YJIT Benchmarks

Details for Benchmarks at 2022-12-03 04:10:23

Overall YJIT is 38.3% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 39.6% 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 5 193 5 193
hexapdf 5 15 5 15
liquid-render 5 154 5 154
mail 5 140 5 140
psych-load 5 15 5 15
railsbench 5 15 5 15
ruby-lsp 5 281 5 281
binarytrees 5 86 5 86
chunky_png 5 32 5 32
erubi 5 69 5 69
erubi_rails 5 1135 5 1135
etanni 5 42 5 42
fannkuchredux 5 15 5 15
lee 5 22 5 22
nbody 5 253 5 253
optcarrot 5 15 5 15
rubykon 5 15 5 15
30k_ifelse 5 50 5 50
30k_methods 5 20 5 20
cfunc_itself 5 494 5 494
fib 5 327 5 327
getivar 5 438 5 438
keyword_args 5 384 5 384
respond_to 5 656 5 656
setivar 5 377 5 377
str_concat 5 483 5 483

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) 2263185 2262171 25 365 6 1% 0 0 0
hexapdf (click) 1706092 1705189 819 10236 216 2% 0 0 0
liquid-render (click) 579675 577803 156 1852 33 1% 0 0 0
mail (click) 1247223 1246349 373 4637 176 3% 0 0 0
psych-load (click) 783019 781652 70 651 19 2% 0 0 0
railsbench (click) 5515208 5513853 1779 14275 482 3% 0 0 0
ruby-lsp (click) 5345695 5345526 6428 41393 1523 3% 14354 0 0
binarytrees (click) 131647 131305 16 139 4 2% 0 0 0
chunky_png (click) 803227 802911 89 1181 31 2% 0 0 0
erubi (click) 716858 715634 13 147 4 2% 0 0 0
erubi_rails (click) 4641468 4641462 329 2784 111 3% 0 0 0
etanni (click) 132439 132046 13 107 4 3% 0 0 0
fannkuchredux (click) 143849 142353 14 272 4 1% 0 0 0
lee (click) 802449 802296 79 812 25 3% 0 0 0
nbody (click) 144443 142840 15 244 6 2% 0 0 0
optcarrot (click) 563020 561187 203 4438 93 2% 0 0 0
rubykon (click) 271554 271037 145 1635 31 1% 0 0 0
30k_ifelse (click) 5280777 5280299 9268 57199 4 0% 0 0 0
30k_methods (click) 2021641 2019300 5787 19426 4 0% 0 0 0
cfunc_itself (click) 129581 127764 13 122 4 3% 0 0 0
fib (click) 121224 119387 13 111 4 3% 0 0 0
getivar (click) 129935 127975 13 138 4 2% 0 0 0
keyword_args (click) 130363 128211 14 124 4 3% 0 0 0
respond_to (click) 130085 128068 13 138 5 3% 0 0 0
setivar (click) 128968 127181 13 118 4 3% 0 0 0
str_concat (click) 122630 120291 15 137 6 4% 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.