[GitHub] [incubator-mxnet] apeforest commented on a change in pull request #18168: Separate GPU kernel for broadcast_axis

[GitBox]

apeforest commented on a change in pull request #18168:
URL: https://github.com/apache/incubator-mxnet/pull/18168#discussion_r417613184



##########
File path: src/operator/tensor/broadcast_reduce_op.h
##########
@@ -1077,6 +1077,42 @@ struct broadcast_kernel {
   }
 };
 
+namespace {
+struct shape_and_stride {
+  index_t in_stride[MXNET_SPECIAL_MAX_NDIM];
+  index_t out_stride[MXNET_SPECIAL_MAX_NDIM];
+  index_t input_shape[MXNET_SPECIAL_MAX_NDIM];
+  index_t output_shape[MXNET_SPECIAL_MAX_NDIM];
+};

Review comment:
       Does it make sense to define a struct with contains only four integers 
and pass in a vector of such struct as aux_data?




----------------------------------------------------------------
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.

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