This repository was archived by the owner on Jan 23, 2023. It is now read-only.
[release/3.1] Port dotnet/runtime#31904 fix for slow UTF-8 transcoding performance#28013
Merged
Anipik merged 1 commit intodotnet:release/3.1from Feb 18, 2020
Merged
Conversation
Remove BMI2 from ASCII and UTF-16 processing hot paths, as not all processors have optimized implementations of pext/pdep
Member
Author
|
Approved via email. |
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 subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Issue dotnet/runtime#2251 reports that on AMD processors, we severely regressed some UTF-8 transcoding code paths on AMD processors. These processors have unoptimized implementations of the
pdep/pextintrinsic. We didn't see this earlier because my benchmarking machine uses an Intel processor.This has already been fixed in .NET 5 (see dotnet/runtime#31904). This PR ports that fix down to release/3.1.
Customer Impact
Significant slowdowns in UTF-8 transcoding performance on 3.0 compared to 2.1 when using an AMD processor and where the input to transcode has mixed ASCII and non-ASCII data.
Regression?
Perf regression between 2.1 and 3.0. Not a functional regression.
Testing
Fix has been perf tested (see dotnet/runtime#31904 (comment)) and fuzzed (see dotnet/runtime#31904 (comment)).
Risk
Low. No functional changes; only performance changes. We have excellent unit test coverage and fuzzer coverage over this code.
Code Reviewer
@tannergooding