#7625: Enable multicore for tilize with padding by default

tenstorrent · May 15, 2024 · b658ebb · b658ebb
1 parent a7636ab
commit b658ebb
Showing 1 changed file with 3 additions and 2 deletions.
diff --git a/ttnn/cpp/ttnn/op_library/to_layout/to_layout_op.cpp b/ttnn/cpp/ttnn/op_library/to_layout/to_layout_op.cpp
@@ -124,13 +124,14 @@ Tensor execute(
 
         } else if (layout == ttnn::TILE_LAYOUT) {
             tensor = unsqueeze_to_4D(tensor);
+            bool use_multicore = true;
             std::vector<uint32_t> padded_4D_output_shape;
             padded_4D_output_shape.push_back(tensor.get_shape()[-4]);
             padded_4D_output_shape.push_back(tensor.get_shape()[-3]);
             padded_4D_output_shape.push_back(ttnn::pad_to_multiple_of_tile_size(tensor.get_shape()[-2]));
             padded_4D_output_shape.push_back(ttnn::pad_to_multiple_of_tile_size(tensor.get_shape()[-1]));
-            tensor =
-                tt::tt_metal::tilize_with_val_padding(tensor, padded_4D_output_shape, 0, output_memory_config, dtype);
+            tensor = tt::tt_metal::tilize_with_val_padding(
+                tensor, padded_4D_output_shape, 0, output_memory_config, dtype, use_multicore);
             return reshape(tensor, ttnn::Shape(tt::tt_metal::Shape{output_shape, padded_output_shape}));
 
         } else {