Markk116/smarm
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

benches: expose preemption knobs + sweep runner

Browse Source 
Config API changes (src/preempt.rs, src/runtime.rs):
- preempt: promote ALLOC_INTERVAL and TIMESLICE_CYCLES from bare consts to
  DEFAULT_ALLOC_INTERVAL / DEFAULT_TIMESLICE_CYCLES; store active values in
  thread-locals set on each actor resume so multiple runtimes can use
  different settings concurrently.
- runtime: add alloc_interval / timeslice_cycles fields to Config; add
  Config::alloc_interval(n) and Config::timeslice_cycles(c) builder methods;
  thread the values through RuntimeInner to the reset_timeslice() call in
  schedule_loop.

Bench changes:
- Add bench_cfg(threads) helper to general/tokio_favored/smarm_favored that
  wraps Config::exact and reads SMARM_ALLOC_INTERVAL / SMARM_TIMESLICE_CYCLES
  env vars, so the sweep script can vary knobs without recompiling.

Sweep tooling (benches/sweep.py):
- 'run':     run the 3-file bench suite once; --save-baseline persists JSON
- 'regress': compare current run against baseline.json, exit 1 on any bench
             that regresses >10% vs stored medians
- 'sweep':   run the full SWEEP_GRID (10 points), print comparison table,
             optional --save-csv; binaries pre-built so no recompile per point

Sweep results (10-point grid, 1-CPU sandbox):
- The preemption knobs have very little effect on this single-CPU machine.
  Most benches move <5% across the entire grid.
- Longer timeslices (tc=600k, tc=1200k) reliably hurt spawn_storm_busy
  (+11-15%) and catch_unwind_panics (+10-12%) because actors hold the
  scheduler mutex longer per timeslice, stalling the storm of joinable tasks.
- Shorter timeslices (tc=150k) give a small improvement on many_timers
  (-3-4%) and a wash everywhere else.
- yield_in_hot_loop and uncontended_channel are essentially flat across all
  knobs — both are scheduling-dominated and call yield_now explicitly, so
  the RDTSC-driven preemption path is irrelevant.
- Conclusion: the knobs matter primarily under contention (multi-core).
  Re-run sweep on a multi-core machine before drawing tuning conclusions.
This commit is contained in:
Bench
2026-05-24 11:48:15 +00:00
committed by smarm
parent 6d1c59fb99
commit 3da6ffaa77
15 changed files with 2315 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

126
benches/baseline-output/sweep/ai128_tc1200k.txt Normal file
View File

@@ -0,0 +1,126 @@
smarm general benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
CHAIN_DEPTH=1000, YIELD_TASKS=200×1000, PRIME_N=400000/64 workers, PP_ROUNDS=1000
================================================================================
chained_spawn: depth 1000
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 8720 | 8526 | 9319
smarm 1-thread | 1000 | 8662 | 8571 | 8991
tokio current_thread | 1000 | 123 | 123 | 152
tokio multi-thread | 1000 | 188 | 184 | 230
================================================================================
yield_many: 200 tasks × 1000 yields
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 200000 | 41530 | 41242 | 43501
smarm 1-thread | 200000 | 41575 | 41187 | 43323
tokio current_thread | 200000 | 15098 | 15020 | 15348
tokio multi-thread | 200000 | 15900 | 15827 | 16012
================================================================================
fan_out_compute: primes in [2, 400000) across 64
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 29573 | 29435 | 31647
smarm 1-thread | 33860 | 29521 | 29453 | 29847
tokio current_thread | 33860 | 28495 | 28441 | 30150
tokio multi-thread | 33860 | 34384 | 34297 | 34745
================================================================================
ping_pong_oneshot: 1000 rounds
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 17190 | 16994 | 17541
smarm 1-thread | 1000 | 17078 | 16916 | 19139
tokio current_thread | 1000 | 899 | 896 | 1000
tokio multi-thread | 1000 | 4198 | 4116 | 4573
smarm tokio-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
STORM_BACKGROUND=8, STORM_SPAWN=10000, MPSC=32×10000, TIMER_ACTORS=10000 (1–10 ms), SCALING_N=400000/64
================================================================================
spawn_storm_busy: 8 bg yielders + 10000 zero-work spawns
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 138556 | 136165 | 140947
smarm 1-thread | 10000 | 140223 | 136325 | 146781
tokio current_thread | 10000 | 2671 | 2622 | 2913
tokio multi-thread | 10000 | 6004 | 4360 | 12576
================================================================================
mpsc_contention: 32 producers × 10000 msgs → 1 consumer
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 320000 | 9051 | 8967 | 11152
smarm 1-thread | 320000 | 9058 | 9008 | 9998
tokio current_thread | 320000 | 17375 | 17131 | 18514
tokio multi-thread | 320000 | 17955 | 17452 | 18508
================================================================================
many_timers: 10000 actors sleeping 1–10 ms
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 156969 | 153124 | 167711
smarm 1-thread | 10000 | 150638 | 146070 | 168286
tokio current_thread | 10000 | 13823 | 13482 | 14796
tokio multi-thread | 10000 | 15034 | 14425 | 15320
================================================================================
multi_thread_scaling: primes in [2, 400000) across 64 workers
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 30075 | 29707 | 30720
tokio multi 1-thread | 33860 | 29060 | 28835 | 44378
smarm smarm-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
RECURSE_DEPTH=500, HOT_YIELDS=500000×2, UNCONT_MSGS=1000000, PANIC_TASKS=10000
================================================================================
deep_recursion: depth 500
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1 | 86 | 79 | 130
smarm 1-thread | 1 | 83 | 78 | 146
tokio current_thread | 1 | 25 | 25 | 31
tokio multi-thread | 1 | 49 | 46 | 85
================================================================================
yield_in_hot_loop: 2 actors × 500000 yields (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 190902 | 187600 | 194333
tokio current_thread | 1000000 | 150279 | 148175 | 188184
================================================================================
uncontended_channel: 1→1, 1000000 msgs (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 27687 | 27198 | 29555
tokio current_thread | 1000000 | 54465 | 54048 | 55954
================================================================================
catch_unwind_panics: 10000 tasks, 50% panic
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 160308 | 154365 | 167009
smarm 1-thread | 10000 | 158662 | 155458 | 168896
tokio current_thread | 10000 | 267762 | 260876 | 294092
tokio multi-thread | 10000 | 275097 | 269344 | 287681

126
benches/baseline-output/sweep/ai128_tc150k.txt Normal file
View File

@@ -0,0 +1,126 @@
smarm general benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
CHAIN_DEPTH=1000, YIELD_TASKS=200×1000, PRIME_N=400000/64 workers, PP_ROUNDS=1000
================================================================================
chained_spawn: depth 1000
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 8596 | 8491 | 8805
smarm 1-thread | 1000 | 8552 | 8461 | 9003
tokio current_thread | 1000 | 125 | 125 | 260
tokio multi-thread | 1000 | 190 | 184 | 338
================================================================================
yield_many: 200 tasks × 1000 yields
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 200000 | 41885 | 41112 | 43292
smarm 1-thread | 200000 | 42174 | 41063 | 43145
tokio current_thread | 200000 | 15195 | 15010 | 15589
tokio multi-thread | 200000 | 16037 | 15869 | 17057
================================================================================
fan_out_compute: primes in [2, 400000) across 64
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 29872 | 29629 | 31596
smarm 1-thread | 33860 | 29776 | 29528 | 30003
tokio current_thread | 33860 | 28705 | 28605 | 30287
tokio multi-thread | 33860 | 34655 | 34503 | 36596
================================================================================
ping_pong_oneshot: 1000 rounds
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 16898 | 16574 | 17386
smarm 1-thread | 1000 | 16871 | 16677 | 18467
tokio current_thread | 1000 | 897 | 857 | 991
tokio multi-thread | 1000 | 4325 | 4228 | 4458
smarm tokio-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
STORM_BACKGROUND=8, STORM_SPAWN=10000, MPSC=32×10000, TIMER_ACTORS=10000 (1–10 ms), SCALING_N=400000/64
================================================================================
spawn_storm_busy: 8 bg yielders + 10000 zero-work spawns
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 133462 | 129526 | 138685
smarm 1-thread | 10000 | 130118 | 127633 | 142344
tokio current_thread | 10000 | 2713 | 2608 | 2831
tokio multi-thread | 10000 | 7367 | 4345 | 11741
================================================================================
mpsc_contention: 32 producers × 10000 msgs → 1 consumer
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 320000 | 9077 | 8944 | 9287
smarm 1-thread | 320000 | 9100 | 9033 | 10604
tokio current_thread | 320000 | 17310 | 17122 | 18616
tokio multi-thread | 320000 | 17484 | 17413 | 17748
================================================================================
many_timers: 10000 actors sleeping 1–10 ms
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 140039 | 135577 | 145123
smarm 1-thread | 10000 | 139931 | 135513 | 143841
tokio current_thread | 10000 | 14524 | 14378 | 14564
tokio multi-thread | 10000 | 15066 | 14677 | 15336
================================================================================
multi_thread_scaling: primes in [2, 400000) across 64 workers
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 29620 | 29511 | 31347
tokio multi 1-thread | 33860 | 29046 | 28817 | 29687
smarm smarm-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
RECURSE_DEPTH=500, HOT_YIELDS=500000×2, UNCONT_MSGS=1000000, PANIC_TASKS=10000
================================================================================
deep_recursion: depth 500
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1 | 94 | 79 | 371
smarm 1-thread | 1 | 183 | 83 | 317
tokio current_thread | 1 | 25 | 25 | 31
tokio multi-thread | 1 | 54 | 41 | 71
================================================================================
yield_in_hot_loop: 2 actors × 500000 yields (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 189034 | 187674 | 192204
tokio current_thread | 1000000 | 151106 | 149564 | 155601
================================================================================
uncontended_channel: 1→1, 1000000 msgs (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 26949 | 26838 | 30868
tokio current_thread | 1000000 | 52984 | 52149 | 55141
================================================================================
catch_unwind_panics: 10000 tasks, 50% panic
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 145860 | 143015 | 152734
smarm 1-thread | 10000 | 144550 | 141592 | 149247
tokio current_thread | 10000 | 267500 | 265301 | 278751
tokio multi-thread | 10000 | 275320 | 268986 | 286891

