* [PATCH net v3] sctp: fix dst refcnt leak in sctp_v4_get_dst
@ 2018-02-05 19:48 ` Tommi Rantala
0 siblings, 0 replies; 14+ messages in thread
From: Tommi Rantala @ 2018-02-05 19:48 UTC (permalink / raw)
To: netdev
Cc: linux-sctp, Neil Horman, Alexey Kodanev, Marcelo Ricardo Leitner,
Tommi Rantala
Fix dst reference count leak in sctp_v4_get_dst() introduced in commit
410f03831 ("sctp: add routing output fallback"):
When walking the address_list, successive ip_route_output_key() calls
may return the same rt->dst with the reference incremented on each call.
The code would not decrement the dst refcount when the dst pointer was
identical from the previous iteration, causing the dst refcnt leak.
Testcase:
ip netns add TEST
ip netns exec TEST ip link set lo up
ip link add dummy0 type dummy
ip link add dummy1 type dummy
ip link add dummy2 type dummy
ip link set dev dummy0 netns TEST
ip link set dev dummy1 netns TEST
ip link set dev dummy2 netns TEST
ip netns exec TEST ip addr add 192.168.1.1/24 dev dummy0
ip netns exec TEST ip link set dummy0 up
ip netns exec TEST ip addr add 192.168.1.2/24 dev dummy1
ip netns exec TEST ip link set dummy1 up
ip netns exec TEST ip addr add 192.168.1.3/24 dev dummy2
ip netns exec TEST ip link set dummy2 up
ip netns exec TEST sctp_test -H 192.168.1.2 -P 20002 -h 192.168.1.1 -p 20000 -s -B 192.168.1.3
ip netns del TEST
In 4.4 and 4.9 kernels this results to:
[ 354.179591] unregister_netdevice: waiting for lo to become free. Usage count = 1
[ 364.419674] unregister_netdevice: waiting for lo to become free. Usage count = 1
[ 374.663664] unregister_netdevice: waiting for lo to become free. Usage count = 1
[ 384.903717] unregister_netdevice: waiting for lo to become free. Usage count = 1
[ 395.143724] unregister_netdevice: waiting for lo to become free. Usage count = 1
[ 405.383645] unregister_netdevice: waiting for lo to become free. Usage count = 1
...
Fixes: 410f03831 ("sctp: add routing output fallback")
Fixes: 0ca50d12f ("sctp: fix src address selection if using secondary addresses")
Signed-off-by: Tommi Rantala <tommi.t.rantala@nokia.com>
---
v2: unconditional dst_release() before breaking out of the loop:
- if dst = NULL, then dst_release() is no-op.
- if dst != &rt->dst, then drop reference to the first dst, we do not
need it after all.
- if dst = &rt->dst, then the refcnt was incremented twice for this
dst, so drop the second ref.
net/sctp/protocol.c | 10 ++++------
1 file changed, 4 insertions(+), 6 deletions(-)
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index ba28d270eb24..aba2381d3cf9 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -509,22 +509,20 @@ static void sctp_v4_get_dst(struct sctp_transport *t, union sctp_addr *saddr,
if (IS_ERR(rt))
continue;
- if (!dst)
- dst = &rt->dst;
-
/* Ensure the src address belongs to the output
* interface.
*/
odev = __ip_dev_find(sock_net(sk), laddr->a.v4.sin_addr.s_addr,
false);
if (!odev || odev->ifindex != fl4->flowi4_oif) {
- if (&rt->dst != dst)
+ if (!dst)
+ dst = &rt->dst;
+ else
dst_release(&rt->dst);
continue;
}
- if (dst != &rt->dst)
- dst_release(dst);
+ dst_release(dst);
dst = &rt->dst;
break;
}
--
2.15.1
^ permalink raw reply related [flat|nested] 14+ messages in thread* Re: [PATCH net v3] sctp: fix dst refcnt leak in sctp_v4_get_dst
2018-02-05 19:48 ` Tommi Rantala
@ 2018-02-05 22:07 ` Marcelo Ricardo Leitner
-1 siblings, 0 replies; 14+ messages in thread
From: Marcelo Ricardo Leitner @ 2018-02-05 22:07 UTC (permalink / raw)
To: Tommi Rantala; +Cc: netdev, linux-sctp, Neil Horman, Alexey Kodanev
On Mon, Feb 05, 2018 at 09:48:14PM +0200, Tommi Rantala wrote:
> Fix dst reference count leak in sctp_v4_get_dst() introduced in commit
> 410f03831 ("sctp: add routing output fallback"):
>
> When walking the address_list, successive ip_route_output_key() calls
> may return the same rt->dst with the reference incremented on each call.
