Commit 2dc6241e authored Apr 11, 2024 by Jakub Kicinski Committed by Zhengchao Shao Apr 11, 2024

net/sched: act_mirred: use the backlog for mirred ingress

mainline inclusion
from mainline-v6.8-rc6
commit 52f671db18823089a02f07efc04efdb2272ddc17
category: bugfix
bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9E2LT
CVE: CVE-2024-26740

Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=52f671db18823089a02f07efc04efdb2272ddc17



--------------------------------

The test Davide added in commit ca22da2f ("act_mirred: use the backlog
for nested calls to mirred ingress") hangs our testing VMs every 10 or so
runs, with the familiar tcp_v4_rcv -> tcp_v4_rcv deadlock reported by
lockdep.

The problem as previously described by Davide (see Link) is that
if we reverse flow of traffic with the redirect (egress -> ingress)
we may reach the same socket which generated the packet. And we may
still be holding its socket lock. The common solution to such deadlocks
is to put the packet in the Rx backlog, rather than run the Rx path
inline. Do that for all egress -> ingress reversals, not just once
we started to nest mirred calls.

In the past there was a concern that the backlog indirection will
lead to loss of error reporting / less accurate stats. But the current
workaround does not seem to address the issue.

Fixes: 53592b36 ("net/sched: act_mirred: Implement ingress actions")
Cc: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Suggested-by: Davide Caratti <dcaratti@redhat.com>
Link: https://lore.kernel.org/netdev/33dc43f587ec1388ba456b4915c75f02a8aae226.1663945716.git.dcaratti@redhat.com/


Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Conflicts:
	net/sched/act_mirred.c

Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>

parent 9b7534a0

net/sched/act_mirred.c

+6 −9

Original line number	Diff line number	Diff line
		@@ -197,18 +197,14 @@ static int tcf_mirred_init(struct net net, struct nlattr nla,
		return ret;
		}

		static bool is_mirred_nested(void)
		{
		return unlikely(__this_cpu_read(mirred_rec_level) > 1);
		}

		static int tcf_mirred_forward(bool want_ingress, struct sk_buff *skb)
		static int
		tcf_mirred_forward(bool at_ingress, bool want_ingress, struct sk_buff *skb)
		{
		int err;

		if (!want_ingress)
		err = dev_queue_xmit(skb);
		else if (is_mirred_nested())
		else if (!at_ingress)
		err = netif_rx(skb);
		else
		err = netif_receive_skb(skb);
		@@ -300,14 +296,15 @@ static int tcf_mirred_act(struct sk_buff skb, const struct tc_action a,
		if (use_reinsert) {
		res->ingress = want_ingress;
		res->qstats = this_cpu_ptr(m->common.cpu_qstats);
		if (tcf_mirred_forward(want_ingress, skb) && res->qstats)
		if (tcf_mirred_forward(skb_at_tc_ingress(skb), want_ingress, skb)
		&& res->qstats)
		qstats_overlimit_inc(res->qstats);
		__this_cpu_dec(mirred_rec_level);
		return TC_ACT_CONSUMED;
		}
		}

		err = tcf_mirred_forward(want_ingress, skb2);
		err = tcf_mirred_forward(skb_at_tc_ingress(skb), want_ingress, skb2);
		if (err) {
		out:
		qstats_overlimit_inc(this_cpu_ptr(m->common.cpu_qstats));

tools/testing/selftests/net/forwarding/tc_actions.sh

+0 −3

Original line number	Diff line number	Diff line
		@@ -183,9 +183,6 @@ mirred_egress_to_ingress_tcp_test()
		check_err $? "didn't mirred redirect ICMP"
		tc_check_packets "dev $h1 ingress" 102 10
		check_err $? "didn't drop mirred ICMP"
		local overlimits=$(tc_rule_stats_get ${h1} 101 egress .overlimits)
		test ${overlimits} = 10
		check_err $? "wrong overlimits, expected 10 got ${overlimits}"

		tc filter del dev $h1 egress protocol ip pref 100 handle 100 flower
		tc filter del dev $h1 egress protocol ip pref 101 handle 101 flower