On Wed, 6 Feb 2013 15:00:58 +0100 (CET), Dag Wieers wrote: > On Wed, 6 Feb 2013, Jean Delvare wrote: > > > Not good. This can mean one of two things: either the chip has no > > register 0x3f and it nacks non-existent register read attempts, or it > > doesn't support SMBus read byte transactions at all. In the latter > > case, I strongly suggest that you leave the chip alone. As for the > > former case, there are a few thermal sensor chips which behave like, > > don't have a register at 0x3f and can use slave address 0x48: DS1621 > > and similar, DS75, LM73. > > It now consistently returns 0x3d, could this be a thermal value, like 61°C ? Could be, but for now all I see is that 0x3d == 0x3f - 0x02, i.e. only one bit changed. You will need to take more samples before you can draw a conclusion. Also compare when CU is idle and when it's busy. > The GPU reports 46°C (coming from nvidia-smi). -- Jean Delvare http://khali.linux-fr.org/wishlist.html _______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
