netdev - [RFC net-next 3/6] drivers: net: bonding: restrict bond mods when rules are offloaded

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 for Android: free password hash cracker in your pocket

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

Date:   Mon,  5 Mar 2018 13:28:31 +0000
From:   John Hurley <john.hurley@...ronome.com>
To:     netdev@...r.kernel.org
Cc:     jiri@...lanox.com, ogerlitz@...lanox.com,
        jakub.kicinski@...ronome.com, simon.horman@...ronome.com,
        John Hurley <john.hurley@...ronome.com>
Subject: [RFC net-next 3/6] drivers: net: bonding: restrict bond mods when rules are offloaded

When a bond has tc rules offloaded to its slaves, prevent new slaves being
added. To remove a slave from a bond, the offloaded rules must first be
deleted. For the case where a slave port on a bond is unregistered from
the kernel, flush all offloaded rules and destroy the bond.

Signed-off-by: John Hurley <john.hurley@...ronome.com>
---
 drivers/net/bonding/bond_main.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index d9e41cf..4c146b1 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1607,6 +1607,15 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev,
 		return -EPERM;
 	}
 
+	/* check for TC offloaded */
+	if (bond_get_offload_cnt(bond)) {
+		NL_SET_ERR_MSG(extack,
+			       "Cannot enslave - bond has offloaded rules.");
+		netdev_err(bond_dev,
+			   "cannot enslave - bond has offloaded rules.\n");
+		return -EPERM;
+	}
+
 	/* vlan challenged mutual exclusion */
 	/* no need to lock since we're protected by rtnl_lock */
 	if (slave_dev->features & NETIF_F_VLAN_CHALLENGED) {
@@ -2237,6 +2246,13 @@ static int __bond_release_one(struct net_device *bond_dev,
 /* A wrapper used because of ndo_del_link */
 int bond_release(struct net_device *bond_dev, struct net_device *slave_dev)
 {
+	if (bond_get_offload_cnt(netdev_priv(bond_dev))) {
+		netdev_err(bond_dev,
+			   "cannot release %s - has offloaded rules.\n",
+			   slave_dev->name);
+		return -EPERM;
+	}
+
 	return __bond_release_one(bond_dev, slave_dev, false, false);
 }
 
@@ -3325,6 +3341,8 @@ static int bond_slave_netdev_event(unsigned long event,
 	case NETDEV_UNREGISTER:
 		if (bond_dev->type != ARPHRD_ETHER)
 			bond_release_and_destroy(bond_dev, slave_dev);
+		else if (bond_get_offload_cnt(bond))
+			unregister_netdevice_queue(bond_dev, NULL);
 		else
 			__bond_release_one(bond_dev, slave_dev, false, true);
 		break;
-- 
2.7.4