Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
For me, allowing LLVM to target native CPU instructions gives a nearly 2x speedup (and it's a single line change!). Before: ``` ------------------------------------------------------------------------------------------------- benchmark: 10 tests -------------------------------------------------------------------------------------------------- Name (time in us) Min Max Mean StdDev Median IQR Outliers OPS Rounds Iterations ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ test_rust_bytes_once 463.7750 (1.0) 2,320.4550 (1.0) 467.5096 (1.0) 44.6473 (1.0) 465.6775 (1.0) 1.9260 (1.0) 15;122 2,138.9934 (1.0) 1830 1 ``` After: ``` ----------------------------------------------------------------------------------------------- benchmark: 10 tests ------------------------------------------------------------------------------------------------ Name (time in us) Min Max Mean StdDev Median IQR Outliers OPS Rounds Iterations -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- test_rust_bytes_once 289.5610 (1.0) 2,190.4260 (1.04) 301.0328 (1.0) 38.1051 (1.0) 299.3540 (1.0) 1.8437 (1.30) 62;823 3,321.8976 (1.0) 2883 1 ```
- Loading branch information