[llvm] [AMDGPU] Define constrained multi-dword scalar load instructions. (PR #96161)

[Jay Foad via llvm-commits]

================
@@ -167,6 +167,20 @@ multiclass SM_Pseudo_Loads<RegisterClass baseClass,
   def _IMM : SM_Load_Pseudo <opName, baseClass, dstClass, IMM_Offset>;
   def _SGPR : SM_Load_Pseudo <opName, baseClass, dstClass, SGPR_Offset>;
   def _SGPR_IMM : SM_Load_Pseudo <opName, baseClass, dstClass, SGPR_IMM_Offset>;
+
+  // The constrained multi-dword load equivalents with early clobber flag at
+  // the dst operand. They are needed only for codegen and there is no need for
+  // their real opcodes.
+  let SubtargetPredicate = isGFX8Plus,
+      Constraints = !if(!gt(dstClass.RegTypes[0].Size, 32),
+                         "@earlyclobber $sdst", "") in {
+    let PseudoInstr = NAME # !cast<OffsetMode>(IMM_Offset).Variant in
+      def _IMM_ec : SM_Load_Pseudo <opName, baseClass, dstClass, IMM_Offset>;
+    let PseudoInstr = NAME # !cast<OffsetMode>(SGPR_Offset).Variant in
+      def _SGPR_ec : SM_Load_Pseudo <opName, baseClass, dstClass, SGPR_Offset>;
+    let PseudoInstr = NAME # !cast<OffsetMode>(SGPR_IMM_Offset).Variant in
+      def _SGPR_IMM_ec : SM_Load_Pseudo <opName, baseClass, dstClass, SGPR_IMM_Offset>;
+  }
----------------
jayfoad wrote:

Can you surround this with `if !gt(dstClass.RegTypes[0].Size, 32)` so that the `_ec` forms are not even defined for dword and smaller?

https://github.com/llvm/llvm-project/pull/96161