Skip to Content.
Sympa Menu

sm-discuss - Re: [SM-Discuss] init woes

sm-discuss AT lists.ibiblio.org

Subject: Public SourceMage Discussion List

List archive

Chronological Thread  
  • From: Seth Woolley <seth AT tautology.org>
  • To: Paul Mahon <dufflebunk AT dufflebunk.homeip.net>
  • Cc: Robert Helgesson <rycee AT home.se>, sm-discuss <sm-discuss AT lists.ibiblio.org>
  • Subject: Re: [SM-Discuss] init woes
  • Date: Thu, 9 Oct 2003 18:53:01 -0700 (PDT)

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

that's what it says. That's for telinit to communicate with it though,
right? Most people would only use it one at a time and never know that.
It does open it O_NONBLOCK, too. Maybe it should be blocking?

Seth

On Thu, 9 Oct 2003, Paul Mahon wrote:

> It uses an actualy FIFO? As in the thing that can get interleaved (and
> thuse useless) data if two processes write to it at the same time?
>
> On Thu, 2003-10-09 at 20:04, Seth Woolley wrote:
> > -----BEGIN PGP SIGNED MESSAGE-----
> > Hash: SHA1
> >
> > Understood, but it's a lot more fun when you don't use the rtc device and
> > just use bash and date.
> >
> > lines 428ff of simpleinit.c from simpleinit-msb-1.1:
> >
> > if ( ( initctl_fd = open (initctl_name, O_RDWR|O_NONBLOCK, 0) ) <
> > 0 ) {
> > mkfifo (initctl_name, S_IRUSR | S_IWUSR);
> > if ( ( initctl_fd = open (initctl_name,
> > O_RDWR|O_NONBLOCK, 0) ) < 0 )
> > err ( _("error opening fifo\n") );
> > }
> > fcntl(initctl_fd,F_SETFD,FD_CLOEXEC); /*make sure no one inherits
> > the fd*/
> >
> > so it tries to open it, otherwise, it creates the fifo and then reopens
> > it, and then it last sets its fcntl options. bash can do as much.
> > (initctl_name is "/dev/initctl")
> >
> > Seth
> >
> > On Thu, 9 Oct 2003, Paul Mahon wrote:
> >
> > > No, the rtc device has a few k of extra memory, that's what I was
> > > talking about.
> > >
> > > On Thu, 2003-10-09 at 17:16, Seth Woolley wrote:
> > > > -----BEGIN PGP SIGNED MESSAGE-----
> > > > Hash: SHA1
> > > >
> > > > Current init uses a fifo/named pipe on /dev/initctl -- I'm not sure
> > > > how
> > > > this works with devfs, but it appears to work. I can look into the
> > > > source
> > > > code on how it does this exactly in a bit if somebody doesn't beat me
> > > > to
> > > > it while I'm at work. I'm sure afrayednot's stuff can use that pipe,
> > > > eh?
> > > >
> > > > If we save the clock time right away, we can minuplate the clock value
> > > > (though it is slightly volatile, esp. in seconds field) and gain some
> > > > few
> > > > extra bits from that.
> > > >
> > > > Seth
> > > >
> > > > P.S. Yes, Java and XML would be great. Then we can have the boot
> > > > screen
> > > > use the gecko rendering engine SVG-enabled and display trees of
> > > > depenedencies and stuff in SVG. A bit custom, but all-in-all a great
> > > > idea and we should look into it.
> > > >
> > > > On Thu, 9 Oct 2003, Paul Mahon wrote:
> > > >
> > > > > No, Python is soo amaturish. We should use Java and change the
> > > > > simpleinit scripts to XML.
> > > > > afrayednot's stuff requires a FS. I'm not sure if it actualy writes
> > > > > to
> > > > > the file, but the file must exist.
> > > > >
> > > > > Could we use the clock memory?
> > > > >
> > > > >
> > > > > On Thu, 2003-10-09 at 16:20, Seth Woolley wrote:
> > > > > > -----BEGIN PGP SIGNED MESSAGE-----
> > > > > > Hash: SHA1
> > > > > >
> > > > > > I'd rather fork off a daemon and do ipc to it... maybe use
> > > > > > afrayednot's
> > > > > > ipc tools if not bash...
> > > > > >
> > > > > > or we could do it all in python ;)
> > > > > >
> > > > > > There's also nvram we could use :)
> > > > > >
> > > > > > Seth
> > > > > >
> > > > > > On Thu, 9 Oct 2003, Paul Mahon wrote:
> > > > > >
> > > > > > > You're right, it's not required for anything. henh, you could
> > > > > > > mount the
> > > > > > > swap partition as ext2 until there's a real FS... ;)
> > > > > > >
> > > > > > > On Thu, 2003-10-09 at 03:22, Robert Helgesson wrote:
> > > > > > > > On Thu, Oct 09, 2003 at 02:23:34 -0500, Paul Mahon wrote:
