fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2026-05-22 21:48:09 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 1
199018 commits 66 branches 936 tags 962 MiB
Commit graph

3472 commits

Author SHA1 Message Date
Samuel Pitoiset
1fa0fe1e0c aco: add support for the trap handler shader on GFX9-GFX10.3 
This has been tested on navi21.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31926>
 2024-11-04 10:48:52 +00:00
Samuel Pitoiset
281eb14df8 aco: fix reading registers from the trap handler shader 
It should read 32-bit values, otherwise some MSB are 0 and it's missing
some information.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31926>
 2024-11-04 10:48:52 +00:00
Samuel Pitoiset
49682fc0cb radv,aco: save SQ_WAVE_GPR_ALLOC from the trap handler 
This would be used to dump SGPRs.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31934>
 2024-11-01 15:40:25 +00:00
Daniel Schürmann
62715984f8 aco/README: add descriptions of recently added passes 
... and less recent ones.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31888>
 2024-10-30 09:23:54 +00:00
Daniel Schürmann
21ceeb22ed aco: move jump threading optimization into separate pass 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31888>
 2024-10-30 09:23:54 +00:00
Daniel Schürmann
87a3c08df1 aco/ssa_elimination: remove some redundant checks during jump threading 
Since phis got already lowered to parallelcopies by this point,
there is no need to cross-check.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31888>
 2024-10-30 09:23:54 +00:00
Daniel Schürmann
a6c38f706d aco/ssa_elimination: perform jump threading after parallelcopy insertion 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31888>
 2024-10-30 09:23:54 +00:00
Daniel Schürmann
10958d04d5 aco: Respect addressible SGPR limit in VS prologs 
On Tonga, the effective SGPR limit is 96, including VCC.

Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31859>
 2024-10-28 11:29:06 +00:00
Georg Lehmann
d01c1ba939 aco: move exec copy out of waterfall loops 
Foz-DB Navi21:
Totals from 348 (0.44% of 79395) affected shaders:
CodeSize: 17944800 -> 17946268 (+0.01%); split: -0.02%, +0.03%
Latency: 29775973 -> 29774369 (-0.01%); split: -0.01%, +0.00%
InvThroughput: 10233380 -> 10232801 (-0.01%); split: -0.01%, +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/19070>
 2024-10-25 16:47:32 +00:00
Georg Lehmann
6c73a8a7f2 aco: optimize conditional divergent breaks at the end of loops 
Removes one branch and one s_mov.

Foz-DB Navi21:
Totals from 1483 (1.87% of 79395) affected shaders:
Instrs: 6424114 -> 6373084 (-0.79%)
CodeSize: 35309320 -> 35091084 (-0.62%); split: -0.63%, +0.01%
Latency: 87950935 -> 88030841 (+0.09%); split: -0.03%, +0.12%
InvThroughput: 24784756 -> 24799536 (+0.06%); split: -0.02%, +0.08%
Copies: 588743 -> 561805 (-4.58%)
Branches: 242521 -> 215578 (-11.11%)
SALU: 877856 -> 850918 (-3.07%)

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/19070>
 2024-10-25 16:47:32 +00:00
Georg Lehmann
075c5818cb aco/ssa_elimination: don't assume exec writes can be removed based on block kind 
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19070>
 2024-10-25 16:47:32 +00:00
Georg Lehmann
61ab33c883 aco/ssa_elimination: add instr_accesses helper 
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19070>
 2024-10-25 16:47:32 +00:00
Georg Lehmann
b79950fc1f aco: remove heuristic that restricts VOP2/C with 2 sgprs 
Looking at the stats, the slightly increased code size isn't a problem
compared to the benefits. This also only affects gfx10+, and those generations
aren't throughput limited by 64bit instructions like early gcn.