126
benches/baseline-output/sweep/ai128_tc300k.txt Normal file
View File

@@ -0,0 +1,126 @@
smarm general benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
CHAIN_DEPTH=1000, YIELD_TASKS=200×1000, PRIME_N=400000/64 workers, PP_ROUNDS=1000
================================================================================
chained_spawn: depth 1000
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 8469 | 8414 | 8717
smarm 1-thread | 1000 | 8625 | 8479 | 10212
tokio current_thread | 1000 | 124 | 123 | 175
tokio multi-thread | 1000 | 194 | 184 | 317
================================================================================
yield_many: 200 tasks × 1000 yields
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 200000 | 41949 | 41419 | 43784
smarm 1-thread | 200000 | 42005 | 41491 | 45224
tokio current_thread | 200000 | 15139 | 15049 | 16352
tokio multi-thread | 200000 | 15985 | 15931 | 16306
================================================================================
fan_out_compute: primes in [2, 400000) across 64
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 29640 | 29515 | 31229
smarm 1-thread | 33860 | 29777 | 29642 | 30056
tokio current_thread | 33860 | 28704 | 28584 | 30317
tokio multi-thread | 33860 | 34870 | 34569 | 35876
================================================================================
ping_pong_oneshot: 1000 rounds
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 17098 | 16968 | 18688
smarm 1-thread | 1000 | 16918 | 16736 | 17326
tokio current_thread | 1000 | 915 | 882 | 1000
tokio multi-thread | 1000 | 4371 | 4265 | 4834
smarm tokio-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
STORM_BACKGROUND=8, STORM_SPAWN=10000, MPSC=32×10000, TIMER_ACTORS=10000 (1–10 ms), SCALING_N=400000/64
================================================================================
spawn_storm_busy: 8 bg yielders + 10000 zero-work spawns
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 127075 | 124760 | 130259
smarm 1-thread | 10000 | 125976 | 125121 | 128728
tokio current_thread | 10000 | 2703 | 2646 | 2807
tokio multi-thread | 10000 | 7201 | 4267 | 12853
================================================================================
mpsc_contention: 32 producers × 10000 msgs → 1 consumer
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 320000 | 9116 | 8985 | 9237
smarm 1-thread | 320000 | 9062 | 8947 | 10648
tokio current_thread | 320000 | 17380 | 17192 | 18363
tokio multi-thread | 320000 | 17854 | 17554 | 18219
================================================================================
many_timers: 10000 actors sleeping 1–10 ms
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 137944 | 132081 | 141862
smarm 1-thread | 10000 | 143773 | 137448 | 153703
tokio current_thread | 10000 | 14174 | 13751 | 15079
tokio multi-thread | 10000 | 15244 | 14625 | 16700
================================================================================
multi_thread_scaling: primes in [2, 400000) across 64 workers
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 30832 | 30082 | 33360
tokio multi 1-thread | 33860 | 29736 | 29321 | 29958
smarm smarm-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
RECURSE_DEPTH=500, HOT_YIELDS=500000×2, UNCONT_MSGS=1000000, PANIC_TASKS=10000
================================================================================
deep_recursion: depth 500
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1 | 84 | 78 | 122
smarm 1-thread | 1 | 90 | 79 | 157
tokio current_thread | 1 | 25 | 25 | 31
tokio multi-thread | 1 | 48 | 47 | 62
================================================================================
yield_in_hot_loop: 2 actors × 500000 yields (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 190830 | 188562 | 196621
tokio current_thread | 1000000 | 151537 | 150038 | 165825
================================================================================
uncontended_channel: 1→1, 1000000 msgs (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 27265 | 26969 | 29317
tokio current_thread | 1000000 | 53894 | 53380 | 56189
================================================================================
catch_unwind_panics: 10000 tasks, 50% panic
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 145006 | 144092 | 149002
smarm 1-thread | 10000 | 144417 | 142000 | 148224
tokio current_thread | 10000 | 265376 | 260227 | 272279
tokio multi-thread | 10000 | 277432 | 270860 | 283266

126
benches/baseline-output/sweep/ai128_tc600k.txt Normal file
View File

@@ -0,0 +1,126 @@
smarm general benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
CHAIN_DEPTH=1000, YIELD_TASKS=200×1000, PRIME_N=400000/64 workers, PP_ROUNDS=1000
================================================================================
chained_spawn: depth 1000
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 8721 | 8398 | 8994
smarm 1-thread | 1000 | 8587 | 8440 | 8810
tokio current_thread | 1000 | 124 | 124 | 294
tokio multi-thread | 1000 | 188 | 184 | 299
================================================================================
yield_many: 200 tasks × 1000 yields
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 200000 | 42588 | 42084 | 45080
smarm 1-thread | 200000 | 42252 | 41963 | 43615
tokio current_thread | 200000 | 15101 | 14994 | 15573
tokio multi-thread | 200000 | 15979 | 15890 | 16356
================================================================================
fan_out_compute: primes in [2, 400000) across 64
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 29686 | 29491 | 31263
smarm 1-thread | 33860 | 29841 | 29586 | 30570
tokio current_thread | 33860 | 28652 | 28510 | 30359
tokio multi-thread | 33860 | 34677 | 34461 | 35318
================================================================================
ping_pong_oneshot: 1000 rounds
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 16909 | 16579 | 20782
smarm 1-thread | 1000 | 16888 | 16537 | 20808
tokio current_thread | 1000 | 925 | 911 | 1021
tokio multi-thread | 1000 | 4192 | 4079 | 4531
smarm tokio-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
STORM_BACKGROUND=8, STORM_SPAWN=10000, MPSC=32×10000, TIMER_ACTORS=10000 (1–10 ms), SCALING_N=400000/64
================================================================================
spawn_storm_busy: 8 bg yielders + 10000 zero-work spawns
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 145813 | 142042 | 152501
smarm 1-thread | 10000 | 145119 | 141282 | 161294
tokio current_thread | 10000 | 2968 | 2899 | 3231
tokio multi-thread | 10000 | 6288 | 4289 | 12226
================================================================================
mpsc_contention: 32 producers × 10000 msgs → 1 consumer
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 320000 | 9662 | 9254 | 11370
smarm 1-thread | 320000 | 9673 | 9331 | 9989
tokio current_thread | 320000 | 18015 | 17334 | 21096
tokio multi-thread | 320000 | 18384 | 17837 | 19534
================================================================================
many_timers: 10000 actors sleeping 1–10 ms
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 160492 | 154795 | 180307
smarm 1-thread | 10000 | 161716 | 156498 | 191986
tokio current_thread | 10000 | 13895 | 13576 | 14913
tokio multi-thread | 10000 | 15074 | 14665 | 16070
================================================================================
multi_thread_scaling: primes in [2, 400000) across 64 workers
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 30001 | 29600 | 38039
tokio multi 1-thread | 33860 | 29419 | 28906 | 30079
smarm smarm-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
RECURSE_DEPTH=500, HOT_YIELDS=500000×2, UNCONT_MSGS=1000000, PANIC_TASKS=10000
================================================================================
deep_recursion: depth 500
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1 | 91 | 79 | 186
smarm 1-thread | 1 | 87 | 81 | 131
tokio current_thread | 1 | 25 | 25 | 103
tokio multi-thread | 1 | 56 | 47 | 64
================================================================================
yield_in_hot_loop: 2 actors × 500000 yields (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 190023 | 188250 | 193824
tokio current_thread | 1000000 | 154681 | 152074 | 187328
================================================================================
uncontended_channel: 1→1, 1000000 msgs (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 27264 | 26772 | 29512
tokio current_thread | 1000000 | 53324 | 51744 | 59282
================================================================================
catch_unwind_panics: 10000 tasks, 50% panic
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 155983 | 152595 | 161438
smarm 1-thread | 10000 | 162122 | 156170 | 200357
tokio current_thread | 10000 | 276303 | 264291 | 296266
tokio multi-thread | 10000 | 271350 | 267654 | 285897

