[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20171123134629.ba5fzofrq37tmt7n@dhcp22.suse.cz>
Date: Thu, 23 Nov 2017 14:46:29 +0100
From: Michal Hocko <mhocko@...nel.org>
To: Tetsuo Handa <penguin-kernel@...ove.SAKURA.ne.jp>
Cc: jack@...e.cz, viro@...iv.linux.org.uk, david@...morbit.com,
linux-kernel@...r.kernel.org, linux-fsdevel@...r.kernel.org,
akpm@...ux-foundation.org
Subject: Re: [PATCH] fs: handle shrinker registration failure in sget_userns
On Thu 23-11-17 22:35:34, Tetsuo Handa wrote:
> Jan Kara wrote:
> > Looks good to me now. You can add:
> >
> > Reviewed-by: Jan Kara <jack@...e.cz>
> >
>
> It does not look good to me, for "goto fail" can call
> destroy_unused_super() before s->s_shrink.list is initialized.
> Also, the comment block saying "this object isn't exposed yet"
> wants to be updated?
>
> ---
> fs/super.c | 10 +++-------
> 1 file changed, 3 insertions(+), 7 deletions(-)
>
> diff --git a/fs/super.c b/fs/super.c
> index 80b118c..44f0c6b 100644
> --- a/fs/super.c
> +++ b/fs/super.c
> @@ -197,6 +197,7 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags,
> if (!s)
> return NULL;
>
> + INIT_LIST_HEAD(&s->s_shrink.list);
> INIT_LIST_HEAD(&s->s_mounts);
> s->s_user_ns = get_user_ns(user_ns);
>
You are right. I will move it.
> @@ -260,9 +261,8 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags,
> s->s_shrink.count_objects = super_cache_count;
> s->s_shrink.batch = 1024;
> s->s_shrink.flags = SHRINKER_NUMA_AWARE | SHRINKER_MEMCG_AWARE;
> - INIT_LIST_HEAD(&s->s_shrink.list);
> - return s;
> -
> + if (register_shrinker(&s->s_shrink) == 0)
> + return s;
> fail:
> destroy_unused_super(s);
> return NULL;
But I am not sure this is correct. So what protects shrinker invocation
while the object is not initialized yet?
--
Michal Hocko
SUSE Labs
Powered by blists - more mailing lists