Minor fixes

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

@@ -43,14 +43,13 @@ struct ComputePtrOffsetOfStridedBatch
         return ds_offset;
     }
 
-    __host__ __device__ constexpr long_index_t GetEPtrOffset(index_t g_idx) const
+    [[maybe_unused]] __host__ __device__ constexpr long_index_t GetEPtrOffset(index_t g_idx) const
     {
         return g_idx * static_cast<long_index_t>(BatchStrideE_);
     }
 
     // alias for kernels without multiple D
-    template <typename enable_if<NumDTensor == 0, bool>::type = false>
-    __host__ __device__ constexpr long_index_t GetCPtrOffset(index_t g_idx) const
+    [[maybe_unused]] __host__ __device__ constexpr long_index_t GetCPtrOffset(index_t g_idx) const
     {
         return g_idx * static_cast<long_index_t>(BatchStrideE_);
     }

test/grouped_convnd_bwd_weight/CMakeLists.txt

@@ -12,7 +12,7 @@ foreach(gpu IN LISTS GPU_TARGETS)
  endif()
  if(gpu IN_LIST gpu_list_wmma AND target EQUAL 0)
    add_gtest_executable(test_grouped_convnd_bwd_weight test_grouped_convnd_bwd_weight.cpp)
-   target_link_libraries(test_grouped_convnd_bwd_weight PRIVATE utility device_grouped_conv3d_bwd_weight_instance)
+   target_link_libraries(test_grouped_convnd_bwd_weight PRIVATE utility device_grouped_conv1d_bwd_weight_instance device_grouped_conv2d_bwd_weight_instance device_grouped_conv3d_bwd_weight_instance)
    add_gtest_executable(test_grouped_convnd_bwd_weight_interface test_grouped_convnd_bwd_weight_interface_wmma.cpp)
    target_link_libraries(test_grouped_convnd_bwd_weight_interface PRIVATE utility)
    set(target 1)