[PATCH] bonding: fix error handling if slave is busy (v2)
From: Stephen Hemminger <hidden>
Date: 2011-12-31 23:26:54
If slave device already has a receive handler registered, then the error unwind of bonding device enslave function is broken. The following will leave a pointer to freed memory in the slave device list, causing a later kernel panic. # modprobe dummy # ip li add dummy0-1 link dummy0 type macvlan # modprobe bonding # echo +dummy0 >/sys/class/net/bond0/bonding/slaves The fix is to detach the slave (which removes it from the list) in the unwind path. Signed-off-by: Stephen Hemminger <redacted> --- v2 - need to keep original err_close for other unwind
--- a/drivers/net/bonding/bond_main.c 2011-12-30 14:20:03.171823181 -0800
+++ b/drivers/net/bonding/bond_main.c 2011-12-31 15:20:16.493379415 -0800@@ -1822,7 +1822,7 @@ int bond_enslave(struct net_device *bond "but new slave device does not support netpoll.\n", bond_dev->name); res = -EBUSY; - goto err_close; + goto err_detach; } } #endif
@@ -1831,7 +1831,7 @@ int bond_enslave(struct net_device *bond res = bond_create_slave_symlinks(bond_dev, slave_dev); if (res) - goto err_close; + goto err_detach; res = netdev_rx_handler_register(slave_dev, bond_handle_frame, new_slave);
@@ -1852,6 +1852,11 @@ int bond_enslave(struct net_device *bond err_dest_symlinks: bond_destroy_slave_symlinks(bond_dev, slave_dev); +err_detach: + write_lock_bh(&bond->lock); + bond_detach_slave(bond, new_slave); + write_unlock_bh(&bond->lock); + err_close: dev_close(slave_dev);