126
benches/baseline-output/sweep/ai256_tc300k.txt Normal file
View File

@@ -0,0 +1,126 @@
smarm general benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
CHAIN_DEPTH=1000, YIELD_TASKS=200×1000, PRIME_N=400000/64 workers, PP_ROUNDS=1000
================================================================================
chained_spawn: depth 1000
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 9130 | 8720 | 10611
smarm 1-thread | 1000 | 8808 | 8617 | 9659
tokio current_thread | 1000 | 126 | 125 | 164
tokio multi-thread | 1000 | 190 | 184 | 329
================================================================================
yield_many: 200 tasks × 1000 yields
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 200000 | 42270 | 41814 | 44737
smarm 1-thread | 200000 | 42999 | 42104 | 45424
tokio current_thread | 200000 | 15441 | 15196 | 16096
tokio multi-thread | 200000 | 16249 | 16070 | 17620
================================================================================
fan_out_compute: primes in [2, 400000) across 64
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 29813 | 29627 | 30176
smarm 1-thread | 33860 | 29613 | 29440 | 31205
tokio current_thread | 33860 | 28637 | 28406 | 29179
tokio multi-thread | 33860 | 34472 | 34389 | 36092
================================================================================
ping_pong_oneshot: 1000 rounds
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 16899 | 16804 | 17017
smarm 1-thread | 1000 | 17001 | 16704 | 19533
tokio current_thread | 1000 | 914 | 893 | 1021
tokio multi-thread | 1000 | 4198 | 4136 | 4297
smarm tokio-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
STORM_BACKGROUND=8, STORM_SPAWN=10000, MPSC=32×10000, TIMER_ACTORS=10000 (1–10 ms), SCALING_N=400000/64
================================================================================
spawn_storm_busy: 8 bg yielders + 10000 zero-work spawns
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 128621 | 126503 | 132268
smarm 1-thread | 10000 | 131316 | 128354 | 133964
tokio current_thread | 10000 | 2763 | 2696 | 2996
tokio multi-thread | 10000 | 6023 | 4300 | 12908
================================================================================
mpsc_contention: 32 producers × 10000 msgs → 1 consumer
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 320000 | 9225 | 9071 | 11272
smarm 1-thread | 320000 | 9174 | 9028 | 9335
tokio current_thread | 320000 | 17210 | 17100 | 18404
tokio multi-thread | 320000 | 17550 | 17413 | 18080
================================================================================
many_timers: 10000 actors sleeping 1–10 ms
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 136396 | 133330 | 142485
smarm 1-thread | 10000 | 137374 | 134345 | 141168
tokio current_thread | 10000 | 13789 | 13499 | 14621
tokio multi-thread | 10000 | 15036 | 14729 | 15359
================================================================================
multi_thread_scaling: primes in [2, 400000) across 64 workers
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 30065 | 29819 | 32418
tokio multi 1-thread | 33860 | 29501 | 28916 | 30057
smarm smarm-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
RECURSE_DEPTH=500, HOT_YIELDS=500000×2, UNCONT_MSGS=1000000, PANIC_TASKS=10000
================================================================================
deep_recursion: depth 500
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1 | 94 | 81 | 257
smarm 1-thread | 1 | 83 | 80 | 134
tokio current_thread | 1 | 25 | 25 | 33
tokio multi-thread | 1 | 57 | 48 | 109
================================================================================
yield_in_hot_loop: 2 actors × 500000 yields (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 188506 | 187971 | 190121
tokio current_thread | 1000000 | 149663 | 148978 | 150733
================================================================================
uncontended_channel: 1→1, 1000000 msgs (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 26945 | 26703 | 29430
tokio current_thread | 1000000 | 52332 | 51838 | 54062
================================================================================
catch_unwind_panics: 10000 tasks, 50% panic
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 146192 | 143776 | 150609
smarm 1-thread | 10000 | 144012 | 140604 | 153892
tokio current_thread | 10000 | 268341 | 260941 | 275404
tokio multi-thread | 10000 | 272691 | 268094 | 307084

126
benches/baseline-output/sweep/ai32_tc150k.txt Normal file
View File

@@ -0,0 +1,126 @@
smarm general benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
CHAIN_DEPTH=1000, YIELD_TASKS=200×1000, PRIME_N=400000/64 workers, PP_ROUNDS=1000
================================================================================
chained_spawn: depth 1000
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 8653 | 8522 | 9163
smarm 1-thread | 1000 | 8908 | 8660 | 10606
tokio current_thread | 1000 | 124 | 123 | 175
tokio multi-thread | 1000 | 244 | 184 | 340
================================================================================
yield_many: 200 tasks × 1000 yields
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 200000 | 42597 | 41857 | 43492
smarm 1-thread | 200000 | 42621 | 42097 | 44386
tokio current_thread | 200000 | 15368 | 15144 | 16484
tokio multi-thread | 200000 | 16120 | 16012 | 19222
================================================================================
fan_out_compute: primes in [2, 400000) across 64
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 30499 | 29657 | 33910
smarm 1-thread | 33860 | 31190 | 30105 | 32675
tokio current_thread | 33860 | 28748 | 28643 | 29398
tokio multi-thread | 33860 | 34714 | 34499 | 36338
================================================================================
ping_pong_oneshot: 1000 rounds
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 16990 | 16853 | 17540
smarm 1-thread | 1000 | 16944 | 16740 | 18603
tokio current_thread | 1000 | 937 | 921 | 1056
tokio multi-thread | 1000 | 4342 | 4205 | 4549
smarm tokio-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
STORM_BACKGROUND=8, STORM_SPAWN=10000, MPSC=32×10000, TIMER_ACTORS=10000 (1–10 ms), SCALING_N=400000/64
================================================================================
spawn_storm_busy: 8 bg yielders + 10000 zero-work spawns
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 130032 | 128075 | 153842
smarm 1-thread | 10000 | 126396 | 125101 | 131406
tokio current_thread | 10000 | 2685 | 2629 | 2841
tokio multi-thread | 10000 | 6014 | 4126 | 11484
================================================================================
mpsc_contention: 32 producers × 10000 msgs → 1 consumer
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 320000 | 9122 | 8987 | 9334
smarm 1-thread | 320000 | 9073 | 8956 | 10151
tokio current_thread | 320000 | 17259 | 17163 | 17673
tokio multi-thread | 320000 | 22771 | 17709 | 24514
================================================================================
many_timers: 10000 actors sleeping 1–10 ms
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 137844 | 134570 | 157034
smarm 1-thread | 10000 | 141200 | 137494 | 156214
tokio current_thread | 10000 | 14809 | 14024 | 16518
tokio multi-thread | 10000 | 15089 | 14704 | 15331
================================================================================
multi_thread_scaling: primes in [2, 400000) across 64 workers
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 30880 | 29931 | 32667
tokio multi 1-thread | 33860 | 29862 | 29116 | 31310
smarm smarm-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
RECURSE_DEPTH=500, HOT_YIELDS=500000×2, UNCONT_MSGS=1000000, PANIC_TASKS=10000
================================================================================
deep_recursion: depth 500
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1 | 90 | 80 | 196
smarm 1-thread | 1 | 87 | 79 | 126
tokio current_thread | 1 | 25 | 25 | 53
tokio multi-thread | 1 | 52 | 47 | 88
================================================================================
yield_in_hot_loop: 2 actors × 500000 yields (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 191187 | 187194 | 198269
tokio current_thread | 1000000 | 152531 | 151113 | 154462
================================================================================
uncontended_channel: 1→1, 1000000 msgs (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 27413 | 27312 | 29463
tokio current_thread | 1000000 | 53620 | 52594 | 55332
================================================================================
catch_unwind_panics: 10000 tasks, 50% panic
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 144199 | 141893 | 157984
smarm 1-thread | 10000 | 144857 | 142722 | 152275
tokio current_thread | 10000 | 268006 | 264666 | 274542
tokio multi-thread | 10000 | 271827 | 268740 | 290301

126
benches/baseline-output/sweep/ai32_tc300k.txt Normal file
View File

@@ -0,0 +1,126 @@
smarm general benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
CHAIN_DEPTH=1000, YIELD_TASKS=200×1000, PRIME_N=400000/64 workers, PP_ROUNDS=1000
================================================================================
chained_spawn: depth 1000
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 8950 | 8591 | 10655
smarm 1-thread | 1000 | 9688 | 8657 | 11720
tokio current_thread | 1000 | 123 | 123 | 256
tokio multi-thread | 1000 | 192 | 177 | 314
================================================================================
yield_many: 200 tasks × 1000 yields
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 200000 | 42965 | 41667 | 44850
smarm 1-thread | 200000 | 42881 | 41634 | 48864
tokio current_thread | 200000 | 15112 | 14986 | 15484
tokio multi-thread | 200000 | 16006 | 15915 | 16647
================================================================================
fan_out_compute: primes in [2, 400000) across 64
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 29931 | 29750 | 31707
smarm 1-thread | 33860 | 29977 | 29670 | 30996
tokio current_thread | 33860 | 28615 | 28441 | 30188
tokio multi-thread | 33860 | 34371 | 34330 | 35176
================================================================================
ping_pong_oneshot: 1000 rounds
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 16753 | 16498 | 18516
smarm 1-thread | 1000 | 16728 | 16599 | 16874
tokio current_thread | 1000 | 940 | 933 | 1037
tokio multi-thread | 1000 | 4317 | 4236 | 4427
smarm tokio-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
STORM_BACKGROUND=8, STORM_SPAWN=10000, MPSC=32×10000, TIMER_ACTORS=10000 (1–10 ms), SCALING_N=400000/64
================================================================================
spawn_storm_busy: 8 bg yielders + 10000 zero-work spawns
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 132575 | 128629 | 136999
smarm 1-thread | 10000 | 130313 | 127372 | 157234
tokio current_thread | 10000 | 2689 | 2611 | 2833
tokio multi-thread | 10000 | 11337 | 4288 | 12635
================================================================================
mpsc_contention: 32 producers × 10000 msgs → 1 consumer
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 320000 | 9122 | 9000 | 11033
smarm 1-thread | 320000 | 9143 | 9015 | 9333
tokio current_thread | 320000 | 17705 | 17250 | 18111
tokio multi-thread | 320000 | 18044 | 17621 | 19484
================================================================================
many_timers: 10000 actors sleeping 1–10 ms
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 141925 | 135531 | 188381
smarm 1-thread | 10000 | 139655 | 134291 | 146458
tokio current_thread | 10000 | 13837 | 13621 | 14877
tokio multi-thread | 10000 | 14992 | 14542 | 15237
================================================================================
multi_thread_scaling: primes in [2, 400000) across 64 workers
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 29687 | 29554 | 31408
tokio multi 1-thread | 33860 | 28963 | 28742 | 30236
smarm smarm-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
RECURSE_DEPTH=500, HOT_YIELDS=500000×2, UNCONT_MSGS=1000000, PANIC_TASKS=10000
================================================================================
deep_recursion: depth 500
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1 | 83 | 80 | 128
smarm 1-thread | 1 | 86 | 77 | 149
tokio current_thread | 1 | 25 | 25 | 50
tokio multi-thread | 1 | 53 | 47 | 84
================================================================================
yield_in_hot_loop: 2 actors × 500000 yields (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 197474 | 194313 | 201690
tokio current_thread | 1000000 | 149289 | 148575 | 154319
================================================================================
uncontended_channel: 1→1, 1000000 msgs (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 26884 | 26675 | 29436
tokio current_thread | 1000000 | 52594 | 51941 | 54495
================================================================================
catch_unwind_panics: 10000 tasks, 50% panic
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 148321 | 146050 | 152943
smarm 1-thread | 10000 | 147961 | 144521 | 152158
tokio current_thread | 10000 | 264487 | 260848 | 274838
tokio multi-thread | 10000 | 272103 | 265687 | 285209

