YJIT Benchmarks

Details for Benchmarks at 2024-03-14 19:08:17 UTC

YJIT metrics from the yjit-bench suite using Ruby c2170e5c2b

Using the geomean of the headline benchmarks for x86 YJIT 3.4.0dev is
  • 72.4% faster than CRuby 3.4.0dev
On railsbench it is
  • 78.6% faster than CRuby 3.4.0dev

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 3.4.0dev warmups YJIT 3.4.0dev iters
activerecord 30 528 30 528
chunky-png 30 31 30 31
erubi-rails 30 19 30 19
hexapdf 30 15 30 15
liquid-c 30 344 30 344
liquid-compile 30 346 30 346
liquid-render 30 258 30 258
lobsters 30 23 30 23
mail 30 162 30 162
psych-load 30 15 30 15
railsbench 30 15 30 15
rubocop 30 184 30 184
ruby-lsp 30 184 30 184
sequel 30 292 30 292
binarytrees 30 89 30 89
blurhash 30 102 30 102
erubi 30 78 30 78
etanni 30 52 30 52
fannkuchredux 30 24 30 24
fluentd 30 15 30 15
graphql 30 15 30 15
graphql-native 30 22 30 22
lee 30 20 30 20
matmul 30 18 30 18
nbody 30 327 30 327
nqueens 30 307 30 307
optcarrot 30 15 30 15
rack 30 184 30 184
ruby-json 30 15 30 15
rubykon 30 15 30 15
sudoku 30 26 30 26
tinygql 30 42 30 42
30k_ifelse 30 55 30 55
30k_methods 30 23 30 23
cfunc_itself 30 827 30 827
fib 30 416 30 416
getivar 30 941 30 941
keyword_args 30 591 30 591
respond_to 30 1480 30 1480
ruby-xor 30 111 30 111
setivar 30 2390 30 2390
setivar_object 30 488 30 488
setivar_young 30 506 30 506
str_concat 30 404 30 404
throw 30 766 30 766

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) 884333 758539 49 525 0 0% 0 0 15.543444
chunky-png (click) 337530 299786 87 1093 1 0% 0 0 32.756941
erubi-rails (click) 1713908 1451368 279 2965 6 0% 0 0 83.291948
hexapdf (click) 1418961 1210960 596 12572 38 0% 0 0 369.425038
liquid-c (click) 523811 439383 119 1754 5 0% 0 0 48.237908
liquid-compile (click) 448121 390406 151 2070 2 0% 0 0 63.055049
liquid-render (click) 624047 549789 136 2284 8 0% 0 0 62.63098
lobsters (click) 7877986 6583851 3218 49993 76 0% 0 0 1705.682799
mail (click) 732377 635954 346 4847 14 0% 0 0 143.581949
psych-load (click) 300866 253380 65 653 3 0% 0 0 20.310932
railsbench (click) 2998971 2566995 1638 15786 54 0% 0 0 454.161561
rubocop (click) 5368186 4568120 2901 47757 106 0% 4 0 1425.424221
ruby-lsp (click) 633211 565327 159 2758 14 0% 0 0 76.656802
sequel (click) 492939 425248 16 119 0 0% 0 0 4.506412
binarytrees (click) 9392 7588 11 75 0 0% 0 0 2.967132
blurhash (click) 52000 47236 32 461 0 0% 0 0 15.542322
erubi (click) 271196 230596 10 111 0 0% 0 0 3.844591
etanni (click) 31686 27819 12 96 0 0% 0 0 3.207028
fannkuchredux (click) 24912 17701 9 257 0 0% 0 0 7.887915
fluentd (click) 387855 311831 11 111 0 0% 0 0 3.610607
graphql (click) 442725 373994 70 673 0 0% 0 0 21.256291
graphql-native (click) 408277 343538 41 271 0 0% 0 0 9.336287
lee (click) 315831 277753 49 731 0 0% 0 0 22.870325
matmul (click) 12586 12175 12 126 0 0% 0 0 4.219472
nbody (click) 15103 7574 11 185 0 0% 0 0 5.023559
nqueens (click) 23609 17866 10 261 0 0% 0 0 7.887164
optcarrot (click) 317087 278432 192 4342 34 0% 0 0 104.364149
rack (click) 272079 234500 31 319 0 0% 0 0 9.121161
ruby-json (click) 28135 23070 12 198 0 0% 0 0 5.854882
rubykon (click) 133625 106471 143 1498 3 0% 0 0 41.902737
sudoku (click) 47007 45195 11 507 0 0% 0 0 15.547309
tinygql (click) 313098 266157 63 787 5 0% 0 0 23.206833
30k_ifelse (click) 5222597 4374236 9264 50813 0 0% 0 0 1496.074764
30k_methods (click) 2019614 1594052 5782 19369 0 0% 0 0 523.440782
cfunc_itself (click) 7589 5810 9 70 0 0% 0 0 2.234842
fib (click) 5017 4633 8 49 0 0% 0 0 1.768557
getivar (click) 6102 6023 8 74 0 0% 0 0 2.230137
keyword_args (click) 8472 6502 10 72 0 0% 0 0 2.544052
respond_to (click) 8341 7760 9 85 0 0% 0 0 2.659598
ruby-xor (click) 8110 7110 9 87 0 0% 0 0 3.158803
setivar (click) 5193 4604 8 54 0 0% 0 0 1.751495
setivar_object (click) 5561 4624 8 54 0 0% 0 0 1.848682
setivar_young (click) 6284 5274 9 62 0 0% 0 0 2.048951
str_concat (click) 7932 7204 11 85 0 0% 0 0 2.822904
throw (click) 7946 6065 10 69 0 0% 0 0 2.480082

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.