gpio_latency_measurements.txt

       
gpio latency measurements 
--------------------------

    * http://wiki.csie.ncku.edu.tw/embedded/xenomai
      => http://wiki.csie.ncku.edu.tw/embedded/xenomai#%E7%A4%BA%E6%B3%A2%E5%99%A8-%E5%AF%A6%E9%A9%97


            Source code

                https://github.com/jacky6016/GPIO-test

                https://github.com/erickitten/GPIO_test_xenomai3_2

            reslut      
                Driver model 	max latency(µs)

                rpi xenomai-2 RTDM        4.71

                beaglebone xenomai-2 RTDM 7.46

                rpi xenomai-3 RTDM        7.022

                rpi xenomai-2 linux native   10.92

                rpi xenomai-2 user-level(python)     244


                => seems uses only rpi 1 for gpio experiments  => no rpi3 in list!


* thesis: Real-Time Linux Testbench on Raspberry Pi 3 using Xenomai by GUSTAV JOHANSSON
      http://kth.diva-portal.org/smash/get/diva2:1251188/FULLTEXT01.pdf
   
  
    using image: 
      http://www.cs.ru.nl/lab/xenomai/raspberrypi.html
        Raspberry Pi image for the Pi zero,1,2,3 with Xenomai 3.0.5 on Raspbian linux 4.1.y Debian 8 jessie
        
  
    Table 7.1: Final results

    +-----------------------------------------------------------------------+
    |         | Together                     | Separate                     |
    |         |------------------------------+------------------------------|
    |         | Min      | Max     | Range   | Min      | Max     | Range   |
    |---------+----------+---------+---------+----------+---------+---------|
    | Reading | -28.10   | 38.10   | 66.20   | -28.10   | 38.10   | 66.20   |
    |         |us        |us       |us       |us        |us       |us       |
    |---------+----------+---------+---------+----------+---------+---------|
    | Writing | -142 us  | 116 us  | 258 us  | -38.10   | 18.10   | 56.20   |
    |         |          |         |         |us        |us       |us       |
    +-----------------------------------------------------------------------+
    
    
    What can be seen in the table is when the task Read and task Write was scheduled together, task Read caused worse
    performance for task Write, as earlier explained. What is interesting to see is that task Write has better
    performance than task Read when they were executed separately.