YJIT Benchmarks

Details for Benchmarks at 2023-07-11 19:08:12

Overall YJIT is 51.3% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 53.2% faster than CRuby!
The basic "faster" measurement is the geomean of all "headlining" x86 benchmarks on this page.

Headlining Benchmarks

These are "headlining" because the "overall" speedup above is based on these benchmarks specifically.

Speed of each Ruby implementation (iterations/second) relative to the CRuby interpreter. Higher is better.

Other Benchmarks

Speed of each Ruby implementation (iterations/second) relative to the CRuby interpreter. Higher is better.

MicroBenchmarks

Speed of each Ruby implementation (iterations/second) relative to the CRuby interpreter. Higher is better.

Want Raw Graphs and CSV?

Benchmarks Speed Details

Benchmark Memory Usage Details

Number of Iterations and Warmups Tested

bench No JIT warmups No JIT iters YJIT warmups YJIT iters
activerecord 30 469 30 469
chunky-png 30 30 30 30
erubi-rails 30 1259 30 1259
hexapdf 30 15 30 15
liquid-c 30 333 30 333
liquid-compile 30 345 30 345
liquid-render 30 193 30 193
mail 30 148 30 148
psych-load 30 15 30 15
railsbench 30 15 30 15
ruby-lsp 30 363 30 363
sequel 30 284 30 284
binarytrees 30 87 30 87
erubi 30 76 30 76
etanni 30 48 30 48
fannkuchredux 30 26 30 26
lee 30 21 30 21
nbody 30 288 30 288
optcarrot 30 15 30 15
rack 15 15 15 15
ruby-json 30 15 30 15
rubykon 30 15 30 15
30k_ifelse 30 55 30 55
30k_methods 30 23 30 23
cfunc_itself 30 518 30 518
fib 30 448 30 448
getivar 30 945 30 945
keyword_args 30 417 30 417
respond_to 30 790 30 790
setivar 30 1667 30 1667
setivar_object 30 474 30 474
setivar_young 30 473 30 473
str_concat 30 420 30 420
throw 30 829 30 829

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
activerecord (click) 677656 623661 51 440 1 0% 0 0 0
chunky-png (click) 258682 249755 86 1063 1 0% 0 0 0
erubi-rails (click) 1620617 1366002 288 2476 1 0% 0 0 0
hexapdf (click) 1164564 1154074 600 10752 22 0% 0 0 0
liquid-c (click) 417487 422865 125 1642 3 0% 0 0 0
liquid-compile (click) 308147 309247 149 1999 2 0% 0 0 0
liquid-render (click) 461611 458349 146 1914 8 0% 0 0 0
mail (click) 638469 599475 346 4654 9 0% 0 0 0
psych-load (click) 213074 203777 64 572 1 0% 0 0 0
railsbench (click) 2285728 2028330 1355 11020 7 0% 0 0 0
ruby-lsp (click) 5112733 4711046 3402 44671 319 0% 4 0 0
sequel (click) 406718 376449 15 95 0 0% 0 0 0
binarytrees (click) 9076 7468 11 75 0 0% 0 0 0
erubi (click) 178711 182146 9 83 0 0% 0 0 0
etanni (click) 12317 10568 11 55 0 0% 0 0 0
fannkuchredux (click) 19589 20370 8 219 0 0% 0 0 0
lee (click) 238160 220701 49 657 0 0% 0 0 0
nbody (click) 16822 16627 10 178 0 0% 0 0 0
optcarrot (click) 321041 382506 197 4341 22 0% 0 0 0
rack (click) 183555 181212 28 259 0 0% 0 0 0
ruby-json (click) 25782 22253 12 192 0 0% 0 0 0
rubykon (click) 142794 149099 145 1627 2 0% 0 0 0
30k_ifelse (click) 5362743 4368508 9263 50806 0 0% 0 0 0
30k_methods (click) 1992158 1528536 5781 19360 0 0% 0 0 0
cfunc_itself (click) 6422 4947 8 61 0 0% 0 0 0
fib (click) 4917 4454 8 49 0 0% 0 0 0
getivar (click) 6058 5980 8 74 0 0% 0 0 0
keyword_args (click) 7197 5541 9 63 0 0% 0 0 0
respond_to (click) 6928 6857 8 76 0 0% 0 0 0
setivar (click) 5209 4541 8 54 0 0% 0 0 0
setivar_object (click) 5544 4561 8 54 0 0% 0 0 0
setivar_young (click) 6335 5537 9 62 0 0% 0 0 0
str_concat (click) 6802 6350 10 76 0 0% 0 0 0
throw (click) 3993 3228 10 37 0 0% 0 0 0

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.