[GitHub] [tvm] tqchen commented on a change in pull request #7983: [PROFILING] Use PAPI to collect hardware performance counters on CPU and CUDA

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

tqchen commented on a change in pull request #7983:
URL: https://github.com/apache/tvm/pull/7983#discussion_r626971170



##########
File path: include/tvm/runtime/c_backend_api.h
##########
@@ -152,6 +152,16 @@ TVM_DLL int TVMBackendParallelBarrier(int task_id, TVMParallelGroupEnv* penv);
  */
 TVM_DLL int TVMBackendRunOnce(void** handle, int (*f)(void*), void* cdata, int nbytes);
 
+/*!
+ * \brief Reset the threads in the pool. All current threads are destroyed and
+ * new ones are created.
+ *
+ * Note that this does nothing when openmp is used.
+ *
+ * \return 0 when no error is thrown, -1 when failure happens
+ */

Review comment:
       Unless needed from the backend generated code, we should avoid adding new functions through C API, as C API needs to remain generally stable. Instead, consider register a packed function to do so




-- 
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