: Fix arbitrary net_device-s corruptions on net_ns stop.
Subsystems (2)
PATCH 1/1 : Fix arbitrary net_device-s corruptions on net_ns stop.
When a net namespace is destroyed, some devices (those, not killed on ns stop explicitly) are moved back to init_net.
The problem, is that this net_ns change has one point of failure - the __dev_alloc_name() may be called if a name collision occurs (and this is easy to trigger). This allocator performs a likely-to-fail GFP_ATOMIC allocation to find a suitable number. Other possible conditions that may cause error (for device being ns local or not registered) are always false in this case.
So, when this call fails, the device is unregistered. But this is *not* the right thing to do, since after this the device may be released (and kfree-ed) improperly. E. g. bridges require more actions (sysfs update, timer disarming, etc.), some other devices want to remove their private areas from lists, etc.
I. e. arbitrary use-after-free cases may occur.
The proposed fix is the following: since the only reason for the dev_change_net_namespace to fail is the name generation, we may give it a unique fall-back name w/o %d-s in it - the dev<ifindex> one, since ifindexes are still unique.
So make this change, raise the failure-case printk loglevel to EMERG and replace the unregister_netdevice call with BUG().
---
net/core/dev.c +5 -3
@@ -4480,17 +4480,19 @@ static void __net_exit default_device_exit(struct net *net)rtnl_lock ();for_each_netdev_safe (net, dev, next) {int err; +char fb_name[IFNAMSIZ ];/* Ignore unmoveable devices (i.e. loopback) */ if (dev->features &NETIF_F_NETNS_LOCAL )continue ;/* Push remaing network devices to init_net */ - err =dev_change_net_namespace (dev, &init_net,"dev%d" ); +sprintf (fb_name,"dev%d" , dev->ifindex); + err =dev_change_net_namespace (dev, &init_net, fb_name);if (err) { -printk (KERN_WARNING "%s: failed to move %s to init_net: %d\n" , +printk (KERN_EMERG "%s: failed to move %s to init_net: %d\n" , __func__, dev->name, err); -unregister_netdevice (dev); +BUG (); } }rtnl_unlock ();
Quoted 21 lines · Pavel Emelyanov
When a net namespace is destroyed, some devices (those, not killed on ns stop explicitly) are moved back to
init_net.The problem, is that this
net_nschange has one point of failure -
the__dev_alloc_name()may be called if a name collision occurs (and
this is easy to trigger). This allocator performs a likely-to-failGFP_ATOMICallocation to find a suitable number. Other possible conditions that may cause error (for device being ns local or not registered) are always false in this case.So, when this call fails, the device is unregistered. But this is
*not* the right thing to do, since after this the device may be
released (and kfree-ed) improperly. E. g. bridges require more
actions (sysfs update, timer disarming, etc.), some other devices want to remove their private areas from lists, etc.I. e. arbitrary use-after-free cases may occur.
The proposed fix is the following: since the only reason for the
dev_change_net_namespaceto fail is the name generation, we may give it a unique fall-back name w/o %d-s in it - the dev<ifindex> one, since ifindexes are still unique.
Quoted 5 lines
So make this change, raise the failure-case printk loglevel to EMERG and replace the
unregister_netdevicecall withBUG().Signed-off-by Pavel Emelyanov <[email protected]>---
Good catch :)
Quoted 7 lines
net/core/dev.c +0
-0@@ -4480,17 +4480,19 @@ static void __net_exit default_device_exit(struct net *net)rtnl_lock ();for_each_netdev_safe (net, dev, next) {
net/core/dev.c static void __net_exit default_device_exit(struct net *net) · line 4490
Quoted 10 lines
net/core/dev.c +2
-1int err; +char fb_name[IFNAMSIZ ];/* Ignore unmoveable devices (i.e. loopback) */ if (dev->features &NETIF_F_NETNS_LOCAL )continue ;/* Push remaing network devices to init_net */ - err =dev_change_net_namespace (dev, &init_net,"dev%d" ); +sprintf (fb_name,"dev%d" , dev->ifindex);
The computed interface name can not exceed IFNAMSIZ, 3 ('dev') + 10 (max int) + 1 ('\0'). In this case there is no risk to corrupt the stack but may be it is more secure to change that to snprintf(fb_name, IFNAMSIZ, "dev%d", dev->ifindex), just in case, no ?
Quoted 15 lines
net/core/dev.c +3
-3+ err =dev_change_net_namespace (dev, &init_net, fb_name);if (err) { -printk (KERN_WARNING "%s: failed to move %s to init_net: %d\n" , +printk (KERN_EMERG "%s: failed to move %s to init_net: %d\n" , __func__, dev->name, err); -unregister_netdevice (dev); +BUG (); } }rtnl_unlock (); --To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 6 lines
net/core/dev.c static void __net_exit default_device_exit(struct net *net) · line 4490
Quoted 1 line
net/core/dev.c +1
-0+sprintf (fb_name,"dev%d" , dev->ifindex);The computed interface name can not exceed IFNAMSIZ, 3 ('dev') + 10 (max int) + 1 ('\0'). In this case there is no risk to corrupt the stack but may be it is more secure to change that to
snprintf(fb_name, IFNAMSIZ, "dev%d", dev->ifindex), just in case, no ?
But you have just noticed, that "there is no risk to corrupt the stack"! What else can be "more secure" then :) ?
Thanks,
Pavel
Quoted 5 lines · Pavel Emelyanov
Quoted 2 lines
net/core/dev.c static void __net_exit default_device_exit(struct net *net) · line 4490
Quoted 1 line
net/core/dev.c +1
-0+sprintf (fb_name,"dev%d" , dev->ifindex);The computed interface name can not exceed IFNAMSIZ, 3 ('dev') + 10 (max int) + 1 ('\0'). In this case there is no risk to corrupt the stack but may be it is more secure to change that to
snprintf(fb_name, IFNAMSIZ, "dev%d", dev->ifindex), just in case, no ?But you have just noticed, that "there is no risk to corrupt the stack"! What else can be "more secure" then :) ?
Just in case, for example, someone changes 'dev' by 'virtdev' or something else.
From: Daniel Lezcano <[email protected]>
Date: Wed, 07 May 2008 18:12:35 +0200
Quoted 7 lines
Quoted 6 lines · Pavel Emelyanov
Quoted 5 lines
net/core/dev.c static void __net_exit default_device_exit(struct net *net) · line 4490
Quoted 1 line
net/core/dev.c +1
-0+sprintf (fb_name,"dev%d" , dev->ifindex);The computed interface name can not exceed IFNAMSIZ, 3 ('dev') + 10 (max int) + 1 ('\0'). In this case there is no risk to corrupt the stack but may be it is more secure to change that to
snprintf(fb_name, IFNAMSIZ, "dev%d", dev->ifindex), just in case, no ?
Quoted 5 lines
Quoted 2 lines
But you have just noticed, that "there is no risk to corrupt the stack"! What else can be "more secure" then :) ?
Just in case, for example, someone changes 'dev' by 'virtdev' or something else.
I've applied Pavel's patch, but I changed it to use snprintf() to address Daniel's concerns.
And this matches the way this is done in the rest of this file too :-)
Thanks.