i multiplied number of nanoseconds to wait * 1000 (i forgot it was not 10^9 not 10^6) and now it works:)  but still i dont get why it was called so many times even with fixed step in 1 step

now i got (result below) and if i coose number of intervals 2 it computes 4 intervals (if i dont include last one), which is kinda odd but I will consider this

simulate(Segway.simulation,cflags="-g",measureTime=true,method="rungekutta")
record SimulationResult
    resultFile = "/home/fei_admin2/NetBeansProjects/OMCklient/testing/Segway.simulation_res.mat",
    simulationOptions = "startTime = 0.0, stopTime = 0.2, numberOfIntervals = 2, tolerance = 1e-06, method = 'rungekutta', fileNamePrefix = 'Segway.simulation', options = '', outputFormat = 'mat', variableFilter = '.*', cflags = '-g', simflags = ''",
    messages = "0.05000000|0.049978|0.000021|49978279|0|27228704|1|0
0.10000000|0.049249|0.050750|49248864|0|27228704|1|0
0.15000000|0.149976|0.000023|149975904|0|27228704|1|0
0.20000000|0.049730|0.150269|49730294|0|27228704|1|0
0.20000000|0.199975|0.000024|199974647|0|1|16|0
",
    timeFrontend = 0.403733309,
    timeBackend = 0.028026882,
    timeSimCode = 0.011971422,
    timeTemplates = 0.017074025,
    timeCompile = 1.061730883,
    timeSimulation = 0.509047586,
    timeTotal = 2.031659746
end SimulationResult;
Warning: The initial conditions are not fully specified. Use +d=initialization for more information.
Warning: There are iteration variables with default zero start attribute. Use +d=initialization for more information.

Hard to tell what went wrong without seeing the code. Did you synchronize on events or in a regular equation? Anyway, the version after OMC 1.9.4 will probably do real-time synchronization on a flag. You can test it out from my branch: https://github.com/sjoelund/OMCompiler/tree/opc-da (passing -rt 1 to synchronize to real-time).

thank you for your response I am developing currently in Linux is that branch also 4 linux ?  will I receive it also if I download latest night build?

well i synchronize it outside modelica I know it is quite messy and a few microseconds out of RT but it works:)
now I figured that why isnt probably example DACSRealTimeSynchronization.mo working - but i havent tested it yet its only a possibility(it fixed my code though).
In external function they are using rt_tick(0); / rt_tock(0); but I found out after a few hours that this timer is being reseted somewhere in modelica probably, so I changed to rt_tick(1); / rt_tock(1); and now it works:)

heres my code
// function which launches only once in algorithm in algorithm if initial() .... calling external function .... endif
double initTimetck(double tn)
{
  rt_tick(1);
  return 0.0;
}

//here things happens (I comunicate via shared memory but that is not important for anyone)

double toktime=rt_tock(1);
    double timesimtogo = (timesim-toktime);
    if (0 < timesimtogo ){
        struct timespec timesleep,timeleft;
        timesleep.tv_nsec=((timesim-toktime) - ((long int)timesimtogo) )  * 1000000000  ;
        timesleep.tv_sec=(long int) timesimtogo;
        int test = nanosleep(&timesleep,&timeleft);

        printf("%.8f|%f|%f|%ld|%ld|%ld|%ld|%d\n", timesim, timesimtogo,toktime,timesleep.tv_nsec, timesleep.tv_sec,timeleft.tv_nsec, timeleft.tv_sec,test);

       printf("%d\n", test);
        }

this function is called each step of simulation
so far it works i just hope there is no other tick(timer reset) for this timer 1 )







       

Yes, the branch works on Linux. And no, you won't get it pre-built from anywhere. rt_tick(1) is used if you use measureTime=true (it is the clock measuring the time for initialization). There won't be any spare timers when measureTime=true is given.