Problems with amd FX 8 core and freq scaling

[Stefan Esser]

Am 11.11.2013 20:52, schrieb Jung-uk Kim:
> On 2013-11-11 13:16:47 -0500, Nicholas McKenzie wrote:
>> But wouldn't this just disable frequency scaling and the whole 
>> point of powerd?
> 
> No.  acpi_throttle (and p4tcc) controls T-state.  "Frequency
> scaling" should be done by changing P-state.
> 
>> On Mon, Nov 11, 2013 at 1:46 AM, Stefan Esser <se at freebsd.org> 
>> wrote:
>>> 
>>> Am 10.11.2013 22:46, schrieb Nicholas Stewart McKenzie:
>>>> My computer crashes if I enable powerd. I can't get cpu freq 
>>>> scaling to work with my cpu:(P.S. I sent this to both
>>>> drivers and amd64 mailing list...
>>> 
>>> Hi,
>>> 
>>> you may want to try booting with the following line added to 
>>> /boot/loader.conf (or entered at the boot menu prompt after 
>>> breaking out of automatic boot):
>>> 
>>> hint.acpi_throttle.0.disabled="1"
>>> 
>>> There have been a number of reports of throttling causing 
>>> crashes. This setting does not prevent powerd from adjusting
>>> your CPU's clock, it just disables some arcane feature which
>>> pre-dates the modern power management methods.
> 
> I rewrote acpi_throttle.c at some point to fix the problem but
> never committed it because nobody was really interested in testing
> the patch.  Also, it is really an arcane and archaic feature:
> 
> http://software.intel.com/en-us/blogs/2013/10/15/c-states-p-states-where-the-heck-are-those-t-states
>
>  Now I think we should disable the feature by default because it
> is causing too much hassle for us (attached).  Any objection?

No objections from me ...

IIRC, there was consensus that throttling should be disabled
when this was last discussed, years ago. There might be some
P-4 systems in use today, but AFAIK they do not need support
from the OS to prevent them from over-heating (which was all
that throttling could achieve).

I'm very much in favour of setting throttling to disabled
even for 10.0. Throttling has no use, but can cause crashes
or undesired behaviour (in combination with P-states the
effective CPU clock may be a magnitude lower than necessary
for the least power consuming P-state).

BTW: Throttling should not only be disabled for amd64 but
also for i386, IMHO.

Regards, STefan