Hi Hans, Just been thinking a while about all these non-standard motherboards, and maybe it's time soon to implement something that enables lm-sensors to behave differently on different boards... One possibility could be to borrow some code from dmidecode to detect the make/model... And the hard part would be the implementation to keep the code-changes to a minimum for all the different drivers if the same driver should behave in a different way on different boards.. But it could at least enable lm-sensors to get if-statements for the sensors.conf depending on the specific mainboard that the person is using, and that small part could still be very useful, and should require just some minimal changes to the parser code and an addition of the dmidecode to detect the actual board. Another thing that this could be used for is to generate a known list of options that are needed for different mainboards, and maybe provide a list of modules for the motherboard instead of having to scan for sensors. Just brainstorming here but maybe they trigger a new idea in some developer that knows the code a bit better... And ignore the spelling... Not a native English-speaker and just about to fall asleep X-) /Patric Hans Edgington wrote: > Hi, > > Thanks for taking a look at the driver. > > >> I hacked the f71882fg driver so it works on my MSI P6N SLI Platinum >> mainboard (Socket 775 with Nvidia 650). >> > > >> The critical fix was to use the address values and logic from the >> existing f71805f driver (see diff). >> > > I haven't had much time to look at your work, but as I understand it your > version is kind of a cross between the f71882fg & f71805f driver. Which > means it now won't work on the Epox mobo's and possibly others. > So somehow we are going to have to detect which mobo is being used. > > >> Note that the driver still causes a kernel oops on rmmod, this is >> probably because f71882fg_exit doesn't call f71882fg_remove but I didn't >> feel like working on that, after all I don't remove the module at all... >> ;-) >> > > Is already fixed. > > > I'll have a closer look at it, as soon as I can. > > Regards, > Hans > > > _______________________________________________ > lm-sensors mailing list > lm-sensors at lm-sensors.org > http://lists.lm-sensors.org/mailman/listinfo/lm-sensors > >
