benches: expose preemption knobs + sweep runner
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
126
benches/baseline-output/sweep/ai128_tc1200k.txt
Normal file
126
benches/baseline-output/sweep/ai128_tc1200k.txt
Normal 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
|
||||
Reference in New Issue
Block a user