Actions
Bug #3162
closedsleep() is not a pthread cancel point?
Description
OS Version: 5.4
When running sleep(5) in one thread, it cannot be canceled.
After change to usleep(5000000), it can be canceled.
So, is sleep() a pthread cancel point on DragonFlyBSD?
I didn't found any information in related manpages.
The binary I tested is linked to /usr/lib/libpthread.so.0,
which symlink to thread/libthread_xu.so.
Thanks
Updated by dillon over 5 years ago
- Status changed from New to Closed
- Assignee set to dillon
sleep is supposed to be a cancel point too. There was a missing strong reference in our pthreads library which caused the wrong sleep() function to be called. I have fixed it in the master and 5.4 release branch.
thanks for the bug report, sorry for the long delay!
-Matt
Actions