Foz-DB Navi21:
Totals from 12377 (15.59% of 79395) affected shaders:
MaxWaves: 269323 -> 269857 (+0.20%); split: +0.23%, -0.03%
Instrs: 16505304 -> 16472552 (-0.20%); split: -0.21%, +0.01%
CodeSize: 89815804 -> 90130344 (+0.35%); split: -0.02%, +0.37%
VGPRs: 661160 -> 658640 (-0.38%); split: -0.40%, +0.02%
SpillSGPRs: 3032 -> 3049 (+0.56%)
SpillVGPRs: 826 -> 796 (-3.63%)
Latency: 145800231 -> 145818568 (+0.01%); split: -0.14%, +0.15%
InvThroughput: 39026010 -> 38892467 (-0.34%); split: -0.36%, +0.02%
VClause: 325693 -> 325992 (+0.09%); split: -0.12%, +0.21%
SClause: 497938 -> 497208 (-0.15%); split: -0.23%, +0.08%
Copies: 1239036 -> 1204045 (-2.82%); split: -2.90%, +0.07%
Branches: 462952 -> 462934 (-0.00%); split: -0.01%, +0.00%
PreSGPRs: 586066 -> 587558 (+0.25%)
PreVGPRs: 550024 -> 547736 (-0.42%)
VALU: 11147608 -> 11114528 (-0.30%); split: -0.31%, +0.01%
SALU: 2105546 -> 2105131 (-0.02%); split: -0.03%, +0.01%
VMEM: 575983 -> 575923 (-0.01%)

Foz-DB Navi31:
Totals from 11544 (14.54% of 79395) affected shaders:
MaxWaves: 319612 -> 319804 (+0.06%)
Instrs: 17563158 -> 17527341 (-0.20%); split: -0.22%, +0.02%
CodeSize: 92366832 -> 92626280 (+0.28%); split: -0.03%, +0.31%
VGPRs: 667620 -> 665484 (-0.32%); split: -0.33%, +0.01%
SpillSGPRs: 3418 -> 3434 (+0.47%)
SpillVGPRs: 896 -> 858 (-4.24%)
Scratch: 4738048 -> 4736512 (-0.03%)
Latency: 141366653 -> 141399756 (+0.02%); split: -0.10%, +0.12%
InvThroughput: 26213994 -> 26165751 (-0.18%); split: -0.21%, +0.03%
VClause: 307956 -> 308124 (+0.05%); split: -0.12%, +0.18%
SClause: 477816 -> 477326 (-0.10%); split: -0.18%, +0.08%
Copies: 1161148 -> 1129386 (-2.74%); split: -2.81%, +0.08%
Branches: 411509 -> 411506 (-0.00%); split: -0.00%, +0.00%
PreSGPRs: 531354 -> 535027 (+0.69%)
PreVGPRs: 525201 -> 521861 (-0.64%)
VALU: 10360363 -> 10330274 (-0.29%); split: -0.30%, +0.01%
SALU: 1778044 -> 1777585 (-0.03%); split: -0.04%, +0.01%
VMEM: 551379 -> 551303 (-0.01%)
VOPD: 3539 -> 3471 (-1.92%); split: +0.14%, -2.06%

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31804>
 2024-10-24 17:44:13 +00:00
Rhys Perry
4579586c66 aco/tests: add tests for VALUReadSGPRHazard 
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/30478>
 2024-10-24 16:08:08 +00:00
Rhys Perry
47e0f468cf aco: workaround VALUReadSGPRHazard 
fossil-db (gfx1200):
Totals from 65112 (82.01% of 79395) affected shaders:
Instrs: 41732906 -> 42987198 (+3.01%); split: -0.00%, +3.01%
CodeSize: 222451964 -> 226942644 (+2.02%); split: -0.01%, +2.03%
Latency: 290411063 -> 290944688 (+0.18%); split: -0.00%, +0.18%
InvThroughput: 45854913 -> 45910275 (+0.12%); split: -0.00%, +0.12%

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/30478>
 2024-10-24 16:08:07 +00:00
Rhys Perry
9ab0c4b047 aco: minor CounterMap::operator== fix 
I don't think this matters for how we use CounterMap::operator==.

The BITSET_TEST() was unnecessary because of the BITSET_EQUAL above.

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/30478>
 2024-10-24 16:08:07 +00:00
Rhys Perry
f5b871f825 aco: split CounterMap off from VGPRCounterMap 
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/30478>
 2024-10-24 16:08:07 +00:00
Rhys Perry
d312220c2d aco,radv,radeonsi: add aco_shader_info::ps::has_prolog 
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/30478>
 2024-10-24 16:08:07 +00:00
Daniel Schürmann
c8348139fd nir: change signature of nir_src_is_divergent() 
Now, it takes nir_src * instead of nir_src.
Also move the implementation to nir_divergence_analysis.c.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30787>
 2024-10-24 10:06:17 +00:00
Daniel Schürmann
ce0a3fe645 nir/opt_uniform_atomics: don't preserve divergence information 
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30787>
 2024-10-24 10:06:17 +00:00
Georg Lehmann
a4b179e445 aco/ssa_elimination: don't avoid saving exec when optimizing branching sequence 
insert_exec_mask will no longer use s_and_saveexec if there was a previous copy
from a sgpr to exec, so this code path is no longer taken.

