[CIR] Implement __builtin_ia32_cmpnleps/cmpnlepd #1893
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
badumbatish
requested review from
andykaylor,
bcardosolopes,
lanza and
xlauko
as code owners
Contributor
Author
|
|
tommymcm
reviewed
Sep 5, 2025
Collaborator
tommymcm
left a comment
tommymcm
left a comment
There was a problem hiding this comment.
There are some soundness issues here. I don't think making VecCmpOp ordered or unordered is the solution, it goes agains the semantics of C/C++.
- Implement createVecCompare, getCIRIntOrFloatBitWidth, getVectorFCmpIR helper for VecCmp op creation. - Add clang/test/CIR/CodeGen/builtin-fcmp-sse.c test.
badumbatish
force-pushed
the
cmpnlepd
branch
from
21aab62 to
a9d273e
Compare
…dered optimization
tommymcm
reviewed
Sep 10, 2025
Collaborator
tommymcm
left a comment
tommymcm
left a comment
There was a problem hiding this comment.
Here's some pointers for matching the skeleton.
tommymcm
reviewed
Sep 10, 2025
bcardosolopes
requested changes
Sep 12, 2025
bcardosolopes
approved these changes
Sep 18, 2025
Member
bcardosolopes
left a comment
bcardosolopes
left a comment
There was a problem hiding this comment.
LGTM with one minot nit.
thanks for your patience!
bcardosolopes
approved these changes
Sep 22, 2025
bcardosolopes
pushed a commit
that referenced
this pull request
Sep 25, 2025
This PR adds supports for __builtin_ia32_cmpnltps/cmpnltpd. Depends on #1893.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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Outdated suggestions cannot be applied.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.
Fixes #1818
VecCmp op creation.
in OG, there is a sext from bool to int before casting to float vector since fcmp's result in llvm ir is boolean-like, while VecCmpOp in CIR returns int in the form of 0 or -1.
There is also a boolean
shouldInvertin CIR since CIR doesn't contain optimized unordered comparison, for example: OLE is the inverse predicate of UGT. So if we need UGT, we have to pass in OLE andshouldInvert = true