Unexcepted shutdown in 2.6.28-rc5

DISCUSSION Maxime Poncet 4 participants 9 messages Nov 20, 2008 – Dec 17, 2008 ACPI

lore

Share & tools

lore

Maxime Poncet

author

2008-11-20 23:14 UTC #0 lore raw

Hi,

I have the same problem in 2.6.28-rc5 and 2.6.27.x :
my laptop shutdown automatically while working on it.

That problem doesn't appear in 2.6.28-rc3.

This terminal output show well the problem
$uname -a
2.6.28-rc5
$while true;
do cat /proc/acpi/thermal_zone/THRM/temperature;
sleep 1;
done

...

temperature:             51 C
temperature:             51 C
temperature:             144 C

And then it shutdowns because critical temperature was reached

Another thing that might be important : the output of dmesg Every 5 seconds this message appears (not in 2.6.28-rc3):

ACPI: Transitioning device [FAN] to D3
ACPI: Unable to turn cooling device [ffff81007fb656d0] 'off'

Hope it will help you.

Maxime

Zhao Yakui

2008-11-21 00:52 UTC #1 lore raw

Quoted 4 lines · Maxime Poncet

Hi,
I have the same problem in 2.6.28-rc5 and 2.6.27.x :
my laptop shutdown automatically while working on it.

Will you please add the boot option of "thermal.nocrt=1" and attach the following output?
a. acpidump
b. dmesg
c. cat /proc/acpi/thermal_zone/*/*

Of course you can file a bug report at
http://bugzilla.kernel.org/enter_bug.cgi?product=ACPI
and attach the output of acpidump, dmesg.

Will you please add the boot option of "acpi.power_nocheck=1" and see whether the following message still exists?

Quoted 2 lines

ACPI: Transitioning device [FAN] to D3
ACPI: Unable to turn cooling device [ffff81007fb656d0] 'off'

Thanks.

Quoted 38 lines

That problem doesn't appear in 2.6.28-rc3.
This terminal output show well the problem
$uname -a
2.6.28-rc5
$while true;
do cat /proc/acpi/thermal_zone/THRM/temperature;
sleep 1;
done
...
temperature:             51 C
temperature:             51 C
temperature:             144 C
And then it shutdowns because critical temperature was reached
Another thing that might be important : the output of dmesg Every 5 seconds this message appears (not in 2.6.28-rc3):
ACPI: Transitioning device [FAN] to D3
ACPI: Unable to turn cooling device [ffff81007fb656d0] 'off'
Hope it will help you.
Maxime
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Maxime Poncet

author

2008-11-23 14:06 UTC #2 lore raw

2008/11/21, Zhao Yakui <[email protected]>:

Quoted 10 lines

Will you please add the boot option of "thermal.nocrt=1" and attach the following output?
a. acpidump
b. dmesg
c. cat /proc/acpi/thermal_zone/*/*
Of course you can file a bug report at
http://bugzilla.kernel.org/enter_bug.cgi?product=ACPI
and attach the output of acpidump, dmesg.

I will try 2.6.28-rc6 to check if this bug disappear.
I have tried several times to send attachement but I can't see my email on http://marc.info/?l=linux-acpi&r=1&b=200811&w=2.
I think it's because I'am using thunderbird.

Quoted 5 lines

Will you please add the boot option of "acpi.power_nocheck=1" and see whether the following message still exists?
Quoted 2 lines
ACPI: Transitioning device [FAN] to D3
ACPI: Unable to turn cooling device [ffff81007fb656d0] 'off'

This message disappear using this boot option.

Zhang Rui

2008-11-21 01:09 UTC #3 lore raw

In reply to Maxime Poncet

Quoted 7 lines · Maxime Poncet

Hi,
I have the same problem in 2.6.28-rc5 and 2.6.27.x :
my laptop shutdown automatically while working on it.
That problem doesn't appear in 2.6.28-rc3.

there are not many ACPI changes between 2.6.28-rc3 and -rc5. it would be great if you can use git-bisect to find out which commit introduces this regression.

thanks,
rui

Quoted 35 lines

This terminal output show well the problem
$uname -a
2.6.28-rc5
$while true;
do cat /proc/acpi/thermal_zone/THRM/temperature;
sleep 1;
done
...
temperature:             51 C
temperature:             51 C
temperature:             144 C
And then it shutdowns because critical temperature was reached
Another thing that might be important : the output of dmesg Every 5 seconds this message appears (not in 2.6.28-rc3):
ACPI: Transitioning device [FAN] to D3
ACPI: Unable to turn cooling device [ffff81007fb656d0] 'off'
Hope it will help you.
Maxime
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Rafael J. Wysocki

2008-12-07 21:17 UTC #4 lore raw

Quoted 6 lines · Zhang Rui

Quoted 5 lines · Maxime Poncet
Hi,
I have the same problem in 2.6.28-rc5 and 2.6.27.x :
my laptop shutdown automatically while working on it.

