Bug 133581

After further testing of this, and by testing I mean just letting my laptop sit for a while and see if it reacts like it previously did in bug #132877, it does not *always* shut itself off. It does, however, continue to get those ACPI-1041 messages very often, regardless of whether the system shut itself off or it lasted for a while until I took it out of screensaver mode (about 40 minutes). 

Could you attach dmesg and acpidmp output, please

Created attachment 57292 [details]
acpidmp text

Created attachment 57293 [details]
dmesg text

This is probably a BIOS issue. Please make sure that you have the latest BIOS installed for this machine before we go on.
As far as I can tell for now, I expect the problem comes from the CPU1 (Hyperthreaded?) that has no C2/C3 state declared, but CPU0 has.

If you find a newer BIOS and still have problems, please repost acpidmp and dmesg output with the new BIOS.

You could also try to disable Hyperthreading in BIOS, that could also help.

I was not able to find a newer BIOS out there, and there are no options in my BIOS to disable hyperthreading (you were correct in saying its HT), just other standard stuff. I'll keep looking around and try to find a possible place to take hyperthreading off, and if there possibly is a new BIOS version that I overlooked.

I just got a bunch of ACPI-1048 errors now, looking like the same as ACPI-1041s from my previous bug except for the numbers at the end. Not sure what this means to you guys, but it is all that is showing up before my laptop just shut itself off after reaching a critical temperature, while I was using it instead of when it is in screensaver mode. I have attached the text from /var/log/messages from a while before it happened up to when it shuts the system down. I apologize if I am adding too much to this as it goes along, feel free to let me know if I should back off on it, as I've never submitted a bug before and just want to let you guys know whats going on.

Created attachment 57596 [details]
/var/log/messages

/var/log/messages after a system shutdown because of a critical temperature reached

Could you try to boot with ec_burst=0, do the cannot agquire semaphore messages vanish?

thomas,

After roughly 24 hours of running with ec_burst=0, here is what I found:
I noticed that the fans don't kick in as much, and it is generally a more quiet experience. That is of course until the screensaver goes on after about 20 minutes of inactivity and the computer begins to heat up and the fans run very fast. I'll move the mouse and take it out of that state and it will calm down after a minute or two and be back to normal using. However, I just got shut down in the middle of browsing the web after a system message was broadcast to the screen and let me know the system was halting (I believe that was the message). This boot parameter seemed to do some good, but I still am seeing decent amounts of ACPI-1048 errors, and just got shut down a few minutes ago.

Is there a significant difference in the ACPI-1048 errors I am seeing recently, and the ACPI-1041 errors I was seeing originally?

Sorry for the late reply. You need to activate the "This comment provides the needed information. Change the status of this bug back to ASSIGNED." button or the bug vanishes from my radar ...

Is it possible for you to install the latest SUSE 10.1?
I doubt this can be fixed for 10.0. However one came up with an interesting patch which I think could fix that.
If you can install a 10.1 you might want to test whether this kernel helps:
ftp.suse.com/pub/people/trenn/kernel_default_Beta10_11_acpi_threaded/kernel-default-2.6.16-28.i586.rpm

Last post of reporter in 2005 -> closing.
Please open a new one if this still happens on 10.2