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/ai32_tc300k.txt
Normal file
126
benches/baseline-output/sweep/ai32_tc300k.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 | 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
|
||||
Reference in New Issue
Block a user