Re: [PATCH 3/4] memcg: fix page migration to reset_owner

[KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>]

On Wed, 28 Dec 2011 16:23:29 -0800 (PST)
Hugh Dickins <hughd@google.com> wrote:

> Usually, migration pages coming to unmap_and_move()'s putback_lru_page()
> have been charged and have pc->mem_cgroup set; but there are several ways
> in which a freshly allocated uncharged page can get there, oopsing when
> added to LRU.  Call mem_cgroup_reset_owner() immediately after allocating.
> 
> Signed-off-by: Hugh Dickins <hughd@google.com>

Ah, ok. It calls putback_lru_page()...

Thank you very much!.
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>


> ---
> Fix N to
>     memcg: clear pc->mem_cgorup if necessary.
> 
>  mm/migrate.c |    2 ++
>  1 file changed, 2 insertions(+)
> 
> --- mmotm.orig/mm/migrate.c	2011-12-22 02:53:31.900041565 -0800
> +++ mmotm/mm/migrate.c	2011-12-28 14:52:37.243034125 -0800
> @@ -841,6 +841,8 @@ static int unmap_and_move(new_page_t get
>  	if (!newpage)
>  		return -ENOMEM;
>  
> +	mem_cgroup_reset_owner(newpage);
> +
>  	if (page_count(page) == 1) {
>  		/* page was freed from under us. So we are done. */
>  		goto out;
> 

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>