Legalize signed comparisons with constants to unsigned comparisons #472
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
mysterymath
reviewed
Aug 13, 2024
| ret i8 %conv2 | ||
| } | ||
|
|
||
| define zeroext i8 @sgt_16(i16 noundef %x) { |
There was a problem hiding this comment.
Is this still faster than the sequence before this change? If not, it seems better to add a TODO to the legalizer code and restrict the optimization to 8 bits for now. We may need some kind of additional combiner pass to clean up the output of the legalizer, but it's not something where I can guess the details offhand.
| auto RHSConst = getIConstantVRegValWithLookThrough(RHS, MRI); | ||
| bool RHSIsZero = RHSConst && RHSConst->Value.isZero(); | ||
|
|
||
| if (Pred == CmpInst::ICMP_SLT && RHSConst && !RHSIsZero) { |
There was a problem hiding this comment.
A short summary comment about why this is faster than the alternative would be helpful. It's difficult to read out of the implementation.
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Outdated suggestions cannot be applied.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.
took a while until i realized i could use a select and let the lower-select pass introduce the branches.
unfortunately there are some artifacts with the comparisons beyond 8 bits, each byte is compared twice, once for equality and once for Unsigned Greater Equal.
fixes #470