From: Patrick McHardy <kaber@trash.net>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Changli Gao <xiaosuo@gmail.com>,
Netfilter Development Mailinglist
<netfilter-devel@vger.kernel.org>
Subject: Re: RFC: netfilter: cache dst_entry in conntrack
Date: Mon, 08 Nov 2010 08:02:56 +0100 [thread overview]
Message-ID: <4CD7A0A0.5010905@trash.net> (raw)
In-Reply-To: <1289199090.2478.326.camel@edumazet-laptop>
On 08.11.2010 07:51, Eric Dumazet wrote:
> Le lundi 08 novembre 2010 à 14:44 +0800, Changli Gao a écrit :
>> On Mon, Nov 8, 2010 at 2:39 PM, Patrick McHardy <kaber@trash.net> wrote:
>>> [resend with netfilter-devel address fixed]
>>>
>>> On 08.11.2010 07:36, Patrick McHardy wrote:
>>>> On 08.11.2010 07:32, Changli Gao wrote:
>>>>> When conntrack is enabled, we can cache dst_entry into the
>>>>> corresponding conntrack to eliminate the subsequence
>>>>> ip_route_input_noref() calls for the same connection. The current
>>>>> implementation is a standalone module. If this idea is welcomed, I'll
>>>>> try to push it up. Thanks.
>>>>>
>>>>> Here it is: https://github.com/xiaosuo/xiaosuo/tree/master/nf_rtcache/
>>>>
>>>> Please send the patch as attachment so it can be discussed.
>>>
>>>
>>
>> Here is a trivial patch to add RTCACHE NF extention.
>>
>>
>> ...
> Adding yet another level of cache seems wrong to me.
>
> Either we fix the first level, or we dont use it at all.
>
> If ip_route_input_noref() is slow, we can work on it to make it faster.
>
> I know dst_use_noref() is the real problem an can be optimized if
> needed.
I think the main idea is to make double use of the existing conntrack
lookup. The most interesting part about this idea in my opinion is
that it should make multipath setups very easy since you don't have
to manually deal with expiring routing cache entries anymore. However
this clearly is a big change to the existing behaviour of the routing
cache and needs a lot more thought, which I will perform after having
some coffee :)
--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2010-11-08 7:03 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <AANLkTimVY870t+ugopoabpe3+QCDC9RFh0nCtgPmJLq8@mail.gmail.com>
2010-11-08 6:34 ` Fwd: RFC: netfilter: cache dst_entry in conntrack Changli Gao
2010-11-08 9:45 ` "Oleg A. Arkhangelsky"
2010-11-08 10:34 ` Jan Engelhardt
2010-11-08 14:29 ` Changli Gao
[not found] ` <4CD79A79.3030103@trash.net>
2010-11-08 6:39 ` Patrick McHardy
2010-11-08 6:44 ` Changli Gao
2010-11-08 6:51 ` Eric Dumazet
2010-11-08 7:02 ` Patrick McHardy [this message]
2010-11-08 7:23 ` Eric Dumazet
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=4CD7A0A0.5010905@trash.net \
--to=kaber@trash.net \
--cc=eric.dumazet@gmail.com \
--cc=netfilter-devel@vger.kernel.org \
--cc=xiaosuo@gmail.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.