> Sorry for the long delay. > I was a little busy here. No problem :) > I got the package. > It patched (no *.rej) and compiled fine. > Attached is a file with some output from my environment. After both of you reported that fan1 was working OK but fan2 wasn't, I had to admit that there could be a problem in the driver. It is theoretically possible that the board maker decides to enable fan1 and disable fan2 (which is why I didn't react after the first report) but two reports by two different testers were too much ;) There was actually a bug in the driver, I wasn't watching the correct registers when deciding whether to ebable fans or not. Same for PWM outputs. This is now fixed in patch tagged beta1: http://khali.linux-fr.org/devel/i2c/linux-2.6/linux-2.6.8-rc2-i2c-smsc47m1-beta1.diff Please test and confirm that both of you now have fan 1 and 2 working, and two pwm file sets as well (whether they have an actual effect on fans is yet another matter). (You can revert the first patch using "patch -R" before applying this one, so that you won't waste your time getting a fresh new tree.) Thanks a lot for testing my code. Among things that can be done to test the driver more actively: 1* Change fan limits in /etc/sensors.conf, run "sensors -s" and check that the limits have been set properly. Ditto for divisors. 2* Change the fan limits through sysfs directly. Testing of corner cases is welcome (large values, negative values, zero). Ditto for divisors. 3* Play with PWM (enable/disable, change speed). Be careful not to stop your fans completely, systems usually don't like that much. Thanks again. -- Jean "Khali" Delvare http://khali.linux-fr.org/
