Hi,
On Mon, Mar 28, 2016 at 10:46:04PM +0700, Adika Bintang Sulaeman wrote:
> Dear all linux-wpan developers and users,
>
> This is my first time exploring 6LoWPAN and I would like to ask you
> some questions. Before getting into questions, these are the devices
> and set up that I use:
> -Raspberry Pi Model B Rev 2
> -Raspbian OS
> -kernel from bluetooth-next ver 4.5.0-rc5+
> -MRF24J40MA Transceiver
>
> After setting up and installing the wpan-tools, I got my interface
> work by prompting "ip link show" and there is something like:
>
> 3: wpan0 <BROADCAST,NOARP> mtu 123 qdisc noop state DOWN mode DEFAULT
> group default qlen 1000
> link/ieee802.15.4 ee:3b:07:84:64:8c:2a:71 brd ff:ff:ff:ff:ff:ff:ff
>
> The first thing I notice strange is why after the command "sudo ip
> link add link wpan0 name lowpan0 type lowpan" the output in "ip link
> show" is a bit different, especially in link/[825]:
>
> 7: lowpan0 <BROADCAST,MULTICAST> mtu 1280 qdisc noop state DOWN mode
> DEFAULT group default qlen 1
> link/[825] ee:3b:07:84:64:8c:2a:71 brd ff:ff:ff:ff:ff:ff:ff
>
> So this is the FIRST QUESTION:
> 1. Is that normal/ok to have link/[825] instead of link/ieee802.15.4?
>
It's okay. lowpan0 != wpan0 these are different interfaces. It should be
link/6lowpan, but this interface type is just too new and iproute2
doesn't know it. That's why it shows the identifier of such type.
We need a patch for iproute2:
diff --git a/lib/ll_types.c b/lib/ll_types.c
index 2c5bf8b..6c71c8b 100644
--- a/lib/ll_types.c
+++ b/lib/ll_types.c
@@ -35,6 +35,7 @@ static const struct {
const char *name;
} arphrd_names[] = {
{ 0, "generic" },
+__PF(6LOWPAN,6lowpan)
__PF(ETHER,ether)
__PF(EETHER,eether)
__PF(AX25,ax25)
Feel free somebody to submit such patch! I cc jukkar here, because this
issue is related to btle 6lowpan also. btw: I didn't test it, just
compile.
The 6lowpan interface is the same type for 802.15.4 6LoWPAN and BTLE
6loWPAN, see slides [0].
> The second thing I notice strange is after "sudo ip link set wpan0 up"
> for both wpan0 and lowpan0, the "ifconfig" command shows that these
> network interfaces have "Link encap:UNSPEC". If I'm not mistaken, Link
> encap shows the hardware family of the interface, right?
>
> So the SECOND QUESTION is:
> 2. Is that normal/ok to have Link encap:UNSPEC? Because this thing
> makes me suspicious about the next problem that I face.
>
I think this requires a similar fix like above for "ifconfig". It's just
eyecandy. Remember: ip > ifconfig, because ifconfig is deprecated.
> I tried to run wpan-ping daemon from /wpan-tools/wpan-ping by
> prompting "wpan-ping --daemon" but the output is "bind: No such
> device". I tried to make my own code to make a server program which
> listen to this socket interface. But when binding with bind(), I got
> the same error status "No device found".
>
> This leads to my THIRD QUESTION:
> 3. Why "no device found" appears? The transceiver is attached, the
> hardware address exists, the interface is up, and I am able to ping
> between two devices using ping6. I am suspicious that the problems
> that I ask you before have caused this problem.
>
> Do you know what's wrong here?
>
Don't know, I never tried the socket interface for 802.15.4. Stefan
Schmidt did some stuff there, maybe he can help here.
btw: the socket interface need a complete rework/cleanup. [1]
What I test is the raw socket interface over AF_PACKET for 802.15.4 interfaces.
- Alex
[0] http://www.netdevconf.org/1.1/proceedings/slides/aring-generic-6lowpan-branch.pdf
[1] https://github.com/linux-wpan/linux-wpan/issues/20
--
To unsubscribe from this list: send the line "unsubscribe linux-wpan" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
