YJIT Benchmarks

Details for Benchmarks at 2024-02-28 19:08:16 UTC

YJIT metrics from the yjit-bench suite using Ruby c5b568bab6

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 70.8% faster than CRuby 3.4.0dev
On railsbench it is
  • 71.7% 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 32 30 32
erubi-rails 30 19 30 19
hexapdf 30 15 30 15
liquid-c 30 348 30 348
liquid-compile 30 340 30 340
liquid-render 30 250 30 250
lobsters 30 23 30 23
mail 30 160 30 160
psych-load 30 15 30 15
railsbench 30 15 30 15
rubocop 30 181 30 181
ruby-lsp 30 182 30 182
sequel 30 286 30 286
binarytrees 30 82 30 82
blurhash 30 101 30 101
erubi 30 76 30 76
etanni 30 50 30 50
fannkuchredux 30 24 30 24
fluentd 30 15 30 15
graphql 30 15 30 15
graphql-native 30 22 30 22
lee 30 20 30 20
matmul 30 18 30 18
nbody 30 327 30 327
nqueens 30 302 30 302
optcarrot 30 15 30 15
rack 30 181 30 181
ruby-json 30 15 30 15
rubykon 30 15 30 15
sudoku 30 26 30 26
tinygql 30 41 30 41
30k_ifelse 30 55 30 55
30k_methods 30 24 30 24
cfunc_itself 30 827 30 827
fib 30 417 30 417
getivar 30 939 30 939
keyword_args 30 591 30 591
respond_to 30 1480 30 1480
ruby-xor 30 90 30 90
setivar 30 2393 30 2393
setivar_object 30 481 30 481
setivar_young 30 506 30 506
str_concat 30 444 30 444
throw 30 747 30 747

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) 885084 766988 49 522 0 0% 0 0 15.530294
chunky-png (click) 338178 292114 87 1101 1 0% 0 0 32.52235
erubi-rails (click) 1721239 1430603 280 2899 6 0% 0 0 82.005065
hexapdf (click) 1427166 1227876 596 12661 38 0% 0 0 368.125942
liquid-c (click) 526131 457683 120 1753 5 0% 0 0 47.015845
liquid-compile (click) 442334 361357 151 2068 2 0% 0 0 62.167942
liquid-render (click) 626508 550947 144 2274 8 0% 0 0 62.524412
lobsters (click) 7885745 6567641 3219 49946 73 0% 0 0 1697.16103
mail (click) 732598 636273 346 4847 14 0% 0 0 140.579437
psych-load (click) 301391 253853 65 655 3 0% 0 0 20.27756
railsbench (click) 3015751 2596930 1643 15666 54 0% 0 0 448.657099
rubocop (click) 5364841 4631437 2902 47742 106 0% 4 0 1406.033354
ruby-lsp (click) 635114 558850 159 2730 14 0% 0 0 76.389083
sequel (click) 495239 427139 16 119 0 0% 0 0 4.439023
binarytrees (click) 9392 7588 11 75 0 0% 0 0 3.161348
blurhash (click) 52000 47236 32 461 0 0% 0 0 15.349946
erubi (click) 271414 222726 10 111 0 0% 0 0 3.803503
etanni (click) 31710 27867 12 96 0 0% 0 0 3.19885
fannkuchredux (click) 24912 17701 9 257 0 0% 0 0 7.829403
fluentd (click) 389820 321061 11 111 0 0% 0 0 3.56856
graphql (click) 444390 375172 70 673 0 0% 0 0 20.99491
graphql-native (click) 409945 344761 41 271 0 0% 0 0 9.15859
lee (click) 316023 278067 49 731 0 0% 0 0 22.649073
matmul (click) 12586 12175 12 126 0 0% 0 0 4.160892
nbody (click) 15103 7574 11 185 0 0% 0 0 5.014293
nqueens (click) 23609 17866 10 261 0 0% 0 0 7.783899
optcarrot (click) 317087 278432 192 4342 34 0% 0 0 103.269136
rack (click) 272543 235008 31 322 0 0% 0 0 9.332287
ruby-json (click) 28135 23070 12 198 0 0% 0 0 5.748647
rubykon (click) 133999 114882 143 1500 3 0% 0 0 41.674728
sudoku (click) 47007 45195 11 507 0 0% 0 0 15.225789
tinygql (click) 313186 266445 63 788 5 0% 0 0 22.567204
30k_ifelse (click) 5222597 4374236 9264 50813 0 0% 0 0 1449.648247
30k_methods (click) 2019614 1594052 5782 19369 0 0% 0 0 515.586359
cfunc_itself (click) 7589 5810 9 70 0 0% 0 0 2.211629
fib (click) 5017 4633 8 49 0 0% 0 0 1.780883
getivar (click) 6102 6023 8 74 0 0% 0 0 2.12316
keyword_args (click) 8472 6502 10 72 0 0% 0 0 2.515995
respond_to (click) 8341 7760 9 85 0 0% 0 0 2.634691
ruby-xor (click) 7940 7126 9 87 0 0% 0 0 2.999846
setivar (click) 5193 4604 8 54 0 0% 0 0 1.725042
setivar_object (click) 5561 4624 8 54 0 0% 0 0 1.813965
setivar_young (click) 6284 5274 9 62 0 0% 0 0 2.05812
str_concat (click) 7932 7204 11 85 0 0% 0 0 2.868106
throw (click) 7946 6065 10 69 0 0% 0 0 2.47184

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.