[Devel] Re: Screamm.. commit f400e198b2ed26ce55b22a1412ded0896e7516ac

[Eric W. Biederman]

"Serge E. Hallyn" <serue at us.ibm.com> writes:

> Yup.  Looks like ambiguous naming once again hid some real (future)
> bugs.  This is of course safe so far in mainline,  but needs to be split
> into
>
> static inline int is_global_init(struct task_struct *tsk)
> {
> 	return (tsk == &init_task);
> }
>
> and
>
> static inline int
> is_container_init(struct task_struct *task, struct pid_namespace *ns)
> {
> 	return (__pid_nr(task, ns) == 1);
> }

Conceptually yes.  The implementation of is_global_init is just wrong.
&init_task is the first processors idle thread.

is_container_init looks correct but I don't know if the ns parameter
makes any sense.

> Where the latter is needed in, for instance, kernel/capability.c.

Yes.

I think more clear cut examples could be made.  It isn't clear to me
why we skip pid == 1 in kernel/capability.c

I believe a good example is that inside a container you should not
be able to send pid == 1 a signal it doesn't have a handler for.
While from outside the container we need that capability.

Eric
_______________________________________________
Containers mailing list
Containers at lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers