[llvm] [AMDGPU] Per-chain MFMA->AGPR conversion (PR #217328)

[Romanov Vlad via llvm-commits]

================
@@ -1399,20 +1457,113 @@ bool RewriteMFMAFormStage::initGCNSchedStage() {
   TII = ST.getInstrInfo();
   SRI = ST.getRegisterInfo();
 
-  std::vector<std::pair<MachineInstr *, unsigned>> RewriteCands;
-  DenseMap<MachineBasicBlock *, std::set<Register>> CopyForUse;
-  SmallPtrSet<MachineInstr *, 8> CopyForDef;
+  // Collect all convertible MFMAs.
+  SmallVector<MachineInstr *, 32> AllCands;
+  for (MachineBasicBlock &MBB : MF)
----------------
romanovvlad wrote:

Added.

https://github.com/llvm/llvm-project/pull/217328