Re: PCI Sanity check breaks on Compaq Deskpro XL 590 traced to 2.1.115

Gabriel Paubert (paubert@iram.es)
Tue, 6 Oct 1998 17:57:42 +0100 (MET)


On Mon, 5 Oct 1998, Martin Giguere wrote:

> I thought I would send this to the kernel mailing list just in case
> someone else has the same problem as me.

Not me, but let us try to help:

> I own a Compaq Deskpro XL 590 and I am having problems with PCI and the
> latest kernels. I would like to point out two things I have notice
> (after many hours of kernel compiling and testing ;> )
>
> At the bottom of the message is as much info as I though possible to
> send. Let me know if you need more.
>
> First the PCI Sanity check intruced in 2.1.115 is invalide on my
> machine. I added a "return 1" in the sanity check function to see, and
> fair enought PCI started working again.

This sanity check is unfortunmatey absolutely necessary on some machines.

> Second the directory address changes from the stable kernel to the
> development kernels (even in 2.1.1) (Ever tryed booting one hundred
> kernels hehe)

What do you mean (privately if you prefer and, a wild guess from your
e-mail address and name, even in french if you find it easier).

> --/proc/pci
>
> PCI devices found:
> Bus 0, device 0, function 0:
> Non-VGA device: Compaq Unknown device (rev 1).
> Vendor id=e11. Device id=1000.
> Medium devsel. Master Capable. Latency=32.
> Non-prefetchable 32 bit memory at 0x2000010 [0x2000010].
> Non-prefetchable 32 bit memory at 0x2000010 [0x2000010].
> I/O at 0x10000e10 [0x10000e11].
> I/O at 0xe2000144 [0xe2000147].

That device is the problem. On most MB, bus 0 device 0 corresponds to the
host bridge: you don't have a very common chipset (Intel or whatever).
These register addresses are obviously bogus, except perhaps the first
one. If you have downloaded the pciutils package, please use lspci -vvx,
else use:

od -tx1 -Ax /proc/bus/pci/00/00.0

under 2.1.1xx, and send back the result.

All the other devices look ok.

Regards,
Gabriel.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/