%
InvThroughput: 1371351 -> 1371267 (-0.01%); split: -0.01%, +0.00%
SClause: 32819 -> 32802 (-0.05%); split: -0.09%, +0.04%
Copies: 33264 -> 33271 (+0.02%); split: -0.01%, +0.03%

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18253>
 2022-09-22 12:35:11 +00:00
Daniel Schürmann
79a8e8b5b2 aco/optimizer: do can_eliminate_and_exec() optimization later 
This will allow to optimize s_not(v_cmp()) safely.

Totals from 1024 (0.76% of 134913) affected shaders: (GFX10.3)
CodeSize: 5695424 -> 5701860 (+0.11%); split: -0.00%, +0.11%
Scratch: 242688 -> 241664 (-0.42%)
Instrs: 1040656 -> 1041635 (+0.09%); split: -0.00%, +0.09%
Latency: 16842282 -> 16922790 (+0.48%); split: -0.06%, +0.54%
InvThroughput: 4772728 -> 4810868 (+0.80%); split: -0.10%, +0.90%
VClause: 20013 -> 20000 (-0.06%); split: -0.12%, +0.05%
Copies: 115057 -> 114384 (-0.58%); split: -1.22%, +0.63%
Branches: 34531 -> 34532 (+0.00%); split: -0.00%, +0.01%
PreSGPRs: 46263 -> 46267 (+0.01%)

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18253>
 2022-09-22 12:35:11 +00:00
Georg Lehmann
84c0529258 aco: Unswizzle v_pk_fma_f16 literals to produce more v_pk_fmac_f16. 
No Foz-DB difference, but it reduces code size in some angle shaders.

Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18676>
 2022-09-21 23:04:06 +00:00
Daniel Schürmann
98e3c446d8 aco/optimizer: disallow can_eliminate_and_exec() with s_not 
Totals from 295 (0.22% of 134913) affected shaders: (GFX10.3)
CodeSize: 1016564 -> 1016896 (+0.03%); split: -0.05%, +0.09%
Instrs: 187659 -> 187724 (+0.03%); split: -0.08%, +0.11%
Latency: 2839516 -> 2839541 (+0.00%); split: -0.01%, +0.01%
Copies: 12301 -> 12305 (+0.03%); split: -0.01%, +0.04%
PreSGPRs: 10266 -> 10268 (+0.02%)

Closes: #7024
Cc: mesa-stable
Tested-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18722>
 2022-09-21 15:33:43 +00:00
Georg Lehmann
94b5225c9b aco: Use v_fmaak/v_fmamk if two operands are the same literal. 
Foz-DB Navi21:
Totals from 5744 (4.26% of 134913) affected shaders:
VGPRs: 237128 -> 237056 (-0.03%); split: -0.04%, +0.01%
CodeSize: 16654484 -> 16620668 (-0.20%); split: -0.23%, +0.03%
MaxWaves: 152838 -> 152846 (+0.01%)
Instrs: 3063214 -> 3058572 (-0.15%); split: -0.17%, +0.02%
Latency: 23935195 -> 23934827 (-0.00%); split: -0.03%, +0.03%
InvThroughput: 5478562 -> 5478160 (-0.01%); split: -0.01%, +0.01%
VClause: 60432 -> 60435 (+0.00%); split: -0.02%, +0.03%
SClause: 121032 -> 120896 (-0.11%); split: -0.20%, +0.09%
Copies: 147865 -> 143144 (-3.19%); split: -3.59%, +0.40%
PreSGPRs: 195722 -> 195661 (-0.03%); split: -0.06%, +0.03%
PreVGPRs: 182849 -> 182787 (-0.03%)

