YJIT Benchmarks

Details for Benchmarks at 2024-08-01 09:39:41 UTC

YJIT metrics from the yjit-bench suite using Ruby e0f4cbd612.

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 75.1% faster than CRuby 3.4.0dev
  • 1.1% faster than YJIT 3.3.4
On railsbench it is
  • 80.3% faster than CRuby 3.4.0dev
  • the same speed as YJIT 3.3.4
x86_64 runtime: 3 hours, 9 minutes
aarch64 runtime: 5 hours, 48 minutes

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 51 10 47 10 113 10 114
chunky-png 10 18 10 15 10 36 10 37
erubi-rails 10 10 10 10 10 20 10 18
hexapdf 10 10 10 10 10 10 10 10
liquid-c 10 346 10 333 10 492 10 487
liquid-compile 10 337 10 328 10 479 10 488
liquid-render 10 137 10 130 10 352 10 366
lobsters 10 14 10 13 10 24 10 23
mail 10 161 10 150 10 234 10 219
psych-load 10 10 10 10 10 10 10 10
railsbench 10 10 10 10 10 10 10 10
rubocop 10 136 10 128 10 230 10 236
ruby-lsp 10 130 10 123 10 213 10 212
sequel 10 322 10 305 10 427 10 411
binarytrees 10 51 10 48 10 119 10 114
blurhash 10 61 10 58 10 120 10 144
erubi 10 83 10 87 10 103 10 109
etanni 10 59 10 57 10 69 10 66
fannkuchredux 10 10 10 10 10 17 10 17
fluentd 10 10 10 10 10 10 10 10
graphql 10 10 10 10 10 10 10 10
graphql-native 10 36 10 32 10 40 10 38
lee 10 11 10 10 10 20 10 19
matmul 10 10 10 10 10 15 10 16
nbody 10 208 10 196 10 407 10 468
nqueens 10 117 10 103 10 112 10 461
optcarrot 10 10 10 10 10 10 10 10
protoboeuf 10 178 10 168 10 603 10 663
protoboeuf-encode 10 123 10 126 10 275 10 363
rack 10 459 10 481 10 797 10 941
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 21 10 51 10 51
30k_ifelse 10 10 10 10 10 70 10 69
30k_methods 10 10 10 10 10 26 10 26
cfunc_itself 10 256 10 239 10 1105 10 1213
fib 10 105 10 97 10 616 10 618
getivar 10 252 10 232 10 1277 10 1740
keyword_args 10 92 10 83 10 720 10 947
object-new 10 227 10 199 10 301 10 257
respond_to 10 94 10 84 10 1728 10 2092
ruby-xor 10 31 10 26 10 98 10 133
setivar 10 421 10 393 10 2354 10 4856
setivar_object 10 264 10 267 10 653 10 703
setivar_young 10 265 10 266 10 650 10 700
str_concat 10 327 10 310 10 593 10 532
throw 10 987 10 902 10 1286 10 1141

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) 1373088 1191412 214 2169 0 0% 0 0 85.543673
chunky-png (click) 297377 258406 83 1054 1 0% 0 0 37.481521
erubi-rails (click) 1651151 1373251 274 2920 6 0% 0 0 106.546639
hexapdf (click) 1266781 1127608 505 11812 39 0% 0 0 430.884583
liquid-c (click) 483833 427556 119 1768 5 0% 0 0 60.885058
liquid-compile (click) 402058 368122 151 2083 2 0% 0 0 76.689219
liquid-render (click) 581407 500455 136 2323 8 0% 0 0 81.106206
lobsters (click) 7564344 6273795 3081 49853 81 0% 0 0 2363.693785
mail (click) 746493 675706 347 5553 14 0% 0 0 200.893924
psych-load (click) 258568 248384 61 592 3 0% 0 0 22.116148
railsbench (click) 2903880 2426549 1636 15957 57 0% 0 0 619.226093
rubocop (click) 5442373 4449713 2892 50925 110 0% 4 0 2039.942811
ruby-lsp (click) 816641 709105 309 4750 24 0% 0 0 169.283769
sequel (click) 467248 376996 16 117 0 0% 0 0 5.352701
binarytrees (click) 8765 7825 10 70 0 0% 0 0 3.374792
blurhash (click) 50455 44132 31 457 0 0% 0 0 18.768727
erubi (click) 235903 210737 10 112 0 0% 0 0 4.599917
etanni (click) 28766 27121 12 95 0 0% 0 0 3.782329
fannkuchredux (click) 24067 31522 5 256 0 0% 0 0 10.386648
fluentd (click) 339955 299935 4 28 0 0% 0 0 1.377467
graphql (click) 398193 353025 66 650 0 0% 0 0 24.3474
graphql-native (click) 369097 311944 41 273 0 0% 0 0 10.989496
lee (click) 259547 219451 34 531 0 0% 0 0 20.700246
matmul (click) 8318 9199 8 86 0 0% 0 0 3.471455
nbody (click) 15375 19248 11 195 0 0% 0 0 6.553758
nqueens (click) 24183 24054 10 283 0 0% 0 0 11.140503
optcarrot (click) 309371 266843 188 4402 34 0% 0 0 124.772229
protoboeuf (click) 140353 122267 17 1368 0 0% 0 0 51.911799
protoboeuf-encode (click) 221090 228639 19 1172 0 0% 0 0 39.247595
rack (click) 241106 225312 37 413 0 0% 0 0 13.952563
ruby-json (click) 24603 21785 8 170 0 0% 0 0 5.878447
rubykon (click) 130798 109409 137 1504 3 0% 0 0 51.521561
sudoku (click) 45653 24997 8 520 0 0% 0 0 20.428497
tinygql (click) 273945 256442 58 735 5 0% 0 0 25.676884
30k_ifelse (click) 5134674 4729436 9259 50778 0 0% 0 0 2027.139042
30k_methods (click) 2016920 1591842 5778 19339 0 0% 0 0 568.937223
cfunc_itself (click) 7533 6248 9 70 0 0% 0 0 2.553415
fib (click) 4970 5023 8 49 0 0% 0 0 2.032128
getivar (click) 5936 6875 8 74 0 0% 0 0 2.528619
keyword_args (click) 8408 6980 10 72 0 0% 0 0 2.87704
object-new (click) 4400 4255 7 45 0 0% 0 0 2.108767
respond_to (click) 8276 8394 9 85 0 0% 0 0 3.132359
ruby-xor (click) 5580 6003 5 62 0 0% 0 0 2.13095
setivar (click) 5087 5551 8 54 0 0% 0 0 2.064876
setivar_object (click) 5470 5459 8 54 0 0% 0 0 2.149629
setivar_young (click) 6180 6181 9 62 0 0% 0 0 2.385078
str_concat (click) 7776 8178 11 85 0 0% 0 0 3.409926
throw (click) 7899 6453 10 69 0 0% 0 0 2.836726

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.