YJIT Benchmarks

Details for Benchmarks at 2024-02-19 19:08:14 UTC

YJIT metrics from the yjit-bench suite using Ruby 391b0bf07b

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 67.4% faster than CRuby 3.4.0dev
On railsbench it is
  • 75.5% faster than CRuby 3.4.0dev

Performance on Headline Benchmarks

Speed of each Ruby implementation relative to the baseline CRuby measurement. Higher is better.

Memory Usage on Headline Benchmarks

Memory usage of each Ruby implementation relative to the baseline CRuby measurement. Lower is better.

Performance on Other Benchmarks

Speed of each Ruby implementation relative to the baseline CRuby measurement. Higher is better.

Memory Usage on Other Benchmarks

Memory usage of each Ruby implementation relative to the baseline CRuby measurement. Lower is better.

Performance on MicroBenchmarks

Speed of each Ruby implementation relative to the baseline CRuby measurement. Higher is better.

Memory Usage on MicroBenchmarks

Memory usage of each Ruby implementation relative to the baseline CRuby measurement. Lower is better.

Want Raw Graphs and CSV?

Benchmarks Speed Details

Benchmark Memory Usage Details

Number of Iterations and Warmups Tested

bench CRuby 3.4.0dev warmups CRuby 3.4.0dev iters YJIT 3.4.0dev warmups YJIT 3.4.0dev iters
activerecord 30 521 30 521
chunky-png 30 31 30 31
erubi-rails 30 19 30 19
hexapdf 30 15 30 15
liquid-c 30 350 30 350
liquid-compile 30 338 30 338
liquid-render 30 251 30 251
lobsters 30 23 30 23
mail 30 161 30 161
psych-load 30 15 30 15
railsbench 30 15 30 15
rubocop 30 184 30 184
ruby-lsp 30 174 30 174
sequel 30 291 30 291
binarytrees 30 82 30 82
blurhash 30 100 30 100
erubi 30 77 30 77
etanni 30 51 30 51
fannkuchredux 30 23 30 23
fluentd 30 15 30 15
graphql 30 15 30 15
graphql-native 30 22 30 22
lee 30 20 30 20
matmul 30 17 30 17
nbody 30 313 30 313
nqueens 30 292 30 292
optcarrot 30 15 30 15
rack 30 172 30 172
ruby-json 30 15 30 15
rubykon 30 15 30 15
sudoku 30 24 30 24
tinygql 30 41 30 41
30k_ifelse 30 55 30 55
30k_methods 30 23 30 23
cfunc_itself 30 827 30 827
fib 30 416 30 416
getivar 30 946 30 946
keyword_args 30 591 30 591
respond_to 30 1480 30 1480
ruby-xor 30 91 30 91
setivar 30 2392 30 2392
setivar_object 30 351 30 351
setivar_young 30 370 30 370
str_concat 30 420 30 420
throw 30 805 30 805

Different Ruby configurations want different amounts of warmup. With no JIT, CRuby needs hardly any. YJIT and MJIT 3.0 both warm up quite quickly, while MJIT in 3.1 often slows down for a time as it compiles, after an unpredictable delay.

Benchmark YJIT Stats

Hover your cursor over the benchmark names for descriptions of each benchmark.

bench Exit Report Inline Outlined Comp iSeqs Comp Blocks Inval Inval Ratio Bind Alloc Bind Set Const Bumps Compile Time MS
activerecord (click) 884247 761474 49 524 0 0% 0 0 15.128014
chunky-png (click) 333600 291081 87 1092 1 0% 0 0 31.771598
erubi-rails (click) 1703732 1439745 280 2891 6 0% 0 0 82.153058
hexapdf (click) 1414286 1181915 596 12600 37 0% 0 0 368.757386
liquid-c (click) 519610 455883 120 1750 5 0% 0 0 47.844191
liquid-compile (click) 442620 381171 151 2055 2 0% 0 0 62.51403
liquid-render (click) 620370 532546 144 2271 8 0% 0 0 62.294769
lobsters (click) 7881138 6599464 3219 49951 73 0% 0 0 1702.027031
mail (click) 719380 614584 346 4846 14 0% 0 0 142.88185
psych-load (click) 296529 261452 65 649 3 0% 0 0 20.01854
railsbench (click) 3001271 2598773 1643 15670 54 0% 0 0 452.869865
rubocop (click) 5360354 4667182 2902 47728 106 0% 4 0 1419.452343
ruby-lsp (click) 625014 516211 159 2816 14 0% 0 0 77.111809
sequel (click) 492760 428184 16 118 0 0% 0 0 4.411664
binarytrees (click) 9392 7588 11 75 0 0% 0 0 2.9426
blurhash (click) 52098 47218 32 461 0 0% 0 0 15.285621
erubi (click) 258400 217372 10 111 0 0% 0 0 3.815112
etanni (click) 31710 27867 12 96 0 0% 0 0 3.265425
fannkuchredux (click) 24912 17701 9 257 0 0% 0 0 7.776414
fluentd (click) 386575 321300 11 111 0 0% 0 0 3.626853
graphql (click) 430381 368998 70 673 0 0% 0 0 20.306248
graphql-native (click) 396367 338863 41 271 0 0% 0 0 9.07974
lee (click) 312793 278189 49 731 0 0% 0 0 22.88126
matmul (click) 12586 12175 12 126 0 0% 0 0 4.208892
nbody (click) 15103 7574 11 185 0 0% 0 0 4.952031
nqueens (click) 23609 17866 10 261 0 0% 0 0 7.668042
optcarrot (click) 317087 278432 192 4342 34 0% 0 0 104.005064
rack (click) 260601 223076 31 341 0 0% 0 0 9.487999
ruby-json (click) 28135 23070 12 198 0 0% 0 0 5.788496
rubykon (click) 133625 106471 143 1498 3 0% 0 0 41.637235
sudoku (click) 47007 45195 11 507 0 0% 0 0 15.377242
tinygql (click) 306073 257268 63 788 5 0% 0 0 23.048436
30k_ifelse (click) 5222597 4374236 9264 50813 0 0% 0 0 1479.070873
30k_methods (click) 2019614 1594052 5782 19369 0 0% 0 0 520.635384
cfunc_itself (click) 7589 5810 9 70 0 0% 0 0 2.299458
fib (click) 5017 4633 8 49 0 0% 0 0 1.762188
getivar (click) 6102 6023 8 74 0 0% 0 0 2.193174
keyword_args (click) 8472 6502 10 72 0 0% 0 0 2.538555
respond_to (click) 8341 7760 9 85 0 0% 0 0 2.659342
ruby-xor (click) 8142 7087 9 87 0 0% 0 0 2.654384
setivar (click) 5193 4604 8 54 0 0% 0 0 1.785878
setivar_object (click) 5561 4624 8 54 0 0% 0 0 1.929663
setivar_young (click) 6284 5274 9 62 0 0% 0 0 2.036445
str_concat (click) 7932 7204 11 85 0 0% 0 0 2.787395
throw (click) 7946 6065 10 69 0 0% 0 0 2.352831

YJIT stats correspond to the YJIT stats exit report.

Note: currently, all stats are collected on x86_64, not ARM.

Raw JSON data files

All graphs and table data in this page comes from processing these data files, which come from benchmark runs.