netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Ben Hutchings <bhutchings@solarflare.com>
To: Bjarke Istrup Pedersen <gurligebis@gentoo.org>
Cc: <linux-kernel@vger.kernel.org>, <netdev@vger.kernel.org>,
	Roger Luethi <rl@hellgate.ch>
Subject: Re: [PATCH 1/1] via-rhine: Fix hanging with high CPU load on low-end broads.
Date: Wed, 28 Dec 2011 18:13:43 +0100	[thread overview]
Message-ID: <1325092423.2327.70.camel@deadeye> (raw)
In-Reply-To: <CACPM=kWpGnBOuKPUK9MgzyETe4VmvEtVXkYtPA_+Xgn=COkCPw@mail.gmail.com>

On Wed, 2011-12-28 at 16:14 +0100, Bjarke Istrup Pedersen wrote:
> 2011/12/28 Ben Hutchings <bhutchings@solarflare.com>:
> > On Wed, 2011-12-28 at 12:28 +0000, Bjarke Istrup Pedersen wrote:
> >> Working around problem causing high CPU load and hanging system when
> >> there is alot of network trafic.
> >>
> >> It is kind of an ugly way to work around it, but it allows the Soekris
> >> net5501 to have trafic between two of it's NICs without hanging so much
> >> that the watchdog kicks in and does a hard reboot of the system.
> >>
> >> There is more info on the problem here:
> >> http://http://lists.soekris.com/pipermail/soekris-tech/2010-October/016889.html
> >>
> >> Tested with positive results on two Soekris net5501-70 boxes.
> >
> > This is completely wrong.  In a UP configuration the extra spinlock
> > calls have no effect (except perhaps a small delay).  In an SMP
> > configuration they will cause rhine_tx() to deadlock when it also tries
> > to acquire the spinlock.
> >
> > Ben.
> 
> Okay, the Soekris net5501-70 boxes are single-core, and I haven't got
> any SMP boxes with that nic.
> Is there a better solution for the problem then, to avoid it hanging
> the box on a non-smp machine with a slow (500mhz) cpu?

If the system actually hangs then I assume there is some bug in the
driver.  I would guess the actual problem is that the interrupt and NAPI
handlers are running constantly so that user processes never run (which
I think counts as soft lockup).

If the hardware supports it, interrupt moderation may help a little by
slightly reducing the per-packet processing cost, but it isn't a full
solution.  Or you can use a real-time kernel, which schedules interrupt
and NAPI handlers as tasks, and adjust priorities so that user processes
can still run.  But that brings its own problems (including generally
lower throughput).

Ben.

-- 
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.

  reply	other threads:[~2011-12-28 17:13 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-12-28 12:28 [PATCH 1/1] via-rhine: Fix hanging with high CPU load on low-end broads Bjarke Istrup Pedersen
2011-12-28 14:37 ` Ben Hutchings
2011-12-28 15:14   ` Bjarke Istrup Pedersen
2011-12-28 17:13     ` Ben Hutchings [this message]
2011-12-28 17:30       ` Bjarke Istrup Pedersen
2011-12-28 18:17         ` Stephen Hemminger
2011-12-28 18:34           ` David Miller
2011-12-28 19:16             ` Bjarke Istrup Pedersen
2011-12-29 15:39               ` Francois Romieu
2011-12-29 18:39                 ` Francois Romieu
2011-12-29 22:00                   ` Bjarke Istrup Pedersen
2011-12-29 22:05                   ` David Miller
2011-12-29 23:19                     ` Bjarke Istrup Pedersen
2011-12-29 23:37                       ` Bjarke Istrup Pedersen
2011-12-29 23:48                         ` David Miller
2011-12-30  1:18                           ` Bjarke Istrup Pedersen
2011-12-29 23:53                         ` Francois Romieu
2011-12-30  1:21                           ` Bjarke Istrup Pedersen
2011-12-30 11:48                           ` Bjarke Istrup Pedersen
2011-12-30 12:56                             ` Francois Romieu
2011-12-30 13:51                               ` Bjarke Istrup Pedersen
2011-12-30 18:03                                 ` Francois Romieu
2011-12-30 18:45                                   ` Bjarke Istrup Pedersen
2011-12-30 20:38                                   ` Bjarke Istrup Pedersen
2011-12-30 21:50                                     ` Francois Romieu
2011-12-30 22:08                                       ` Bjarke Istrup Pedersen
2011-12-30 22:08                                         ` Francois Romieu
2011-12-30 22:13                                       ` Bjarke Istrup Pedersen
2011-12-30 22:19                                         ` Francois Romieu
2011-12-30 22:41                                           ` Bjarke Istrup Pedersen
2011-12-31  1:37                                             ` Bjarke Istrup Pedersen
2011-12-31 12:17                                               ` Francois Romieu
2012-01-01 12:09                                                 ` Bjarke Istrup Pedersen
2012-01-01 21:15                                                 ` Bjarke Istrup Pedersen
2012-01-03 19:20                                                 ` Bjarke Istrup Pedersen
2012-01-03 22:32                                                   ` Francois Romieu
2012-01-04  7:59                                                     ` Bjarke Istrup Pedersen
2012-01-05 12:01                                                       ` Bjarke Istrup Pedersen
2012-01-05 16:20                                                         ` Francois Romieu
2012-01-05 16:46                                                           ` Bjarke Istrup Pedersen
2012-01-05 17:12                                                         ` 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=1325092423.2327.70.camel@deadeye \
    --to=bhutchings@solarflare.com \
    --cc=gurligebis@gentoo.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=rl@hellgate.ch \
    /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).