netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Ding Tianhong <dingtianhong@huawei.com>
To: David Miller <davem@davemloft.net>
Cc: <ben@decadent.org.uk>, <john.r.fastabend@intel.com>,
	<kaber@trash.net>, <netdev@vger.kernel.org>,
	<f.fainelli@gmail.com>
Subject: Re: [PATCH net-next] vlan: Fix lockdep warning when vlan dev handle notification
Date: Fri, 7 Mar 2014 09:55:25 +0800	[thread overview]
Message-ID: <5319270D.4030602@huawei.com> (raw)
In-Reply-To: <20140306.145126.497070034648321818.davem@davemloft.net>

On 2014/3/7 3:51, David Miller wrote:
> From: Ding Tianhong <dingtianhong@huawei.com>
> Date: Thu, 6 Mar 2014 22:13:33 +0800
> 
>> When I open the LOCKDEP config and run these steps:
>>
>> modprobe 8021q
>> vconfig add eth2 20
>> vconfig add eth2.20 30
>> ifconfig eth2 xx.xx.xx.xx
>>
>> then the Call Trace happened:
>  ...
>> The reason is that all of the vlan dev have the same class key for dev_lock_list,
>> if we up or down the real dev, the notification will change the state for every
>> vlan dev in the vlan group, then the vlan dev will hold netif_addr_lock and the
>> real dev also hold its own netif_addr_lock together, so the warning happened.
>>
>> The best way to fix the problem is add a new class key for the vlan dev if its
>> real dev is a vlan dev too.
>>
>> Signed-off-by: Ding Tianhong <dingtianhong@huawei.com>
> 
> So what happens when one does:
> 
> vconfig add eth2 20
> vconfig add eth2.20 30
> vconfig add eth2.30 40
> ifconfig eth2 xx.xx.xx.xx
> 
> 

the old vlan list:

eth2---------------------->eth2.20-------------------->eth2.20.30 (xxx_addr_lock_key)
			(xxx_addr_lock_key)|
					    ---------->eth2.20.50 (xxx_addr_lock_key)

after this patch:

eth2---------------------->eth2.20-------------------->eth2.20.30 (xxx_addr_lock_subkey)
			(xxx_addr_lock_key)|
					    ---------->eth2.20.50 (xxx_addr_lock_subkey)


The eth2.20 will only sync mc ddress with eth2.20.30 or eth2.20.50, and
the eth2.20.30 would never sync mc with eth2.20.50, so the xxx_addr_lock_subkey
will not meet each other.

But I think if I vconfig add eth2.20.30 50, the warning would happen again, but till now
I could not find any scene that use 3 vlan id on a virtual device, so I think it is not a 
problem.

  reply	other threads:[~2014-03-07  1:55 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-03-06 14:13 [PATCH net-next] vlan: Fix lockdep warning when vlan dev handle notification Ding Tianhong
2014-03-06 19:51 ` David Miller
2014-03-07  1:55   ` Ding Tianhong [this message]
2014-03-07  6:49     ` David Miller
2014-03-10  2:08       ` Ding Tianhong

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5319270D.4030602@huawei.com \
    --to=dingtianhong@huawei.com \
    --cc=ben@decadent.org.uk \
    --cc=davem@davemloft.net \
    --cc=f.fainelli@gmail.com \
    --cc=john.r.fastabend@intel.com \
    --cc=kaber@trash.net \
    --cc=netdev@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).