YJIT Benchmarks

Details for Benchmarks at 2024-08-23 19:08:31 UTC

YJIT metrics from the yjit-bench suite using Ruby 3f6be01bfc

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 74.1% faster than CRuby 3.4.0dev
  • 2.9% faster than YJIT 3.3.4
On railsbench it is
  • 81.3% faster than CRuby 3.4.0dev
  • 1.6% faster than YJIT 3.3.4

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 3.3.4 warmups CRuby 3.3.4 iters CRuby 3.4.0dev warmups CRuby 3.4.0dev iters YJIT 3.3.4 warmups YJIT 3.3.4 iters YJIT 3.4.0dev warmups YJIT 3.4.0dev iters
activerecord 10 49 10 49 10 117 10 120
chunky-png 10 18 10 18 10 36 10 40
erubi-rails 10 10 10 10 10 21 10 19
hexapdf 10 10 10 10 10 10 10 10
liquid-c 10 340 10 350 10 488 10 514
liquid-compile 10 344 10 347 10 503 10 501
liquid-render 10 139 10 134 10 357 10 373
lobsters 10 14 10 13 10 24 10 23
mail 10 159 10 158 10 238 10 228
psych-load 10 10 10 10 10 10 10 10
railsbench 10 10 10 10 10 10 10 10
rubocop 10 135 10 132 10 237 10 239
ruby-lsp 10 127 10 124 10 217 10 216
sequel 10 317 10 322 10 438 10 433
binarytrees 10 51 10 49 10 118 10 115
blurhash 10 62 10 62 10 120 10 147
erubi 10 79 10 98 10 99 10 123
etanni 10 60 10 64 10 69 10 76
fannkuchredux 10 10 10 10 10 17 10 18
fluentd 10 10 10 10 10 10 10 10
graphql 10 10 10 10 10 10 10 10
graphql-native 10 28 10 31 10 33 10 38
lee 10 11 10 12 10 20 10 21
matmul 10 10 10 10 10 15 10 18
nbody 10 203 10 214 10 421 10 509
nqueens 10 113 10 114 10 113 10 431
optcarrot 10 10 10 10 10 10 10 10
protoboeuf 10 182 10 182 10 572 10 685
protoboeuf-encode 10 124 10 161 10 283 10 639
rack 10 452 10 475 10 813 10 899
ruby-json 10 10 10 10 10 10 10 10
rubykon 10 10 10 10 10 10 10 10
sudoku 10 10 10 10 10 28 10 32
tinygql 10 23 10 22 10 51 10 50
30k_ifelse 10 10 10 10 10 71 10 69
30k_methods 10 10 10 10 10 26 10 26
cfunc_itself 10 248 10 245 10 1108 10 1213
fib 10 102 10 104 10 616 10 617
getivar 10 214 10 241 10 1247 10 1749
keyword_args 10 89 10 84 10 729 10 947
object-new 10 228 10 217 10 308 10 287
respond_to 10 90 10 93 10 1785 10 2096
ruby-xor 10 30 10 25 10 92 10 155
setivar 10 349 10 360 10 2319 10 4823
setivar_object 10 216 10 228 10 536 10 518
setivar_young 10 216 10 227 10 528 10 533
str_concat 10 318 10 357 10 658 10 682
throw 10 977 10 909 10 1342 10 1214

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) 1359068 1257715 214 2169 0 0% 0 0 81.358056
chunky-png (click) 289629 259069 83 1054 1 0% 0 0 35.797044
erubi-rails (click) 1640199 1363703 275 2944 6 0% 0 0 102.340639
hexapdf (click) 1246425 1125419 505 11812 38 0% 0 0 412.342758
liquid-c (click) 461536 456238 119 1768 5 0% 0 0 58.02099
liquid-compile (click) 383402 337355 151 2070 2 0% 0 0 72.916534
liquid-render (click) 554255 491553 136 2323 8 0% 0 0 76.75738
lobsters (click) 7562962 6388062 3081 49982 81 0% 0 0 2260.012829
mail (click) 720959 669516 347 5554 14 0% 0 0 189.032735
psych-load (click) 224181 209111 61 594 3 0% 0 0 21.376889
railsbench (click) 2900804 2473534 1637 15973 57 0% 0 0 589.761159
rubocop (click) 5451713 4653415 2894 50970 114 0% 4 0 1939.863298
ruby-lsp (click) 787133 687831 309 4750 24 0% 0 0 160.875308
sequel (click) 432035 366858 16 116 0 0% 0 0 4.96479
binarytrees (click) 8404 7594 9 66 0 0% 0 0 3.012373
blurhash (click) 50094 43901 30 453 0 0% 0 0 17.707419
erubi (click) 210021 195156 10 112 0 0% 0 0 4.491137
etanni (click) 28405 26890 11 91 0 0% 0 0 3.506148
fannkuchredux (click) 23706 31291 4 252 0 0% 0 0 9.774915
fluentd (click) 338257 273739 4 28 0 0% 0 0 1.279216
graphql (click) 371411 320245 66 650 0 0% 0 0 22.999644
graphql-native (click) 344164 312718 42 278 0 0% 0 0 10.715521
lee (click) 234910 187753 34 531 0 0% 0 0 19.51669
matmul (click) 7957 8968 7 82 0 0% 0 0 3.238549
nbody (click) 15014 19017 10 191 0 0% 0 0 6.166148
nqueens (click) 23822 23823 9 279 0 0% 0 0 10.34832
optcarrot (click) 309010 266612 187 4398 34 0% 0 0 118.511644
protoboeuf (click) 139992 122036 16 1364 0 0% 0 0 48.465159
protoboeuf-encode (click) 212046 174944 18 1189 0 0% 0 0 36.521468
rack (click) 240185 216085 37 413 0 0% 0 0 13.038717
ruby-json (click) 24237 21549 7 166 0 0% 0 0 5.572918
rubykon (click) 128223 123238 136 1480 3 0% 0 0 48.371578
sudoku (click) 45292 24766 7 516 0 0% 0 0 19.084566
tinygql (click) 266197 232489 58 735 5 0% 0 0 24.319854
30k_ifelse (click) 5134313 4729205 9258 50774 0 0% 0 0 1919.43835
30k_methods (click) 2016559 1591611 5777 19335 0 0% 0 0 533.415873
cfunc_itself (click) 7172 6017 8 66 0 0% 0 0 2.34392
fib (click) 4609 4792 7 45 0 0% 0 0 1.841806
getivar (click) 5575 6644 7 70 0 0% 0 0 2.324806
keyword_args (click) 8042 6744 9 68 0 0% 0 0 2.632459
object-new (click) 4039 4024 6 41 0 0% 0 0 2.161187
respond_to (click) 7910 8158 8 81 0 0% 0 0 2.874349
ruby-xor (click) 5219 5772 4 58 0 0% 0 0 1.952179
setivar (click) 4726 5320 7 50 0 0% 0 0 1.87326
setivar_object (click) 5109 5228 7 50 0 0% 0 0 1.96107
setivar_young (click) 5819 5950 8 58 0 0% 0 0 2.193976
str_concat (click) 7415 7947 10 81 0 0% 0 0 3.230788
throw (click) 7538 6222 9 65 0 0% 0 0 2.69008

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.