ACPI Warning (nspredef-0858): \_TZ_.THRM._PSL: Return Package type mismatch at index 0 on ASUS recent BIOS

[Julien Aubin]

Hi all,

My computer :
Asus M2N-SLI Deluxe
BIOS 1701
AMD Athlon 3600+ X2 (will be upgraded to Phenom 9850 soon)

On startup, I get the following error message :
[ 1.210514] ACPI Warning (nspredef-0858): \_TZ_.THRM._PSL: Return Package type mismatch at index 0 - found [NULL ObjectDescriptor], expected Reference [20080926]

This problem is linked to my recent BIOS update, as with previous BIOS the issue did not appear. However the issue seems to be widespread on recent Asus AMD boards, or at least with recent BIOSes issued from them :
http://lkml.indiana.edu/hypermail/linux/kernel/0901.1/01674.html
http://www.nvnews.net/vbulletin/showthread.php?t=132601
http://<email address hidden>/msg22295.html
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/305023

Release :
Description: Ubuntu 9.04
Release: 9.04

Linux version :
Linux pcathlon64 2.6.28-11-generic #42-Ubuntu SMP Fri Apr 17 01:57:59 UTC 2009 i686 GNU/Linux

[Julien Aubin (gojulgarbmail)]

Added : I could not experience any issue currently but it seems to raise instabilities in some cases, slows down the boot process and seems to prevent suspend from working properly. (Not experienced there, just saw this on the internet while looking for solutions upon this issue on the Internet)

[Timmy Weerwag (timmy-weerwag)]

I get this message too and can verify the very slow boot and broken suspend.

Disabling acpi (acpi=off) resolves the problem. Disabling apic (noapic) also resolves it, but then the computer locks up after a couple of minutes.

Suprisingly, when the boot stalls after the message, pressing keys or the power button resumes the boot process.

However, when I shut down Kubuntu, the screen goes black and it hangs. After holding the power button for four seconds it powers off. Although, it won´t start properly anymore (no BIOS splash, no beep, nothing). I have to switch the power button of the PSU, wait till the powerled on the mobo goes dark, and then turn on the power again.

Computer:
Asus M3N78 Pro with BIOS version 0801
AMD Phenom X4 9950 Black Edition
Kubuntu 9.04 (64-bit) with 2.6.28-11-generic kernel

[Timmy Weerwag (timmy-weerwag)]

It seems that the DSDT is compiled with the Microsoft compiler, and thus only works correctly on Windows. I still couldn't get a fixed DSDT to work. Some portions of the DSDT (disassembled with the Intel compiler) don't make any sense at all. Including the definition of _PSL. It refers to an external object (/_PR_.CPU0) or something, instead to the processor objects defined in the source (/_PR_.C00x, with x the number of the core). The only reference to CPU0 is in that line, so I really don't know what it is.

[hawthornso23 (hawthorn)]

my DSDT had (/_PR.CPU0) instead of (/_PR_.CPU0)

/_PR_.CPU0 is at least defined as an external object.
/_PR.CPU0 just seems to be a typo.

May I assume from your comment that you found
and fixed a similar error and STILL couldn't get it
to work?

I am reluctant to try using a changed DSDT as I
really don't know what I'm doing here.

[hawthornso23 (hawthorn)]

OK - the CPU0 entry is indeed the error. But my conjecture about the underscore I
think was just a mistake. However I have fixed the problem on my machine.

Discussion at this location
http://<email address hidden>/msg11969.html
is relevant. This analyses a case where this error first appeared due to an update in BIOS.
Comparisons of the BIOS before and after showed that the processor names were changed
from CPU0 and CPU1 to C000 to C001 through C003 , presumably to cope with
quad core. However this entry was missed.

Here are the changes I made to my dsdt.dsl
-------------------------------------------------------------
Before>>>>
            Name (_PSL, Package (0x01)
            {
                \_PR.CPU0
            })

After>>>>
            Name (_PSL, Package (0x01)
            {
                \_PR.C000
            })

>>>>

I believe it is also necessary to delete the line
    External (\_PR_.CPU0)
--------------------------------------------------------------

Here is the relevant portion of dmesg before the change

[ 1.194521] fan PNP0C0B:00: registered as cooling_device0
[ 1.194525] ACPI: Fan [FAN] (on)
[ 1.194687] processor ACPI_CPU:00: registered as cooling_device1
[ 1.194712] processor ACPI_CPU:01: registered as cooling_device2
[ 1.196442] ACPI Warning (nspredef-0858): \_TZ_.THRM._PSL: Return Package type mismatch at index 0 - found [NULL Object Descriptor], expected Reference [20080926]
[ 1.196448] ACPI: Expecting a [Reference] package element, found type 0
[ 1.196629] thermal LNXTHERM:01: registered as thermal_zone0
[ 1.196864] ACPI: Thermal Zone [THRM] (40 C)
[ 1.224705] Linux agpgart interface v0.103
[ 1.224715] Serial: 8250/16550 driver4 ports, IRQ sharing enabled

Here is the relevant portion of dmesg after the change

[ 1.206285] fan PNP0C0B:00: registered as cooling_device0
[ 1.206289] ACPI: Fan [FAN] (on)
[ 1.206445] processor ACPI_CPU:00: registered as cooling_device1
[ 1.206470] processor ACPI_CPU:01: registered as cooling_device2
[ 1.208151] thermal LNXTHERM:01: registered as thermal_zone0
[ 1.208361] ACPI: Thermal Zone [THRM] (40 C)
[ 1.236701] Linux agpgart interface v0.103
[ 1.236712] Serial: 8250/16550 driver4 ports, IRQ sharing enabled

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

In terms of performance it is hard to say what difference it makes. I have
in the past had some infrequent but rather mysterious complete lockups
requiring reboot while doing nothing in particular. This may however have
been a video driver problem.

This really isn't an Ubuntu bug.
It is a buggy DSDT file in the BIOS for these cards.

[Timmy Weerwag (timmy-weerwag)]

Well, I think \_PR is just a short-hand for \_PR_. Not that I am an ACPI expert...

Besides the CPU0 problem, the real mess seems to be in the Hybrid SLI part. On my first attempts I just removed that whole part and than it would compile. But without changing anything else the kernel just panics. So I haven't had the possibillity to check if CPU0 is the cause of slow boot and suspend problems.

It is indeed not really an Ubuntu bug, but the kernel devs state that if it works on Windows, it should be regarded as a Linux bug.

[Jeremy Foshee (jeremyfoshee)]

Hi Julien,

This bug was reported a while ago and there hasn't been any activity in it recently. We were wondering if this is still an issue? Can you try with the latest development release of Ubuntu? ISO CD images are available from http://cdimage.ubuntu.com/releases/ .

If it remains an issue, could you run the following command from a Terminal (Applications->Accessories->Terminal). It will automatically gather and attach updated debug information to this report.

apport-collect -p linux 379940

Also, if you could test the latest upstream kernel available that would be great. It will allow additional upstream developers to examine the issue. Refer to https://wiki.ubuntu.com/KernelMainlineBuilds . Once you've tested the upstream kernel, please remove the 'needs-upstream-testing' tag. This can be done by clicking on the yellow pencil icon next to the tag located at the bottom of the bug description and deleting the 'needs-upstream-testing' text. Please let us know your results.

Thanks in advance.

[This is an automated message. Apologies if it has reached you inappropriately; please just reply to this message indicating so.]

[Julien Aubin (gojulgarbmail)]

Hi,

Sorry not to have closed this bug, but it no longer appears under Ubuntu Karmic.

Rgds,

Julien.