>
> The code would not decrement the dst refcount when the dst pointer was
> identical from the previous iteration, causing the dst refcnt leak.
>
> Testcase:
> ip netns add TEST
> ip netns exec TEST ip link set lo up
> ip link add dummy0 type dummy
> ip link add dummy1 type dummy
> ip link add dummy2 type dummy
> ip link set dev dummy0 netns TEST
> ip link set dev dummy1 netns TEST
> ip link set dev dummy2 netns TEST
> ip netns exec TEST ip addr add 192.168.1.1/24 dev dummy0
> ip netns exec TEST ip link set dummy0 up
> ip netns exec TEST ip addr add 192.168.1.2/24 dev dummy1
> ip netns exec TEST ip link set dummy1 up
> ip netns exec TEST ip addr add 192.168.1.3/24 dev dummy2
> ip netns exec TEST ip link set dummy2 up
> ip netns exec TEST sctp_test -H 192.168.1.2 -P 20002 -h 192.168.1.1 -p 20000 -s -B 192.168.1.3
> ip netns del TEST
>
> In 4.4 and 4.9 kernels this results to:
> [ 354.179591] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 364.419674] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 374.663664] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 384.903717] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 395.143724] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 405.383645] unregister_netdevice: waiting for lo to become free. Usage count = 1
> ...
>
> Fixes: 410f03831 ("sctp: add routing output fallback")
> Fixes: 0ca50d12f ("sctp: fix src address selection if using secondary addresses")
> Signed-off-by: Tommi Rantala <tommi.t.rantala@nokia.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Thanks
> ---
>
> v2: unconditional dst_release() before breaking out of the loop:
> - if dst == NULL, then dst_release() is no-op.
> - if dst != &rt->dst, then drop reference to the first dst, we do not
> need it after all.
> - if dst == &rt->dst, then the refcnt was incremented twice for this
> dst, so drop the second ref.
>
> net/sctp/protocol.c | 10 ++++------
> 1 file changed, 4 insertions(+), 6 deletions(-)
>
> diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
> index ba28d270eb24..aba2381d3cf9 100644
> --- a/net/sctp/protocol.c
> +++ b/net/sctp/protocol.c
> @@ -509,22 +509,20 @@ static void sctp_v4_get_dst(struct sctp_transport *t, union sctp_addr *saddr,
> if (IS_ERR(rt))
> continue;
>
> - if (!dst)
> - dst = &rt->dst;
> -
> /* Ensure the src address belongs to the output
> * interface.
> */
> odev = __ip_dev_find(sock_net(sk), laddr->a.v4.sin_addr.s_addr,
> false);
> if (!odev || odev->ifindex != fl4->flowi4_oif) {
> - if (&rt->dst != dst)
> + if (!dst)
> + dst = &rt->dst;
> + else
> dst_release(&rt->dst);
> continue;
> }
>
> - if (dst != &rt->dst)
> - dst_release(dst);
> + dst_release(dst);
> dst = &rt->dst;
> break;
> }
> --
> 2.15.1
>
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH net v3] sctp: fix dst refcnt leak in sctp_v4_get_dst
@ 2018-02-05 22:07 ` Marcelo Ricardo Leitner
0 siblings, 0 replies; 14+ messages in thread
From: Marcelo Ricardo Leitner @ 2018-02-05 22:07 UTC (permalink / raw)
To: Tommi Rantala; +Cc: netdev, linux-sctp, Neil Horman, Alexey Kodanev
On Mon, Feb 05, 2018 at 09:48:14PM +0200, Tommi Rantala wrote:
> Fix dst reference count leak in sctp_v4_get_dst() introduced in commit
> 410f03831 ("sctp: add routing output fallback"):
>
> When walking the address_list, successive ip_route_output_key() calls
> may return the same rt->dst with the reference incremented on each call.
>
> The code would not decrement the dst refcount when the dst pointer was
> identical from the previous iteration, causing the dst refcnt leak.
