]> git.ipfire.org Git - thirdparty/kernel/stable.git/commit
ipv6: prevent infinite loop in rt6_nlmsg_size()
authorEric Dumazet <edumazet@google.com>
Fri, 25 Jul 2025 14:07:23 +0000 (14:07 +0000)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Fri, 15 Aug 2025 10:05:00 +0000 (12:05 +0200)
commit6d345136c9b875f065d226908a29c25cdf9343f8
tree3be2b2444b66b3d5594e1152e566ff8e37d8dc17
parente254e034daa69d6d561dcc0d77e6ed55d4833d51
ipv6: prevent infinite loop in rt6_nlmsg_size()

[ Upstream commit 54e6fe9dd3b0e7c481c2228782c9494d653546da ]

While testing prior patch, I was able to trigger
an infinite loop in rt6_nlmsg_size() in the following place:

list_for_each_entry_rcu(sibling, &f6i->fib6_siblings,
fib6_siblings) {
rt6_nh_nlmsg_size(sibling->fib6_nh, &nexthop_len);
}

This is because fib6_del_route() and fib6_add_rt2node()
uses list_del_rcu(), which can confuse rcu readers,
because they might no longer see the head of the list.

Restart the loop if f6i->fib6_nsiblings is zero.

Fixes: d9ccb18f83ea ("ipv6: Fix soft lockups in fib6_select_path under high next hop churn")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250725140725.3626540-3-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
net/ipv6/ip6_fib.c
net/ipv6/route.c