[llvm] [LLVM][NVPTX] Add movmatrix intrinsic and PTX instruction support (PR #190109)

[Varad Rahul Kamthe via llvm-commits]

================
@@ -2802,6 +2802,11 @@ foreach transposed = [0, 1] in {
   }
 }
 
+// movmatrix
+def int_nvvm_movmatrix_sync_aligned_m8n8_trans_b16
+  : DefaultAttrsIntrinsic<[llvm_i32_ty], [llvm_i32_ty],
----------------
varadk27 wrote:

Switched to `Intrinsic `to avoid the implicit `WillReturn`.  Kept `IntrNoCallback `and `IntrNoFree`, but dropped `IntrNoSync `since movmatrix with .sync does synchronize threads, similar to shuffle intrinsics.

https://github.com/llvm/llvm-project/pull/190109