YJIT Benchmarks

Details for Benchmarks at 2024-06-17 06:08:17 UTC

YJIT metrics from the yjit-bench suite using Ruby ef423dd6d4

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 70.1% faster than CRuby 3.4.0dev
  • 8.7% faster than YJIT 3.3.1
On railsbench it is
  • 78.3% faster than CRuby 3.4.0dev
  • 6.9% 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 3.3.1 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 44 10 49 10 103 10 124
chunky-png 10 17 10 16 10 34 10 37
erubi-rails 10 10 10 10 10 18 10 20
hexapdf 10 10 10 10 10 10 10 10
liquid-c 10 332 10 336 10 465 10 485
liquid-compile 10 329 10 339 10 461 10 496
liquid-render 10 129 10 136 10 331 10 361
lobsters 10 13 10 13 10 22 10 23
mail 10 149 10 151 10 211 10 219
psych-load 10 10 10 10 10 10 10 10
railsbench 10 10 10 10 10 10 10 10
rubocop 10 128 10 131 10 213 10 243
ruby-lsp 10 166 10 195 10 282 10 257
sequel 10 298 10 302 10 388 10 423
binarytrees 10 47 10 49 10 94 10 113
blurhash 10 58 10 60 10 125 10 144
erubi 10 78 10 85 10 96 10 108
etanni 10 53 10 58 10 61 10 67
fannkuchredux 10 10 10 10 10 14 10 19
fluentd 10 10 10 10 10 10 10 10
graphql 10 10 10 10 10 10 10 10
graphql-native 10 31 10 35 10 36 10 25
lee 10 10 10 10 10 16 10 20
matmul 10 10 10 10 10 15 10 18
nbody 10 203 10 209 10 454 10 498
nqueens 10 107 10 108 10 110 10 454
optcarrot 10 10 10 10 10 10 10 10
protoboeuf 10 167 10 180 10 540 10 659
rack 10 424 10 469 10 740 10 890
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 33
tinygql 10 21 10 23 10 46 10 50
30k_ifelse 10 10 10 10 10 70 10 69
30k_methods 10 10 10 10 10 26 10 26
cfunc_itself 10 252 10 251 10 1104 10 1211
fib 10 106 10 104 10 616 10 617
getivar 10 258 10 245 10 1274 10 1368
keyword_args 10 86 10 87 10 720 10 945
respond_to 10 85 10 91 10 1727 10 2147
ruby-xor 10 30 10 29 10 94 10 144
setivar 10 415 10 419 10 2353 10 3580
setivar_object 10 226 10 271 10 447 10 734
setivar_young 10 229 10 271 10 501 10 743
str_concat 10 327 10 318 10 638 10 568
throw 10 915 10 831 10 1165 10 1042

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) 1407270 1200478 214 2182 0 0% 0 0 77.417406
chunky-png (click) 329203 266875 82 1053 1 0% 0 0 33.580895
erubi-rails (click) 1674007 1445754 272 2912 6 0% 0 0 97.12695
hexapdf (click) 1286148 1078781 504 11684 40 0% 0 0 372.553741
liquid-c (click) 515760 439648 119 1761 5 0% 0 0 53.130768
liquid-compile (click) 439482 354054 151 2075 2 0% 0 0 69.755169
liquid-render (click) 610965 507363 136 2305 8 0% 0 0 70.590802
lobsters (click) 7514018 6213309 3076 49040 82 0% 0 0 2079.961204
mail (click) 721149 650672 346 4881 14 0% 0 0 156.386171
psych-load (click) 291104 233680 61 594 3 0% 0 0 20.380126
railsbench (click) 2932999 2456423 1633 15992 57 0% 0 0 558.796228
rubocop (click) 5430855 4392368 2891 50372 111 0% 4 0 1789.810478
ruby-lsp (click) 624110 567766 159 2903 28 0% 0 0 89.970648
sequel (click) 481425 427697 16 117 0 0% 0 0 5.240303
binarytrees (click) 8756 7777 10 70 0 0% 0 0 2.982007
blurhash (click) 50434 43304 31 457 0 0% 0 0 16.659685
erubi (click) 268155 220694 10 112 0 0% 0 0 4.450625
etanni (click) 30236 28701 12 96 0 0% 0 0 3.45596
fannkuchredux (click) 21198 26809 5 228 0 0% 0 0 7.787881
fluentd (click) 373163 334684 4 28 0 0% 0 0 1.212022
graphql (click) 429187 343899 66 648 0 0% 0 0 22.116361
graphql-native (click) 401230 345469 42 278 0 0% 0 0 10.57444
lee (click) 291718 236870 34 531 0 0% 0 0 18.208472
matmul (click) 8315 983 8 86 0 0% 0 0 3.126889
nbody (click) 14798 17933 11 185 0 0% 0 0 5.608695
nqueens (click) 22145 28339 10 257 0 0% 0 0 8.919311
optcarrot (click) 304463 277860 188 4310 34 0% 0 0 109.173574
protoboeuf (click) 140419 133760 17 1349 0 0% 0 0 43.661829
rack (click) 274126 202353 37 414 0 0% 0 0 12.600682
ruby-json (click) 30222 27890 8 167 0 0% 0 0 5.148238
rubykon (click) 129933 105991 137 1498 3 0% 0 0 45.610319
sudoku (click) 41994 57542 8 473 0 0% 0 0 15.945579
tinygql (click) 305847 240807 58 734 5 0% 0 0 23.100648
30k_ifelse (click) 5134674 4729436 9259 50778 0 0% 0 0 2138.602985
30k_methods (click) 2016920 1591842 5778 19339 0 0% 0 0 533.458332
cfunc_itself (click) 7534 6180 9 70 0 0% 0 0 2.35276
fib (click) 4970 5003 8 49 0 0% 0 0 1.967562
getivar (click) 5963 6375 8 74 0 0% 0 0 2.235851
keyword_args (click) 8409 6912 10 72 0 0% 0 0 2.589389
respond_to (click) 8277 8274 9 85 0 0% 0 0 2.796892
ruby-xor (click) 5565 5673 5 62 0 0% 0 0 1.885586
setivar (click) 5102 5375 8 54 0 0% 0 0 1.890322
setivar_object (click) 5470 5395 8 54 0 0% 0 0 1.956487
setivar_young (click) 6182 6093 9 62 0 0% 0 0 2.18999
str_concat (click) 7774 7946 11 85 0 0% 0 0 3.285187
throw (click) 7895 6377 10 69 0 0% 0 0 2.653447

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.