Hi, I have been wondering why inxi -Cxxx output displays cpu 3 always at 2667mhz, while the other cpu’s are at 2000mhz. Is there a possible issue with this cpu 3?
Thanks
Your “always” may not constitute a large enough sample size:
# cat /proc/cpuinfo | grep MHz cpu MHz : 3400.000 cpu MHz : 800.068 cpu MHz : 3400.000 cpu MHz : 3400.000 # cat /proc/cpuinfo | grep MHz cpu MHz : 800.005 cpu MHz : 3400.000 cpu MHz : 3400.000 cpu MHz : 800.025 # cat /proc/cpuinfo | grep MHz cpu MHz : 3400.000 cpu MHz : 3400.000 cpu MHz : 800.063 cpu MHz : 3400.000 # cat /proc/cpuinfo | grep MHz cpu MHz : 800.018 cpu MHz : 3400.000 cpu MHz : 3400.000 cpu MHz : 800.015 # cat /proc/cpuinfo | grep MHz cpu MHz : 3400.000 cpu MHz : 3400.000 cpu MHz : 800.084 cpu MHz : 3400.000 # cat /proc/cpuinfo | grep MHz cpu MHz : 3400.000 cpu MHz : 3400.000 cpu MHz : 2391.149 cpu MHz : 2631.298 # cat /proc/cpuinfo | grep MHz cpu MHz : 800.038 cpu MHz : 3400.000 cpu MHz : 3400.000 cpu MHz : 3400.000 # cat /proc/cpuinfo | grep MHz cpu MHz : 824.194 cpu MHz : 803.817 cpu MHz : 3400.000 cpu MHz : 3400.000 #
I’m guessing /proc/cpuinfo is the primary if not only source of inxi -C output.
Thanks for your response.
Thinkcentre-M57p:~/bin> cat /proc/cpuinfo | grep MHz
cpu MHz : 1995.178
cpu MHz : 1995.173
cpu MHz : 1996.520
cpu MHz : 2000.000
Thinkcentre-M57p:~/bin> inxi -C
CPU:
Info: quad core model: Intel Core2 Quad Q8400 bits: 64 type: MCP cache:
L2: 4 MiB
Speed (MHz): avg: 2265 min/max: 2000/2667 cores: 1: 1995 2: 1995 3: 2464
4: 2608
Thinkcentre-M57p:~/bin> inxi -Cxxx
CPU:
Info: quad core model: Intel Core2 Quad Q8400 bits: 64 type: MCP
smt: <unsupported> arch: Penryn rev: A cache: L1: 256 KiB L2: 4 MiB
Speed (MHz): avg: 2166 high: 2667 min/max: 2000/2667 cores: 1: 2000
2: 2000 3: 2667 4: 2000 bogomips: 21288
Flags: ht lm nx pae sse sse2 sse3 sse4_1 ssse3 vmx
Thinkcentre-M57p:~/bin>
paul@paul-Thinkcentre-M57p:~/bin> cat /proc/cpuinfo | grep MHz
cpu MHz : 2000.000
cpu MHz : 2000.000
cpu MHz : 2042.513
cpu MHz : 2336.826
paul@paul-Thinkcentre-M57p:~/bin> cat /proc/cpuinfo | grep MHz
cpu MHz : 1995.178
cpu MHz : 1995.145
cpu MHz : 2524.616
cpu MHz : 2584.166
paul@paul-Thinkcentre-M57p:~/bin> cat /proc/cpuinfo | grep MHz
cpu MHz : 1995.178
cpu MHz : 1995.193
cpu MHz : 1996.797
cpu MHz : 2000.000
paul@paul-Thinkcentre-M57p:~/bin> cat /proc/cpuinfo | grep MHz
cpu MHz : 2000.000
cpu MHz : 1995.178
cpu MHz : 1997.031
cpu MHz : 2079.743
paul@paul-Thinkcentre-M57p:~/bin> cat /proc/cpuinfo | grep MHz
cpu MHz : 1995.174
cpu MHz : 1994.740
cpu MHz : 2500.408
cpu MHz : 2000.000
paul@paul-Thinkcentre-M57p:~/bin> cat /proc/cpuinfo | grep MHz
cpu MHz : 1995.178
cpu MHz : 1995.178
cpu MHz : 1995.245
cpu MHz : 1995.210
paul@paul-Thinkcentre-M57p:~/bin> cat /proc/cpuinfo | grep MHz
cpu MHz : 1995.175
cpu MHz : 1995.142
cpu MHz : 1995.178
cpu MHz : 2007.388
paul@paul-Thinkcentre-M57p:~/bin> cat /proc/cpuinfo | grep MHz
cpu MHz : 1995.178
cpu MHz : 1995.163
cpu MHz : 1994.936
cpu MHz : 2000.000
paul@paul-Thinkcentre-M57p:~/bin> cat /proc/cpuinfo | grep MHz
cpu MHz : 1995.170
cpu MHz : 1995.178
cpu MHz : 2163.581
cpu MHz : 2516.965
paul@paul-Thinkcentre-M57p:~/bin> cat /proc/cpuinfo | grep MHz
cpu MHz : 1995.212
cpu MHz : 1995.173
cpu MHz : 2003.248
cpu MHz : 1999.613
paul@paul-Thinkcentre-M57p:~/bin> cat /proc/cpuinfo | grep MHz
cpu MHz : 1995.176
cpu MHz : 1995.178
cpu MHz : 1996.178
cpu MHz : 2232.576
paul@paul-Thinkcentre-M57p:~/bin>
What are your thoughts on this? To me it doesn’t look like it’s jammed up like when passing:
