YJIT Benchmarks

Details for Benchmarks at 2024-10-09 04:00:07 UTC

YJIT metrics from the yjit-bench suite using Ruby 3e98abc49d

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 73.8% faster than CRuby 3.4.0dev
  • 3.9% faster than YJIT 3.3.4
On railsbench it is
  • 80.9% faster than CRuby 3.4.0dev
  • 2.8% faster than YJIT 3.3.4

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 57 10 58 10 146 10 153
chunky-png 10 18 10 17 10 35 10 40
erubi-rails 10 10 10 10 10 23 10 22
hexapdf 10 10 10 10 10 10 10 10
liquid-c 10 341 10 350 10 501 10 520
liquid-compile 10 345 10 359 10 501 10 534
liquid-render 10 137 10 136 10 353 10 369
lobsters 10 13 10 13 10 24 10 23
mail 10 158 10 162 10 238 10 234
psych-load 10 10 10 10 10 10 10 10
railsbench 10 10 10 10 10 10 10 10
rubocop 10 135 10 134 10 234 10 241
ruby-lsp 10 128 10 131 10 215 10 217
sequel 10 322 10 324 10 438 10 441
binarytrees 10 51 10 51 10 118 10 111
blurhash 10 62 10 62 10 120 10 147
erubi 10 80 10 99 10 100 10 125
etanni 10 59 10 69 10 69 10 81
fannkuchredux 10 59 10 58 10 123 10 133
fluentd 10 38 10 37 10 42 10 49
graphql 10 62 10 60 10 75 10 75
graphql-native 10 28 10 40 10 34 10 46
lee 10 11 10 11 10 20 10 21
matmul 10 38 10 44 10 72 10 84
nbody 10 203 10 216 10 421 10 523
nqueens 10 113 10 119 10 113 10 447
optcarrot 10 10 10 10 10 10 10 10
protoboeuf 10 158 10 158 10 582 10 640
protoboeuf-encode 10 132 10 169 10 293 10 663
rack 10 458 10 499 10 831 10 907
ruby-json 10 64 10 66 10 76 10 78
rubykon 10 13 10 13 10 37 10 39
sudoku 10 48 10 50 10 139 10 156
tinygql 10 23 10 24 10 51 10 53
30k_ifelse 10 26 10 26 10 230 10 207
30k_methods 10 32 10 33 10 344 10 342
cfunc_itself 10 247 10 255 10 1108 10 1212
fib 10 102 10 99 10 616 10 623
getivar 10 214 10 230 10 1248 10 1739
keyword_args 10 90 10 88 10 729 10 947
object-new 10 230 10 187 10 309 10 234
respond_to 10 91 10 89 10 1785 10 2218
ruby-xor 10 186 10 190 10 497 10 841
setivar 10 348 10 421 10 2319 10 3127
setivar_object 10 216 10 234 10 536 10 489
setivar_young 10 216 10 236 10 529 10 494
str_concat 10 319 10 349 10 658 10 659
throw 10 973 10 834 10 1342 10 1043

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) 1475493 1196936 214 2301 0 0% 0 0 74.300682
chunky-png (click) 294102 258388 82 1090 1 0% 0 0 35.221665
erubi-rails (click) 1906566 1639092 245 2795 8 0% 0 0 84.184196
hexapdf (click) 1362295 1233964 507 12566 40 0% 0 0 394.799859
liquid-c (click) 490350 389169 119 1859 5 0% 0 0 56.88612
liquid-compile (click) 412219 344524 150 2202 2 0% 0 0 74.337637
liquid-render (click) 603866 533373 136 2473 12 0% 0 0 75.411164
lobsters (click) 8041622 6609863 3088 51861 183 0% 0 0 1900.534481
mail (click) 739825 617378 346 5481 16 0% 0 0 176.910398
psych-load (click) 254484 223945 60 613 3 0% 0 0 21.026328
railsbench (click) 3313044 2785421 1631 16671 41 0% 0 0 526.160603
rubocop (click) 5808573 5012792 2892 52168 110 0% 4 0 1686.598022
ruby-lsp (click) 886678 764594 308 4991 26 0% 0 0 156.013535
sequel (click) 466388 376216 15 124 0 0% 0 0 4.986336
binarytrees (click) 9627 8546 8 75 0 0% 0 0 3.332471
blurhash (click) 57136 49713 29 501 0 0% 0 0 18.521296
erubi (click) 224785 176903 9 108 0 0% 0 0 4.019724
etanni (click) 28241 26371 10 86 0 0% 0 0 3.215745
fannkuchredux (click) 24582 31311 6 262 0 0% 0 0 8.778131
fluentd (click) 342595 300598 11 114 0 0% 0 0 4.250559
graphql (click) 390482 334067 70 700 0 0% 0 0 23.685502
graphql-native (click) 375039 313449 41 275 0 0% 0 0 9.912682
lee (click) 255151 245278 33 567 0 0% 0 0 19.450547
matmul (click) 7323 8338 6 76 0 0% 0 0 2.863411
nbody (click) 15428 19810 9 201 0 0% 0 0 6.010945
nqueens (click) 21699 28649 8 252 0 0% 0 0 8.709547
optcarrot (click) 324183 308580 186 4508 34 0% 0 0 114.527075
protoboeuf (click) 163936 155521 15 1615 0 0% 0 0 52.889825
protoboeuf-encode (click) 235128 231105 17 1254 0 0% 0 0 36.520949
rack (click) 237925 207276 36 417 0 0% 0 0 12.261144
ruby-json (click) 27649 24400 11 194 0 0% 0 0 6.367838
rubykon (click) 137360 133862 132 1535 3 0% 0 0 48.687707
sudoku (click) 44908 40454 6 507 0 0% 0 0 17.200574
tinygql (click) 268287 243550 57 743 5 0% 0 0 23.515505
30k_ifelse (click) 6316209 5792190 9262 56478 0 0% 0 0 1818.922726
30k_methods (click) 2236110 1594189 5781 19364 0 0% 0 0 531.135576
cfunc_itself (click) 6912 5554 7 62 0 0% 0 0 2.227252
fib (click) 4585 4605 6 42 0 0% 0 0 1.79856
getivar (click) 5037 6014 6 64 0 0% 0 0 2.0517
keyword_args (click) 8181 6507 8 65 0 0% 0 0 2.529515
object-new (click) 3469 3394 5 35 0 0% 0 0 1.583983
respond_to (click) 9866 2524 7 100 0 0% 0 0 3.341534
ruby-xor (click) 6981 7365 7 77 0 0% 0 0 2.765669
setivar (click) 4188 4690 6 44 0 0% 0 0 1.696961
setivar_object (click) 4717 4838 6 46 0 0% 0 0 1.902469
setivar_young (click) 5427 5560 7 54 0 0% 0 0 2.11857
str_concat (click) 7646 7893 9 81 0 0% 0 0 3.097247
throw (click) 7384 5592 8 59 0 0% 0 0 2.336193

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.