[Mlir-commits] [mlir] [mlir][xegpu] Remove chunk_size attribute; infer from types (PR #205122)

[Jianhui Li]

================
@@ -837,14 +831,9 @@ def XeGPU_StoreScatterOp : XeGPU_Op<"store", [MemoryEffects<[MemWrite]>, AnchorL
   let summary = "store data to scattered memory locations.";
   let description =
       [{ It (aka. store) stores data to scattered memory locations. The value is
-  typically a 1D vector. But when the chunk size of the TensorDesc is larger than 1, it will be
-  a 2D vector instead. For the later case, dim-1 of the value correspods to the simd lanes
-  and the dim-0 of the value corresponds to the chunk size stored per lane. So `store_scatter`
-  has transpose effect, which is similar to `load_gather`. Therefore, a transpose attribute is
-  introduced on purpose, making sure users are aware of this implicit transformation.
-
-  In lane level, the result is a 1D vector that represents the data to be stored by
-  each lane. If size is not 1, size should be equal to the chunk size.
+  typically a 1D vector. At lane level each lane may store a 1D vector of
----------------
Jianhui-Li wrote:

The original comment is a bit outdated.
please describe the op not only from lane level. See comments above for the load op. 


https://github.com/llvm/llvm-project/pull/205122