[llvm] [AMDGPU] Allocate i1 argument to SGPRs (PR #72461)

[Matt Arsenault via llvm-commits]

================
@@ -15887,7 +15889,14 @@ static bool isCopyFromRegForI1Return(const SDNode *N) {
   SDNode *N2 = N1->getOperand(0).getNode();
   if (N2->getOpcode() != ISD::CopyFromReg)
     return false;
-  SDNode *N3 = N2->getOperand(0).getNode();
+
----------------
arsenm wrote:

Yes, this whole function should be unnecessary. isArgPassedInSGPR needs to be updated, and the use for uniformity needs to be updated to account for this non-uniform, but also in SGPR case 

https://github.com/llvm/llvm-project/pull/72461