Open
Conversation
Contributor
Author
⏱️ Benchmark resultsClick to expand:
|
Contributor
|
We didn't target Clang performance much in the past. We didn't build and run benchmark for Clang until #5533. I intended #5533 as a preparation for #5591 (the 6-byte color case in mismatch that can be vectorized only for Clang due to MSVC lacking the ability to detect such case). Whereas #5590 was later than that, Clang numbers weren't obtained. Sure, the Clang numbers can be collected now. We can build with |
StephanTLavavej
approved these changes
Feb 6, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR adds a Neon path for the semi-vectorized implementation of
includes.Similar to #5590, it is a mixed bag of results. However, most significantly for this PR the new Neon path is slower than the Clang scalar codegen for 64-bit types. I am not sure if this was also the case for the x86 vectorization, as that PR looks as though it only reports MSVC numbers (?)
I guess we could avoid 64-bit vectorization - the tradeoff here would be improving Clang perf at the cost of MSVC perf.