perf triage for 2021-07-20. #930
Merged
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
rylev
approved these changes
Jul 21, 2021
| Replace associated item bound vars with placeholders when projecting [#86993](https://github.com/rust-lang/rust/issues/86993) | ||
| - Moderate regression in [instruction counts](https://perf.rust-lang.org/compare.html?start=b1f8e27b74c541d3d555149c8efa4bfe9385cd56&end=27e42058811e448b1a7dd8630d86ab247fbfcb9b&stat=instructions:u) (up to 1.5% on `incr-patched: b9b3e592dd cherry picked` builds of `style-servo-debug`) | ||
| - A perf run was [done on the PR originally](https://perf.rust-lang.org/compare.html?start=fdfe819580062a441024d713b49340cd3f7d7efc&end=7baa78ec683fd14db4d4c1869dbef5cbbc5b774d). The regression flagged here seems to be on a different set of benchmarks: `style-servo-debug` and `wf-projection-stress-65510-*`, neither of which was flagged as a regression in the original runs. | ||
| - Results overall seem pretty mixed and hard to interpret. Left a [comment](https://github.com/rust-lang/rust/pull/86993#issuecomment-883624762) on the PR. |
There was a problem hiding this comment.
I'm not sure why the triage report labeled this as a regression when the results seem so mixed... I will investigate this.
|
|
||
| Remove special case for `ExprKind::Paren` in `MutVisitor` [#87284](https://github.com/rust-lang/rust/issues/87284) | ||
| - Moderate regression in [instruction counts](https://perf.rust-lang.org/compare.html?start=014026d1a7ca991f82f12efa95ef4dffb29dc8af&end=6535449a002264ee08dec8e741f1aadd97428fae&stat=instructions:u) (up to 1.0% on `full` builds of `coercions-debug`) | ||
| - This is on the borderline for "significant regression", and was the *only* regression associated with this PR. |
There was a problem hiding this comment.
Eventually I'd like to make the logic for labeling runs as regressions to be a bit smarter in these cases. We should be able to encode that if there is only a single noisy benchmark which regressed and that regression is barely significant, we shouldn't count it as a regression.
…t already got tags added to them.
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.
No description provided.