netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Maciej Żenczykowski" <zenczykowski@gmail.com>
To: David Miller <davem@davemloft.net>
Cc: therbert@google.com, netdev@vger.kernel.org, lorenzo@google.com
Subject: Re: [PATCH] IPv6: fix rt_lookup in pmtu_discovery
Date: Thu, 7 Jan 2010 17:05:36 -0800	[thread overview]
Message-ID: <55a4f86e1001071705i33f8c58cubae56f5616216de4@mail.gmail.com> (raw)
In-Reply-To: <20100107.012701.257511338.davem@davemloft.net>

> This needs to do what the IPV4 side does, iterate over specific then
> "any" device index, doing a lookup for each case until the route is
> found, therefore starting from more specific and going towards less
> specific routes.

I've spoken with Tom and we can't quite seem to figure out what
exactly the code should be attempting to accomplish here.
Is checking the specific device index meant to deal with link local IPs?

As for the v4 code, I assume you're referring to ip_rt_frag_needed in
net/ipv4/route.c.

If so, shouldn't this sort of route lookup be abstracted away into
some function?
Path mtu discovery / fragmentation handling functions don't seem to be
the right place to be implementing route lookup policy.

Are you suggesting the following logic for ipv6:

rt = rt6_lookup(net, daddr, saddr, dev->ifindex, 0);
if (rt == NULL)
        rt = rt6_lookup(net, daddr, saddr, 0, 0);
if (rt == NULL)
        rt = rt6_lookup(net, daddr, 0, dev->ifindex, 0);
if (rt == NULL)
        rt = rt6_lookup(net, daddr, 0, 0, 0);
if (rt == NULL)
        return;

It's not clear to me what the last two lookups with saddr replaced
with 0 are for.

Maciej

  reply	other threads:[~2010-01-08  1:05 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-01-07  4:43 [PATCH] IPv6: fix rt_lookup in pmtu_discovery Tom Herbert
2010-01-07  9:27 ` David Miller
2010-01-08  1:05   ` Maciej Żenczykowski [this message]
2010-01-08  1:10     ` David Miller
2010-01-09  0:12       ` Lorenzo Colitti
2010-01-10 21:15         ` David Miller
2010-01-14  0:51           ` Maciej Żenczykowski
2010-01-20 22:55             ` Maciej Żenczykowski
2010-01-20 22:57               ` David Miller
2010-01-20 23:33                 ` Maciej Żenczykowski
2010-01-23 10:20             ` David Miller
2010-09-27 10:05             ` [PATCH] net: Fix IPv6 PMTU disc. w/ asymmetric routes Maciej Żenczykowski
     [not found]               ` <AANLkTikPOHy79E1ZG=iJ-rHj0vzS+AY-mGqCEtWoXp2o@mail.gmail.com>
2010-09-27 18:11                 ` David Miller
2010-09-28 20:58               ` David Miller
2010-09-28 22:37                 ` Maciej Żenczykowski
2010-09-30  7:41                   ` David Miller
2010-10-03 21:49                     ` David Miller
2010-10-04  0:21                       ` Maciej Żenczykowski

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=55a4f86e1001071705i33f8c58cubae56f5616216de4@mail.gmail.com \
    --to=zenczykowski@gmail.com \
    --cc=davem@davemloft.net \
    --cc=lorenzo@google.com \
    --cc=netdev@vger.kernel.org \
    --cc=therbert@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).