126
benches/baseline-output/sweep/ai512_tc300k.txt Normal file
View File

@@ -0,0 +1,126 @@
smarm general benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
CHAIN_DEPTH=1000, YIELD_TASKS=200×1000, PRIME_N=400000/64 workers, PP_ROUNDS=1000
================================================================================
chained_spawn: depth 1000
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 8574 | 8421 | 8729
smarm 1-thread | 1000 | 8675 | 8401 | 12686
tokio current_thread | 1000 | 125 | 125 | 148
tokio multi-thread | 1000 | 188 | 184 | 291
================================================================================
yield_many: 200 tasks × 1000 yields
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 200000 | 42389 | 41316 | 46466
smarm 1-thread | 200000 | 41776 | 41342 | 48940
tokio current_thread | 200000 | 15168 | 15094 | 15658
tokio multi-thread | 200000 | 15953 | 15862 | 17408
================================================================================
fan_out_compute: primes in [2, 400000) across 64
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 29680 | 29572 | 30661
smarm 1-thread | 33860 | 29816 | 29597 | 30401
tokio current_thread | 33860 | 28657 | 28581 | 29488
tokio multi-thread | 33860 | 34837 | 34529 | 37270
================================================================================
ping_pong_oneshot: 1000 rounds
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 16735 | 16601 | 17444
smarm 1-thread | 1000 | 16702 | 16500 | 17184
tokio current_thread | 1000 | 898 | 873 | 994
tokio multi-thread | 1000 | 4343 | 4241 | 4448
smarm tokio-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
STORM_BACKGROUND=8, STORM_SPAWN=10000, MPSC=32×10000, TIMER_ACTORS=10000 (1–10 ms), SCALING_N=400000/64
================================================================================
spawn_storm_busy: 8 bg yielders + 10000 zero-work spawns
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 128408 | 126199 | 133268
smarm 1-thread | 10000 | 131599 | 129387 | 135080
tokio current_thread | 10000 | 2718 | 2661 | 2981
tokio multi-thread | 10000 | 7264 | 4608 | 11583
================================================================================
mpsc_contention: 32 producers × 10000 msgs → 1 consumer
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 320000 | 9289 | 9039 | 9751
smarm 1-thread | 320000 | 9510 | 9157 | 9677
tokio current_thread | 320000 | 17550 | 17290 | 18578
tokio multi-thread | 320000 | 18336 | 17527 | 18989
================================================================================
many_timers: 10000 actors sleeping 1–10 ms
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 139111 | 136105 | 146606
smarm 1-thread | 10000 | 137302 | 133316 | 141350
tokio current_thread | 10000 | 13720 | 13455 | 14607
tokio multi-thread | 10000 | 14964 | 14546 | 15400
================================================================================
multi_thread_scaling: primes in [2, 400000) across 64 workers
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 30048 | 29705 | 31530
tokio multi 1-thread | 33860 | 28894 | 28682 | 30094
smarm smarm-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
RECURSE_DEPTH=500, HOT_YIELDS=500000×2, UNCONT_MSGS=1000000, PANIC_TASKS=10000
================================================================================
deep_recursion: depth 500
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1 | 93 | 81 | 161
smarm 1-thread | 1 | 103 | 80 | 178
tokio current_thread | 1 | 25 | 25 | 28
tokio multi-thread | 1 | 53 | 47 | 74
================================================================================
yield_in_hot_loop: 2 actors × 500000 yields (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 188726 | 187640 | 192658
tokio current_thread | 1000000 | 149332 | 148133 | 155745
================================================================================
uncontended_channel: 1→1, 1000000 msgs (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 27630 | 27086 | 29749
tokio current_thread | 1000000 | 54225 | 53355 | 56307
================================================================================
catch_unwind_panics: 10000 tasks, 50% panic
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 144934 | 143038 | 163552
smarm 1-thread | 10000 | 146614 | 143653 | 151325
tokio current_thread | 10000 | 266330 | 263523 | 271639
tokio multi-thread | 10000 | 274729 | 266323 | 285114

126
benches/baseline-output/sweep/ai64_tc150k.txt Normal file
View File

@@ -0,0 +1,126 @@
smarm general benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
CHAIN_DEPTH=1000, YIELD_TASKS=200×1000, PRIME_N=400000/64 workers, PP_ROUNDS=1000
================================================================================
chained_spawn: depth 1000
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 8849 | 8486 | 9224
smarm 1-thread | 1000 | 8841 | 8477 | 9108
tokio current_thread | 1000 | 124 | 124 | 219
tokio multi-thread | 1000 | 187 | 184 | 283
================================================================================
yield_many: 200 tasks × 1000 yields
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 200000 | 41681 | 41278 | 43685
smarm 1-thread | 200000 | 41721 | 41218 | 42261
tokio current_thread | 200000 | 14969 | 14940 | 15051
tokio multi-thread | 200000 | 16004 | 15868 | 17569
================================================================================
fan_out_compute: primes in [2, 400000) across 64
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 29679 | 29516 | 30105
smarm 1-thread | 33860 | 29677 | 29594 | 31365
tokio current_thread | 33860 | 28656 | 28572 | 29239
tokio multi-thread | 33860 | 34783 | 34617 | 36531
================================================================================
ping_pong_oneshot: 1000 rounds
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 17009 | 16822 | 17418
smarm 1-thread | 1000 | 16866 | 16723 | 17315
tokio current_thread | 1000 | 880 | 871 | 1035
tokio multi-thread | 1000 | 4263 | 4178 | 4391
smarm tokio-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
STORM_BACKGROUND=8, STORM_SPAWN=10000, MPSC=32×10000, TIMER_ACTORS=10000 (1–10 ms), SCALING_N=400000/64
================================================================================
spawn_storm_busy: 8 bg yielders + 10000 zero-work spawns
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 126566 | 124995 | 130402
smarm 1-thread | 10000 | 128278 | 126209 | 135156
tokio current_thread | 10000 | 2680 | 2640 | 2787
tokio multi-thread | 10000 | 7411 | 4393 | 12421
================================================================================
mpsc_contention: 32 producers × 10000 msgs → 1 consumer
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 320000 | 9073 | 8937 | 9324
smarm 1-thread | 320000 | 9120 | 9018 | 9263
tokio current_thread | 320000 | 17245 | 17180 | 17574
tokio multi-thread | 320000 | 18518 | 17685 | 19621
================================================================================
many_timers: 10000 actors sleeping 1–10 ms
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 141855 | 135415 | 145810
smarm 1-thread | 10000 | 138265 | 135535 | 142346
tokio current_thread | 10000 | 14441 | 13453 | 14650
tokio multi-thread | 10000 | 14956 | 14529 | 15451
================================================================================
multi_thread_scaling: primes in [2, 400000) across 64 workers
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 30033 | 29659 | 31803
tokio multi 1-thread | 33860 | 29078 | 28963 | 30231
smarm smarm-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
RECURSE_DEPTH=500, HOT_YIELDS=500000×2, UNCONT_MSGS=1000000, PANIC_TASKS=10000
================================================================================
deep_recursion: depth 500
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1 | 83 | 79 | 132
smarm 1-thread | 1 | 85 | 78 | 146
tokio current_thread | 1 | 25 | 25 | 73
tokio multi-thread | 1 | 51 | 47 | 64
================================================================================
yield_in_hot_loop: 2 actors × 500000 yields (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 191352 | 188830 | 196235
tokio current_thread | 1000000 | 152382 | 150674 | 187815
================================================================================
uncontended_channel: 1→1, 1000000 msgs (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 27552 | 27099 | 30612
tokio current_thread | 1000000 | 53160 | 52436 | 55255
================================================================================
catch_unwind_panics: 10000 tasks, 50% panic
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 145243 | 143291 | 173727
smarm 1-thread | 10000 | 145242 | 142819 | 148457
tokio current_thread | 10000 | 266471 | 262904 | 269145
tokio multi-thread | 10000 | 274195 | 269312 | 286111

