Hello,
On Mon, Aug 13, 2018 at 09:58:10AM +0300, Konstantin Khlebnikov wrote:
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 19a4348974a4..7ef6ea9d5e4a 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -333,6 +333,7 @@ struct cgroup_subsys_state *mem_cgroup_css_from_page(struct page *page)
> /**
> * page_cgroup_ino - return inode number of the memcg a page is charged to
> * @page: the page
> + * @online: return closest online ancestor
> *
> * Look up the closest online ancestor of the memory cgroup @page is charged to
> * and return its inode number or 0 if @page is not charged to any cgroup. It
> @@ -343,14 +344,14 @@ struct cgroup_subsys_state *mem_cgroup_css_from_page(struct page *page)
> * after page_cgroup_ino() returns, so it only should be used by callers that
> * do not care (such as procfs interfaces).
> */
> -ino_t page_cgroup_ino(struct page *page)
> +ino_t page_cgroup_ino(struct page *page, bool online)
> {
> struct mem_cgroup *memcg;
> unsigned long ino = 0;
>
> rcu_read_lock();
> memcg = READ_ONCE(page->mem_cgroup);
> - while (memcg && !(memcg->css.flags & CSS_ONLINE))
> + while (memcg && online && !(memcg->css.flags & CSS_ONLINE))
> memcg = parent_mem_cgroup(memcg);
> if (memcg)
> ino = cgroup_ino(memcg->css.cgroup);
We pin the ino till the cgroup is actually released now but that's an
implementation detail which may change in the future, so I'm not sure
this is a good idea. Can you instead use the 64bit filehandle exposed
by kernfs? That's currently also based on ino (+gen) but it's
something guarnateed to stay unique per cgroup and you can easily get
to the cgroup using the fh too.
Thanks.
--
tejun
