sgilinux
[Top] [All Lists]

Re: 1600SW

To: sparker@xxxxxxxxxxx, Thomas Duffy <tduffy@xxxxxxxxxxxxxxxxxxxx>
Subject: Re: 1600SW
From: sparker@xxxxxxxxxxx (Steven G. Parker)
Date: Wed, 6 Sep 2000 15:44:41 -0600 (MDT)
Cc: sgilinux@xxxxxxxxxxx
Sender: owner-sgilinux@xxxxxxxxxxx
> On Wed, 6 Sep 2000 sparker@xxxxxxxxxxx wrote:
> 
> > I have a new 330 with a 1600SW and the multilink adapter.  Has anybody
> > successfully made that work with XFree86?
> > 
> > Easy question:
> >  - I can get it to work with the "nv" driver and a regular monitor.
> >    However, I cannot get it to sync up with the 1600SW.  Does anyone
> >    have a modeline?
> 
> Have you tried the modelines listed on Xfree86.org for the flatpanel?
> 

[snip]
Yes.  If I use these, I get "unsupported mode" from the multilink.
If I use the ones in the XF86Config_sgivpro, the multilink doesn't
complain, but I get blue and yellow fuzz.   If you cross your eyes
just right, you can see a glimmer of an xterm in there somewhere.

The multilink is also supposedly able to lock onto VESA 1280x1024
(and other) resolutions.  However, I get the same result - blue/yellow
fuzz.  It works great in windows though (at 1280x1024 - haven't figured
out 1600x1024 yet but that is a different story).   Win must know
something that I don't.

> > Harder question:
> >  - The sgivpro and nvidia drivers both hang.  The sgi one core dumps
> >    the x server at startup, and the nvidia driver hangs the whole
> >    machine.
> 
> do you have the kernel module, sgivpro loaded at the time you start X?
Yes.  This is the message from X, just before it chokes:
(==) NVIDIA(0): Write-combining range (0x88000000,0x4000000)

It loads sgivpro, mouse, vgahw,  and then it recognizes quadro.  Then
it sets up the ramdac, loads cfb, mfb, xaa and ramdac.  Then it prints
out the above message and hangs the machine.

As far as I can tell, the behavior is the same whether I load the
module or not.

Steve



<Prev in Thread] Current Thread [Next in Thread>
  • 1600SW, Steven G. Parker
    • Re: 1600SW, Steven G. Parker <=