You are not logged in.
Hi!
lm_sensors/coretemp reports constant temperature for both cores:
uname -a
Linux nas 2.6.37-ARCH #1 SMP PREEMPT Tue Mar 15 09:21:17 CET 2011 x86_64 Intel(R) Celeron(R) CPU E3200 @ 2.40GHz GenuineIntel GNU/Linux
# sensors
coretemp-isa-0000
Adapter: ISA adapter
Core 0: +41.0°C (high = +76.0°C, crit = +100.0°C)
coretemp-isa-0001
Adapter: ISA adapter
Core 1: +35.0°C (high = +76.0°C, crit = +100.0°C)
...
Other values (fan speed, voltage,...) are live. But those +41.0°C and +35.0°C for cores never change.
Where to dig in?
Last edited by student975 (2011-03-21 19:39:11)
"I exist" is the best myth I know..
Offline
Run mprime or better yet, linpack. They should affect the temps. URL in my sig.
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
Run mprime or better yet, linpack. They should affect the temps. URL in my sig.
No-no! - it's NAS with OS on a flash USB stick And at any case cores' temperatures are always live. They change constantly. Every next 'sensors' run must show (at least slightly) another values. But those values (41C and 35C) are always the same (even after reboot).
"I exist" is the best myth I know..
Offline
...right, increased stress on cpu = increased heat
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
...right, increased stress on cpu = increased heat
Without any dedicated loading coretemp values are always slightly "moving". OK, in sense with your opinion I have tried to change governor with cpufrequtils (changing CPU freq from 1.2GHz to 2.4GHz and back). On my workstation I have plasmoid with coretemp graphics and have good sense what is it.
"I exist" is the best myth I know..
Offline
I know that some cpu/board combos that make that behaviour... DTS sensors stops working. But probably the motherboard sensor works no? (use lm_sensors). You could try to search for a bios update to see if it solves that problem
.::. TigTex @ Portugal .::.
Offline
I know that some cpu/board combos that make that behaviour... DTS sensors stops working. But probably the motherboard sensor works no? (use lm_sensors). You could try to search for a bios update to see if it solves that problem
I have already searched all was able to imagine , but have not found any similar problems. BIOS has the only (first) version yet. Also I have sent request to motherboard manufacturer (and it is GIGABYTE), will wait for answer..
"I exist" is the best myth I know..
Offline
OK, can some Celeron E3200 owner verify coretemp does work at all for this CPU? To see a core temp you don't need to install lm-sensors. Just
sudo modprobe coretemp
to load the module. Then, to see first core temperature, repeat few times
sudo cat /sys/devices/platform/coretemp.0/temp1_input
You will see a temp in millidegree.
"I exist" is the best myth I know..
Offline
Up
"I exist" is the best myth I know..
Offline
Have you tried checking your temperatures with windows? (just to see if it really is a bios bug and not a linux one)
.::. TigTex @ Portugal .::.
Offline
OK, it seems this CPU has absolutely different behavior rather I know about other CPUs. At last I have found a way to change core temps - it is just 'stress' package.
After ~30min of testing I have got +39C and +48C. And there are two big differences in comparison with I saw with other processors:
- temperatures reproting has very high inertia (it seems values are integrated with long time period)
- there is a big difference between two cores
Probably, it is CPU-specific behavior..
@TigTex
I haven't got windows at all during last 10 years
"I exist" is the best myth I know..
Offline
@op - how did you stress test? Are you sure both cores are being tested? Run htop to verify.
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
@op - how did you stress test? Are you sure both cores are being tested? Run htop to verify.
Of course, I have checked with 'top' I have tried 'stress -c N' with different N (2, 3, 4, 5) with the same result.
"I exist" is the best myth I know..
Offline