Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Perform single copy + pointer select instead of double write for smal…
…l types A previous optimization is no longer deemed worth it, new test results and understanding suggest that a single call to ptr::copy_nonoverlapping per element is more efficient for large input that don't fit into the last level cache. As well as for types like f128.
- Loading branch information