so, i gather that this dedicated pci bus connects to the graphics
chipset.
and since it appears as another pci device, it must implement the pci configuration space, and thus it has an expansion ROM base address register which we can determine if a BIOS exist or not. If it has one, then it is likely that the initialization of the graphics chip is done by the this BIOS instead of the system BIOS.
Unfortunately it may not be that simple on motherboards with integrated graphics. There is only one NVRAM chip. This NVRAM is owned by the PIIX4E bridge. I am not sure how they would do the expansion ROM base address for the AGP stuff.
But it is certainly worth a look.
If there is only one NVRAM the Video BIOS must be included in the system BIOS.
Maybe you can do the following: - start with your original BIOS - copy the Video BIOS to a file (it starts at 0xC0000) - include the file into your BIOS - move the Video BIOS to 0xC0000 - start it
I think that should work.
Marcus - To unsubscribe: send mail to majordomo@freiburg.linux.de with 'unsubscribe openbios' in the body of the message
On Tue, 22 Feb 2000, Marcus Gossner wrote:
Maybe you can do the following:
- start with your original BIOS
- copy the Video BIOS to a file (it starts at 0xC0000)
- include the file into your BIOS
- move the Video BIOS to 0xC0000
- start it
Good idea and easily done.
Thanks
ron
- To unsubscribe: send mail to majordomo@freiburg.linux.de with 'unsubscribe openbios' in the body of the message
The problem with this idea is that the video BIOS, while included in the flash device, is probably compressed and won't be easy to identify. Many BIOS implementations need to use compression to fit all of the necessary code into the flash part. The code at the reset vector is uncompressed so that the system can start booting. Once RAM is available the BIOS can uncompress initialization code for things like PnP and PCI and execute them out of RAM. Since this code isn't used at runtime it can easily be discarded. Near the end of POST the final runtime BIOS is shadowed into F000:0000h and the system will boot an OS. Option ROMs like video and SCSI, graphic splash screens, multi-language support, etc. will almost always be compressed.
Dave
-----Original Message----- From: owner-openbios@elvis.informatik.uni-freiburg.de [mailto:owner-openbios@elvis.informatik.uni-freiburg.de]On Behalf Of Marcus Gossner Sent: Tuesday, February 22, 2000 4:54 AM To: 'openbios@freiburg.linux.de' Subject: RE: [OpenBIOS] Video (particularly 440GX)
so, i gather that this dedicated pci bus connects to the graphics
chipset.
and since it appears as another pci device, it must implement the pci configuration space, and thus it has an expansion ROM base address register which we can determine if a BIOS exist or not. If it has one, then it is likely that the initialization of the graphics chip is done by the this BIOS instead of the system BIOS.
Unfortunately it may not be that simple on motherboards with integrated graphics. There is only one NVRAM chip. This NVRAM is owned by the PIIX4E bridge. I am not sure how they would do the expansion ROM base address for the AGP stuff.
But it is certainly worth a look.
If there is only one NVRAM the Video BIOS must be included in the system BIOS.
Maybe you can do the following: - start with your original BIOS - copy the Video BIOS to a file (it starts at 0xC0000) - include the file into your BIOS - move the Video BIOS to 0xC0000 - start it
I think that should work.
Marcus - To unsubscribe: send mail to majordomo@freiburg.linux.de with 'unsubscribe openbios' in the body of the message
- To unsubscribe: send mail to majordomo@freiburg.linux.de with 'unsubscribe openbios' in the body of the message
On Tue, 22 Feb 2000, David Christensen wrote:
The problem with this idea is that the video BIOS, while included in the flash device, is probably compressed and won't be easy to identify.
interesting. Do you think they compressed it even in the 1M part?
I'm going to try not to use it anyway, but it would be good to know ...
thanks ron
- To unsubscribe: send mail to majordomo@freiburg.linux.de with 'unsubscribe openbios' in the body of the message
-
David Christensen -
Marcus Gossner -
Ronald G. Minnich