i386/kernel/pci-pc.c and tdfxfb.c

i386/kernel/pci-pc.c and tdfxfb.c

[Perry Gilfillan]

I've got two questions, and I'm not sure if they will be related in any way.

My equipment:  I have a Voodoo3 and VIA MVP3/Pro133x AGP and Apollo VP3

The first is that in 2.4.18, arch/i386/kernel/pci-pc.c had pci_fixup for the VIA VT82C597, 598, and 691 bridge chip sets.  

All referenc to the VT82C5xx chips seem to be removed.

If I recall corectly, ( I might not ) on 2.4.18, when I removed the lines from the pcibios_fixups struct, and teh procedures associated with them, it cleared up the corupted boot logo.

The logo is corupted again.

The second, and more critical problem is that if tdfxfb is compiled in or set as a module, all consoles go black on black when X comes up.

I didn't see any refrence to tdfxfb.c in the 2.4.19 patch, and I don't know where else to look.

With tdfxfb out of the picture, the consoles are fine.

If anyone can point me in the right direction, I'm glad to try any sugestions, including patches, to resolve this problem.

Regads,
           Perry Gilfillan

-- 
Get your free email from www.linuxmail.org 


Powered by Outblaze

Re: i386/kernel/pci-pc.c and tdfxfb.c

[Alan Cox]

On Sat, 2002-08-03 at 23:36, Perry Gilfillan wrote:
> The first is that in 2.4.18, arch/i386/kernel/pci-pc.c had pci_fixup for 
> the VIA VT82C597, 598, and 691 bridge chip sets.  
> 
> All referenc to the VT82C5xx chips seem to be remoed.

Actually I'm quite fond of my data so I'd prefer to leave them present.
2.4.19 knows about the onboard video corner case which 2.4.18 didnt.

The tdfx one I can't help on. Although if its X that is making a mess
probably there is an interaction between XFree86 and the kernel driver
that wants looking at by both parties to see what code is forgetting to 
restore what values