YJIT Benchmarks

Details for Benchmarks at 2022-11-18 20:10:16

Overall YJIT is 31.6% faster than interpreted CRuby!
On Railsbench specifically, YJIT is 37.6% 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 YJIT warmups YJIT iters
activerecord 5 206 5 206
hexapdf 5 15 5 15
liquid-render 5 176 5 176
mail 5 122 5 122
psych-load 5 15 5 15
railsbench 5 15 5 15
ruby-lsp 5 242 5 242
binarytrees 5 80 5 80
chunky_png 5 32 5 32
erubi 5 68 5 68
erubi_rails 5 1123 5 1123
etanni 5 42 5 42
fannkuchredux 5 15 5 15
lee 5 21 5 21
nbody 5 233 5 233
optcarrot 5 15 5 15
rubykon 5 15 5 15
30k_ifelse 5 50 5 50
30k_methods 5 20 5 20
cfunc_itself 5 494 5 494
fib 5 327 5 327
getivar 5 427 5 427
keyword_args 5 386 5 386
respond_to 5 714 5 714
setivar 5 417 5 417
str_concat 5 483 5 483

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) 2303528 2302671 25 380 6 1% 0 0 0
hexapdf (click) 1774002 1772449 820 10303 219 2% 0 0 0
liquid-render (click) 589504 589460 156 1842 34 1% 0 0 0
mail (click) 1331087 1329752 373 5175 176 3% 0 0 0
psych-load (click) 800668 798812 70 638 19 2% 0 0 0
railsbench (click) 5610457 5609794 1779 13952 484 3% 0 0 0
ruby-lsp (click) 5425199 5423548 6425 40572 1521 3% 14254 0 0
binarytrees (click) 134623 133361 16 139 4 2% 0 0 0
chunky_png (click) 821472 820702 89 1183 31 2% 0 0 0
erubi (click) 746136 745725 13 147 4 2% 0 0 0
erubi_rails (click) 4729541 4728193 329 2686 111 4% 0 0 0
etanni (click) 127958 126418 13 107 4 3% 0 0 0
fannkuchredux (click) 146649 144465 13 269 4 1% 0 0 0
lee (click) 826601 824400 79 804 25 3% 0 0 0
nbody (click) 140895 140375 15 248 6 2% 0 0 0
optcarrot (click) 556602 554056 203 4133 93 2% 0 0 0
rubykon (click) 273760 272523 146 1659 32 1% 0 0 0
30k_ifelse (click) 5403089 5402946 9268 57196 4 0% 0 0 0
30k_methods (click) 2094588 2092547 5787 19425 4 0% 0 0 0
cfunc_itself (click) 125116 124327 13 122 4 3% 0 0 0
fib (click) 123656 123235 13 111 4 3% 0 0 0
getivar (click) 132937 132287 13 138 4 2% 0 0 0
keyword_args (click) 133344 132513 14 124 4 3% 0 0 0
respond_to (click) 133068 132355 13 138 5 3% 0 0 0
setivar (click) 133337 132513 13 111 4 3% 0 0 0
str_concat (click) 132771 132287 15 137 6 4% 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.


Far Too Many Details

These are links to HTML reports on GitHub with far, far too much detail about the benchmarking runs per-platform.


Autogenerated by continuous_reporting/generate_and_upload_reports.rb script.