paul@paul-Thinkcentre-M57p:~/bin> inxi -C
CPU:
Info: quad core model: Intel Core2 Quad Q8400 bits: 64 type: MCP cache:
L2: 4 MiB
Speed (MHz): avg: 1996 min/max: 2000/2667 cores: 1: 2000 2: 1995 3: 1997
4: 1995
paul@paul-Thinkcentre-M57p:~/bin>
Appreciate your input on this, thanks.
I don’t recall ever seeing all 4 or more CPUs pegged at once in /proc/cpuinfo. IOW, varied reported speeds among cores is expected behavior, but probably more likely with 1 or 2 cores, or with hyperthreading.
The machine itself seems to be running smoothly, from what I can tell. Do you think that inxi output for sensors:
Sensors:
System Temperatures: cpu: 58.0 C mobo: N/A gpu: nvidia temp: 60 C
Fan Speeds (rpm): N/A gpu: nvidia fan: 30%
cpu temperature is high? I did install this cpu onto this board from another machine and did use new thermal paste while doing so. It’s fans are real quiet during normal operation and when coming out of suspend it cycles the fan up (right after login to plasma very quickly) and then the fan returns to a constant smooth sounding state.
Thanks
@panorain Does seem high to me, but sure there isn’t a stuck process running? What does top, or btop show?
My workstation shows;
Sensors:
System Temperatures: cpu: 27.0 C mobo: N/A gpu: nvidia temp: 39 C
Fan Speeds (rpm): N/A gpu: nvidia fan: 38%
It’s good to see the results of your system temperature for comparison, (it is a desktop correct)? Can you reference me to a program that could increase the designated speed for cpu fan? Looking into the machines case more deeply I can see the designated cpu fan is not spinning very fast at all (this may be typical ). I’ve been assessing some typical mainboard temp’s and it seems like the mainboard here is being reported as on the ‘up temp level’ of 'like compared
’ mainboard’s typical temperatures.
This email includes recent top output (1 zombie process). ← is 1 zombie process common?
Also just for the sake of documentation about this. The machine itself is free from any heavy dust buildup inside. The machine’s power supply fan is spinning and the additional rear case fan is also spinning.
-Regards
@panorain I have a HP Z440 workstation, it has a rear fan, memory coolers fans, CPU fan and power supply fan all controlled by the hardware and spins up as needed to date… My T400 has a card controlled fan, my Tesla P4 has a self install PWM fan and connected/controlled by the M/B front fan controller. The cpu is a Xeon 18 core (36 threads), memory cooler is for 8 x 16GB ECC RAM modules.
If you run (as root user) pwmconfig does it provide control/configuration?
I powercycled the machine and reviewed the BIOS settings available under ‘Processor Power Management’. All the options consider 'how to deal with ‘C states’. The following are the 4 available settings.
- Disabled
- GV1/GV3 Only
- C states only
- Enabled
The setting was at “GV1/GV3 only”, switching now to “Enabled” then saving. Seems temps are up a bit.
Sensors:
System Temperatures: cpu: 52.0 C mobo: N/A gpu: nvidia temp: 58 C
Fan Speeds (rpm): N/A gpu: nvidia fan: 29%
Is there a recommendation on dealing with “c states” you may be able to suggest?
paul@paul-Thinkcentre-M57p:~> sudo pwmconfig
[sudo] password for root:
# pwmconfig version 3.6.0
This program will search your sensors for pulse width modulation (pwm)
controls, and test each one to see if it controls a fan on
your motherboard. Note that many motherboards do not have pwm
circuitry installed, even if your sensor chip supports pwm.
