clean

include/ck/tensor_operation/gpu/device/impl/device_grouped_gemm_xdl.hpp

@@ -617,7 +617,7 @@ struct DeviceGroupedGemm_Xdl : public DeviceGroupedGemm<ALayout,
 
         // If we use padding we do not support vector loads for dimensions not divisible by vector
         // load size.
-        // if constexpr(GemmSpec != GemmSpecialization::Default)
+        if constexpr(GemmSpec != GemmSpecialization::Default)
         {
             // [A|B]BlockTransferSrcVectorDim value define dimension in the block {K0,M,K1} layout,
             // thus we have to adapt it to the {M,K} or {N,K} layout.
@@ -630,12 +630,10 @@ struct DeviceGroupedGemm_Xdl : public DeviceGroupedGemm<ALayout,
                 const auto b_vector_dim = arg.b_mtx_nraw_kraw_[i].At(Number<b_raw_vector_dim>{});
 
                 if(a_vector_dim % ABlockTransferSrcScalarPerVector != 0)
-                    ;
-                return false;
+                    return false;
 
                 if(b_vector_dim % BBlockTransferSrcScalarPerVector != 0)
-                    ;
-                return false;
+                    return false;
             }
         }