On Mon, 1 Oct 2001, Alan Cox wrote:
> > I'm asking for application of a patch I'm using since 2.4.0. Issue was the
> > BIOS of my OB-800 mapped the memory regions of the cardbus bridges into
> > legacy 1M-area (0xe6000 e.g.). Despite being pretty bogus this used to
> > work after a reboot just until the first pm-suspend/resume, where the
> > hostbridge somehow looses access to this area.
>
> At a guess your bios doesnt restore the shadow ram disables right.
In fact, there is a (IIRC single) byte changed in the hostbridge's config
space after resume. Blindly writing back the old value re-enables access
to the cardbus memory resource below 1M. But I agree solving the real
cause (i.e. bogus mapping to legacy area) is better then trying to cure
whatever triggers the problem with that.
> Would a generic
>
> pci_fixup_device()
>
> type function not be more appropriate
IIRC there was some discussion on l-k some time ago, whether such fixups
should stay close to were the issue appeared or better be placed at
generic pci quirk location. In this particular case I personally tend to
the latter one as well, particularly because there may be other BIOS'
doing similar things with non-cardbus memory bar's.
The patch however was simply what I'm using since 2.4.0 just in case the
cardbus-specific fixup would be the prefered way.
Martin
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Sep 30 2001 - 21:01:15 EST