[NFC] Improve DeduplicateExecutables bucketing #19601
Open
Conversation
This file contains 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
Use the result types as a part of the hash when grouping ops. This vastly improves the performance of this pass when there are several similar objects that have different types. This pass can become very slow because checking structural equivalence is a costly procedure pass is O(n^2) with respect to the bucket size. Signed-off-by: Ian Wood <[email protected]>
hanhanW
approved these changes
Jan 6, 2025
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Outdated suggestions cannot be applied.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 change uses the result types as a part of the hash when grouping ops. This vastly improves the performance of this pass when there are several similar objects that consist of ops with the same names but differ in the number/type of results. However, this may increase the overhead of hashing when bucketing isn't effective.
Although a sample size of one, I found that for 405b tp8, the number of buckets went from 35 -> 140. This brought the time of this pass from a few minutes to several seconds.