YJIT Benchmarks

Details for Benchmarks at 2024-05-06 06:08:15 UTC

YJIT metrics from the yjit-bench suite using Ruby a5cb8c8c5b

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 68.9% faster than CRuby 3.4.0dev
On railsbench it is
  • 79.5% 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 10 344 10 734
chunky-png 10 17 10 38
erubi-rails 10 10 10 19
hexapdf 10 10 10 10
liquid-c 10 350 10 506
liquid-compile 10 339 10 508
liquid-render 10 135 10 356
lobsters 10 13 10 23
mail 10 156 10 228
psych-load 10 10 10 10
railsbench 10 10 10 10
rubocop 10 131 10 245
ruby-lsp 10 176 10 235
sequel 10 315 10 415
binarytrees 10 51 10 117
blurhash 10 60 10 141
erubi 10 84 10 106
etanni 10 25 10 28
fannkuchredux 10 10 10 17
fluentd 10 10 10 10
graphql 10 10 10 10
graphql-native 10 35 10 41
lee 10 11 10 21
matmul 10 10 10 16
nbody 10 207 10 471
nqueens 10 115 10 465
optcarrot 10 10 10 10
protoboeuf 10 187 10 730
rack 10 205 10 255
ruby-json 10 10 10 10
rubykon 10 10 10 10
sudoku 10 10 10 31
tinygql 10 23 10 52
30k_ifelse 10 10 10 70
30k_methods 10 10 10 26
cfunc_itself 10 254 10 1211
fib 10 104 10 618
getivar 10 220 10 1348
keyword_args 10 89 10 945
respond_to 10 90 10 2149
ruby-xor 10 29 10 143
setivar 10 422 10 3582
setivar_object 10 233 10 590
setivar_young 10 236 10 607
str_concat 10 334 10 635
throw 10 968 10 1166

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) 879223 716160 49 528 0 0% 0 0 16.870138
chunky-png (click) 338761 273627 82 1048 1 0% 0 0 31.513467
erubi-rails (click) 1693263 1382065 272 2903 6 0% 0 0 82.657777
hexapdf (click) 1295303 1132786 504 11518 41 0% 0 0 339.305251
liquid-c (click) 525934 465131 119 1757 5 0% 0 0 49.57559
liquid-compile (click) 451209 381087 151 2082 2 0% 0 0 65.74613
liquid-render (click) 623961 528746 136 2292 8 0% 0 0 64.81617
lobsters (click) 7555903 6212537 3110 48893 91 0% 0 0 1662.120232
mail (click) 731588 650671 346 4865 14 0% 0 0 146.64288
psych-load (click) 303929 244228 61 617 3 0% 0 0 19.891648
railsbench (click) 2954304 2561947 1633 15944 57 0% 0 0 466.967356
rubocop (click) 5332391 4512983 2901 48277 112 0% 4 0 1466.801491
ruby-lsp (click) 622665 530329 159 2808 28 0% 0 0 78.535962
sequel (click) 489410 418187 16 118 0 0% 0 0 4.423246
binarytrees (click) 8756 7777 10 70 0 0% 0 0 2.789827
blurhash (click) 50429 43273 31 455 0 0% 0 0 15.486996
erubi (click) 277116 235415 10 112 0 0% 0 0 3.854186
etanni (click) 29692 28216 11 91 0 0% 0 0 3.001373
fannkuchredux (click) 21198 26809 5 228 0 0% 0 0 7.161903
fluentd (click) 385230 328274 4 28 0 0% 0 0 1.121918
graphql (click) 438895 384067 66 644 0 0% 0 0 20.071509
graphql-native (click) 410104 360491 42 278 0 0% 0 0 9.486495
lee (click) 300573 267787 34 528 0 0% 0 0 17.200818
matmul (click) 8300 983 8 85 0 0% 0 0 2.951861
nbody (click) 14798 17933 11 185 0 0% 0 0 5.198859
nqueens (click) 22703 29059 10 260 0 0% 0 0 8.117138
optcarrot (click) 307949 264223 188 4285 34 0% 0 0 103.268768
protoboeuf (click) 146719 135165 17 1366 0 0% 0 0 42.389463
rack (click) 267837 202314 31 324 0 0% 0 0 9.604278
ruby-json (click) 30222 27890 8 167 0 0% 0 0 4.987281
rubykon (click) 127405 128474 137 1461 3 0% 0 0 42.113151
sudoku (click) 42479 58142 8 475 0 0% 0 0 15.141434
tinygql (click) 315372 256249 58 741 5 0% 0 0 22.031078
30k_ifelse (click) 5134674 4729436 9259 50778 0 0% 0 0 1533.143549
30k_methods (click) 2016920 1591842 5778 19339 0 0% 0 0 520.587404
cfunc_itself (click) 7534 6180 9 70 0 0% 0 0 2.279957
fib (click) 4970 5003 8 49 0 0% 0 0 1.803448
getivar (click) 5963 6375 8 74 0 0% 0 0 2.240799
keyword_args (click) 8409 6912 10 72 0 0% 0 0 2.588942
respond_to (click) 8277 8274 9 85 0 0% 0 0 2.702268
ruby-xor (click) 5593 5796 5 62 0 0% 0 0 1.778138
setivar (click) 5102 5375 8 54 0 0% 0 0 1.836542
setivar_object (click) 5470 5395 8 54 0 0% 0 0 1.917736
setivar_young (click) 6182 6093 9 62 0 0% 0 0 2.130964
str_concat (click) 7774 7946 11 85 0 0% 0 0 2.869785
throw (click) 7895 6377 10 69 0 0% 0 0 2.667865

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.