netdev - Re: [PATCH net] macvlan: Fix the bug when destroy macvlan dev port

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, 14 Aug 2014 20:32:43 +0000
From:	"Keller, Jacob E" <jacob.e.keller@...el.com>
To:	"eric.dumazet@...il.com" <eric.dumazet@...il.com>
CC:	"dingtianhong@...wei.com" <dingtianhong@...wei.com>,
	"davem@...emloft.net" <davem@...emloft.net>,
	"cwang@...pensource.com" <cwang@...pensource.com>,
	"kaber@...sh.net" <kaber@...sh.net>,
	"netdev@...r.kernel.org" <netdev@...r.kernel.org>
Subject: Re: [PATCH net] macvlan: Fix the bug when destroy macvlan dev port

On Thu, 2014-08-14 at 10:38 -0700, Eric Dumazet wrote:
> On Thu, 2014-08-14 at 09:58 -0700, Cong Wang wrote:
> > On Thu, Aug 14, 2014 at 2:12 AM, Ding Tianhong <dingtianhong@...wei.com> wrote:
> > >
> > > To fix this problem, we need to check whether the port is exist when calling
> > > macvlan_port_destroy(), do nothing if the port has been destroyed yet.
> > >
> > 
> > As I said, this will make the first call of macvlan_port_destroy()
> > free the port,
> > which was freed by the last one before.
> > 
> > Why not just revert your change? It doesn't give any benefit.
> > --
> 
> Yes please, I think patch should be reverted.
> 
> 

As far as I can tell from the original patch, the only advantage it has
is the removal of a single int from a structure size. I'm not sure
that's very advantageous.

The other fix I saw might work is move the list_del so that vlans is
emptied inside uninit instead of dellink. This would be the same place
that count was decremented. However, I think this has many potential
side effects, and is not worth the saving of a single integer's worth of
space from a structure.

Thanks,
Jake