Foz-DB Vega10:
Totals from 5290 (3.92% of 135041) affected shaders:
SGPRs: 357952 -> 359616 (+0.46%); split: -0.11%, +0.57%
VGPRs: 204048 -> 203928 (-0.06%); split: -0.08%, +0.02%
CodeSize: 14043176 -> 14003100 (-0.29%); split: -0.29%, +0.00%
MaxWaves: 39401 -> 39398 (-0.01%); split: +0.01%, -0.02%
Instrs: 2636739 -> 2631246 (-0.21%); split: -0.21%, +0.00%
Latency: 25264088 -> 25256482 (-0.03%); split: -0.05%, +0.02%
InvThroughput: 12039643 -> 12039346 (-0.00%); split: -0.00%, +0.00%
VClause: 55603 -> 55584 (-0.03%); split: -0.04%, +0.00%
SClause: 101577 -> 101342 (-0.23%); split: -0.30%, +0.07%
Copies: 213344 -> 207929 (-2.54%); split: -2.58%, +0.05%
Branches: 34053 -> 34054 (+0.00%)
PreSGPRs: 172405 -> 172260 (-0.08%)

Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18645>
 2022-09-21 12:16:47 +00:00
Daniel Schürmann
3d6ea4f666 aco: use std::vector::reserve() more often 
This removes the majority of vector re-allocations.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18105>
 2022-08-30 16:03:26 +00:00
Rhys Perry
f60cb8d0af aco: rename is_cmp to is_fp_cmp 
The old name is no longer accurate.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18077>
 2022-08-16 17:31:33 +00:00
Daniel Schürmann
66d46a23fb aco: fix packed 16bit fneg/fsat optimization 
Make sure that the Operand is '1.0.xx'.

Fixes: b03be30e07 ('aco: optimize packed fneg')
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17395>
 2022-07-11 10:10:00 +00:00
Rhys Perry
6642f2fd74 aco: handle subtractions in parse_base_offset 
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
Rhys Perry
52934f6cdb aco: combine additions and constants into scratch load/store 
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
Georg Lehmann
68db0a079b aco: Fix swapping sources in SOPC -> SOPK optimization. 
Fixes: 2d6b0a4177 ("aco/optimizer: Optimize SOPC with literal to SOPK.")

Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17407>
 2022-07-08 09:43:51 +00:00
Georg Lehmann
2d6b0a4177 aco/optimizer: Optimize SOPC with literal to SOPK. 
Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15999>
 2022-07-06 09:54:54 +00:00
Georg Lehmann
52f8167b25 aco/optimizer: Convert s_add_u32 with literals to s_add_i32 if carry is not used. 
To allow further optimizations to s_addk_i32.

Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15999>
 2022-07-06 09:54:54 +00:00
Rhys Perry
84b404d34d aco: don't use 32-bit fp inline constants for fp16 vop3p literals 
If we're applying the literal 0x3f800000 to a fp16 vop3p instruction, we
shouldn't use the 1.0 inline constant, because the hardware will use the
16-bit 1.0: 0x00003c00.

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/16296>
 2022-07-05 16:39:56 +00:00
Rhys Perry
994f9b5a39 aco: try sign-extending or shifting constants in propagate_constants_vop3p 
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/16296>
 2022-07-05 16:39:56 +00:00
Rhys Perry
33befb58b0 aco: fix redirect combine in propagate_constants_vop3p() with negatives 
This previously didn't correctly consider negative integers when bits=16
(which sign-extend) and would have combined 0xfffe0000.xy as -2.yx. Now it
combines 0xfffeffff.xy as that instead. It was also skipped when bits=32.

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/16296>
 2022-07-05 16:39:56 +00:00
Rhys Perry
fc39c3a0b1 aco: don't use opsel to fold constants into dot accumulation sources 
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/16296>
 2022-07-05 16:39:56 +00:00
Rhys Perry
ae74474509 aco: fix propagate_constants_vop3p with integer vop3p and 16-bit constants 
This would have created a 1.0.xx operand from 0x3c00.xx or 0x3c003c00.xy
for vop3p instructions which have 32-bit operands.

