[llvm] [AMDGPU] Insert before and after instructions that always use GDS (PR #131338)

[Stephen Thomas via llvm-commits]

================
@@ -1982,6 +2003,65 @@ static bool isCacheInvOrWBInst(MachineInstr &Inst) {
          Opc == AMDGPU::GLOBAL_WBINV;
 }
 
+// Return true if the next instruction is S_ENDPGM, following fallthrough
+// blocks if necessary.
+bool SIInsertWaitcnts::isNextENDPGM(MachineBasicBlock::instr_iterator It,
+                                    MachineBasicBlock *Block) const {
+  auto E = Block->instr_end();
+
+  while (true) {
+    if (It == E) {
----------------
stepthomas wrote:

It looks like `It..sEnd()` does exist, `It->getParent()` fails if `It` is at the end of the iterator, so it's not possible get the next block without knowing the current block already.

https://github.com/llvm/llvm-project/pull/131338