Re: 2.6.0-test4-mm6

[Matt Mackall <mpm@selenic.com>]

On Fri, Sep 05, 2003 at 11:29:14PM -0500, Matt Mackall wrote:
> On Sat, Sep 06, 2003 at 03:07:58AM +0200, Jan Ischebeck wrote:
> > Am Fr, 2003-09-05 um 16.51 schrieb
> > viro@parcelfarce.linux.theplanet.co.uk:
> > > On Fri, Sep 05, 2003 at 02:46:40PM +0200, Jan Ischebeck wrote:
> > > > Seems like I got the reason for X not starting:
> > > > 
> > > > pseudo terminals can't be acquired and only two consoles are
> > running.
> > > > 
> > > > -> X11 can't get console Vt7
> > > > -> pppd doesn't work either
> > > > 
> > > > This definitely worked with -mm5.
> > > 
> > > Grr...  Dumb typo.  Patch below should fix that...
> > > 
> > > diff -urN B4-misc3/drivers/char/tty_io.c
> > B4-current/drivers/char/tty_io.c
> > > +++ B4-current/drivers/char/tty_io.c  Fri Sep  5 10:46:59 2003
> > > @@ -1334,7 +1334,7 @@
> > >               return -ENODEV;
> > >       }
> > >  
> > > -     if (device == MKDEV(TTY_MAJOR,2)) {
> > > +     if (device == MKDEV(TTYAUX_MAJOR,2)) {
> > >  #ifdef CONFIG_UNIX98_PTYS
> > >               /* find a device that is not in use. */
> > >               retval = -1;
> > 
> > Thank you, that solved the problem with X.
> > 
> > But PPP is still broken, I get the following errors when I've tried to
> > setup a connection. And after trying to use ppp the machine oops at
> > shutdown.
> 
> Probably related badness: ttyS[01] have disappeared from
> /proc/interrupts (though still show up in dmesg) and while serial
> console appears to still work, kgdb over serial appears to have gone
> south as well. Last tested in -mm4.

Ok, kgdb over serial works again if you boot without the gdbeth
options (the comment on kgdb_net_interrupt is a bit misleading here).
This causes KGDB-stub to show up again in /proc/interrupts, but my
other serial interrupt is still missing.

-- 
Matt Mackall : http://www.selenic.com : of or relating to the moon