[Mlir-commits] [mlir] [MLIR][XeGPU] Add transformation pattern for vector.broadcast in Wg to Sg pass (PR #144417)

[Nishant Patel]

================
@@ -170,6 +170,22 @@ gpu.func @dpas_no_sg_data(%a: memref<24x32xf32>, %b: memref<32x24xf32>) {
     gpu.return
   }
 
+  // CHECK-LABEL: broadcast
+  // CHECK-SAME: %[[ARG_0:.*]]: memref<24x1xf32>
+  gpu.func @broadcast(%src: memref<24x1xf32>) {
+    %tdesc = xegpu.create_nd_tdesc %src[0, 0] : memref<24x1xf32>
+      -> !xegpu.tensor_desc<24x1xf32, #xegpu.layout<sg_layout = [2, 1], sg_data = [12, 1], lane_layout = [2, 1], lane_data = [1, 1]>>
----------------
nbpatel wrote:

added the test case

https://github.com/llvm/llvm-project/pull/144417