YJIT Benchmarks

Details for Benchmarks at 2024-06-26 06:08:16 UTC

YJIT metrics from the yjit-bench suite using Ruby b4503d7964

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 70.6% faster than CRuby 3.4.0dev
  • the same speed as YJIT 3.3.3
On railsbench it is
  • 76.6% faster than CRuby 3.4.0dev
  • 1.5% slower than YJIT 3.3.3

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 3.3.3 warmups CRuby 3.3.3 iters CRuby 3.4.0dev warmups CRuby 3.4.0dev iters YJIT 3.3.3 warmups YJIT 3.3.3 iters YJIT 3.4.0dev warmups YJIT 3.4.0dev iters
activerecord 10 51 10 47 10 116 10 117
chunky-png 10 18 10 15 10 36 10 36
erubi-rails 10 10 10 10 10 21 10 20
hexapdf 10 10 10 10 10 10 10 10
liquid-c 10 352 10 333 10 504 10 480
liquid-compile 10 341 10 332 10 490 10 481
liquid-render 10 136 10 134 10 354 10 351
mail 10 162 10 149 10 238 10 218
psych-load 10 10 10 10 10 10 10 10
railsbench 10 10 10 10 10 10 10 10
rubocop 10 136 10 126 10 230 10 235
ruby-lsp 10 199 10 190 10 263 10 246
sequel 10 325 10 309 10 436 10 410
binarytrees 10 51 10 48 10 117 10 113
blurhash 10 61 10 58 10 118 10 145
erubi 10 83 10 84 10 102 10 105
etanni 10 59 10 58 10 68 10 68
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 34 10 34 10 39 10 39
lee 10 11 10 10 10 19 10 21
matmul 10 10 10 10 10 15 10 17
nbody 10 212 10 191 10 385 10 502
nqueens 10 115 10 102 10 115 10 434
optcarrot 10 10 10 10 10 10 10 10
protoboeuf 10 189 10 174 10 681 10 657
rack 10 472 10 465 10 811 10 883
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 29
tinygql 10 24 10 22 10 52 10 48
30k_ifelse 10 10 10 10 10 70 10 69
30k_methods 10 10 10 10 10 26 10 26
cfunc_itself 10 257 10 231 10 1105 10 1210
fib 10 103 10 100 10 616 10 618
getivar 10 255 10 242 10 1279 10 1364
keyword_args 10 90 10 83 10 726 10 944
respond_to 10 89 10 83 10 1782 10 2146
ruby-xor 10 31 10 25 10 100 10 133
setivar 10 424 10 406 10 2353 10 3572
setivar_object 10 264 10 265 10 653 10 734
setivar_young 10 263 10 265 10 651 10 738
str_concat 10 327 10 288 10 592 10 520
throw 10 974 10 837 10 1276 10 1034

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) 1402864 1179425 214 2166 0 0% 0 0 76.786683
chunky-png (click) 331257 268947 82 1053 1 0% 0 0 33.876317
erubi-rails (click) 1676145 1438804 272 2899 6 0% 0 0 96.825835
hexapdf (click) 1288545 1097551 504 11686 41 0% 0 0 375.861773
liquid-c (click) 518410 434198 120 1762 5 0% 0 0 53.360732
liquid-compile (click) 440081 387228 151 2060 2 0% 0 0 68.841553
liquid-render (click) 613161 509414 137 2303 8 0% 0 0 70.645381
mail (click) 723226 644597 346 4881 14 0% 0 0 157.547866
psych-load (click) 293176 243984 61 594 3 0% 0 0 20.496088
railsbench (click) 2924350 2545652 1634 15918 57 0% 0 0 557.954256
rubocop (click) 5434533 4518701 2891 50396 110 0% 4 0 1792.263052
ruby-lsp (click) 635601 567764 159 2903 28 0% 0 0 90.216848
sequel (click) 487435 417501 16 116 0 0% 0 0 4.642682
binarytrees (click) 8756 7777 10 70 0 0% 0 0 2.982223
blurhash (click) 50429 43324 31 457 0 0% 0 0 16.98974
erubi (click) 270214 214579 10 112 0 0% 0 0 4.389882
etanni (click) 30236 28701 12 96 0 0% 0 0 3.463483
fannkuchredux (click) 21198 26809 5 228 0 0% 0 0 7.888156
fluentd (click) 375259 336801 4 28 0 0% 0 0 1.302768
graphql (click) 430898 362014 66 648 0 0% 0 0 22.077336
graphql-native (click) 402926 330696 42 278 0 0% 0 0 10.743789
lee (click) 293772 238942 34 531 0 0% 0 0 18.233924
matmul (click) 8315 983 8 86 0 0% 0 0 3.032907
nbody (click) 14798 17933 11 185 0 0% 0 0 5.573007
nqueens (click) 22145 28339 10 257 0 0% 0 0 8.976288
optcarrot (click) 304463 277860 188 4310 34 0% 0 0 109.148329
protoboeuf (click) 140419 133760 17 1349 0 0% 0 0 43.208876
rack (click) 275741 228654 37 408 0 0% 0 0 12.028301
ruby-json (click) 30222 27890 8 167 0 0% 0 0 5.197704
rubykon (click) 129933 122375 137 1498 3 0% 0 0 45.707925
sudoku (click) 41994 57542 8 473 0 0% 0 0 16.143203
tinygql (click) 298487 261389 58 734 5 0% 0 0 23.417237
30k_ifelse (click) 5134674 4729436 9259 50778 0 0% 0 0 1895.367541
30k_methods (click) 2016920 1591842 5778 19339 0 0% 0 0 537.235912
cfunc_itself (click) 7534 6180 9 70 0 0% 0 0 2.441405
fib (click) 4970 5003 8 49 0 0% 0 0 2.019112
getivar (click) 5963 6375 8 74 0 0% 0 0 2.282599
keyword_args (click) 8409 6912 10 72 0 0% 0 0 2.638465
respond_to (click) 8277 8274 9 85 0 0% 0 0 2.786317
ruby-xor (click) 5565 5673 5 62 0 0% 0 0 1.894428
setivar (click) 5102 5375 8 54 0 0% 0 0 1.934421
setivar_object (click) 5470 5395 8 54 0 0% 0 0 2.040856
setivar_young (click) 6182 6093 9 62 0 0% 0 0 2.269255
str_concat (click) 7774 7946 11 85 0 0% 0 0 3.115287
throw (click) 7895 6377 10 69 0 0% 0 0 2.593065

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.