[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Fri, 8 Aug 2014 03:13:58 +0000
From: Dexuan Cui <decui@...rosoft.com>
To: Richard Weinberger <richard.weinberger@...il.com>,
David Miller <davem@...emloft.net>,
"Yue Zhang (OSTC DEV)" <yuezha@...rosoft.com>
CC: "olaf@...fle.de" <olaf@...fle.de>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"driverdev-devel@...uxdriverproject.org"
<driverdev-devel@...uxdriverproject.org>,
LKML <linux-kernel@...r.kernel.org>,
Greg KH <gregkh@...uxfoundation.org>,
"jasowang@...hat.com" <jasowang@...hat.com>,
Haiyang Zhang <haiyangz@...rosoft.com>,
KY Srinivasan <kys@...rosoft.com>,
Thomas Shao <huishao@...rosoft.com>
Subject: RE: [PATCH] Hyperv: Trigger DHCP renew after host hibernation
> -----Original Message-----
> From: Richard Weinberger [mailto:richard.weinberger@...il.com]
> Sent: Friday, August 8, 2014 6:37 AM
> To: David Miller; Yue Zhang (OSTC DEV)
> Cc: olaf@...fle.de; netdev@...r.kernel.org; driverdev-
> devel@...uxdriverproject.org; LKML; Greg KH; jasowang@...hat.com;
> Haiyang Zhang; KY Srinivasan; Thomas Shao; Dexuan Cui
> Subject: Re: [PATCH] Hyperv: Trigger DHCP renew after host hibernation
>
> On Mon, Jul 21, 2014 at 11:32 PM, David Miller <davem@...emloft.net>
> wrote:
> > From: Olaf Hering <olaf@...fle.de>
> > Date: Mon, 21 Jul 2014 11:18:51 +0200
> >
> >> On Mon, Jul 21, Richard Weinberger wrote:
> >>
> >>> My concern is that 10 seconds is maybe not a the right choice.
> >>> (As we cannot know all implementations)
> >>
> >> Until someone reports an issue with it, 10 is fine. Just like 20 or 666.
> >
> > Wrong, this is policy and belongs in userspace.
>
> The "/etc/init.d/network restart" nonsense now hit Linus' tree.
> Yue, what is your proposal to fix that?
>
> //richard
Hi Richard and all,
Sorry for the late response -- actually we have been trying to
figure out a solution that's acceptable to all.
IMO the most feasible and need-the-least-change solution may be:
the hyperv network VSC driver passes the event
RNDIS_STATUS_NETWORK_CHANGE to the udev daemon?
In this way, every distro only needs to add a udev rule, which should
be simple.
Any comment?
-- Dexuan
Powered by blists - more mailing lists