[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20150918183957.GC18191@tuxdriver.com>
Date: Fri, 18 Sep 2015 14:39:57 -0400
From: "John W. Linville" <linville@...driver.com>
To: Jesse Gross <jesse@...ira.com>
Cc: Eric Dumazet <eric.dumazet@...il.com>,
netdev <netdev@...r.kernel.org>,
Stephen Hemminger <stephen@...workplumber.org>,
Pravin B Shelar <pshelar@...ira.com>
Subject: Re: [PATCH] iplink_geneve: add UDP destination port configuration at
link creation
On Fri, Sep 18, 2015 at 09:15:56AM -0700, Jesse Gross wrote:
> On Fri, Sep 18, 2015 at 7:45 AM, Eric Dumazet <eric.dumazet@...il.com> wrote:
> > On Fri, 2015-09-18 at 10:26 -0400, John W. Linville wrote:
> >> On Thu, Sep 17, 2015 at 01:49:49PM -0700, Eric Dumazet wrote:
> >> > On Thu, 2015-09-17 at 15:27 -0400, John W. Linville wrote:
> >> > > Signed-off-by: John W. Linville <linville@...driver.com>
> >> > > ---
> >> >
> >> > > }
> >> > >
> >> > > @@ -150,6 +159,10 @@ static void geneve_print_opt(struct link_util *lu, FILE *f, struct rtattr *tb[])
> >> > > else
> >> > > fprintf(f, "tos %#x ", tos);
> >> > > }
> >> > > +
> >> > > + if (tb[IFLA_GENEVE_PORT])
> >> > > + fprintf(f, "dstport %u ",
> >> > > + ntohs(rta_getattr_u16(tb[IFLA_GENEVE_PORT])));
> >> >
> >> > This looks strange.
> >> >
> >> > Kernel does :
> >> >
> >> > if (nla_put_u16(skb, IFLA_GENEVE_PORT, ntohs(geneve->dst_port)))
> >> > goto nla_put_failure;
> >>
> >> Indeed, you are right. I had essentially copied some vxlan code when
> >> I did my version of adding the port attribute, and didn't take much
> >> care when adapting that code for the version that actually got merged.
> >>
> >> The current geneve code is using host byte-order for the UDP port in
> >> the netlink messages. But, I see that vxlan, gre, iptnl, etc are using
> >> network byte order for specifying UDP ports in their netlink stuff.
> >> Should geneve follow that practice as well? Or does it matter?
> >
> > It might be too late to change the ABI, as geneve is in linux-4.2
>
> We might be in luck actually. Geneve itself came earlier but I think
> the ability to specify the dest port is in 4.3 only, so we still have
> a chance to make it consistent.
Right, that's what I was thinking. So we are agreed to use network
byte order? I'll be happy to post a patch.
John
--
John W. Linville Someday the world will need a hero, and you
linville@...driver.com might be all we have. Be ready.
--
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
Powered by blists - more mailing lists