[Mlir-commits] [mlir] [mlir][ROCDL] Add tensor load and store instructions to ROCDL (PR #165016)

[Ravil Dorozhinskii]

================
@@ -663,6 +663,70 @@ def ROCDL_GlobalLoadLDSOp :
   }];
 }
 
+//===---------------------------------------------------------------------===//
+// Tensor load/store intrinsics (available in GFX1250)
+//===---------------------------------------------------------------------===//
+
+// Base class for tensor load/store operations with 4 descriptor groups.
+class ROCDL_TensorLDSIntrOp<string mnemonic> :
+  ROCDL_IntrOp<mnemonic, [], [], [], 0, 0, 1, 0, [4], ["cachePolicy"]> {
+  dag args = (ins LLVM_VectorOf<I32>:$dgroup0, LLVM_VectorOf<I32>:$dgroup1,
+                  LLVM_VectorOf<I32>:$dgroup2, LLVM_VectorOf<I32>:$dgroup3,
----------------
ravil-mobile wrote:

Input types could be changed to `ROCDL_V4I32Type` and `ROCDL_V8I32Type`. You will need to add the following two definitions:

```
def ROCDL_V4I32Type : ROCDL_ConcreteVector<I32, 4>;
def ROCDL_V8I32Type : ROCDL_ConcreteVector<I32, 8>;
```

This approach constrains types of input operands better

https://github.com/llvm/llvm-project/pull/165016