No Foz-DB changes.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31567>
 2024-10-23 19:34:53 +00:00
Georg Lehmann
d2dcaf1f5e aco/insert_exec: reuse old exec temp instead using s_and_saveexec 
This means the v_cmpx optimization in ssa_elimination no longer
needs to insert a copy to save exec.

Foz-DB Navi31:
Totals from 13816 (17.40% of 79395) affected shaders:
Instrs: 23694267 -> 23670199 (-0.10%); split: -0.11%, +0.01%
CodeSize: 124559288 -> 124457508 (-0.08%); split: -0.09%, +0.01%
SpillSGPRs: 5324 -> 5354 (+0.56%); split: -1.00%, +1.56%
Latency: 207245846 -> 207213681 (-0.02%); split: -0.03%, +0.01%
InvThroughput: 35442657 -> 35437220 (-0.02%); split: -0.02%, +0.01%
VClause: 444672 -> 444670 (-0.00%); split: -0.00%, +0.00%
SClause: 639419 -> 639373 (-0.01%); split: -0.04%, +0.03%
Copies: 1529008 -> 1515871 (-0.86%); split: -1.02%, +0.16%
Branches: 557201 -> 557701 (+0.09%); split: -0.00%, +0.09%
PreSGPRs: 682840 -> 686048 (+0.47%)
VALU: 13978010 -> 13978032 (+0.00%); split: -0.00%, +0.00%
SALU: 2214600 -> 2197061 (-0.79%); split: -0.81%, +0.02%
VOPD: 5561 -> 5560 (-0.02%)

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31567>
 2024-10-23 19:34:53 +00:00
Georg Lehmann
0471522377 aco/insert_exec: reuse old exec temp in loop pre-header 
Avoid an exec copy.

Foz-DB Navi31:
Totals from 2315 (2.92% of 79395) affected shaders:
Instrs: 9082831 -> 9058990 (-0.26%); split: -0.27%, +0.00%
CodeSize: 48017244 -> 47858064 (-0.33%); split: -0.34%, +0.01%
SpillSGPRs: 1680 -> 1684 (+0.24%); split: -0.48%, +0.71%
Latency: 109511718 -> 109525041 (+0.01%); split: -0.01%, +0.02%
InvThroughput: 20287085 -> 20289370 (+0.01%); split: -0.00%, +0.02%
VClause: 192259 -> 192260 (+0.00%)
SClause: 234082 -> 234124 (+0.02%); split: -0.01%, +0.03%
Copies: 667271 -> 645577 (-3.25%); split: -3.27%, +0.02%
Branches: 264086 -> 264088 (+0.00%)
PreSGPRs: 136831 -> 136966 (+0.10%)
VALU: 5234735 -> 5234740 (+0.00%); split: -0.00%, +0.00%
SALU: 949283 -> 927327 (-2.31%); split: -2.32%, +0.01%
VOPD: 1529 -> 1535 (+0.39%)

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31567>
 2024-10-23 19:34:53 +00:00
Georg Lehmann
31f62a6123 aco/insert_exec: don't always reset top exec 
This allows to re-use previous temporaries in case exec was restored
from a Temp, rather than having to create a new copy of exec.

Foz-DB Navi31:
Totals from 545 (0.69% of 79395) affected shaders:
Instrs: 216563 -> 215698 (-0.40%)
CodeSize: 1183536 -> 1180076 (-0.29%)
Latency: 1135269 -> 1135294 (+0.00%); split: -0.00%, +0.00%
Copies: 11933 -> 11072 (-7.22%)
SALU: 18990 -> 18129 (-4.53%)

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31567>
 2024-10-23 19:34:53 +00:00
Georg Lehmann
4f04e6f0c4 aco/insert_exec: avoid phis for masks in exec 
Exec always contains the same value as the top of stack, even if the
top of stack is a temporary/constant. So if the predecessors have different
top of stack operands, don't insert a phi and use exec as the new top of stack.

No Foz-DB changes.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31567>
 2024-10-23 19:34:53 +00:00
Georg Lehmann
0338bb9ae8 aco/ssa_elimination: also optimize branching sequence with s_and without saveexec 
insert_exec will start using this in the future, handle it the same
just without the path to save exec before the v_cmpx instruction.

No Foz-DB changes.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31567>
 2024-10-23 19:34:53 +00:00
Daniel Schürmann
e8472d484f aco/spill: use float division for score() calculation rather than integers 
This was the original intention and should result in more fine-grained
and thus better decisions.

