Boot sequence speed-up?

Dan Nicholson dbn.lists at gmail.com
Sat Feb 3 11:28:48 PST 2007


On 2/3/07, Ken Moffat <ken at linuxfromscratch.org> wrote:
> On Sat, Feb 03, 2007 at 10:45:58AM -0800, Dan Nicholson wrote:
> >
> > At the minimum, things could be vastly sped up by not serializing the
> > whole operation. Read this article from IBM for an example.
> >
> > http://www-128.ibm.com/developerworks/linux/library/l-boot.html
> >

> For me, the big delays in bringing up userspace
> are ntp, and to lesser extents dhcp, starting cups (varies -
> sometimes seems very slow, probably version dependant), and mounting
> an nfs share.  Oh, and starting X itself - it doesn't matter how
> fast the machine gets, X always takes too long to start (several
> seconds, feels like an eternity!).
>
>  Sure, for my desktops I could save time by running ntp in the
> background, but I like the comfort (or not) of seeing how far out
> the time was.

I don't think it's really in the background, though. At least on SuSE
(which I had for a bit on my laptop when I first got it), all the
bootscripts run as they always do, it's just not serial. It still has
to wait until all the scripts complete before init returns. So, you'd
still see the message coming from ntp, but it wouldn't stall the whole
process while that was happening. It only stalls bootscripts which are
dependent on ntp. Somehow, the messages from the different bootscripts
weren't getting garbled. I don't know how, though.

> Still, I await your initial hint with interest ;)

It's one of the next things I want to play with :) I'll let you know
when I make some progress, but I imagine that others have already done
it.

--
Dan



More information about the lfs-dev mailing list