YDL 3.0 Crashes 800 MHz Titaniam Laptop

John Howland yellowdog-general@lists.terrasoftsolutions.com
Fri Jun 20 16:22:01 2003


On Fri, 20 Jun 2003, John Howland wrote:

> I have a G4 Titanium laptop which repeatededly crashes
> while running YDL 3.0.  The machine runs the Apple diagonistics
> and seems to run OSX in reliable fashion.  Does anyone else
> have this experience?

This is one of the kinds of failures I have seen:

Hello,
does this (see following between snip/snap) during startup of the first CD:

------------snip--------------
Oops: Kernel access of bad area, sig. 11
NIP: C01E5670  XER: 20000000  LR: C01E64A0  SP: C0273BC0  REGS: c0273d10
TRAP: 0300 Not tainted
MSR: 00009032  EF: 1  PR: 0  FP: 0  ME: 1  IR/DR: 11
DAR: 20000000, DSISR: 40000000
TASK=c0272000[11] 'loader' last syscall: 3
last math c0272000  last activec 00000000
....
----------snap----------------

mean, my machine (NewWorld iMac) has a hardware problem? Like RAM?
(I tried all the install CD images I found up to the latest, May 30th)

Best Regards and Have a nice day,
Igor

> 
> I have installed everything and found that
> pwrctl and pwrctl-local in /etc/power were not executable.
> Various pmud features were not available until file modes
> were changed.  I found that by default pmud runs the
> processor at 667 MHz rather than 800 MHz even when running
> on AC power.  I introduced the code:
> 
> case "$2" in
> ac)
>         $logger -p daemon.info -t pwrctl-local "setting cpu speed 800000 MHz"
>         echo -n "800000:800000:performance" > /proc/cpufreq
> ;;
> battery)
>         $logger -p daemon.info -t pwrctl-local "setting cpu speed 667000 MHz"
>         echo -n "667000:667000:powersave" > /proc/cpufreq
> ;;
> esac
> 
> in pwrctl-local to run faster when on AC.  I cannot trace the kernel
> crashes to this change as the machine crashes when pwrctl and
> pwrctl-local and pmud are disabled.
> 
> I have compiled kernels but have been unable to achieve the same
> kernel features as YDL 3.0 and the kernels I've built also
> crash the machine.  The crashes are easiest to produce by compiling
> a kernel, but sometimes come when the machine is not being used.
> 
> I would appreciate any input or suggestions before I send the
> machine back to Apple (for the second time) for repairs.
> 
> 

-- 
_______________________________________________________________
John E. Howland       url: http://www.cs.trinity.edu/~jhowland/
Computer Science    email: jhowland@ariel.cs.trinity.edu
Trinity University  voice: (210) 999-7364
One Trinity Place     fax: (210) 999-7477
San Antonio, Texas  78212-7200