[llvm] [AMDGPU] Implement IR expansion for frem instruction (PR #130988)

[Matt Arsenault via llvm-commits]

================
@@ -654,8 +1035,10 @@ static bool runImpl(Function &F, const TargetLowering &TLI) {
 
   while (!Replace.empty()) {
     Instruction *I = Replace.pop_back_val();
-    if (I->getOpcode() == Instruction::FPToUI ||
-        I->getOpcode() == Instruction::FPToSI) {
+    if (I->getOpcode() == Instruction::FRem)
+      expandFRem(llvm::cast<BinaryOperator>(*I));
----------------
arsenm wrote:

```suggestion
      expandFRem(cast<BinaryOperator>(*I));
```

https://github.com/llvm/llvm-project/pull/130988