>
> Testcase:
> ip netns add TEST
> ip netns exec TEST ip link set lo up
> ip link add dummy0 type dummy
> ip link add dummy1 type dummy
> ip link add dummy2 type dummy
> ip link set dev dummy0 netns TEST
> ip link set dev dummy1 netns TEST
> ip link set dev dummy2 netns TEST
> ip netns exec TEST ip addr add 192.168.1.1/24 dev dummy0
> ip netns exec TEST ip link set dummy0 up
> ip netns exec TEST ip addr add 192.168.1.2/24 dev dummy1
> ip netns exec TEST ip link set dummy1 up
> ip netns exec TEST ip addr add 192.168.1.3/24 dev dummy2
> ip netns exec TEST ip link set dummy2 up
> ip netns exec TEST sctp_test -H 192.168.1.2 -P 20002 -h 192.168.1.1 -p 20000 -s -B 192.168.1.3
> ip netns del TEST
>
> In 4.4 and 4.9 kernels this results to:
> [ 354.179591] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 364.419674] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 374.663664] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 384.903717] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 395.143724] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 405.383645] unregister_netdevice: waiting for lo to become free. Usage count = 1
> ...
>
> Fixes: 410f03831 ("sctp: add routing output fallback")
> Fixes: 0ca50d12f ("sctp: fix src address selection if using secondary addresses")
> Signed-off-by: Tommi Rantala <tommi.t.rantala@nokia.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Thanks
> ---
>
> v2: unconditional dst_release() before breaking out of the loop:
> - if dst = NULL, then dst_release() is no-op.
> - if dst != &rt->dst, then drop reference to the first dst, we do not
> need it after all.
> - if dst = &rt->dst, then the refcnt was incremented twice for this
> dst, so drop the second ref.
>
> net/sctp/protocol.c | 10 ++++------
> 1 file changed, 4 insertions(+), 6 deletions(-)
>
> diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
> index ba28d270eb24..aba2381d3cf9 100644
> --- a/net/sctp/protocol.c
> +++ b/net/sctp/protocol.c
> @@ -509,22 +509,20 @@ static void sctp_v4_get_dst(struct sctp_transport *t, union sctp_addr *saddr,
> if (IS_ERR(rt))
> continue;
>
> - if (!dst)
> - dst = &rt->dst;
> -
> /* Ensure the src address belongs to the output
> * interface.
> */
> odev = __ip_dev_find(sock_net(sk), laddr->a.v4.sin_addr.s_addr,
> false);
> if (!odev || odev->ifindex != fl4->flowi4_oif) {
> - if (&rt->dst != dst)
> + if (!dst)
> + dst = &rt->dst;
> + else
> dst_release(&rt->dst);
> continue;
> }
>
> - if (dst != &rt->dst)
> - dst_release(dst);
> + dst_release(dst);
> dst = &rt->dst;
> break;
> }
> --
> 2.15.1
>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH net v3] sctp: fix dst refcnt leak in sctp_v4_get_dst
2018-02-05 19:48 ` Tommi Rantala
@ 2018-02-05 23:20 ` David Ahern
-1 siblings, 0 replies; 14+ messages in thread
From: David Ahern @ 2018-02-05 23:20 UTC (permalink / raw)
To: Tommi Rantala, netdev
Cc: linux-sctp, Neil Horman, Alexey Kodanev, Marcelo Ricardo Leitner
On 2/5/18 12:48 PM, Tommi Rantala wrote:
> Fix dst reference count leak in sctp_v4_get_dst() introduced in commit
> 410f03831 ("sctp: add routing output fallback"):
>
> When walking the address_list, successive ip_route_output_key() calls
> may return the same rt->dst with the reference incremented on each call.
>
> The code would not decrement the dst refcount when the dst pointer was
> identical from the previous iteration, causing the dst refcnt leak.
>
...
> ...
>
> Fixes: 410f03831 ("sctp: add routing output fallback")
> Fixes: 0ca50d12f ("sctp: fix src address selection if using secondary addresses")
any syzbot references for this bug?
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH net v3] sctp: fix dst refcnt leak in sctp_v4_get_dst
@ 2018-02-05 23:20 ` David Ahern
0 siblings, 0 replies; 14+ messages in thread
From: David Ahern @ 2018-02-05 23:20 UTC (permalink / raw)
To: Tommi Rantala, netdev
Cc: linux-sctp, Neil Horman, Alexey Kodanev, Marcelo Ricardo Leitner
On 2/5/18 12:48 PM, Tommi Rantala wrote:
> Fix dst reference count leak in sctp_v4_get_dst() introduced in commit
> 410f03831 ("sctp: add routing output fallback"):
>
> When walking the address_list, successive ip_route_output_key() calls
> may return the same rt->dst with the reference incremented on each call.
