User Tools

Site Tools


realtime:documentation:howto:tools:cyclictest:start

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
realtime:documentation:howto:tools:cyclictest:start [2023/08/05 05:18]
costa.shul [More Information] + manpage
realtime:documentation:howto:tools:cyclictest:start [2025/01/19 17:42] (current)
costa.shul remove link to empty result-examples
Line 1: Line 1:
 ====== Cyclictest ====== ====== Cyclictest ======
  
-Cyclictest accurately and repeatedly measures the difference between a thread'​s intended wake-up time and the time at which it actually wakes up in order to provide statistics about the system'​s latencies. It can measure latencies in real-time systems caused by the hardware, the firmware, and the operating system.+Cyclictest accurately and repeatedly measures the difference between a thread'​s intended wake-up time and the time at which it actually wakes up from [[https://​man7.org/​linux/​man-pages/​man2/​clock_nanosleep.2.html|clock_nanosleep]] ​in order to provide statistics about the system'​s latencies. It can measure latencies in real-time systems caused by the hardware, the firmware, and the operating system.
  
 The original test was written by Thomas Gleixner (tglx), but several people have subsequently contributed modifications. Cyclictest is currently maintained by Clark Williams and John Kacur and is part of the test suite [[realtime:​documentation:​howto:​tools:​rt-tests|rt-tests]]. The original test was written by Thomas Gleixner (tglx), but several people have subsequently contributed modifications. Cyclictest is currently maintained by Clark Williams and John Kacur and is part of the test suite [[realtime:​documentation:​howto:​tools:​rt-tests|rt-tests]].
Line 50: Line 50:
 The table below contains a description of the abbreviated labels in the results: The table below contains a description of the abbreviated labels in the results:
  
-<WRAP center ​400px+<WRAP center>​ 
-^Abbreviation ​^ Label    ^ Description ​                                                      | +abr.  ​^ Label     ​^ Description ​                                                       ^ 
-|T            | Thread ​  ​| Thread index and thread ID                                        +| T             ​| Thread ​   | Thread index and thread ID                                         ​
-|P            | Priority | RT thread priority ​                                               +| P             ​| Priority ​ | RT thread priority ​                                                ​
-|I            | Interval | Intended wake up period for the latency measuring threads (in us) | +| I             ​| Interval ​ | Intended wake up period for the latency measuring threads (in us)  
-|C            | Count    | Number of times the latency was measured i.e. iteration count     ​+| C             ​| Count     ​| Number of times the latency was measured i.e. iteration count      
-|Min          | Minimum ​ | Minimum latency that was measured (in us)                         ​+| Min           ​| Minimum ​  ​| Minimum latency that was measured (in us)                          
-|Act          | Actual ​  ​| Latency measured during the latest completed iteration (in us)    +| Act           ​| Actual ​   | Latency measured during the latest completed iteration (in us)     ​
-|Max          | Maximum ​ | Maximum latency that was measured (in us)                         ​|+| Max           ​| Maximum ​  ​| Maximum latency that was measured (in us)                          |
 </​WRAP>​ </​WRAP>​
  
Line 78: Line 78:
  
 It can also be interesting to compare with results from other systems that were running loads, using various Cyclictest parameters, and running different real-time and non real-time kernels. These results give a general idea of the order of magnitude that the latencies should be. They also show the effect that different test conditions and parameters can have on the Cyclictest results. It can also be interesting to compare with results from other systems that were running loads, using various Cyclictest parameters, and running different real-time and non real-time kernels. These results give a general idea of the order of magnitude that the latencies should be. They also show the effect that different test conditions and parameters can have on the Cyclictest results.
- 
-<WRAP rightalign>​[[realtime:​documentation:​howto:​tools:​cyclictest:​result-examples|Read detailed examples of results]]</​WRAP>​ 
  
 ===== Test Design ===== ===== Test Design =====
Line 97: Line 95:
   * [[realtime:​documentation:​howto:​tools:​cyclictest:​faq|Cyclictest FAQ]]   * [[realtime:​documentation:​howto:​tools:​cyclictest:​faq|Cyclictest FAQ]]
   * [[https://​man.archlinux.org/​man/​cyclictest.8.en|cyclictest manpage]]   * [[https://​man.archlinux.org/​man/​cyclictest.8.en|cyclictest manpage]]
 +  * [[https://​man.archlinux.org/​man/​get_cyclictest_snapshot.8.en|get_cyclictest_snapshot]] – get a snapshot of running instances of cyclictest
  
realtime/documentation/howto/tools/cyclictest/start.1691212736.txt.gz · Last modified: 2023/08/05 05:18 by costa.shul