On Sun, 2 Mar 2003, Christoph Pittracher wrote:
> On Saturday 01 March 2003 03:18, Mark Vojkovich wrote:
> > Can you VT switch back to the X-server OK, and startx again after
> > quitting?
>
> Yes, I can.
>
> > If so, it's probably just a matter of not enough stuff
> > being saved and restored in the UnLoad/LoadStateExt functions in
> > riva_hw.c.
>
> Hmm, how can I test that? Is there a list of all registers that
> exists/should be saved?
If I knew of more that should be saved, it would be saving
them already.
>
> > If you wanted to collect some data you could try reading
> > chip->PMC[0x00000004/4] at the top of nv10GetConfig() just to
> > see what state that was in before the X-server touched it.
>
> What should I do with the data found at chip->PMC[0x00000004/4]? I
> didn't take a closer look at the source code, is chip->PMC a struct
> including all registers or how do I know which byte is which register?
chip->PMC[0x00000004/4] is a 32 word. You can just print it out.
ErrorF("chip->PMC[0x00000004/4] = 0x%x\n", chip->PMC[0x00000004/4]);
will send it to the /var/log/XFree86.0.log.
That's just a shot in the dark. There's probably nothing wrong there
since it starts succesfully the second time.
I'll ask some people about what's different between the
Apple and PC mobile configurations.
Mark.
_______________________________________________
Devel mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/devel
