Re: ebtables v 2.4.26?

scottmackay-/[email protected] said:
> I noticed the latest 2.4 kernel is 2.4.26.  Since I have issues with
> 2.6.x and some of the packet filtering needs I have, I am continuing
> development in the 2.4.x environment and am preparing to compile up
> the 2.4.26 for use.  I see the latest ebtables patch is made against
> 2.4.25.  To save some time DLing 2.4.25 source and comparing 2.4.25 to
> 2.4.26 (evil sateillite internet with their funky DL rate rules), can
> someone tell me if ebtables-brnf-5_vs_2.4.25 works vs .26?  Thanks in
> advance!!!

It almost applies, but not quite. There's a reject in
net/8021q/vlan_dev.c, in vlan_dev_hard_start_xmit():

--- linux-2.4.25/net/8021q/vlan_dev.c   2004-02-18 14:36:32.000000000 +0100
+++ linux-2.4.25-ebt-brnf/net/8021q/vlan_dev.c  2004-02-23 15:54:16.000000000 +0100
@@ -507,6 +507,10 @@ int vlan_dev_hard_start_xmit(struct sk_b
        stats->tx_packets++; /* for statics only */
        stats->tx_bytes += skb->len;

+       skb->protocol = __constant_htons(ETH_P_8021Q);
+       skb->mac.raw -= VLAN_HLEN;
+       skb->nh.raw -= VLAN_HLEN;
+
        dev_queue_xmit(skb);

        return 0;

The vlan code seems to have had significant changes between 2.4.25 and
2.4.26. It looks like this change isn't needed anymore.

At any rate, if you're not handling hardware-accelerated tagged vlan
frames, it's unlikely this will affect you. So you can probably delete
the above hunk from the patch (or ignore the reject).

-- 
Jason Lunz			Reflex Security
[email protected]		http://www.reflexsecurity.com/



-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
digital self defense, top technical experts, no vendor pitches, 
unmatched networking opportunities. Visit www.blackhat.com