We will attempt to briefly stop each fan using the pwm controls.
The program will attempt to restore each fan to full speed
after testing. However, it is ** very important ** that you
physically verify that the fans have been to full speed
after the program has completed.
/usr/sbin/pwmconfig: There are no pwm-capable sensor modules installed
paul@paul-Thinkcentre-M57p:~>
Thanks
paul@paul-Thinkcentre-M57p:~> sensors
coretemp-isa-0000
Adapter: ISA adapter
Core 0: +47.0°C (crit = +100.0°C)
Core 1: +40.0°C (crit = +100.0°C)
Core 2: +44.0°C (crit = +100.0°C)
Core 3: +41.0°C (crit = +100.0°C)
paul@paul-Thinkcentre-M57p:~>
@panorain So all controlled by the Motherboard then…
If you run sensors-detect --auto does it add any new ones?
paul@paul-Thinkcentre-M57p:~> sensors-detect --auto
Absolute path to 'sensors-detect' is '/usr/sbin/sensors-detect', so running it may require superuser privileges (eg. root).
paul@paul-Thinkcentre-M57p:~> sudo sensors-detect --auto
[sudo] password for root:
# sensors-detect version 3.6.0
# System: LENOVO 9088A83 [ThinkCentre M57p]
# Board: LENOVO LENOVO
# Kernel: 6.6.10-1-default x86_64
# Processor: Intel(R) Core(TM)2 Quad CPU Q8400 @ 2.66GHz (6/23/10)
Running in automatic mode, default answers to all questions
are assumed.
Some south bridges, CPUs or memory controllers contain embedded sensors.
Do you want to scan for them? This is totally safe. (YES/no):
Module cpuid loaded successfully.
Silicon Integrated Systems SIS5595... No
VIA VT82C686 Integrated Sensors... No
VIA VT8231 Integrated Sensors... No
AMD K8 thermal sensors... No
AMD Family 10h thermal sensors... No
AMD Family 11h thermal sensors... No
AMD Family 12h and 14h thermal sensors... No
AMD Family 15h thermal sensors... No
AMD Family 16h thermal sensors... No
AMD Family 17h thermal sensors... No
AMD Family 15h power sensors... No
AMD Family 16h power sensors... No
Hygon Family 18h thermal sensors... No
Intel digital thermal sensor... Success!
(driver `coretemp')
Intel AMB FB-DIMM thermal sensor... No
Intel 5500/5520/X58 thermal sensor... No
VIA C7 thermal sensor... No
VIA Nano thermal sensor... No
Some Super I/O chips contain embedded sensors. We have to write to
standard I/O ports to probe them. This is usually safe.
Do you want to scan for Super I/O sensors? (YES/no):
Probing for Super-I/O at 0x2e/0x2f
Trying family `National Semiconductor/ITE'... No
Trying family `SMSC'... Yes
Found unknown chip with ID 0x8b00
Probing for Super-I/O at 0x4e/0x4f
Trying family `National Semiconductor/ITE'... No
Trying family `SMSC'... No
Trying family `VIA/Winbond/Nuvoton/Fintek'... No
Trying family `ITE'... No
Some systems (mainly servers) implement IPMI, a set of common interfaces
through which system health data may be retrieved, amongst other things.
We first try to get the information from SMBIOS. If we don't find it
there, we have to read from arbitrary I/O ports to probe for such
interfaces. This is normally safe. Do you want to scan for IPMI
interfaces? (YES/no):
Probing for `IPMI BMC KCS' at 0xca0... No
Probing for `IPMI BMC SMIC' at 0xca8... No
Some hardware monitoring chips are accessible through the ISA I/O ports.
We have to write to arbitrary I/O ports to probe them. This is usually
safe though. Yes, you do have ISA I/O ports even if you do not have any
ISA slots! Do you want to scan the ISA I/O ports? (YES/no):
Probing for `National Semiconductor LM78' at 0x290... No
Probing for `National Semiconductor LM79' at 0x290... No
Probing for `Winbond W83781D' at 0x290... No
Probing for `Winbond W83782D' at 0x290... No
Lastly, we can probe the I2C/SMBus adapters for connected hardware
monitoring devices. This is the most risky part, and while it works
reasonably well on most systems, it has been reported to cause trouble
on some systems.
