Re: [PATCH 4/5] bpf: sockmap, sockhash: return file descriptors from privileged lookup

[Lorenz Bauer <lmb@cloudflare.com>]

On Wed, 11 Mar 2020 at 23:27, John Fastabend <john.fastabend@gmail.com> wrote:
>
> Lorenz Bauer wrote:
> > Allow callers with CAP_NET_ADMIN to retrieve file descriptors from a
> > sockmap and sockhash. O_CLOEXEC is enforced on all fds.
> >
> > Without this, it's difficult to resize or otherwise rebuild existing
> > sockmap or sockhashes.
> >
> > Suggested-by: Jakub Sitnicki <jakub@cloudflare.com>
> > Signed-off-by: Lorenz Bauer <lmb@cloudflare.com>
> > ---
> >  net/core/sock_map.c | 19 +++++++++++++++++++
> >  1 file changed, 19 insertions(+)
> >
> > diff --git a/net/core/sock_map.c b/net/core/sock_map.c
> > index 03e04426cd21..3228936aa31e 100644
> > --- a/net/core/sock_map.c
> > +++ b/net/core/sock_map.c
> > @@ -347,12 +347,31 @@ static void *sock_map_lookup(struct bpf_map *map, void *key)
> >  static int __sock_map_copy_value(struct bpf_map *map, struct sock *sk,
> >                                void *value)
> >  {
> > +     struct file *file;
> > +     int fd;
> > +
> >       switch (map->value_size) {
> >       case sizeof(u64):
> >               sock_gen_cookie(sk);
> >               *(u64 *)value = atomic64_read(&sk->sk_cookie);
> >               return 0;
> >
> > +     case sizeof(u32):
> > +             if (!capable(CAP_NET_ADMIN))
> > +                     return -EPERM;
> > +
> > +             fd = get_unused_fd_flags(O_CLOEXEC);
> > +             if (unlikely(fd < 0))
> > +                     return fd;
> > +
> > +             read_lock_bh(&sk->sk_callback_lock);
> > +             file = get_file(sk->sk_socket->file);
> > +             read_unlock_bh(&sk->sk_callback_lock);
> > +
> > +             fd_install(fd, file);
> > +             *(u32 *)value = fd;
> > +             return 0;
> > +
>
> Hi Lorenz, Can you say something about what happens if the sk
> is deleted from the map or the sock is closed/unhashed ideally
> in the commit message so we have it for later reference. I guess
> because we are in an rcu block here the sk will be OK and psock
> reference will exist until after the rcu block at least because
> of call_rcu(). If the psock is destroyed from another path then
> the fd will still point at the sock. correct?

This is how I understand it:
* sk is protected by rcu_read_lock (as you point out)
* sk->sk_callback_lock protects against sk->sk_socket being
  modified by sock_orphan, sock_graft, etc. via sk_set_socket
* get_file increments the refcount on the file

I'm not sure how the psock figures into this, maybe you can
elaborate a little?

-- 
Lorenz Bauer  |  Systems Engineer
6th Floor, County Hall/The Riverside Building, SE1 7PB, UK

www.cloudflare.com