netdev - Re: [PATCH net-next] openvswitch: fix vport-netdev unregister

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Thu, 10 Oct 2013 13:48:31 -0700
From:	Alexei Starovoitov <ast@...mgrid.com>
To:	Pravin Shelar <pshelar@...ira.com>
Cc:	"David S. Miller" <davem@...emloft.net>,
	Jesse Gross <jesse@...ira.com>, Jiri Pirko <jiri@...nulli.us>,
	"dev@...nvswitch.org" <dev@...nvswitch.org>,
	netdev <netdev@...r.kernel.org>
Subject: Re: [PATCH net-next] openvswitch: fix vport-netdev unregister

On Thu, Oct 10, 2013 at 11:21 AM, Pravin Shelar <pshelar@...ira.com> wrote:
> On Wed, Oct 9, 2013 at 11:26 PM, Alexei Starovoitov <ast@...mgrid.com> wrote:
>> On Wed, Oct 9, 2013 at 11:07 PM, Pravin Shelar <pshelar@...ira.com> wrote:
>>> On Wed, Oct 9, 2013 at 9:11 PM, Alexei Starovoitov <ast@...mgrid.com> wrote:
>>>> On Wed, Oct 9, 2013 at 8:02 PM, Pravin Shelar <pshelar@...ira.com> wrote:
>>>>> On Tue, Oct 8, 2013 at 8:07 PM, Alexei Starovoitov <ast@...mgrid.com> wrote:
>>>>>> The combination of two commits
>>>>>>
>>>>>> commit 8e4e1713e4
>>>>>> ("openvswitch: Simplify datapath locking.")
>>>>>>
>>>>>> and
>>>>>>
>>>>>> commit 2537b4dd0a
>>>>>> ("openvswitch:: link upper device for port devices")
>>>>>>
>>>>>> introduced a bug where upper_dev wasn't unlinked upon
>>>>>> netdev_unregister notification
>>>>>>
>>>>>> The following steps:
>>>>>>
>>>>>>   modprobe openvswitch
>>>>>>   ovs-dpctl add-dp test
>>>>>>   ip tuntap add dev tap1 mode tap
>>>>>>   ovs-dpctl add-if test tap1
>>>>>>   ip tuntap del dev tap1 mode tap
>>>>>>
>>>>>> are causing multiple warnings:
>>>>>> diff --git a/net/openvswitch/dp_notify.c b/net/openvswitch/dp_notify.c
>>>>>> index c323567..e9380bd 100644
>>>>>> --- a/net/openvswitch/dp_notify.c
>>>>>> +++ b/net/openvswitch/dp_notify.c
>>>>>> @@ -88,6 +88,11 @@ static int dp_device_event(struct notifier_block *unused, unsigned long event,
>>>>>>                 return NOTIFY_DONE;
>>>>>>
>>>>>>         if (event == NETDEV_UNREGISTER) {
>>>>>> +               /* rx_handler_unregister and upper_dev_unlink immediately */
>>>>>> +               if (dev->reg_state == NETREG_UNREGISTERING)
>>>>>> +                       ovs_netdev_unlink_dev(vport);
>>>>>> +
>>>>>
>>>>> Rather than doing vport destroy here, we can just unlink upper device
>>>>> and let workq do rest of work.
>>>>
>>>> isn't it what it's doing?
>>>
>>> I meant just call netdev_upper_dev_unlink() here in event handler and
>>> rest of vport destroy can be done in workq.
>>
>> netdev_upper_dev_unlink() without netdev_rx_handler_unregister() ?!
>> that's dangerous.
> why is it dangerous? ovs already had ref to net-device.

comment from dev.c:
                /* Notify protocols, that we are about to destroy
                   this device. They should clean all the things.
                */
                call_netdevice_notifiers(NETDEV_UNREGISTER, dev);

so here you're suggesting to just netdev_upper_dev_unlink() to silence
the warning,
but then do dev_set_promisc(-1) in workqueue?

well, as a minimum audit of promiscuity will be wrong.
ndo_change_rx_flags will be called after ndo_uninit,
all sorts of other cleanups are done.
I cannot track all possible scenarios, but it seems much safer to
cleanup everything possible
as soon as ovs received NETDEV_UNREGISTER event.

May be all these risks are worth taking, then please explain what is
the problem with the proposed patch?

Thanks
Alex
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html