mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Andi Kleen <ak@suse.de>
To: Peter Williams <peterw@aurema.com>
Cc: Andi Kleen <ak@suse.de>, linux-kernel@vger.kernel.org, johnl@aurema.com
Subject: Re: [RFC][PATCH] O(1) Entitlement Based Scheduler
Date: Wed, 3 Mar 2004 11:13:37 +0100	[thread overview]
Message-ID: <20040303101336.GC8008@wotan.suse.de> (raw)
In-Reply-To: <404554D8.5040800@aurema.com>

On Wed, Mar 03, 2004 at 02:45:28PM +1100, Peter Williams wrote:
> Andi Kleen wrote:
> >Peter Williams <peterw@aurema.com> writes:
> >
> >One comment on the patches: could you remove the zillions of numerical 
> >Kconfig
> >options and just make them sysctls? I don't think it makes any sense 
> >to require a reboot to change any of that. And the user is unlikely
> >to have much idea yet on what he wants on them while configuring.
> 
> The default initial values should be fine and the default configuration 
> allows the scheduling tuning parameters (i.e. half life and time slice 
>       ) to be changed on a running system via the /proc file system. 
> These are mainly there so that different settings can be used with 
> various benchmarks to determine what are the best settings for various 
> types of loads.  If good default values that work well for a wide 
> variety of load types can be found as a result of these experiments then 
> these parameters may be made constants in the code.  If not they 
> probably should be made settable via system calls rather than via /proc 
> as you suggest.

No doubt that there are different settings that make sense 
for different workloads. But there is no reason one has to recompile
to set them - it's much nicer to just run a script at boot time to set
them, instead of recompiling/rebooting. This will also make benchmarking
much easier, because you can just write a script that sets the
various parameters, runs workloads, sets other parameters, runs
workload again etc. Requiring a recompile and reboot makes this
much harder.

Also I suspect most people who are not heavily into scheduling
theory will go with the defaults at compile time, so it's important
that they already work well.

And consider it from the side of a standard distribution: users 
don't normally recompile their kernels there, so everything that
makes sense to be tunable should be settable without recompiling.

IMHO CONFIG_* should not be used for anything except for kernel binary
size tuning and possible compiler tuning.

-Andi


  reply	other threads:[~2004-03-03 10:13 UTC|newest]

