https://bugzilla.novell.com/show_bug.cgi?id=404245 User daugirdas@gmail.com added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c2 --- Comment #2 from Daugirdas Racys 2008-06-27 03:34:12 MDT --- Created an attachment (id=224775) --> (https://bugzilla.novell.com/attachment.cgi?id=224775) messages - shows the poweroff messages -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=404245 User daugirdas@gmail.com added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c4 --- Comment #4 from Daugirdas Racys 2008-06-28 04:50:14 MDT --- ignore the previous comment. You have ondemand built in to the kernel. Now, my system doesn't support throttling, but it does support scaling (1800, 1600 and 800 Mhz). I would be wise to make use of that. I suppose you could modify one of the powersave scripts for acer laptops (and similarly affected systems) so that you scale them down when they reach PASSIVE point. You had this working in the past anyway. I would happily do that but I really don't have the knowledge of programming. I am looking forward for the fix. Many thanks. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=404245 User daugirdas@gmail.com added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c5 --- Comment #5 from Daugirdas Racys 2008-07-01 07:53:50 MDT --- Update: Kernel vmlinuz-2.6.25.5-1.1-vanilla is NOT affected by the bug. The scaling works fine, and once the passive point is reached it goes down to 800 MHz until the system cools down to 75C. Suse patched kernel shuts down instead. Ubuntu 8.04 32 and 64 bit are affected. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=404245 User daugirdas@gmail.com added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c7 --- Comment #7 from Daugirdas Racys 2008-07-01 18:35:02 MDT --- I did some source comparison between suse and vanilla 2.6.25.5 kernels. The /drivers/acpi/processor_thermal.c and processor_throttling.c were identical. Thermal.c were different (vanilla on the left): daugirdas@dtrsuse64:~/Desktop/linux-2.6.25.5/drivers/acpi> diff thermal.c thermal.cs 443,445c443 < if (ACPI_FAILURE(status)) < tz->trips.passive.flags.valid = 0; < else ---

if (ACPI_SUCCESS(status)) { 447,452c445,454 < < if (memcmp(&tz->trips.passive.devices, &devices, < sizeof(struct acpi_handle_list))) { < memcpy(&tz->trips.passive.devices, &devices, < sizeof(struct acpi_handle_list)); < ACPI_THERMAL_TRIPS_EXCEPTION(flag, "device");

if (memcmp(&tz->trips.passive.devices, &devices, sizeof(struct acpi_handle_list))) { memcpy(&tz->trips.passive.devices, &devices, sizeof(struct acpi_handle_list)); ACPI_THERMAL_TRIPS_EXCEPTION(flag, "device"); } } else { tz->trips.passive.flags.valid = 0; ACPI_EXCEPTION((AE_INFO, status, "Invalid passiv trip" " point\n"));

I can't read this but hopefully this would suggest something. Especially since thermal.c contains these lines further down: /* take no action if nocrt is set */ if(!nocrt) { printk(KERN_EMERG "Critical temperature reached (%ld C), shutting down.\n", KELVIN_TO_CELSIUS(tz->temperature)); orderly_poweroff(true); } Another point: THRC critical point on my system is 97C. 90C is PASSIVE, but I get shutdowns at 90C. That may also mean kernel confuses CRITICAL with PASSIVE! -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=404245 User trenn@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c9 Thomas Renninger changed: What |Removed |Added ---------------------------------------------------------------------------- Severity|Blocker |Major Status|NEW |ASSIGNED Priority|P0 - Crit Sit |P5 - None Summary|Acer laptop poweroffs intermintently after |Acer laptop powers off due to overheating |overheating | --- Comment #9 from Thomas Renninger 2008-07-07 09:15:23 MDT --- Hi Daugirdas, you got more help than all the people on the launchpad bug, right. Anyway your help is appreciated, I am sorry I missed track on this one. I removed the patch you showed up the differences in drivers/acpi/thermal.c in comment #7. It takes a while until the changes are built, put into upload directory and finally find its way to the ftp server. Also there is a lot other work to do, therefore I forgot about this one. A simple "ping!" would have been enough, though. The kernel is here: ftp://ftp.suse.com/pub/projects/kernel/kotd/SL110_BRANCH/i386/kernel-default.rpm (take the x86_64 directory for a 64 bit kernel). Please do not modify severity and priority flags. Priority is used for other things. I see that this bug should have high severity, but blocker doesn't make much sense after the product is out, the shipment can't be blocked anyway anymore. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=404245 User daugirdas@gmail.com added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c10 --- Comment #10 from Daugirdas Racys 2008-07-08 01:27:21 MDT --- Hello, Thanks for getting back. I tried the kernel last night. Although I *think* I saw more freq scaling take place (mainly 1800 to 1600Mhz), once it hit 90C PASSIVE it quickly went overboard and powered off (on stock kernel it stayed at 800Mhz at that point until it cooled down by ~20C). Could there be another patch that I've missed? Thanks, Daugirdas -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=404245 User jnelson-suse@jamponi.net added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c12 Jon Nelson changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |jnelson-suse@jamponi.net --- Comment #12 from Jon Nelson 2008-07-11 08:51:01 MDT --- As part as Acer laptops go, go to the acer europe website and get the latest BIOS. Acer has recently released a number of BIOS updates that made a difference for my friend's 7720. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=404245 User daugirdas@gmail.com added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c14 --- Comment #14 from Daugirdas Racys 2008-07-11 12:06:40 MDT --- UPDATE: Vanilla is also "faulty". I must admit my CPU burning methodology was wrong. zip doesn't use 100% CPU, hence the impression of scaling. Sorry for confusion. I ran gnome-appearance-properties [Bug 405154], and got a shut down with both kernels in less than 30s at 91C (passive displayed since 90C). So back to the issue. I believe it all started with suse after some update to 10.3 kernel. I have been just moderately using it, but I couldn't use ubuntu at that time at all due to this issue. I guess the best idea is to try 10.3 kernel and test any updates. How could I install that kernel ALONGSIDE 11.0 kernel? P.S. kacpid was on the top, but CPU usage was constantly only about 0.7% -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=404245 User hmacht@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c16 --- Comment #16 from Holger Macht 2008-07-12 06:51:00 MDT --- Commit history shows something like: ------------------------------------------------------------------------ r3015 | hmacht | 2007-08-07 16:23:27 +0200 (Tue, 07 Aug 2007) | 2 lines do not enable thermal polling per default, as requested by intel -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=404245 User daugirdas@gmail.com added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c18 --- Comment #18 from Daugirdas Racys 2008-07-13 04:44:01 MDT --- for x in /proc/acpi/thermal_zone/*/polling_frequency;do echo 5 > $x done it works! Thanks a lot. How could I enable this by default at boot time please? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=404245 User jnelson-suse@jamponi.net added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c20 --- Comment #20 from Jon Nelson 2008-07-13 06:51:02 MDT --- (In reply to comment #15 from Thomas Renninger)

