netdev - Re: [PATCH RFC] netns: introduce NETREG_NETNS_MOVING reg

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]

Message-ID: <20100826082145.GD235835@jupiter.n2.diac24.net>
Date:	Thu, 26 Aug 2010 10:21:45 +0200
From:	David Lamparter <equinox@...c24.net>
To:	"Eric W. Biederman" <ebiederm@...ssion.com>
Cc:	David Lamparter <equinox@...c24.net>,
	Daniel Lezcano <daniel.lezcano@...e.fr>,
	netdev@...r.kernel.org, Patrick McHardy <kaber@...sh.net>
Subject: Re: [PATCH RFC] netns: introduce NETREG_NETNS_MOVING reg_state

On Wed, Aug 25, 2010 at 10:39:01AM -0700, Eric W. Biederman wrote:
> David Lamparter <equinox@...c24.net> writes:
> > No; since NETDEV_UNREGISTER is still sent in all places where it is
> > currently sent, none of the stacks would require changing. However,
> > users of the actual, physical device (independent of its network
> > namespace) would need changing to NETDEV_UNREGISTER_PHYSICAL, which is a
> > new notification that is only sent if the device really disappears.
> >
> > TBH, I'm not quite sure which is the better solution here;
> > NETDEV_UNREGISTER_PHYSICAL or NETREG_NETNS_MOVING...
> > ... or the initial one?
> 
> Before we get too far down any of these paths, what is it that caused
> you to notice that moving the physical device broke the connection to
> the vlan and macvlan devices?
> 
> I just want to understand this case a little better before we walk down
> any of these paths.

I'm doing system configuration software that allows you to manage
interfaces on a virtual routing platform. I can work around the
vlan-disappearings, but they kind of disrupt operation, especially as
they can affect namespaces not even related to what the user is changing
in the config.

If nothing else, I'd try to push my original patch. It is pretty
nonintrusive and should be OK with maybe a few more lines of comments.


David

--
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