Re: Re (4): xorg-server failing on IBM NetVista with Intel 82815 video
On Thu, 20 May 2010 13:43:54 -0400 (EDT), Peter Easthope wrote:
> Stephen Powell wrote:
>> I suspect that something not
>> directly related to X is causing X to have a problem of some sort.
>> Sorry I can't be more specific at this point.
>
> http://carnot.pathology.ubc.ca/dalton.dmesg
> and
> http://carnot.pathology.ubc.ca/dalton.syslog
> are accessable now.
>
> For me these lines in dmesg are interesting.
>
> [ 0.134235] vgaarb: device added: PCI:0000:00:02.0,decodes=io+mem,owns=io+mem,locks=none
> [ 0.134328] vgaarb: loaded
> [ 0.134626] PCI: Using ACPI for IRQ routing
> [ 0.135002] Switching to clocksource tsc
> [ 0.136001] pnp: PnP ACPI init
> [ 0.136001] ACPI: bus type pnp registered
> [ 0.142666] pnp: PnP ACPI: found 16 devices
> [ 0.143884] ACPI: ACPI bus type pnp unregistered
> [ 0.143947] PnPBIOS: Disabled by ACPI PNP
> ...
> [ 91.894410] [drm] Initialized i810 1.4.0 20030605 for 0000:00:02.0 on minor 0
> [ 91.895429] mtrr: base(0xf8000000) is not aligned on a size(0x258000) boundary
> [ 92.045703] [drm] Using v1.4 init.
> [ 95.240017] [drm:drm_reclaim_locked_buffers] *ERROR* reclaim_buffers_locked() deadlock. Please rework this
> [ 95.240025] driver to use reclaim_buffers_idlelocked() instead.
> [ 95.240029] I will go on reclaiming the buffers anyway.
They are interesting to me too, especially the drm and mtrr messages.
This may be a kernel bug.
>
> In the BIOS Setup I have
> Advanced Setup
> Plug and Play Control
> Plug and Play Operating System [Yes]
>
> Power Management
> ACPI BIOS IRQ [IRQ 9]
> ACPI Standby Mode [S3]
> .
>
> Changeing to "Plug and Play Operating System [No]" doesn't help.
That's for operating systems that don't have drivers to initialize
plug and play devices, such as DOS.
>
> Can anyone comment on
> PnPBIOS: Disabled by ACPI PNP
> or
> mtrr: base ... not aligned
> ?
PnPBIOS: Disabled by ACPI PNP
is a normal message when ACPI is active. The mtrr message I've
never seen before. Or at least I haven't noticed it before.
There is a newer kernel available in squeeze now, 2.6.32-5.
Maybe it will fix the problem.
# aptitude update
# aptitude full-upgrade
This will also be a good test of your kernel installation environment
changes. If everything was customized correctly, the symlinks should
be automatically updated and lilo should be automatically run. Since
only the two most recently installed kernels will normally be bootable,
you will normally want to manually purge any older kernels that are
still installed to save disk space *after* the upgrade. For example:
aptitude purge linux-image-2.6.32-trunk-686
--
.''`. Stephen Powell
: :' :
`. `'`
`-
Reply to: