huajsj commented on code in PR #11599:
URL: https://github.com/apache/tvm/pull/11599#discussion_r902997516


##########
src/runtime/threading_backend.cc:
##########
@@ -101,13 +109,19 @@ class QuRTThread {
 };
 #endif  // __hexagon__
 thread_local int max_concurrency = 0;
+
 class ThreadGroup::Impl {
  public:
   Impl(int num_workers, std::function<void(int)> worker_callback, bool 
exclude_worker0)
       : num_workers_(num_workers) {
     ICHECK_GE(num_workers, 1) << "Requested a non-positive number of worker 
threads.";
     for (int i = exclude_worker0; i < num_workers_; ++i) {
-      threads_.emplace_back([worker_callback, i] { worker_callback(i); });
+      threads_.emplace_back([worker_callback, i, this] {
+#ifndef __hexagon__
+        SetTid();

Review Comment:
   SetTid running in a worker thread, that means the execution order is random, 
but currently logic assume it follow a order to execution and this would cause 
logic issue.
   
   we may need to figure out  if there is any better and light weight solution.



-- 
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-unsubscr...@tvm.apache.org

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

Reply via email to