From: Helge Hafting <helge.hafting@aitel.hist.no>
To: Kyle Moffett <mrmacman_g4@mac.com>
Cc: Lee Revell <rlrevell@joe-job.com>,
Patrick McFarland <pmcfarland@downeast.net>,
linux-kernel@vger.kernel.org, Tim Bird <tim.bird@am.sony.com>,
Prakash Punnoor <prakashp@arcor.de>,
Paolo Ciarrocchi <paolo.ciarrocchi@gmail.com>,
linux-hotplug-devel@lists.sourceforge.net,
Greg KH <gregkh@suse.de>, Roland Dreier <roland@topspin.com>
Subject: Re: [OT] speeding boot process (was Re: [ANNOUNCE] hotplug-ng 001
Date: Tue, 15 Feb 2005 08:55:24 +0000 [thread overview]
Message-ID: <4211B8FC.8000600@aitel.hist.no> (raw)
In-Reply-To: <4B923A81-7EF3-11D9-86CC-000393ACC76E@mac.com>
Kyle Moffett wrote:
> On Feb 14, 2005, at 20:17, Lee Revell wrote:
>
>> On Mon, 2005-02-14 at 16:16 -0800, Tim Bird wrote:
>>
>>> Lee Revell wrote:
>>>
>>>> But, I was referring more to things like GDM not being started
>>>> until all
>>>> the other init scripts are done. Why not start it first, and let the
>>>> network initialize while the user is logging in?
>>>
>>>
>>> There are a number of techniques used by CE vendors to get fast bootup
>>> time. Some CE products boot Linux in under 1 second. Sony's
>>> best Linux boot time in the lab (from power on to user space)
>>> was 148 milliseconds, on an ARM chip (running at 200 MHZ I believe).
>>
>>
>> The reason I marked by response OT is that the time from power on to
>> userspace does not seem to be a big problem. It's the amount of time
>> from user space to presenting a login prompt that's way too long. My
>> distro (Debian) runs all the init scripts one at a time, and GDM is the
>> last thing that gets run. There is just no reason for this. We should
>> start X and initialize the display and get the login prompt up there
>> ASAP, and let the system acquire the DHCP lease and start sendmail and
>> apache and get the date from the NTP server *in the background while I
>> am logging in*. It's not rocket science.
>
>
> Such a system needs a drastically different bootup process than currently
> exists, including the ability to specify init-script dependencies. (Like
The init-script dependencies are specifies already - at least on debian.
Look at the most used runlevel, ls /etc/rc2.d:
S10sysklogd@
S11klogd
S14ppp
S18portmap
S19slapd
S20alsa
S20cupsys
S20dhcp3-server
S20exim
...
The numbers specify ordering constraint, low numbers must start before
high numbers.
But plenty of scripts have no interdependencies, I have 18 scripts
numbered S20.
Today they start sequentially anyway, in alphabetical order. Several of
them simply
wait for hardware, which could be done in parallel. So parallel
execution would be a win.
One could always have a user-settable limit on how much to run in
parallel, to help
out those low-memory machines.
> for example user login via GDM (and with our setup, GDM working at all)
> requires that AFS is mounted and NIS is working, which both require the
> network to be available, which requires... You can see where this is
Yes. I don't think the X system will start before other things in a
standard
installation, precisely because it *might* need the network and a bunch
of other
things to validate the user. So most of the gain will come from the
parallel starting
of earlier scripts.
I think one can safely say that the xserver it won't need web, email
or the printer subsystems to get up though, and that could also save
some time.
> going. I think eventually we need a better /sbin/init, one that can use
> a traditional legacy /etc/inittab file in addition to a newfangled
> simultaneous boot process with lots of ways to start various kinds of
> services. Unfortunately such a system will need a _LOT_ of work and
> testing to make sure it doesn't break existing setups. Oh well, I can
> dream, can't I? :-D
It isn't all that hard. An init with one change - scripts with the same
number
runs in parallel instead of sequentially. (And possibly some
configurable limit
on parallel scripts). Then the rest of the work lies in an efficient
and correct numbering of those
scripts. The paranoid can renumber them so they run in the same
order as they used to. Most people can use the existing numbering,
perhaps they
discover an ordering bug or two which is trivially fixed using "mv" to
renumber scripts.
And the more daring can renumber their scripts for maximum parallelism.
Even with today's system the goal of starting X early is achievable. If you
know that X doesn't need the network on _your_ machine - renumber it to
start really early. The modification is easy enough to do. :-)
If it still starts slow, swithc to a light session manager (like xdm) and
a lightweight window manager that doesn't bring in kde or gnome.
