On 15.07.2011 20:22, Oliver Hartkopp wrote: > On 15.07.2011 20:01, Mike Brown wrote:
>> >> What might need to do to get the Flexcan interface to behave the same as the >> virtual can interface? > > There was a problem that has been introduced in the 2.6.35 in the networking > stack that has been fixed in the stable series, see: > > http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.35.y.git;a=commitdiff;h=a5f18b1cca50987eb7773bb0be0d8730dfb10306 > > Please check, if this patch is also part of your 2.6.35.13 tree. Some additional hint: You can check, if your current setup has the described problem, with a test application called tst-rcv-own-msgs (in the SVN at socketcan/trunk/test): The major thing is to load the vcan driver with echo on driver level to identify your described problem. This is the test script from the Linux Test Project (ltp): # load needed CAN networklayer modules modprobe -f can modprobe -f can_raw # ensure the vcan driver to perform the ECHO on driver level modprobe -r vcan modprobe -f vcan echo=1 VCAN=vcan0 # create virtual CAN device ip link add dev $VCAN type vcan || exit 1 ifconfig $VCAN up # check precondition for CAN frame flow test HAS_ECHO=`ip link show $VCAN | grep -c ECHO` if [ $HAS_ECHO -ne 1 ] then exit 1 fi # test of CAN filters on af_can.c ./tst-filter $VCAN || exit 1 # test of CAN frame flow down to the netdevice and up again ./tst-rcv-own-msgs $VCAN || exit 1 exit 0 --- If this script ends with Starting PF_CAN frame flow test. checking socket default settings ... ok. check loopback 0 recv_own_msgs 0 ... ok. check loopback 0 recv_own_msgs 1 ... ok. check loopback 1 recv_own_msgs 0 ... ok. check loopback 1 recv_own_msgs 1 ... ok. PF_CAN frame flow test was successful. Then everything is fine in your kernel :-) Regards, Oliver _______________________________________________ Socketcan-users mailing list [email protected] https://lists.berlios.de/mailman/listinfo/socketcan-users
