Re: [lm-sensors] hwmon strangeness, no fan control on w83627dhg

[Guenter Roeck <linux@roeck-us.net>]

On 02/14/2015 02:56 AM, Pavel Machek wrote:
> Hi!
>
>>> I don't have much time to spend on this right now but you should start
>>> by comparing the datasheets of the W83627DHG and the W83627DHG-P
>>> looking for differences.
>>
>> I got the datasheets, (-P version is novuton, the other one is
>> winbond), and there are pretty similar.
>>
>> The hardware monitoring chapter is pretty similar. Section 5 (PECI) is
>> completely rewritten. Section 6.3 has some differences, it looks -P
>> has new smart fan III+ feature. Section 6.3.4 (smart fan III+) is
>> new. There are differences in 7.1.3.1. section, but that is some SMI
>> interrupt stuff, not interesting for us.
>>
>> Chapter 8/9 (hardware monitor register set) has completely different
>> graphics... but seems to describe same hardware. Hmm, it has even the
>> same value (5CA3) in the identification register. Sections .70-.84 seems to
>> describe new registers for smart fan III+ mode and more. And there are
>> more new registers near section .98 and .119.
>>
>> But it seems hardware should be backwards-compatible...
>
> Ok, it seems it is. But by default it starts in (non-compatible) smart
> fan III+ mode. I can do port(0x15e) &= ~0x20 to kick it back into
> compatible mode, and I can then control CPU fan speed in 800..1700 RPM
> range as expected; but my goal was to stop the fan to make the machine
> quiet... No success with that so far.
>

Not all CPU fans can be stopped. Maybe that is the case here ?

Guenter


_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors