[llvm] [AggressiveInstCombine] Use APInt and avoid truncation when folding loads (PR #118525)
via llvm-commits
llvm-commits at lists.llvm.org
Tue Dec 3 10:52:54 PST 2024
github-actions[bot] wrote:
<!--LLVM CODE FORMAT COMMENT: {clang-format}-->
:warning: C/C++ code formatter, clang-format found issues in your code. :warning:
<details>
<summary>
You can test this locally with the following command:
</summary>
``````````bash
git-clang-format --diff fed3a9b8f81f5f4450e515f4499ecfda95804e95 232d5e95f58d92aede12917c9bcbe71cc22cba3f --extensions cpp -- llvm/lib/Transforms/AggressiveInstCombine/AggressiveInstCombine.cpp
``````````
</details>
<details>
<summary>
View the diff from clang-format here.
</summary>
``````````diff
diff --git a/llvm/lib/Transforms/AggressiveInstCombine/AggressiveInstCombine.cpp b/llvm/lib/Transforms/AggressiveInstCombine/AggressiveInstCombine.cpp
index 96e768f3c7..45ee2d472a 100644
--- a/llvm/lib/Transforms/AggressiveInstCombine/AggressiveInstCombine.cpp
+++ b/llvm/lib/Transforms/AggressiveInstCombine/AggressiveInstCombine.cpp
@@ -803,8 +803,7 @@ static bool foldConsecutiveLoads(Instruction &I, const DataLayout &DL,
APInt Offset1(DL.getIndexTypeSizeInBits(Load1Ptr->getType()), 0);
Load1Ptr = Load1Ptr->stripAndAccumulateConstantOffsets(
DL, Offset1, /* AllowNonInbounds */ true);
- Load1Ptr = Builder.CreatePtrAdd(Load1Ptr,
- Builder.getInt(Offset1));
+ Load1Ptr = Builder.CreatePtrAdd(Load1Ptr, Builder.getInt(Offset1));
}
// Generate wider load.
NewLoad = Builder.CreateAlignedLoad(WiderType, Load1Ptr, LI1->getAlign(),
``````````
</details>
https://github.com/llvm/llvm-project/pull/118525
More information about the llvm-commits
mailing list