From: Stephen Hemminger <shemminger@vyatta.com>
To: Daniel Hazelton <dhazelton@enter.net>
Cc: Berck Nash <flyboy@gmail.com>,
Andrew Morton <akpm@linux-foundation.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
netdev@vger.kernel.org
Subject: Re: sky2 panic in 2.6.32.1 under load
Date: Thu, 24 Dec 2009 14:21:46 -0800 [thread overview]
Message-ID: <20091224142146.700e4ac8@nehalam> (raw)
In-Reply-To: <200912241128.57599.dhazelton@enter.net>
On Thu, 24 Dec 2009 11:28:57 -0500
Daniel Hazelton <dhazelton@enter.net> wrote:
> On Thursday 24 December 2009 11:03:56 am Berck Nash wrote:
> > Andrew Morton wrote:
> > > On Mon, 21 Dec 2009 16:52:10 -0700 "Berck E. Nash" <flyboy@gmail.com>
> wrote:
> > >> Since 2.6.32, I've been getting kernel panics under heavy network load
> > >> (bittorrent usage).
> > >
> > > Let's cc the right list and developer.
> > >
> > > This is a 2.6.31->2.6.32 regression?
> >
> > I believe so. Since it's intermittent and difficult to reproduce, it's
> > possible (but unlikely) that I simply never triggered it under 2.6.31.
>
> This is far from new. I have seen this under 2.6.27 when at least one botnet
> has been pointed at a server of mine and told to gain access. It has happened
> four times in the last six to eight months - and I have no easy way to capture
> the logs. But the oops that was posted looks very, very similar to what I've
> seen.
>
> It's always an allocation error in the transmit path that leads to the panic.
> Because this is a production machine that I do not have a way to take down and
> do testing with I've not reported the problem before.
>
Even though I wrote/maintain the sky driver, I don't work for SysKonnect, and
only have access to a limited set of information:
the technical manuals (under NDA), and the vendor sk98lin driver. The sky2 driver
imitates the receiver timeout of the sk98lin driver; other people have told me
that the FIFO hardware implementation is buggy and when it gets full, it gets
stuck. Probably the equivalent of a software FIFO where the developer forgets to
reserve a slot so that head == tail can mean both empty and full!
The workaround with a timer is prone to errors when traffic keeps going, also
the vendor doesn't really provide clear instructions on how to unlock it. I do
not have access to the hardware errata describing the problem. If I did a more
minimal solution would be possible.
The easiest advice is avoid sky2 chips with FIFO for any heavy traffic, the next
advice is make sure receive flow control is enabled so that receiver doesn't
get overrun. If tx timeouts are an issue use a rate limiter like TBF. Do not use
the chip with 10 or 100 mbit since the transmitter is more prone to get overrun.
next prev parent reply other threads:[~2009-12-24 22:22 UTC|newest]
Thread overview: 140+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <4B300A2A.8040305@gmail.com>
2009-12-24 6:58 ` sky2 panic in 2.6.32.1 under load Andrew Morton
2009-12-24 16:03 ` Berck Nash
2009-12-24 16:28 ` Daniel Hazelton
2009-12-24 22:21 ` Stephen Hemminger [this message]
2009-12-24 22:42 ` Michael Breuer
2009-12-25 0:06 ` Daniel Hazelton
2009-12-24 16:10 ` Michael Breuer
2009-12-24 16:16 ` Berck Nash
2009-12-24 16:26 ` Michael Breuer
[not found] ` <4B300E30.9090707@majjas.com>
[not found] ` <4B3114E3.1070602@majjas.com>
[not found] ` <4B329FA3.9090904@majjas.com>
2009-12-24 7:01 ` sky2 panic in 2.6.32.1 under load (new oops) Andrew Morton
2009-12-24 19:18 ` Michael Breuer
2009-12-24 22:27 ` Stephen Hemminger
2009-12-25 16:28 ` Michael Breuer
2009-12-25 23:22 ` Stephen Hemminger
2009-12-26 3:23 ` Michael Breuer
2009-12-26 17:57 ` Stephen Hemminger
2009-12-26 20:37 ` Michael Breuer
2009-12-26 22:05 ` [PATCH] sky2: make sure ethernet header is in transmit skb Stephen Hemminger
2009-12-27 3:44 ` David Miller
2009-12-27 4:11 ` David Miller
2010-01-04 5:32 ` David Miller
2010-01-04 16:40 ` Stephen Hemminger
2010-01-04 17:02 ` Michael Breuer
2010-01-05 23:07 ` [PATCH] af_packet: Don't use skb after dev_queue_xmit() Jarek Poplawski
2010-01-05 23:16 ` Michael Breuer
2010-01-05 23:29 ` Jarek Poplawski
2010-01-06 2:36 ` Michael Breuer
2010-01-06 7:22 ` Jarek Poplawski
2010-01-06 9:15 ` [PATCH alt.2] " Jarek Poplawski
2010-01-06 14:49 ` Stephen Hemminger
