From: ebiederm@xmission•com (Eric W. Biederman)
To: Eric Dumazet <eric.dumazet@gmail•com>
Cc: Stephen Hemminger <stephen@networkplumber•org>,
Benoit Lourdelet <blourdel@juniper•net>,
Serge Hallyn <serge.hallyn@ubuntu•com>,
"netdev\@vger.kernel.org" <netdev@vger•kernel.org>
Subject: Re: [RFC][PATCH] iproute: Faster ip link add, set and delete
Date: Thu, 28 Mar 2013 17:25:46 -0700 [thread overview]
Message-ID: <87ppyj6ohh.fsf@xmission.com> (raw)
In-Reply-To: <1364516016.15753.59.camel@edumazet-glaptop> (Eric Dumazet's message of "Thu, 28 Mar 2013 17:13:36 -0700")
Eric Dumazet <eric.dumazet@gmail•com> writes:
> On Thu, 2013-03-28 at 16:52 -0700, Eric W. Biederman wrote:
>
>> On my microbenchmark of just creating 5000 veth pairs this takes pairs
>> 16s instead of 13s of my earlier hacks but that is well down in the
>> usable range.
>
> I guess most of the time is taken by sysctl_check_table()
All of the significant sysctl slowdowns were fixed in 3.4. If you see
something of sysctl show up in a trace I would be happy to talk about
it. The kernel side seems to be creating N network devices seems to
take NlogN time now. Both sysfs and sysctl store directories as
rbtrees removing their previous bottlenecks.
The loop I timed at 16s was just:
time for i in $(seq 1 5000) ; do ip link add a$i type veth peer name b$i; done
There is plenty of room for inefficiencies in 10000 network devices and
5000 forks+execs.
Eric
next prev parent reply other threads:[~2013-03-29 0:25 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-22 22:23 [RFC][PATCH] iproute: Faster ip link add, set and delete Eric W. Biederman
2013-03-22 22:27 ` Stephen Hemminger
2013-03-26 11:51 ` Benoit Lourdelet
2013-03-26 12:40 ` Eric W. Biederman
2013-03-26 14:17 ` Serge Hallyn
2013-03-26 14:33 ` Serge Hallyn
2013-03-27 13:37 ` Benoit Lourdelet
2013-03-27 15:11 ` Eric W. Biederman
2013-03-27 17:47 ` Stephen Hemminger
2013-03-28 0:46 ` Eric W. Biederman
2013-03-28 3:20 ` Serge Hallyn
2013-03-28 3:44 ` Eric W. Biederman
2013-03-28 4:28 ` Serge Hallyn
2013-03-28 5:00 ` Eric W. Biederman
2013-03-28 13:36 ` Serge Hallyn
2013-03-28 13:42 ` Benoit Lourdelet
2013-03-28 15:04 ` Serge Hallyn
2013-03-28 15:21 ` Benoit Lourdelet
2013-03-28 22:20 ` Stephen Hemminger
2013-03-28 23:52 ` Eric W. Biederman
2013-03-29 0:13 ` Eric Dumazet
2013-03-29 0:25 ` Eric W. Biederman [this message]
2013-03-29 0:43 ` Eric Dumazet
2013-03-29 1:06 ` Eric W. Biederman
2013-03-29 1:10 ` Eric Dumazet
2013-03-29 1:29 ` Eric W. Biederman
2013-03-29 1:38 ` Eric Dumazet
2013-03-30 10:09 ` Benoit Lourdelet
2013-03-30 14:44 ` Eric Dumazet
2013-03-30 16:07 ` Benoit Lourdelet
2013-03-28 20:27 ` Benoit Lourdelet
2013-03-26 15:31 ` Eric Dumazet
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87ppyj6ohh.fsf@xmission.com \
--to=ebiederm@xmission$(echo .)com \
--cc=blourdel@juniper$(echo .)net \
--cc=eric.dumazet@gmail$(echo .)com \
--cc=netdev@vger$(echo .)kernel.org \
--cc=serge.hallyn@ubuntu$(echo .)com \
--cc=stephen@networkplumber$(echo .)org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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