126
benches/baseline-output/sweep/ai64_tc300k.txt Normal file
View File

@@ -0,0 +1,126 @@
smarm general benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
CHAIN_DEPTH=1000, YIELD_TASKS=200×1000, PRIME_N=400000/64 workers, PP_ROUNDS=1000
================================================================================
chained_spawn: depth 1000
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 8735 | 8508 | 9314
smarm 1-thread | 1000 | 8808 | 8506 | 10346
tokio current_thread | 1000 | 123 | 123 | 172
tokio multi-thread | 1000 | 190 | 184 | 273
================================================================================
yield_many: 200 tasks × 1000 yields
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 200000 | 41619 | 41255 | 43489
smarm 1-thread | 200000 | 41544 | 41196 | 43259
tokio current_thread | 200000 | 15382 | 15233 | 16007
tokio multi-thread | 200000 | 16095 | 15999 | 16296
================================================================================
fan_out_compute: primes in [2, 400000) across 64
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 30032 | 29838 | 31744
smarm 1-thread | 33860 | 29782 | 29653 | 30601
tokio current_thread | 33860 | 28754 | 28614 | 30700
tokio multi-thread | 33860 | 34988 | 34570 | 36871
================================================================================
ping_pong_oneshot: 1000 rounds
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000 | 17088 | 16868 | 18654
smarm 1-thread | 1000 | 16951 | 16797 | 17783
tokio current_thread | 1000 | 932 | 899 | 1019
tokio multi-thread | 1000 | 4340 | 4273 | 5245
smarm tokio-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
STORM_BACKGROUND=8, STORM_SPAWN=10000, MPSC=32×10000, TIMER_ACTORS=10000 (1–10 ms), SCALING_N=400000/64
================================================================================
spawn_storm_busy: 8 bg yielders + 10000 zero-work spawns
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 129009 | 127353 | 132990
smarm 1-thread | 10000 | 128009 | 126554 | 140472
tokio current_thread | 10000 | 2666 | 2624 | 2794
tokio multi-thread | 10000 | 5974 | 4368 | 11517
================================================================================
mpsc_contention: 32 producers × 10000 msgs → 1 consumer
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 320000 | 9044 | 8970 | 10788
smarm 1-thread | 320000 | 9087 | 8995 | 12500
tokio current_thread | 320000 | 17185 | 17072 | 18440
tokio multi-thread | 320000 | 17720 | 17394 | 19182
================================================================================
many_timers: 10000 actors sleeping 1–10 ms
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 145819 | 140671 | 150512
smarm 1-thread | 10000 | 139046 | 135846 | 146127
tokio current_thread | 10000 | 13866 | 13522 | 14670
tokio multi-thread | 10000 | 14900 | 14471 | 16378
================================================================================
multi_thread_scaling: primes in [2, 400000) across 64 workers
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 33860 | 30695 | 29720 | 33196
tokio multi 1-thread | 33860 | 29261 | 28895 | 31013
smarm smarm-favored benchmarks
available parallelism: 1 threads
ITERS=15 (+1 warmup, discarded)
RECURSE_DEPTH=500, HOT_YIELDS=500000×2, UNCONT_MSGS=1000000, PANIC_TASKS=10000
================================================================================
deep_recursion: depth 500
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1 | 82 | 79 | 113
smarm 1-thread | 1 | 85 | 78 | 143
tokio current_thread | 1 | 25 | 25 | 56
tokio multi-thread | 1 | 50 | 47 | 63
================================================================================
yield_in_hot_loop: 2 actors × 500000 yields (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 188698 | 187922 | 192263
tokio current_thread | 1000000 | 150231 | 148746 | 151723
================================================================================
uncontended_channel: 1→1, 1000000 msgs (single thread)
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 1000000 | 28461 | 27638 | 30283
tokio current_thread | 1000000 | 52224 | 51880 | 54732
================================================================================
catch_unwind_panics: 10000 tasks, 50% panic
================================================================================
runtime | result | median µs | min µs | max µs
--------------------------------------------------------------------------------
smarm 1-thread | 10000 | 144604 | 143246 | 145585
smarm 1-thread | 10000 | 148208 | 142691 | 151076
tokio current_thread | 10000 | 265255 | 260637 | 271065
tokio multi-thread | 10000 | 273131 | 271313 | 300420

224
benches/baseline.json Normal file
View File

@@ -0,0 +1,224 @@
{
"chained_spawn": {
"smarm 1-thread": {
"result": 1000,
"median": 8637,
"min": 8553,
"max": 8933
},
"tokio current_thread": {
"result": 1000,
"median": 124,
"min": 124,
"max": 153
},
"tokio multi-thread": {
"result": 1000,
"median": 188,
"min": 183,
"max": 229
}
},
"yield_many": {
"smarm 1-thread": {
"result": 200000,
"median": 41622,
"min": 41063,
"max": 44973
},
"tokio current_thread": {
"result": 200000,
"median": 15085,
"min": 15013,
"max": 15274
},
"tokio multi-thread": {
"result": 200000,
"median": 15964,
"min": 15880,
"max": 17959
}
},
"fan_out_compute": {
"smarm 1-thread": {
"result": 33860,
"median": 29727,
"min": 29491,
"max": 31634
},
"tokio current_thread": {
"result": 33860,
"median": 28503,
"min": 28391,
"max": 28866
},
"tokio multi-thread": {
"result": 33860,
"median": 34542,
"min": 34396,
"max": 36111
}
},
"ping_pong_oneshot": {
"smarm 1-thread": {
"result": 1000,
"median": 16848,
"min": 16633,
"max": 17301
},
"tokio current_thread": {
"result": 1000,
"median": 879,
"min": 868,
"max": 973
},
"tokio multi-thread": {
"result": 1000,
"median": 4328,
"min": 4223,
"max": 4461
}
},
"spawn_storm_busy": {
"smarm 1-thread": {
"result": 10000,
"median": 130058,
"min": 126790,
"max": 134475
},
"tokio current_thread": {
"result": 10000,
"median": 2772,
"min": 2641,
"max": 4367
},
"tokio multi-thread": {
"result": 10000,
"median": 7462,
"min": 4469,
"max": 12892
}
},
"mpsc_contention": {
"smarm 1-thread": {
"result": 320000,
"median": 9260,
"min": 9095,
"max": 10081
},
"tokio current_thread": {
"result": 320000,
"median": 17570,
"min": 17213,
"max": 18276
},
"tokio multi-thread": {
"result": 320000,
"median": 17593,
"min": 17452,
"max": 19564
}
},
"many_timers": {
"smarm 1-thread": {
"result": 10000,
"median": 135806,
"min": 132573,
"max": 141651
},
"tokio current_thread": {
"result": 10000,
"median": 14462,
"min": 13555,
"max": 15457
},
"tokio multi-thread": {
"result": 10000,
"median": 15011,
"min": 14655,
"max": 15368
}
},
"multi_thread_scaling": {
"smarm 1-thread": {
"result": 33860,
"median": 30029,
"min": 29720,
"max": 31351
},
"tokio multi 1-thread": {
"result": 33860,
"median": 28983,
"min": 28908,
"max": 29323
}
},
"deep_recursion": {
"smarm 1-thread": {
"result": 1,
"median": 83,
"min": 78,
"max": 587
},
"tokio current_thread": {
"result": 1,
"median": 25,
"min": 25,
"max": 33
},
"tokio multi-thread": {
"result": 1,
"median": 59,
"min": 47,
"max": 205
}
},
"yield_in_hot_loop": {
"smarm 1-thread": {
"result": 1000000,
"median": 188753,
"min": 187007,
"max": 194366
},
"tokio current_thread": {
"result": 1000000,
"median": 153929,
"min": 152712,
"max": 158749
}
},
"uncontended_channel": {
"smarm 1-thread": {
"result": 1000000,
"median": 26811,
"min": 26498,
"max": 29069
},
"tokio current_thread": {
"result": 1000000,
"median": 51888,
"min": 51530,
"max": 52708
}
},
"catch_unwind_panics": {
"smarm 1-thread": {
"result": 10000,
"median": 142215,
"min": 140189,
"max": 143570
},
"tokio current_thread": {
"result": 10000,
"median": 682295,
"min": 670281,
"max": 700774
},
"tokio multi-thread": {
"result": 10000,
"median": 662688,
"min": 641453,
"max": 681868
}
}
}

