OL> In that case, the atomic_inc for the first task should occur
OL> earlier - as soon as it is attached to the task _and_ inserted
OL> into the objhash.

Okay, fair enough.  By my calculations, it belongs just a few lines
above, in restore_sem_undo() before a successful termination.  Then we
leave the refcount increment in restore_obj_sem_undo(), but only for
second-and-later tasks.  Agreed?

-- 
Dan Smith
IBM Linux Technology Center
email: [email protected]
_______________________________________________
Containers mailing list
[email protected]
https://lists.linux-foundation.org/mailman/listinfo/containers

_______________________________________________
Devel mailing list
[email protected]
https://openvz.org/mailman/listinfo/devel

Reply via email to