Thread overview: 66+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <fa.fi4j08o.17nchps@ifi.uio.no.suse.lists.linux.kernel>
     [not found] ` <fa.ctat17m.8mqa3c@ifi.uio.no.suse.lists.linux.kernel>
     [not found]   ` <yydjishqw10p.fsf@galizur.uio.no.suse.lists.linux.kernel>
     [not found]     ` <40426E1C.8010806@aurema.com.suse.lists.linux.kernel>
2004-03-03  2:48       ` Andi Kleen
2004-03-03  3:45         ` Peter Williams
2004-03-03 10:13           ` Andi Kleen [this message]
2004-03-03 23:46             ` Peter Williams
2004-03-03 15:57           ` Andi Kleen
2004-03-04  0:41             ` Peter Williams
2004-03-05  3:55               ` Andi Kleen
     [not found] <1vuMd-5jx-5@gated-at.bofh.it>
     [not found] ` <1vuMd-5jx-7@gated-at.bofh.it>
     [not found]   ` <1vuMd-5jx-9@gated-at.bofh.it>
     [not found]     ` <1vuMd-5jx-11@gated-at.bofh.it>
     [not found]       ` <1vuMd-5jx-3@gated-at.bofh.it>
     [not found]         ` <1vvyx-6jy-13@gated-at.bofh.it>
     [not found]           ` <1vBE2-48V-21@gated-at.bofh.it>
2004-03-03 21:38             ` Bill Davidsen
     [not found] <fa.ftul5bl.nlk3pr@ifi.uio.no>
     [not found] ` <fa.cvc8vnj.ahebjd@ifi.uio.no>
2004-03-01  9:18   ` Joachim B Haga
2004-03-01 10:18     ` Paul Wagland
2004-03-01 19:11       ` Mike Fedyk
     [not found] <fa.jgj0bdi.b3u6qk@ifi.uio.no>
2004-03-01  1:54 ` Andy Lutomirski
2004-03-01  2:54   ` Peter Williams
2004-03-01  3:46     ` Andy Lutomirski
2004-03-01  4:18       ` Peter Williams
2004-03-02 23:36   ` Peter Williams
     [not found] <894006121@toto.iv>
2004-03-01  0:00 ` Peter Chubb
2004-03-02  1:25   ` Peter Williams
     [not found] <fa.fi4j08o.17nchps@ifi.uio.no>
     [not found] ` <fa.ctat17m.8mqa3c@ifi.uio.no>
2004-02-29 11:58   ` Joachim B Haga
2004-02-29 20:39     ` Paul Jackson
2004-02-29 22:56     ` Peter Williams
     [not found] <1t8wp-qF-11@gated-at.bofh.it>
     [not found] ` <1th6J-az-13@gated-at.bofh.it>
     [not found]   ` <403E2929.2080705@tmr.com>
2004-02-27  3:44     ` Rik van Riel
2004-02-28 21:27       ` Bill Davidsen
2004-02-28 23:55         ` Peter Williams
2004-03-04 21:08           ` Timothy Miller
     [not found] <1tfy0-7ly-29@gated-at.bofh.it>
     [not found] ` <1thzJ-A5-13@gated-at.bofh.it>
     [not found]   ` <1tjrN-2m5-1@gated-at.bofh.it>
     [not found]     ` <1tjLa-2Ab-9@gated-at.bofh.it>
     [not found]       ` <1tlaf-3OY-11@gated-at.bofh.it>
     [not found]         ` <1tljX-3Wf-5@gated-at.bofh.it>
     [not found]           ` <1tznd-CP-35@gated-at.bofh.it>
     [not found]             ` <1tzQe-10s-25@gated-at.bofh.it>
2004-02-26 20:14               ` Bill Davidsen
2004-02-26  3:30 Albert Cahalan
2004-02-26  6:19 ` Peter Williams
2004-02-26 17:57   ` Albert Cahalan
2004-02-26 23:24     ` Peter Williams
2004-03-01  3:47       ` Peter Williams
     [not found] <fa.f12rt3d.c0s9rt@ifi.uio.no>
     [not found] ` <fa.ishajoq.q5g90m@ifi.uio.no>
2004-02-25 23:33   ` Junio C Hamano
2004-02-26  8:15     ` Catalin BOIE
  -- strict thread matches above, loose matches on Subject: below --
2004-02-25 14:35 John Lee
2004-02-25 17:09 ` Timothy Miller
2004-02-25 22:12   ` John Lee
2004-02-26  0:31     ` Timothy Miller
2004-02-26  2:04       ` John Lee
2004-02-26  2:18       ` Peter Williams
2004-02-26  2:42         ` Mike Fedyk
2004-02-26  4:10           ` Peter Williams
2004-02-26  4:19             ` Mike Fedyk
2004-02-26 19:23               ` Shailabh Nagar
2004-02-26 19:46                 ` Mike Fedyk
2004-02-26 20:42                   ` Shailabh Nagar
2004-02-26 16:10           ` Timothy Miller
2004-02-26 19:47             ` Mike Fedyk
2004-02-26 22:51             ` Peter Williams
2004-02-27 10:06               ` Helge Hafting
2004-02-27 11:04                 ` Peter Williams
2004-02-26 16:08         ` Timothy Miller
2004-02-26 16:51           ` Rik van Riel
2004-02-26 20:15           ` Peter Williams
2004-02-27 14:46             ` Timothy Miller
2004-02-28  5:00               ` Peter Williams
2004-03-04 21:18         ` Robert White
2004-03-04 23:15           ` Peter Williams
2004-02-26  2:48       ` Nuno Silva
2004-02-26  4:25         ` Peter Williams
2004-02-26 15:57           ` Rik van Riel
2004-02-26 19:28             ` Shailabh Nagar
2004-02-26 16:12         ` Timothy Miller
2004-02-25 22:51 ` Pavel Machek
2004-02-26  3:14   ` John Lee
2004-02-25 23:45 ` Rik van Riel
2004-02-26  7:18 ` John Lee

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=20040303101336.GC8008@wotan.suse.de \
    --to=ak@suse.de \
    --cc=johnl@aurema.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=peterw@aurema.com \
    /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

all inboxes | Powered by JetHome®