442
benches/general.rs Normal file
View File

@@ -0,0 +1,442 @@
//! General benchmarks — workloads where neither runtime has a structural
//! advantage. Both should be competitive; large gaps here indicate a real
//! difference in per-task or per-yield overhead.
//!
//! Workloads:
//! 1. chained_spawn — task N spawns N+1, depth 1000. Spawn+exit overhead in
//! a serial chain. Adapted from tokio's bench of the same
//! name.
//! 2. yield_many — 200 actors × 1000 yields. Pure scheduling throughput
//! with no allocation, no IO. Adapted from tokio.
//! 3. fan_out_compute— count primes in [2, 400_000) across 64 workers. Same
//! shape as multi_scheduler::primes but lives here for
//! completeness.
//! 4. ping_pong_oneshot — N rounds of (spawn pair, send oneshot, await).
//! Closer to a request/response workload than channel
//! ping-pong.
use std::sync::atomic::{AtomicU64, Ordering};
use std::sync::Arc;
use std::time::Instant;
// ---------------------------------------------------------------------------
// Shared harness
// ---------------------------------------------------------------------------
const ITERS: u32 = 15;
fn available_threads() -> usize {
std::thread::available_parallelism().map(|n| n.get()).unwrap_or(1)
}
fn print_header(title: &str) {
println!("\n{}", "=".repeat(80));
println!(" {title}");
println!("{}", "=".repeat(80));
println!(
"{:>26} | {:>12} | {:>10} | {:>10} | {:>10}",
"runtime", "result", "median µs", "min µs", "max µs"
);
println!("{}", "-".repeat(80));
}
fn run_n<F: FnMut() -> (u64, u128)>(name: &str, n: u32, mut f: F) {
let mut times = Vec::new();
let mut last = 0u64;
// One warmup iteration, discarded.
let _ = f();
for _ in 0..n {
let (v, t) = f();
times.push(t);
last = v;
}
times.sort_unstable();
let median = times[times.len() / 2];
let min = *times.iter().min().unwrap();
let max = *times.iter().max().unwrap();
println!(
"{:>26} | {:>12} | {:>10} | {:>10} | {:>10}",
name, last, median, min, max
);
}
// ---------------------------------------------------------------------------
// 1. chained_spawn — depth 1000
// ---------------------------------------------------------------------------
const CHAIN_DEPTH: u64 = 1_000;
fn bench_chained_smarm(threads: usize) -> (u64, u128) {
let counter = Arc::new(AtomicU64::new(0));
let c2 = counter.clone();
let start = Instant::now();
smarm::runtime::init(bench_cfg(threads)).run(move || {
// Fire-and-forget chain, matching tokio's bench shape: each link
// spawns the next link and exits immediately; depth 0 signals done
// via a channel. Crucially this does *not* nest joins on the
// spawner's stack — important because smarm actor stacks are a
// fixed 64 KiB.
let (tx, rx) = smarm::channel::<()>();
fn iter(c: Arc<AtomicU64>, tx: smarm::Sender<()>, n: u64) {
if n == 0 {
tx.send(()).unwrap();
} else {
let cc = c.clone();
smarm::spawn(move || {
cc.fetch_add(1, Ordering::Relaxed);
iter(cc.clone(), tx, n - 1);
});
// Caller exits; JoinHandle dropped, no parking.
}
}
iter(c2, tx, CHAIN_DEPTH);
rx.recv().unwrap();
});
(counter.load(Ordering::Relaxed), start.elapsed().as_micros())
}
fn bench_chained_tokio_current() -> (u64, u128) {
let counter = Arc::new(AtomicU64::new(0));
let c2 = counter.clone();
let rt = tokio::runtime::Builder::new_current_thread().build().unwrap();
let start = Instant::now();
let local = tokio::task::LocalSet::new();
local.block_on(&rt, async move {
// Use a oneshot done channel like tokio's own chained_spawn bench.
let (done_tx, done_rx) = tokio::sync::oneshot::channel();
fn iter(
c: Arc<AtomicU64>,
done: tokio::sync::oneshot::Sender<()>,
n: u64,
) {
if n == 0 {
let _ = done.send(());
} else {
tokio::task::spawn_local(async move {
c.fetch_add(1, Ordering::Relaxed);
iter(c, done, n - 1);
});
}
}
iter(c2, done_tx, CHAIN_DEPTH);
let _ = done_rx.await;
});
(counter.load(Ordering::Relaxed), start.elapsed().as_micros())
}
fn bench_chained_tokio_multi() -> (u64, u128) {
let counter = Arc::new(AtomicU64::new(0));
let c2 = counter.clone();
let rt = tokio::runtime::Builder::new_multi_thread()
.worker_threads(available_threads())
.build()
.unwrap();
let start = Instant::now();
rt.block_on(async move {
let (done_tx, done_rx) = tokio::sync::oneshot::channel();
fn iter(c: Arc<AtomicU64>, done: tokio::sync::oneshot::Sender<()>, n: u64) {
if n == 0 {
let _ = done.send(());
} else {
tokio::spawn(async move {
c.fetch_add(1, Ordering::Relaxed);
iter(c, done, n - 1);
});
}
}
iter(c2, done_tx, CHAIN_DEPTH);
let _ = done_rx.await;
});
(counter.load(Ordering::Relaxed), start.elapsed().as_micros())
}
// ---------------------------------------------------------------------------
// 2. yield_many — 200 actors × 1000 yields
// ---------------------------------------------------------------------------
const YIELD_TASKS: u64 = 200;
const YIELD_ROUNDS: u64 = 1_000;
fn bench_yield_smarm(threads: usize) -> (u64, u128) {
let start = Instant::now();
smarm::runtime::init(bench_cfg(threads)).run(|| {
let mut handles = Vec::new();
for _ in 0..YIELD_TASKS {
handles.push(smarm::spawn(|| {
for _ in 0..YIELD_ROUNDS {
smarm::yield_now();
}
}));
}
for h in handles {
h.join().unwrap();
}
});
(YIELD_TASKS * YIELD_ROUNDS, start.elapsed().as_micros())
}
fn bench_yield_tokio_current() -> (u64, u128) {
let rt = tokio::runtime::Builder::new_current_thread().build().unwrap();
let start = Instant::now();
let local = tokio::task::LocalSet::new();
local.block_on(&rt, async move {
let mut handles = Vec::new();
for _ in 0..YIELD_TASKS {
handles.push(tokio::task::spawn_local(async move {
for _ in 0..YIELD_ROUNDS {
tokio::task::yield_now().await;
}
}));
}
for h in handles {
let _ = h.await;
}
});
(YIELD_TASKS * YIELD_ROUNDS, start.elapsed().as_micros())
}
fn bench_yield_tokio_multi() -> (u64, u128) {
let rt = tokio::runtime::Builder::new_multi_thread()
.worker_threads(available_threads())
.build()
.unwrap();
let start = Instant::now();
rt.block_on(async move {
let mut handles = Vec::new();
for _ in 0..YIELD_TASKS {
handles.push(tokio::spawn(async move {
for _ in 0..YIELD_ROUNDS {
tokio::task::yield_now().await;
}
}));
}
for h in handles {
let _ = h.await;
}
});
(YIELD_TASKS * YIELD_ROUNDS, start.elapsed().as_micros())
}
// ---------------------------------------------------------------------------
// 3. fan_out_compute — primes, same shape as multi_scheduler::primes
// ---------------------------------------------------------------------------
const PRIME_N: u64 = 400_000;
const PRIME_WORKERS: u64 = 64;
fn is_prime(n: u64) -> bool {
if n < 2 { return false; }
if n < 4 { return true; }
if n % 2 == 0 { return false; }
let mut i = 3u64;
while i * i <= n { if n % i == 0 { return false; } i += 2; }
true
}
fn count_primes(lo: u64, hi: u64) -> u64 {
(lo..hi).filter(|&n| is_prime(n)).count() as u64
}
fn primes_slice(w: u64) -> (u64, u64) {
let per = PRIME_N / PRIME_WORKERS;
let lo = w * per;
let hi = if w + 1 == PRIME_WORKERS { PRIME_N } else { lo + per };
(lo, hi)
}
fn bench_primes_smarm(threads: usize) -> (u64, u128) {
let total = Arc::new(AtomicU64::new(0));
let t2 = total.clone();
let start = Instant::now();
smarm::runtime::init(bench_cfg(threads)).run(move || {
let mut handles = Vec::new();
for w in 0..PRIME_WORKERS {
let (lo, hi) = primes_slice(w);
let tc = t2.clone();
handles.push(smarm::spawn(move || {
tc.fetch_add(count_primes(lo, hi), Ordering::Relaxed);
}));
}
for h in handles { h.join().unwrap(); }
});
(total.load(Ordering::Relaxed), start.elapsed().as_micros())
}
fn bench_primes_tokio_current() -> (u64, u128) {
let total = Arc::new(AtomicU64::new(0));
let t2 = total.clone();
let rt = tokio::runtime::Builder::new_current_thread().build().unwrap();
let start = Instant::now();
let local = tokio::task::LocalSet::new();
local.block_on(&rt, async move {
let mut handles = Vec::new();
for w in 0..PRIME_WORKERS {
let (lo, hi) = primes_slice(w);
let tc = t2.clone();
handles.push(tokio::task::spawn_local(async move {
tc.fetch_add(count_primes(lo, hi), Ordering::Relaxed);
}));
}
for h in handles { let _ = h.await; }
});
(total.load(Ordering::Relaxed), start.elapsed().as_micros())
}
fn bench_primes_tokio_multi() -> (u64, u128) {
let total = Arc::new(AtomicU64::new(0));
let t2 = total.clone();
let rt = tokio::runtime::Builder::new_multi_thread()
.worker_threads(available_threads())
.build()
.unwrap();
let start = Instant::now();
rt.block_on(async move {
let mut handles = Vec::new();
for w in 0..PRIME_WORKERS {
let (lo, hi) = primes_slice(w);
let tc = t2.clone();
handles.push(tokio::spawn(async move {
tc.fetch_add(count_primes(lo, hi), Ordering::Relaxed);
}));
}
for h in handles { let _ = h.await; }
});
(total.load(Ordering::Relaxed), start.elapsed().as_micros())
}
// ---------------------------------------------------------------------------
// 4. ping_pong_oneshot — 1000 rounds of spawn-pair-await
// ---------------------------------------------------------------------------
const PP_ROUNDS: u64 = 1_000;
fn bench_pp_smarm(threads: usize) -> (u64, u128) {
let start = Instant::now();
smarm::runtime::init(bench_cfg(threads)).run(|| {
for _ in 0..PP_ROUNDS {
// smarm has no oneshot, so use a channel<()> per round — both
// sides spawn, A sends ping, B replies pong, A joins B.
let (tx_ping, rx_ping) = smarm::channel::<()>();
let (tx_pong, rx_pong) = smarm::channel::<()>();
let hb = smarm::spawn(move || {
rx_ping.recv().unwrap();
tx_pong.send(()).unwrap();
});
let ha = smarm::spawn(move || {
tx_ping.send(()).unwrap();
rx_pong.recv().unwrap();
});
ha.join().unwrap();
hb.join().unwrap();
}
});
(PP_ROUNDS, start.elapsed().as_micros())
}
fn bench_pp_tokio_current() -> (u64, u128) {
let rt = tokio::runtime::Builder::new_current_thread().build().unwrap();
let start = Instant::now();
let local = tokio::task::LocalSet::new();
local.block_on(&rt, async move {
for _ in 0..PP_ROUNDS {
let (tx1, rx1) = tokio::sync::oneshot::channel::<()>();
let (tx2, rx2) = tokio::sync::oneshot::channel::<()>();
let hb = tokio::task::spawn_local(async move {
rx1.await.unwrap();
tx2.send(()).unwrap();
});
let ha = tokio::task::spawn_local(async move {
tx1.send(()).unwrap();
rx2.await.unwrap();
});
let _ = ha.await;
let _ = hb.await;
}
});
(PP_ROUNDS, start.elapsed().as_micros())
}
fn bench_pp_tokio_multi() -> (u64, u128) {
let rt = tokio::runtime::Builder::new_multi_thread()
.worker_threads(available_threads())
.build()
.unwrap();
let start = Instant::now();
rt.block_on(async move {
for _ in 0..PP_ROUNDS {
let (tx1, rx1) = tokio::sync::oneshot::channel::<()>();
let (tx2, rx2) = tokio::sync::oneshot::channel::<()>();
let hb = tokio::spawn(async move {
rx1.await.unwrap();
tx2.send(()).unwrap();
});
let ha = tokio::spawn(async move {
tx1.send(()).unwrap();
rx2.await.unwrap();
});
let _ = ha.await;
let _ = hb.await;
}
});
(PP_ROUNDS, start.elapsed().as_micros())
}
// ---------------------------------------------------------------------------
// main
// ---------------------------------------------------------------------------
// ---------------------------------------------------------------------------
// Knob helper — reads SMARM_ALLOC_INTERVAL / SMARM_TIMESLICE_CYCLES env vars
// so the sweep script can override the preemption knobs without recompiling.
// ---------------------------------------------------------------------------
fn bench_cfg(threads: usize) -> smarm::runtime::Config {
let mut cfg = smarm::runtime::Config::exact(threads);
if let Ok(v) = std::env::var("SMARM_ALLOC_INTERVAL") {
if let Ok(n) = v.parse::<u32>() { cfg = cfg.alloc_interval(n); }
}
if let Ok(v) = std::env::var("SMARM_TIMESLICE_CYCLES") {
if let Ok(n) = v.parse::<u64>() { cfg = cfg.timeslice_cycles(n); }
}
cfg
}
fn main() {
let n = available_threads();
println!("smarm general benchmarks");
println!("available parallelism: {n} threads");
println!("ITERS={ITERS} (+1 warmup, discarded)");
println!(
"CHAIN_DEPTH={CHAIN_DEPTH}, YIELD_TASKS={YIELD_TASKS}×{YIELD_ROUNDS}, \
PRIME_N={PRIME_N}/{PRIME_WORKERS} workers, PP_ROUNDS={PP_ROUNDS}"
);
// ---- 1. chained_spawn ----
print_header(&format!("chained_spawn: depth {CHAIN_DEPTH}"));
run_n("smarm 1-thread", ITERS, || bench_chained_smarm(1));
run_n(&format!("smarm {n}-thread"), ITERS, || bench_chained_smarm(n));
run_n("tokio current_thread", ITERS, bench_chained_tokio_current);
run_n("tokio multi-thread", ITERS, bench_chained_tokio_multi);
// ---- 2. yield_many ----
print_header(&format!("yield_many: {YIELD_TASKS} tasks × {YIELD_ROUNDS} yields"));
run_n("smarm 1-thread", ITERS, || bench_yield_smarm(1));
run_n(&format!("smarm {n}-thread"), ITERS, || bench_yield_smarm(n));
run_n("tokio current_thread", ITERS, bench_yield_tokio_current);
run_n("tokio multi-thread", ITERS, bench_yield_tokio_multi);
// ---- 3. fan_out_compute ----
print_header(&format!("fan_out_compute: primes in [2, {PRIME_N}) across {PRIME_WORKERS}"));
run_n("smarm 1-thread", ITERS, || bench_primes_smarm(1));
run_n(&format!("smarm {n}-thread"), ITERS, || bench_primes_smarm(n));
run_n("tokio current_thread", ITERS, bench_primes_tokio_current);
run_n("tokio multi-thread", ITERS, bench_primes_tokio_multi);
// ---- 4. ping_pong_oneshot ----
print_header(&format!("ping_pong_oneshot: {PP_ROUNDS} rounds"));
run_n("smarm 1-thread", ITERS, || bench_pp_smarm(1));
run_n(&format!("smarm {n}-thread"), ITERS, || bench_pp_smarm(n));
run_n("tokio current_thread", ITERS, bench_pp_tokio_current);
run_n("tokio multi-thread", ITERS, bench_pp_tokio_multi);
}

