FrozenGene commented on a change in pull request #4747: [ThreadPool] Solve ARM
BIG.LITTLE heterogeneous multicores
URL: https://github.com/apache/incubator-tvm/pull/4747#discussion_r368745458
##########
File path: src/runtime/threading_backend.cc
##########
@@ -148,19 +146,36 @@ class ThreadGroup::Impl {
CPU_SET(sorted_order_[0], &cpuset);
}
pthread_setaffinity_np(pthread_self(), sizeof(cpu_set_t), &cpuset);
+ } else {
Review comment:
Hmm...How about call SetFullCpuAffinity(reverse) directly like Android? I
think make master thread could have free migration on needed cores is better
than bind just one specific core. Thanks.
----------------------------------------------------------------
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:
[email protected]
With regards,
Apache Git Services
