[llvm] [AMDGPU] Fix operand definitions for atomic scalar memory instructions. (PR #71799)

[Jay Foad via llvm-commits]

================
@@ -909,6 +909,10 @@ class AMDGPUOperand : public MCParsedAsmOperand {
   bool isWaitVDST() const;
   bool isWaitEXP() const;
 
+  auto getPredicate(std::function<bool(const AMDGPUOperand &Op)> P) const {
----------------
jayfoad wrote:

OK. In the long term do you think we should aim for a deisgn where all operand predicate strings in .td files are expressions involving `Op` (like `Op.isCPol()`) instead of just names of predicate methods (like `isCPol`)?

https://github.com/llvm/llvm-project/pull/71799