[PATCH] cgroup: fix a race condition in manipulating tsk->cg_list

Paul Menage menage at google.com
Wed Apr 16 21:17:34 PDT 2008


On Wed, Apr 16, 2008 at 9:11 PM, Andrew Morton
<akpm at linux-foundation.org> wrote:
>
>  I don't fully understand the race.  Both paths hold css_set_lock.
>
>  Can you describe it in more detail please?

Task A starts exiting, passes the check for unlinking current->cg_list.

Before it completely exits task B does the very first
cgroup_iter_begin() call (via reading a cgroups tasks file) which
links all tasks in to their css_set objects via tsk->cg_list.

Then task A finishes exiting and is freed, but doesn't unlink from the cg_list.

>
>  afacit the task at *p could set PF_EXITING immediately after this code has
>  tested PF_EXITING and then the task at *p could proceed until we hit the
>  same race (whatever that is).

The important fact there is that the task sets PF_EXITING *before* it
checks whether it needs to unlink from current->cg_list.

Paul


More information about the Containers mailing list