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

PowerMac7,3 insists it has 0 batteries and no AC power!



Hi,

I have a PowerMac7,3 machine (dual G5 970FX). It does not run Linux
continously just yet. So I installed anacron as I do for all such
computers. This is a stock Debian sarge installation (in particular, I
do not compile my own kernel).

Over the last few weeks I've noticed the dlocate database is always
out of date....it seems the kernel insists I am running on battery
power with zero batteries.

------------------------------------------------------------------------
shyamal@turtle:~$ sudo /etc/init.d/anacron start
Starting anac(h)ronistic cron: deferred while on battery power.

shyamal@turtle:~$ cat /proc/pmu/info
PMU driver version     : 2
PMU firmware version   : 0c
AC Power               : 0
Battery count          : 0

shyamal@turtle:~$ dmesg | grep -i pmu
via-pmu: Server Mode is disabled
PMU driver 2 initialized for Core99, firmware: 0c

shyamal@turtle:~$ grep -i pmu /boot/config-2.6.8-power4-smp
CONFIG_ADB_PMU=y

shyamal@turtle:~$ grep CONFIG_PMAC_PBOOK /boot/config-2.6.8-power4-smp
CONFIG_PMAC_PBOOK=y

shyamal@turtle:~$ uname -r
2.6.8-power4-smp
shyamal@turtle:~$ dpkg -l | grep kernel-image
ii  kernel-image-2 2.6.8-9        Linux kernel image for 2.6.8-power4
ii  kernel-image-2 2.6.8-9        Linux kernel image for 2.6.8-power4-smp
------------------------------------------------------------------------

I did poke around in via-pmu.c for a while, but between the fact that
I am recovering from some serious RSI and this comment at the top of
the file

 * THIS DRIVER IS BECOMING A TOTAL MESS !

I decided to immediately ask for help here :-) 

It does seem that the battery request simply does not set the bit for
AC power on this hardware (in done_battery_state_smart), and this
machine is not behaving like a typical power book (for which the
Debian kernel compiles code in), so, well......are there different
types of core99 (Key Largo) PMUs? Or is this the Debian "one kernel to
rule them all" approach showing the limits of how the driver is
written?

I'd be happy to volunteer to build and/or test fixes or provide more
information, etc. etc. within limits of the fact that I am recovering
from a spectacular bout with RSI.

I suspect I should file a bug against the Debian powerpc kernel -
right??? Or did I miss the forest for the trees?

Cheers!
Shyamal



Reply to: