lm_sensors on ECS GeForce 6100PM-M2

[mittfh]

So I've compiled and installed lm_sensors (latest version) on my box (ECS GeForce 6100 PM-M2) and attempted to decipher the sensor names, but many are still a mystery.

If you're up to the challenge of a spot of detective work, read on...

 

Here's my "sensors" output:

[root@LinuxBox etc]# sensors
acpitz-virtual-0
Adapter: Virtual device
temp1:	   +40.0°C  (crit = +124.0°C)				  

k8temp-pci-00c3
Adapter: PCI adapter
Core0 Temp:  +12.0°C									
Core0 Temp:   +0.0°C									
Core1 Temp:   +2.0°C									
Core1 Temp:   +0.0°C									

it8716-isa-0290
Adapter: ISA adapter
in0:		 +1.30 V  (min =  +0.00 V, max =  +0.06 V)   ALARM
in1:		 +2.48 V  (min =  +0.00 V, max =  +0.06 V)   ALARM
in2:		 +1.84 V  (min =  +2.05 V, max =  +0.00 V)   ALARM
in3:		 +3.31 V  (min =  +0.03 V, max =  +0.00 V)   ALARM
in4:		 +1.20 V  (min =  +0.00 V, max =  +0.03 V)   ALARM
in5:		 +1.15 V  (min =  +0.54 V, max =  +0.00 V)   ALARM
in6:		 +2.94 V  (min =  +0.00 V, max =  +0.00 V)   ALARM
in7:		 +3.06 V  (min =  +0.00 V, max =  +2.69 V)   ALARM
VBat:		+2.99 V
fan1:	   2360 RPM  (min = 21093 RPM)  ALARM
fan2:		  0 RPM  (min = 5273 RPM)  ALARM
temp1:	   +15.0°C  (low  = +64.0°C, high =  +3.0°C)  ALARM  sensor = thermal diode
temp2:	   +34.0°C  (low  = +38.0°C, high =  +0.0°C)  ALARM  sensor = transistor
temp3:	   +25.0°C  (low  =  +0.0°C, high = +128.0°C)  sensor = transistor
cpu0_vid:   +1.525 V

 

My BIOS reports the following:

CPU Tcontrol: 40°C
System temperature: 34°C
CPU Fan Speed: 2360 RPM
CPU Vcore: 1.29 V
VDIMM: 1.84V

 

So the virtual adaptor temp1 is what the BIOS reports as "CPU Tcontrol" (overall temperature of the CPU die?)

 

In k8temp, temp1 and temp3 are misreporting core temperatures (IIRC, someone's written a shell script to correct this),

and temp2 and temp4 are non-reporting, so can be ignored (presumably the pins are there for the benefit of quad core Phenoms)

 

In it87, in0 is probably CPU Vcore, and in2 probably VDIMM.

temp1 may be CPU temp (i.e. should be reporting CPU Tcontrol)

temp2 is probably "System temperature"

temp3 is interesting - I didn't know I'd got another thermistor, so is it real or a fake?

 

Updating sensors3.conf with my findings results in this sensors output:

k8temp-pci-00c3
Adapter: PCI adapter
Core0 Temp:  +13.0°C									
Core1 Temp:   +4.0°C									

acpitz-virtual-0
Adapter: Virtual device
temp1:	   +17.0°C  (crit = +70.0°C)				  

it8716-isa-0290
Adapter: ISA adapter
Vcore:	   +1.30 V  (min =  +0.00 V, max =  +0.06 V)   ALARM
in1:		 +2.48 V  (min =  +0.00 V, max =  +0.06 V)   ALARM
VDIMM:	   +1.84 V  (min =  +2.05 V, max =  +0.00 V)   ALARM
in3:		 +3.31 V  (min =  +0.03 V, max =  +0.00 V)   ALARM
in4:		 +1.20 V  (min =  +0.00 V, max =  +0.03 V)   ALARM
in5:		 +1.15 V  (min =  +0.54 V, max =  +0.00 V)   ALARM
in6:		 +2.94 V  (min =  +0.00 V, max =  +0.00 V)   ALARM
in7:		 +3.06 V  (min =  +0.00 V, max =  +2.69 V)   ALARM
VBat:		+2.99 V
fan1:	   2368 RPM  (min = 21093 RPM)  ALARM
fan2:		  0 RPM  (min = 5273 RPM)  ALARM
CPU temp:	+17.0°C  (low  = +64.0°C, high =  +3.0°C)  ALARM  sensor = thermal diode
M/B temp:	+34.0°C  (low  = +38.0°C, high =  +0.0°C)  ALARM  sensor = transistor
temp3:	   +25.0°C  (low  =  +0.0°C, high = +128.0°C)  sensor = transistor
cpu0_vid:   +1.525 V

 

But what about the rest? Or the scaling formulae for the 5V and 12V rails?

