5.10-stable review patch. If anyone has any objections, please let me know.
------------------
From: Florian Westphal fw@strlen.de
[ Upstream commit de788b2e6227462b6dcd0e07474e72c089008f74 ]
There is a reference count leak in ctnetlink_dump_table(): if (res < 0) { nf_conntrack_get(&ct->ct_general); // HERE cb->args[1] = (unsigned long)ct; ...
While its very unlikely, its possible that ct == last. If this happens, then the refcount of ct was already incremented. This 2nd increment is never undone.
This prevents the conntrack object from being released, which in turn keeps prevents cnet->count from dropping back to 0.
This will then block the netns dismantle (or conntrack rmmod) as nf_conntrack_cleanup_net_list() will wait forever.
This can be reproduced by running conntrack_resize.sh selftest in a loop. It takes ~20 minutes for me on a preemptible kernel on average before I see a runaway kworker spinning in nf_conntrack_cleanup_net_list.
One fix would to change this to: if (res < 0) { if (ct != last) nf_conntrack_get(&ct->ct_general);
But this reference counting isn't needed in the first place. We can just store a cookie value instead.
A followup patch will do the same for ctnetlink_exp_dump_table, it looks to me as if this has the same problem and like ctnetlink_dump_table, we only need a 'skip hint', not the actual object so we can apply the same cookie strategy there as well.
Fixes: d205dc40798d ("[NETFILTER]: ctnetlink: fix deadlock in table dumping") Signed-off-by: Florian Westphal fw@strlen.de Signed-off-by: Pablo Neira Ayuso pablo@netfilter.org Signed-off-by: Sasha Levin sashal@kernel.org --- net/netfilter/nf_conntrack_netlink.c | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-)
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index b2b06033ef2c..f622fcad3f50 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -839,8 +839,6 @@ ctnetlink_conntrack_event(unsigned int events, struct nf_ct_event *item)
static int ctnetlink_done(struct netlink_callback *cb) { - if (cb->args[1]) - nf_ct_put((struct nf_conn *)cb->args[1]); kfree(cb->data); return 0; } @@ -1112,19 +1110,26 @@ static int ctnetlink_filter_match(struct nf_conn *ct, void *data) return 0; }
+static unsigned long ctnetlink_get_id(const struct nf_conn *ct) +{ + unsigned long id = nf_ct_get_id(ct); + + return id ? id : 1; +} + static int ctnetlink_dump_table(struct sk_buff *skb, struct netlink_callback *cb) { unsigned int flags = cb->data ? NLM_F_DUMP_FILTERED : 0; struct net *net = sock_net(skb->sk); - struct nf_conn *ct, *last; + unsigned long last_id = cb->args[1]; struct nf_conntrack_tuple_hash *h; struct hlist_nulls_node *n; struct nf_conn *nf_ct_evict[8]; + struct nf_conn *ct; int res, i; spinlock_t *lockp;
- last = (struct nf_conn *)cb->args[1]; i = 0;
local_bh_disable(); @@ -1160,7 +1165,7 @@ ctnetlink_dump_table(struct sk_buff *skb, struct netlink_callback *cb) continue;
if (cb->args[1]) { - if (ct != last) + if (ctnetlink_get_id(ct) != last_id) continue; cb->args[1] = 0; } @@ -1173,8 +1178,7 @@ ctnetlink_dump_table(struct sk_buff *skb, struct netlink_callback *cb) NFNL_MSG_TYPE(cb->nlh->nlmsg_type), ct, true, flags); if (res < 0) { - nf_conntrack_get(&ct->ct_general); - cb->args[1] = (unsigned long)ct; + cb->args[1] = ctnetlink_get_id(ct); spin_unlock(lockp); goto out; } @@ -1187,12 +1191,10 @@ ctnetlink_dump_table(struct sk_buff *skb, struct netlink_callback *cb) } out: local_bh_enable(); - if (last) { + if (last_id) { /* nf ct hash resize happened, now clear the leftover. */ - if ((struct nf_conn *)cb->args[1] == last) + if (cb->args[1] == last_id) cb->args[1] = 0; - - nf_ct_put(last); }
while (i) {