[RFC] Patch to enable temperature ceiling in powerd

[Johannes Dieterich]

Hi,

Hajimu UMEMOTO wrote:
> Hi,
> 
>>>>>> On Sat, 16 Feb 2008 18:06:12 +0100
>>>>>> Johannes Dieterich <dieterich.joh at googlemail.com> said:
> 
>> dieterich.joh> # hw.acpi.thermal.tz1.passive_cooling=1
>> dieterich.joh> su: hw.acpi.thermal.tz1.passive_cooling=1: command not found
>>
>> It should be `sysctl hw.acpi.thermal.tz1.passive_cooling=1'.
> dieterich.joh> Sorry. Just a typing error. Works now. Should I try to make buildworld?
> 
> No problem.  Please make sure hw.acpi.thermal.tz1.passive_cooling is 1
> by typing `sysctl hw.acpi.thermal.tz1.passive_cooling' before trying.
> If hw.acpi.thermal.tz1.passive_cooling is 1, passive cooling should be
> working.
I can confirm that it works, I can get buildworld through.

> 
> You can see if passive cooling is actually working by setting
> hw.acpi.verbose to 1 using sysctl.  While hw.acpi.verbose is 1,
> changing cpufreq by passive cooling is logged into /var/log/messages
> like:
> 
> Feb 17 02:26:19 kasuga kernel: acpi_tz0: temperature 86.8C: decreasing clock speed from 1200 MHz to 1000 MHz
> Feb 17 02:26:23 kasuga kernel: acpi_tz0: temperature 78.8C: resuming previous clock speed (1200 MHz)
> 
# sysctl hw.acpi.verbose
hw.acpi.verbose: 1
but no information is written to /var/log/messages. I've tailed it
during buildworld and the only informations written there are from my
atheros card.

Another question: Ian Smith asked here on the list why the fan levels in
acpi_ibm would not have not been implemented yet to 6.x. Since my
notebook started overheating when moving to 7, could that be a reason?
On my first mail to freebsd-stable I also quoted a case from the LKML
where someone with a X60s reported overheating and the problem was in
the fan levels of the (linux) acpi_ibm module, if I recall it correct.

Regards,

Johannes