ipv6: Handle missing host route in __ipv6_ifa_notify
Subsystems (3)
David Ahern
maintainerPATCH 1/1 ipv6: Handle missing host route in __ipv6_ifa_notify
From: David Ahern <[email protected]>
Rajendra reported a kernel panic when a link was taken down:
[ 6870.263084] BUG: unable to handle kernel NULL pointer dereference at 00000000000000a8
[ 6870.271856] IP: [<ffffffff8efc5764>] __ipv6_ifa_notify+0x154/0x290 <snip>
[ 6870.570501] Call Trace:
[ 6870.573238] [<ffffffff8efc58c6>] ? ipv6_ifa_notify+0x26/0x40
[ 6870.579665] [<ffffffff8efc98ec>] ? addrconf_dad_completed+0x4c/0x2c0
[ 6870.586869] [<ffffffff8efe70c6>] ? ipv6_dev_mc_inc+0x196/0x260
[ 6870.593491] [<ffffffff8efc9c6a>] ? addrconf_dad_work+0x10a/0x430
[ 6870.600305] [<ffffffff8f01ade4>] ? __switch_to_asm+0x34/0x70
[ 6870.606732] [<ffffffff8ea93a7a>] ? process_one_work+0x18a/0x430
[ 6870.613449] [<ffffffff8ea93d6d>] ? worker_thread+0x4d/0x490
[ 6870.619778] [<ffffffff8ea93d20>] ? process_one_work+0x430/0x430
[ 6870.626495] [<ffffffff8ea99dd9>] ? kthread+0xd9/0xf0
[ 6870.632145] [<ffffffff8f01ade4>] ? __switch_to_asm+0x34/0x70
[ 6870.638573] [<ffffffff8ea99d00>] ? kthread_park+0x60/0x60
[ 6870.644707] [<ffffffff8f01ae77>] ? ret_from_fork+0x57/0x70
[ 6870.650936] Code: 31 c0 31 d2 41 b9 20 00 08 02 b9 09 00 00 0addrconf_dad_work is kicked to be scheduled when a device is brought up. There is a race between addrcond_dad_work getting scheduled and taking the rtnl lock and a process taking the link down (under rtnl). The latter removes the host route from the inet6_addr as part of addrconf_ifdown which is run for NETDEV_DOWN. The former attempts to use the host route in __ipv6_ifa_notify. If the down event removes the host route due to the race to the rtnl, then the BUG listed above occurs.
Since the DAD sequence can not be aborted, add a check for the missing host route in __ipv6_ifa_notify. The only way this should happen is due to the previously mentioned race. The host route is created when the address is added to an interface; it is only removed on a down event where the address is kept. Add a warning if the host route is missing AND the device is up; this is a situation that should never happen.
--- net/ipv6/addrconf.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-)
net/ipv6/addrconf.c +12 -5
@@ -5964,13 +5964,20 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)switch (event) {case RTM_NEWADDR :/* -* If the address was optimistic -* we inserted the route at the start of -* our DAD process, so we don't need -* to do it again +* If the address was optimistic we inserted the route at the +* start of our DAD process, so we don't need to do it again. +* If the device was taken down in the middle of the DAD +* cycle there is a race where we could get here without a +* host route, so nothing to insert. That will be fixed when +* the device is brought up. */ -if (!rcu_access_pointer (ifp->rt->fib6_node)) +if (ifp->rt && !rcu_access_pointer (ifp->rt->fib6_node)) {ip6_ins_rt (net, ifp->rt); + }else if (!ifp->rt && (ifp->idev->dev->flags &IFF_UP )) { +pr_warn ("BUG: Address %pI6c on device %s is missing its host route.\n" , + &ifp->addr, ifp->idev->dev->name); + } +if (ifp->idev->cnf.forwarding)addrconf_join_anycast (ifp);if (!ipv6_addr_any (&ifp->peer_addr))
Signature · 1 line
2.11.0
Quoted 22 lines · David Ahern
From: David Ahern <[email protected]>
Rajendra reported a kernel panic when a link was taken down:
up. There is a race between
addrcond_dad_workgetting scheduled and taking the rtnl lock and a process taking the link down (under rtnl). The latter removes the host route from theinet6_addras part ofaddrconf_ifdownwhich is run forNETDEV_DOWN. The former attempts to use the host route in__ipv6_ifa_notify. If the down event removes the host route due to the race to the rtnl, then the BUG listed above occurs.Since the DAD sequence can not be aborted, add a check for the missing host route in
__ipv6_ifa_notify. The only way this should happen is due to the previously mentioned race. The host route is created when the address is added to an interface; it is only removed on a down event where the address is kept. Add a warning if the host route is missing AND the device is up; this is a situation that should never happen.
