Hm... OK, I'm sorry, Jean is right that lm85 is the correct driver. Have you posted the output from i2cdetect? Are there devices at the locations we expect to find the adm1027/lm85? The auto-detect code isn't spitting anything out to /var/log/messages, but it might not be called. It will print more information if you compile with DEBUG enabled in the lm85 driver. Add '#define DEBUG 1' at the beginning of lm85.c to enable debugging in that driver, then recompile and reload. Just to be sure that the devices are not on the i2c-amd756 bus, can you send the output of i2cdetect if you haven't already... :v) Charles Galpin wrote: > Hi Philip > > I can't find a adm1027 module, source or binary. When I modprobe lm85 I > get no errors. > > [root at opteron root]# modprobe lm85 > [root at opteron root]# lsmod > Module Size Used by Not tainted > lm85 24052 0 (unused) > i2c-proc 9800 0 [lm85] > i2c-dev 5696 0 > i2c-amd756 5138 0 > i2c-core 25124 0 [lm85 i2c-proc i2c-dev i2c-amd756] > ... > > Mar 1 07:44:30 opteron kernel: i2c-proc.o version 2.8.4 (20040207) > Mar 1 07:44:30 opteron kernel: lm85: Version 2.8.4 (20040207) > Mar 1 07:44:30 opteron kernel: lm85: See > http://www.penguincomputing.com/lm_sensors for more info. > > [root at opteron root]# uname -a > Linux opteron 2.4.22-1.2135.nptlsmp #1 SMP Mon Dec 15 15:45:38 EST 2003 > x86_64 x86_64 x86_64 GNU/Linux > > [root at opteron root]# ls > /lib/modules/2.4.22-1.2135.nptlsmp/kernel/drivers/i2c/chips/adm* > /lib/modules/2.4.22-1.2135.nptlsmp/kernel/drivers/i2c/chips/adm1021.o > /lib/modules/2.4.22-1.2135.nptlsmp/kernel/drivers/i2c/chips/adm1024.o > /lib/modules/2.4.22-1.2135.nptlsmp/kernel/drivers/i2c/chips/adm1025.o > /lib/modules/2.4.22-1.2135.nptlsmp/kernel/drivers/i2c/chips/adm1026.o > /lib/modules/2.4.22-1.2135.nptlsmp/kernel/drivers/i2c/chips/adm9240.o > > thanks, > > charles > > On Mar 1, 2004, at 10:20 AM, Philip Pokorny wrote: > >> The ADM1027 is a monitoring chip. There is a driver written for it. >> "adm1027" >> >> What happens when you "modprobe adm1027" or "insmod >> kernel/chips/adm1027.o"? >> >> If there is an error message, post that and the output from "dmesg | >> tail -30" >> >> Thanks, >> :v) >> >> Charles Galpin wrote: >> >>> >>> On Feb 29, 2004, at 5:14 PM, Jean Delvare wrote: >>> >>>> Hi Charles, >>>> >>>> The Super-I/O chip on your board seems to be a PC87366 by National >>>> Semiconductor. At least the device ID matches. That said, it still >>>> looks >>>> like a disabled chip (at least the hardware monitoring part of it). >>>> >>>> Please confirm this information if you can (docs, manufacturer, >>>> visual... whatever). >>> >>> >>> >>> Hi Jean >>> >>> I was trying not to bother everyone before emailing again. I have >>> requested more info from the vendor who has a close relationship with >>> MSI and *should* be able to provide the info. >>> >>> I did get to look at the hardware last night though. It turns out the >>> motherboard is a MS-9145 - which I cannot find any info on. The only >>> winbond chip I see is a W986432DH-6 which is SDRAM. I'm honestly >>> not exactly sure what I'm looking for so I wrote down the info for >>> any chips i could see. I did find info on one marked 'ADM1027 51'. >>> google pointed me to >>> >>> http://www.analog.com/Analog_Root/productPage/productHome/ >>> 0%2C%2CADM1027%2C00.html >>> >>> Which looks right on the money. Not sure it jives with your PC87366 >>> identification though? >>> >>> Everything else was clearly not a sensor, and I did not see anything >>> with PC87366 on it. >>> >>> Does this help? >>> >>> I'll let you know what I get from the vendor. Thanks again for >>> continuing to look into this! >>> >>> charles >>> >> > -- Philip Pokorny, Director of Engineering Tel: 415-954-2823 Fax: 415-954-2899 Toll Free: 888-PENGUIN PENGUIN COMPUTING, INC. www.penguincomputing.com
