[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#637659: Linux kernel: 2.6.32.42: (Debian Bug#637659): pci 0000:00:00.0: BAR 0: address space collision on of device [0xc8000000-0xcfffffff]



Dear Jonathan,


as always, thank you very much for your prompt help and responses.

Am Montag, den 10.10.2011, 04:46 -0500 schrieb Jonathan Nieder:

> Paul Menzel wrote:
> 
> > Commit 29cf7a30 [3] addresses this issue for my board by adding a quirk.
> >
> >         commit 29cf7a30f8a0ce4af2406d93d5a332099be26923
> >         Author: Paul Menzel <paulepanter@users.sourceforge.net>
> >         Date:   Wed Aug 31 17:07:10 2011 +0200
> >
> >             x86/PCI: use host bridge _CRS info on ASUS M2V-MX SE
> [...]
> > Unfortunately I do not have outputs from `dmesg` anywhere from older
> > Linux versions so I do not know when this problem started. I am pretty
> > sure though that they have not been there since the beginning.
> 
> Yep, the regression fixed by 29cf7a30 was likely introduced by
> v2.6.31-rc1~1 (Revert "PCI: use ACPI _CRS data by default",
> 2009-06-24).

The following commit is the reverted one.

        $ git show 9e9f46c44
        commit 9e9f46c44e487af0a82eb61b624553e2f7118f5b
        Author: Jesse Barnes <jbarnes@virtuousgeek.org>
        Date:   Thu Jun 11 10:58:28 2009 -0700
        
            PCI: use ACPI _CRS data by default

        $ git describe 9e9f46c44
        v2.6.30-22-g9e9f46c

> A partial fix was v2.6.34-rc1~287^2~2 (x86/PCI: use host bridge _CRS
> info by default on 2008 and newer machines, 2010-02-23), which does
> not seem to be part of v2.6.32.y (and Debian squeeze) but probably
> should be.

Additionally that would not be enough for my board as this is an “old“
machine from 2006.

Additionally as guessed in Linux 2.6.32.x there seems to be no way to
force the use of _CRS information using the Linux kernel command line.

        $ more /proc/cmdline 
        BOOT_IMAGE=/vmlinuz-2.6.32-5-amd64 root=/dev/mapper/storage-root ro quiet pci=use_crs

Furthermore searching for »Linux 2.6.32 address space collision on of
device« on the Web there are several hits like Debian bug report #567435
[5] so it would be worth to fix in my opinion.

Ben and Maximilian state in the report thread [5] that the message is
harmless. But since my system occasionally freezes it would be great if
you could confirm that.


Thanks,

Paul


[5] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=567435

Attachment: signature.asc
Description: This is a digitally signed message part


Reply to: