YJIT Benchmarks

Details for Benchmarks at 2024-06-18 19:08:17 UTC

YJIT metrics from the yjit-bench suite using Ruby eb215c8dc6

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 72.1% faster than CRuby 3.4.0dev
  • the same speed as YJIT 3.3.3
On railsbench it is
  • 77.0% 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 10 115 10 118
chunky-png 10 18 10 15 10 36 10 36
erubi-rails 10 10 10 10 10 20 10 19
hexapdf 10 10 10 10 10 10 10 10
liquid-c 10 352 10 333 10 504 10 473
liquid-compile 10 340 10 330 10 490 10 485
liquid-render 10 137 10 134 10 353 10 350
mail 10 162 10 147 10 238 10 219
psych-load 10 10 10 10 10 10 10 10
railsbench 10 10 10 10 10 10 10 10
rubocop 10 137 10 127 10 231 10 234
ruby-lsp 10 204 10 191 10 261 10 280
sequel 10 325 10 302 10 437 10 404
binarytrees 10 51 10 47 10 117 10 112
blurhash 10 61 10 56 10 118 10 143
erubi 10 83 10 82 10 103 10 103
etanni 10 58 10 57 10 68 10 67
fannkuchredux 10 10 10 10 10 18 10 16
fluentd 10 10 10 10 10 10 10 10
graphql 10 10 10 10 10 10 10 10
graphql-native 10 34 10 33 10 39 10 38
lee 10 11 10 10 10 19 10 21
matmul 10 10 10 10 10 15 10 16
nbody 10 211 10 187 10 385 10 470
nqueens 10 115 10 101 10 115 10 449
optcarrot 10 10 10 10 10 10 10 10
protoboeuf 10 189 10 170 10 681 10 658
rack 10 468 10 471 10 804 10 851
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 31
tinygql 10 25 10 22 10 52 10 50
30k_ifelse 10 10 10 10 10 70 10 69
30k_methods 10 10 10 10 10 26 10 26
cfunc_itself 10 257 10 235 10 1104 10 1211
fib 10 102 10 94 10 616 10 618
getivar 10 255 10 223 10 1280 10 1353
keyword_args 10 90 10 84 10 726 10 944
respond_to 10 89 10 84 10 1783 10 2148
ruby-xor 10 31 10 25 10 100 10 131
setivar 10 428 10 396 10 2347 10 3566
setivar_object 10 264 10 270 10 654 10 737
setivar_young 10 264 10 269 10 651 10 729
str_concat 10 327 10 280 10 593 10 528
throw 10 974 10 816 10 1275 10 1029

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) 1403001 1138624 214 2166 0 0% 0 0 76.308259
chunky-png (click) 331257 268947 82 1053 1 0% 0 0 33.912631
erubi-rails (click) 1676338 1422797 272 2887 6 0% 0 0 95.969428
hexapdf (click) 1288449 1081167 505 11688 41 0% 0 0 375.368663
liquid-c (click) 517841 425381 119 1761 5 0% 0 0 53.513211
liquid-compile (click) 440653 379877 151 2069 2 0% 0 0 69.292474
liquid-render (click) 612809 533640 136 2302 8 0% 0 0 70.692664
mail (click) 723226 644565 346 4881 14 0% 0 0 156.854059
psych-load (click) 293176 243984 61 594 3 0% 0 0 20.511695
railsbench (click) 2927101 2507554 1633 15942 57 0% 0 0 559.05035
rubocop (click) 5434540 4493968 2891 50368 110 0% 4 0 1792.871383
ruby-lsp (click) 635601 567796 159 2903 28 0% 0 0 90.23148
sequel (click) 485653 415510 16 117 0 0% 0 0 5.335087
binarytrees (click) 8756 7777 10 70 0 0% 0 0 2.977581
blurhash (click) 50429 43324 31 457 0 0% 0 0 17.047797
erubi (click) 270214 214579 10 112 0 0% 0 0 4.397807
etanni (click) 30236 28701 12 96 0 0% 0 0 3.49468
fannkuchredux (click) 21198 26809 5 228 0 0% 0 0 7.939582
fluentd (click) 375007 342770 4 28 0 0% 0 0 1.245255
graphql (click) 430898 353752 66 648 0 0% 0 0 22.20829
graphql-native (click) 402926 330733 42 278 0 0% 0 0 10.764528
lee (click) 293365 214005 34 531 0 0% 0 0 18.301666
matmul (click) 8315 983 8 86 0 0% 0 0 3.009294
nbody (click) 14798 17933 11 185 0 0% 0 0 5.582984
nqueens (click) 22145 28339 10 257 0 0% 0 0 8.970386
optcarrot (click) 304463 277860 188 4310 34 0% 0 0 108.859816
protoboeuf (click) 140419 133760 17 1349 0 0% 0 0 43.391875
rack (click) 278736 232096 37 445 0 0% 0 0 13.130892
ruby-json (click) 30222 27890 8 167 0 0% 0 0 5.190887
rubykon (click) 128391 79892 137 1484 3 0% 0 0 45.291217
sudoku (click) 41994 57542 8 473 0 0% 0 0 16.104764
tinygql (click) 307906 259268 58 734 5 0% 0 0 23.38348
30k_ifelse (click) 5134674 4729436 9259 50778 0 0% 0 0 1909.65981
30k_methods (click) 2016920 1591842 5778 19339 0 0% 0 0 531.393027
cfunc_itself (click) 7534 6180 9 70 0 0% 0 0 2.343122
fib (click) 4970 5003 8 49 0 0% 0 0 1.982173
getivar (click) 5963 6375 8 74 0 0% 0 0 2.241247
keyword_args (click) 8409 6912 10 72 0 0% 0 0 2.600887
respond_to (click) 8277 8274 9 85 0 0% 0 0 2.776231
ruby-xor (click) 5565 5673 5 62 0 0% 0 0 1.864295
setivar (click) 5102 5375 8 54 0 0% 0 0 1.921893
setivar_object (click) 5470 5395 8 54 0 0% 0 0 1.996459
setivar_young (click) 6182 6093 9 62 0 0% 0 0 2.218097
str_concat (click) 7774 7946 11 85 0 0% 0 0 3.088601
throw (click) 7895 6377 10 69 0 0% 0 0 2.646862

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.