[lm-sensors] Answer Epox - EPoX EP1308 sensor-chip

[khali@linux-fr.org (Jean Delvare)]

Hi Hans,

> I have been trying to find what some of the other fields mean this is what I
> have found out so far.
> 
> The f8 and f9 seem to contain the temperature at which the system shuts
> down.
> 5d-5e are the fintek ID, discovered by Jean.
> 
> Tried to find the fields for fan control and found the following:
> When setting the fan control, was disabled the field 96 a3-a5 a7-a9 and
> ab-ae changed. A3 and ab contain the same value, which I set in the bios,
> and is the percentage of max rpm at which the cpu fan can operate. Seems
> that the fields a4 and a5 are the max rpm of the fan a4 is the MSB and a5
> LSB. The fields get updated when a higher rpm is detected.
> Ab is  fast fan speed percnatge of max, ac field is the half fan speed, ad
> and ae are the lowest fanspeed setting which I can't change but are 30% of
> max.
> A6, a7 and a8 are the temperatures at which the fanspeed will increase.
> 
> Haven't found out why field 96 changed maybe to say that fan temps and
> speeds have been set?
> 
> Below are two isadumps which I used two discover the changes, first one
> before setting fan control second is after.
> 
>      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
> 00: ff 03 00 00 ff ff ff ff ff ff ff ff ff ff ff ff 
> 10: 00 00 00 00 ff ff ff ff ff ff ff ff ff ff ff ff 
> 20: d3 af ab 83 7a 8a 76 d6 c3 ff ff ff ff ff ff ff 
> 30: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff 
> 40: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff 
> 50: ff ff ff ff ff ff ff ff ff ff 03 04 10 19 34 ff 
> 60: 00 00 00 00 ff ff 02 74 00 00 ff 06 40 24 ff 00 
> 70: ff ff 2e ff 1f ff 00 ff ff ff ff ff ff ff ff ff 
> 80: ff ff 65 64 64 55 55 46 ff ff ff ff ff ff a8 ff 
> 90: 00 0c 0c 00 15 ff 57 ff 44 22 ff aa 55 55 ff 0a 
> a0: 01 d5 00 ff 02 23 3c 32 28 1e ff d9 b2 99 80 01 
> b0: 02 31 00 99 02 1b 3c 32 28 1e ff d9 b2 99 80 02 
> c0: 0f ff 00 80 03 ff 3c 32 28 1e ff d9 b2 99 80 03 
> d0: 0f ff 00 80 03 ff 3c 32 28 1e ff d9 b2 99 80 03 
> e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff 
> f0: 00 00 00 00 00 00 37 ff 22 ff ff 02 ff ff ff ff
> 
>      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
> 00: ff 03 00 00 ff ff ff ff ff ff ff ff ff ff ff ff 
> 10: 00 00 00 00 ff ff ff ff ff ff ff ff ff ff ff ff 
> 20: d3 af ab 84 7a 8a 76 d6 c7 ff ff ff ff ff ff ff 
> 30: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff 
> 40: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff 
> 50: ff ff ff ff ff ff ff ff ff ff 03 04 10 19 34 ff 
> 60: 00 00 00 00 ff ff 02 74 00 00 ff 06 40 24 ff 00 
> 70: ff ff 30 ff 1e ff 00 ff ff ff ff ff ff ff ff ff 
> 80: ff ff 65 64 64 55 55 46 ff ff ff ff ff ff a8 ff 
> 90: 00 0c 0c 00 15 ff 55 ff 44 22 ff aa 55 55 ff 0a 
> a0: 02 31 00 cc 01 bf 3c 28 14 0a ff cc 7f 4a 4a 01 
> b0: 02 1b 00 99 02 1b 3c 32 28 1e ff d9 b2 99 80 02 
> c0: 0f ff 00 80 03 ff 3c 32 28 1e ff d9 b2 99 80 03 
> d0: 0f ff 00 80 03 ff 3c 32 28 1e ff d9 b2 99 80 03 
> e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff 
> f0: 00 00 00 00 00 00 37 ff 02 ff ff 02 ff ff ff ff

That kind of reverse-engineering work is interesting (even though in a
perfect world we shouldn't have to do it), however I think it is more
important to first implement a driver with the basic monitoring
functions for which Epox provided the specifications to Age. It will
also be easier for you to start with something which is documented,
even lightly.

Note that I am unlikely to accept undocumented fan control functions in
the mainline kernel anyway. I don't want to put the user's system at
risk, and I don't want to encourage manufacturers to keep their
documentation secret.

Thanks,
-- 
Jean Delvare