YJIT Benchmarks

Details for Benchmarks at 2024-03-07 06:08:17 UTC

YJIT metrics from the yjit-bench suite using Ruby d3279a0c11

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 69.7% faster than CRuby 3.4.0dev
On railsbench it is
  • 80.1% 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 30 519 30 519
chunky-png 30 31 30 31
erubi-rails 30 19 30 19
hexapdf 30 15 30 15
liquid-c 30 349 30 349
liquid-compile 30 340 30 340
liquid-render 30 253 30 253
lobsters 30 23 30 23
mail 30 163 30 163
psych-load 30 15 30 15
railsbench 30 15 30 15
rubocop 30 181 30 181
ruby-lsp 30 188 30 188
sequel 30 291 30 291
binarytrees 30 84 30 84
blurhash 30 100 30 100
erubi 30 76 30 76
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 30 316 30 316
nqueens 30 300 30 300
optcarrot 30 15 30 15
rack 30 176 30 176
ruby-json 30 15 30 15
rubykon 30 15 30 15
sudoku 30 25 30 25
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 30 941 30 941
keyword_args 30 591 30 591
respond_to 30 1480 30 1480
ruby-xor 30 109 30 109
setivar 30 2393 30 2393
setivar_object 30 488 30 488
setivar_young 30 507 30 507
str_concat 30 396 30 396
throw 30 716 30 716

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) 884642 767118 49 524 0 0% 0 0 15.395239
chunky-png (click) 338380 292265 87 1098 1 0% 0 0 32.89761
erubi-rails (click) 1707163 1428848 279 2890 6 0% 0 0 82.66703
hexapdf (click) 1419499 1244223 596 12577 38 0% 0 0 369.647559
liquid-c (click) 519073 435374 119 1754 5 0% 0 0 47.531281
liquid-compile (click) 447966 390660 151 2075 2 0% 0 0 63.172625
liquid-render (click) 624034 558006 136 2283 8 0% 0 0 62.898809
lobsters (click) 7887132 6616374 3217 50068 73 0% 0 0 1725.053493
mail (click) 732395 635986 346 4847 14 0% 0 0 143.150122
psych-load (click) 301215 253619 65 655 3 0% 0 0 20.566371
railsbench (click) 2997246 2606069 1638 15752 54 0% 0 0 453.480813
rubocop (click) 5368309 4478067 2901 47762 106 0% 4 0 1423.697005
ruby-lsp (click) 634011 541577 159 2735 14 0% 0 0 76.677514
sequel (click) 494689 426374 16 118 0 0% 0 0 4.368142
binarytrees (click) 9392 7588 11 75 0 0% 0 0 3.047952
blurhash (click) 52000 47236 32 461 0 0% 0 0 15.483425
erubi (click) 271214 230596 10 111 0 0% 0 0 3.92848
etanni (click) 31686 27819 12 96 0 0% 0 0 3.259132
fannkuchredux (click) 24912 17701 9 257 0 0% 0 0 7.849705
fluentd (click) 387873 311763 11 111 0 0% 0 0 3.663613
graphql (click) 443736 382823 70 673 0 0% 0 0 20.965152
graphql-native (click) 409288 344379 41 271 0 0% 0 0 9.384179
lee (click) 315849 277753 49 731 0 0% 0 0 23.043129
matmul (click) 12586 12175 12 126 0 0% 0 0 4.205169
nbody (click) 15103 7574 11 185 0 0% 0 0 5.060082
nqueens (click) 23609 17866 10 261 0 0% 0 0 7.901272
optcarrot (click) 317087 278432 192 4342 34 0% 0 0 104.328461
rack (click) 272097 234500 31 319 0 0% 0 0 9.183176
ruby-json (click) 28135 23070 12 198 0 0% 0 0 5.829566
rubykon (click) 133625 106471 143 1498 3 0% 0 0 41.982438
sudoku (click) 47007 45195 11 507 0 0% 0 0 15.648178
tinygql (click) 313116 266157 63 787 5 0% 0 0 23.217168
30k_ifelse (click) 5222597 4374236 9264 50813 0 0% 0 0 1476.109879
30k_methods (click) 2019614 1594052 5782 19369 0 0% 0 0 523.663344
cfunc_itself (click) 7589 5810 9 70 0 0% 0 0 2.2395
fib (click) 5017 4633 8 49 0 0% 0 0 1.790884
getivar (click) 6102 6023 8 74 0 0% 0 0 2.211162
keyword_args (click) 8472 6502 10 72 0 0% 0 0 2.573476
respond_to (click) 8341 7760 9 85 0 0% 0 0 2.676448
ruby-xor (click) 8110 7110 9 87 0 0% 0 0 3.163377
setivar (click) 5193 4604 8 54 0 0% 0 0 1.748603
setivar_object (click) 5561 4624 8 54 0 0% 0 0 1.840213
setivar_young (click) 6284 5274 9 62 0 0% 0 0 2.056096
str_concat (click) 7932 7204 11 85 0 0% 0 0 2.868356
throw (click) 7946 6065 10 69 0 0% 0 0 2.504306

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.