Totals from 63 (0.08% of 79395) affected shaders: (Navi31)
Instrs: 3173500 -> 3174012 (+0.02%); split: -0.01%, +0.02%
CodeSize: 16345348 -> 16349288 (+0.02%); split: -0.01%, +0.03%
Latency: 18528036 -> 18526082 (-0.01%); split: -0.02%, +0.01%
InvThroughput: 3619125 -> 3618709 (-0.01%); split: -0.02%, +0.01%
VClause: 82654 -> 82648 (-0.01%)
SClause: 61256 -> 61257 (+0.00%); split: -0.00%, +0.01%
Copies: 250037 -> 250158 (+0.05%); split: -0.06%, +0.11%
Branches: 101302 -> 101303 (+0.00%)
VALU: 1791447 -> 1791435 (-0.00%); split: -0.00%, +0.00%
SALU: 401898 -> 402007 (+0.03%); split: -0.03%, +0.06%
VOPD: 730 -> 741 (+1.51%)

Totals from 40 (0.06% of 63053) affected shaders: (Vega10)

Instrs: 161584 -> 161567 (-0.01%); split: -0.04%, +0.03%
CodeSize: 891168 -> 891004 (-0.02%); split: -0.04%, +0.03%
Latency: 3550766 -> 3549770 (-0.03%); split: -0.05%, +0.03%
InvThroughput: 2627028 -> 2626484 (-0.02%); split: -0.03%, +0.01%
VClause: 2970 -> 2971 (+0.03%)
SClause: 4203 -> 4205 (+0.05%); split: -0.26%, +0.31%
Copies: 19923 -> 19893 (-0.15%); split: -0.44%, +0.29%
VALU: 116045 -> 116054 (+0.01%); split: -0.01%, +0.02%
SALU: 22100 -> 22066 (-0.15%); split: -0.39%, +0.24%
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31769>
 2024-10-23 14:35:29 +00:00
Daniel Schürmann
30d85b23ef aco/spill: fix faulty assertions 
By unintentionally using integer division for score(), these
assertions were likely to be raised by accident.

Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31769>
 2024-10-23 14:35:29 +00:00
Daniel Schürmann
d5581b1124 aco/live_var_analysis: check isFixed() for definitions in order to set needs_vcc 
In rare cases, it could happen that during post-RA validation,
live-var-analysis sets needs_vcc = false after if was true
before register allocation.

Fixes: bb5eace0dc ('aco/live_var_analysis: check for isPrecolored flag rather than isFixed')
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31791>
 2024-10-23 09:42:26 +00:00
Georg Lehmann
5da34ebee4 aco/insert_exec: remove get_exec_op 
We used to only store Temps in the stack, so undef meant exec.
Then the stack was changed to operands, and some places started storing exec
directly, drop the undef handling by replacing everything with Operand(exec, lm)

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31560>
 2024-10-22 17:03:26 +00:00
Georg Lehmann
8d148401cb aco/ir: rework Operand equality to return true for equal fixed non-temp ops 
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31560>
 2024-10-22 17:03:26 +00:00
Georg Lehmann
6716fb08d8 aco/insert_exec: remove unused includes 
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31560>
 2024-10-22 17:03:26 +00:00
Georg Lehmann
23fb0883eb aco/insert_exec: untangle add_branch_code control flow 
All of the single ifs with return hide that this is effectively almost an
if-else chain, so convert it to one for real.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31560>
 2024-10-22 17:03:26 +00:00
Georg Lehmann
de7d931962 aco/insert_exec: remove stray break_cond variable 
This was always trivial since some discard rework.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31560>
 2024-10-22 17:03:26 +00:00
Georg Lehmann
ade7f1a203 aco/insert_exec: replace pair with a named struct 
.first and .second everywhere was hard to read.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31560>
 2024-10-22 17:03:26 +00:00
Georg Lehmann
a3054499ba aco/insert_exec: don't pretend WQMState is a bit mask 
It's a simple enum.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31560>
 2024-10-22 17:03:26 +00:00
Daniel Schürmann
ef47cce51c aco/ra: always block register file for precolored operands 
so that they don't accidentally get renamed.

