Ticket 1834 Update
Jean Delvare
khali at linux-fr.org
Sat Dec 4 16:23:19 CET 2004
> The chip on the MSI KT8 Neo is actually a Winbond w83697hf and is not
> detected correctly during sensors-detect.
What was sensors-detect detecting? Please attach the complete log.
> pwmconfig can't detect a usable a correlation:
>
> Testing pwm control 1-0290/fan1_pwm ...
> 1-0290/fan1_input ... speed was 3229 now 0
> It appears that fan 1-0290/fan1_input
> is controlled by pwm 1-0290/fan1_pwm
> Would you like to generate a detailed correlation (y)?
> PWM 255 FAN 3229
> PWM 240 FAN 3214
> PWM 225 FAN 3214
> PWM 210 FAN 3229
> PWM 195 FAN 3229
> PWM 180 FAN 3229
> PWM 165 FAN 3229
> PWM 150 FAN 3229
> PWM 135 FAN 3229
> PWM 120 FAN 3229
> PWM 105 FAN 3214
> PWM 90 FAN 3214
> PWM 75 FAN 3229
> PWM 60 FAN 3214
> PWM 45 FAN 3214
> PWM 30 FAN 3214
> PWM 15 FAN 3169
> PWM 0 FAN 0
> Fan Stopped at PWM = 0
This is admittedly a strange behavior, I can't remember of a similar
report. Maybe you need to adjust the PWM clock. This is done by register
0 of the chip (which you can write to by running "isaset 0x295 0x296 0
<value>" as root). Default value is 1 and results in a 125kHz clock. You
can increase the value up to 127, resulting frequency being
125kHz/value. If you set the MSB to 1, the base clock changes from
125kHz to 1MHz (and again you can divide it by changing the 7 other
bits). I'd suggest you try this
I doubt it'll really help, but if you have some spare time and really
want it to work, you may want to give it a try.
--
Jean Delvare
http://khali.linux-fr.org/
More information about the lm-sensors
mailing list