The it87 section of sensor3.conf as supplied just labelled in8 as VBat. Full stop.

 

Aside: is it my imagination or do the alarm limits look rather suspicious?

[Guest Dublinwesley]

Hi Mittfh,

 

I'm facing the same problem you were facing with another Linux distribution on an ECS GeForce 6100 PM-M2. I was wondering how far you got with a solution and if you have a copy of your updated sensors.conf or sensors3.conf file?

 

From everything I've read so far - I believe that the sensors-detect is picking up the wrong chip driver for our mobo - or the sensors command is outputing too much information at the moment.

 

The sensors.conf file at the moment is a default one - with every chip supplied, and the program as I undersand will reference the last matching directives for output on usage.

 

I took a look at the motherboard specs from

http://www.ecs.com.tw/ECSWebSite/Product/Product_Detail.aspx?DetailID=988&CategoryID=1&MenuID=20&LanID=0

 

In the downloadable manual there is only 2 fans specified: 1. CPU_FAN and 2. SYS_FAN (mine is not connected) - and the motherboard is using the GeForce MP61 chipset not it87 (e.g. the chip driver should have "nforce" in it somewhere - per lm-sensors.org Device listings).

 

Also of note are the voltage limits for the fan connectors:

 

CPU_FAN Power Connector has 4 pins:

 

1. System Ground

2. +12V

3. Sensor

4. Control

 

SYS_FAN Power Connect has 3 pins:

1. System Ground

2. +12V

3. Sensor

 

My default sensors output is as follows:

acpitz-virtual-0
Adapter: Virtual device
temp1:       +30.0°C  (crit = +110.0°C)                  

it8716-isa-0a10
Adapter: ISA adapter
in0:         +1.12 V  (min =  +0.00 V, max =  +4.08 V)   
in1:         +2.46 V  (min =  +0.00 V, max =  +4.08 V)   
in2:         +1.86 V  (min =  +0.00 V, max =  +4.08 V)   
+5V:         +2.98 V  (min =  +0.00 V, max =  +4.08 V)   
in4:         +1.20 V  (min =  +3.82 V, max =  +4.08 V)   ALARM
in5:         +1.15 V  (min =  +0.00 V, max =  +4.08 V)   
in6:         +2.91 V  (min =  +0.00 V, max =  +4.08 V)   
5VSB:        +2.82 V  (min =  +0.00 V, max =  +4.08 V)   
Vbat:        +3.07 V
fan1:       3229 RPM  (min =    0 RPM)
fan2:          0 RPM  (min =    0 RPM)
fan3:          0 RPM  (min =    0 RPM)
temp1:       +14.0°C  (low  = +127.0°C, high = -53.0°C)  ALARM  sensor = thermal diode
temp2:       +33.0°C  (low  =  -1.0°C, high = +127.0°C)  sensor = thermistor
temp3:      -128.0°C  (low  =  -1.0°C, high = +127.0°C)  sensor = disabled

k10temp-pci-00c3
Adapter: PCI adapter
temp1:       +19.8°C  (high = +70.0°C)              

 

Clearly there is something wrong with in my output with IN4 as it is just a voltage setting, yet it is alarming at the moment (signified by the word ALARM after the line of text for the output). There is also something wrong with TEMP1 - which specifies a high of negative 53 degrees, and low of plus 127 degrees - it has "ALARM" set also - and might explain the high noise output from my PC (which is why I'm investigating this). I believe that the system is actively trying to reduce the temperature of the CPU to as close as possible to negative 53 degrees all the time (e.g. fan on constantly)

 

I think the solution to my problem would be to reduce my sensors.conf file so that only the chips that I know are being used are referenced, then following this try to update in4 with the correct limits and then add a compute line for temp1 so that the values set as upper and lower limits are actually more believable (probably includes a negative multiplier)

 

The reason I ask if you have an updated sensors.conf file for your machine is so that I can try and avoid clipping and pruning mine to match (since we're both using the same motherboard)

 

Thanks in Advance,

 

Wez

[Guest Dublinwesley]

Ok here's what I came up with after some messing around with sensors-detect

 

