Re: Very serious cooling issues CURRENT/STABLE

On 12/21/08, Nathan Lay <nslay_at_comcast.net> wrote:
> Paul B. Mahol wrote:
>> On 12/21/08, Nathan Lay <nslay_at_comcast.net> wrote:
>>
>>> acpi.thermal.min_runtime: 0
>>> hw.acpi.thermal.polling_rate: 10
>>> hw.acpi.thermal.user_override: 0
>>> hw.acpi.thermal.tz0.temperature: 37.0C
>>> hw.acpi.thermal.tz0.active: -1
>>>
>> Is this one ever changed?
>>
>>
>>> hw.acpi.thermal.tz0.passive_cooling: 1
>>> hw.acpi.thermal.tz0.thermal_flags: 0
>>> hw.acpi.thermal.tz0._PSV: 89.5C
>>> hw.acpi.thermal.tz0._HOT: -1
>>> hw.acpi.thermal.tz0._CRT: 93.0C
>>> hw.acpi.thermal.tz0._ACx: -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
>>>
>>
>> This one means that coling will never be used, why:
>> my output looks like this:
>> hw.acpi.thermal.tz0._ACx: 85.0C 75.0C 60.0C 50.0C -1 -1 -1 -1 -1 -1
>>
>>
>>> hw.acpi.thermal.tz0._TC1: 5
>>> hw.acpi.thermal.tz0._TC2: 4
>>> hw.acpi.thermal.tz0._TSP: 600
>>>
>>
>> You can play with all thermal values once you enable:
>> hw.acpi.thermal.user_override
>>
>> But acpi may redo such values again after some time.
>> You only real workaround is to use modified acpi ASL:
>> it is explained in handbook.
>>
>> In my case I fixed in that way bogus kernel
>> message "_CRT value is absurd, ignored".
>>
>>
>>
> hw.acpi never displayed thermal values for some reason.  However, after
> loading acpi_ibm, I can query those values without a problem
> dev.acpi_ibm.0.thermal: 49 41 33 48 27 -1 22 -1
>
> I'm not sure the critical temperature (99C) is a problem, but what I
> have observed is you should never be near it.  None of these thinkpads
> got over 80C under load with FreeBSD installed until recently.  ACPI's
> ASL does not appear to be the problem as it has worked correctly in the
> past.

Until recenty when, can you point into svn revision?

If the same overheat happens with acpi disabled that I dont see how
freebsd acpi can help you.


-- 
Paul