25
benches/smarm_favored.rs
View File

@@ -84,7 +84,7 @@ fn bench_recurse_smarm(threads: usize) -> (u64, u128) {
let total = Arc::new(AtomicU64::new(0));
let t2 = total.clone();
let start = Instant::now();
smarm::runtime::init(smarm::runtime::Config::exact(threads)).run(move || {
smarm::runtime::init(bench_cfg(threads)).run(move || {
// Plain Rust recursion on the actor's own (growable) stack.
fn recurse(c: &AtomicU64, n: u64) -> u64 {
if n == 0 {
@@ -170,7 +170,7 @@ const HOT_YIELDS: u64 = 500_000;
fn bench_hot_smarm() -> (u64, u128) {
let start = Instant::now();
smarm::runtime::init(smarm::runtime::Config::exact(1)).run(|| {
smarm::runtime::init(bench_cfg(1)).run(|| {
let ha = smarm::spawn(|| {
for _ in 0..HOT_YIELDS {
smarm::yield_now();
@@ -216,7 +216,7 @@ const UNCONT_MSGS: u64 = 1_000_000;
fn bench_unc_smarm() -> (u64, u128) {
let start = Instant::now();
smarm::runtime::init(smarm::runtime::Config::exact(1)).run(|| {
smarm::runtime::init(bench_cfg(1)).run(|| {
let (tx, rx) = smarm::channel::<u64>();
let consumer = smarm::spawn(move || {
let mut count = 0u64;
@@ -273,7 +273,7 @@ fn bench_panic_smarm(threads: usize) -> (u64, u128) {
let ok2 = ok.clone();
let err2 = err.clone();
let start = Instant::now();
smarm::runtime::init(smarm::runtime::Config::exact(threads)).run(move || {
smarm::runtime::init(bench_cfg(threads)).run(move || {
let mut handles = Vec::new();
for i in 0..PANIC_TASKS {
handles.push(smarm::spawn(move || {
@@ -355,6 +355,23 @@ fn bench_panic_tokio_multi() -> (u64, u128) {
// main
// ---------------------------------------------------------------------------
// ---------------------------------------------------------------------------
// Knob helper — reads SMARM_ALLOC_INTERVAL / SMARM_TIMESLICE_CYCLES env vars
// so the sweep script can override the preemption knobs without recompiling.
// ---------------------------------------------------------------------------
fn bench_cfg(threads: usize) -> smarm::runtime::Config {
let mut cfg = smarm::runtime::Config::exact(threads);
if let Ok(v) = std::env::var("SMARM_ALLOC_INTERVAL") {
if let Ok(n) = v.parse::<u32>() { cfg = cfg.alloc_interval(n); }
}
if let Ok(v) = std::env::var("SMARM_TIMESLICE_CYCLES") {
if let Ok(n) = v.parse::<u64>() { cfg = cfg.timeslice_cycles(n); }
}
cfg
}
fn main() {
let n = available_threads();
println!("smarm smarm-favored benchmarks");