> > > > > > > > > I don't suppse it would be possible to mount a tmpfs system
> > > > > > > > > right at the
> > > > > > > > > start?
> > > > > > > >
> > > > > > > > Yes, that would probably work, but I'm not sure that it's
> > > > > > > > safe to assume
> > > > > > > > tmpfs being available on all smgl installations. Is tmpfs
> > > > > > > > required by
> > > > > > > > anything else?
> > > > > > > >
> > > > > > > > > On Thu, 2003-10-09 at 00:25, Robert Helgesson wrote:
> > > > > > > > > > On Wed, Oct 08, 2003 at 23:41:53 -0500, Paul Mahon wrote:
> > > > > > > > > > > I don't know how the init stuff actualy works, so the
> > > > > > > > > > > no FS limitation
> > > > > > > > > > > may not be as bad. Good ideas though.
> > > > > > > > > >
> > > > > > > > > > It is as bad as you can imagine and worse. Being limited
> > > > > > > > > > to no
> > > > > > > > > > read/write fs access and no /usr can be quite frustrating
> > > > > > > > > > at times :)
> > > > > > > --
> > > > > > > De mortuis nihil nisi bonum.
> > > > > > >
> > > > > > > What is that funny looking stuff in the attachment?
> > > > > > > http://www.technomom.com/cynthia/pgp/
> > > > > > >
> > > > > >
> > > > > > - --
> > > > > > Seth Alan Woolley <seth at tautology.org>, SPAM/UCE is
> > > > > > unauthorized
> > > > > > Key id 7BEACC7D = 2978 0BD1 BA48 B671 C1EB 93F7 EDF4 3CDF 7BEA
> > > > > > CC7D
> > > > > > Full Key at seth.tautology.org and pgp.mit.edu. info:
> > > > > > www.gnupg.org
> > > > > > -----BEGIN PGP SIGNATURE-----
> > > > > > Version: GnuPG v1.2.2 (FreeBSD)
> > > > > >
> > > > > > iD8DBQE/hdEx7fQ833vqzH0RApU7AJ457uQtnkLbUEmDssyXgywRnvcaeQCfThhD
> > > > > > PJMiIKnfD5Gu7sha4y7RqcY=
> > > > > > =ACup
> > > > > > -----END PGP SIGNATURE-----
> > > > > >
> > > > > > _______________________________________________
> > > > > > SM-Discuss mailing list
> > > > > > SM-Discuss AT lists.ibiblio.org
> > > > > > http://lists.ibiblio.org/mailman/listinfo/sm-discuss
> > > > > --
> > > > > De mortuis nihil nisi bonum.
> > > > >
> > > > > What is that funny looking stuff in the attachment?
> > > > > http://www.technomom.com/cynthia/pgp/
> > > > >
> > > >
> > > > - --
> > > > Seth Alan Woolley <seth at tautology.org>, SPAM/UCE is unauthorized
> > > > Key id 7BEACC7D = 2978 0BD1 BA48 B671 C1EB 93F7 EDF4 3CDF 7BEA CC7D
> > > > Full Key at seth.tautology.org and pgp.mit.edu. info: www.gnupg.org
> > > > -----BEGIN PGP SIGNATURE-----
> > > > Version: GnuPG v1.2.2 (FreeBSD)
> > > >
> > > > iD8DBQE/hd5e7fQ833vqzH0RAnfpAJ9fpTCSZa5+vcZ3c1/FhTnT5gHPjgCfVVc+
> > > > xZ+dT5JM/g8utcAAdQJt7sE=
> > > > =Zms6
> > > > -----END PGP SIGNATURE-----
> > > >
> > > > _______________________________________________
> > > > SM-Discuss mailing list
> > > > SM-Discuss AT lists.ibiblio.org
> > > > http://lists.ibiblio.org/mailman/listinfo/sm-discuss
> > > --
> > > De mortuis nihil nisi bonum.
> > >
> > > What is that funny looking stuff in the attachment?
> > > http://www.technomom.com/cynthia/pgp/
> > >
> >
> > - --
> > Seth Alan Woolley <seth at tautology.org>, SPAM/UCE is unauthorized
> > Key id 7BEACC7D = 2978 0BD1 BA48 B671 C1EB 93F7 EDF4 3CDF 7BEA CC7D
> > Full Key at seth.tautology.org and pgp.mit.edu. info: www.gnupg.org
> > -----BEGIN PGP SIGNATURE-----
> > Version: GnuPG v1.2.2 (FreeBSD)
> >
> > iD8DBQE/hgWn7fQ833vqzH0RAnfCAKCyPXk1XCj+SzV18Ox3LYxOKUsHXgCeOJ1a
> > qD+6aHn7rnTgMvxlJK43yUc=
> > =m4TB
> > -----END PGP SIGNATURE-----
> --
> De mortuis nihil nisi bonum.
>
> What is that funny looking stuff in the attachment?
> http://www.technomom.com/cynthia/pgp/
>

- --
Seth Alan Woolley <seth at tautology.org>, SPAM/UCE is unauthorized
Key id 7BEACC7D = 2978 0BD1 BA48 B671 C1EB 93F7 EDF4 3CDF 7BEA CC7D
Full Key at seth.tautology.org and pgp.mit.edu. info: www.gnupg.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (FreeBSD)

iD8DBQE/hhEC7fQ833vqzH0RAh2PAJ9kaUvRL9aTphq0NbifPcx8TwtG8wCcC4Ss
XIljo8PMXCtAoxtZbqosMp4=
=IhUV
-----END PGP SIGNATURE-----





Archive powered by MHonArc 2.6.24.

Top of Page