Tag Info

Hot answers tagged

27

Wut?!? This is crazy, you know? It's unsupported and likely not good for your hardware. You should be using equally-spec'd processors in your server. I suppose you couple simply execute your application in a cgroup or shield (or via taskset) that only contains the CPU cores you wish to use. You can also pin memory access to that CPU with numactl. But ...


11

There's two entries for each core, because Nehalem (Xeon 55## series) supports two hardware threads per core, which to Linux appears as two processors. The CPU frequency varies because each core can be independently clocked; going by the physical id value the two you've posted appear to be different cores. Edit: Yes, frequency scaling is why the chips have ...


7

Your CPU and operating system are utilising Intel's SpeedStep technology. This allows the CPU to dynamically vary it's clock speed depending on load, etc. The 3GHz you see is the maximum that your CPU is capable of and the 2GHz amount is what it is currently running at.


6

On OpenBSD you have: sysctl hw.ncpu or sysctl hw.ncpufound As explained in sysctl(3), ncpu is the number of CPU used by system and ncpufound is the number of CPU found by the system. By the way, devio.us provides free shell account on OpenBSD servers.


5

An E5620 CPU has 4 cores on the die. With Hyper-Threading turned on, that gives you 8 threads. Perhaps your machine has 2 physical E5620 CPUs? If it's not powered on, try opening the case and counting them. If it's powered up and you can log in, try this: cat /proc/cpuinfo | grep “physical id” | sort | uniq | wc -l 2 physical CPUs x 4 cores/CPU x 2 ...


5

I'm not sure of the exact details as it is an issue from long ago, possibly as far back as when 386 based machines were common. The HLT instruction can only be called in "ring 0" when the CPU is not in "real" mode, so it should only called by the kernel in a modern OS. It instructs the processor to pause until the next interrupt is received. Modern CPUs ...


5

The lm flag indicates support for x86-64. For comparison, the flags of an older Intel Xeon: 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 pbe pebs bts cid xtpr


5

The processor speed in the model name line is part of the processor's model name. The processor's actual name is "Intel(R) Xeon(R) CPU     X5675 @ 3.07GHz". The processor speed in the "cpu MHz" line is the current CPU clock speed. Presumably, the CPU is not very busy now and is running at a reduced speed to save energy and keep cool. If ...


3

It's quite possible that this is just completely normal for that model. In some case designs one CPU gets a bit more airflow than the other - but they both get at least enough to keep them cool, which seems to be the case. With 12 cores, it's also possible one set is being used more heavily than the other, especially when the server isn't running flat out ...


3

My guess is that it's related to the processor affinity of the processes and applications that are running. If more processes/applications are running on one CPU than the other then it's going to naturally run hotter. If the temps of both CPU's are within "safe" ranges then this isn't something I'd be particularly worried about.


3

Original answer: A little bit of everything I would say. The AMD 3280 CPU product description specifies 8 cores and a base frequency of 2400 Hz. which doesn't match with your /proc/cpuinfo nor with the 8x2.3 GHz on your invoice... How to do CPU frequency scaling to conserve power? Paraphrased from https://wiki.debian.org/HowTo/CpuFrequencyScaling ...


3

Flag 'lm' -- long mode


3

RE: Unexpected CPU Speed You don't need to worry about this, it's automatic, a power/heat saving feature. When your computer needs full CPU power, it will automatically ramp up the processors to full speed, then ramp them back down when it's done. You can test this to verify it for yourself. Set up a simple number crunching app without any kind of ...


3

On FreeBSD, it's sysctl -n hw.ncpu and on Darwin sysctl -n hw.availcpu So perhaps it's similar? There is a system call, I think, but I'd poke around with the command line first.


2

The internet tells me to try this dmesg | grep -i CPU or sysctl -a | egrep -i 'hw.machine|hw.model|hw.ncpu' is cpuinfo available?? I know you can have it on BSD variants..


2

I encountered one! My first computer was a Soviet Iskra EVM (basically an IBM PC/XT with the Iron Curtain's own bus, but fully software compatible). On some rare instances it was freezing, sometimes producing garbage on the screen. Upon closer investigation I discovered: The system had a Siemens SAB 8086 CPU running at 8 Mhz. The culprit was the HLT ...


2

Have you tried booting a different kernel/distribution? Perhaps it's just a kernel bug. Try a fedora LiveCD, that should be suitably different from your Ubuntu.


2

The CPU has 4 Cores and 8 Threads. cpuinfo shows you logical CPUs so it shows you the 8 Threads of which every two share a core.


2

The package "cpuid" exists on FreeBSD, but this is not part of the base system. Output example: root@freeBSD > cpuid eax in eax ebx ecx edx 00000000 00000005 756e6547 6c65746e 49656e69 00000001 000106e5 01040800 00000201 178bfbff 00000002 55035a01 00f0b2e4 00000000 09ca212c 00000003 00000000 00000000 00000000 00000000 00000004 ...


2

This kind of thing is quite normal. Airflow, the arrangement of heat piping, and imperfectly balanced load can cause this. Even different types of loads can cause different power usage and thus heat dissipation between different processors. Also, it's totally inconsequential as all your processors should be running well below their operational maximum ...


2

Your processor is called Intel(R) Xeon(R) CPU X5675 @ 3.07GHz and contains a tech called Intel SpeedStep. Speedstep lowers the clock frequency of the CPU when the CPU has no work to do. This is mainly a power saving feature, and can normally be switched off in the bios.


1

Fix me if I'm wrong slot - imagine that you have 4 processors - slots are numbered from CPU 1 to CPU 4. This is physical slot - not a core or thread size - honestly I'm not sure but looks like FSB which can be treated as size of the processor multiplied by processor multiplier (for example 3.10GHz is 1600Mhz FSB * 1.93) capacity - I would not treat ...


1

Use lscpu for more useful information: $ sudo lscpu Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian CPU(s): 8 On-line CPU(s) list: 0-7 Thread(s) per core: 2 Core(s) per socket: 4 Socket(s): 1 NUMA node(s): 1 Vendor ID: GenuineIntel CPU family: ...


1

The speed reported in /proc/cpuinfo is the current speed of the processor. Intel processors include a technology called Speedstep. Speedstep dynamically adjusts the speed of the processor based on CPU utilization. It is fully supported in the Linux kernel since v2.6. I suspect that you're simply looking at /proc/cpuinfo when there is little to no load on ...


1

Try the dmidecode command to compare results.


1

Your kernel may be configured without SMP support. See the source for proc.c: unsigned int cpu = 0; ... #ifdef CONFIG_SMP cpu = c->cpu_index; #endif This cpu variable is then used in the output of the processor field in /proc/cpuinfo.


1

Looks like the second CPU has these extra flags (comments from linux-2.6.38-rc1/arch/x86/include/asm/cpufeature.h): dtes64 - 64-bit debug store ds_cpl - CPL Qual. Debug Store vmx - hardware virtualization smx - safer mode est - enhanced speedstep cx16 - CMPXCHG16B xtpr - send tax priority messages pdcm - performance capabilities sse4_1 - sse xsave - ...


1

What kernel are you on? there have been issues with reporting CPU speeds with some earlier versions - basically it's to do with CPU power management technologies - you may find you have some set in your BIOS.



Only top voted, non community-wiki answers of a minimum length are eligible