On 6/20/08, Paolo Minazzi <paolo.minazzi@gmail.com> wrote:
Hi Jonathan,
can you do an other simple test ?

Run colinux on vista.
************************
Then digit :

# n1=$(cat /proc/interrupts | grep "0:" | awk '{ print $2 }')

wait <n> seconds on you hand-clock (Don't use sleep command ! we are
studying the PC timer taht seems broken! ). Then exec a new command

# n2=$(cat /proc/interrupts | grep "0:" | awk '{ print $2 }')

Now digit
# echo $(expr $n2 - $n1)

Get the value and divide it by <n>.

You should have a value very closed to 100. Is it ?


Run an other linux distribution that works
***********************************************
Use the same procedure written above.

You should have 100 or 1000 (or 250) .... Depends on the HZ defined
into kernel ....
What is the value you see ?

Bye,

Paolo


-------------------------------------------------------------------------

I've got a Athlon64 x2 with same bios problem (which it had not before last bios update). I can confirm Ubuntu 7.10 did not pass booting without acpi=off parameter. 
Should this help debugging I can send more data|run some test.
My 2 c

Michelangelo