On 05/06/2019 12:04 PM, Daniel Borkmann wrote:
> On 05/04/2019 06:06 PM, Björn Töpel wrote:
>> From: Björn Töpel <bjorn.topel@xxxxxxxxx>
>>
>> When an AF_XDP socket is released/closed the XSKMAP still holds a
>> reference to the socket in a "released" state. The socket will still
>> use the netdev queue resource, and block newly created sockets from
>> attaching to that queue, but no user application can access the
>> fill/complete/rx/tx rings. This results in that all applications need
>> to explicitly clear the map entry from the old "zombie state"
>> socket. This should be done automatically.
>>
>> After this patch, when a socket is released, it will remove itself
>> from all the XSKMAPs it resides in, allowing the socket application to
>> remove the code that cleans the XSKMAP entry.
>>
>> This behavior is also closer to that of SOCKMAP, making the two socket
>> maps more consistent.
>>
>> Reported-by: Bruce Richardson <bruce.richardson@xxxxxxxxx>
>> Signed-off-by: Björn Töpel <bjorn.topel@xxxxxxxxx>
> [...]
>
>
>> +static void __xsk_map_delete_elem(struct xsk_map *map,
>> + struct xdp_sock **map_entry)
>> +{
>> + struct xdp_sock *old_xs;
>> +
>> + spin_lock_bh(&map->lock);
>> + old_xs = xchg(map_entry, NULL);
>> + if (old_xs)
>> + xsk_map_del_node(old_xs, map_entry);
>> + spin_unlock_bh(&map->lock);
>> +
>> +}
>> +
>> static void xsk_map_free(struct bpf_map *map)
>> {
>> struct xsk_map *m = container_of(map, struct xsk_map, map);
>> @@ -78,15 +142,16 @@ static void xsk_map_free(struct bpf_map *map)
>> bpf_clear_redirect_map(map);
>> synchronize_net();
>>
>> + spin_lock_bh(&m->lock);
>> for (i = 0; i < map->max_entries; i++) {
>> + struct xdp_sock **entry = &m->xsk_map[i];
>> struct xdp_sock *xs;
>>
>> - xs = m->xsk_map[i];
>> - if (!xs)
>> - continue;
>> -
>> - sock_put((struct sock *)xs);
>> + xs = xchg(entry, NULL);
>> + if (xs)
>> + __xsk_map_delete_elem(m, entry);
>> }
>> + spin_unlock_bh(&m->lock);
>>
>
> Was this tested? Doesn't the above straight run into a deadlock?
>
> From xsk_map_free() you iterate over the map with m->lock held. Once you
> xchg'ed the entry and call into __xsk_map_delete_elem(), you attempt to
> call map->lock on the same map once again. What am I missing?
(It also does the xchg() twice so we'd leak the xs since it's NULL in the
second one.)
> Thanks,
> Daniel
