Hi Pekka,
> The isimodem GPRS connections (PHONET_PIPEs) have ARP header type 821.
> ---
> src/rtnl.c | 5 +++++
> 1 files changed, 5 insertions(+), 0 deletions(-)
>
> diff --git a/src/rtnl.c b/src/rtnl.c
> index 83f7952..ec9a5a7 100644
> --- a/src/rtnl.c
> +++ b/src/rtnl.c
> @@ -38,6 +38,10 @@
>
> #include "connman.h"
>
> +#ifndef ARPHDR_PHONET_PIPE
> +#define ARPHDR_PHONET_PIPE (821)
> +#endif
> +
> #define print(arg...) do { } while (0)
> //#define print(arg...) connman_info(arg)
>
> @@ -398,6 +402,7 @@ static void process_newlink(unsigned short type, int
> index, unsigned flags,
> switch (type) {
> case ARPHRD_ETHER:
> case ARPHRD_LOOPBACK:
> + case ARPHDR_PHONET_PIPE:
> case ARPHRD_NONE:
> __connman_ipconfig_newlink(index, type, flags,
> str, mtu, &stats);
we don't really need this since we are detecting ISI GPRS connections
via oFono actually. Do you want this for the statistics counting? If so,
then yes, I forgot to think about that part. Even PPP and HSO might be
affected here. Unless both are actually using ARPHRD_NONE which I have
to double-check.
Does this have any side effect that we might have not thought about when
it comes to device detection.
In the end I wanted to remove the udev based detection from ConnMan and
do everything via RTNL. udev is not really needed by ConnMan at all
anymore since all 3G and Bluetooth detection is done via oFono and BlueZ
these days. And only oFono needs udev since it deals with TTYs.
Regards
Marcel
_______________________________________________
connman mailing list
[email protected]
http://lists.connman.net/listinfo/connman