Vanilla/mainline kernel is also affected? That would make sense.

It could come from the fact that we do not do thermal polling anymore. Seife, Holger do you remember the exact date (more or less at least the distribution) when we stopped thermal polling by default?

Daugirdas: Can you do: for x in /proc/acpi/thermal_zone/*/polling_frequency;do echo 5 > $x done Does it help? Can you still provide: cat /proc/acpi/thermal_zone/*/trip_points pls. There are some values in the passive line which could be interesting.

(In reply to comment #18 from Daugirdas Racys)

for x in /proc/acpi/thermal_zone/*/polling_frequency;do echo 5 > $x done

it works! Thanks a lot.

How could I enable this by default at boot time please?

Edit /etc/init.d/boot.local and add it to the bottom.

-- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=404245 User jnelson-suse@jamponi.net added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c22 --- Comment #22 from Jon Nelson 2008-07-23 17:02:49 MDT --- Is this your machine (the Aspire 1520 series?) This is probably the BIOS (released December of '07) ftp://ftp.work.acer-euro.com/notebook/aspire_1520/bios/v1.15.zip I've had much better luck with the latest BIOSes when it comes to Acer laptops. If you aren't running 1.15 and this is an update for your BIOS, consider giving it a try! -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=404245 User trenn@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c24 Thomas Renninger changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |NEEDINFO Info Provider| |daugirdas@gmail.com --- Comment #24 from Thomas Renninger 2008-07-24 02:40:21 MDT --- A BIOS update is a good point. Is there a new one or is there not? Most abilities to alter thermal trip points at runtime have been removed. You can pass a lower passive trip point via boot param. But it would be interesting to know whether the latest BIOS is still affected. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=404245 User trenn@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c27 --- Comment #27 from Thomas Renninger 2008-10-21 06:07:06 MDT ---

Or maybe better, could I set some custom powersave scheme, to say for example limit speed to 1600 (max is 1800) when temperature reaches {passive - 10C} I demanded that in the ACPI kernel interface with no luck until now. Since passive cooling or say temp monitoring moves more and more to hwmon and userspace, I'll try again. But this is not trivial. Hmm, there should still be boot params to modify trip points. Does thermal.psv=60 boot param + setting polling frequency help?

-- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=404245 User jnelson-suse@jamponi.net added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c29 --- Comment #29 from Jon Nelson 2009-02-24 11:53:43 MST --- This is still a problem with openSUSE 11.1, in fact the problem appears much worse. The thermo gauge never reads above 50 but the laptop appears to be hard-hanging several times a week, instead of "never" or "rarely" beforehand. It appears to be temperature related. I will look into using some of the items above shortly. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=404245 User trenn@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=404245#c31 --- Comment #31 from Thomas Renninger 2009-02-25 02:43:21 MST --- The problem is that general activation of thermal polling will let the machine suffer which correctly provide thermal events or already provide a sane polling value for the machine. A list of affected machines could be added where polling gets activated, I try to bring that up. Until this may hit kernels (or if it does not get accepted), you have to add the thermal.tzp=10 (ten second polling should be enough) boot parameter. Best do that at installation time or add it in /boot/grub/menu.lst. Updating kernels will take over that boot parameter and everything works fine. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.