[Mlir-commits] [mlir] [mlir][Interfaces] DestinationStyleOpInterface: Improve documentation (PR #65927)

[llvmlistbot at llvm.org]

llvmbot wrote:

@llvm/pr-subscribers-mlir

<details>
<summary>Changes</summary>

Mention that sizes of dynamic dims of tied OpResults/operands match at runtime.
--
Full diff: https://github.com/llvm/llvm-project/pull/65927.diff

1 Files Affected:

- (modified) mlir/include/mlir/Interfaces/DestinationStyleOpInterface.td (+2-1) 


<pre>
diff --git a/mlir/include/mlir/Interfaces/DestinationStyleOpInterface.td b/mlir/include/mlir/Interfaces/DestinationStyleOpInterface.td
index ff2da985c53d7c4..8558e0279e3e5bb 100644
--- a/mlir/include/mlir/Interfaces/DestinationStyleOpInterface.td
+++ b/mlir/include/mlir/Interfaces/DestinationStyleOpInterface.td
@@ -29,7 +29,8 @@ def DestinationStyleOpInterface : OpInterface<"DestinationStyleOpInterface"> {
     ranked tensors and every tensor init is tied to a corresponding tensor
     OpResult in a 1-to-1 fashion. The i-th init tensor is tied to the i-th
     OpResult. The op may not have any additional OpResults. Init operands and
-    their tied OpResults have the same type.
+    their tied OpResults have the same type. Dynamic dimension sizes also match
+    at runtime.
 
     If the op has "buffer semantics", then the input operands are either ranked
     memrefs or other non-tensor/memref types ("scalar" types). Furthermore, the
</pre>

</details>

https://github.com/llvm/llvm-project/pull/65927