[llvm] [AMDGPU] Clear no convergence flag on onperand folding. NFCI (PR #179438)

[via llvm-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-backend-amdgpu

Author: Stanislav Mekhanoshin (rampitec)

<details>
<summary>Changes</summary>

Clear the flag. It fails verification if set, only convergent
operations may have NoConvergent flag. NFCI as it is now because
it just does not happen.


---
Full diff: https://github.com/llvm/llvm-project/pull/179438.diff


1 Files Affected:

- (modified) llvm/lib/Target/AMDGPU/SIFoldOperands.cpp (+2) 


``````````diff
diff --git a/llvm/lib/Target/AMDGPU/SIFoldOperands.cpp b/llvm/lib/Target/AMDGPU/SIFoldOperands.cpp
index 3f72b65e8148a..a2fe31bd849c3 100644
--- a/llvm/lib/Target/AMDGPU/SIFoldOperands.cpp
+++ b/llvm/lib/Target/AMDGPU/SIFoldOperands.cpp
@@ -1445,6 +1445,7 @@ void SIFoldOperandsImpl::foldOperand(
           return;
 
         UseMI->setDesc(TII->get(AMDGPU::S_MOV_B32));
+        UseMI->clearFlag(MachineInstr::NoConvergent);
 
         if (OpToFold.isImm()) {
           UseMI->getOperand(1).ChangeToImmediate(
@@ -1476,6 +1477,7 @@ void SIFoldOperandsImpl::foldOperand(
         UseMI->getOperand(1).setSubReg(OpToFold.getSubReg());
         UseMI->getOperand(1).setIsKill(false);
         UseMI->removeOperand(2); // Remove exec read (or src1 for readlane)
+        UseMI->clearFlag(MachineInstr::NoConvergent);
         return;
       }
     }

``````````

</details>


https://github.com/llvm/llvm-project/pull/179438