Draft: Fix double shufti's false positive #316
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
Signed-off-by: Yoan Picchi <yoan.picchi@arm.com>
If we look for a pattern like "ab" and the letter 'a' fall at the end of the vector, then it was reporting a positive match, regardless of the second letter. This patch fix this false positive, but slows shufti down by 16%. Signed-off-by: Yoan Picchi <yoan.picchi@arm.com>
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Outdated suggestions cannot be applied.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.
When one have a search pattern with two characters (eg "ab"), and the first character is the last character of a vector, then it triggers a match, regardless of whether or not the second character appear in the next vector. (so only 'a' is require to match)
This patch fix it.
The bug have been reproduced on neon, sve and x86.
To experience the bug, just reverse the order of the pattern in the benchmark ("ba" instead of "ab" and you'll notice it returns almost instantaneously)
It does produce a 16% slowdown though. It might be faster to process the string 15 chars at a time instead of 16 and thus saving a load each time, but it requires more change than I can do just now.