YJIT Benchmarks

Details for Benchmarks at 2022-07-13 20:14:45

Overall YJIT is 36.2% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 31.2% 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 MJIT warmups MJIT iters YJIT warmups YJIT iters
activerecord 5 191 75 191 20 191
hexapdf 5 15 20 15
liquid-render 5 157 75 157 20 157
mail 5 126 75 126 20 126
psych-load 5 15 75 15 20 15
railsbench 5 15 75 15 20 15
binarytrees 5 66 75 66 20 66
chunky_png 5 30 75 30 20 30
erubi 5 53 75 53 20 53
erubi_rails 5 995 75 995 20 995
fannkuchredux 5 15 57 15 20 15
lee 5 21 75 21 20 21
nbody 5 275 75 275 20 275
optcarrot 5 15 75 15 20 15
rubykon 5 15 33 15 20 15
30k_ifelse 5 55 75 55 20 55
30k_methods 5 22 29 22 20 22
cfunc_itself 5 431 75 431 20 431
fib 5 314 75 314 20 314
getivar 5 763 75 763 20 763
keyword_args 5 375 75 375 20 375
respond_to 5 105 75 105 20 105
setivar 5 1857 75 1857 20 1857
str_concat 5 366 75 366 20 366

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) 1027931 763635 131 1262 0 0% 0 0 0
hexapdf (click) 1123355 881405 683 7753 9 0% 0 0 0
liquid-render (click) 513883 394342 150 1653 2 0% 0 0 0
mail (click) 905563 653724 374 5372 15 0% 0 0 0
psych-load (click) 344667 258194 68 462 1 0% 0 0 0
railsbench (click) 2638875 1928932 1436 9942 16 0% 0 0 0
binarytrees (click) 172763 129315 13 58 0 0% 0 0 0
chunky_png (click) 387163 288044 88 1031 0 0% 0 0 0
erubi (click) 321179 243865 10 79 0 0% 0 0 0
erubi_rails (click) 1942299 1396995 300 1987 4 0% 0 0 0
fannkuchredux (click) 186971 140148 10 188 0 0% 0 0 0
lee (click) 382619 286477 58 623 0 0% 0 0 0
nbody (click) 182427 136651 12 158 0 0% 0 0 0
optcarrot (click) 514395 442558 207 3585 20 0% 0 0 0
rubykon (click) 317467 250283 144 1530 2 0% 0 0 0
30k_ifelse (click) 5562011 4349774 9265 57804 0 0% 0 0 0
30k_methods (click) 2177179 1658236 5784 19361 0 0% 0 0 0
cfunc_itself (click) 171995 128408 10 49 0 0% 0 0 0
fib (click) 169947 127687 10 38 0 0% 0 0 0
getivar (click) 172251 130527 10 65 0 0% 0 0 0
keyword_args (click) 172827 129049 11 51 0 0% 0 0 0
respond_to (click) 174107 129862 10 64 0 0% 0 0 0
setivar (click) 171675 128722 10 38 0 0% 0 0 0
str_concat (click) 171611 129097 12 62 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.