[Mlir-commits] [mlir] [MLIR][Linalg] Fix crash on duplicate dimensions in linalg.broadcast (PR #211203)

[Andrzej Warzyński]

================
@@ -478,6 +478,11 @@ def BroadcastOp : LinalgStructuredBase_Op<"broadcast", [
   let description = [{
     Broadcast the input into the given shape by adding `dimensions`.
 
+    Each index in `dimensions` attribute refers to a dimension of the init
+    tensor that is added by the operation. The indices must be unique and
+    within the init rank; the sizes of the remaining (non-added) init
+    dimensions must match the input shape.
----------------
banach-space wrote:

1. "init" could be either a tensor or a memref
2. init -> `init` (to make it clear that you are referring to `$init`)
3. input -> `input` (to make it clear that you are referring to `$input`)

https://github.com/llvm/llvm-project/pull/211203