All of lore.kernel.org
 help / color / mirror / Atom feed
From: Max Krasnyansky <maxk@qualcomm.com>
To: Ingo Oeser <ioe-lkml@rameria.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Paul Jackson <pj@sgi.com>,
	linux-kernel@vger.kernel.org, Con Kolivas <kernel@kolivas.org>,
	"Derek L. Fults" <dfults@sgi.com>, devik <devik@cdi.cz>,
	Dimitri Sivanich <sivanich@sgi.com>,
	Dinakar Guniguntala <dino@in.ibm.com>,
	Emmanuel Pacaud <emmanuel.pacaud@univ-poitiers.fr>,
	Frederik Deweerdt <deweerdt@free.fr>, Ingo Molnar <mingo@elte.hu>,
	Matthew Dobson <colpatch@us.ibm.com>,
	Nick Piggin <nickpiggin@yahoo.com.au>,
	rostedt@goodmis.org, Oleg Nesterov <oleg@tv-sign.ru>,
	"Paul E. McKenney" <paulmck@us.ibm.com>,
	Paul Menage <menage@google.com>,
	"Randy.Dunlap" <rddunlap@osdl.org>,
	suresh.b.siddha@intel.com, Thomas Gleixner <tglx@linutronix.de>
Subject: Re: Inquiry: Should we remove "isolcpus= kernel boot option? (may have realtime uses)
Date: Mon, 02 Jun 2008 20:32:17 -0700	[thread overview]
Message-ID: <4844BB41.7000605@qualcomm.com> (raw)
In-Reply-To: <200806030156.00155.ioe-lkml@rameria.de>



Ingo Oeser wrote:
> Hi Max,
> Hi Peter,
> 
> On Tuesday 03 June 2008, Max Krasnyansky wrote:
>> Ingo, I just wanted to elaborate on what Peter is saying. That CPU will still
>> have to be _booted_ properly. It may be used for hard- and soft- interrupt
>> processing, workqueues (internal kernel queuing mechanism) and kernel timers.
> 
> Oh! Didn't know that user process scheduling is so much 
Not sure what you meant here. Stuff that I listed has nothing to do with user
process scheduling.

>> In your particular case you're much much much better off with doing
>> 	echo 0 > /sys/devices/system/cpuN/online
>> either during initrd stage or as a first init script.
>> That way bad cpu will be _completely_ disabled.
> 
> The initrd is from the distribution. I have no sane way to change it
> fast and permanent. Can I change the initrd and still have a certified
> RHEL or SLES? Are there initrd hooks, which survive packet installation?
That's why I mentioned "first init" script. You can create a simple init.d
compliant script that runs with priority 0 (see /etc/init.d/network for
example). That should be early enough.

> I would really appreciate some way to keep the kernel from using 
> a CPU at all to do fault isolation. If possible not even booting it.
How does isolcpu= boot option helps in this case ?
I suppose the closes option is maxcpus=. We can probably add ignorecpus= or
something to handle your use case but it has nothing to do with isolcpus=.

> Bootparameters survived all distro fiddling so far. I love them!
So do custom init.d scripts.

> 
> Try to convince a hardware vendor, that you don't have a software bug.
> Try to convince him that you didn't break the hardware by swapping it around.
> 
> So I'll ACK removing isolcpus, if we get a better replacement boot option.
I think you're missing the point here. It's like saying
"Lets not switch to electric cars because I use gasoline to kill weeds".

As I mentioned before, cpus listed in the isolcpus= boot option will still
handle hard-/soft- irqs, kernel work, kernel timers. You are much better off
using cpu hotplug (ie putting bad cpu offline). Feel free to propose
ignorecpus= option in a separate thread.

