Skip to Content.
Sympa Menu

baslinux - [BL] BL3 xvesa on a 386

baslinux AT lists.ibiblio.org

Subject: Baslinux mailing list

List archive

Chronological Thread  
  • From: James Miller <jamtat AT mailsnare.net>
  • To: baslinux AT lists.ibiblio.org
  • Subject: [BL] BL3 xvesa on a 386
  • Date: Thu, 5 Feb 2004 23:54:05 -0600 (CST)

Ok. I can start spending some more "spare time" on this project now that
we're moved in and more or less settled. I did set up a static network
for the 386, btw.

On Sat, 17 Jan 2004 3aoo-cvfd AT dea.spamcon.org wrote:

> James Miller wrote:
> >
> > > ---------------------------------------------
> > > http://basiclinux.com.ru/download/x-initrd.gz
> > > ---------------------------------------------
> >
> > Managed to sneak in a test during final stages of packing.
> > did startx: screen goes blank black and just sits there.
> > After a short time, if I do ctrl-alt-F2
>
> ctrl-alt-backspace is the standard exit from X.

I tried that. It doesn't work. Only ctrl-alt-F2, then back to the first
tty by ctrl-alt-F1 gets any sort of display back (console mode).
ctrl-alt-bkspc just doesn't do anything.

> > with some error messages on the screen as follows:
> >
> > sigaction for smart scheduler: Function not implemented
> > Interrupt pointer doesn't point at ROM
>
> I've seen that "Interrupt pointer doesn't point at ROM".
> It happened when my DOS configuration was interferring with
> Xvesa. I suggest you boot a bare DOS, without CONFIG.SYS
> and AUTOEXEC.BAT, and see if that makes a difference.

Tried a bare DOS. With either empty, or missing autoexec.bat and
config.sys, I get the same problem: blank dark screen. Even ctrl-alt-F2
won't work in this case, though. I can't get any sort of display back no
matter what I try/have tried. Funny that when I use normal autoexec.bat
and config.sys files, I can at least get to a console by shifting between
virtual consoles.

> > > Fatal server error:
> > no screens found
> > Memory fault
> >
> > Too little RAM, perhaps?
>
> That seems to be what the "Memory fault" is indicating.
> There are several ways we could try to get around this:
> (1) using a swap partition -or-
> (2) turning off shadow RAM in the BIOS -or-
> (3) rawriting x-initrd to a floppy -or-
> (4) getting x-initrd onto an HD partition
>
> Once you are settled in after your move, we can have
> another look at this.

I do have a swap partition on the HD on this machine (10MB), so 1) is an
option. Thus far, I have been unable to get into BIOS setup (I've tried
many key combinations, but I haven't hit the right one or at the exact
right moment yet), so 2) looks less feasible for now. I know I've been in
this BIOS before, and with enough trial and error I can get in again. 3)
is also a definite possibility, but I don't quite understand why it would
work while the current setup (loading from HD) doesn't. As for 4), I
don't fully understand: x-initrd *is* on the HD (the DOS partition).
Should it be on the Linux partition under this scenario?

Just to be sure I understand fundamentally what's going on here: I'm
expecting to get the cross-hatched screen that you get under Xwindows when
you have X installed, but no window desktop is specified, with the mouse
cursor in it. Is that what I should be expecting? I know eventually it
should get an xsession offer from another machine on the network, but I'm
not quite clear on how to set that up (it was in one of the init scripts
in another BL xterminal project I did: X -query 192.168.x.x - something
like that). But I'm thinking of that as some further stage in the
project. For right now, I'm guessing I should just be looking for the
typical signs that a window mgr is loading - the crosshatched screen with
the mouse pointer in it. Am I expecting correctly? I hope so, because
not only have I not told the xterm where to look for an xsession display
export, but I have yet to set up the other machine to offer the xsession.

To be continued tomorrow evening, I hope.

James




Archive powered by MHonArc 2.6.24.

Top of Page