Update reference to RegisterOperators in error message in Convolution

Signed-off-by: Edward Z. Yang <ezyang@fb.com>

ghstack-source-id: a99e00be9b56d6f8fd5555604236ebdd5195c4f7
Pull Request resolved: pytorch#36389

aten/src/ATen/native/Convolution.cpp

@@ -550,7 +550,7 @@ at::Tensor convolution_overrideable(
     const Tensor& input, const Tensor& weight, const Tensor& bias,
     IntArrayRef stride, IntArrayRef padding, IntArrayRef dilation,
     bool transposed, IntArrayRef output_padding, int64_t groups) {
-  AT_ERROR("You are likely triggering this with tensor backend other than CPU/CUDA/MKLDNN, if this is intended, please use torch::RegisterOperators() to override this function ");
+  AT_ERROR("You are likely triggering this with tensor backend other than CPU/CUDA/MKLDNN, if this is intended, please use TORCH_LIBRARY_IMPL to override this function ");
 }
 
 at::Tensor _convolution(
@@ -798,7 +798,7 @@ std::tuple<Tensor, Tensor, Tensor> convolution_backward_overrideable(
         const Tensor& grad_output, const Tensor& input, const Tensor& weight,
         IntArrayRef stride, IntArrayRef padding, IntArrayRef dilation,
         bool transposed, IntArrayRef output_padding, int64_t groups, std::array<bool, 3> output_mask) {
-  AT_ERROR("You are likely triggering this with tensor backend other than CPU/CUDA/MKLDNN, if this is intended, please use torch::RegisterOperators() to override this function ");
+  AT_ERROR("You are likely triggering this with tensor backend other than CPU/CUDA/MKLDNN, if this is intended, please use TORCH_LIBRARY_IMPL to override this function ");
   return std::tuple<Tensor, Tensor, Tensor>(
           at::empty_like(input, LEGACY_CONTIGUOUS_MEMORY_FORMAT),
           at::empty_like(weight, LEGACY_CONTIGUOUS_MEMORY_FORMAT),