CPU statistics not updating properly
This document (7009576) is provided subject to the disclaimer at the end of this document.
Environment
SUSE Linux Enterprise Server 11 Service Pack 2
SUSE Linux Enterprise Server 11 Service Pack 1
SUSE Linux Enterprise Server 11 Service Pack 1
Situation
Processor statistics show all zeros:
root@dev01:/ # mpstat -P 7
Linux 2.6.32.19-0.3-default (dev01) 07/21/2011 _x86_64_
09:09:29 AM CPU %usr %nice %sys %iowait %irq %soft %steal %guest %idle
09:09:30 AM 7 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00
09:09:31 AM 7 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
09:09:32 AM 7 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00
root@dev01:/ # mpstat -P 7
Linux 2.6.32.19-0.3-default (dev01) 07/21/2011 _x86_64_
09:09:29 AM CPU %usr %nice %sys %iowait %irq %soft %steal %guest %idle
09:09:30 AM 7 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00
09:09:31 AM 7 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
09:09:32 AM 7 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00
Resolution
Update to kernel-3.0.26-0.7.6 or later.
Additional Information
In prior releases, the linux kernel periodically interrupted each CPU
on a system at a predetermined frequency (100 Hz, 250 Hz, or 1000Hz)
depending on the platform. SLES 10 used 250 Hz on x86_64.
The kernel queried the CPU about the processes that it was executing and
used the results for process accounting and load balancing.
This is known as the "timer tick" and the kernel performed this interrupt
regardless of the power state of the CPU. Therefore, even an "idle" CPU
was responding to up to 1000 of these queries every second.
On systems that implemented power saving for idle CPUs, the timer tick
prevented the CPU from remaining idle long enough for the system to
benefit from power savings.
In the mainline linux kernel 2.6.21 and later, these periodic interrupts were
replaced with on demand interrupts. This is referred to as a "tickless" kernel.
This allows idle CPUs to remain idle long enough to benefit from power
savings resulting in lower core temperatures and lower power consumption.
Due to the tickless kernel, the statistics in /proc/stat don't get updated for the
CPU, which gets reported as zero due to the way that most software report
processor statistics.
on a system at a predetermined frequency (100 Hz, 250 Hz, or 1000Hz)
depending on the platform. SLES 10 used 250 Hz on x86_64.
The kernel queried the CPU about the processes that it was executing and
used the results for process accounting and load balancing.
This is known as the "timer tick" and the kernel performed this interrupt
regardless of the power state of the CPU. Therefore, even an "idle" CPU
was responding to up to 1000 of these queries every second.
On systems that implemented power saving for idle CPUs, the timer tick
prevented the CPU from remaining idle long enough for the system to
benefit from power savings.
In the mainline linux kernel 2.6.21 and later, these periodic interrupts were
replaced with on demand interrupts. This is referred to as a "tickless" kernel.
This allows idle CPUs to remain idle long enough to benefit from power
savings resulting in lower core temperatures and lower power consumption.
Due to the tickless kernel, the statistics in /proc/stat don't get updated for the
CPU, which gets reported as zero due to the way that most software report
processor statistics.
Disclaimer
This Support Knowledgebase provides a valuable tool for SUSE customers and parties interested in our products and solutions to acquire information, ideas and learn from one another. Materials are provided for informational, personal or non-commercial use within your organization and are presented "AS IS" WITHOUT WARRANTY OF ANY KIND.
- Document ID:7009576
- Creation Date: 17-Oct-2011
- Modified Date:14-Oct-2022
-
- SUSE Linux Enterprise Server
For questions or concerns with the SUSE Knowledgebase please contact: tidfeedback[at]suse.com
