YJIT Benchmarks

Details for Benchmarks at 2021-12-04 07:11:01

Overall YJIT is 32.1% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 29.4% 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 MJIT3.0 warmups MJIT3.0 iters MJIT warmups MJIT iters YJIT warmups YJIT iters
activerecord 5 187 20 187 75 187 20 187
hexapdf 5 15 20 15 20 15
liquid-render 5 174 20 174 75 174 20 174
mail 5 136 20 136 75 136 20 136
psych-load 5 15 20 15 75 15 20 15
railsbench 5 15 20 15 75 15 20 15
binarytrees 5 87 20 87 75 87 20 87
erubi 5 51 20 51 75 51 20 51
erubi_rails 5 596 20 596 75 596 20 596
fannkuchredux 5 15 20 15 53 15 20 15
jekyll 5 15 20 15 35 15 20 15
lee 5 26 20 26 75 26 20 26
nbody 5 343 20 343 75 343 20 343
optcarrot 5 15 20 15 75 15 20 15
rubykon 5 15 20 15 43 15 20 15
30k_ifelse 5 65 20 65 75 65 20 65
30k_methods 5 27 20 27 43 27 20 27
cfunc_itself 5 508 20 508 75 508 20 508
fib 5 383 20 383 75 383 20 383
getivar 5 562 20 562 75 562 20 562
keyword_args 5 432 20 432 75 432 20 432
respond_to 5 125 20 125 75 125 20 125
setivar 5 476 20 476 75 476 20 476

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) 990378 758722 185 1194 0 0% 0 0 0
hexapdf (click) 1096426 898327 596 7996 614 7% 0 0 321
liquid-render (click) 491205 390268 143 1432 95 6% 0 0 1
mail (click) 892522 667722 325 5429 124 2% 0 0 39
psych-load (click) 311585 247284 57 428 1 0% 0 0 0
railsbench (click) 2520098 1915643 1295 9251 263 2% 0 0 26
binarytrees (click) 164019 127726 9 49 0 0% 0 0 0
erubi (click) 290465 230150 8 35 0 0% 0 0 0
erubi_rails (click) 2451810 1775634 260 1995 17 0% 0 0 3
fannkuchredux (click) 179635 139520 6 203 0 0% 0 0 0
jekyll (click) 1879274 1520577 336 3791 185 4% 0 0 0
lee (click) 350122 277047 44 562 67 11% 0 0 12
nbody (click) 173427 135097 8 149 0 0% 0 0 0
optcarrot (click) 502259 449943 196 3567 20 0% 0 0 0
rubykon (click) 305139 249969 139 1529 2 0% 0 0 0
30k_ifelse (click) 5552627 4367761 9261 57795 0 0% 0 0 0
30k_methods (click) 2175091 1658897 5780 19398 0 0% 0 0 0
cfunc_itself (click) 162035 126006 6 40 0 0% 0 0 0
fib (click) 160499 125614 6 29 0 0% 0 0 0
getivar (click) 162227 128325 6 56 0 0% 0 0 0
keyword_args (click) 162867 126647 7 42 0 0% 0 0 0
respond_to (click) 164147 127460 6 55 0 0% 0 0 0
setivar (click) 162163 126649 6 29 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.