Svante Signell a écrit : > On Thu, 2008-01-03 at 22:49 +0100, Aurelien Jarno wrote: >> Svante Signell a écrit : >>> Package: lm-sensors >>> Version: 3.0.0 >>> Severity: serious >>> Tags: patch >>> >>> The following sensors calculate wrong max and min values for negative >>> inputs in sensors3.conf: >>> chip "lm78-*" "lm79-*" "w83781d-*" >>> chip "as99127f-*" >>> chip "lm87-*" "adm1024-*" >>> >>> The max and min values should be swapped as shown for lm78 below: >>> Change: >>> # set in5_max -12 * 1.05 >>> # set in5_min -12 * 0.95 >>> # set in6_max -5 * 1.05 >>> # set in6_min -5 * 0.95 >> Those values are actually correct, the important thing is that they >> match the logic of the sensor, so that the alarm is working. >> >> If you revert it as in your patch, the alarm is not working anymore as >> max_value_raw < min_value_raw. >> > I saw that the alarm does not work with the changes I proposed. However, > in my opinion the code logic in libsensors4/lm-sensors should be changed > accordingly to be consistent for all sensors. I don't believe that a > sensor has any logic built-in, the interpretation of the readings is > most probably in the libsensors4/lm-sensors code. I have not looked at > the code yet, but can create a patch if of interest. >
The logic of the sensors applies on the raw values (the ones from /sys): min < input < max. Then you apply a formula to get the real value. Depending on the formula you use, that is on your motherboard and not on the sensors itself, the min and max order could be reverted or not. -- .''`. Aurelien Jarno | GPG: 1024D/F1BCDB73 : :' : Debian developer | Electrical Engineer `. `' [EMAIL PROTECTED] | [EMAIL PROTECTED] `- people.debian.org/~aurel32 | www.aurel32.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
