On Thu, Jan 28, 2010 at 4:16 PM, Wolfgang Grandegger <[email protected]> wrote: > Daniel Baluta wrote: >> On Thu, Jan 28, 2010 at 3:17 PM, Wolfgang Grandegger <[email protected]> >> wrote: >>> Daniel Baluta wrote: >>>> Hi, >>>> >>>> I'm writing tests for an Intel based CAN controller and I've now >>>> reached the part with bus off. >>>> Can you give me some hints about how to generate a deterministic test >>>> for bus off. >>>> >>>> I have a testcase where two nodes are communicating correctly using a >>>> given baud rate, >>>> and a third one tries to transmit using a different baud rate. Using >>>> this test I can generate >>>> bus off, but its not deterministic. >>> What do you mean with "it's not deterministic"? >> >> Not deterministic means that when I run the test multiple times I get >> different results. (e.g I don't get >> bus off each time ). > > I normally short-circuit CAN low and high to trigger bus-off.
thanks, this works just fine. anyhow, this implies either a manual configuration either a power switch. it must be another way of generating this bus off condition. > >>>> I'm reading the CAN specification, but still not reaching to clear >>>> testcase. >>> I don't understand yet what your problem is. >> >> Sorry, perhaps I didn't explained it very well. >> So. As state here ([1]), page 25 a CAN unit maybe in three states. My >> question is how CAN I design a test >> that always brings my CAN unit in bus off state. > > See above. Maybe there is a betetr method to realibly get a bus-off. > >> Is there any irc channel with topic CAN ? > > I personally don't use IRC _______________________________________________ Socketcan-users mailing list [email protected] https://lists.berlios.de/mailman/listinfo/socketcan-users