Quoted 7 lines
Fixes f1705ec197e7 ("net: ipv6: Make address flushing on ifdown optional")Reported-by Rajendra Dendukuri <[email protected]>Signed-off-by David Ahern <[email protected]>--- net/ipv6/addrconf.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-)
This seems goot to me, thanks.
David Miller
maintainerFrom: David Ahern <[email protected]>
Date: Fri, 4 Oct 2019 08:03:09 -0700
Quoted 40 lines
From: David Ahern <[email protected]>
Rajendra reported a kernel panic when a link was taken down:
[ 6870.263084] BUG: unable to handle kernel NULL pointer dereference at 00000000000000a8 [ 6870.271856] IP: [<ffffffff8efc5764>] __ipv6_ifa_notify+0x154/0x290<snip> [ 6870.570501] Call Trace: [ 6870.573238] [<ffffffff8efc58c6>] ? ipv6_ifa_notify+0x26/0x40 [ 6870.579665] [<ffffffff8efc98ec>] ? addrconf_dad_completed+0x4c/0x2c0 [ 6870.586869] [<ffffffff8efe70c6>] ? ipv6_dev_mc_inc+0x196/0x260 [ 6870.593491] [<ffffffff8efc9c6a>] ? addrconf_dad_work+0x10a/0x430 [ 6870.600305] [<ffffffff8f01ade4>] ? __switch_to_asm+0x34/0x70 [ 6870.606732] [<ffffffff8ea93a7a>] ? process_one_work+0x18a/0x430 [ 6870.613449] [<ffffffff8ea93d6d>] ? worker_thread+0x4d/0x490 [ 6870.619778] [<ffffffff8ea93d20>] ? process_one_work+0x430/0x430 [ 6870.626495] [<ffffffff8ea99dd9>] ? kthread+0xd9/0xf0 [ 6870.632145] [<ffffffff8f01ade4>] ? __switch_to_asm+0x34/0x70 [ 6870.638573] [<ffffffff8ea99d00>] ? kthread_park+0x60/0x60 [ 6870.644707] [<ffffffff8f01ae77>] ? ret_from_fork+0x57/0x70 [ 6870.650936] Code: 31 c0 31 d2 41 b9 20 00 08 02 b9 09 00 00 0
addrconf_dad_workis kicked to be scheduled when a device is brought up. There is a race betweenaddrcond_dad_workgetting scheduled and taking the rtnl lock and a process taking the link down (under rtnl). The latter removes the host route from theinet6_addras part ofaddrconf_ifdownwhich is run forNETDEV_DOWN. The former attempts to use the host route in__ipv6_ifa_notify. If the down event removes the host route due to the race to the rtnl, then the BUG listed above occurs.Since the DAD sequence can not be aborted, add a check for the missing host route in
__ipv6_ifa_notify. The only way this should happen is due to the previously mentioned race. The host route is created when the address is added to an interface; it is only removed on a down event where the address is kept. Add a warning if the host route is missing AND the device is up; this is a situation that should never happen.
Quoted 3 lines
Fixes f1705ec197e7 ("net: ipv6: Make address flushing on ifdown optional")Reported-by Rajendra Dendukuri <[email protected]>Signed-off-by David Ahern <[email protected]>
Applied and queued up for -stable.