From mboxrd@z Thu Jan  1 00:00:00 1970
From: Hangbin Liu <liuhangbin@gmail.com>
Subject: Re: [PATCH net] ipv6: use rt6_info members when dst is set in
 rt6_fill_node
Date: Tue, 11 Sep 2018 11:41:06 +0800
Message-ID: <20180911034106.GB24677@leo.usersys.redhat.com>
References: <840f0c017c670ae9cc8fd8330ca24fcca1207918.1536398641.git.lucien.xin@gmail.com>
 <4abf5ec6-862b-f88e-b0ce-ddcd4308fd59@cumulusnetworks.com>
 <CADvbK_fgFM+VZ=kew4QkuM1xP90T2rWetXo3Awu48AEjJ+nvkg@mail.gmail.com>
 <b13be408-a734-e959-3299-bdd49a1318e7@cumulusnetworks.com>
 <CADvbK_fJCA9PUqzsuFjVYgMtfM1tvtq32=OU90PPykq8CPP=PA@mail.gmail.com>
 <c519ac67-a32e-4b68-76d4-49378e1440df@cumulusnetworks.com>
 <20180911010449.GA24677@leo.usersys.redhat.com>
 <879245ec-fd99-3b7f-6fc2-2d8d25cf2c57@cumulusnetworks.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: Xin Long <lucien.xin@gmail.com>,
        network dev <netdev@vger.kernel.org>,
        davem <davem@davemloft.net>,
        Roopa Prabhu <roopa@cumulusnetworks.com>
To: David Ahern <dsa@cumulusnetworks.com>
Return-path: <netdev-owner@vger.kernel.org>
Received: from mail-pl1-f194.google.com ([209.85.214.194]:35963 "EHLO
        mail-pl1-f194.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1726251AbeIKIib (ORCPT
        <rfc822;netdev@vger.kernel.org>); Tue, 11 Sep 2018 04:38:31 -0400
Received: by mail-pl1-f194.google.com with SMTP id p5-v6so3208697plk.3
        for <netdev@vger.kernel.org>; Mon, 10 Sep 2018 20:41:17 -0700 (PDT)
Content-Disposition: inline
In-Reply-To: <879245ec-fd99-3b7f-6fc2-2d8d25cf2c57@cumulusnetworks.com>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

Hi David,
On Mon, Sep 10, 2018 at 08:39:34PM -0600, David Ahern wrote:
> On 9/10/18 7:04 PM, Hangbin Liu wrote:
> 
> > diff --git a/net/ipv6/route.c b/net/ipv6/route.c
> > index 18e00ce..62621b4 100644
> > --- a/net/ipv6/route.c
> > +++ b/net/ipv6/route.c
> > @@ -998,6 +998,21 @@ static void ip6_rt_copy_init(struct rt6_info *rt, struct fib6_info *ort)
> >  	rt->rt6i_prefsrc = ort->fib6_prefsrc;
> >  }
> >  
> > +static void rt6_update_info(struct rt6_info *rt)
> > +{
> > +	struct fib6_info *from;
> > +
> > +	rcu_read_lock();
> > +	from = rcu_dereference(rt->from);
> > +	fib6_info_hold(from);
> > +	rcu_read_unlock();
> > +
> > +	from->fib6_flags = rt->rt6i_flags;
> > +	from->fib6_nh.nh_gw = rt->rt6i_gateway;
> 
> As I mentioned on your last patch, redirects do *not* update fib
> entries. Exceptions, yes, but not core data of a fib entry.

Thanks for the comments, I understand that we should not update original route.
And here I know the redirect (should?) do not update fib entries.

So Xin Long's patch looks more reasonable.

Thanks
Hangbin