From: Peter Zijlstra <peterz@infradead.org>
To: raz ben yehuda <raziebe@gmail.com>
Cc: Maxim Levitsky <maximlevitsky@gmail.com>,
Christoph Lameter <cl@linux-foundation.org>,
Chris Friesen <cfriesen@nortel.com>,
Mike Galbraith <efault@gmx.de>,
riel@redhat.com, mingo@elte.hu,
andrew motron <akpm@linux-foundation.org>,
wiseman@macs.biu.ac.il, lkml <linux-kernel@vger.kernel.org>,
linux-rt-users@vger.kernel.org
Subject: Re: RFC: THE OFFLINE SCHEDULER
Date: Wed, 26 Aug 2009 17:30:25 +0200 [thread overview]
Message-ID: <1251300625.18584.18.camel@twins> (raw)
In-Reply-To: <1251298443.4791.7.camel@raz>
On Wed, 2009-08-26 at 17:54 +0300, raz ben yehuda wrote:
> This is exactly what offsched (sos) is. you got it. SOS was partly
> inspired by the notion of a GPU.
It is not, GPUs and other paired chips form a hybrid system. Linux is
known to run on one or more of such chips and communicate through
whatever means these chips have.
But what you propose here is hard partitioning a homogeneous system,
totally different.
> Processors are to become more and more redundant and Linux as an
> evolutionary system must use it. why not offload raid5 write engine ?
> why not encrypt in a different processor ?
Why waste a whole cpu for something that could be done by part of one?
> Also , having so many processors in a single OS means a bug prone
> system , with endless contention points when two or more OS processors
> interacts.
You're bound to have interaction between the core os and these
partitions you want, non of it different from how threads in the kernel
would interact, other than that you're going to re-invent everything
already present in the kernel.
> let's make things simpler.
You don't, you make things more complex by introducing duplicate
functionality.
What's more, you burden the user with having to configure such a system,
and make choices on having to give up parts of his system, nothing like
that should be needed on a homogeneous system.
Work spend on trimming fat of of the core kernel helps everybody, even
users not otherwise interested in things like giving up a whole cpu for
some odd purpose.
There is no reason something could be done more efficiently on a
dedicated CPU than not when you assume a homogeneous system (which is
all Linux supports in the single image sense).
If you think the kernel is too fat and does superfluous things for your
needs, help trim it.
next prev parent reply other threads:[~2009-08-26 15:31 UTC|newest]
Thread overview: 81+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-08-22 23:27 RFC: THE OFFLINE SCHEDULER raz ben yehuda
2009-08-23 5:21 ` Mike Galbraith
2009-08-23 9:09 ` raz ben yehuda
2009-08-23 7:30 ` Mike Galbraith
2009-08-23 11:05 ` raz ben yehuda
2009-08-23 9:52 ` Mike Galbraith
2009-08-25 15:23 ` Christoph Lameter
2009-08-25 17:56 ` Mike Galbraith
2009-08-25 18:03 ` Christoph Lameter
2009-08-25 18:12 ` Mike Galbraith
[not found] ` <5d96567b0908251522m3fd4ab98n76a52a34a11e874c@mail.gmail.com>
2009-08-25 22:32 ` Fwd: " Raz
2009-08-25 19:08 ` Peter Zijlstra
2009-08-25 19:18 ` Christoph Lameter
2009-08-25 19:22 ` Chris Friesen
2009-08-25 20:35 ` Sven-Thorsten Dietrich
2009-08-26 5:31 ` Peter Zijlstra
2009-08-26 10:29 ` raz ben yehuda
2009-08-26 8:02 ` Mike Galbraith
2009-08-26 8:16 ` Raz
2009-08-26 13:47 ` Christoph Lameter
2009-08-26 14:45 ` Maxim Levitsky
2009-08-26 14:54 ` raz ben yehuda
2009-08-26 15:06 ` Pekka Enberg
2009-08-26 15:11 ` raz ben yehuda
2009-08-26 15:30 ` Peter Zijlstra [this message]
2009-08-26 15:41 ` Christoph Lameter
2009-08-26 16:03 ` Peter Zijlstra
2009-08-26 16:16 ` Pekka Enberg
2009-08-26 16:20 ` Christoph Lameter
2009-08-26 18:04 ` Ingo Molnar
2009-08-26 19:15 ` Christoph Lameter
2009-08-26 19:32 ` Ingo Molnar
2009-08-26 20:40 ` Christoph Lameter
2009-08-26 20:50 ` Andrew Morton
2009-08-26 21:09 ` Christoph Lameter
2009-08-26 21:15 ` Chris Friesen
2009-08-26 21:37 ` raz ben yehuda
2009-08-27 16:51 ` Chris Friesen
2009-08-27 17:04 ` Christoph Lameter
2009-08-27 21:09 ` Thomas Gleixner
2009-08-27 22:22 ` Gregory Haskins
2009-08-28 2:15 ` Rik van Riel
2009-08-28 3:33 ` Gregory Haskins
2009-08-28 4:27 ` Gregory Haskins
2009-08-28 10:26 ` Thomas Gleixner
2009-08-28 18:57 ` Christoph Lameter
2009-08-28 19:23 ` Thomas Gleixner
2009-08-28 19:52 ` Christoph Lameter
2009-08-28 20:00 ` Thomas Gleixner
2009-08-28 20:21 ` Christoph Lameter
2009-08-28 20:34 ` Thomas Gleixner
2009-08-31 19:19 ` Christoph Lameter
2009-08-31 17:44 ` Roland Dreier
2009-09-01 18:42 ` Christoph Lameter
2009-09-01 16:15 ` Roland Dreier
2009-08-29 17:03 ` jim owens
2009-08-31 19:22 ` Christoph Lameter
2009-08-31 15:33 ` Peter Zijlstra
2009-09-01 18:46 ` Christoph Lameter
2009-08-28 6:14 ` Peter Zijlstra
2009-08-27 23:51 ` Chris Friesen
2009-08-28 0:44 ` Thomas Gleixner
2009-08-28 21:20 ` Chris Friesen
2009-08-28 18:43 ` Christoph Lameter
2009-08-27 21:33 ` raz ben yehuda
2009-08-27 22:05 ` Thomas Gleixner
2009-08-28 8:38 ` raz ben yehuda
2009-08-28 10:05 ` Thomas Gleixner
2009-08-28 13:25 ` Rik van Riel
2009-08-28 13:37 ` jim owens
2009-08-28 15:22 ` raz ben yehuda
2009-08-26 21:34 ` Ingo Molnar
2009-08-27 2:55 ` Frank Ch. Eigler
2009-08-26 21:34 ` raz ben yehuda
2009-08-26 21:08 ` Ingo Molnar
2009-08-26 21:26 ` Christoph Lameter
2009-08-26 21:32 ` raz ben yehuda
2009-08-27 7:15 ` Mike Galbraith
2009-08-26 15:37 ` Chetan.Loke
2009-08-26 15:21 ` Pekka Enberg
2009-08-25 21:09 ` Éric Piel
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=1251300625.18584.18.camel@twins \
--to=peterz@infradead.org \
--cc=akpm@linux-foundation.org \
--cc=cfriesen@nortel.com \
--cc=cl@linux-foundation.org \
--cc=efault@gmx.de \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-rt-users@vger.kernel.org \
--cc=maximlevitsky@gmail.com \
--cc=mingo@elte.hu \
--cc=raziebe@gmail.com \
--cc=riel@redhat.com \
--cc=wiseman@macs.biu.ac.il \
/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