YJIT Benchmarks

Details for Benchmarks at 2024-08-09 06:08:34 UTC

YJIT metrics from the yjit-bench suite using Ruby f57167d338

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 73.2% faster than CRuby 3.4.0dev
  • 2.0% faster than YJIT 3.3.4
On railsbench it is
  • 79.0% faster than CRuby 3.4.0dev
  • 2.1% 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 48 10 49 10 114 10 115
chunky-png 10 18 10 16 10 36 10 36
erubi-rails 10 10 10 10 10 20 10 19
hexapdf 10 10 10 10 10 10 10 10
liquid-c 10 347 10 342 10 491 10 502
liquid-compile 10 343 10 336 10 482 10 498
liquid-render 10 140 10 132 10 353 10 365
lobsters 10 14 10 13 10 24 10 23
mail 10 161 10 151 10 233 10 222
psych-load 10 10 10 10 10 10 10 10
railsbench 10 10 10 10 10 10 10 10
rubocop 10 136 10 131 10 227 10 237
ruby-lsp 10 131 10 123 10 209 10 212
sequel 10 320 10 307 10 430 10 415
binarytrees 10 51 10 49 10 118 10 121
blurhash 10 62 10 59 10 120 10 144
erubi 10 79 10 86 10 98 10 108
etanni 10 59 10 57 10 69 10 67
fannkuchredux 10 10 10 10 10 18 10 17
fluentd 10 10 10 10 10 10 10 10
graphql 10 10 10 10 10 10 10 10
graphql-native 10 35 10 33 10 41 10 25
lee 10 11 10 10 10 20 10 19
matmul 10 10 10 10 10 15 10 17
nbody 10 207 10 206 10 407 10 486
nqueens 10 117 10 108 10 117 10 465
optcarrot 10 10 10 10 10 10 10 10
protoboeuf 10 179 10 175 10 603 10 697
protoboeuf-encode 10 123 10 137 10 275 10 462
rack 10 457 10 466 10 801 10 878
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 31 10 32
tinygql 10 24 10 22 10 52 10 51
30k_ifelse 10 10 10 10 10 70 10 69
30k_methods 10 10 10 10 10 26 10 26
cfunc_itself 10 254 10 242 10 1105 10 1212
fib 10 105 10 103 10 616 10 618
getivar 10 256 10 225 10 1277 10 1733
keyword_args 10 92 10 85 10 720 10 946
object-new 10 228 10 211 10 300 10 267
respond_to 10 95 10 87 10 1728 10 2096
ruby-xor 10 31 10 28 10 99 10 151
setivar 10 427 10 414 10 2353 10 4875
setivar_object 10 263 10 259 10 652 10 702
setivar_young 10 265 10 261 10 651 10 698
str_concat 10 327 10 312 10 595 10 549
throw 10 987 10 878 10 1276 10 1183

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) 1365011 1200951 214 2169 0 0% 0 0 80.941701
chunky-png (click) 289629 259069 83 1054 1 0% 0 0 35.513272
erubi-rails (click) 1641365 1389078 274 2920 6 0% 0 0 101.895539
hexapdf (click) 1264757 1101143 505 11813 39 0% 0 0 409.164487
liquid-c (click) 467247 420398 119 1768 5 0% 0 0 58.129593
liquid-compile (click) 389196 371238 151 2082 2 0% 0 0 73.025256
liquid-render (click) 573747 468380 136 2323 8 0% 0 0 76.36447
lobsters (click) 7529475 6406388 3082 49792 81 0% 0 0 2245.689921
mail (click) 739359 709503 347 5554 14 0% 0 0 189.136682
psych-load (click) 250820 216259 61 592 3 0% 0 0 21.005946
railsbench (click) 2894846 2500793 1636 15965 57 0% 0 0 587.388973
rubocop (click) 5443875 4702802 2894 50965 114 0% 4 0 1925.70066
ruby-lsp (click) 807350 684752 309 4745 24 0% 0 0 160.077451
sequel (click) 433221 410909 16 116 0 0% 0 0 4.947364
binarytrees (click) 8404 7594 9 66 0 0% 0 0 2.969531
blurhash (click) 50094 43901 30 453 0 0% 0 0 17.68338
erubi (click) 221158 187774 10 112 0 0% 0 0 4.388177
etanni (click) 28405 26890 11 91 0 0% 0 0 3.481843
fannkuchredux (click) 23706 31291 4 252 0 0% 0 0 9.771888
fluentd (click) 333141 277261 4 28 0 0% 0 0 1.387615
graphql (click) 391317 362805 66 650 0 0% 0 0 23.050653
graphql-native (click) 348383 305860 41 273 0 0% 0 0 10.213903
lee (click) 235493 187424 34 531 0 0% 0 0 19.529484
matmul (click) 7957 8968 7 82 0 0% 0 0 3.238507
nbody (click) 15014 19017 10 191 0 0% 0 0 6.161148
nqueens (click) 23822 23823 9 279 0 0% 0 0 10.320057
optcarrot (click) 309010 266612 187 4398 34 0% 0 0 118.76939
protoboeuf (click) 139992 122036 16 1364 0 0% 0 0 48.33985
protoboeuf-encode (click) 212046 158560 18 1189 0 0% 0 0 36.485527
rack (click) 232430 224506 37 408 0 0% 0 0 12.9889
ruby-json (click) 22939 19989 7 164 0 0% 0 0 5.363378
rubykon (click) 128223 123238 136 1480 3 0% 0 0 48.177786
sudoku (click) 45292 24766 7 516 0 0% 0 0 19.108039
tinygql (click) 249477 240302 58 735 5 0% 0 0 24.254417
30k_ifelse (click) 5134313 4729205 9258 50774 0 0% 0 0 1914.413035
30k_methods (click) 2016559 1591611 5777 19335 0 0% 0 0 530.257487
cfunc_itself (click) 7172 6017 8 66 0 0% 0 0 2.375272
fib (click) 4609 4792 7 45 0 0% 0 0 1.869235
getivar (click) 5575 6644 7 70 0 0% 0 0 2.349758
keyword_args (click) 8042 6744 9 68 0 0% 0 0 2.675366
object-new (click) 4039 4024 6 41 0 0% 0 0 1.93873
respond_to (click) 7910 8158 8 81 0 0% 0 0 2.885212
ruby-xor (click) 5219 5772 4 58 0 0% 0 0 1.971804
setivar (click) 4726 5320 7 50 0 0% 0 0 1.926831
setivar_object (click) 5109 5228 7 50 0 0% 0 0 2.006613
setivar_young (click) 5819 5950 8 58 0 0% 0 0 2.178995
str_concat (click) 7415 7947 10 81 0 0% 0 0 3.2426
throw (click) 7538 6222 9 65 0 0% 0 0 2.695962

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.