[GitHub] [incubator-mxnet] anko-intel commented on a change in pull request #20624: [Master] Decouple OneDNN data structures in MXNet C++ API

[GitBox <gi...@apache.org>]

anko-intel commented on a change in pull request #20624:
URL: https://github.com/apache/incubator-mxnet/pull/20624#discussion_r818801158



##########
File path: include/mxnet/ndarray.h
##########
@@ -766,36 +763,36 @@ class NDArray {
   /*
    * This function returns dnnl::memory with the default primitive_desc.
    */
-  const dnnl::memory* GetDNNLData() const;
+  const void* GetDNNLData() const;

Review comment:
       I agree with @PawelGlomski-Intel  that we can use _dnnl::memory*_ here and below:
   ```suggestion
     const dnnl::memory* GetDNNLData() const;
   ```
   with forward declaration near line 77
   ```
   namespace dnnl{
     struct memory;
   }
   ```
   This way we can avoid most of the static_cast in this change.
   
   

##########
File path: src/operator/nn/dnnl/dnnl_softmax-inl.h
##########
@@ -81,7 +81,6 @@ class DNNLSoftmaxFwd {
   std::shared_ptr<linear_t> temperature_fwd;
 };
 
-

Review comment:
       why do you remove this empty line ?




-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscribe@mxnet.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org