Thanks, Porkchop and Chris and Michael! Porkchop wrote: > The chip, an IT8718, isn't responsible for whats on each line. Each line > can be attached to whatever the designer of your motherboard wanted to > attach it to. You need to find the exact model of mobo and start > googleing that instead of the chip type. > The only docs I could find so far on my MB are (IMO) inadequate. There seem to be common practices for connecting the sensor chip's inputs, and following those gives reasonable values for all the voltages except the two I mentioned. I miss the '80s, when you could actually look at the motherboard schematics to answer things like this.
Chris Knadle wrote: >> [EMAIL PROTECTED] ~]$ sudo modprobe coretemp >> FATAL: Error inserting coretemp >> (/lib/modules/2.6.22.15-desktop-1.uc1mdv/kernel/drivers/hwmon/coretemp.ko): >> Unknown symbol in module, or unknown parameter (see dmesg) >> > My best guess is that modprobe can't determine which of the lower-level > device modules are necessary to load, or it's also possible that modprobe > knows which one to load but the kernel as delivered by Mandriva didn't > include the required module. > Because of several problems, I ended up doing an OS reinstall, and I'm back to: [EMAIL PROTECTED] ~]$ sudo modprobe coretemp FATAL: Error inserting coretemp (/lib/modules/2.6.22.15-desktop-1.uc1mdv/kernel/drivers/hwmon/coretemp.ko.gz): No such device > If you want to fix your 'coretemp' issue, you don't need the source for > just the 'coretemp' module, you need the kernel source so that you can > recompile either the whole thing, or just the unerlying dependencies that may > be missing. You can either get the kernel source from Mandriva which has > their patches applied and options removed, or you can get the "vanilla" > kernel source from www.kernel.org. If you get the source from kernel.org, I > recommend the "full" download of 2.6.23.12, the latest "stable" kernel. > At this point, compiling an entire kernel may be beyond my abilities. Rather than go through the instabilities of the past few days, I think I'll just leave the core temp a mystery, as there are absolutely no signs of any overheating. > After expanding the kernel source, if you want a look, there's > documentation on hardware monitoring in /Documentation/hwmon. The "it87" > text file covers the IT8718F. There are even two downloads for compressed > PDFs that list which each pin of the IT8718F is for, though you probably > don't need it. > I found docs for the IT8718F, so that's under control. There seems to be standard practices for connecting the IT87* chips, and following those give reasonable values for all but two of the voltages it measures. "Typically" Vin5 and Vin6 are -12 and -5, scaled, but on my system in5 is always 0 and in6 is always 0.704V, and those don't jibe with the usual equations. Oh well. Michael Muller wrote: > I'm not at all familiar with the hardware monitoring stuff, but if you get it > working I'd appreciate it if you could post your results because I'm trying > to > track down some elusive hardware issues of my own ;-). > lm_sensors does a nice job of finding the sensors and displaying what the hardware monitor chip reports -- system voltages, fan speeds, temperatures. The problem can be figuring out what each input to that chip represents and how to scale it to the correct value. If you have any problems getting lm_sensors running, I'll be glad to share what little I know about it. Version 3.0.0 just came out, but I can't (yet) get it to compile, so I'm still using 2.10.4. Adam _______________________________________________ Mid-Hudson Valley Linux Users Group http://mhvlug.org http://mhvlug.org/cgi-bin/mailman/listinfo/mhvlug Upcoming Meetings (6pm - 8pm) MHVLS Auditorium Feb 6 - DBUS Mar 5 - Setting up a platform-independent home/small office network using Linux
