> > I suspect you have an old config file at /etc/sensors.conf, with no > > section for the adm1025. If I'm right, you should copy the new one > > (etc/sensors.conf.eg from lm_sensors-2.8.0) to /etc/sensors.conf, > > run"sensors -s" and run "sensors" again. You should at least see > > VCCP labeled VCore. Maybe the limits will change too. Please let me > > know what they are. > > OK, I did it. Now, line with VCCP is > VCore: +1.46 V (min = +1.40 V, max = +1.55 V). Far better isnt' it :) This is because the driver (as all our drivers) default to an old version of the VRM standard (8.x). The config file changes this to use the new standard (9.0). > > In the meantime, I'll fix our driver to support your configuration. > > I have an idea on how to do that, I'll think of it while sleeping. > > great :o) OK, just finished. I fixed a pair of actual or potential bugs and cleaned the code up (over 10% reduction in source code size). Could you please give it a try and tell me how it works for you? Remplace the file in kernel/chips with this new one, recompile, reinstall and you're done. What you should expect to see: 1* +12V reading = 0 (because it's not wired on your board, and that's why it doesn't appear in Intel's software). 2* Slightly higher default voltage limits (both min and max). 3* No more ALARM for VCore nor +12V. 4* No need to force the driver anymore. You may have to reboot your system (sigh) in order to see 1* and 3*. That's because our driver used to overwrite a configuration bit. I fixed that but if it has already been overwritten by the previous version of the driver, it won't help. I admit the changes aren't very important from the user-side, but I have indeed changed many things in the simplification process so I'd like to see how it turns out, and that's why I'm very interested in you testing the new driver. I'll also appreciate a dump of the /proc/sys/dev/sensors/adm1025-i2c-0-2d/* files. You could play a bit with limits in the config file too, and see if it does what it is supposed to do. > > BTW, is the data at address Ox44 on your I2C bus still the same? ;) > > Just curious. > > you think this magic line? 00: 00 00 0f 3f 00 00 02 00 50 00 00 00 00 > 00 00 00 It's constantly. I tested something and after 2-3 reboots > looks this line still same, so maybe you have new fun ... No fun, it just confirms it must be some kind of EEPROM. I don't have any idea of what it can stand for however, so I won't go any further unless the other guy can help (not heard from him yet). -- Jean Delvare http://www.ensicaen.ismra.fr/~delvare/ -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: adm1025.c Url: http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20030728/742432f2/attachment.c
