ipv6: restore the behavior of ipv6_sock_ac_drop()
authorWANG Cong <xiyou.wangcong@gmail.com>
Fri, 5 Sep 2014 21:33:00 +0000 (14:33 -0700)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 15 Oct 2014 06:36:41 +0000 (08:36 +0200)
commitd7c5b263df9a794934589b372120e8043707f82e
tree857df022ce237c89483ac91bbd3b4104236f3b81
parent9ede8fd5c6a7530d99ce93416f47e72872b052d2
ipv6: restore the behavior of ipv6_sock_ac_drop()

[ Upstream commit de185ab46cb02df9738b0d898b0c3a89181c5526 ]

It is possible that the interface is already gone after joining
the list of anycast on this interface as we don't hold a refcount
for the device, in this case we are safe to ignore the error.

What's more important, for API compatibility we should not
change this behavior for applications even if it were correct.

Fixes: commit a9ed4a2986e13011 ("ipv6: fix rtnl locking in setsockopt for anycast and multicast")
Cc: Sabrina Dubroca <sd@queasysnail.net>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
net/ipv6/anycast.c