347
benches/sweep.py Executable file
View File

@@ -0,0 +1,347 @@
#!/usr/bin/env python3
"""
smarm bench sweep + regression checker.
Usage:
# Run a full knob sweep and print a comparison table:
python3 benches/sweep.py sweep
# Check the current build against the committed baseline:
python3 benches/sweep.py regress
# Run all benches once (default knobs) and print results:
python3 benches/sweep.py run
The sweep grid is defined in SWEEP_GRID below.
The regression baseline is loaded from benches/baseline.json.
"""
import argparse
import json
import os
import re
import subprocess
import sys
from pathlib import Path
# ---------------------------------------------------------------------------
# Configuration
# ---------------------------------------------------------------------------
REPO = Path(__file__).resolve().parent.parent
# Bench files to run (primes + multi_scheduler omitted — legacy harness,
# not part of the 12-bench suite, and insensitive to the preemption knobs).
BENCHES = ["general", "tokio_favored", "smarm_favored"]
# Knob sweep grid: (alloc_interval, timeslice_cycles)
# alloc_interval: lower = check RDTSC more often = finer preemption
# timeslice_cycles: lower = shorter timeslice = more cooperative
SWEEP_GRID = [
(32, 150_000),
(64, 150_000),
(128, 150_000), # default interval, shorter slice
(32, 300_000),
(64, 300_000),
(128, 300_000), # <<< baseline (defaults)
(256, 300_000),
(512, 300_000),
(128, 600_000),
(128, 1_200_000),
]
# Regression threshold: warn if median is more than this % worse than baseline.
REGRESSION_THRESHOLD_PCT = 10
# ---------------------------------------------------------------------------
# Parsing
# ---------------------------------------------------------------------------
# Match lines like:
# " smarm 1-thread | 1000000 | 31473 | 28719 | 33113"
ROW_RE = re.compile(
r"^\s*(?P<name>[^|]+?)\s*\|\s*(?P<result>\d+)\s*\|\s*(?P<median>\d+)\s*\|\s*(?P<min>\d+)\s*\|\s*(?P<max>\d+)\s*$"
)
# Match section headers like:
# " chained_spawn: depth 1000"
HEADER_RE = re.compile(r"^\s{2}(?P<bench>[a-z_]+)[:—]")
def parse_output(text: str) -> dict[str, dict[str, dict]]:
"""
Returns {bench_name: {runtime_label: {median, min, max, result}}}.
bench_name is the snake_case name extracted from the section header.
"""
results: dict[str, dict[str, dict]] = {}
current_bench = None
for line in text.splitlines():
hm = HEADER_RE.match(line)
if hm:
current_bench = hm.group("bench")
results.setdefault(current_bench, {})
continue
if current_bench is None:
continue
rm = ROW_RE.match(line)
if rm:
label = rm.group("name").strip()
results[current_bench][label] = {
"result": int(rm.group("result")),
"median": int(rm.group("median")),
"min": int(rm.group("min")),
"max": int(rm.group("max")),
}
return results
# ---------------------------------------------------------------------------
# Running
# ---------------------------------------------------------------------------
def run_benches(env_extra: dict[str, str] | None = None) -> dict[str, dict[str, dict]]:
"""Run all BENCHES and return merged parsed results."""
env = os.environ.copy()
if env_extra:
env.update(env_extra)
all_results: dict[str, dict[str, dict]] = {}
for bench in BENCHES:
cmd = ["cargo", "bench", "--bench", bench]
proc = subprocess.run(
cmd,
cwd=REPO,
env=env,
capture_output=True,
text=True,
)
if proc.returncode != 0:
print(f" ERROR running {bench}:\n{proc.stderr[-800:]}", file=sys.stderr)
continue
parsed = parse_output(proc.stdout)
all_results.update(parsed)
return all_results
# ---------------------------------------------------------------------------
# Baseline JSON
# ---------------------------------------------------------------------------
BASELINE_PATH = REPO / "benches" / "baseline.json"
def load_baseline() -> dict:
if not BASELINE_PATH.exists():
sys.exit(
f"No baseline found at {BASELINE_PATH}.\n"
"Run: python3 benches/sweep.py run then save the output manually,\n"
"or use --save-baseline with the run subcommand."
)
return json.loads(BASELINE_PATH.read_text())
def save_baseline(results: dict) -> None:
BASELINE_PATH.write_text(json.dumps(results, indent=2))
print(f"Baseline saved to {BASELINE_PATH}")
# ---------------------------------------------------------------------------
# Regression check
# ---------------------------------------------------------------------------
def check_regressions(current: dict, baseline: dict) -> bool:
"""
Compare current results to baseline. Print warnings for regressions.
Returns True if any regression found.
"""
any_regression = False
for bench, runtimes in baseline.items():
cur_bench = current.get(bench, {})
for label, base_data in runtimes.items():
cur_data = cur_bench.get(label)
if cur_data is None:
print(f" MISSING {bench}/{label} — not present in current run")
any_regression = True
continue
base_med = base_data["median"]
cur_med = cur_data["median"]
if base_med == 0:
continue
pct = (cur_med - base_med) / base_med * 100
if pct > REGRESSION_THRESHOLD_PCT:
print(
f" REGRESSION {bench}/{label}: "
f"{base_med} → {cur_med} µs ({pct:+.1f}%)"
)
any_regression = True
elif pct < -REGRESSION_THRESHOLD_PCT:
print(
f" IMPROVEMENT {bench}/{label}: "
f"{base_med} → {cur_med} µs ({pct:+.1f}%)"
)
return any_regression
# ---------------------------------------------------------------------------
# Pretty print
# ---------------------------------------------------------------------------
def print_results(results: dict, label: str = "") -> None:
if label:
print(f"\n{'='*70}")
print(f" {label}")
print(f"{'='*70}")
for bench, runtimes in sorted(results.items()):
print(f"\n [{bench}]")
print(f" {'runtime':>28} | {'result':>10} | {'median µs':>10} | {'min':>8} | {'max':>8}")
print(f" {'-'*75}")
for rt_label, data in runtimes.items():
print(
f" {rt_label:>28} | {data['result']:>10} | "
f"{data['median']:>10} | {data['min']:>8} | {data['max']:>8}"
)
def print_sweep_table(sweep_results: list[tuple[int, int, dict]]) -> None:
"""Print a compact comparison across sweep points for each bench/runtime."""
# Collect all bench/label pairs
all_keys: list[tuple[str, str]] = []
for _, _, results in sweep_results:
for bench, runtimes in results.items():
for label in runtimes:
key = (bench, label)
if key not in all_keys:
all_keys.append(key)
# Header
col_w = 12
print(f"\n{'bench/runtime':<45}", end="")
for interval, cycles, _ in sweep_results:
tag = f"ai={interval}/tc={cycles//1000}k"
print(f" {tag:>{col_w}}", end="")
print()
print("-" * (45 + (col_w + 2) * len(sweep_results)))
for bench, label in all_keys:
key_str = f"{bench}/{label}"
print(f" {key_str:<43}", end="")
for _, _, results in sweep_results:
val = results.get(bench, {}).get(label, {}).get("median")
cell = str(val) if val is not None else "—"
print(f" {cell:>{col_w}}", end="")
print()
# ---------------------------------------------------------------------------
# Subcommands
# ---------------------------------------------------------------------------
def cmd_run(args) -> None:
print("Building release binaries…")
subprocess.run(
["cargo", "build", "--release", "--benches"],
cwd=REPO, check=True, capture_output=True,
)
print("Running benches…")
results = run_benches()
print_results(results, "Results (default knobs)")
if args.save_baseline:
save_baseline(results)
def cmd_regress(args) -> None:
baseline = load_baseline()
print("Building release binaries…")
subprocess.run(
["cargo", "build", "--release", "--benches"],
cwd=REPO, check=True, capture_output=True,
)
print("Running benches…")
current = run_benches()
print_results(current, "Current results")
print(f"\nRegression check (threshold: >{REGRESSION_THRESHOLD_PCT}% slower than baseline)")
print("-" * 60)
found = check_regressions(current, baseline)
if not found:
print(" No regressions detected.")
sys.exit(1 if found else 0)
def cmd_sweep(args) -> None:
print("Building release binaries (once)…")
subprocess.run(
["cargo", "build", "--release", "--benches"],
cwd=REPO, check=True, capture_output=True,
)
# Benches are pre-built; env vars change runtime behaviour, no recompile needed.
sweep_results: list[tuple[int, int, dict]] = []
for interval, cycles in SWEEP_GRID:
tag = f"alloc_interval={interval}, timeslice_cycles={cycles}"
print(f" Running: {tag} …", flush=True)
env_extra = {
"SMARM_ALLOC_INTERVAL": str(interval),
"SMARM_TIMESLICE_CYCLES": str(cycles),
}
results = run_benches(env_extra)
sweep_results.append((interval, cycles, results))
print_sweep_table(sweep_results)
if args.save_csv:
import csv
rows = []
for interval, cycles, results in sweep_results:
for bench, runtimes in results.items():
for label, data in runtimes.items():
rows.append({
"alloc_interval": interval,
"timeslice_cycles": cycles,
"bench": bench,
"runtime": label,
**data,
})
with open(args.save_csv, "w", newline="") as f:
writer = csv.DictWriter(f, fieldnames=rows[0].keys())
writer.writeheader()
writer.writerows(rows)
print(f"\nCSV saved to {args.save_csv}")
# ---------------------------------------------------------------------------
# Entry point
# ---------------------------------------------------------------------------
def main() -> None:
parser = argparse.ArgumentParser(description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter)
sub = parser.add_subparsers(dest="cmd", required=True)
p_run = sub.add_parser("run", help="Run benches once with default knobs")
p_run.add_argument("--save-baseline", action="store_true",
help="Save results as the regression baseline")
p_run.set_defaults(func=cmd_run)
p_reg = sub.add_parser("regress", help="Check current results against baseline")
p_reg.set_defaults(func=cmd_regress)
p_sw = sub.add_parser("sweep", help="Sweep preemption knobs and compare")
p_sw.add_argument("--save-csv", metavar="FILE",
help="Write full sweep results to a CSV file")
p_sw.set_defaults(func=cmd_sweep)
args = parser.parse_args()
args.func(args)
if __name__ == "__main__":
main()

25
benches/tokio_favored.rs
View File

@@ -83,7 +83,7 @@ fn bench_storm_smarm(threads: usize) -> (u64, u128) {
let s2 = stop.clone();
let start = Instant::now();
smarm::runtime::init(smarm::runtime::Config::exact(threads)).run(move || {
smarm::runtime::init(bench_cfg(threads)).run(move || {
// Background actors: yield in a tight loop until told to stop.
let mut bg_handles = Vec::new();
for _ in 0..STORM_BACKGROUND {
@@ -189,7 +189,7 @@ const MPSC_PER_PRODUCER: u64 = 10_000;
fn bench_mpsc_smarm(threads: usize) -> (u64, u128) {
let start = Instant::now();
smarm::runtime::init(smarm::runtime::Config::exact(threads)).run(|| {
smarm::runtime::init(bench_cfg(threads)).run(|| {
let (tx, rx) = smarm::channel::<u64>();
let mut prod_handles = Vec::new();
for p in 0..MPSC_PRODUCERS {
@@ -289,7 +289,7 @@ fn timer_delay_ms(i: u64) -> u64 {
fn bench_timers_smarm(threads: usize) -> (u64, u128) {
let start = Instant::now();
smarm::runtime::init(smarm::runtime::Config::exact(threads)).run(|| {
smarm::runtime::init(bench_cfg(threads)).run(|| {
let mut handles = Vec::new();
for i in 0..TIMER_ACTORS {
let ms = timer_delay_ms(i);
@@ -373,7 +373,7 @@ fn bench_scaling_smarm(threads: usize) -> (u64, u128) {
let total = Arc::new(AtomicU64::new(0));
let t2 = total.clone();
let start = Instant::now();
smarm::runtime::init(smarm::runtime::Config::exact(threads)).run(move || {
smarm::runtime::init(bench_cfg(threads)).run(move || {
let mut handles = Vec::new();
for w in 0..SCALING_WORKERS {
let (lo, hi) = scaling_slice(w);
@@ -413,6 +413,23 @@ fn bench_scaling_tokio_multi(threads: usize) -> (u64, u128) {
// main
// ---------------------------------------------------------------------------
// ---------------------------------------------------------------------------
// Knob helper — reads SMARM_ALLOC_INTERVAL / SMARM_TIMESLICE_CYCLES env vars
// so the sweep script can override the preemption knobs without recompiling.
// ---------------------------------------------------------------------------
fn bench_cfg(threads: usize) -> smarm::runtime::Config {
let mut cfg = smarm::runtime::Config::exact(threads);
if let Ok(v) = std::env::var("SMARM_ALLOC_INTERVAL") {
if let Ok(n) = v.parse::<u32>() { cfg = cfg.alloc_interval(n); }
}
if let Ok(v) = std::env::var("SMARM_TIMESLICE_CYCLES") {
if let Ok(n) = v.parse::<u64>() { cfg = cfg.timeslice_cycles(n); }
}
cfg
}
fn main() {
let n = available_threads();
println!("smarm tokio-favored benchmarks");