>
> The code would not decrement the dst refcount when the dst pointer was
> identical from the previous iteration, causing the dst refcnt leak.
>
...
> ...
>
> Fixes: 410f03831 ("sctp: add routing output fallback")
> Fixes: 0ca50d12f ("sctp: fix src address selection if using secondary addresses")
any syzbot references for this bug?
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH net v3] sctp: fix dst refcnt leak in sctp_v4_get_dst
2018-02-05 23:20 ` David Ahern
@ 2018-02-06 5:06 ` Xin Long
-1 siblings, 0 replies; 14+ messages in thread
From: Xin Long @ 2018-02-06 5:06 UTC (permalink / raw)
To: David Ahern
Cc: Tommi Rantala, network dev, linux-sctp, Neil Horman,
Alexey Kodanev, Marcelo Ricardo Leitner, Dmitry Vyukov
On Tue, Feb 6, 2018 at 7:20 AM, David Ahern <dsahern@gmail.com> wrote:
> On 2/5/18 12:48 PM, Tommi Rantala wrote:
>> Fix dst reference count leak in sctp_v4_get_dst() introduced in commit
>> 410f03831 ("sctp: add routing output fallback"):
>>
>> When walking the address_list, successive ip_route_output_key() calls
>> may return the same rt->dst with the reference incremented on each call.
>>
>> The code would not decrement the dst refcount when the dst pointer was
>> identical from the previous iteration, causing the dst refcnt leak.
>>
> ...
>> ...
>>
>> Fixes: 410f03831 ("sctp: add routing output fallback")
>> Fixes: 0ca50d12f ("sctp: fix src address selection if using secondary addresses")
>
> any syzbot references for this bug?
In Dmitry Vyukov mail, there was no syzbot reference provided.
Not sure if there's another way to tell syzbot.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-sctp" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH net v3] sctp: fix dst refcnt leak in sctp_v4_get_dst
@ 2018-02-06 5:06 ` Xin Long
0 siblings, 0 replies; 14+ messages in thread
From: Xin Long @ 2018-02-06 5:06 UTC (permalink / raw)
To: David Ahern
Cc: Tommi Rantala, network dev, linux-sctp, Neil Horman,
Alexey Kodanev, Marcelo Ricardo Leitner, Dmitry Vyukov
On Tue, Feb 6, 2018 at 7:20 AM, David Ahern <dsahern@gmail.com> wrote:
> On 2/5/18 12:48 PM, Tommi Rantala wrote:
>> Fix dst reference count leak in sctp_v4_get_dst() introduced in commit
>> 410f03831 ("sctp: add routing output fallback"):
>>
>> When walking the address_list, successive ip_route_output_key() calls
>> may return the same rt->dst with the reference incremented on each call.
>>
>> The code would not decrement the dst refcount when the dst pointer was
>> identical from the previous iteration, causing the dst refcnt leak.
>>
> ...
>> ...
>>
>> Fixes: 410f03831 ("sctp: add routing output fallback")
>> Fixes: 0ca50d12f ("sctp: fix src address selection if using secondary addresses")
>
> any syzbot references for this bug?
In Dmitry Vyukov mail, there was no syzbot reference provided.
Not sure if there's another way to tell syzbot.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-sctp" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH net v3] sctp: fix dst refcnt leak in sctp_v4_get_dst
2018-02-06 5:06 ` Xin Long
@ 2018-02-06 8:27 ` Dmitry Vyukov
-1 siblings, 0 replies; 14+ messages in thread
From: Dmitry Vyukov @ 2018-02-06 8:27 UTC (permalink / raw)
To: Xin Long
Cc: David Ahern, Tommi Rantala, network dev, linux-sctp, Neil Horman,
Alexey Kodanev, Marcelo Ricardo Leitner
On Tue, Feb 6, 2018 at 6:06 AM, Xin Long <lucien.xin@gmail.com> wrote:
> On Tue, Feb 6, 2018 at 7:20 AM, David Ahern <dsahern@gmail.com> wrote:
>> On 2/5/18 12:48 PM, Tommi Rantala wrote:
>>> Fix dst reference count leak in sctp_v4_get_dst() introduced in commit
>>> 410f03831 ("sctp: add routing output fallback"):
>>>
>>> When walking the address_list, successive ip_route_output_key() calls
>>> may return the same rt->dst with the reference incremented on each call.
