> >>> Hi Jörn, > >>> > >>> please don't top-post [1] (I'm already seeing too much of > >>> this @work... > >>> :->). > >>> > >>> Jörn Hoffmann wrote: > >>>> Thank you Jan! > >>>> Because i want to develop against the posix skin, i now > >>> added the posix wrapper options > >>>> to the linker call (posix.wrappers) and use the function > >>> calls without rt_dev prefix. > >>>> It works! > >>>> Unfortunately a new problem or misunderstanding occured. I > >>> want to set a socket to timeout > >>>> behaviour for receiving and used ioctl.......argh..... > >>> Everytime i tried to ioctl the socket > >>>> behaviour, no matter if blocking, nonblocking, timeouting > >>> (;)), recvfrom returned with error. > >>> > >>> What error do you get precisely (this time in errno again)? > >>> > >>>> If i rt_dev_ioctl the socket it works..hmm. But i've seen, > >>> that ioctl ist also wrapped by > >>>> Xennomai. Is there something more i've not noticed?? > >>> Did you use RTNET_RTIOC_TIMEOUT for setting the timeout? > >>> > >>> Jan > >>> > >>> [1] http://en.wikipedia.org/wiki/Top-posting#Top-posting > >>> > >>> > >> Ok. This time no top-posting. Hope thats better now?! > >> Oh..Oh...confusion's back again. Yes, i used RTNET_RTIOC_TIMEOUT. > >> Although ioctl should be wrapped, it seems i'm switching back > >> to plain linux, > >> because errno is now again setted by recvfrom(). I got errno=11. > >> The returncode of recvfrom() is -1. > >> > >> Jörn > >> > > Hmm....I now did it the native way but the problem is the > same. As long as > > i don't touch the socket via rt_dev_ioctl (using > RTNET_RTIOC_TIMEOUT) > > i can send and receive udp packets. When i use rt_dev_ioctl > with RTNET_RTIOC_TIMEOUT, > > no matter if i set the socket to blocking, nonblocking or > timeout, i can't receive > > any message with rt_dev_recv. The function always returned > with errorcode -11 immediately. > > Does anybody have any idea?? > > "Interesting" effect that I didn't came across yet. Could you > post your > test code (the simpler, the better)? > > Jan > Hi Jan, i hope it's ok to post my test code as an attachment. Hope it works with the mailing list. Since i inserted a while loop for the rt_dev_recvfrom(), to try to receive again if EAGAIN is returned, i now receive the right message (it seems i get it always with the second try). Maybe rt_dev_ioctl() takes some time!? But than i don't understand why the rt_dev_sendto() doesn't fail although it is called before the receive. Next is that it doesn't matter which socket behaviour i request. The receive always returns immediately.
Regards, Jörn
just_a_test.c
Description: just_a_test.c
------------------------------------------------------------------------- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________ RTnet-users mailing list RTnet-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/rtnet-users