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31387>
 2024-10-22 12:29:18 +00:00
Daniel Schürmann
18e7e8d8f0 aco/ra: make use of Precolored flag 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31387>
 2024-10-22 12:29:18 +00:00
Daniel Schürmann
bb5eace0dc aco/live_var_analysis: check for isPrecolored flag rather than isFixed 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31387>
 2024-10-22 12:29:18 +00:00
Daniel Schürmann
e2705a9d85 aco: set Precolored flag before register allocation 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31387>
 2024-10-22 12:29:18 +00:00
Daniel Schürmann
c2ed4b474a aco: introduce 'isPrecolored' flag for Operand and Definition 
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31387>
 2024-10-22 12:29:18 +00:00
Georg Lehmann
10951bb11a aco: fix 64bit extract_i8/extract_i16 
The old code only sign extended to 32bit, with a zero hi half.

Fixes: 1f2518ef9f ("aco: implement nir_op_extract/nir_op_insert")
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31734>
 2024-10-21 07:13:57 +00:00
Georg Lehmann
aabadb30fc aco/print_ir: use parse_depctr_wait 
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31132>
 2024-10-17 11:16:16 +00:00
Georg Lehmann
ced7a01954 aco/statistics: update branch issue cycles 
Foz-DB Navi31:
Totals from 14319 (18.04% of 79395) affected shaders:
Instrs: 20064495 -> 20062876 (-0.01%)
CodeSize: 105334568 -> 105327704 (-0.01%)

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31132>
 2024-10-17 11:16:16 +00:00
Georg Lehmann
ec11cfc69d aco/insert_delay_alu: do not delay lane mask fast forwarding 
The delay actually hurts performance in this case.

Foz-DB Navi31:
Totals from 30340 (38.21% of 79395) affected shaders:
Instrs: 30778999 -> 30726605 (-0.17%); split: -0.17%, +0.00%
CodeSize: 162380180 -> 162170808 (-0.13%); split: -0.13%, +0.00%
Latency: 228185562 -> 228186976 (+0.00%); split: -0.00%, +0.00%
InvThroughput: 39001151 -> 39000897 (-0.00%); split: -0.00%, +0.00%

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31132>
 2024-10-17 11:16:16 +00:00
Georg Lehmann
e4889fd4b5 aco/insert_delay_alu: consider more implicit waits 
Foz-DB Navi31:
Totals from 37961 (47.81% of 79395) affected shaders:
Instrs: 34175286 -> 33978599 (-0.58%)
CodeSize: 180059352 -> 179190076 (-0.48%); split: -0.48%, +0.00%
Latency: 259826196 -> 259798474 (-0.01%); split: -0.01%, +0.00%
InvThroughput: 42792700 -> 42789298 (-0.01%); split: -0.01%, +0.00%

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31132>
 2024-10-17 11:16:16 +00:00
Georg Lehmann
840b5841d3 aco: do not track ALU delay across jumps 
This assumes that the best case jump latency is higher than the worst case
ALU latency.

Foz-DB Navi31:
Totals from 17720 (22.32% of 79395) affected shaders:
Instrs: 26009663 -> 25929989 (-0.31%); split: -0.31%, +0.00%
CodeSize: 136571496 -> 136254420 (-0.23%); split: -0.23%, +0.00%
Latency: 215731308 -> 215722059 (-0.00%); split: -0.01%, +0.00%
InvThroughput: 36534197 -> 36532070 (-0.01%); split: -0.01%, +0.00%

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31132>
 2024-10-17 11:16:16 +00:00
Georg Lehmann
977f435f4c aco/ir: add function to parse depctr waits 
No Foz-DB changes on Navi31.

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/31132>
 2024-10-17 11:16:16 +00:00
Rhys Perry
33eb2d7fe4 aco: skip uniformization of certain merge phis 
If a source is a VGPR, then skip if it's safe. This fixes the regressions
from the previous commit.

fossil-db (navi31):
Totals from 5118 (6.45% of 79395) affected shaders:
MaxWaves: 159560 -> 159520 (-0.03%); split: +0.01%, -0.03%
Instrs: 2165351 -> 2138456 (-1.24%); split: -1.26%, +0.02%
CodeSize: 11260340 -> 11152460 (-0.96%); split: -0.98%, +0.02%
VGPRs: 218124 -> 225144 (+3.22%); split: -0.13%, +3.35%
Latency: 11059208 -> 11116102 (+0.51%); split: -0.18%, +0.69%
InvThroughput: 1252148 -> 1230193 (-1.75%); split: -1.77%, +0.01%
VClause: 39513 -> 39518 (+0.01%); split: -0.48%, +0.49%
SClause: 59434 -> 59378 (-0.09%); split: -0.11%, +0.02%
Copies: 165997 -> 156172 (-5.92%); split: -6.68%, +0.76%
PreSGPRs: 181203 -> 181094 (-0.06%)
PreVGPRs: 139393 -> 139731 (+0.24%)
VALU: 1244301 -> 1220769 (-1.89%); split: -1.91%, +0.02%
SALU: 200240 -> 199567 (-0.34%); split: -0.34%, +0.00%

