[Mlir-commits] [mlir] [MLIR] [XeGPU] Add distribution pattern for vector.constant_mask from Wg To Sg (PR #168118)

[Charitha Saumya]

================
@@ -130,5 +130,14 @@ gpu.module @test_distribution {
     %trans = vector.transpose %load, [1, 0] {layout_result_0 = #xegpu.layout<sg_layout = [4, 8], sg_data = [16, 32], lane_layout = [1, 16], lane_data = [1, 1], order =[1, 0]>} : vector<256x128xf32> to vector<128x256xf32>
       gpu.return
   }
+
+  // CHECK-LABEL: vector_mask_2D
+  gpu.func @vector_mask_2D() {
+    %cst16 = arith.constant 16 : index
----------------
charithaintc wrote:

in all the test this constant value is unused. why not remove it? 

https://github.com/llvm/llvm-project/pull/168118