From: David Miller <davem@davemloft.net>
To: alexander.duyck@gmail.com
Cc: eric.dumazet@gmail.com, netdev@vger.kernel.org
Subject: Re: [PATCH 00/16] Remove the ipv4 routing cache
Date: Wed, 25 Jul 2012 16:17:32 -0700 (PDT) [thread overview]
Message-ID: <20120725.161732.91008692477078715.davem@davemloft.net> (raw)
In-Reply-To: <CAKgT0UdcqE52WFCT-t2om0SN3Yq5GqiwN+JbwCcnBv4LDOB6BQ@mail.gmail.com>
From: Alexander Duyck <alexander.duyck@gmail.com>
Date: Wed, 25 Jul 2012 16:02:45 -0700
> Since your patches are in I have started to re-run my tests. I am
> seeing a significant drop in throughput with 8 flows which I expected,
> however it looks like one of the biggest issues I am seeing is that
> the dst_hold and dst_release calls seem to be causing some serious
> cache thrash. I was at 12.5Mpps w/ 8 flows before the patches, after
> your patches it drops to 8.3Mpps.
Yes, this is something we knew would start happening.
One idea is to make cached dsts be per-cpu in the nexthops.
> I am also seeing routing fail periodically.
Every 30 seconds by chance? :-)
> I will be moving at rates listed above and suddenly drop to single
> digits packets per second. When this occurs the trace completely
> changes and __write_lock_failed jumps to over 90% of the CPU cycles.
It's probably happening when the nexthop ARP entry expires.
next prev parent reply other threads:[~2012-07-25 23:17 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-07-20 21:25 [PATCH 00/16] Remove the ipv4 routing cache David Miller
2012-07-20 22:05 ` Eric Dumazet
2012-07-20 22:42 ` Eric Dumazet
2012-07-20 22:50 ` David Miller
2012-07-20 22:54 ` David Miller
2012-07-20 23:13 ` David Miller
2012-07-21 5:40 ` Eric Dumazet
2012-07-22 7:47 ` Vijay Subramanian
2012-07-22 19:42 ` David Miller
2012-07-23 0:39 ` David Miller
2012-07-23 7:15 ` Eric Dumazet
2012-07-23 17:54 ` Paweł Staszewski
2012-07-23 20:10 ` David Miller
2012-07-26 17:02 ` Eric Dumazet
2012-07-25 23:02 ` Alexander Duyck
2012-07-25 23:17 ` David Miller [this message]
2012-07-25 23:39 ` David Miller
2012-07-26 0:54 ` David Miller
2012-07-26 2:30 ` Alexander Duyck
2012-07-26 5:32 ` David Miller
2012-07-26 8:13 ` Eric Dumazet
2012-07-26 8:18 ` David Miller
2012-07-26 8:27 ` Eric Dumazet
2012-07-26 8:47 ` David Miller
2012-07-26 9:12 ` Eric Dumazet
2012-07-26 17:18 ` Alexander Duyck
2012-07-26 17:30 ` Eric Dumazet
2012-07-26 17:36 ` Eric Dumazet
2012-07-26 17:43 ` Eric Dumazet
2012-07-26 17:48 ` Eric Dumazet
2012-07-26 18:26 ` Alexander Duyck
2012-07-26 21:06 ` David Miller
2012-07-26 22:03 ` Alexander Duyck
2012-07-26 22:13 ` Stephen Hemminger
2012-07-26 22:19 ` Eric Dumazet
2012-07-26 22:48 ` David Miller
2012-07-26 22:53 ` David Miller
2012-07-27 2:14 ` Alexander Duyck
2012-07-27 3:08 ` David Miller
2012-07-27 6:02 ` David Miller
2012-07-27 10:01 ` Eric Dumazet
2012-07-27 14:53 ` Eric W. Biederman
2012-07-27 15:12 ` Eric Dumazet
2012-07-27 16:23 ` Eric W. Biederman
2012-07-27 16:28 ` Eric Dumazet
2012-07-27 19:06 ` Alexander Duyck
2012-07-28 4:15 ` David Miller
2012-07-28 5:45 ` Alexander Duyck
2012-07-26 18:06 ` Alexander Duyck
2012-07-26 21:00 ` David Miller
2012-07-26 20:59 ` David Miller
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=20120725.161732.91008692477078715.davem@davemloft.net \
--to=davem@davemloft.net \
--cc=alexander.duyck@gmail.com \
--cc=eric.dumazet@gmail.com \
--cc=netdev@vger.kernel.org \
/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).