Quoted 31 lines

Quoted 2 lines
That problem doesn't appear in 2.6.28-rc3.
there are not many ACPI changes between 2.6.28-rc3 and -rc5. it would be great if you can use git-bisect to find out which commit introduces this regression.
thanks,
rui
Quoted 22 lines
This terminal output show well the problem
$uname -a
2.6.28-rc5
$while true;
do cat /proc/acpi/thermal_zone/THRM/temperature;
sleep 1;
done
...
temperature:             51 C
temperature:             51 C
temperature:             144 C
And then it shutdowns because critical temperature was reached
Another thing that might be important : the output of dmesg Every 5 seconds this message appears (not in 2.6.28-rc3):
ACPI: Transitioning device [FAN] to D3
ACPI: Unable to turn cooling device [ffff81007fb656d0] 'off'
Hope it will help you.

I wonder why _may_ be the reason of this. Alex?

Rafael

Zhao Yakui

2008-12-08 01:08 UTC #5 lore raw

Quoted 2 lines · Rafael J. Wysocki

On Friday, 21 of November 2008, Zhang Rui wrote:
From the test it seems that the shutdown is caused by that the

temperature is above the critical threshold.

Quoted 2 lines

CRT threshold 100
the temperature 144C

But the problem is why the temperature is above the critical threshold.

Quoted 1 line

From the acpidump the temperature is obtained by evaluating the _TMP

object.
Method (_TMP, 0, Serialized)

            {
                If (DTSE)
                {
                    If (LGreaterEqual (DTS1, DTS2))
                    {
                        Return (Add (0x0AAC, Multiply (DTS1, 0x0A)))
                    }

                    Return (Add (0x0AAC, Multiply (DTS2, 0x0A)))
                }

                If (ECON)
                {
                    Return (Add (0x0AAC, Multiply
(\_SB.PCI0.LPC.WBEC.DTMP, 0x0A)))
                }

                Return (0x0BB8)
            }

Quoted 1 line

From the above definition of _TMP object the temperature will be related

with the following factor:
a. DTSE . If the DTSE is non-zero, the temperature will be the
maixium of DTSE1/DTS2. But the DTSE/DTS1/DTS2 is defined in ACPI NVS memory region. If the DTSE is non-zero, in theory the shutdown will happen on the kernel of 2.6.28-rc3.
b. DTMP. When the DTSE is zero, the temperature will be obtained from
DTMP, which is an EC internal register.

Hi, Maxime
As there exist some changes related with EC between 2.6.28-rc3/rc5,
Will you please use the git-bisect to identify which commit causes the regression?
Thanks.

Quoted 18 lines

Quoted 7 lines
Quoted 7 lines
...
temperature:             51 C
temperature:             51 C
temperature:             144 C
And then it shutdowns because critical temperature was reached
Quoted 7 lines
Quoted 7 lines
Another thing that might be important : the output of dmesg Every 5 seconds this message appears (not in 2.6.28-rc3):
ACPI: Transitioning device [FAN] to D3
ACPI: Unable to turn cooling device [ffff81007fb656d0] 'off'
Hope it will help you.
I wonder why _may_ be the reason of this. Alex?
Rafael

Maxime Poncet

author

2008-12-10 23:35 UTC #6 lore raw

Quoted 5 lines · Zhao Yakui

Hi, Maxime
As there exist some changes related with EC between 2.6.28-rc3/rc5,
Will you please use the git-bisect to identify which commit causes the regression?
Thanks.

Hi Zhao,
This issue is resolved for me in 2.6.28-rc7

But I am still getting the following message every 5 seconds
ACPI: Transitioning device [FAN] to D3
ACPI: Unable to turn cooling device [ffff88007fb63590] 'off'

I am using bitsect to find out which commit causes this regression.

Zhao Yakui

2008-12-15 02:14 UTC #7 lore raw

Quoted 8 lines · Maxime Poncet

Quoted 5 lines · Zhao Yakui
Hi, Maxime
As there exist some changes related with EC between 2.6.28-rc3/rc5,
Will you please use the git-bisect to identify which commit causes the regression?
Thanks.
Hi Zhao,
This issue is resolved for me in 2.6.28-rc7

Sorry for the late response.

It is good news that the shutdown issue is fixed in 2.6.28-rc7. But the following issue still exists.

Quoted 2 lines

ACPI: Transitioning device [FAN] to D3
ACPI: Unable to turn cooling device [ffff88007fb63590] 'off'

Will you please add the boot option of "acpi.power_nocheck=1" and see whether the above issue still exists?
Thanks.

Quoted 8 lines

But I am still getting the following message every 5 seconds
ACPI: Transitioning device [FAN] to D3
ACPI: Unable to turn cooling device [ffff88007fb63590] 'off'
I am using bitsect to find out which commit causes this regression.

Maxime Poncet

author

2008-12-17 10:44 UTC #8 lore raw

Quoted 10 lines · Zhao Yakui

Sorry for the late response.
It is good news that the shutdown issue is fixed in 2.6.28-rc7. But the following issue still exists.
Quoted 2 lines
ACPI: Transitioning device [FAN] to D3
ACPI: Unable to turn cooling device [ffff88007fb63590] 'off'
Will you please add the boot option of "acpi.power_nocheck=1" and see whether the above issue still exists?
Thanks.

This boot option resolve this issue.