Bug#490694: Extra video card seems to be the problem
Hi Jonathan,
I have a yes and no answer for you. Yes, I'm not sure whether the problem still exists and
consequently am assuming it does. And no since unfortunately my powermac is not hooked up
right now and cannot easily be set up since I'm without monitor for the time being. I'm living
from a laptop at the moment. If I could get my hands on a monitor, then I'd be willing to set it
up (and track down my extra graphics card that I had removed from the system to get it to work).
One little piece of into that I'm not sure I gave previously is that I think this problem, or a very similar one,
appeared in the circa 2002 time frame. At that point I had Yellow Dog or Mandrake (I would switch between
the two) installed on it. At some point, it broke similarly and my solution was a custom build of the kernel
where I had gone in and reordered the graphics card "attempts", i.e. probes. (iirc) It was along the lines of the graphics
card (whick seems odd) rather than the PCI bus. That is odd enough to suspect my memory has become
fouled with age, but anyway that's the way I remember it. One fact is, it did break, it was graphics card
related, and modifying the kermel "fixed" it.
So, I'm not sure if I've helped in terrms of what to do with this bug (at this moment). If you'd really like
to get to the bottom of the issue, then I might (just might) be able to track a monitor down for some
testing purposes.
regards,
-jeff
On Feb 15, 2012, at 6:34 PM, Jonathan Nieder wrote:
> tags 490694 + moreinfo
> quit
>
> Hi Jeff,
>
> Jeffrey B. Green wrote:
>
>> After doing a "lspci -vt" and noticing the extra video card that I've
>> not used for years, i.e. forgot totally about it, I tried the 2.6.25
>> kernel with the extra video card removed (0001:11:04.0). It booted just
>> fine.
>
> Thanks for the detective work.
>
> Are you still interested in pursuing this? No promises, but if this
> is still reproducible with current (3.x.y) kernels, someone at
> linuxppc-dev@lists.ozlabs.org might be able to give some hints. I'd
> suggest coordinating work at [1] so it's easy for interested parties
> to track.
>
> If you do not have access to this hardware or are no longer interested
> in pursuing this for some other reason, that's fine too, but please
> let us know so we can stop tracking it.
>
> Thanks,
> Jonathan
>
> [1] https://bugzilla.kernel.org/show_bug.cgi?id=12436
Reply to: