On Tue, Jul 14, 2009 at 3:16 AM, Balbir Singh<balbir@xxxxxxxxxxxxxxxxxx> wrote: > * menage@xxxxxxxxxx <menage@xxxxxxxxxx> [2009-07-13 23:49:16]: >> As a first cut, we were planning to add an rwsem that gets taken for >> read in cgroup_fork(), released in cgroup_post_fork(), and taken for >> write when moving an entire process to a new cgroup; not ideal >> performance-wise, but safe. >> >> If adding a field to task_struct is an option, then the rwsem could be >> per thread-group leader, which would reduce contention. That would indeed help, but would only improve system-wide performance in the case with write-contention (no improvement at all if, say, the threadgroup is the only one forking, or when nobody is writing to the procs file). While that's preferable to a global lock, if we can add a field to task_struct, a (lockless) flag-based approach might be possible. > We should also document that moving large processes with several > threads can be expensive. Indeed. Or more specifically, that moving fast-growing threadgroups can be expensive. _______________________________________________ Containers mailing list Containers@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/containers
