Answering myself: I discovered a problem with the patch in 2.6.30-stable, see:
http://marc.info/?l=linux-netdev&m=125978891705230&w=2 Oliver Hartkopp wrote: > Matthias Fuchs wrote: >> Hi, >> >> what happened to this? >> >> I observed this behavior with Socket-CAN SVN r1091 on 2.6.30 kernels. > > Hi Matthias, > > this problem is *not* inside SocketCAN aka /linux/net/can/* ... > > It's inside the surrounding networking infrastructure. > > It has been fixed here: > http://patchwork.ozlabs.org/patch/28993/ > > Committed to Mainline Linux here (for 2.6.31) > http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=d55d87fdff8252d0e2f7c28c2d443aee17e9d70f > > Committed for 2.6.30-stable here (should be ok in 2.6.30.5) > http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.30.y.git;a=commitdiff;h=172570a224fe66d560c097e48fca15b620c76e72 > > So if you upgrade to 2.6.30.5 it should work as the problem is only between > 2.6.30 and 2.6.30.4 . > > Does that info help you? > > Regards, > Oliver > >> On Saturday 20 June 2009 14:42, Oliver Hartkopp wrote: >>> Oliver Hartkopp wrote: >>>> Michel Marti wrote: >>>> >>>>> I recently updated my Kernel from 2.6.26 to 2.6.30 and since then, CAN >>>>> frames sent over >>>>> the virtual CAN device (vcan) get looped back to the sender of the frame, >>>>> as if the >>>>> CAN_RAW_RECV_OWN_MSGS socket option was enabled (it isn't). Socketcan >>>>> from SVN-trunk shows >>>>> the same behaviour... Am I missing something or is this a bug? >>>> I'll take a look at it. >>>> >>>> Stay tuned :-) >>> Hi Michel, >>> >>> the needed sk reference for omitting the own received messages is killed by >>> a >>> new skb_orphan() call in net/core/dev.c introduced in >>> >>> http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.30.y.git;a=commitdiff;h=9a279bcbe347496799711155ed41a89bc40f79c5 >>> >>> So it first occurs in 2.6.30. >>> >>> Thanks for pointing me to this issue! >>> >>> I'll try to get some feedback from Herbert Xu about this and how we can >>> solve >>> these kinds of problems in the future. >>> >>> Additionally i'll create some test-application for LTP to be sure that >>> kernel >>> changes do not kill established CAN functionalities. >>> >>> Regards, >>> Oliver >>> >>> _______________________________________________ >>> Socketcan-users mailing list >>> [email protected] >>> https://lists.berlios.de/mailman/listinfo/socketcan-users >>> >>> > > _______________________________________________ > Socketcan-users mailing list > [email protected] > https://lists.berlios.de/mailman/listinfo/socketcan-users _______________________________________________ Socketcan-users mailing list [email protected] https://lists.berlios.de/mailman/listinfo/socketcan-users