Helge Hafting
-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_ide95&alloc_id\x14396&op=click
_______________________________________________
Linux-hotplug-devel mailing list http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel
next prev parent reply other threads:[~2005-02-15 8:55 UTC|newest]
Thread overview: 148+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-02-11 0:40 [ANNOUNCE] hotplug-ng 001 release Greg KH
2005-02-11 0:52 ` Greg KH
2005-02-11 1:30 ` Kasper Sandberg
2005-02-11 6:41 ` Greg KH
2005-02-11 11:47 ` Kasper Sandberg
2005-02-11 17:06 ` Greg KH
2005-02-12 0:02 ` Kasper Sandberg
2005-02-11 1:07 ` Patrick McFarland
2005-02-11 1:16 ` Greg KH
2005-02-11 20:40 ` Vojtech Pavlik
2005-02-14 4:06 ` Lee Revell
2005-02-14 8:32 ` Paolo Ciarrocchi
2005-02-14 8:51 ` Prakash Punnoor
2005-02-14 23:04 ` Lee Revell
2005-02-14 23:16 ` Greg KH
2005-02-14 23:28 ` Lee Revell
2005-02-15 15:15 ` Stefan Seyfried
2005-02-15 20:12 ` kernel
2005-02-14 23:21 ` Roland Dreier
2005-02-14 23:45 ` [OT] speeding boot process (was Re: [ANNOUNCE] hotplug-ng 001 Lee Revell
2005-02-15 0:16 ` Tim Bird
2005-02-15 1:17 ` Lee Revell
2005-02-15 1:45 ` [OT] speeding boot process (was Re: [ANNOUNCE] hotplug-ng 001 release) Kyle Moffett
2005-02-15 7:32 ` Gábor Lénárt
2005-02-15 8:34 ` Paolo Ciarrocchi
2005-02-15 9:27 ` [OT] speeding boot process Bernd Petrovitsch
2005-02-15 8:55 ` Helge Hafting [this message]
2005-02-15 9:33 ` [OT] speeding boot process (was Re: [ANNOUNCE] hotplug-ng 001 Bernd Petrovitsch
2005-02-15 13:20 ` Helge Hafting
2005-02-15 13:28 ` Paulo Marques
2005-02-15 13:50 ` Bernd Petrovitsch
2005-02-15 3:38 ` Jim Crilly
2005-02-15 5:52 ` Nigel Cunningham
2005-02-15 6:15 ` Jim Crilly
2005-02-15 6:39 ` Nigel Cunningham
2005-02-15 19:32 ` Lee Revell
2005-02-17 18:37 ` [OT] speeding boot process (was Re: [ANNOUNCE] hotplug-ng 001 release) jlnance
2005-02-17 19:18 ` Chris Larson
2005-02-19 5:53 ` [OT] speeding boot process (was Re: [ANNOUNCE] hotplug-ng 001 Jim Crilly
2005-02-17 19:58 ` [OT] speeding boot process (was Re: [ANNOUNCE] hotplug-ng 001 release) Helge Hafting
2005-02-19 5:56 ` [OT] speeding boot process (was Re: [ANNOUNCE] hotplug-ng 001 Jim Crilly
2005-02-19 22:47 ` [OT] speeding boot process (was Re: [ANNOUNCE] hotplug-ng 001 release) Helge Hafting
2005-02-20 2:09 ` [OT] speeding boot process (was Re: [ANNOUNCE] hotplug-ng 001 Jim Crilly
2005-02-15 8:33 ` [OT] speeding boot process (was Re: [ANNOUNCE] hotplug-ng 001 release) Paolo Ciarrocchi
2005-02-14 23:43 ` [ANNOUNCE] hotplug-ng 001 release Diego Calleja
2005-02-15 19:51 ` Lee Revell
2005-02-15 21:02 ` Diego Calleja
2005-02-15 21:32 ` Chris Friesen
2005-02-15 21:09 ` Valdis.Kletnieks
2005-02-15 19:56 ` Optimizing disk-I/O [was Re: [ANNOUNCE] hotplug-ng 001 release] Linas Vepstas
2005-02-15 20:46 ` Adam Goode
2005-02-15 21:11 ` Diego Calleja
2005-02-15 21:21 ` Valdis.Kletnieks
2005-02-11 3:18 ` [ANNOUNCE] hotplug-ng 001 release Bill Nottingham
2005-02-11 6:46 ` Greg KH
2005-02-11 16:19 ` Christian Bornträger
2005-02-11 17:01 ` Greg KH
2005-02-11 19:01 ` Erik Andersen
2005-02-11 19:23 ` Greg KH
2005-02-11 21:37 ` Erik van Konijnenburg
2005-02-11 21:49 ` Greg KH
2005-02-11 22:06 ` Erik van Konijnenburg
