bpf, sockmap: fix map elem deletion race with smap_stop_sock

The smap_start_sock() and smap_stop_sock() are each protected under
the sock->sk_callback_lock from their call-sites except in the case
of sock_map_delete_elem() where we drop the old socket from the map
slot. This is racy because the same sock could be part of multiple
sock maps, so we run smap_stop_sock() in parallel, and given at that
point psock->strp_enabled might be true on both CPUs, we might for
example wrongly restore the sk->sk_data_ready / sk->sk_write_space.
Therefore, hold the sock->sk_callback_lock as well on delete. Looks
like 2f857d0 ("bpf: sockmap, remove STRPARSER map_flags and add
multi-map support") had this right, but later on e9db4ef ("bpf:
sockhash fix omitted bucket lock in sock_close") removed it again
from delete leaving this smap_stop_sock() instance unprotected.

Fixes: e9db4ef ("bpf: sockhash fix omitted bucket lock in sock_close")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

kernel/bpf/sockmap.c

@@ -1786,8 +1786,11 @@ static int sock_map_delete_elem(struct bpf_map *map, void *key)
 	if (!psock)
 		goto out;
 
-	if (psock->bpf_parse)
+	if (psock->bpf_parse) {
+		write_lock_bh(&sock->sk_callback_lock);
 		smap_stop_sock(psock, sock);
+		write_unlock_bh(&sock->sk_callback_lock);
+	}
 	smap_list_map_remove(psock, &stab->sock_map[k]);
 	smap_release_sock(psock, sock);
 out: