[v3.20]: Missing /sys/class/ACAD directory OR Cannot release Mutex

[v3.20]: Missing /sys/class/ACAD directory OR Cannot release Mutex

[Jörg Otte]

Starting with kernel 3.19.0-05184-g18320f2, I often find my notebook
running with 'powersave' policy even if it is on AC.

It turned out that /sys/class/ACAD directory is missing and in the
logs I see:

ACPI Error: Cannot release Mutex [MUT0], not acquired (20150204/exmutex-376)
ACPI Error: Method parse/execution failed [\_SB_.ACAD._PSR] (Node
ffff880216028d48), AE_AML_MUTEX_NOT_ACQUIRED (20150204/psparse-536)
ACPI Exception: AE_AML_MUTEX_NOT_ACQUIRED, Error reading AC Adapter
state (20150204/ac-131)

At the last 25 startups since 3.19.0-05184-g18320f2 it occured 13 times.

It does not occur with kernels older than 3.19.0-05184-g18320f2.

Thanks, Jörg
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [v3.20]: Missing /sys/class/ACAD directory OR Cannot release Mutex

[Rafael J. Wysocki]

On Sunday, February 22, 2015 01:33:09 PM Jörg Otte wrote:
> Starting with kernel 3.19.0-05184-g18320f2, I often find my notebook
> running with 'powersave' policy even if it is on AC.
> 
> It turned out that /sys/class/ACAD directory is missing and in the
> logs I see:
> 
> ACPI Error: Cannot release Mutex [MUT0], not acquired (20150204/exmutex-376)
> ACPI Error: Method parse/execution failed [\_SB_.ACAD._PSR] (Node
> ffff880216028d48), AE_AML_MUTEX_NOT_ACQUIRED (20150204/psparse-536)
> ACPI Exception: AE_AML_MUTEX_NOT_ACQUIRED, Error reading AC Adapter
> state (20150204/ac-131)
> 
> At the last 25 startups since 3.19.0-05184-g18320f2 it occured 13 times.
> 
> It does not occur with kernels older than 3.19.0-05184-g18320f2.

Which mainline commit is that?

The only changes in 4.0-rc1 that can affect that are the ACPI EC updates IIRC.

Is there a chance to try the acpi-ec branch from linux-pm.git and see if the
problem is there?

Rafael

--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [v3.20]: Missing /sys/class/ACAD directory OR Cannot release Mutex

[Jörg Otte]

2015-02-24 0:24 GMT+01:00 Rafael J. Wysocki <rjw@rjwysocki.net>:
> On Sunday, February 22, 2015 01:33:09 PM Jörg Otte wrote:
>> Starting with kernel 3.19.0-05184-g18320f2, I often find my notebook
>> running with 'powersave' policy even if it is on AC.
>>
>> It turned out that /sys/class/ACAD directory is missing and in the
>> logs I see:
>>
>> ACPI Error: Cannot release Mutex [MUT0], not acquired (20150204/exmutex-376)
>> ACPI Error: Method parse/execution failed [\_SB_.ACAD._PSR] (Node
>> ffff880216028d48), AE_AML_MUTEX_NOT_ACQUIRED (20150204/psparse-536)
>> ACPI Exception: AE_AML_MUTEX_NOT_ACQUIRED, Error reading AC Adapter
>> state (20150204/ac-131)
>>
>> At the last 25 startups since 3.19.0-05184-g18320f2 it occured 13 times.
>>
>> It does not occur with kernels older than 3.19.0-05184-g18320f2.
>
> Which mainline commit is that?
18320f2 Merge tag 'pm+acpi-3.20-rc1-2' of
git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm

>
> The only changes in 4.0-rc1 that can affect that are the ACPI EC updates IIRC.
>
> Is there a chance to try the acpi-ec branch from linux-pm.git and see if the
> problem is there?
Should be no problem. I will give it a try.

Thanks, Jörg
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [v3.20]: Missing /sys/class/ACAD directory OR Cannot release Mutex

[Jörg Otte]

2015-02-24 12:03 GMT+01:00 Jörg Otte <jrg.otte@gmail.com>:
> 2015-02-24 0:24 GMT+01:00 Rafael J. Wysocki <rjw@rjwysocki.net>:
>> On Sunday, February 22, 2015 01:33:09 PM Jörg Otte wrote:
>>> Starting with kernel 3.19.0-05184-g18320f2, I often find my notebook
>>> running with 'powersave' policy even if it is on AC.
>>>
>>> It turned out that /sys/class/ACAD directory is missing and in the
>>> logs I see:
>>>
>>> ACPI Error: Cannot release Mutex [MUT0], not acquired (20150204/exmutex-376)
>>> ACPI Error: Method parse/execution failed [\_SB_.ACAD._PSR] (Node
>>> ffff880216028d48), AE_AML_MUTEX_NOT_ACQUIRED (20150204/psparse-536)
>>> ACPI Exception: AE_AML_MUTEX_NOT_ACQUIRED, Error reading AC Adapter
>>> state (20150204/ac-131)
>>>
>>> At the last 25 startups since 3.19.0-05184-g18320f2 it occured 13 times.
>>>
>>> It does not occur with kernels older than 3.19.0-05184-g18320f2.
>>
>> Which mainline commit is that?
> 18320f2 Merge tag 'pm+acpi-3.20-rc1-2' of
> git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
>
>>
>> The only changes in 4.0-rc1 that can affect that are the ACPI EC updates IIRC.
>>
>> Is there a chance to try the acpi-ec branch from linux-pm.git and see if the
>> problem is there?
> Should be no problem. I will give it a try.

I booted kernel 3.19.0-rc5-00560-g92e4b1b from linux-pm/acpi-ec 5 times
and it didn't occure.

Today I noticed 2 other things:
1) There are at least 2 slightly different error cases
     a) ACPI Error: Cannot release Mutex [MUT0], not acquired
     b) ACPI Error: Thread 370114560 cannot release Mutex [MUT0]
acquired by thread ..

2) Boot times are different between good and bad case, bad case takes
significally longer:
     a)  Good case
[    0.285892] intel_idle: MWAIT substates: 0x21120
[    0.285893] intel_idle: v0.4 model 0x3A
[    0.285894] intel_idle: lapic_timer_reliable_states 0xffffffff
[    0.288461] ACPI: AC Adapter [ACAD] (off-line)

     b) Bad case
[    0.284061] intel_idle: MWAIT substates: 0x21120
[    0.284062] intel_idle: v0.4 model 0x3A
[    0.284063] intel_idle: lapic_timer_reliable_states 0xffffffff
[    1.279089] tsc: Refined TSC clocksource calibration: 2494.334 MHz
[    2.280355] Switched to clocksource tsc
[    5.194933] ACPI Error: Thread 370114560 cannot release Mutex [MUT0]

Thanks, Jörg
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html