Max






  reply	other threads:[~2008-06-03  3:32 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-06-02  2:30 Inquiry: Should we remove "isolcpus= kernel boot option? (may have realtime uses) Paul Jackson
2008-06-02 16:42 ` Dimitri Sivanich
2008-06-02 18:39   ` Max Krasnyansky
2008-06-02 21:41     ` Dimitri Sivanich
2008-06-02 21:59       ` Max Krasnyansky
2008-06-03 14:40         ` Dimitri Sivanich
2008-06-03 17:57           ` Max Krasnyanskiy
2008-06-04 14:00           ` Dimitri Sivanich
2008-06-04 18:07             ` Stop machine threads are getting preemted by the rt period enforcement Max Krasnyansky
2008-06-04 18:18               ` Peter Zijlstra
2008-06-04 18:24                 ` Max Krasnyansky
2008-06-04 18:55                   ` Peter Zijlstra
2008-06-04 20:14                     ` Max Krasnyansky
2008-06-02 22:35 ` Inquiry: Should we remove "isolcpus= kernel boot option? (may have realtime uses) Ingo Oeser
2008-06-02 22:45   ` Peter Zijlstra
2008-06-02 23:04     ` Max Krasnyansky
2008-06-02 23:55       ` Ingo Oeser
2008-06-03  3:32         ` Max Krasnyansky [this message]
2008-06-03 23:47           ` Max Krasnyanskiy
2008-06-04  0:41             ` Paul Jackson
2008-06-04  4:32               ` Max Krasnyansky
2008-06-04  4:47                 ` Paul Jackson
2008-06-04 12:18                 ` Andi Kleen
2008-06-04 17:41                   ` Paul Jackson
2008-06-04 18:29                     ` Max Krasnyansky
2008-06-04 18:56                       ` Peter Zijlstra
2008-06-04 19:34                         ` Max Krasnyansky
2008-06-04 18:58                       ` Paul Jackson
2008-06-04 19:31                         ` Max Krasnyansky
2008-06-04 19:37                           ` Paul Jackson
2008-06-04 19:45                             ` Max Krasnyansky
2008-06-04 20:05                       ` Andi Kleen
2008-06-04 20:23                         ` Paul Jackson
2008-06-04 20:03                     ` Andi Kleen
2008-06-04 20:16                       ` Paul Jackson
2008-06-04 20:33                         ` Andi Kleen
2008-06-04 20:38                           ` Paul Jackson
2008-06-04 21:16                             ` Max Krasnyansky
2008-06-04 21:17                               ` Paul Jackson
2008-06-04 21:20                                 ` Max Krasnyansky
2008-06-04 21:26                                   ` Paul Jackson
2008-06-04  1:18             ` Nick Piggin
2008-06-04  3:00               ` Max Krasnyansky
2008-06-04 16:18             ` Ingo Oeser
2008-06-04 17:47               ` Max Krasnyansky
2008-06-03  6:03     ` Nick Piggin
2008-06-04  9:58       ` Mark Hounschell
2008-06-04 17:26         ` Paul Jackson
2008-06-04 21:00           ` Mark Hounschell
2008-06-04 21:03             ` Paul Jackson
2008-06-04 19:26         ` Max Krasnyansky
2008-06-04 20:25           ` Peter Zijlstra
2008-06-04 21:44             ` Michael Trimarchi
2008-06-04 21:52               ` Peter Zijlstra
2008-06-05 11:16                 ` Michael Trimarchi
2008-06-05 12:07                   ` Peter Zijlstra
2008-06-05 14:57                     ` Michael Trimarchi
2009-05-08  2:48               ` GeunSik Lim
2008-06-05 11:44             ` Mark Hounschell
2008-06-06 22:28               ` Max Krasnyanskiy

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=4844BB41.7000605@qualcomm.com \
    --to=maxk@qualcomm.com \
    --cc=a.p.zijlstra@chello.nl \
    --cc=colpatch@us.ibm.com \
    --cc=devik@cdi.cz \
    --cc=deweerdt@free.fr \
    --cc=dfults@sgi.com \
    --cc=dino@in.ibm.com \
    --cc=emmanuel.pacaud@univ-poitiers.fr \
    --cc=ioe-lkml@rameria.de \
    --cc=kernel@kolivas.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=menage@google.com \
    --cc=mingo@elte.hu \
    --cc=nickpiggin@yahoo.com.au \
    --cc=oleg@tv-sign.ru \
    --cc=paulmck@us.ibm.com \
    --cc=pj@sgi.com \
    --cc=rddunlap@osdl.org \
    --cc=rostedt@goodmis.org \
    --cc=sivanich@sgi.com \
    --cc=suresh.b.siddha@intel.com \
    --cc=tglx@linutronix.de \
    /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.