imx_3.14.28_1.0.0_ga + preempt-rt patch patch-3.14.28-rt25 on MCIMX6Q-SDP

[m_c]

Below is test program run on MCIMX6Q-SDP with imx_3.14.28_1.0.0_ga + preempt-rt patch patch-3.14.28-rt25. The red numbers are good, but others seems not good. Is there anything can be improve?

root@imx6qsabresd:~# ./testrun

1. 0.000010333 sec
2. 0.000020333 sec
3. 0.000004667 sec
4. 0.000005666 sec
5. 0.000021000 sec
6. 0.000019667 sec
7. 0.000004666 sec
8. 0.000020333 sec
9. 0.000004334 sec
10. 0.000004666 sec
11. 0.000004666 sec
12. 0.000004667 sec
13. 0.000004333 sec
14. 0.000005000 sec
15. 0.000005000 sec
16. 0.000018666 sec
17. 0.000004666 sec
18. 0.000004666 sec
19. 0.000020666 sec
20. 0.000005000 sec

Board: imx6qsabresd

Linux: imx_3.14.28_1.0.0_ga + preempt-rt patch patch-3.14.28-rt25 

Toolchain: poky 1.7

RootFS: core-image-minimal

1. testrun.c

#include <stdio.h>

#include <time.h>

#include <sched.h>

#include <string.h>

#include <unistd.h>

#include <sys/mman.h>

struct timespec diff(struct timespec start, struct timespec end)

{

struct timespec temp;

if ((end.tv_nsec-start.tv_nsec)<0) {

temp.tv_sec = end.tv_sec-start.tv_sec-1;

temp.tv_nsec = 1000000000+end.tv_nsec-start.tv_nsec;

} else {

temp.tv_sec = end.tv_sec-start.tv_sec;

temp.tv_nsec = end.tv_nsec-start.tv_nsec;

}

return temp;

}

int main(){

struct timespec time1, time2;

struct sched_param schedp;

if (mlockall(MCL_CURRENT | MCL_FUTURE))

perror("mlockall failed:");

memset(&schedp, 0, sizeof(schedp));

schedp.sched_priority = 90;

sched_setscheduler(0, SCHED_FIFO, &schedp);

while(1) {

  clock_gettime(CLOCK_REALTIME, &time1);

usleep(1);

  clock_gettime(CLOCK_REALTIME, &time2);

  printf("%ld.%09ld sec\n", diff(time1,time2).tv_sec, diff(time1,time2).tv_nsec);

  sleep(1);   

}

return 0;

}

[igorpadykov]

Hi m.c.

one can look at

Re: Long max latencies on Wandbaord / Freescale i.MX6 (Linux Real Time)

in general disabling low power modes/dvfs and running on one core should

improve latencies.

Best regards

igor