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/16296>
 2022-07-05 16:39:56 +00:00
Rhys Perry
5d8f5615d0 aco: ignore precise flag when optimizing integer clamps 
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/16296>
 2022-07-05 16:39:56 +00:00
Rhys Perry
61eb632775 aco: include _e64 variants of 16-bit min/max in minmax optimizations 
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/16296>
 2022-07-05 16:39:56 +00:00
Rhys Perry
f2a346eb40 aco: don't accept med3 opcodes in get_minmax_info() 
I don't think the presence of med3 here breaks anything, but it shouldn't
be here anyway.

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/16296>
 2022-07-05 16:39:56 +00:00
Rhys Perry
f937c5be7c aco: add and use constantValue16() 
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/16296>
 2022-07-05 16:39:56 +00:00
Rhys Perry
40db52488b aco: consider fma with multiplication by power-of-two unfused 
fossil-db (Sienna Cichlid):
Totals from 700 (0.43% of 162353) affected shaders:
MaxWaves: 18986 -> 18990 (+0.02%)
Instrs: 546475 -> 539729 (-1.23%); split: -1.24%, +0.00%
CodeSize: 2823716 -> 2808504 (-0.54%); split: -0.55%, +0.01%
VGPRs: 25304 -> 25288 (-0.06%)
Latency: 2180102 -> 2168187 (-0.55%); split: -0.55%, +0.01%
InvThroughput: 466223 -> 457326 (-1.91%)
VClause: 6768 -> 6797 (+0.43%); split: -0.01%, +0.44%
SClause: 12235 -> 12237 (+0.02%); split: -0.22%, +0.24%
Copies: 34498 -> 34479 (-0.06%); split: -0.21%, +0.15%
PreVGPRs: 20968 -> 20958 (-0.05%)

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/15862>
 2022-06-06 19:06:01 +00:00
Marek Olšák
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
Daniel Schürmann
d70688492c aco/optimizer: re-combine and copy-propagate p_create_vector(p_split_vector) 
Totals from 309 (0.23% of 134913) affected shaders: (GFX10.3)
CodeSize: 1853812 -> 1857732 (+0.21%); split: -0.05%, +0.27%
Instrs: 340810 -> 341789 (+0.29%); split: -0.07%, +0.36%
Latency: 3301814 -> 3301774 (-0.00%); split: -0.02%, +0.02%
InvThroughput: 590473 -> 590914 (+0.07%); split: -0.00%, +0.08%
Copies: 28751 -> 29731 (+3.41%); split: -0.87%, +4.28%
PreSGPRs: 14010 -> 14028 (+0.13%); split: -0.01%, +0.14%

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15414>
 2022-05-06 14:52:07 +00:00
Daniel Schürmann
12d7f911c9 aco/optimizer: prevent any overflow between SGPR and const offset on MUBUF 
Apparently, if the SGPR offset + const offset overflows,
it doesn't work.

Totals from 145 (0.11% of 134913) affected shaders: (GFX10.3)
SpillSGPRs: 134 -> 104 (-22.39%)
CodeSize: 1632676 -> 1645916 (+0.81%); split: -0.03%, +0.84%
Instrs: 316920 -> 320252 (+1.05%); split: -0.01%, +1.07%
Latency: 1456285 -> 1459686 (+0.23%); split: -0.02%, +0.25%
InvThroughput: 165785 -> 166086 (+0.18%); split: -0.02%, +0.20%
VClause: 6815 -> 6875 (+0.88%); split: -0.03%, +0.91%
SClause: 19089 -> 19079 (-0.05%); split: -0.06%, +0.01%
PreSGPRs: 7302 -> 7304 (+0.03%); split: -0.01%, +0.04%

Fixes: KHR-GL45.shader_storage_buffer_object.basic-operations-case1-cs
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15866>
 2022-04-29 15:57:57 +00:00
Rhys Perry
5aa5af7776 aco: handle read2st64/write2st64 in optimizer 
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/13778>
 2022-04-13 23:08:07 +00:00