fossil-db (navi21):
Totals from 35520 (44.74% of 79395) affected shaders:
MaxWaves: 951870 -> 951830 (-0.00%)
Instrs: 20229388 -> 20227776 (-0.01%); split: -0.01%, +0.00%
CodeSize: 105379916 -> 105513740 (+0.13%); split: -0.01%, +0.13%
VGPRs: 1375232 -> 1375400 (+0.01%)
Latency: 81046435 -> 81013986 (-0.04%); split: -0.04%, +0.00%
InvThroughput: 15269166 -> 15273295 (+0.03%); split: -0.01%, +0.04%
VClause: 354314 -> 354310 (-0.00%); split: -0.00%, +0.00%
SClause: 417049 -> 417047 (-0.00%); split: -0.00%, +0.00%
Copies: 1699445 -> 1699488 (+0.00%); split: -0.01%, +0.01%
Branches: 591274 -> 591269 (-0.00%); split: -0.00%, +0.00%
PreSGPRs: 1371062 -> 1370567 (-0.04%)
PreVGPRs: 1100716 -> 1100953 (+0.02%)
VALU: 11076189 -> 11075167 (-0.01%); split: -0.01%, +0.00%
SALU: 3648002 -> 3647378 (-0.02%); split: -0.02%, +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/30211>
 2024-10-10 14:59:27 +00:00
Rhys Perry
ce33ffd03a aco: ensure phis uniformized by divergence analysis are SGPR 
Otherwise, they might not actually be uniform when divergence analysis
claimed they are.

fossil-db (navi31):
Totals from 5118 (6.45% of 79395) affected shaders:
MaxWaves: 159520 -> 159560 (+0.03%); split: +0.03%, -0.01%
Instrs: 2138456 -> 2165351 (+1.26%); split: -0.02%, +1.28%
CodeSize: 11152460 -> 11260340 (+0.97%); split: -0.02%, +0.98%
VGPRs: 225144 -> 218124 (-3.12%); split: -3.25%, +0.13%
Latency: 11116102 -> 11059208 (-0.51%); split: -0.69%, +0.18%
InvThroughput: 1230193 -> 1252148 (+1.78%); split: -0.01%, +1.80%
VClause: 39518 -> 39513 (-0.01%); split: -0.49%, +0.48%
SClause: 59378 -> 59434 (+0.09%); split: -0.02%, +0.11%
Copies: 156172 -> 165997 (+6.29%); split: -0.81%, +7.10%
PreSGPRs: 181094 -> 181203 (+0.06%)
PreVGPRs: 139731 -> 139393 (-0.24%)
VALU: 1220769 -> 1244301 (+1.93%); split: -0.02%, +1.95%
SALU: 199567 -> 200240 (+0.34%); split: -0.00%, +0.34%

fossil-db (navi21):
Totals from 35520 (44.74% of 79395) affected shaders:
MaxWaves: 951830 -> 951870 (+0.00%)
Instrs: 20227773 -> 20229388 (+0.01%); split: -0.00%, +0.01%
CodeSize: 105513724 -> 105379916 (-0.13%); split: -0.13%, +0.01%
VGPRs: 1375400 -> 1375232 (-0.01%)
Latency: 81013985 -> 81046435 (+0.04%); split: -0.00%, +0.04%
InvThroughput: 15273291 -> 15269166 (-0.03%); split: -0.04%, +0.01%
VClause: 354310 -> 354314 (+0.00%); split: -0.00%, +0.00%
SClause: 417047 -> 417049 (+0.00%); split: -0.00%, +0.00%
Copies: 1699486 -> 1699445 (-0.00%); split: -0.01%, +0.01%
Branches: 591269 -> 591274 (+0.00%); split: -0.00%, +0.00%
PreSGPRs: 1370567 -> 1371062 (+0.04%)
PreVGPRs: 1100953 -> 1100716 (-0.02%)
VALU: 11075164 -> 11076189 (+0.01%); split: -0.00%, +0.01%
SALU: 3647378 -> 3648002 (+0.02%); split: -0.00%, +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/30211>
 2024-10-10 14:59:26 +00:00