Re: Userspace testsuite issues
Paolo Mantegazza <[email protected]>
| Newsgroups | gmane.linux.real-time.rtai |
|---|---|
| Message-ID | <[email protected]> |
On 01/06/2016 07:24 PM, Emmanuel Pacaud wrote: > Hi, > > Le lun. 14 déc. 2015 à 18:42, Paolo Mantegazza > <[email protected]> a écrit : >> At the moment I've no idea. >> Just configure you latency dat by hand, e.g. 4000, 0,0 and go on. >> Run the latency test and adjust for a better figure, as it should >> have been done before. > > I have done that, but it doesn't help in making testsuite pass. > > I have tried to debug things a little, and found the problem is in the > rt_make_hard_realtime call. > That's interesting, you found a possible bug in something that has worked without problems for decades. So, my problem is now to understand what's going wrong in your environment. Not a simple task as I cannot reproduce it on any of the four machines(from P III to dual core I5, + an old OPTERON) I've available. Paolo > If I comment the call to this function, in latency.c for example, the > test seems to run (with overruns though): > > RTAI Testsuite - USER latency (all data in nanoseconds) > 2016/01/6 19:21:23 > RTH| lat min| ovl min| lat avg| lat max| ovl max| overruns > RTD| -1052| -1052| -1006| 7465| 7465| 17 > RTD| -1052| -1052| -1010| 7834| 7834| 17 > RTD| -1054| -1054| -1010| 7176| 7834| 17 > RTD| -1050| -1054| -1010| 7168| 7834| 17 > RTD| -1052| -1054| -1010| 8478| 8478| 17 > RTD| -1053| -1054| -1013| 7644| 8478| 17 > RTD| -1049| -1054| -1012| 7281| 8478| 17 > RTD| -1052| -1054| -1011| 7843| 8478| 17 > > If I let the rt_make_hard_realtime call, the program is blocked in > this call and never continue to the next line (rt_task_make_periodic > call). > > Any idea ? > > Cheers, > > Emmanuel. > _______________________________________________ Rtai mailing list [email protected] https://mail.rtai.org/cgi-bin/mailman/listinfo/rtai