>>>
>>> The code would not decrement the dst refcount when the dst pointer was
>>> identical from the previous iteration, causing the dst refcnt leak.
>>>
>> ...
>>> ...
>>>
>>> Fixes: 410f03831 ("sctp: add routing output fallback")
>>> Fixes: 0ca50d12f ("sctp: fix src address selection if using secondary addresses")
>>
>> any syzbot references for this bug?
> In Dmitry Vyukov mail, there was no syzbot reference provided.
> Not sure if there's another way to tell syzbot.
If we are talking about "net: hang in unregister_netdevice: waiting
for lo to become free":
https://groups.google.com/d/msg/syzkaller/-06_laheMF0/xqezy58kAwAJ
Then there is no syzbot tag. It was found with syzkaller, but not
reported by syzbot because the manifestation is too tricky, it could
have been reported as "no output from test machine" with no additional
details, which is not too actionable.
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH net v3] sctp: fix dst refcnt leak in sctp_v4_get_dst
@ 2018-02-06 8:27 ` Dmitry Vyukov
0 siblings, 0 replies; 14+ messages in thread
From: Dmitry Vyukov @ 2018-02-06 8:27 UTC (permalink / raw)
To: Xin Long
Cc: David Ahern, Tommi Rantala, network dev, linux-sctp, Neil Horman,
Alexey Kodanev, Marcelo Ricardo Leitner
On Tue, Feb 6, 2018 at 6:06 AM, Xin Long <lucien.xin@gmail.com> wrote:
> On Tue, Feb 6, 2018 at 7:20 AM, David Ahern <dsahern@gmail.com> wrote:
>> On 2/5/18 12:48 PM, Tommi Rantala wrote:
>>> Fix dst reference count leak in sctp_v4_get_dst() introduced in commit
>>> 410f03831 ("sctp: add routing output fallback"):
>>>
>>> When walking the address_list, successive ip_route_output_key() calls
>>> may return the same rt->dst with the reference incremented on each call.
>>>
>>> The code would not decrement the dst refcount when the dst pointer was
>>> identical from the previous iteration, causing the dst refcnt leak.
>>>
>> ...
>>> ...
>>>
>>> Fixes: 410f03831 ("sctp: add routing output fallback")
>>> Fixes: 0ca50d12f ("sctp: fix src address selection if using secondary addresses")
>>
>> any syzbot references for this bug?
> In Dmitry Vyukov mail, there was no syzbot reference provided.
> Not sure if there's another way to tell syzbot.
If we are talking about "net: hang in unregister_netdevice: waiting
for lo to become free":
https://groups.google.com/d/msg/syzkaller/-06_laheMF0/xqezy58kAwAJ
Then there is no syzbot tag. It was found with syzkaller, but not
reported by syzbot because the manifestation is too tricky, it could
have been reported as "no output from test machine" with no additional
details, which is not too actionable.
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH net v3] sctp: fix dst refcnt leak in sctp_v4_get_dst
2018-02-05 19:48 ` Tommi Rantala
@ 2018-02-06 1:17 ` Neil Horman
-1 siblings, 0 replies; 14+ messages in thread
From: Neil Horman @ 2018-02-06 1:17 UTC (permalink / raw)
To: Tommi Rantala; +Cc: netdev, linux-sctp, Alexey Kodanev, Marcelo Ricardo Leitner
On Mon, Feb 05, 2018 at 09:48:14PM +0200, Tommi Rantala wrote:
> Fix dst reference count leak in sctp_v4_get_dst() introduced in commit
> 410f03831 ("sctp: add routing output fallback"):
>
> When walking the address_list, successive ip_route_output_key() calls
> may return the same rt->dst with the reference incremented on each call.
>
> The code would not decrement the dst refcount when the dst pointer was
> identical from the previous iteration, causing the dst refcnt leak.
