Re: [RFC PATCH] cgroup: use root_mem_cgroup as css when current is not enabled

[Zhaoyang Huang]

On Sat, Aug 13, 2022 at 3:06 AM Tejun Heo <tj@xxxxxxxxxx> wrote:
>
> On Fri, Aug 12, 2022 at 06:09:26PM +0800, zhaoyang.huang wrote:
> > From: Zhaoyang Huang <zhaoyang.huang@xxxxxxxxxx>
> >
> > Memory charged on group B abserved on belowing v2 hierarchy where we just would
> > like to only have group E's memory be controlled and B's descendants compete freely
> > for memory. This should be the consequences of unified hierarchy. Solve this by
> > have the cgroup without valid memory css alloced use root_mem_cgroup instead of
> > its ancestor's.
> >
> >  A(subtree_control = memory) - B(subtree_control = NULL) - C()
> >                                                          \ D()
> >                            - E(subtree_control = memory) - F()
> >                                                          \ G()
> >
> > Signed-off-by: Zhaoyang Huang <zhaoyang.huang@xxxxxxxxxx>
> > ---
> >  kernel/cgroup/cgroup.c | 8 ++++++++
> >  1 file changed, 8 insertions(+)
> >
> > diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
> > index 1779ccd..b29b3f6 100644
> > --- a/kernel/cgroup/cgroup.c
> > +++ b/kernel/cgroup/cgroup.c
> > @@ -533,6 +533,14 @@ static struct cgroup_subsys_state *cgroup_e_css_by_mask(struct cgroup *cgrp,
> >        * can't test the csses directly.  Test ss_mask.
> >        */
> >       while (!(cgroup_ss_mask(cgrp) & (1 << ss->id))) {
> > +             /*
> > +              * charging to the parent cgroup which hasn't distribute
> > +              * memory control to its descendants doesn't make sense
> > +              * especially on cgroup v2, where the parent could be configured
> > +              * to use memory controller as its sibling want to use it
> > +              */
> > +             if (memory_cgrp_id == ss->id)
> > +                     return &root_mem_cgroup->css;
>
> This is gonna be a hard nack. A given cgroup always encompasses all the
> resources consumed in its self-including subtree.
>
> Thanks.
IMHO, I would like to say if it makes more sense as "A given cgroup
always encompasses all the resources consumed in its ENABLED
self-including subtree." Otherwise, how should I couple with the
scenarios I raised in the commit message which I prefer parts of the
subtrees compete for "memory" while others are free for it. The free
here is not only without "min/low/high watermarks" but also not
charged to their own LRU.
>
> --
> tejun