[Mlir-commits] [mlir] [mlir][linalg] Enable masked vectorisation for depthwise convolutions (PR #81625)

[Cullen Rhodes]

================
@@ -19,6 +19,25 @@ module attributes {transform.with_named_sequence} {
 
 // -----
 
+func.func @depthwise_conv1d_nwc_wc_dyn_w_dim(%input: memref<3x?x4xf32>, %filter: memref<?x4xf32>, %output: memref<3x?x4xf32>) {
+  // expected-error @+1 {{Attempted to vectorize, but failed}}
+  linalg.depthwise_conv_1d_nwc_wc
+    {dilations = dense<2> : tensor<1xi64>, strides = dense<1> : tensor<1xi64>}
----------------
c-rhodes wrote:

nit: simplify test by dropping these attributes?

https://github.com/llvm/llvm-project/pull/81625