YJIT Benchmarks

Details for Benchmarks at 2024-05-29 19:08:17 UTC

YJIT metrics from the yjit-bench suite using Ruby a760e21bc1

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 68.7% faster than CRuby 3.4.0dev
  • 9.1% faster than YJIT 3.3.1
On railsbench it is
  • 76.3% faster than CRuby 3.4.0dev
  • 5.0% faster than YJIT 3.3.1

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 3.3.1 warmups CRuby 3.3.1 iters CRuby 3.4.0dev warmups CRuby 3.4.0dev iters YJIT 3.3.1 warmups YJIT 3.3.1 iters YJIT 3.4.0dev warmups YJIT 3.4.0dev iters
activerecord 10 45 10 48 10 103 10 113
chunky-png 10 17 10 16 10 34 10 36
erubi-rails 10 10 10 10 10 17 10 19
hexapdf 10 10 10 10 10 10 10 10
liquid-c 10 332 10 338 10 469 10 481
liquid-compile 10 327 10 337 10 465 10 491
liquid-render 10 129 10 136 10 331 10 357
lobsters 10 13 10 13 10 22 10 23
mail 10 149 10 153 10 211 10 220
psych-load 10 10 10 10 10 10 10 10
railsbench 10 10 10 10 10 10 10 10
rubocop 10 128 10 131 10 212 10 245
ruby-lsp 10 167 10 193 10 233 10 276
sequel 10 299 10 308 10 389 10 403
binarytrees 10 47 10 49 10 94 10 113
blurhash 10 58 10 60 10 125 10 141
erubi 10 78 10 85 10 96 10 105
etanni 10 53 10 24 10 62 10 26
fannkuchredux 10 10 10 10 10 14 10 16
fluentd 10 10 10 10 10 10 10 10
graphql 10 10 10 10 10 10 10 10
graphql-native 10 31 10 32 10 36 10 38
lee 10 10 10 11 10 16 10 21
matmul 10 10 10 10 10 15 10 15
nbody 10 202 10 206 10 454 10 466
nqueens 10 109 10 110 10 110 10 421
optcarrot 10 10 10 10 10 10 10 10
protoboeuf 10 167 10 183 10 540 10 652
rack 10 421 10 481 10 750 10 859
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 26 10 28
tinygql 10 21 10 22 10 45 10 49
30k_ifelse 10 10 10 10 10 70 10 70
30k_methods 10 10 10 10 10 26 10 26
cfunc_itself 10 252 10 245 10 1104 10 1210
fib 10 106 10 106 10 616 10 618
getivar 10 258 10 244 10 1279 10 1367
keyword_args 10 86 10 88 10 721 10 945
respond_to 10 85 10 88 10 1727 10 2148
ruby-xor 10 30 10 27 10 94 10 128
setivar 10 416 10 419 10 2349 10 3580
setivar_object 10 226 10 266 10 472 10 735
setivar_young 10 229 10 273 10 501 10 743
str_concat 10 328 10 310 10 638 10 565
throw 10 914 10 861 10 1164 10 1046

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) 1421677 1227949 214 2184 0 0% 0 0 67.740634
chunky-png (click) 331874 253630 82 1044 1 0% 0 0 31.612704
erubi-rails (click) 1685335 1394352 272 2905 6 0% 0 0 83.740391
hexapdf (click) 1290468 1089628 504 11526 41 0% 0 0 343.058906
liquid-c (click) 521280 414011 119 1757 5 0% 0 0 49.238377
liquid-compile (click) 444596 320065 151 2069 2 0% 0 0 65.3678
liquid-render (click) 618609 452654 136 2292 8 0% 0 0 65.111371
lobsters (click) 7561151 6388947 3110 48944 89 0% 0 0 1675.279027
mail (click) 725115 598526 346 4860 14 0% 0 0 147.287845
psych-load (click) 298577 241884 61 617 3 0% 0 0 20.132882
railsbench (click) 2949194 2552104 1633 15958 57 0% 0 0 472.279494
rubocop (click) 5328641 4446795 2901 48293 112 0% 4 0 1483.073044
ruby-lsp (click) 629084 552118 159 2818 28 0% 0 0 80.627529
sequel (click) 484568 448336 16 116 0 0% 0 0 4.158115
binarytrees (click) 8756 7777 10 70 0 0% 0 0 2.81495
blurhash (click) 50429 43273 31 455 0 0% 0 0 15.884414
erubi (click) 271362 208084 10 112 0 0% 0 0 4.057669
etanni (click) 29692 28216 11 91 0 0% 0 0 3.048983
fannkuchredux (click) 21198 26809 5 228 0 0% 0 0 7.414022
fluentd (click) 379812 326146 4 28 0 0% 0 0 1.130262
graphql (click) 433409 381760 66 644 0 0% 0 0 20.677865
graphql-native (click) 404678 349551 41 273 0 0% 0 0 9.555212
lee (click) 294824 256825 34 528 0 0% 0 0 17.259508
matmul (click) 8300 983 8 85 0 0% 0 0 2.929654
nbody (click) 14798 17933 11 185 0 0% 0 0 5.386267
nqueens (click) 22703 29059 10 260 0 0% 0 0 8.124442
optcarrot (click) 307949 264223 188 4285 34 0% 0 0 104.106395
protoboeuf (click) 146719 135165 17 1366 0 0% 0 0 42.889813
rack (click) 277534 222960 37 418 0 0% 0 0 11.683262
ruby-json (click) 30222 27890 8 167 0 0% 0 0 4.984291
rubykon (click) 128574 113214 137 1471 3 0% 0 0 42.961261
sudoku (click) 42479 58142 8 475 0 0% 0 0 15.233368
tinygql (click) 309879 278475 58 741 5 0% 0 0 22.572284
30k_ifelse (click) 5134674 4729436 9259 50778 0 0% 0 0 1535.629478
30k_methods (click) 2016920 1591842 5778 19339 0 0% 0 0 524.636334
cfunc_itself (click) 7534 6180 9 70 0 0% 0 0 2.242852
fib (click) 4970 5003 8 49 0 0% 0 0 1.797942
getivar (click) 5963 6375 8 74 0 0% 0 0 2.229132
keyword_args (click) 8409 6912 10 72 0 0% 0 0 2.546011
respond_to (click) 8277 8274 9 85 0 0% 0 0 2.667475
ruby-xor (click) 5593 5796 5 62 0 0% 0 0 1.786754
setivar (click) 5102 5375 8 54 0 0% 0 0 1.810705
setivar_object (click) 5470 5395 8 54 0 0% 0 0 1.854338
setivar_young (click) 6182 6093 9 62 0 0% 0 0 2.09396
str_concat (click) 7774 7946 11 85 0 0% 0 0 3.016554
throw (click) 7895 6377 10 69 0 0% 0 0 2.611212

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.