>
> Testcase:
> ip netns add TEST
> ip netns exec TEST ip link set lo up
> ip link add dummy0 type dummy
> ip link add dummy1 type dummy
> ip link add dummy2 type dummy
> ip link set dev dummy0 netns TEST
> ip link set dev dummy1 netns TEST
> ip link set dev dummy2 netns TEST
> ip netns exec TEST ip addr add 192.168.1.1/24 dev dummy0
> ip netns exec TEST ip link set dummy0 up
> ip netns exec TEST ip addr add 192.168.1.2/24 dev dummy1
> ip netns exec TEST ip link set dummy1 up
> ip netns exec TEST ip addr add 192.168.1.3/24 dev dummy2
> ip netns exec TEST ip link set dummy2 up
> ip netns exec TEST sctp_test -H 192.168.1.2 -P 20002 -h 192.168.1.1 -p 20000 -s -B 192.168.1.3
> ip netns del TEST
>
> In 4.4 and 4.9 kernels this results to:
> [ 354.179591] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 364.419674] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 374.663664] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 384.903717] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 395.143724] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 405.383645] unregister_netdevice: waiting for lo to become free. Usage count = 1
> ...
>
> Fixes: 410f03831 ("sctp: add routing output fallback")
> Fixes: 0ca50d12f ("sctp: fix src address selection if using secondary addresses")
> Signed-off-by: Tommi Rantala <tommi.t.rantala@nokia.com>
> ---
>
> v2: unconditional dst_release() before breaking out of the loop:
> - if dst == NULL, then dst_release() is no-op.
> - if dst != &rt->dst, then drop reference to the first dst, we do not
> need it after all.
> - if dst == &rt->dst, then the refcnt was incremented twice for this
> dst, so drop the second ref.
>
> net/sctp/protocol.c | 10 ++++------
> 1 file changed, 4 insertions(+), 6 deletions(-)
>
> diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
> index ba28d270eb24..aba2381d3cf9 100644
> --- a/net/sctp/protocol.c
> +++ b/net/sctp/protocol.c
> @@ -509,22 +509,20 @@ static void sctp_v4_get_dst(struct sctp_transport *t, union sctp_addr *saddr,
> if (IS_ERR(rt))
> continue;
>
> - if (!dst)
> - dst = &rt->dst;
> -
> /* Ensure the src address belongs to the output
> * interface.
> */
> odev = __ip_dev_find(sock_net(sk), laddr->a.v4.sin_addr.s_addr,
> false);
> if (!odev || odev->ifindex != fl4->flowi4_oif) {
> - if (&rt->dst != dst)
> + if (!dst)
> + dst = &rt->dst;
> + else
> dst_release(&rt->dst);
> continue;
> }
>
> - if (dst != &rt->dst)
> - dst_release(dst);
> + dst_release(dst);
> dst = &rt->dst;
> break;
> }
> --
> 2.15.1
>
>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH net v3] sctp: fix dst refcnt leak in sctp_v4_get_dst
@ 2018-02-06 1:17 ` Neil Horman
0 siblings, 0 replies; 14+ messages in thread
From: Neil Horman @ 2018-02-06 1:17 UTC (permalink / raw)
To: Tommi Rantala; +Cc: netdev, linux-sctp, Alexey Kodanev, Marcelo Ricardo Leitner
On Mon, Feb 05, 2018 at 09:48:14PM +0200, Tommi Rantala wrote:
> Fix dst reference count leak in sctp_v4_get_dst() introduced in commit
> 410f03831 ("sctp: add routing output fallback"):
>
> When walking the address_list, successive ip_route_output_key() calls
> may return the same rt->dst with the reference incremented on each call.
>
> The code would not decrement the dst refcount when the dst pointer was
> identical from the previous iteration, causing the dst refcnt leak.
>
> Testcase:
> ip netns add TEST
> ip netns exec TEST ip link set lo up
> ip link add dummy0 type dummy
> ip link add dummy1 type dummy
> ip link add dummy2 type dummy
> ip link set dev dummy0 netns TEST
> ip link set dev dummy1 netns TEST
> ip link set dev dummy2 netns TEST
> ip netns exec TEST ip addr add 192.168.1.1/24 dev dummy0
> ip netns exec TEST ip link set dummy0 up
> ip netns exec TEST ip addr add 192.168.1.2/24 dev dummy1
> ip netns exec TEST ip link set dummy1 up
> ip netns exec TEST ip addr add 192.168.1.3/24 dev dummy2
> ip netns exec TEST ip link set dummy2 up
> ip netns exec TEST sctp_test -H 192.168.1.2 -P 20002 -h 192.168.1.1 -p 20000 -s -B 192.168.1.3
> ip netns del TEST
>
> In 4.4 and 4.9 kernels this results to:
> [ 354.179591] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 364.419674] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 374.663664] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 384.903717] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 395.143724] unregister_netdevice: waiting for lo to become free. Usage count = 1
> [ 405.383645] unregister_netdevice: waiting for lo to become free. Usage count = 1
> ...
