YJIT Benchmarks

Details for Benchmarks at 2024-03-12 19:08:14 UTC

YJIT metrics from the yjit-bench suite using Ruby 97b5f35ac4

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 71.8% faster than CRuby 3.4.0dev
On railsbench it is
  • 82.6% 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 30 528 30 528
chunky-png 30 31 30 31
erubi-rails 30 19 30 19
hexapdf 30 15 30 15
liquid-c 30 352 30 352
liquid-compile 30 344 30 344
liquid-render 30 257 30 257
lobsters 30 23 30 23
mail 30 161 30 161
psych-load 30 15 30 15
railsbench 30 15 30 15
rubocop 30 185 30 185
ruby-lsp 30 178 30 178
sequel 30 290 30 290
binarytrees 30 87 30 87
blurhash 30 102 30 102
erubi 30 77 30 77
etanni 30 52 30 52
fannkuchredux 30 24 30 24
fluentd 30 15 30 15
graphql 30 15 30 15
graphql-native 30 22 30 22
lee 30 19 30 19
matmul 30 18 30 18
nbody 30 330 30 330
nqueens 30 312 30 312
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 42 30 42
30k_ifelse 30 55 30 55
30k_methods 30 23 30 23
cfunc_itself 30 827 30 827
fib 30 417 30 417
getivar 30 946 30 946
keyword_args 30 591 30 591
respond_to 30 1480 30 1480
ruby-xor 30 104 30 104
setivar 30 2392 30 2392
setivar_object 30 488 30 488
setivar_young 30 506 30 506
str_concat 30 394 30 394
throw 30 775 30 775

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) 884351 758543 49 525 0 0% 0 0 15.749262
chunky-png (click) 337548 299786 87 1093 1 0% 0 0 32.611329
erubi-rails (click) 1714121 1443168 279 2948 6 0% 0 0 83.497281
hexapdf (click) 1418817 1243523 595 12569 38 0% 0 0 372.191658
liquid-c (click) 519073 435374 119 1754 5 0% 0 0 47.792002
liquid-compile (click) 447474 390164 151 2068 2 0% 0 0 63.662401
liquid-render (click) 623835 541382 136 2284 8 0% 0 0 62.959863
lobsters (click) 7876633 6596525 3219 49954 73 0% 0 0 1718.217527
mail (click) 732395 635954 346 4847 14 0% 0 0 144.01396
psych-load (click) 301215 253619 65 655 3 0% 0 0 20.593703
railsbench (click) 2996749 2556848 1638 15763 54 0% 0 0 454.487073
rubocop (click) 5372070 4563070 2901 47786 106 0% 4 0 1432.961004
ruby-lsp (click) 634011 541577 159 2735 14 0% 0 0 77.074477
sequel (click) 496603 428626 16 117 0 0% 0 0 4.176534
binarytrees (click) 9392 7588 11 75 0 0% 0 0 2.989805
blurhash (click) 52000 47236 32 461 0 0% 0 0 15.583028
erubi (click) 271214 230596 10 111 0 0% 0 0 3.863696
etanni (click) 31686 27819 12 96 0 0% 0 0 3.242578
fannkuchredux (click) 24912 17701 9 257 0 0% 0 0 7.88956
fluentd (click) 387873 311751 11 111 0 0% 0 0 3.624202
graphql (click) 443736 382930 70 673 0 0% 0 0 21.154644
graphql-native (click) 409288 344276 41 271 0 0% 0 0 9.45821
lee (click) 315849 277753 49 731 0 0% 0 0 23.185239
matmul (click) 12586 12175 12 126 0 0% 0 0 4.215105
nbody (click) 15103 7574 11 185 0 0% 0 0 5.074432
nqueens (click) 23609 17866 10 261 0 0% 0 0 7.893891
optcarrot (click) 317087 278432 192 4342 34 0% 0 0 105.340682
rack (click) 272367 234752 31 322 0 0% 0 0 9.392319
ruby-json (click) 28135 23070 12 198 0 0% 0 0 5.854983
rubykon (click) 133625 106471 143 1498 3 0% 0 0 41.871412
sudoku (click) 47007 45195 11 507 0 0% 0 0 15.82171
tinygql (click) 313116 266157 63 787 5 0% 0 0 23.020317
30k_ifelse (click) 5222597 4374236 9264 50813 0 0% 0 0 1503.790612
30k_methods (click) 2019614 1594052 5782 19369 0 0% 0 0 530.415891
cfunc_itself (click) 7589 5810 9 70 0 0% 0 0 2.253386
fib (click) 5017 4633 8 49 0 0% 0 0 1.82067
getivar (click) 6102 6023 8 74 0 0% 0 0 2.191445
keyword_args (click) 8472 6502 10 72 0 0% 0 0 2.581885
respond_to (click) 8341 7760 9 85 0 0% 0 0 2.687778
ruby-xor (click) 8110 7110 9 87 0 0% 0 0 3.202137
setivar (click) 5193 4604 8 54 0 0% 0 0 1.77208
setivar_object (click) 5561 4624 8 54 0 0% 0 0 1.917522
setivar_young (click) 6284 5274 9 62 0 0% 0 0 2.054488
str_concat (click) 7932 7204 11 85 0 0% 0 0 2.876039
throw (click) 7946 6065 10 69 0 0% 0 0 2.502158

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.