[PATCH] D78464: [mlir][Linalg] Add support for fusing linalg.tensor_reshape with linalg.generic operations.

[Lei Zhang via Phabricator via llvm-commits]

antiagainst accepted this revision.
antiagainst added a comment.

Cool stuff!



================
Comment at: mlir/lib/Dialect/Linalg/Transforms/Fusion.cpp:580
 
+/// Linearize the expressions in `sourceMap` based on the `reassociationMaps`
+/// provided, given the shape of the source tensor that corresponds to the
----------------
Linearizes


================
Comment at: mlir/lib/Dialect/Linalg/Transforms/Fusion.cpp:592
+/// %1 = linalg.tensor_reshape %0 [affine_map<(i, j, k, l) -> (i)>,
+///                                  affine_map<(i, j, k, l) -> (j, k, l)>] :
+///     tensor<?x?x4x5xf32> into tensor<?x?xf32>
----------------
Neat: alignment :)


================
Comment at: mlir/lib/Dialect/Linalg/Transforms/Fusion.cpp:661
+      if (d.isa<AffineSymbolExpr>())
+        hasSymbol = true;
+    });
----------------
Can return WalkResult::interrupt() after this to avoid further walking.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78464/new/

https://reviews.llvm.org/D78464