[llvm] [AMDGPU] Eliminate redundant s_round_mode writes (PR #216305)

[Shilei Tian via llvm-commits]

================
@@ -514,6 +521,68 @@ bool SIPreEmitPeephole::removeExeczBranch(MachineInstr &MI,
   return true;
 }
 
+/// Remove writes to the FP round mode that can never be observed: either the
+/// value written is already live in MODE, or a later s_round_mode overwrites
+/// the whole round mode field before anything reads it.
+///
+/// This is a purely intra-block analysis: the round mode on entry to \p MBB is
+/// unknown, and a write that is still live at the end of the block is kept for
+/// the benefit of the successors.
+bool SIPreEmitPeephole::removeRedundantRoundMode(
+    MachineBasicBlock &SrcMBB) const {
+  MachineInstr *PendingWrite = nullptr;
+  std::optional<int64_t> CurrentRoundMode;
+  std::optional<int64_t> RoundModeBeforePendingWrite;
+  bool Changed = false;
+
+  for (MachineInstr &MI : make_early_inc_range(SrcMBB.instrs())) {
+    if (MI.isDebugInstr())
+      continue;
+
+    if (MI.getOpcode() == AMDGPU::S_ROUND_MODE) {
+      int64_t NewRoundMode = MI.getOperand(0).getImm();
+
+      if (PendingWrite && CurrentRoundMode != NewRoundMode) {
+        LLVM_DEBUG(dbgs() << "Removing dead round mode write: "
+                          << *PendingWrite);
+        PendingWrite->eraseFromParent();
+        ++NumRoundModeWritesRemoved;
+        Changed = true;
+        PendingWrite = nullptr;
+        CurrentRoundMode = RoundModeBeforePendingWrite;
+      }
+
+      if (CurrentRoundMode == NewRoundMode) {
+        LLVM_DEBUG(dbgs() << "Removing redundant round mode write: " << MI);
+        MI.eraseFromParent();
+        ++NumRoundModeWritesRemoved;
+        Changed = true;
+        continue;
+      }
+
+      RoundModeBeforePendingWrite = CurrentRoundMode;
+      PendingWrite = &MI;
+      CurrentRoundMode = NewRoundMode;
+      continue;
+    }
+
+    if (!PendingWrite && !CurrentRoundMode)
+      continue;
+
+    if (MI.isCall() || MI.isInlineAsm() ||
+        MI.modifiesRegister(AMDGPU::MODE, TRI)) {
+      PendingWrite = nullptr;
+      CurrentRoundMode.reset();
+      continue;
+    }
+
+    if (MI.readsRegister(AMDGPU::MODE, TRI) || MI.hasUnmodeledSideEffects()) {
+      PendingWrite = nullptr;
+    }
----------------
shiltian wrote:

```suggestion
    if (MI.readsRegister(AMDGPU::MODE, TRI) || MI.hasUnmodeledSideEffects())
      PendingWrite = nullptr;
```

https://github.com/llvm/llvm-project/pull/216305