PCI: Direct Access and STPC

From: Fabrice Gautier (gautier@email.enst.fr)
Date: Thu May 24 2001 - 04:28:20 EST

Next message: john slee: "Re: Just FYI..."
Previous message: Malcolm Beattie: "Re: Why side-effects on open(2) are evil. (was Re: [RFD w/info-PATCH]device arguments from lookup)"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Hi,

I noticed that when building a kernel with CONFIG_PCI_DIRECT and not
CONFIG_PCI_BIOS, the kernel doesn't detect any PCI bus when runing on a
STPC embeded CPU.

The STPC include a PCI chipset but the configuration methods usednin the
kernel does not seems to match the STPC docs.

The STPC use 0xCF8 and 0xCFC registers to access PCI configuration space.
This seems to match the code of pci_conf1_read_config_byte but the
sanity_check fail pci_check_direct.

The fact is that the north bridge in the STPC have a class code of 0,
so is not detected as a host bridge. For my own tests i've removed the
sanity check and so i get the following result in /proc/pci:

PCI devices found:
  Bus 0, device 11, function 0:
    Non-VGA unclassified device: PCI device 100e:0564 (Weitek) (rev 0).
  Bus 0, device 12, function 0:
    ISA bridge: PCI device 100e:55cc (Weitek) (rev 0).
  Bus 0, device 12, function 1:
    IDE interface: PCI device 100e:55cc (Weitek) (rev 0).
      I/O at 0xfc00 [0xfc07].
      I/O at 0xfc20 [0xfc23].
      I/O at 0xfc40 [0xfc47].
      I/O at 0xfc60 [0xfc63].
      I/O at 0xfc80 [0xfc8f].

I've seen some fixup functions that seems to correct some IDs. I guess
in my case I would need one to correct the device CLASS for the north
bridge. Is that correct ?

Thanks.

-- Fabrice Gautier <gautier@email.enstfr>

- 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/

Next message: john slee: "Re: Just FYI..."
Previous message: Malcolm Beattie: "Re: Why side-effects on open(2) are evil. (was Re: [RFD w/info-PATCH]device arguments from lookup)"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

This archive was generated by hypermail 2b29 : Thu May 31 2001 - 21:00:14 EST