Hi,
I've had trouble with ACPI support under SuSE 9.3 and 10.0 before (kernel
versions: 2.6.11 and 2.6.13), so that I switched acpi support off altogether
with a boot parameter. I hoped ACPI would work better with SuSE 10.1 and
waited...
I've updated my laptop a few days ago and noticed that now I can at least
leave the boot parameter away (in earlier versions that resulted in a CPU
load of 100% and tons of identical debug messages written
to /var/log/messages). Now the CPU load is ok, but I don't get a login on
console 1, because loading stops when acpid is started. Once I kill that
process from the X console, booting continues and the (text) login on console
1 appears eventually.
My next try was to use that Intel tool to dump the DSDT from /proc and
disassemble it. Worked fine. I did not change anything in the text file I got
out of the disassembly, but compiled it directly again, without encountering
any errors. I've put the freshly compiled DSDT file into the initrd (via
parameter in /etc/sysconfig/kernel) by calling mk_initrd and rebooted.
Now, acpid starts, but there are still plenty of error messages (that do not
enlighten me very much about the origin of the problem) appearing
in /var/log/messages. I can't adjust display brightness, the powersave
subsystem doesn't appear to know much about the battery state, I can't adjust
the CPU performance and "echo 1 > /proc/acpi/asus/[wm]led" doesn't have any
effect either (the latter is supposed to switch on the wireless/mail LEDs).
I've put the asus_acpi kernel module into the initrd. Seems to load fine
(from /var/log/boot.msg):
Asus Laptop ACPI Extras version 0.30-SUSE
M6R model detected, supported
Now, can anyone here give me a pointer what I should do to get things running?
I'm using the machine quite often with the battery and it happend several
times a week that I don't notice the blinking battery LED until the laptop
just switches off... I understand ACPI could help me avoid this by triggering
some event such as playing a warning sound or something?
I had a quick look through the disassembled DSDT code and noticed some
"Windows NT" string there. I remember having played (long time ago) with some
boot paramenter that fools the DSDT into believing that Windows NT would be
running. However, I can't find the webpage anymore... Would that be a
possible solution? If yes, what is that boot parameter?
All required files are attached.
Thanks for any help,
Raimund Eimann
Today's wisdom:
Engineering: 'How will this work?'
Science: 'Why will this work?
Management: 'When will this work?'
Liberal Arts: 'Do you want fries with that?'
Attachment:
acpidump.txt.bz2
Description: BZip2 compressed data
Attachment:
lspci.txt.bz2
Description: BZip2 compressed data
Attachment:
proc_interrupts.txt.bz2
Description: BZip2 compressed data
Attachment:
var_log_messages.txt.bz2
Description: BZip2 compressed data
Attachment:
dmidecode.txt.bz2
Description: BZip2 compressed data
