I hacked some more, it's probably cleaner now.
Current output :
dell: {119} prog/sensors/sensors
w83627hf-i2c-0-0290
Adapter: Dummy adapter
Algorithm: Dummy algorithm
VCore 1: +0.00 V (min = +26.08 V, max = +0.00 V)
VCore 2: +0.00 V (min = +26.08 V, max = +0.00 V)
+3.3V: +0.00 V (min = +26.08 V, max = +0.00 V)
+5V: +0.00 V (min = +43.81 V, max = +0.00 V)
+12V: +0.00 V (min = +99.10 V, max = +0.00 V)
-12V: -14.91 V (min = +119.14 V, max = -14.91 V) ALARM
-5V: -7.71 V (min = +74.18 V, max = -7.71 V)
V5SB: +0.00 V (min = +43.81 V, max = +0.00 V)
VBat: +0.00 V (min = +26.08 V, max = +0.00 V)
fan1: 0 RPM (min = 2608 RPM, div = 2608)
fan2: 0 RPM (min = 2608 RPM, div = 0)
fan3: 0 RPM (min = 2608 RPM, div = 0) ALARM
temp1: +0?C (limit = +261?C, hysteresis = +0?C) sensor =
thermistor ALARM
temp2: +0.0?C (limit = +89254816?C, hysteresis = +1?C) sensor =
thermistor ALARM
temp3: +0.0?C (limit = +88919504?C, hysteresis = +1?C) sensor =
thermistor
vid: +808792.881 V
alarms:
beep_enable:
Sound alarm enabled
New patch attached (direct output from "cvs -z9 diff").
Obviously there's still something wrong :-)
But I've added code to locate the sysfs mount point, and I believe I'm
reading sufficient info to get rid of hardcoded device numbers in my
earlier hacks.
Am I in the right direction ?
BTW /sys/bus/i2c/devices/0-0290 looks like this :
dell: {2} ls /sys/bus/i2c/devices/0-0290
alarms fan_min1 in_input7 in_max8 name temp_input3
beep_enable fan_min2 in_input8 in_min0 power temp_max1
beep_mask fan_min3 in_max0 in_min1 pwm1 temp_max2
detach_state in_input0 in_max1 in_min2 pwm2 temp_max3
fan_div1 in_input1 in_max2 in_min3 pwm_enable2 temp_min1
fan_div2 in_input2 in_max3 in_min4 sensor1 temp_min2
fan_div3 in_input3 in_max4 in_min5 sensor2 temp_min3
fan_input1 in_input4 in_max5 in_min6 sensor3 vid
fan_input2 in_input5 in_max6 in_min7 temp_input1 vrm
fan_input3 in_input6 in_max7 in_min8 temp_input2
And no, this is no "dell" system, it's an ABIT BE-7 board.
Danny
On Fri, 2003-10-10 at 20:44, Danny Backx wrote:
> On Thu, 2003-10-09 at 23:11, Greg KH wrote:
> > On Thu, Oct 09, 2003 at 08:01:07PM +0200, Danny Backx wrote:
> > > I think I've submitted some patches twice. I've been
> > > trying to get libsensors and prog/sensors to work
> > > based on linux 2.6-test4 and test6. But I don't
> > > see my submissions in the support db, nor did
> > > I get a reply.
> > >
> > > Is anyone interested in this?
> >
> > Yes, have you emailed them to us?
>
> No, I didn't discover this link until yesterday. It's somewhat hidden.
> I tried copy/pasting this into the support website but apparently
> that failed twice.
>
> Anyway. Stuff attached. This is no miracle yet, I've hacked both
> libsensors and prog/sensors a bit to get them to talk to sysfs.
>
> I probably need to look into all the stuff tracking bus numbers
> and such, I've basically added some #if 0 lines here and there
> to get sensors to produce output (even if that output is
> currently nonsense for obvious reasons).
>
> I've done this on linux-2.6test4. On test6 the output is rather
> empty :
> dell: {32} setenv LD_LIBRARY_PATH `pwd`/lib
> dell: {33} prog/sensors/sensors
> w83627hf
> -i2c--2-ffffffff
> Adapter: Dummy adapter
> Algorithm: Dummy algorithm
>
> Need to look into the reason, but it's probably one of the
> hacks that fails now.
>
> I could use help with sorting out the issues with bus numbers
> and such, because I am completely new to the lm_sensors source
> and this looks rather confusing.
>
> Danny
>
> P.S. If I'm not helping but delaying you with all this, just
> tell me and I'll wait until you come up with a version that
> works :-)
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: diff-to-cvs.gz
Type: application/x-gzip
Size: 5724 bytes
Desc: not available
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20031011/2112ae58/attachment.gz