2010-01-06 19:40 ` Jarek Poplawski
2010-01-06 19:49 ` [PATCH] " Michael Breuer
2010-01-06 20:22 ` Jarek Poplawski
2010-01-06 20:33 ` Michael Breuer
2010-01-06 21:09 ` Jarek Poplawski
2010-01-06 21:32 ` Michael Breuer
2010-01-06 21:10 ` Stephen Hemminger
2010-01-06 21:20 ` Michael Breuer
2010-01-06 23:26 ` Michael Breuer
2010-01-07 2:42 ` Michael Breuer
2010-01-07 4:00 ` Michael Breuer
2010-01-07 4:53 ` Stephen Hemminger
2010-01-07 5:10 ` Michael Breuer
2010-01-07 5:32 ` Michael Breuer
2010-01-07 5:54 ` Michael Breuer
2010-01-07 7:20 ` Michael Breuer
2010-01-07 7:47 ` Jarek Poplawski
2010-01-07 7:55 ` Michael Breuer
2010-01-07 8:21 ` Jarek Poplawski
2010-01-07 15:03 ` Michael Breuer
2010-01-07 17:56 ` Jarek Poplawski
2010-01-07 18:17 ` Jarek Poplawski
2010-01-07 15:05 ` Michael Breuer
2010-01-07 18:01 ` Jarek Poplawski
2010-01-07 18:19 ` Michael Breuer
2010-01-07 18:35 ` Jarek Poplawski
2010-01-07 18:40 ` Michael Breuer
2010-01-07 18:43 ` Michael Breuer
2010-01-07 18:50 ` Jarek Poplawski
2010-01-07 19:36 ` Jarek Poplawski
2010-01-07 19:55 ` Michael Breuer
2010-01-07 20:22 ` Jarek Poplawski
2010-01-07 23:11 ` Michael Breuer
2010-01-08 7:45 ` Jarek Poplawski
2010-01-08 16:40 ` Michael Breuer
2010-01-08 21:29 ` Jarek Poplawski
2010-01-08 21:48 ` Michael Breuer
2010-01-08 22:02 ` Jarek Poplawski
2010-01-09 4:45 ` Michael Breuer
2010-01-09 5:44 ` Michael Breuer
2010-01-09 12:28 ` Jarek Poplawski
2010-01-09 18:34 ` Michael Breuer
2010-01-13 20:39 ` Michael Breuer
2010-01-13 21:09 ` Jarek Poplawski
2010-01-13 21:16 ` Michael Breuer
2010-01-13 21:34 ` Jarek Poplawski
2010-01-17 16:26 ` Michael Breuer
2010-01-17 22:17 ` Jarek Poplawski
2010-01-17 22:34 ` Michael Breuer
2010-01-17 23:05 ` Jarek Poplawski
2010-01-17 23:15 ` Michael Breuer
2010-01-18 7:30 ` Jarek Poplawski
2010-01-18 16:29 ` Michael Breuer
2010-01-18 20:46 ` Jarek Poplawski
2010-01-18 20:56 ` Michael Breuer
2010-01-18 21:00 ` Stephen Hemminger
2010-01-18 21:06 ` Jarek Poplawski
2010-01-18 21:24 ` Michael Breuer
2010-01-18 21:50 ` Jarek Poplawski
2010-01-18 21:25 ` Jarek Poplawski
2010-01-18 21:39 ` Michael Breuer
2010-01-18 22:08 ` Jarek Poplawski
2010-01-18 22:17 ` Jarek Poplawski
2010-01-18 22:47 ` Michael Breuer
2010-01-19 5:46 ` Michael Breuer
2010-01-19 8:41 ` Jarek Poplawski
2010-01-19 15:28 ` Michael Breuer
2010-01-21 19:48 ` Michael Breuer
2010-01-19 10:47 ` Jarek Poplawski
2010-01-19 15:47 ` Michael Breuer
2010-01-19 19:59 ` Jarek Poplawski
2010-01-19 20:06 ` Michael Breuer
2010-01-19 20:29 ` Jarek Poplawski
2010-01-19 22:45 ` Jarek Poplawski
2010-01-20 1:01 ` Michael Breuer
2010-01-20 1:10 ` Stephen Hemminger
2010-01-21 16:14 ` Stefan Richter
2010-01-21 16:50 ` Stefan Richter
2010-01-18 22:25 ` Michael Breuer
2010-01-18 22:40 ` Jarek Poplawski
2009-12-27 17:03 ` sky2 panic in 2.6.32.1 under load (new oops) Michael Breuer
2009-12-27 18:22 ` Stephen Hemminger
2009-12-27 19:39 ` Michael Breuer
2009-12-29 17:30 ` Stephen Hemminger
2009-12-29 17:39 ` Michael Breuer
2009-12-29 18:38 ` Michael Breuer
2009-12-29 18:54 ` Michael Breuer
2009-12-29 19:49 ` Stephen Hemminger
2009-12-29 20:41 ` Michael Breuer
2009-12-30 7:23 ` Michael Breuer
2009-12-30 7:58 ` Stephen Hemminger
2009-12-30 17:49 ` Michael Breuer
2009-12-30 19:15 ` audit.c skb - tty race condition - was " Michael Breuer
2009-12-30 20:44 ` Michael Breuer
2009-12-30 21:15 ` Michael Breuer
2009-12-30 21:21 ` Michael Breuer
2009-12-30 7:59 ` Stephen Hemminger
2009-12-30 15:40 ` Michael Breuer
2009-12-30 18:10 ` Stephen Hemminger
2009-12-30 18:37 ` Michael Breuer
2009-12-31 18:09 ` Michael Breuer
2009-12-31 18:24 ` Stephen Hemminger
2010-01-01 17:42 ` Michael Breuer
2010-01-01 19:26 ` sky2 panic in 2.6.32.1 under load (tty NULL write) Michael Breuer
2010-01-01 20:34 ` Michael Breuer
2010-01-02 21:42 ` Michael Breuer
2009-12-29 19:15 ` sky2 panic in 2.6.32.1 under load (new oops) Jarek Poplawski
2009-12-29 19:20 ` Michael Breuer
2009-12-30 8:07 ` Stephen Hemminger
2009-12-30 15:36 ` Michael Breuer
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=20091224142146.700e4ac8@nehalam \
--to=shemminger@vyatta.com \
--cc=akpm@linux-foundation.org \
--cc=dhazelton@enter.net \
--cc=flyboy@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--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).