Do you want to probe the I2C/SMBus adapters now? (YES/no):
Using driver `i2c-i801' for device 0000:00:1f.3: Intel ICH9
Module i2c-dev loaded successfully.
Next adapter: SMBus I801 adapter at 1c20 (i2c-0)
Do you want to scan it? (YES/no/selectively):
Client found at address 0x50
Handled by driver `at24' (already loaded), chip type `spd'
(note: this is probably NOT a sensor chip!)
Client found at address 0x51
Handled by driver `at24' (already loaded), chip type `spd'
(note: this is probably NOT a sensor chip!)
Client found at address 0x52
Handled by driver `at24' (already loaded), chip type `spd'
(note: this is probably NOT a sensor chip!)
Client found at address 0x53
Handled by driver `at24' (already loaded), chip type `spd'
(note: this is probably NOT a sensor chip!)
Client found at address 0x5c
Probing for `Analog Devices ADT7462'... No
Probing for `SMSC EMC1072'... No
Probing for `SMSC EMC1073'... No
Probing for `SMSC EMC1074'... No
Next adapter: NVIDIA i2c adapter 0 at 1:00.0 (i2c-1)
Do you want to scan it? (yes/NO/selectively):
Next adapter: NVIDIA i2c adapter 1 at 1:00.0 (i2c-2)
Do you want to scan it? (yes/NO/selectively):
Next adapter: NVIDIA i2c adapter 6 at 1:00.0 (i2c-3)
Do you want to scan it? (yes/NO/selectively):
Next adapter: NVIDIA i2c adapter 10 at 1:00.0 (i2c-4)
Do you want to scan it? (yes/NO/selectively):
Now follows a summary of the probes I have just done.
Driver `coretemp':
* Chip `Intel digital thermal sensor' (confidence: 9)
Do you want to generate /etc/sysconfig/lm_sensors? (YES/no):
Created symlink /etc/systemd/system/multi-user.target.wants/lm_sensors.service → /usr/lib/systemd/system/lm_sensors.service.
Unloading i2c-dev... OK
Unloading cpuid... OK
paul@paul-Thinkcentre-M57p:~>
@panorain nothing new there then… looks like it is what it is, any room for another fan…
You need to find the Service manual for the system and see what the settings can/can’t do.
1 Like
Is my thought of
Driver `coretemp':
* Chip `Intel digital thermal sensor' (confidence: 9)
mean lm_sensors.service a service installed to system.d now then?
Ok I will check for more documentation and such.
Thanks
I do see that a fan can be added behind the front side of the machines case (on the inside). There is a connector for an additional fan on the mainboard.
Is it because the system was cheaply as possible that it doesn’t have fan inside front of case?
Which direction should fan inside front of case? Pull air in or push air out in this instance? I think pull air in right?
Too much pushing air out without enough unrestricted access for air to get in can lower the air pressure within the case, causing the air inside it to be less effective in drawing heat. In fans vs. out fans can be a juggling act in a poorly engineered case. All mere users can do is experiment. Anyway:
CPU yours mine
model Q8400 E8400
release Q2'09 Q1'08
lithography 45nm 45nm
cores 4 2
base freq 2.666 3.000
TDP 95 65
socket LGA775 LGA775
# inxi -CSs
System:
Host: gx780 Kernel: 6.5.9-1-default arch: x86_64 bits: 64 Console: pty pts/0
Distro: openSUSE Tumbleweed 20231114
CPU:
Info: dual core model: Intel Core2 Duo E8400 bits: 64 type: MCP cache:
L2: 6 MiB
Speed (MHz): avg: 2000 min/max: 2000/3000 cores: 1: 2000 2: 2000
Sensors:
System Temperatures: cpu: 37.0 C mobo: N/A
Fan Speeds (rpm): N/A
Mine described above is in a SFF Dell GX 780, which has one 80mm fan only for dual purpose CPU heat sink and air into the case, plus 60mm fan in PSU blowing out of the case. It was available with Core2Quad. Fan noise is usually nil except at power-up briefly. On occasion I have heard fan speed up rather slightly under load, but it doesn’t often experience much load.
