YJIT Benchmarks

Details for Benchmarks at 2022-12-18 20:09:50

Overall YJIT is 41.2% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 40.0% 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 5 214 5 214
hexapdf 5 15 5 15
liquid-render 5 174 5 174
mail 5 139 5 139
psych-load 5 15 5 15
railsbench 5 15 5 15
ruby-lsp 5 289 5 289
binarytrees 5 87 5 87
chunky_png 5 32 5 32
erubi 5 69 5 69
erubi_rails 5 1146 5 1146
etanni 5 42 5 42
fannkuchredux 5 21 5 21
lee 5 21 5 21
nbody 5 281 5 281
optcarrot 5 15 5 15
rubykon 5 15 5 15
30k_ifelse 5 50 5 50
30k_methods 5 20 5 20
cfunc_itself 5 488 5 488
fib 5 318 5 318
getivar 5 436 5 436
keyword_args 5 384 5 384
respond_to 5 707 5 707
setivar 5 1360 5 1360
setivar_object 5 367 5 367
str_concat 5 475 5 475

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) 2266301 2263960 25 390 5 1% 0 0 0
hexapdf (click) 1715534 1714074 819 10438 217 2% 0 0 0
liquid-render (click) 564753 561848 156 1848 32 1% 0 0 0
mail (click) 1241159 1239502 373 4669 176 3% 0 0 0
psych-load (click) 777271 774704 70 651 19 2% 0 0 0
railsbench (click) 5520122 5519184 1779 14307 483 3% 0 0 0
ruby-lsp (click) 5397158 5396466 6428 41841 1536 3% 14404 0 0
binarytrees (click) 96267 94002 16 140 3 2% 0 0 0
chunky_png (click) 798229 796830 89 1210 31 2% 0 0 0
erubi (click) 738681 738044 13 149 4 2% 0 0 0
erubi_rails (click) 4662204 4661018 329 2832 111 3% 0 0 0
etanni (click) 112867 110494 13 108 3 2% 0 0 0
fannkuchredux (click) 108481 107742 14 273 3 1% 0 0 0
lee (click) 799909 797870 79 813 24 2% 0 0 0
nbody (click) 109148 108114 15 245 5 2% 0 0 0
optcarrot (click) 519950 518430 203 4439 92 2% 0 0 0
rubykon (click) 235447 233164 146 1639 30 1% 0 0 0
30k_ifelse (click) 5238111 5235722 9268 57200 3 0% 0 0 0
30k_methods (click) 1993486 1991930 5787 19427 3 0% 0 0 0
cfunc_itself (click) 93752 92274 13 123 3 2% 0 0 0
fib (click) 92325 91458 13 112 3 2% 0 0 0
getivar (click) 94108 92482 13 150 3 2% 0 0 0
keyword_args (click) 94538 93002 14 125 3 2% 0 0 0
respond_to (click) 87067 85164 13 150 4 2% 0 0 0
setivar (click) 93149 91962 13 130 3 2% 0 0 0
setivar_object (click) 93488 92118 13 130 3 2% 0 0 0
str_concat (click) 94031 92462 15 149 5 3% 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.