[llvm] [AMDGPU] Add encoding/decoding support for non-result-returning ATOMIC_CSUB instructions (PR #68197)

[Jay Foad via llvm-commits]

================
@@ -160,6 +160,7 @@ class GCNSubtarget final : public AMDGPUGenSubtargetInfo,
   bool HasAtomicFaddNoRtnInsts = false;
   bool HasAtomicBufferGlobalPkAddF16NoRtnInsts = false;
   bool HasAtomicBufferGlobalPkAddF16Insts = false;
+  bool HasAtomicCsubNoRtnInsts = false;
----------------
jayfoad wrote:

Nit: I'd prefer to capitalise it like "CSub" throughout to match the existing hasAtomicCSub method.

https://github.com/llvm/llvm-project/pull/68197