Cache both head and tail index in both Consumer and Producer (again) #132
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
|
According to #39 (comment), this indeed significantly improved the performance on an AMD CPU! I think I will add a comment to the source code (so that the caching isn't removed in the future again), but other than that, it's ready to merge. |
951d3d6 to
5d6e900
Compare
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Outdated suggestions cannot be applied.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 reverts PR #48.
This may or may not be faster on AMD CPUs.
Performance comparisons in #39 (comment) and #39 (comment) show that
rtrbis significantly slower (in the contended case) than the crossbeam PR it is based on (crossbeam-rs/crossbeam#338).There shouldn't be too many differences between the two code bases, but one of them is #48, so maybe that's the cause for the regression?
It might also be unrelated, though. This PR should only be merged if it shows a clear improvement on AMD CPUs. Ideally, it should have the same performance as (or even better than) crossbeam PR 338.
If this PR doesn't show any improvements on AMD, we'll have to keep looking ...