YJIT Benchmarks

Details for Benchmarks at 2022-09-08 20:12:02

Overall YJIT is 33.9% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 32.6% 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 MJIT warmups MJIT iters YJIT warmups YJIT iters
activerecord 5 195 75 195 20 195
hexapdf 5 15 20 15
liquid-render 5 160 75 160 20 160
mail 5 127 75 127 20 127
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 61 75 61 20 61
erubi_rails 5 1043 75 1043 20 1043
fannkuchredux 5 15 56 15 20 15
lee 5 21 75 21 20 21
nbody 5 263 75 263 20 263
optcarrot 5 15 75 15 20 15
rubykon 5 15 32 15 20 15
30k_ifelse 5 53 68 53 20 53
30k_methods 5 25 23 25 20 25
cfunc_itself 5 446 75 446 20 446
fib 5 309 75 309 20 309
getivar 5 469 75 469 20 469
keyword_args 5 384 75 384 20 384
respond_to 5 103 75 103 20 103
setivar 5 379 75 379 20 379
str_concat 5 421 75 421 20 421

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) 1078199 782073 131 1280 0 0% 0 0 0
hexapdf (click) 1193847 923640 682 8012 9 0% 0 0 0
liquid-render (click) 536311 401884 150 1672 2 0% 0 0 0
mail (click) 977463 678866 375 5755 17 0% 0 0 0
psych-load (click) 357751 261659 68 480 1 0% 0 0 0
railsbench (click) 2766903 1972235 1436 10075 16 0% 0 0 0
binarytrees (click) 176375 127327 13 58 0 0% 0 0 0
chunky_png (click) 405559 291572 88 1043 0 0% 0 0 0
erubi (click) 343287 251661 10 79 0 0% 0 0 0
erubi_rails (click) 2047607 1433293 300 2036 4 0% 0 0 0
fannkuchredux (click) 190519 137095 10 196 0 0% 0 0 0
lee (click) 399927 290613 58 634 0 0% 0 0 0
nbody (click) 188471 136946 12 171 0 0% 0 0 0
optcarrot (click) 534967 441707 207 3588 20 0% 0 0 0
rubykon (click) 332599 256882 144 1578 1 0% 0 0 0
30k_ifelse (click) 5703991 4347466 9265 57804 0 0% 0 0 0
30k_methods (click) 2209847 1656681 5784 19361 0 0% 0 0 0
cfunc_itself (click) 175735 126474 10 49 0 0% 0 0 0
fib (click) 174071 126082 10 38 0 0% 0 0 0
getivar (click) 174583 127466 10 65 0 0% 0 0 0
keyword_args (click) 176439 127165 11 51 0 0% 0 0 0
respond_to (click) 177271 127554 10 64 0 0% 0 0 0
setivar (click) 175799 126989 10 38 0 0% 0 0 0
str_concat (click) 174775 126996 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.