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

[Jianhui Li]

================
@@ -770,6 +787,33 @@ class LoadStoreToXeVMPattern : public OpConversionPattern<OpType> {
                                           basePtrI64);
     }
     Value mask = adaptor.getMask();
+
+    // Coalesce a chunked access into one contiguous block access.
+    //
+    // Distribution gives every element its own offset and mask bit. A lane
+    // whose `lane_data` is wider than one element therefore arrives here with
+    // `vector<D>` offsets and mask beside a `vector<D>` value, and those D
+    // offsets are one ascending run: `lane_data` is by definition the number of
+    // neighbouring elements the lane takes, and `inst_data` blocks the op so
----------------
Jianhui-Li wrote:

nit: Does these comments contribute any value? //, and `inst_data` blocks the op so
    // that no lane is given more. A single element converts to a scalar offset
    // instead, so a surviving vector offset always means a contiguous run.

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