Weird readings for mtp008 chip (Examined/Solved - 2.5.5 (CVS update on Jan 4))

[contact]

Hello!

I am using a Tyan 2510 Thunder LE motherboard with the ServerWorks? chipset on

it. Linux version: 2.2.14-5.

I downloaded the last CVS version of lm_sensors (January 4th, 5PM Eastern time)

and compiled both i2C and lm_sensors. Sensors_Detect does detect the mtp008

chip on the i2C bus. But when I run "sensors" I get data that does not seem

real to me. Here is the output from sensors:

mtp008-i2c-0-2e

Adapter: SMBus OSB4 adapter at 0580

Algorithm: Non-I2C SMBus adapter

VCore1: +1.72 V (min = +2.52 V, max = +3.08 V) ALARM

+3.3V: +2.70 V (min = +2.97 V, max = +3.63 V) ALARM

+12V: +15.50 V (min = +10.79 V, max = +13.11 V) ALARM

Vcore2: +4.08 V (min = +2.52 V, max = +3.08 V) ALARM

-12V: +0.49 V (min = -12.52 V, max = -11.54 V) ALARM

Vtt: +0.00 V (min = +1.34 V, max = +1.64 V) ALARM

CPUFAN1: 0 RPM (min = 3000 RPM, div = 2) ALARM

CPUFAN2: 0 RPM (min = 3000 RPM, div = 2) ALARM

CPUFAN3: 0 RPM (min = 3000 RPM, div = 2) ALARM

CPU1 Temp:+127.0C (min = +5C, max = +70C)

CPU2 Temp:+17.0C (min = +5C, max = +7C)

vid: +2.80 V

Is there any configuration I need to do to get it to work?

Thanks for your help!

Denis

P.S. I think this chip has just been added to the list. If you need beta

testers, we intend to do a lot of tests, so we could help. --- True that we added the mtp008 driver recently.

But your results are pretty ugly compared to others we have seen,

not sure why.

First of all are you sure there is a MTP008 on the board (look and see)?

Second of all did you run 'sensors -s' to set the limits?

(And then 'sensors;sleep 2; sensors' to clear the alarms)?

Third, for the temperatures, you can try changing the sensor

type by doing echo [1-3] > sensor[1-2]. See the code for

details, I don't think we've documented it yet.

Fourth, you can send us the output of 'i2cdump 0 0x2e'

so we can see the chip registers if necessary.

Not sure what the voltage problems are.

Follow up by sending mail to sensors@…

and referencing your ticket #.

MDS 1/5/01

---

Changed driver to preserve BIOS pin configuration,

hopefully this will make things better.

Fixed in CVS and what will be 2.6.0.

MDS 2/24/01