YJIT Benchmarks

Details for Benchmarks at 2024-04-01 19:08:14 UTC

YJIT metrics from the yjit-bench suite using Ruby a9658b6409.

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 73.8% faster than CRuby 3.4.0dev
On railsbench it is
  • 79.3% 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 30 525 30 525
chunky-png 30 32 30 32
erubi-rails 30 20 30 20
hexapdf 30 15 30 15
liquid-c 30 346 30 346
liquid-compile 30 354 30 354
liquid-render 30 251 30 251
lobsters 30 23 30 23
mail 30 160 30 160
psych-load 30 15 30 15
railsbench 30 15 30 15
rubocop 30 182 30 182
ruby-lsp 30 207 30 207
sequel 30 289 30 289
binarytrees 30 82 30 82
blurhash 30 99 30 99
erubi 30 76 30 76
etanni 30 24 30 24
fannkuchredux 30 23 30 23
fluentd 30 15 30 15
graphql 30 15 30 15
graphql-native 30 33 30 33
lee 30 19 30 19
matmul 30 17 30 17
nbody 30 309 30 309
nqueens 30 288 30 288
optcarrot 30 15 30 15
protoboeuf 30 413 30 413
rack 30 174 30 174
ruby-json 30 15 30 15
rubykon 30 15 30 15
sudoku 30 23 30 23
tinygql 30 39 30 39
30k_ifelse 30 55 30 55
30k_methods 30 24 30 24
cfunc_itself 30 827 30 827
fib 30 417 30 417
getivar 30 944 30 944
keyword_args 30 592 30 592
respond_to 30 1480 30 1480
ruby-xor 30 108 30 108
setivar 30 2393 30 2393
setivar_object 30 488 30 488
setivar_young 30 505 30 505
str_concat 30 414 30 414
throw 30 769 30 769

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) 883577 765584 49 526 0 0% 0 0 15.665049
chunky-png (click) 337635 282733 87 1087 1 0% 0 0 32.805349
erubi-rails (click) 1712891 1447563 276 2962 6 0% 0 0 83.960249
hexapdf (click) 1418346 1175541 595 12569 38 0% 0 0 371.623984
liquid-c (click) 524652 455230 119 1754 5 0% 0 0 48.307302
liquid-compile (click) 447874 397661 151 2068 2 0% 0 0 63.626545
liquid-render (click) 623922 564499 136 2285 8 0% 0 0 63.241769
lobsters (click) 7895706 6616006 3212 50212 73 0% 0 0 1720.620297
mail (click) 731013 625387 346 4833 14 0% 0 0 142.838294
psych-load (click) 301393 252981 65 653 3 0% 0 0 20.390886
railsbench (click) 2992357 2534668 1634 15779 54 0% 0 0 458.781065
rubocop (click) 5368821 4599283 2901 47762 106 0% 4 0 1435.82098
ruby-lsp (click) 633865 556837 159 2735 14 0% 0 0 76.631339
sequel (click) 495891 435576 16 118 0 0% 0 0 4.444122
binarytrees (click) 9384 7556 11 75 0 0% 0 0 3.0408
blurhash (click) 51809 46977 32 459 0 0% 0 0 15.539932
erubi (click) 271655 230266 10 111 0 0% 0 0 3.898616
etanni (click) 31131 27054 11 91 0 0% 0 0 3.021104
fannkuchredux (click) 24443 9175 9 257 0 0% 0 0 7.8118
fluentd (click) 388157 327800 11 111 0 0% 0 0 3.599025
graphql (click) 443950 381331 70 673 0 0% 0 0 20.344818
graphql-native (click) 411003 344853 42 276 0 0% 0 0 9.240796
lee (click) 316487 277439 49 732 0 0% 0 0 23.14435
matmul (click) 12555 12107 12 126 0 0% 0 0 4.249585
nbody (click) 15072 7506 11 185 0 0% 0 0 5.054539
nqueens (click) 23418 17741 10 260 0 0% 0 0 7.904662
optcarrot (click) 314912 252069 192 4322 34 0% 0 0 105.441522
protoboeuf (click) 153790 132358 17 1378 0 0% 0 0 42.087609
rack (click) 272408 234245 31 319 0 0% 0 0 9.235777
ruby-json (click) 34095 28619 12 196 0 0% 0 0 5.795335
rubykon (click) 133943 122908 143 1500 3 0% 0 0 42.213226
sudoku (click) 46892 53197 11 507 0 0% 0 0 15.676308
tinygql (click) 313430 265839 63 786 5 0% 0 0 23.183911
30k_ifelse (click) 5222589 4374204 9264 50813 0 0% 0 0 1483.547378
30k_methods (click) 2019606 1594020 5782 19369 0 0% 0 0 530.425138
cfunc_itself (click) 7581 5778 9 70 0 0% 0 0 2.263747
fib (click) 5009 4601 8 49 0 0% 0 0 1.826984
getivar (click) 6094 5991 8 74 0 0% 0 0 2.140352
keyword_args (click) 8464 6470 10 72 0 0% 0 0 2.558744
respond_to (click) 8333 7728 9 85 0 0% 0 0 2.651912
ruby-xor (click) 8062 6902 9 87 0 0% 0 0 3.088434
setivar (click) 5185 4572 8 54 0 0% 0 0 1.762141
setivar_object (click) 5553 4592 8 54 0 0% 0 0 1.842382
setivar_young (click) 6276 5242 9 62 0 0% 0 0 2.064574
str_concat (click) 7876 7217 11 85 0 0% 0 0 2.857966
throw (click) 7938 6033 10 69 0 0% 0 0 2.46738

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.