2005-02-11 22:13 ` Greg KH
2005-02-12 0:48 ` Ingo Oeser
2005-02-14 22:43 ` Greg KH
2005-02-11 8:10 ` Roman Kagan
2005-02-11 8:17 ` Greg KH
2005-02-11 9:52 ` Olivier Galibert
2005-02-11 17:08 ` Greg KH
2005-02-11 10:53 ` Roman Kagan
2005-02-11 17:41 ` Christian Bornträger
2005-02-11 18:15 ` Greg KH
2005-02-11 18:47 ` Marco d'Itri
2005-02-11 19:33 ` Greg KH
2005-02-11 19:36 ` Greg KH
2005-02-11 19:40 ` Marco d'Itri
2005-02-11 19:57 ` Greg KH
2005-02-11 20:06 ` Harald Dunkel
2005-02-11 21:01 ` Greg KH
2005-02-12 8:30 ` Harald Dunkel
2005-02-14 22:36 ` Greg KH
2005-02-15 5:39 ` Harald Dunkel
2005-02-15 7:14 ` Greg KH
2005-02-11 20:10 ` Marco d'Itri
2005-02-11 20:11 ` Roman Kagan
2005-02-11 20:20 ` Marco d'Itri
2005-02-11 20:44 ` Roman Kagan
2005-02-11 20:46 ` Roman Kagan
2005-02-11 20:56 ` Marco d'Itri
2005-02-11 21:01 ` Kay Sievers
2005-02-11 21:10 ` Greg KH
2005-02-17 6:46 ` Michael Tokarev
2005-02-11 21:21 ` Kay Sievers
2005-02-11 21:30 ` Greg KH
2005-02-11 21:33 ` Greg KH
2005-02-11 21:54 ` Marco d'Itri
2005-02-11 21:57 ` Greg KH
2005-02-11 22:05 ` Bill Nottingham
2005-02-11 22:12 ` Greg KH
2005-02-11 22:13 ` Roman Kagan
2005-02-11 22:16 ` Marco d'Itri
2005-02-11 22:18 ` Marco d'Itri
2005-02-11 22:45 ` Roman Kagan
2005-02-11 22:55 ` Roman Kagan
2005-02-11 23:05 ` Erik van Konijnenburg
2005-02-11 23:17 ` Marco d'Itri
2005-02-11 23:29 ` Roman Kagan
2005-02-11 23:35 ` Marco d'Itri
2005-02-12 0:10 ` Roman Kagan
2005-02-12 0:17 ` Marco d'Itri
2005-02-12 0:34 ` Roman Kagan
2005-02-12 11:43 ` Pozsár Balázs
2005-02-12 12:10 ` Roman Kagan
2005-02-12 16:27 ` Alexander E. Patrakov
2005-02-14 22:42 ` Greg KH
2005-02-15 7:34 ` Greg KH
2005-02-15 8:15 ` Christian Zoz
2005-02-15 10:25 ` Alexander E. Patrakov
2005-02-15 12:20 ` [sane-devel] " Julien BLACHE
2005-02-16 10:51 ` Roman Kagan
2005-02-16 11:07 ` Roman Kagan
2005-02-16 11:11 ` Marco d'Itri
2005-02-16 11:18 ` Roman Kagan
2005-02-16 11:27 ` Roman Kagan
2005-02-16 12:37 ` Kay Sievers
2005-02-16 15:17 ` Roman Kagan
2005-02-16 16:08 ` Kay Sievers
2005-02-16 23:02 ` Willem Riede
2005-02-16 23:17 ` Roman Kagan
2005-02-16 23:32 ` Willem Riede
2005-02-18 17:17 ` Patrick Mansfield
2005-02-18 18:11 ` Roman Kagan
2005-02-18 18:33 ` Patrick Mansfield
2005-02-18 19:41 ` Roman Kagan
2005-02-18 20:17 ` Patrick Mansfield
2005-02-17 9:12 ` Roman Kagan
2005-02-21 20:40 ` Erik van Konijnenburg
2005-02-22 10:35 ` Roman Kagan
2005-02-23 0:44 ` Erik van Konijnenburg
2005-02-23 11:40 ` Roman Kagan
2005-02-23 13:04 ` Erik van Konijnenburg
2005-02-24 6:27 ` Roman Kagan
2005-02-24 19:28 ` Erik van Konijnenburg
2005-02-25 22:07 ` Greg KH
2005-02-25 22:07 ` Greg KH
2005-02-25 22:11 ` Greg KH
2005-02-25 22:12 ` Greg KH
2005-02-26 17:13 ` Roman Kagan
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=4211B8FC.8000600@aitel.hist.no \
--to=helge.hafting@aitel.hist.no \
--cc=gregkh@suse.de \
--cc=linux-hotplug-devel@lists.sourceforge.net \
--cc=linux-kernel@vger.kernel.org \
--cc=mrmacman_g4@mac.com \
--cc=paolo.ciarrocchi@gmail.com \
--cc=pmcfarland@downeast.net \
--cc=prakashp@arcor.de \
--cc=rlrevell@joe-job.com \
--cc=roland@topspin.com \
--cc=tim.bird@am.sony.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;
as well as URLs for NNTP newsgroup(s).