[Mlir-commits] [mlir] [mlir][SCF] Allow using a custom operation to generate loops with `mlir::tileUsingSCF`. (PR #159660)

[llvmlistbot at llvm.org]

================
@@ -33,6 +33,14 @@ using SCFTileSizeComputationFunction =
 
 /// Options to use to control tiling.
 struct SCFTilingOptions {
+  /// Specify which loop construct to use for tile and fuse.
+  enum class LoopType { ForOp, ForallOp, CustomOp };
+  LoopType loopType = LoopType::ForOp;
----------------
MaheshRavishankar wrote:

I agree... I think the main issue with `scf.forall` is the `num_threads` path is a bit wonky. I want to see if I can remove that option using this method. Then things will be much more "aligned" and I probably dont need to split it even further.

https://github.com/llvm/llvm-project/pull/159660