On Thu, Dec 06, 2001 at 09:23:17AM +0100, Kauder Thorsten wrote:
> Hi James,
> I guess you know that there's a problem sniifing mpls-packets if the machine
> running ethereal is the destination of these packets.
> I posted a message to the ethereal-user-mailinglist and that's what I've got as
> answer.
>
> >Well, if you can get the Linux networking stack to send *raw Ethernet*
> >frames to PF_PACKET/SOCK_RAW sockets *AND*, if the Linux MPLS code
> >modifies those frames in place, get it to do a copy-on-write of the
> >frame, so that it doesn't damage the copy of the frame sent to the
> >PF_PACKET/SOCK_RAW socket, that might help.
> >
> >I don't know how one would do that, however.
Yes. I know what they mean but I haven't had a chance to implement it.
I've seen some of the other networking stack use skb_cow() (copy on write).
It's a guess, but I think I need to do that after receiving the packet
in mpls_rcv().
> Maybe you can use that.
Thanks for the info,
Jim
> greetings
> Thorsten
>
> _______________________________________________
> mpls-linux-general mailing list
> mpl...@li...
> https://lists.sourceforge.net/lists/listinfo/mpls-linux-general
--
James R. Leu
jl...@mi...
|
