On Sat, May 07, 2011 at 08:44:06PM +0200, Oliver Hartkopp wrote:
> Some more tryouts of the j1939 name stuff ...
> 
> # IP addr add dev vcan0 j1939 name 1122334455667788
> # jacd -r 100,80-120 -c /tmp/1122334455667788.jacd 1122334455667788 vcan0
> 
> and then
> 
> echo hello234 | jsr -v -v vcan0:20,3333 1122334455667788
> 
> i got with 'jspy -P' this:
> 
> vcan0:20,03300 1122334455667788 !6 [9] 68656c6c 6f323334 0a
> 
> candump:
I'll do a interpretation here:
> 
> (2305.450438)  vcan0  18EAFFFE  [3] 00 EE 00
request for all nodes' address claims (00ee00) on the bus. wait 1250 msec, 
everyone
can respond, and jacd will pick address.
>  (001.250165)  vcan0  18EEFF50  [8] 88 77 66 55 44 33 22 11
no one responed. no problem, pick 0x50.
$ ip addr will now list 0x50 (before, no address was assigned).
> 
TP.RTS
>  (013.149226)  vcan0  18EC5020  [8] 10 09 00 02 02 00 33 00
TP.CTS
>  (000.000050)  vcan0  18EC2050  [8] 11 02 01 FF FF 00 33 00
TP.DAT
>  (000.000036)  vcan0  18EB5020  [8] 01 68 65 6C 6C 6F 32 33
>  (000.000012)  vcan0  18EB5020  [3] 02 34 0A
TP.'EOF' (EOF is not right, it's called different, but it means
"we're done here"
>  (000.000057)  vcan0  18EC2050  [8] 13 09 00 02 FF 00 33 00
> 
> and 'jsr vcan0:1122334455667788' still produces 'hell' ;-)
ok, I had no time to look yet :-)
> 
> # IP addr show vcan0
> 4: vcan0: <NOARP,UP,LOWER_UP> mtu 16 qdisc noqueue state UNKNOWN
>     link/can
>     can-j1939 0x20 scope link
>     can-j1939 0x30 scope link
>     can-j1939 0x50 name 1122334455667788 scope link dynamic
> 
> Is this, what you would have expected
Yes, this looks very well (not also the mix between static & dynamic addressing 
...)
> (especially the byte order of the 'name' in the CAN frame)?
J1939 is little endian on the network :-)
> 
> Regards,
> Oliver

Regards,
Kurt

PS: my email access next week may be interrupted. I still try
to address this kind of questions! This kind of 'totally different view'
is very important.
_______________________________________________
Socketcan-core mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/socketcan-core

Reply via email to