# inxi -CSs
System:
Host: gx78b.ij.net Kernel: 6.5.9-1-default arch: x86_64 bits: 64
Console: pty pts/0 Distro: openSUSE Tumbleweed 20231114
CPU:
Info: dual core model: Intel Core2 Duo E8400 bits: 64 type: MCP cache:
L2: 6 MiB
Speed (MHz): avg: 2493 min/max: N/A cores: 1: 2992 2: 1995
Sensors:
Src: /sys System Temperatures: cpu: 44.0 C mobo: N/A gpu: radeon
temp: 36.5 C
Fan Speeds (rpm): N/A
Above, taken immediately after booting TW, differs in having half as much RAM, a passively cooled discrete GPU, HDD instead of SSD, but is otherwise an identical GX 780 twin. After a several minutes’ uptime, Radeon temp went up a bunch, while CPU temp went down:
# inxi -CSs
System:
Host: gx78b.ij.net Kernel: 6.5.9-1-default arch: x86_64 bits: 64
Console: pty pts/0 Distro: openSUSE Tumbleweed 20231114
CPU:
Info: dual core model: Intel Core2 Duo E8400 bits: 64 type: MCP cache:
L2: 6 MiB
Speed (MHz): avg: 1995 min/max: N/A cores: 1: 1995 2: 1995
Sensors:
Src: /sys System Temperatures: cpu: 42.0 C mobo: N/A gpu: radeon
temp: 50.0 C
Fan Speeds (rpm): N/A
If yours was mine, I’d thorougly remove the existing CPU/heatsink thermal compound or tape and apply new compound before deciding another cooling fan is needed.
Can thermal paste fail in say 4 years or there are bad batches out there? When I salvaged the CPU I did reapply a ‘pea’ sized amount to the cleaned (now installed) cpu. Your testing is more than excellent and i’m reviewing this and finally getting a few more answers put together. It’s worth a try though with new compound. Do you have a preferred compound or are you aware of what to obtain or better yet what not to obtain , as far as the thermal paste goes? Maybe even a fancy thermal paste would help. I experience the brief power-up and unsuspend fan activity increase at least similarity to what you are describing above.
Thinkcentre-M57p:~> inxi -CSs
System:
Host: paul-Thinkcentre-M57p Kernel: 6.6.10-1-default arch: x86_64 bits: 64
Desktop: KDE Plasma v: 5.27.10 Distro: openSUSE Tumbleweed 20240109
CPU:
Info: quad core model: Intel Core2 Quad Q8400 bits: 64 type: MCP cache:
L2: 4 MiB
Speed (MHz): avg: 998 min/max: 2000/2667 cores: 1: 2000 2: 1995 3: 0 4: 0
Sensors:
System Temperatures: cpu: 52.0 C mobo: N/A gpu: nvidia temp: 61 C
Fan Speeds (rpm): N/A gpu: nvidia fan: 30%
Thinkcentre-M57p:~>
I wonder if I should time this above command. Shy away from any suspending (just leaving the machine on in the OS) pass inxi -CSs again even after hours and hours. Thing is I just installed an SSD on this machine too, so prior it was operating with a platter drive.
- I guess the biggest question I have for you is which thermal compound would be something worth trying?
- Do you think the rear fan speed should possibly be increased?
I just don’t get why the cpu fan spins so slowly cpu fan is real close to the back of the case, so the rear fan can suck out the hot air with from more towards the rear. The rear case fan is real close to it. Rear case fan must be an 80mm while cpu fan is 60mm. Plus I recently added this NVIDIA K00 to it so hmm.
-Thanks
@panorain I use Arctic Silver Ceramique 2 here. Are the fans and the spare header 3 or 4 wire? You can always look at different fans and a controller to get things running along. But as @mrmazda indicated test before looking at another fan. An aftermarket CPU heatsink/fan assembly may improve the temperatures.
IMO proper application of thermal compound may be more important than the type. Peas come in different sizes, but the amount I use seems like about 1/12 or less of an average pea. What I use likely amounts to no more than a 3/32" ball, or less than the size of a grain of rice. I’ve always suspected the efficacy of thermal compound if not judiciously applied. A pea sized blob not thoroughly spread across the whole CPU surface might act as a better insulator than heat transfer agent. IOW, too much could well be worse than none if both surfaces are truly flat.
My previous acquisition of thermal compound was only 2 grams and lasted more than a decade of application to an ample but unknown number of CPUs and GPUs. The 4 gram MX-4 tube since acquired I expect to be the last needed (I’m 72).
60C on the GPU to me suggests its putting lots of heat into the case unless its fan directs exhaust straight outside, and if it does that, the case may need more forced into it than it’s getting for CPU needs.
I’ve never seen a cooler made for LGA775 with a fan size less than 80mm. Many use larger. If yours really is only 60mm, I’d find a more capable cooler, but still verify correct compound application first.