[Mlir-commits] [mlir] [OpenMP][MLIR] Add omp.canonical_loop operation, !omp.cli type, omp.new_cli operation (PR #71712)

[Kiran Chandramohan]

kiranchandramohan wrote:

The only concern I have is that the dependence between the loop transformation operation and the canonical loop is indirectly expressed through the CLI. This has a possible issue that the MLIR transformation passes might think that there is no direct dependence between the canonical loop and the transformation operation and could cause some issues.

```
    %outer = omp.new_cli : !omp.cli
    %inner = omp.new_cli : !omp.cli
    omp.canonical_loop %iv1 : i32 in [0, %tripcount), %outer : !omp.cli{
      omp.canonical_loop %iv2 : i32 in [0, %tc), %inner : !omp.cli {
        %a = load %arrA[%iv1, %iv2] : memref<?x?xf32>
        store %a, %arrB[%iv1, %iv2] : memref<?x?xf32>
      }
    }
    omp.tile(%outer, %inner : !omp.cli, !omp.cli)
 ```

https://github.com/llvm/llvm-project/pull/71712