Bug 202389 - broken temperature ACPI thermal_zones hp nc8430
Summary: broken temperature ACPI thermal_zones hp nc8430
Status: RESOLVED DUPLICATE of bug 179702
Alias: None
Product: SUSE Linux 10.1
Classification: openSUSE
Component: Kernel (show other bugs)
Version: Final
Hardware: x86 Other
: P5 - None : Critical (vote)
Target Milestone: ---
Assignee: Thomas Renninger
QA Contact: E-mail List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2006-08-29 11:56 UTC by rossano de
Modified: 2007-01-10 11:47 UTC (History)
3 users (show)

See Also:
Found By: Other
Services Priority:
Business Priority:
Blocker: ---
Marketing QA Status: ---
IT Deployment: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description rossano de 2006-08-29 11:56:28 UTC
I have a notebook hp compaq nc8430, the temperature remains always highest ( > 60°C), also with powersave governor, the acpi sensors seems to be broken because
if i turn on all fan the temperature increase to 100° C.
Comment 1 Thomas Renninger 2006-08-30 14:27:43 UTC
Yes, something is seriously broken in 2.6.16 and mainly HPs seem to be affected.
These may be related:
https://bugzilla.novell.com/show_bug.cgi?id=200169
http://bugzilla.kernel.org/show_bug.cgi?id=5534
http://bugzilla.kernel.org/show_bug.cgi?id=6455

All HPs with EC or other ACPI problems.

Can you try whether it works better with ec_intr=0 boot param, please.
If yes it's related to the last of above bugs and there is already a solution for a 2.6.16 kernel.
If not it's a bit harder. I am currently working on this, but I have 1 1/2 weeks of holidays now. I hope to be able to provide something soon.
Comment 2 rossano de 2006-08-30 18:09:06 UTC
ec_intr=0 don't work,
I have noticed that acpitool produces a segmentation fault and that if i turn off all funs the temperature (TZ4) become 0°C if I turn on all funs the temperature increase to 100°C. Excuse for my bad english.
Comment 3 Thomas Renninger 2006-11-08 14:52:55 UTC
Rainer, you have the same laptop and you tested the kernel from:
ftp.suse.com/pub/people/trenn/sle10_SP1_hp_fix_cpufreq_ACPI_events_kernels_third_try/

Can you confirm the kernel has this fixed? If yes 10.2 should have this fixed also and i like to close this one.
Comment 4 Thomas Renninger 2006-11-08 14:59:03 UTC
Oh sorry, you tested this kernel:
ftp.suse.com/pub/people/trenn/sle10_SP1_hp_fix_cpufreq_ACPI_events_kernels_third_try/
This has some additional fixes which are not in SP1 kernel yet. Some temperature stuff isn't even fixed mainline.

Just ignore my last comment.
However, be sure I am working on this.
I would be really glad if you guys could give the SP1 kernel a try as soon as i have all fixes for HPs collected and submitted. I will come back as soon as I committed the stuff. Thanks any feedback.
Comment 5 Rainer Klier 2006-11-08 15:13:46 UTC
ok, i will try it when you made the new kernel.
just give me a message.

question: which of these "test-kernels" is newer and has more fixes:
from ftp.suse.com/pub/people/trenn/sle10_SP1_hp_fix_cpufreq_ACPI_events_kernels_third_try/
or from
http://ftp.gwdg.de/linux/suse/projects/kernel/kotd/sles10-i386/SLES10_SP1_BRANCH/ ???

and which one should i try?
Comment 6 Thomas Renninger 2006-11-08 16:08:30 UTC
This one will be the one that gets shipped and I like you to test at later time:
http://ftp.gwdg.de/linux/suse/projects/kernel/kotd/sles10-i386/SLES10_SP1_BRANCH/

I will integrate patches from the other (third_try) kernel which I put to my ftp for testing. There will also be added some other patches. I mainly wait for comments on this bug:
http://bugzilla.kernel.org/show_bug.cgi?id=7122
whether the last issues are really fixed now.

Thanks a lot for your feedback, it's very much appreciated!
Comment 7 Rainer Klier 2006-11-08 16:28:00 UTC
ok, i wait for your message to test it.

but it seems the kernel from "SLES10_SP1_BRANCH" is a sles10 kernel and not a suse10.1 kernel.
to who will the sles10 kernel be shipped?
will it be included in the sp1 for sles10?

the only problem is, that the "third_try" kernel fits better to my current setup (suse10.1).
i have some self built-kernel modules for some hardware inside the nc8430 (tg3 network card, ati x1600 gfx-card, ti card reader, cisco-vpn-client) which will complain about sles10 kernel and will not work any more. and need to be rebuilt.
but with your "third_try" kernel everything works (ok, more or less ... ;-)
currently i am writing this running on this "third_try" kernel.
it has the same version-number as the latest official suse/novell-kernel update for suse10.1: 2.6.16.21-0.25
and therefor it also fits to the latest kernel-source for suse10.1

so after testing the sles10 sp1 kernel, will all new patches, bugfixes and enhancements also be integrated in your "third_try" kernel or in the next kernel-update for suse10.1?

and what should i do if suse/novell releases a kernel update for suse10.1?
will your fixes for the hp-notebooks (all that stuff with battery, cpu-thotteling and thermal_zones,...) be integrated in the next kernel update?
or will i loose all enhancements you integrated in the "third_try" kernel when i update?
Comment 8 Thomas Renninger 2006-11-08 17:48:28 UTC
ok, I understand. Hmm not sure whether I can force the build no. to fit, but I try.

I am not sure whether SLE10 SP1 will also be offered as some kind of 10.1 SP1. In general 10.1 has nearly the same code base at least the same kernel (the latest updated released ones).

Maybe Andreas could comment whether you get SLE10 SP1 update packages for your 10.1 dist.
Comment 9 Andreas Jaeger 2006-11-08 17:52:50 UTC
We have not decided about SLE10 SP1 and 10.1 backports yet.

Please test that openSUSE 10.2 works for you.
Comment 10 Rainer Klier 2006-11-09 14:34:06 UTC
ok, what should i do now?
should i test another patched kernel, or not?
currently i am running on suse10.1 with the kernel from ftp.suse.com/pub/people/trenn/sle10_SP1_hp_fix_cpufreq_ACPI_events_kernels_third_try/

and with that setup i can see progress.
thermalzones seem ok: TZ0 is now about 66c - 72c, which was about 84c before.
battery-status seem ok: counting down, when running on battery
cpu-throtteling is nearly ok: when patching manually the highest freq to /sys-filesystem with following commands:
echo 2000000 >/sys/devices/system/cpu/cpu1/cpufreq/scaling_max_freq
echo 2000000 >/sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq
then cpus are clocked from 1ghz up to 2ghz.
Comment 11 Thomas Renninger 2006-11-09 15:14:14 UTC
Please give me some days. 10.2 Beta2 is as good as out, I hope to come to it at beginning of next week.
Comment 12 Thomas Renninger 2007-01-10 11:47:54 UTC

*** This bug has been marked as a duplicate of bug 179702 ***