[Mlir-commits] [mlir] [mlir][linalg] Fix padding shape computation in PadTilingInterface for convs (PR #149576)

[Fabian Mora]

================
@@ -63,6 +85,13 @@ getFullRankPaddingSizes(Builder &b, ArrayRef<OpFoldResult> indexingSizes,
 /// The `indexingMap` + `indexingSizes` encoding suits StructuredOps.
 /// The implementaiton below iteratively combines increases from contributing
 /// dimensions using affine.apply operations.
+/// The padded shape is computed by evaluating the maximum accessed index per
+/// dimension, which may involve multiplying by constant factors derived from
+/// the affine indexing expressions. Currently, only a limited set of projected
+/// permutation indexing map is supported, such as
----------------
fabianmcg wrote:

u-nit.

```suggestion
/// permutation indexing maps are supported, such as
```

https://github.com/llvm/llvm-project/pull/149576