netdev - [PATCH net-next v2] ipv6: try not to take rtnl_lock in ip6mr_sk

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <20171108192347.0E52D40383@fruggeri-Arora18_1.sjc.aristanetworks.com>
Date:   Wed, 08 Nov 2017 11:23:46 -0800
From:   fruggeri@...sta.com (Francesco Ruggeri)
To:     yoshfuji@...ux-ipv6.org, kuznet@....inr.ac.ru, davem@...emloft.net,
        fruggeri@...sta.com, netdev@...r.kernel.org
Subject: [PATCH net-next v2] ipv6: try not to take rtnl_lock in
 ip6mr_sk_done

Avoid traversing the list of mr6_tables (which requires the
rtnl_lock) in ip6mr_sk_done(), when we know in advance that
a match will not be found.
This can happen when rawv6_close()/ip6mr_sk_done() is invoked
on non-mroute6 sockets.
This patch helps reduce rtnl_lock contention when destroying
a large number of net namespaces, each having a non-mroute6
raw socket.

v2: same patch, only fixed subject line and expanded comment.

Signed-off-by: Francesco Ruggeri <fruggeri@...sta.com>

diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c
index f5500f5..e1bb2d8 100644
--- a/net/ipv6/ip6mr.c
+++ b/net/ipv6/ip6mr.c
@@ -1617,6 +1617,10 @@ int ip6mr_sk_done(struct sock *sk)
 	struct net *net = sock_net(sk);
 	struct mr6_table *mrt;

+	if (sk->sk_type != SOCK_RAW ||
+	    inet_sk(sk)->inet_num != IPPROTO_ICMPV6)
+		return err;
+
 	rtnl_lock();
 	ip6mr_for_each_table(mrt, net) {
 		if (sk == mrt->mroute6_sk) {