[bpf PATCH] bpf: sockmap missing NULL psock check

[bpf PATCH] bpf: sockmap missing NULL psock check

[John Fastabend]

Add psock NULL check to handle a racing sock event that can get the
sk_callback_lock before this case but after xchg happens causing the
refcnt to hit zero and sock user data (psock) to be null and queued
for garbage collection.

Also add a comment in the code because this is a bit subtle and
not obvious in my opinion.

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
---
 kernel/bpf/sockmap.c |    7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c
index 5ee2e41..dfbbde2 100644
--- a/kernel/bpf/sockmap.c
+++ b/kernel/bpf/sockmap.c
@@ -591,6 +591,13 @@ static void sock_map_free(struct bpf_map *map)
 
 		write_lock_bh(&sock->sk_callback_lock);
 		psock = smap_psock_sk(sock);
+		/* This check handles a racing sock event that can get the
+		 * sk_callback_lock before this case but after xchg happens
+		 * causing the refcnt to hit zero and sock user data (psock)
+		 * to be null and queued for garbage collection.
+		 */
+		if (unlikely(!psock))
+			continue;
 		smap_list_remove(psock, &stab->sock_map[i]);
 		smap_release_sock(psock, sock);
 		write_unlock_bh(&sock->sk_callback_lock);

Re: [bpf PATCH] bpf: sockmap missing NULL psock check

[Daniel Borkmann]

On 01/03/2018 04:47 AM, John Fastabend wrote:
> Add psock NULL check to handle a racing sock event that can get the
> sk_callback_lock before this case but after xchg happens causing the
> refcnt to hit zero and sock user data (psock) to be null and queued
> for garbage collection.
> 
> Also add a comment in the code because this is a bit subtle and
> not obvious in my opinion.
> 
> Signed-off-by: John Fastabend <john.fastabend@gmail.com>
> ---
>  kernel/bpf/sockmap.c |    7 +++++++
>  1 file changed, 7 insertions(+)
> 
> diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c
> index 5ee2e41..dfbbde2 100644
> --- a/kernel/bpf/sockmap.c
> +++ b/kernel/bpf/sockmap.c
> @@ -591,6 +591,13 @@ static void sock_map_free(struct bpf_map *map)
>  
>  		write_lock_bh(&sock->sk_callback_lock);
>  		psock = smap_psock_sk(sock);
> +		/* This check handles a racing sock event that can get the
> +		 * sk_callback_lock before this case but after xchg happens
> +		 * causing the refcnt to hit zero and sock user data (psock)
> +		 * to be null and queued for garbage collection.
> +		 */
> +		if (unlikely(!psock))
> +			continue;

Don't we need a write_unlock_bh(&sock->sk_callback_lock) before the continue?

>  		smap_list_remove(psock, &stab->sock_map[i]);
>  		smap_release_sock(psock, sock);
>  		write_unlock_bh(&sock->sk_callback_lock);
>