Re: [PATCH v3 1/6] mm: memcontrol: introduce mem_cgroup_ino() and mem_cgroup_get_from_ino()

[Roman Gushchin <roman.gushchin@linux.dev>]

On Sun, May 22, 2022 at 03:05:33PM +0800, Muchun Song wrote:
> On Mon, May 09, 2022 at 11:38:15AM -0700, Roman Gushchin wrote:
> > Shrinker debugfs requires a way to represent memory cgroups without
> > using full paths, both for displaying information and getting input
> > from a user.
> > 
> > Cgroup inode number is a perfect way, already used by bpf.
> > 
> > This commit adds a couple of helper functions which will be used
> > to handle memcg-aware shrinkers.
> > 
> > Signed-off-by: Roman Gushchin <roman.gushchin@linux.dev>
> > ---
> >  include/linux/memcontrol.h | 21 +++++++++++++++++++++
> >  mm/memcontrol.c            | 23 +++++++++++++++++++++++
> >  2 files changed, 44 insertions(+)
> > 
> > diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
> > index fe580cb96683..a6de9e5c1549 100644
> > --- a/include/linux/memcontrol.h
> > +++ b/include/linux/memcontrol.h
> > @@ -831,6 +831,15 @@ static inline unsigned short mem_cgroup_id(struct mem_cgroup *memcg)
> >  }
> >  struct mem_cgroup *mem_cgroup_from_id(unsigned short id);
> >  
> > +#ifdef CONFIG_SHRINKER_DEBUG
> > +static inline unsigned long mem_cgroup_ino(struct mem_cgroup *memcg)
> > +{
> > +	return memcg ? cgroup_ino(memcg->css.cgroup) : 0;
> > +}
> > +
> > +struct mem_cgroup *mem_cgroup_get_from_ino(unsigned long ino);
> > +#endif
> > +
> >  static inline struct mem_cgroup *mem_cgroup_from_seq(struct seq_file *m)
> >  {
> >  	return mem_cgroup_from_css(seq_css(m));
> > @@ -1324,6 +1333,18 @@ static inline struct mem_cgroup *mem_cgroup_from_id(unsigned short id)
> >  	return NULL;
> >  }
> >  
> > +#ifdef CONFIG_SHRINKER_DEBUG
> > +static inline unsigned long mem_cgroup_ino(struct mem_cgroup *memcg)
> > +{
> > +	return 0;
> > +}
> > +
> > +static inline struct mem_cgroup *mem_cgroup_get_from_ino(unsigned long ino)
> > +{
> > +	return NULL;
> > +}
> > +#endif
> > +
> >  static inline struct mem_cgroup *mem_cgroup_from_seq(struct seq_file *m)
> >  {
> >  	return NULL;
> > diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> > index 04cea4fa362a..e6472728fa66 100644
> > --- a/mm/memcontrol.c
> > +++ b/mm/memcontrol.c
> > @@ -5018,6 +5018,29 @@ struct mem_cgroup *mem_cgroup_from_id(unsigned short id)
> >  	return idr_find(&mem_cgroup_idr, id);
> >  }
> >  
> > +#ifdef CONFIG_SHRINKER_DEBUG
> > +struct mem_cgroup *mem_cgroup_get_from_ino(unsigned long ino)
> > +{
> > +	struct cgroup *cgrp;
> > +	struct cgroup_subsys_state *css;
> > +	struct mem_cgroup *memcg;
> > +
> > +	cgrp = cgroup_get_from_id(ino);
> > +	if (!cgrp)
> > +		return ERR_PTR(-ENOENT);
> > +
> > +	css = cgroup_get_e_css(cgrp, &memory_cgrp_subsys);
> > +	if (css)
> > +		memcg = container_of(css, struct mem_cgroup, css);
> > +	else
> > +		memcg = ERR_PTR(-ENOENT);
> > +
> > +	cgroup_put(cgrp);
> 
> I think it's better to use css_put() here since the refcount is get
> via cgroup_get_e_css() which returns a css struct.

cgroup_put() is matching cgroup_get_from_id().

The reference grabbed by cgroup_get_e_css() shouldn't be dropped
because mem_cgroup_get_from_ino() has a "get" semantics.

Thanks!