[llvm] [AArch64] Use isKnownNonZero to optimize eligible compares to cmn (PR #96349)

[via llvm-commits]

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 d2c817df846a4fa141778560e2258ab674426f44 369567fe89f6418c047789714a43424191b85e98 -- llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
``````````

</details>

<details>
<summary>
View the diff from clang-format here.
</summary>

``````````diff
diff --git a/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp b/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
index d96276c8b6..7c585ebdf4 100644
--- a/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
+++ b/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
@@ -3902,8 +3902,12 @@ static SDValue getAArch64Cmp(SDValue LHS, SDValue RHS, ISD::CondCode CC,
                                     !isLegalArithImmed(-RHS->getAsZExtVal()))) {
     SDValue TheLHS =
         isCMN(LHS, LHS.getOperand(1), CC, DAG) ? LHS.getOperand(1) : LHS;
-    SDValue TheRHS = !isa<ConstantSDNode>(RHS) && isCMN(RHS, RHS.getOperand(1), CC, DAG) ? RHS.getOperand(1) : RHS;
-    if (getCmpOperandFoldingProfit(TheLHS) > getCmpOperandFoldingProfit(TheRHS)) {
+    SDValue TheRHS =
+        !isa<ConstantSDNode>(RHS) && isCMN(RHS, RHS.getOperand(1), CC, DAG)
+            ? RHS.getOperand(1)
+            : RHS;
+    if (getCmpOperandFoldingProfit(TheLHS) >
+        getCmpOperandFoldingProfit(TheRHS)) {
       std::swap(LHS, RHS);
       CC = ISD::getSetCCSwappedOperands(CC);
     }

``````````

</details>


https://github.com/llvm/llvm-project/pull/96349