[Devel] Re: Question : memrlimit cgroup's task_move (2.6.26-rc5-mm3)

[Balbir Singh]

KAMEZAWA Hiroyuki wrote:
> I used memrlimit cgroup at the first time.
> 
> May I ask a question about memrlimit cgroup ?
> 
> In following 
> ==
> static void memrlimit_cgroup_move_task(struct cgroup_subsys *ss,
>                                         struct cgroup *cgrp,
>                                         struct cgroup *old_cgrp,
>                                         struct task_struct *p)
> {
>         struct mm_struct *mm;
>         struct memrlimit_cgroup *memrcg, *old_memrcg;
> 
> <snip>
>         if (res_counter_charge(&memrcg->as_res, (mm->total_vm << PAGE_SHIFT)))
>                 goto out;
>         res_counter_uncharge(&old_memrcg->as_res, (mm->total_vm << PAGE_SHIFT));
> ==
> This is a callback for task_attach(). and this never fails.
> 
> What happens when the moved task, which move-of-charge fails, exits ?
> 

Good question - I am working on this, some of the logic should move to
can_attach(). I'll try and experiment with it and send out a fix.

> ==
> % mkdir /dev/cgroup/memrlimit/group_01
> % mkdir /dev/cgroup/memrlimit/group_02
> % echo 1G > /dev/cgroup/memrlimit/group_01/memrlimit.limit_in_bytes
> % echo 0 >  /dev/cgroup/memrlimit/group_02/memrlimit.limit_in_bytes
> % echo $$ > /dev/cgroup/memrlimit/group_01/tasks
> % echo $$ > /dev/cgroup/memrlimit/group_02/tasks
> % exit
> == you'll see WARNING ==
> 
> I think the charge of the new group goes to minus. right ?
> (and old group's charge never goes down.)
> I don't think this is "no problem".
> 
> What kind of patch is necessary to fix this ?
> task_attach() should be able to fail in future ?
> 
> I'm sorry if I misunderstand something or this is already in TODO list.
> 

It's already on the TODO list. Thanks for keeping me reminded about it.

-- 
	Warm Regards,
	Balbir Singh
	Linux Technology Center
	IBM, ISTL
_______________________________________________
Containers mailing list
Containers at lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers