perf: use uint MAX_INDEX
on arm
/ aarch64
for uint SIMD
#54
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Enhances the performance for
arm
/aarch64
uint operations by using uint asMAX_INDEX
instead of int.This results in 2x fewer exits from the inner SIMD loop (and thus 2x less - much slower - horizontal SIMD ops) 🎉
Illustration of the performance gain
main branch ⬇️
this PR ⬇️
We see here 30-40% better performance. This improvement becomes smaller when increasing in bit-size of the underlying uint datatype.