[llvm] [AMDGPU] Do not allow M0 as v_readfirstlane_b32 dst (PR #128851)

[Matt Arsenault via llvm-commits]

================
@@ -243,7 +243,7 @@ defm V_MOV_B64 : VOP1Inst <"v_mov_b64", VOP_I64_I64>;
 } // End isMoveImm = 1
 
 def VOP_READFIRSTLANE : VOPProfile <[i32, i32, untyped, untyped]> {
-  let DstRC = RegisterOperand<SReg_32>;
+  let DstRC = RegisterOperand<SReg_32_XM0>;
----------------
arsenm wrote:

Not writelane 

https://github.com/llvm/llvm-project/pull/128851