[llvm] [AMDGPU] Introduce iglp_opt(2): Generalized exp/mfma interleaving for select kernels (PR #81342)

[Austin Kerbow via llvm-commits]

================
@@ -11,10 +11,17 @@
 
 #include "llvm/CodeGen/ScheduleDAGMutation.h"
 #include <memory>
+#include <vector>
 
 namespace llvm {
 
-std::unique_ptr<ScheduleDAGMutation> createIGroupLPDAGMutation(bool IsReentry);
+// Components of the mask that determines which instruction types may be may be
+// classified into a SchedGroup.
+enum class IGLPPhase { Initial = 0u, PreRAReentry = 1u << 0, PostRA = 1u << 1 };
----------------
kerbowa wrote:

What's the usecase for this being a mask?

https://github.com/llvm/llvm-project/pull/81342