[CIR] Skip generation of a continue block when flattening TernaryOp #1651
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
We used to insert a continue Block at the end of a flattened ternary op that only contained a branch to the remaing operation of the remaining Block. This patch removes that continue block and changes the true/false blocks to directly jump to the remaining ops. With this patch the CIR now generates exactly the same LLVM IR as the original codegen.
mmha
commented
May 29, 2025
Comment on lines
896
to
902
| if(auto rt = op.getResultTypes(); rt.size()) { | ||
| auto args = remainingOpsBlock->addArguments(rt, op.getLoc()); | ||
| SmallVector<mlir::Value, 2> values; | ||
| llvm::copy(args, std::back_inserter(values)); | ||
| rewriter.replaceOpUsesWithinBlock(op, values, remainingOpsBlock); | ||
| } | ||
| rewriter.eraseOp(op); |
There was a problem hiding this comment.
All of this felt a bit... unergonomic. Is there no better way to replace an op and insert arguments into the block?
|
✅ With the latest revision this PR passed the C/C++ code formatter. |
andykaylor
approved these changes
May 29, 2025
bcardosolopes
changed the title
mmha
added a commit
to mmha/llvm-project
that referenced
this pull request
May 30, 2025
We used to insert a continue Block at the end of a flattened ternary op that only contained a branch to the remaing operation of the remaining Block. This patch removes that continue block and changes the true/false blocks to directly jump to the remaining ops. With this patch the CIR now generates exactly the same LLVM IR as the original codegen. This upstreams llvm/clangir#1651.
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Outdated suggestions cannot be applied.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.
We used to insert a continue Block at the end of a flattened ternary op that only contained a branch to the remaing operation of the remaining Block. This patch removes that continue block and changes the true/false blocks to directly jump to the remaining ops.
With this patch the CIR now generates exactly the same LLVM IR as the original codegen.