>
> Fixes: 410f03831 ("sctp: add routing output fallback")
> Fixes: 0ca50d12f ("sctp: fix src address selection if using secondary addresses")
> Signed-off-by: Tommi Rantala <tommi.t.rantala@nokia.com>
> ---
>
> v2: unconditional dst_release() before breaking out of the loop:
> - if dst = NULL, then dst_release() is no-op.
> - if dst != &rt->dst, then drop reference to the first dst, we do not
> need it after all.
> - if dst = &rt->dst, then the refcnt was incremented twice for this
> dst, so drop the second ref.
>
> net/sctp/protocol.c | 10 ++++------
> 1 file changed, 4 insertions(+), 6 deletions(-)
>
> diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
> index ba28d270eb24..aba2381d3cf9 100644
> --- a/net/sctp/protocol.c
> +++ b/net/sctp/protocol.c
> @@ -509,22 +509,20 @@ static void sctp_v4_get_dst(struct sctp_transport *t, union sctp_addr *saddr,
> if (IS_ERR(rt))
> continue;
>
> - if (!dst)
> - dst = &rt->dst;
> -
> /* Ensure the src address belongs to the output
> * interface.
> */
> odev = __ip_dev_find(sock_net(sk), laddr->a.v4.sin_addr.s_addr,
> false);
> if (!odev || odev->ifindex != fl4->flowi4_oif) {
> - if (&rt->dst != dst)
> + if (!dst)
> + dst = &rt->dst;
> + else
> dst_release(&rt->dst);
> continue;
> }
>
> - if (dst != &rt->dst)
> - dst_release(dst);
> + dst_release(dst);
> dst = &rt->dst;
> break;
> }
> --
> 2.15.1
>
>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH net v3] sctp: fix dst refcnt leak in sctp_v4_get_dst
2018-02-05 19:48 ` Tommi Rantala
@ 2018-02-06 2:22 ` David Miller
-1 siblings, 0 replies; 14+ messages in thread
From: David Miller @ 2018-02-06 2:22 UTC (permalink / raw)
To: tommi.t.rantala
Cc: netdev, linux-sctp, nhorman, alexey.kodanev, marcelo.leitner
From: Tommi Rantala <tommi.t.rantala@nokia.com>
Date: Mon, 5 Feb 2018 21:48:14 +0200
> Fix dst reference count leak in sctp_v4_get_dst() introduced in commit
> 410f03831 ("sctp: add routing output fallback"):
>
> When walking the address_list, successive ip_route_output_key() calls
> may return the same rt->dst with the reference incremented on each call.
>
> The code would not decrement the dst refcount when the dst pointer was
> identical from the previous iteration, causing the dst refcnt leak.
...
> Fixes: 410f03831 ("sctp: add routing output fallback")
> Fixes: 0ca50d12f ("sctp: fix src address selection if using secondary addresses")
> Signed-off-by: Tommi Rantala <tommi.t.rantala@nokia.com>
Applied and queued up for -stable, thanks.
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH net v3] sctp: fix dst refcnt leak in sctp_v4_get_dst
@ 2018-02-06 2:22 ` David Miller
0 siblings, 0 replies; 14+ messages in thread
From: David Miller @ 2018-02-06 2:22 UTC (permalink / raw)
To: tommi.t.rantala
Cc: netdev, linux-sctp, nhorman, alexey.kodanev, marcelo.leitner
From: Tommi Rantala <tommi.t.rantala@nokia.com>
Date: Mon, 5 Feb 2018 21:48:14 +0200
> Fix dst reference count leak in sctp_v4_get_dst() introduced in commit
> 410f03831 ("sctp: add routing output fallback"):
>
> When walking the address_list, successive ip_route_output_key() calls
> may return the same rt->dst with the reference incremented on each call.
>
> The code would not decrement the dst refcount when the dst pointer was
> identical from the previous iteration, causing the dst refcnt leak.
...
> Fixes: 410f03831 ("sctp: add routing output fallback")
> Fixes: 0ca50d12f ("sctp: fix src address selection if using secondary addresses")
> Signed-off-by: Tommi Rantala <tommi.t.rantala@nokia.com>
Applied and queued up for -stable, thanks.
^ permalink raw reply [flat|nested] 14+ messages in thread