Re: [PATCH 0/2] Support hugetlb charge moving at task migration

[Baolin Wang]

On 2021/10/8 19:55, Michal Hocko wrote:
> On Fri 08-10-21 17:17:12, Baolin Wang wrote:
> > On 2021/10/8 15:12, Michal Hocko wrote:
> > > On Thu 07-10-21 23:39:15, Baolin Wang wrote:
> > > > Hi Michal,
> > > >
> > > > (Sorry for late reply due to my holidays)
> > > > On 2021/9/30 18:46, Michal Hocko wrote:
> > > > > On Wed 29-09-21 18:19:26, Baolin Wang wrote:
> > > > > > Hi,
> > > > > >
> > > > > > Now in the hugetlb cgroup, charges associated with a task aren't moved
> > > > > > to the new hugetlb cgroup at task migration, which is odd for hugetlb
> > > > > > cgroup usage.
> > > > >
> > > > > Could you elaborate some more about the usecase and/or problems you see
> > > > > with the existing semantic?
> > > >
> > > > The problems is that, it did not check if the tasks can move to the new
> > > > hugetlb cgroup if the new hugetlb cgroup has a limitation, and the hugetlb
> > > > cgroup usage is incorrect when moving tasks among hugetlb cgroups.
> > >
> > > Could you be more specific please? What do you mean by cgroup usage is
> > > incorrect? Ideally could you describe your usecase?
> >
> > Sorry for confusing, what I mean is, when tasks from one hugetlb cgroup are
> > migrated to a new hugetlb cgroup, the new hugetlb cgroup's hugetlb page
> > usage is not increased accordingly.
>
> Which is a perferctly reasonable behavior as the memory has been
> consumed from the original cgroup and it will be freed there as well.
> Migrating to a new cgroup doesn't imply all the resources to be migrated
> as well.

OK.

> > The issue I found is just from my
> > testing for the hugetlb cgroup, and I think this is not resonable if we've
> > already set a hugetlb limitation for a cgroup, but we always ignore it when
> > tasks migration among hugetlb cgroups.
>
> I would like to learn more about why you consider this unreasonable.
> This will likely depend on the reason why you want/need to migrate task.
> If you want to move a task to completely new resource domain (read a

Yes.

> completely different cgroup subtree) then I can imagine you want to
> leave all the resources but this will have problems with other resource
> controllers - e.g. memory cgroup v2 which doesn't allow charge migration
> either.

OK. I see. Thanks.