[Fwd: xeontemp vs. adm1021]

Philip Edelbrock phil at edgedesign.us
Thu Jul 29 20:47:11 CEST 2004


That's a good question.  It certainly looks as if it is the CPU temp 
since it is quite a bit higher than the board temp.  Strange that the 
two CPU's are so different in temp readings, though.  It's also possible 
that those readings are actually for something else (like the power supply).

What I would try is to load the machine and monitor the temperatures. 
The board temps should rise slowly, but the CPU temps should rise 
quickly.  That should give you a easy and definitive answer in just a 
minute or so.


Phil

PS- Nice to see our stuff at work at Fermilab! ;')

-------- Original Message --------
Date: Thu, 29 Jul 2004 13:27:01 -0500 (CDT)
From: Steven Timm <timm at fnal.gov>
To: frodol at dds.nl, phil at netroedge.com, mdsxyz123 at yahoo.com
Subject: xeontemp vs. adm1021
Message-ID: <Pine.LNX.4.58.0407291316090.10290 at boxer.fnal.gov>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Quarentine-Loop: Processed
X-Spambayes-Classification: ham; 0.00


Hi Guys..

Was reading your documentation at

http://www2.lm-sensors.nu/~lm78/cvs/lm_sensors2/doc/chips/xeontemp

and

http://www2.lm-sensors.nu/~lm78/cvs/lm_sensors2/doc/chips/adm1021

My hardware, Supermicro X5DPA-GG motherboard,
dual Xeon 2.66GHz 533 FSB chips,512kb L2 cache
and
dual Xeon 3.06GHz 533 FSB chips,512kb L2 cache

Running kernel 2.4.20-30.7smp from Redhat with
lm_sensors 2.8.3 and i2c 2.8.3 modules loaded over top of it.

In modules.conf
alias char-major-89 i2c-dev

sensors-detect finds the following:
MODULE_0=i2c-i801
MODULE_1=adm1021
MODULE_2=w83781d
MODULE_3=eeprom
MODULE_4=smbus-arp

My question is the following:

The docs on xeontemp which I mention above say that 533 MHz
xeons don't have the external sensor for the die temp.
"unfortunately the 533 MHz FSB Xeons dropped the on-board sensor chip"

but the adm1021 docs say " If you have a Xeon and the adm1021
module loads and both temperatures appear valid, then things are good"

In my case the adm1021 module does load, and I get this:


root at fnpc151 root]# sensors
max1617-i2c-0-18
Adapter: SMBus I801 adapter at 0540
Algorithm: Non-I2C SMBus adapter
Board:       +47C  (low  =   -55C, high =  +127C)
CPU:         +53C  (low  =   -55C, high =   +78C)

max1617-i2c-0-1a
Adapter: SMBus I801 adapter at 0540
Algorithm: Non-I2C SMBus adapter
Board:       +48C  (low  =   -55C, high =  +127C)
CPU:         +73C  (low  =   -55C, high =   +78C)

The CPU temperatures go up and down with CPU activity, as
expected.

My question is--is this "CPU" temperature that I am
reading out the temperature of the die?  The vendor
is telling me that it is.  The documentation you have
hints that it should be too, but I am just confused by
the statement that 533MHZ FSB chips don't have this sensor.

At first with lm_sensors 2.6 we did not see the sensor
but once we ran the sensors-detect from lm_sensors 2.8.3
we could see it just fine and it appears to be functioning.

Is this sensor the die temperature?

Thanks for your help.

Steve Timm




[root at fnpc151 root]# cat /proc/cpuinfo
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 15
model           : 2
model name      : Intel(R) Xeon(TM) CPU 3.06GHz
stepping        : 9
cpu MHz         : 3066.856
cache size      : 512 KB
physical id     : 0
siblings        : 1
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 2
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca
cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm
bogomips        : 6121.06

processor       : 1
vendor_id       : GenuineIntel
cpu family      : 15
model           : 2
model name      : Intel(R) Xeon(TM) CPU 3.06GHz
stepping        : 9
cpu MHz         : 3066.856
cache size      : 512 KB
physical id     : 0
siblings        : 1
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 2
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca
cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm
bogomips        : 6121.06


------------------------------------------------------------------
Steven C. Timm, Ph.D  (630) 840-8525  timm at fnal.gov 
http://home.fnal.gov/~timm/
Fermilab Computing Division/Core Support Services Dept.
Assistant Group Leader, Scientific Computing Support Group
Lead of Computing Farms Team



More information about the lm-sensors mailing list