[Mlir-commits] [mlir] [MLIR][XeGPU] Extend propagation and sg_to_lane distribution pass support broadcast with low rank and scalar source input (PR #170409)

[Charitha Saumya]

================
@@ -511,6 +518,20 @@ bool SliceAttr::isSliceOf(const xegpu::DistributeLayoutAttr &other) {
                       [&](int64_t dim) { return thisDims.contains(dim); });
 }
 
+bool SliceAttr::isIdentical(const xegpu::DistributeLayoutAttr &other) {
+  if (dyn_cast<xegpu::LayoutAttr>(other))
+    return false;
+
+  auto flattenedThis = flatten();
+  auto flattenedOther = dyn_cast<xegpu::SliceAttr>(other).flatten();
+
+  if ((flattenedThis.getParent() == flattenedOther.getParent()) &&
+      (flattenedThis.getDims() == flattenedOther.getDims())) {
+    return true;
+  }
+  return false;
----------------
charithaintc wrote:


```suggestion
  return (flattenedThis.getParent() == flattenedOther.getParent()) &&
      (flattenedThis.getDims() == flattenedOther.getDims());
```

https://github.com/llvm/llvm-project/pull/170409