[llvm] [AMDGPU] Track LDS DMA VGPR source reads in expert scheduling (PR #203292)

[Austin Kerbow via llvm-commits]

================
@@ -2858,7 +2858,7 @@ bool SIInsertWaitcnts::generateWaitcnt(AMDGPU::Waitcnt Wait,
 
 std::optional<WaitEventType>
 SIInsertWaitcnts::getExpertSchedulingEventType(const MachineInstr &Inst) const {
-  if (TII.isVALU(Inst)) {
----------------
kerbowa wrote:

I suggest adding a TODO about DMA being classified as VALU. We need to look into avoiding this category of issue.

https://github.com/llvm/llvm-project/pull/203292