Bug 217175

Summary: VID no read out correctly (Pentium D, Intel motherboard)
Product: Red Hat Enterprise Linux 5 Reporter: Aleksandar Milivojevic <alex>
Component: lm_sensorsAssignee: Phil Knirsch <pknirsch>
Status: CLOSED INSUFFICIENT_DATA QA Contact: David Lawrence <dkl>
Severity: medium Docs Contact:
Priority: medium    
Version: 5.0CC: rvokal, sghosh
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2009-03-11 16:43:24 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Aleksandar Milivojevic 2006-11-24 17:21:36 UTC
Description of problem:
Not sure if this is lm_sensors or kernel problem.  Anyhow here's description of
the problem.

I've 3.2GHz Pentium D (I beleive it's sSpec number is SL94Q) and Intel D945GNT
motherboard.  Sensors-detect recognized the onboard chip as lm85.  The specs for
the processor indicate that VRD 10.1 specification should be used (seems to have
same decoding of core voltages as VRD 10.0).

After setting "set vrm 10.0" and running sensors -s; sensors, I get in the output:

VCore:      +1.22 V  (min =  +1.03 V, max =  +1.14 V)   ALARM
vid:      +1.088 V  (VRM Version 10.0)

The VCore line is correct (BIOS reports almost the same values).  The voltage is
good (Intel's specs for the processor say VCore should be between 1.25V and 1.4V
for all versions of 3.2GHz Pentium D, currently there's two of them).

Looking at the VRD spec, 1.088V would correspond to all VID pins set to 0 (in
both VRM 10.0 and 10.1).  So I guess lm_sensors silently failed to read out VID,
and assumed all VID pins on processor were zero.

Version-Release number of selected component (if applicable):
lm_sensors-2.10.0-3.1
kernel-2.6.18-1.2747.el5

How reproducible:
Allways

Steps to Reproduce:
1. sensors-detect
2. set vrm 10.0 in appropriate section of /etc/sensors.conf
3. sensors -s
4. sensors
  
Actual results:
Incorrect readout of processor's VID pins (as if they were all zero)

Expected results:


Additional info:

Comment 1 Phil Knirsch 2009-03-11 12:02:23 UTC
Is this still a problem with the most current version of the kernel and lm_sensors as with Red Hat Enterprise Linux 5 Update 3 lm_sensors was rebased to a newer version.

Thanks & regards.

Comment 2 Aleksandar Milivojevic 2009-03-11 16:01:37 UTC
I don't have the particular hardware to test on anymore.  Unless somebody else chimes in, feel free to close.

Comment 3 Phil Knirsch 2009-03-11 16:43:24 UTC
Thanks for the info.

I'll close this bug then and in case it reappears we can easily reopen it.

Regards, Phil