bpf, sockmap: Fix update element with same

commit 75e072a390da9a22e7ae4a4e8434dfca5da499fb upstream.

Consider a sockmap entry being updated with the same socket:

	osk = stab->sks[idx];
	sock_map_add_link(psock, link, map, &stab->sks[idx]);
	stab->sks[idx] = sk;
	if (osk)
		sock_map_unref(osk, &stab->sks[idx]);

Due to sock_map_unref(), which invokes sock_map_del_link(), all the
psock's links for stab->sks[idx] are torn:

	list_for_each_entry_safe(link, tmp, &psock->link, list) {
		if (link->link_raw == link_raw) {
			...
			list_del(&link->list);
			sk_psock_free_link(link);
		}
	}

And that includes the new link sock_map_add_link() added just before
the unref.

This results in a sockmap holding a socket, but without the respective
link. This in turn means that close(sock) won't trigger the cleanup,
i.e. a closed socket will not be automatically removed from the sockmap.

Stop tearing the links when a matching link_raw is found.

Tracked-On: OAM-128620
Fixes: 604326b ("bpf, sockmap: convert to generic sk_msg interface")
Signed-off-by: Michal Luczaj <[email protected]>
Signed-off-by: Daniel Borkmann <[email protected]>
Reviewed-by: John Fastabend <[email protected]>
Link: https://lore.kernel.org/bpf/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

net/core/sock_map.c

@@ -156,6 +156,7 @@ static void sock_map_del_link(struct sock *sk,
 				verdict_stop = true;
 			list_del(&link->list);
 			sk_psock_free_link(link);
+			break;
 		}
 	}
 	spin_unlock_bh(&psock->link_lock);