# lm_sensors configuration file for the ECS GeForce MP 6100 MP V:3 Motherboard
#CPU Temperature
chip "k10temp-*"
#label in0 "CPU Temp"
label temp1 "CPU Temp"
compute temp1 @*2,@/2	

chip "it8716-*"
### Voltages

label in0 "CPU Vcore"
ignore in1
label in2 "VDIMM"
ignore in3
label in4 "Fan Voltage V12"
ignore in5
ignore in6
ignore in7
ignore in8

compute  in4  @ * 10, @ / 10
set in4_min  10
set in4_max 13

### Temperatures

label  temp1  "UNKNOWN"
label  temp2  "System Temp"
ignore temp3

set temp1_type 4

compute temp1 @/2,@ *2
#set sensor3 0
#set in4_min 10
#set in4_max 13
set temp1_min 30
set temp1_max 80
#set temp2_min 10
#set temp2_max 40
### Fans
label  fan1  "CPU Fan"
label  fan2  "Case Fan - Disconnected"
ignore fan3
# Adjust for your own fans
set fan1_min 600
#set fan1_max 4000

 

Unfortunately even after configuring the Sensors for the motherboard to read out realistic values, I later discovered that the speed (RPM) of the fan can only be controlled if the fan is a 4 pin fan. Guess what - I had a stock 3 pin fan on my board which took the 12V and was constantly running at just over 3200 RPM. (the forth pin, as listed in the Mobo docs is for "Control")

 

I found a fan from an old CPU cooler I had with 4 pins and bolted it onto the stock heatsink, it's running fine with only 2 screws securing it (rest assured I'll go and buy a proper heatsink & cpu cooler asap). I made a lot more progress with this other fan installed.

 

Output of sensors for me just now is:

acpitz-virtual-0
Adapter: Virtual device
temp1:       +30.0°C  (crit = +110.0°C)                  

it8716-isa-0a10
Adapter: ISA adapter
CPU Vcore:               +1.12 V  (min =  +0.00 V, max =  +4.08 V)   
VDIMM:                   +1.86 V  (min =  +0.00 V, max =  +4.08 V)   
Fan Voltage V12:        +12.00 V  (min = +10.08 V, max = +12.96 V)   
CPU Fan:                 679 RPM  (min =  600 RPM)
Case Fan - Disconnected:   0 RPM  (min =    0 RPM)
UNKNOWN:                 +42.5°C  (low  = +30.0°C, high = +63.5°C)  sensor = thermistor
System Temp:             +36.0°C  (low  =  -1.0°C, high = +127.0°C)  sensor = thermistor

k10temp-pci-00c3
Adapter: PCI adapter
CPU Temp:    +53.0°C  (high = +140.0°C) 

 

After connecting a 4 pin fan I was then able to use pwmconfig to create a config file relating the sensor k10temp reading to the device it8716 fan speed.

(There is still some deduction to do for which device in sensors3.conf maps to which "hwmon" device from pwmconfig)

 

Found the following devices:
  hwmon0 is acpitz
  hwmon1/device is it8716
  hwmon2/device is k10temp

Found the following PWM controls:
  hwmon1/device/pwm1
  hwmon1/device/pwm2
  hwmon1/device/pwm3

Giving the fans some time to reach full speed...
Found the following fan sensors:
  hwmon1/device/fan1_input     current speed: 3245 RPM
  hwmon1/device/fan2_input     current speed: 0 ... skipping!
  hwmon1/device/fan3_input     current speed: 0 ... skipping!

Warning!!! This program will stop your fans, one at a time,
for approximately 5 seconds each!!!
This may cause your processor temperature to rise!!!
If you do not want to do this hit control-C now!!!
Hit return to continue: 

Testing pwm control hwmon1/device/pwm1 ...
 hwmon1/device/fan1_input ... speed was 3245 now 677
   It appears that fan hwmon1/device/fan1_input
   is controlled by pwm hwmon1/device/pwm1
Would you like to generate a detailed correlation (y)? 
Note: If you had gnuplot installed, I could generate a graphical plot.
   PWM 255 FAN 3245
   PWM 240 FAN 3245
   PWM 225 FAN 3125
   PWM 210 FAN 2973
   PWM 195 FAN 2743
etc...

 

After this it was a short step to create an /etc/fancontrol config file and run fancontrol to tame the fan!

 

I'm off to read the AMD documentation and double check the safe operating temperatures for this CPU. And now I have a bit of control over the speed of one fan I might be able to get some work done! :)

 

Wez