YJIT Benchmarks

Details for Benchmarks at 2022-04-23 19:12:03

Overall YJIT is 32.0% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 30.7% 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 5 184 75 184 20 184
hexapdf 5 15 20 15
liquid-render 5 156 75 156 20 156
mail 5 121 75 121 20 121
psych-load 5 15 75 15 20 15
railsbench 5 15 75 15 20 15
binarytrees 5 63 75 63 20 63
chunky_png 5 30 75 30 20 30
discourse 5 39 75 39 20 39
erubi 5 47 75 47 20 47
erubi_rails 5 980 75 980 20 980
fannkuchredux 5 15 73 15 20 15
lee 5 21 75 21 20 21
nbody 5 274 75 274 20 274
optcarrot 5 15 75 15 20 15
rubykon 5 15 28 15 20 15
30k_ifelse 5 56 75 56 20 56
30k_methods 5 22 29 22 20 22
cfunc_itself 5 452 75 452 20 452
fib 5 320 75 320 20 320
getivar 5 770 75 770 20 770
keyword_args 5 379 75 379 20 379
respond_to 5 105 75 105 20 105
setivar 5 1782 75 1782 20 1782

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) 994961 816089 185 1185 0 0% 0 0 0
hexapdf (click) 1100241 923614 600 7912 427 5% 0 0 321
liquid-render (click) 503441 425443 143 1535 92 5% 0 0 5
mail (click) 844305 655277 324 4965 96 1% 0 0 111
psych-load (click) 341841 286772 59 434 1 0% 0 0 0
railsbench (click) 2597329 2102588 1297 9390 224 2% 0 0 26
binarytrees (click) 168401 138123 11 58 0 0% 0 0 0
chunky_png (click) 376337 309841 85 864 0 0% 0 0 0
discourse (click) 6201937 5571258 3089 26535 3134 11% 2 0 140
erubi (click) 329681 277241 10 79 0 0% 0 0 0
erubi_rails (click) 1956817 1596267 262 1888 12 0% 0 0 3
fannkuchredux (click) 183441 149524 8 188 0 0% 0 0 0
lee (click) 379025 316065 45 582 68 11% 0 0 13
nbody (click) 177809 145437 10 158 0 0% 0 0 0
optcarrot (click) 505425 451283 198 3575 20 0% 0 0 0
rubykon (click) 311505 259972 140 1543 2 0% 0 0 0
30k_ifelse (click) 5558161 4358874 9263 57806 0 0% 0 0 0
30k_methods (click) 2172881 1665724 5782 19362 0 0% 0 0 0
cfunc_itself (click) 167377 137081 8 49 0 0% 0 0 0
fib (click) 165969 136689 8 38 0 0% 0 0 0
getivar (click) 168465 139574 8 65 0 0% 0 0 0
keyword_args (click) 167761 137393 9 51 0 0% 0 0 0
respond_to (click) 170321 138909 8 64 0 0% 0 0 0
setivar (click) 167697 137724 8 38 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.