From: stephen mulcahy <smulcahy@gmail.com>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Ben Hutchings <bhutchings@solarflare.com>,
netdev <netdev@vger.kernel.org>,
Ben Hutchings <ben@decadent.org.uk>,
Ayaz Abdulla <aabdulla@nvidia.com>,
572201@bugs.debian.org
Subject: Re: forcedeth driver hangs under heavy load
Date: Tue, 13 Apr 2010 16:00:25 +0100 [thread overview]
Message-ID: <4BC48709.7060600@gmail.com> (raw)
In-Reply-To: <4BC48460.4040001@gmail.com>
stephen mulcahy wrote:
>> Now some brave fouls to check the 6410 lines of this driver ? ;)
>>
>> Question of the day : Why TSO is broken in forcedeth ?
>> Is it generically broken or is it broken for specific NICS ?
>>
>
> Actually, it is only when tx-checksumming is turned off that the problem
> doesn't occur (so I'm not sure TSO is the problem).
>
> Additionally, a google also turns up this existing Debian bug
> http://bugs.debian.org/506419 which seems to be related.
As mentioned in the original Debian bug - I can reproduce this by
running Hadoop[1] TeraSort[2] but I haven't identified a simpler
reproducer. I tried to recreate this with iperf and ping -f but neither
helped - it may be that the problem only occurs when systems are passing
large amounts of traffic and have very high cpu utilisation (when
running the Hadoop TeraSort all 8 cores run at 70-100% utilisation as
measure with htop - I plan to instrument the nodes with something like
Zabbix or Ganglia but it hasn't happened yet).
-stephen
[1] http://hadoop.apache.org/
[2]
http://hadoop.apache.org/common/docs/current/api/org/apache/hadoop/examples/terasort/package-summary.html
next prev parent reply other threads:[~2010-04-13 15:00 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <4B9E6C60.7030300@atlanticlinux.ie>
[not found] ` <20100315182220.GQ2763@decadent.org.uk>
[not found] ` <4B9F5E5E.2060209@atlanticlinux.ie>
[not found] ` <1270393967.8341.11.camel@localhost>
[not found] ` <4BBCA19C.5080204@atlanticlinux.ie>
2010-04-10 23:36 ` forcedeth driver hangs under heavy load Ben Hutchings
2010-04-12 10:01 ` Bug#572201: " stephen mulcahy
2010-04-12 12:39 ` stephen mulcahy
2010-04-12 12:47 ` Eric Dumazet
2010-04-12 13:05 ` stephen mulcahy
2010-04-12 13:19 ` stephen mulcahy
2010-04-12 15:24 ` Eric Dumazet
2010-04-12 16:11 ` stephen mulcahy
2010-04-12 16:59 ` Eric Dumazet
2010-04-13 10:03 ` stephen mulcahy
2010-04-13 10:49 ` Eric Dumazet
2010-04-13 11:00 ` stephen mulcahy
2010-04-13 12:04 ` Ben Hutchings
2010-04-13 14:27 ` stephen mulcahy
2010-04-13 14:42 ` Eric Dumazet
2010-04-13 14:49 ` stephen mulcahy
2010-04-13 15:00 ` stephen mulcahy [this message]
2010-04-13 15:05 ` Eric Dumazet
2010-04-13 15:08 ` stephen mulcahy
2010-04-13 15:22 ` Eric Dumazet
2010-04-13 15:25 ` stephen mulcahy
2010-04-13 20:01 ` Eric Dumazet
2010-04-13 21:43 ` David Miller
2010-04-13 21:46 ` Eric Dumazet
2010-04-14 5:33 ` Bug#572201: " Ayaz Abdulla
2010-04-14 1:41 ` David Miller
2010-04-14 14:30 ` stephen mulcahy
2010-04-14 5:31 ` Bug#572201: [PATCH] forcedeth: fix tx limit2 flag check Ayaz Abdulla
2010-04-14 10:14 ` stephen mulcahy
2010-04-13 17:22 forcedeth driver hangs under heavy load Xose Vazquez Perez
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=4BC48709.7060600@gmail.com \
--to=smulcahy@gmail.com \
--cc=572201@bugs.debian.org \
--cc=aabdulla@nvidia.com \
--cc=ben@decadent.org.uk \
--cc=bhutchings@solarflare.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 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.