YJIT Benchmarks

Details for Benchmarks at 2024-05-21 06:08:14 UTC

YJIT metrics from the yjit-bench suite using Ruby c773453c77

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 71.6% faster than CRuby 3.4.0dev
  • 10.8% faster than YJIT 3.3.1
On railsbench it is
  • 78.9% faster than CRuby 3.4.0dev
  • 9.9% 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 3.4.0dev 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 49 10 103 10 117
chunky-png 10 16 10 34 10 38
erubi-rails 10 10 10 18 10 20
hexapdf 10 10 10 10 10 10
liquid-c 10 340 10 470 10 501
liquid-compile 10 338 10 465 10 496
liquid-render 10 134 10 332 10 360
lobsters 10 13 10 22 10 23
mail 10 153 10 211 10 228
psych-load 10 10 10 10 10 10
railsbench 10 10 10 10 10 10
rubocop 10 129 10 212 10 245
ruby-lsp 10 181 10 232 10 254
sequel 10 315 10 389 10 421
binarytrees 10 50 10 94 10 124
blurhash 10 56 10 125 10 143
erubi 10 84 10 97 10 106
etanni 10 25 10 61 10 28
fannkuchredux 10 10 10 14 10 16
fluentd 10 10 10 10 10 10
graphql 10 10 10 10 10 10
graphql-native 10 35 10 36 10 40
lee 10 11 10 16 10 19
matmul 10 10 10 15 10 16
nbody 10 188 10 454 10 467
nqueens 10 103 10 110 10 459
optcarrot 10 10 10 10 10 10
protoboeuf 10 174 10 540 10 771
rack 10 459 10 762 10 831
ruby-json 10 10 10 10 10 10
rubykon 10 10 10 10 10 10
sudoku 10 10 10 26 10 31
tinygql 10 23 10 45 10 51
30k_ifelse 10 10 10 70 10 70
30k_methods 10 10 10 26 10 26
cfunc_itself 10 240 10 1104 10 1211
fib 10 98 10 616 10 618
getivar 10 210 10 1280 10 1340
keyword_args 10 87 10 721 10 944
respond_to 10 93 10 1727 10 2146
ruby-xor 10 27 10 94 10 155
setivar 10 344 10 2350 10 3526
setivar_object 10 250 10 472 10 702
setivar_young 10 249 10 501 10 736
str_concat 10 310 10 638 10 630
throw 10 923 10 1164 10 1134

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) 1421248 1186556 214 2184 0 0% 0 0 68.195111
chunky-png (click) 333404 263089 82 1048 1 0% 0 0 31.938264
erubi-rails (click) 1684220 1401285 272 2893 6 0% 0 0 83.941356
hexapdf (click) 1290147 1105927 504 11519 41 0% 0 0 345.149209
liquid-c (click) 521560 406158 119 1757 5 0% 0 0 50.427576
liquid-compile (click) 443932 368868 151 2067 2 0% 0 0 64.795294
liquid-render (click) 618663 526446 136 2292 8 0% 0 0 65.274681
lobsters (click) 7577363 6287090 3109 49113 89 0% 0 0 1693.194477
mail (click) 726292 623847 346 4865 14 0% 0 0 147.978894
psych-load (click) 298240 233446 61 615 3 0% 0 0 19.964237
railsbench (click) 2949731 2503019 1633 15936 57 0% 0 0 474.447617
rubocop (click) 5326888 4600897 2901 48278 112 0% 4 0 1487.400376
ruby-lsp (click) 628380 576000 159 2808 28 0% 0 0 80.967647
sequel (click) 484810 440225 16 120 0 0% 0 0 4.747655
binarytrees (click) 8756 7777 10 70 0 0% 0 0 2.830652
blurhash (click) 50429 43273 31 455 0 0% 0 0 15.911514
erubi (click) 271818 208559 10 112 0 0% 0 0 3.944928
etanni (click) 29692 28216 11 91 0 0% 0 0 3.061035
fannkuchredux (click) 21198 26809 5 228 0 0% 0 0 7.268028
fluentd (click) 379930 326280 4 28 0 0% 0 0 1.166399
graphql (click) 433168 373147 66 644 0 0% 0 0 20.594506
graphql-native (click) 405276 350177 42 278 0 0% 0 0 9.617099
lee (click) 295280 281896 34 528 0 0% 0 0 17.500945
matmul (click) 8300 983 8 85 0 0% 0 0 2.985634
nbody (click) 14798 17933 11 185 0 0% 0 0 5.258534
nqueens (click) 22703 29059 10 260 0 0% 0 0 8.180987
optcarrot (click) 307949 264223 188 4285 34 0% 0 0 104.939869
protoboeuf (click) 146719 143337 17 1366 0 0% 0 0 42.896639
rack (click) 277990 207039 37 418 0 0% 0 0 11.612904
ruby-json (click) 30222 27890 8 167 0 0% 0 0 5.034424
rubykon (click) 127405 128474 137 1461 3 0% 0 0 42.729305
sudoku (click) 42479 58142 8 475 0 0% 0 0 15.364179
tinygql (click) 310010 262090 58 741 5 0% 0 0 22.495821
30k_ifelse (click) 5134674 4729436 9259 50778 0 0% 0 0 1541.571141
30k_methods (click) 2016920 1591842 5778 19339 0 0% 0 0 533.453344
cfunc_itself (click) 7534 6180 9 70 0 0% 0 0 2.284719
fib (click) 4970 5003 8 49 0 0% 0 0 1.789338
getivar (click) 5963 6375 8 74 0 0% 0 0 2.162237
keyword_args (click) 8409 6912 10 72 0 0% 0 0 2.553227
respond_to (click) 8277 8274 9 85 0 0% 0 0 2.721625
ruby-xor (click) 5593 5796 5 62 0 0% 0 0 1.772928
setivar (click) 5102 5375 8 54 0 0% 0 0 1.852945
setivar_object (click) 5470 5395 8 54 0 0% 0 0 1.894702
setivar_young (click) 6182 6093 9 62 0 0% 0 0 2.112621
str_concat (click) 7774 7946 11 85 0 0% 0 0 2.974857
throw (click) 7895 6377 10 69 0 0% 0 0 2.628074

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.