Hi Timothy, > Yep, I'll have a go at backporting then, mind you, the last time I > touched any C code was in a "Hello World" program! haha! ... So I'll > start reading around how to do this, of course any hints/tips would > be appreciated (though I recognise your time is important, and I dont > expect to be hand-fed). > > I've now downloaded a 22-June-05 build of "w83627ehf.c" (by Greg) > from GIT (which I dont really get understand)... though not sure > how/where to complile it yet ... (I'm just wing'n it! haha) >From there, create a Makefile with this line: obj-m += w83627ehf.o Then run: make -C /usr/src/linux SUBDIRS=$PWD modules (replace /usr/src/linux with the path to your real kernel source tree). That should do it. If there are warnings or errors, you'll have to figure out how to fix them (or ask us). Then modprobe the new driver and see how it goes. > >This should give us the exact ID. > > In the mean time, here's the "isadump" you asked for ... I built this > from the CVS version. > > [root at localhost dump]# ./isadump -y -k 0x87,0x87 0x2e 0x2f 0x0b > 0 1 2 3 4 5 6 7 8 9 a b c d e f > 00: ff ff ff ff ff ff ff 0b ff ff ff ff ff ff ff ff > 10: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff > 20: 88 63 ff 00 44 00 00 ff 50 04 00 00 9a 21 00 ff For the records, your chip has ID 0x8863, which matches the W83627EHF datasheet (it says 0x886x), but oddly enough not the ID of the only W83627EHF chip we saw so far (which had 0x8853). I have good hope that both chips will be compatible as far as hardware monitoring is concerned. Oh, and remember that my w83627ehf driver is not finished, it only implements half of the features the chip has, due to a lack of time and hardware to test my work. Someone else is working on improving it. Or I might do if someone considers a hardware donation. -- Jean Delvare
