2002-01-17 13:03:26

by Andre Hedrick

[permalink] [raw]
Subject: XP PCI Contamination, GURR (Re: Care?)


It appears the folks up in redmond have succeeded in having the BIOS
people default disable PCI resources. Since XP will reject, or assume a
device is in use should the BAR's be allocated, the various archs may need
to have a broader setup table or a more generic ruleset.

Any thoughts on how best to address good hardare, which the BIOS does not
setup per redmond-rules.


On Mon, 17 Dec 2001, Jeremy Freeman wrote:

> Not sure if you care about this or not.. Probably seen it from someone else
> already.. Dual Athlon 1800+ MPs on a Tyain Thunder K7 BIOS 2.09.
>
> 2.4.16 kernel.
>
> <snip>
> AMD7411: IDE controller on PCI bus 00 dev 39
> PCI: Device 00:07.1 not available because of resource collisions
> AMD7411: chipset revision 1
> AMD7411: not 100% native mode: will probe irqs later
> AMD7411: IO baseregs (BIOS) are reported as MEM, report to
> <[email protected]>.
> AMD7411: simplex device: DMA disabled
> ide1: AMD7411 Bus-Master DMA disabled (BIOS)
> hda: ATAPI 52X CDROM, ATAPI CD/DVD-ROM drive
> ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
> hda: ATAPI 48X CD-ROM drive, 128kB Cache
> <snip>
>

Andre Hedrick
Linux Disk Certification Project Linux ATA Development


2002-01-17 13:14:42

by Jeff Garzik

[permalink] [raw]
Subject: Re: XP PCI Contamination, GURR (Re: Care?)

Andre Hedrick wrote:
>
> It appears the folks up in redmond have succeeded in having the BIOS
> people default disable PCI resources. Since XP will reject, or assume a
> device is in use should the BAR's be allocated, the various archs may need
> to have a broader setup table or a more generic ruleset.
>
> Any thoughts on how best to address good hardare, which the BIOS does not
> setup per redmond-rules.

pci_enable_enable, pci_set_master, and all the tools that the kernel
already provides...

--
Jeff Garzik | Alternate titles for LOTR:
Building 1024 | Fast Times at Uruk-Hai
MandrakeSoft | The Took, the Elf, His Daughter and Her Lover
| Samwise Gamgee: International Hobbit of Mystery

2002-01-17 13:18:02

by Dave Jones

[permalink] [raw]
Subject: Re: XP PCI Contamination, GURR (Re: Care?)

On Thu, 17 Jan 2002, Andre Hedrick wrote:

> It appears the folks up in redmond have succeeded in having the BIOS
> people default disable PCI resources. Since XP will reject, or assume a
> device is in use should the BAR's be allocated, the various archs may need
> to have a broader setup table or a more generic ruleset.
> Any thoughts on how best to address good hardare, which the BIOS does not
> setup per redmond-rules.

Why do I have this sneaking suspicion that ACPI will play a part here.

--
| Dave Jones. http://www.codemonkey.org.uk
| SuSE Labs