[lm-sensors] F8000 support added to f71882fg driver, test please.
ludovic at lebegue.org
Mon Dec 15 23:20:03 CET 2008
I compiled and tested the driver.
Here is the kernel log at startup :
Dec 15 23:07:46 xxxx kernel: [82199.773523] f71882fg: Found f8000 chip
at 0xa00, revision 41
Dec 15 23:07:46 xxxx kernel: [82199.773742] f71882fg f71882fg.2560:
Fan: 1 is in duty-cycle mode
Dec 15 23:07:46 xxxx kernel: [82199.773745] f71882fg f71882fg.2560:
Fan: 2 is in duty-cycle mode
Dec 15 23:07:46 xxxx kernel: [82199.773748] f71882fg f71882fg.2560:
Fan: 3 is in duty-cycle mode
Here is the "sensors" (sensors version 3.0.2 with libsensors version
3.0.2) output :
Adapter: ISA adapter
in0: +1.68 V
in1: +1.67 V
in2: +1.64 V
fan1: 1486 RPM
fan2: 894 RPM
fan3: 0 RPM ALARM
fan4: 34464 RPM
temp1: +46.0°C (high = +70.0°C, hyst = +60.0°C)
temp2: +41.0°C (high = +100.0°C, hyst = +85.0°C) sensor = Intel PECI
temp3: +40.0°C (high = +100.0°C, hyst = +85.0°C)
There are only two fans on my computer.
"fan3: 0 RPM ALARM " seems alright as there is no fan.
"fan4: 34464 RPM" is a bit strange
Voltage indication seems to be divided by two. I guess sensors3.conf
should be modified for f8000 chip.
On Mon, Dec 15, 2008 at 7:03 PM, Chuan-kai Lin <cklin at cs.pdx.edu> wrote:
> On Mon, Dec 15, 2008 at 04:32:09PM +0100, Hans de Goede wrote:
>> I got your email address from Jean Delvare. I'm the f71882fg maintainer
>> and I've been working on adding f8000 support to the f71882fg driver.
>> That work is now finished.
> Hi Hans,
> Thank you for your great work! The module builds and loads fine in
> 18.104.22.168 on my Dell Studio Slim 540s desktop computer, and sensors shows
> fan speed, voltage, and temperature measurements. Very cool!! I have
> attached two files showing the driver in action:
> idle.txt - sysfs contents on an idle system right after bootup
> loaded.txt - sysfs contents + sensors output on a loaded system
> Here are some issues I found:
> 1. There are only two fans in my system. The driver reports 0 RPM for
> the third fan (which sounds right), and some absurdly high RPM for
> the fourth fan (which sounds wrong).
> 2. When the driver loads into the kernel, it reports that there are
> three fans, and they are all in duty-cycle mode. However, only the
> CPU fan (fan2) uses a 4-pin connector; the case fan (fan1) uses a
> 3-pin connector. I seem to recall that a fan on a 3-pin connector
> can be controlled through only DC voltage?
> 3. The full-speed number seems to record the maximum measured fan speed
> instead of the maximum possible fan speed.
> 4. I cannot change fan speed by altering tempX_auto_pointY_pwm values.
> However, both the CPU fan and the case fan do speed up when the
> system gets warm.
> I am happy to provide more information to help you diagnose the
> Chuan-kai Lin
More information about the lm-sensors