[llvm-branch-commits] [llvm] [WIP] AMDGPU: Handle v_add* in eliminateFrameIndex (PR #102346)

[via llvm-branch-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 b9a85cef41e26f6f468c87a00f2072ea798ec628 af30ca7d438747ce58dd79ef293a9914ceef592f --extensions cpp -- llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
``````````

</details>

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

``````````diff
diff --git a/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp b/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
index 0427b021d4..505802c679 100644
--- a/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
@@ -2534,15 +2534,11 @@ bool SIRegisterInfo::eliminateFrameIndex(MachineBasicBlock::iterator MI,
         if (isVGPRClass(getPhysRegBaseClass(MaterializedReg))) {
           // If we know we have a VGPR already, it's more likely the other
           // operand is a legal vsrc0.
-          AddI32
-            .add(*OtherOp)
-            .addReg(MaterializedReg, MaterializedRegFlags);
+          AddI32.add(*OtherOp).addReg(MaterializedReg, MaterializedRegFlags);
         } else {
           // Commute operands to avoid violating VOP2 restrictions. This will
           // typically happen when using scratch.
-          AddI32
-            .addReg(MaterializedReg, MaterializedRegFlags)
-            .add(*OtherOp);
+          AddI32.addReg(MaterializedReg, MaterializedRegFlags).add(*OtherOp);
         }
 
         if (MI->getOpcode() == AMDGPU::V_ADD_CO_U32_e64 ||

``````````

</details>


https://github.com/llvm/llvm-project/pull/102346