Comparing
DaniPopes:dani/perf-optimizations
(75444af
) withmain
(c147f59
)-14%
- Improvement
- 81
- Regression
- 1
- Untouched
- 304
Benchmarks
386 total
root/2/256/256
benches/bench.rs::benches::benches::group+13%
11.3 ms10 ms
wrapping_shl/128
benches/bench.rs::benches::benches::group+13%
281.9 µs248.8 µs
root/3/192/192
benches/bench.rs::benches::benches::group+11%
8.4 ms7.6 ms
to/f32/4096
benches/bench.rs::benches::benches::group+11%
29.2 µs26.3 µs
to/f64/4096
benches/bench.rs::benches::benches::group+11%
29.4 µs26.4 µs
most_significant_bits/4096/4096
benches/bench.rs::benches::benches::group+11%
28.9 µs26.1 µs
root/3/256/256
benches/bench.rs::benches::benches::group+10%
11.4 ms10.3 ms
fmt/octal/384
benches/bench.rs::benches::benches::group+9%
12.2 ms11.2 ms
overflowing_shl/128
benches/bench.rs::benches::benches::group+9%
346.1 µs318.5 µs
root/5/256/256
benches/bench.rs::benches::benches::group+8%
11.2 ms10.4 ms
root/5/192/192
benches/bench.rs::benches::benches::group+8%
8 ms7.4 ms
div_rem/192/64
benches/bench.rs::benches::benches::group+7%
1.2 ms1.1 ms
div_rem/128/64
benches/bench.rs::benches::benches::group+7%
819.1 µs763.7 µs
reduce_mod/192
benches/bench.rs::benches::benches::group+7%
1.1 ms1 ms
div_rem/128/128
benches/bench.rs::benches::benches::group+7%
821.4 µs768.5 µs
div_rem/256/64
benches/bench.rs::benches::benches::group+6%
1.4 ms1.4 ms
reduce_mod/256
benches/bench.rs::benches::benches::group+6%
1.3 ms1.2 ms
div_rem/192/96
benches/bench.rs::benches::benches::group+6%
1.4 ms1.3 ms
div_rem/256/128
benches/bench.rs::benches::benches::group+4%
1.6 ms1.5 ms
div_rem/256/256
benches/bench.rs::benches::benches::group+4%
1.6 ms1.5 ms
neg/4096
benches/bench.rs::benches::benches::group+4%
72.9 µs70 µs
root/127/256/256
benches/bench.rs::benches::benches::group+4%
13.7 ms13.1 ms
root/127/192/192
benches/bench.rs::benches::benches::group+4%
18.6 ms18 ms
div_rem/192/192
benches/bench.rs::benches::benches::group+4%
1.3 ms1.3 ms
sub/4096
benches/bench.rs::benches::benches::group+3%
96.1 µs93.1 µs
Commits
Click on a commit to change the comparison rangeBase
main
c147f59
-29.84%
perf: specialize hot paths for common sizes
52a8393
3 days ago
by DaniPopes+15.71%
perf: revert cmp U256 specialization, use push_byte in fmt
75444af
3 days ago
by DaniPopes