[clang] [llvm] [AMDGPU][WIP] Add support for i64/f64 readlane, writelane and readfirstlane operations. (PR #89217)

[Matt Arsenault via llvm-commits]

================
@@ -4822,6 +4822,111 @@ static MachineBasicBlock *lowerWaveReduce(MachineInstr &MI,
   return RetBB;
 }
 
+static MachineBasicBlock *lowerPseudoLaneOp(MachineInstr &MI,
----------------
arsenm wrote:

You should try to do this before selection. Doing it after just adds a lot of complexity and interferes with combine opportunities 

https://github.com/llvm/llvm-project/pull/89217