Improve IndexOfAnyAsciiSearcher ARM throughput #78739
Merged
Conversation
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
Also solves an edge-case bug for byte overloads on ARM64 where we'd get false positives (negatives for Except) when haystack values were exactly 128 above a value in the needle
ghost
assigned
|
Tagging subscribers to this area: @dotnet/area-system-memory Issue DetailsAlso solves an edge-case bug for byte overloads on ARM64 where we'd get false positives (negatives for Except) when haystack values were exactly 128 above a value in the needle. Before
After
|
MihaZupan
commented
Nov 23, 2022
| haystackWithOffsetNeedle[i] = (byte)(128 + needleValues[i % needleValues.Length]); | ||
| } | ||
|
|
||
| Assert.Equal(-1, haystackWithOffsetNeedle.IndexOfAny(needle)); |
Member
Author
There was a problem hiding this comment.
The test would blow up here on ARM before this change
stephentoub
reviewed
Nov 23, 2022
| Vector128<byte> ascii = Sse2.IsSupported | ||
| ? Sse2.PackSignedSaturate(ascii0, ascii1).AsByte() | ||
| : AdvSimd.Arm64.UnzipEven(ascii0.AsByte(), ascii1.AsByte()); | ||
| Vector128<byte> ascii = Sse2.PackSignedSaturate(ascii0, ascii1).AsByte(); |
Member
There was a problem hiding this comment.
Nit: does the ascii local temporary help with anything?
Member
Author
There was a problem hiding this comment.
I don't think so, I just found it a bit more readable when initially writing this logic
stephentoub
approved these changes
Nov 23, 2022
Member
Author
|
/azp run runtime-libraries-coreclr outerloop |
|
Azure Pipelines successfully started running 1 pipeline(s). |
EgorBo
approved these changes
Nov 23, 2022
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
4 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.
Also solves an edge-case bug for byte overloads on ARM64 where we'd get false positives (negatives for Except) when haystack values were exactly 128 above a value in the needle.
Closes #78718
Before
After