* Problems with w83781d module
2005-05-19 6:24 Problems with w83781d module Lee, Bruce
@ 2005-05-19 6:24 ` Lee, Bruce
2005-05-19 6:24 ` Jean Delvare
` (2 subsequent siblings)
3 siblings, 0 replies; 5+ messages in thread
From: Lee, Bruce @ 2005-05-19 6:24 UTC (permalink / raw)
To: lm-sensors
I tried all of the methods below and still no success. I have attached
a file containing the results from dmesg.
I appreciate the help.
Thanks
Bruce
-----Original Message-----
From: Jean Delvare [mailto:khali@linux-fr.org]
Sent: Friday, November 14, 2003 12:42 PM
To: Lee, Bruce
Cc: sensors@stimpy.netroedge.com
Subject: Re: Problems with w83781d module
> I am trying to monitor sensors on a gateway 980 with a Winbond 83791d
> chip. I sent an earlier message and it was suggested that I download
> the latest code from CVS. I did this and still cannot resolve the
> problems that I am seeing. I have looked at the motherboard and can
> see a Winbond 83791d chip. The chip is detected with sensors-detect
> and after loading the w83781d module I get no results when running
> sensors. Looking in /proc/sys/dev/sensors, the chips file is empty and
> there are no sensor directories.
The driver fails to recognize the chip as one of its owns. I took a look
at the chipset's dump and can't understand why this happens.
I fixed a small, probabably unrelated bug in that driver. Please update
your CVS checkout and recompile the driver with DEBUG=1 (edit the
Makefile). Run "make install".
Then, try loading the driver again (don't forget to unload it before
doing so) and watch dmesg for details. See if sensors returns something.
If it doesn't help, unload the driver and reload it using force=0,0x2d.
Look at dmesg again. Check wether sensors returns the information you
are looking for.
If it still doesn't help, unload the driver once again and reload it
using force_w83791d=0x2d. Look at dmesg again. Run sensors again.
> Is it possible that the information from that chip is not available?
The chip is there at 0x2d, doubtlessly.
> Do I need to do a force of the chip and subclients?
You shouldn't, but it might help understanding where the problems
stands.
--
Jean Delvare
http://www.ensicaen.ismra.fr/~delvare/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: dmesg.out
Type: application/octet-stream
Size: 8229 bytes
Desc: dmesg.out
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20031114/6818703f/dmesg.obj
^ permalink raw reply [flat|nested] 5+ messages in thread* Problems with w83781d module
2005-05-19 6:24 Problems with w83781d module Lee, Bruce
2005-05-19 6:24 ` Lee, Bruce
@ 2005-05-19 6:24 ` Jean Delvare
2005-05-19 6:24 ` Jean Delvare
2005-05-19 6:24 ` Jean Delvare
3 siblings, 0 replies; 5+ messages in thread
From: Jean Delvare @ 2005-05-19 6:24 UTC (permalink / raw)
To: lm-sensors
From your sensors-detect output:
Driver `adm1021' (should be inserted but causes problems):
Detects correctly:
* Bus `SMBus I801 adapter at 0540' (Non-I2C SMBus adapter)
Busdriver `i2c-i801', I2C address 0x29
Chip `Maxim MAX1617A' (confidence: 7)
* Bus `SMBus I801 adapter at 0540' (Non-I2C SMBus adapter)
Busdriver `i2c-i801', I2C address 0x2a
Chip `Maxim MAX1617A' (confidence: 7)
Misdetects:
* Bus `SMBus I801 adapter at 0540' (Non-I2C SMBus adapter)
Busdriver `i2c-i801', I2C address 0x29
Chip `Maxim MAX1617' (confidence: 3)
* Bus `SMBus I801 adapter at 0540' (Non-I2C SMBus adapter)
Busdriver `i2c-i801', I2C address 0x2a
Chip `Maxim MAX1617' (confidence: 3)
I thought I had fixed that case (a driver returning both a detection and
a misdetetion for the same address) in mid-September. Could you please
double-check that the latest sensors-detect you got from CVS still does
the same? I read the code again and it really shouldn't.
--
Jean Delvare
http://www.ensicaen.ismra.fr/~delvare/
^ permalink raw reply [flat|nested] 5+ messages in thread* Problems with w83781d module
2005-05-19 6:24 Problems with w83781d module Lee, Bruce
2005-05-19 6:24 ` Lee, Bruce
2005-05-19 6:24 ` Jean Delvare
@ 2005-05-19 6:24 ` Jean Delvare
2005-05-19 6:24 ` Jean Delvare
3 siblings, 0 replies; 5+ messages in thread
From: Jean Delvare @ 2005-05-19 6:24 UTC (permalink / raw)
To: lm-sensors
> I tried all of the methods below and still no success. I have
> attached a file containing the results from dmesg.
Looks like a problem with the bus driver, not the client driver. With
the force_w83791d option, the w83781d driver shouldn't possibly refuse
to load, unless the bus itself is broken.
Could you try loading a driver for one of the other chips you have on
that bus (you have so many, it's just amazing)? For example, try
adm1026, adm9240, adm1021 and/or lm75. If all of them fail to work, this
means that the bus driver isn't working as intended.
> I appreciate the help.
You're welcome. We appreciate the feedback :)
--
Jean Delvare
http://www.ensicaen.ismra.fr/~delvare/
^ permalink raw reply [flat|nested] 5+ messages in thread
* Problems with w83781d module
2005-05-19 6:24 Problems with w83781d module Lee, Bruce
` (2 preceding siblings ...)
2005-05-19 6:24 ` Jean Delvare
@ 2005-05-19 6:24 ` Jean Delvare
3 siblings, 0 replies; 5+ messages in thread
From: Jean Delvare @ 2005-05-19 6:24 UTC (permalink / raw)
To: lm-sensors
> I am trying to monitor sensors on a gateway 980 with a Winbond 83791d
> chip. I sent an earlier message and it was suggested that I download
> the latest code from CVS. I did this and still cannot resolve the
> problems that I am seeing. I have looked at the motherboard and can
> see a Winbond 83791d chip. The chip is detected with sensors-detect
> and after loading the w83781d module I get no results when running
> sensors. Looking in /proc/sys/dev/sensors, the chips file is empty and
> there are no sensor directories.
The driver fails to recognize the chip as one of its owns. I took a look
at the chipset's dump and can't understand why this happens.
I fixed a small, probabably unrelated bug in that driver. Please update
your CVS checkout and recompile the driver with DEBUG=1 (edit the
Makefile). Run "make install".
Then, try loading the driver again (don't forget to unload it before
doing so) and watch dmesg for details. See if sensors returns something.
If it doesn't help, unload the driver and reload it using force=0,0x2d.
Look at dmesg again. Check wether sensors returns the information you
are looking for.
If it still doesn't help, unload the driver once again and reload it
using force_w83791d=0x2d. Look at dmesg again. Run sensors again.
> Is it possible that the information from that chip is not available?
The chip is there at 0x2d, doubtlessly.
> Do I need to do a force of the chip and subclients?
You shouldn't, but it might help understanding where the problems
stands.
--
Jean Delvare
http://www.ensicaen.ismra.fr/~delvare/
^ permalink raw reply [flat|nested] 5+ messages in thread