[GitHub] [incubator-tvm] u99127 commented on a change in pull request #4274: [µTVM] Enable AutoTVM for ARM STM32F746XX Boards

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

u99127 commented on a change in pull request #4274: [µTVM] Enable AutoTVM for ARM STM32F746XX Boards
URL: https://github.com/apache/incubator-tvm/pull/4274#discussion_r349798137
 
 

 ##########
 File path: src/runtime/micro/micro_session.h
 ##########
 @@ -260,6 +279,125 @@ struct MicroDevSpace {
   ObjectPtr<MicroSession> session;
 };
 
+// TODO(weberlo): maybe templatize serialization to reduce redundancy
+
+/*! \brief TVM array for serialization to 32-bit devices */
+struct TVMArray32 {
+  TVMArray32(
+      TargetVal data,
+      DLContext ctx,
+      int32_t ndim,
+      DLDataType dtype,
+      TargetVal shape,
+      TargetVal strides,
+      TargetVal byte_offset)
+    : data(data.val32),
+      ctx(ctx),
+      ndim(ndim),
+      pad0(0),
+      dtype(dtype),
+      shape(shape.val32),
+      strides(strides.val32),
+      pad1(0),
+      byte_offset(byte_offset.val32),
+      pad2(0) { }
+
+  /*!
+   * \brief The opaque data pointer points to the allocated data.
+   *  This will be CUDA device pointer or cl_mem handle in OpenCL.
 
 Review comment:
   Is CUDA or Open_CL reference here a result of copy-pastism ? 
   
   

----------------------------------------------------------------
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:
users@infra.apache.org


With regards,
Apache Git Services