Conversation
Eliminate all of the wasteful allocation, replace it with a 3-pass algorithm that inserts indices/values directly into their final positions. Signed-off-by: Andrew Duffy <andrew@a10y.dev>
4d3533b to
c9e1e79
Compare
Merging this PR will improve performance by 26.44%
Performance Changes
Comparing 
Footnotes
|
0ax1
reviewed
Mar 9, 2026
| } | ||
|
|
||
| // Loop over patches, writing them to final positions | ||
| let indices_out = indices_buffer.spare_capacity_mut(); |
Contributor
There was a problem hiding this comment.
Should we add a broader test coverage for transpose?
- with non-zero offset
- different value types, in particular 64bit as that changes the lane count to 16
- array len % 1024 != 0
0ax1
approved these changes
Mar 9, 2026
Contributor
There was a problem hiding this comment.
Looks great! Added one comment about adding more tests inline.
Signed-off-by: Andrew Duffy <andrew@a10y.dev>
Contributor
Author
|
I just pushed a test case generator, will merge once I verify it doesn't hurt CI times |
Signed-off-by: Andrew Duffy <andrew@a10y.dev>
Signed-off-by: Andrew Duffy <andrew@a10y.dev>
Signed-off-by: Andrew Duffy <andrew@a10y.dev>
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Eliminate all of the wasteful allocating and deallocating, replace it with a 3-pass algorithm that reuses the buffers.
Drops the CPU transpose time by a factor of 5-10 in my benchmarks on the other branch
Tested using existing test suite