MP-BIOS bug: 8254 timer; laptop won't stay connected to AC

[Russell Reiter]

I'm trying to get a Toshiba Satellite Pro A110 working with Debian
Wheezy. I get
this error -  MP-BIOS bug: 8254 timer not connected to IO-APIC

When I startup and log into the desktop, the AC is connected and the
battery charges normally then in a few moments disconnects. I can
navigate the desktop a little but if I open a terminal the resulting
kernel events trigger the switch to battery power.

This notebook is circa 2006 and the bios is Phoenix  from 2004.

dmesg shows the  ACPI 's DSDT was compiled by MS and there is a known
issue with the miss-referenced location of the 8254 timer.

I made a copy of /sys/firmware/acpi/tables/DSDT and decompiled it with iasl.
I recompiled after I fixed the various method issues and updated grub
to use the modified table.

udevadm shows
KERNEL[4638.288193] change
/devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A03:00/device:00/ACPI0003:00/power_supply/ACAD
(power_supply)
UDEV  [4638.297613] change
/devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A03:00/device:00/ACPI0003:00/power_supply/ACAD
(power_supply)
KERNEL[4639.300775] change
/devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A03:00/device:00/PNP0C0A:00/power_supply/BAT1
(power_supply)
UDEV  [4639.302402] change
/devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A03:00/device:00/PNP0C0A:00/power_supply/BAT1
(power_supply)

If I leave the system to it's own devices, something on loopback, I
guess, triggers the disconnect from the AC. If I run udevadm monitor,
it seems to run as a tunnel and keeps the AC alive and the battery
charges. If I open a new terminal the AC disconnects.

Has anyone slogged all the way through the southbridge to get to the clock?

This is beginning to feel like that old WW2 movie "The Bridges At
Remagin" someone changed all the signs.

Thanks
Russell
--
The Toronto Linux Users Group.      Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists