[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1486322541-8206-137-git-send-email-w@1wt.eu>
Date: Sun, 5 Feb 2017 20:20:58 +0100
From: Willy Tarreau <w@....eu>
To: linux-kernel@...r.kernel.org, stable@...r.kernel.org,
linux@...ck-us.net
Cc: Mahesh Bandewar <maheshb@...gle.com>,
"David S . Miller" <davem@...emloft.net>, Willy Tarreau <w@....eu>
Subject: [PATCH 3.10 236/319] bonding: Fix bonding crash
From: Mahesh Bandewar <maheshb@...gle.com>
commit 24b27fc4cdf9e10c5e79e5923b6b7c2c5c95096c upstream.
Following few steps will crash kernel -
(a) Create bonding master
> modprobe bonding miimon=50
(b) Create macvlan bridge on eth2
> ip link add link eth2 dev mvl0 address aa:0:0:0:0:01 \
type macvlan
(c) Now try adding eth2 into the bond
> echo +eth2 > /sys/class/net/bond0/bonding/slaves
<crash>
Bonding does lots of things before checking if the device enslaved is
busy or not.
In this case when the notifier call-chain sends notifications, the
bond_netdev_event() assumes that the rx_handler /rx_handler_data is
registered while the bond_enslave() hasn't progressed far enough to
register rx_handler for the new slave.
This patch adds a rx_handler check that can be performed right at the
beginning of the enslave code to avoid getting into this situation.
Signed-off-by: Mahesh Bandewar <maheshb@...gle.com>
Acked-by: Eric Dumazet <edumazet@...gle.com>
Signed-off-by: David S. Miller <davem@...emloft.net>
Signed-off-by: Willy Tarreau <w@....eu>
---
drivers/net/bonding/bond_main.c | 7 ++++---
include/linux/netdevice.h | 1 +
net/core/dev.c | 16 ++++++++++++++++
3 files changed, 21 insertions(+), 3 deletions(-)
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index c0ed7c8..ce41616 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1565,9 +1565,10 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
bond_dev->name, slave_dev->name);
}
- /* already enslaved */
- if (slave_dev->flags & IFF_SLAVE) {
- pr_debug("Error, Device was already enslaved\n");
+ /* already in-use? */
+ if (netdev_is_rx_handler_busy(slave_dev)) {
+ netdev_err(bond_dev,
+ "Error: Device is in use and cannot be enslaved\n");
return -EBUSY;
}
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 4d2e041..45a618b 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2223,6 +2223,7 @@ static inline void napi_free_frags(struct napi_struct *napi)
napi->skb = NULL;
}
+bool netdev_is_rx_handler_busy(struct net_device *dev);
extern int netdev_rx_handler_register(struct net_device *dev,
rx_handler_func_t *rx_handler,
void *rx_handler_data);
diff --git a/net/core/dev.c b/net/core/dev.c
index 1ccfc49..408f6ee 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3346,6 +3346,22 @@ out:
#endif
/**
+ * netdev_is_rx_handler_busy - check if receive handler is registered
+ * @dev: device to check
+ *
+ * Check if a receive handler is already registered for a given device.
+ * Return true if there one.
+ *
+ * The caller must hold the rtnl_mutex.
+ */
+bool netdev_is_rx_handler_busy(struct net_device *dev)
+{
+ ASSERT_RTNL();
+ return dev && rtnl_dereference(dev->rx_handler);
+}
+EXPORT_SYMBOL_GPL(netdev_is_rx_handler_busy);
+
+/**
* netdev_rx_handler_register - register receive handler
* @dev: device to register a handler for
* @rx_handler: receive handler to register
--
2.8.0.rc2.1.gbe9624a
Powered by blists - more mailing lists