[llvm] [AMDGPU] Add new llvm.amdgcn.wave.shuffle intrinsic (PR #167372)

[via llvm-commits]

================
@@ -2718,6 +2718,15 @@ def int_amdgcn_call_whole_wave:
              llvm_vararg_ty], // The arguments to the callee.
             [IntrConvergent]>;
 
+// <result>
+//   llvm.amdgcn.subgroup.shuffle <value> <id>
+// value and result can be any scalar of floating-point, integer,
+// or Boolean types, but must be the same type
----------------
saxlungs wrote:

I added some more info about that and about what happens if index values are outside valid range

https://github.com/llvm/llvm-project/pull/167372