Re: [IPv6 Users] Multicast Loopback on DAD

2006-10-26 Thread YOSHIFUJI Hideaki / 吉藤英明 
In article <[EMAIL PROTECTED]> (at Fri, 27 Oct 2006 11:26:15 +0900), Suresh 
<[EMAIL PROTECTED]> says:

> But, again the same issue. Has Linux not implemented "LOOPBACK
> SUPPRESSION"? If the Neighbour Solicitation is just looped-back, will
> the DAD will fail? How the NS is differentiated between the loop-back and
> the NS from other node?

On Linux, if DAD ever looped-back, the DAD will fail.
Note that we assume the multicast will not loop-back in device driver level,
and we disable loop-back for ND messages (incl. DAD) in IP-layer level,
NS for DAD will not loop-back on Linux.

--yoshfuji
___
Users mailing list
Users@ipv6.org
https://lists.ipv6.org/mailman/listinfo/users

Re: [IPv6 Users] Multicast Loopback on DAD

2006-10-26 Thread Suresh 
Hi,

> This will only be evaluated on Token Ring devices.

I understood, Thank you.

But, again the same issue. Has Linux not implemented "LOOPBACK
SUPPRESSION"? If the Neighbour Solicitation is just looped-back, will
the DAD will fail? How the NS is differentiated between the loop-back and
the NS from other node?

with regards,
G Suresh


On Thu, 26 Oct 2006 10:18:28 -0400
Vlad Yasevich <[EMAIL PROTECTED]> wrote:

> Suresh wrote:
> > Hi,
> > 
> > I have doubt on the Multicast Loopback handling on DAD.
> > Does Linux-IPv6 confirm "9.  APPENDIX A: LOOPBACK SUPPRESSION &
> > DUPLICATE ADDRESS DETECTION" of RFC2462 ? According to the following
> > code, DAD will not work where two interfaces are using the same
> > link-layer address, since it simply discards the packet from other node
> > also. Am I right?
> > 
> > 
> >  if (dev->type == ARPHRD_IEEE802_TR) {
> >  unsigned char *sadr = skb->mac.raw;
> >  if (((sadr[8] ^ dev->dev_addr[0]) &
> > 0x7f) == 0 &&
> >  sadr[9] == dev->dev_addr[1] &&
> >  sadr[10] == dev->dev_addr[2] &&
> >  sadr[11] == dev->dev_addr[3] &&
> >  sadr[12] == dev->dev_addr[4] &&
> >  sadr[13] == dev->dev_addr[5]) {
> >  /* looped-back to us */
> >  goto out;
> >  }
> 
> This will only be evaluated on Token Ring devices.
> 
> -vlad


___
Users mailing list
Users@ipv6.org
https://lists.ipv6.org/mailman/listinfo/users

Re: [IPv6 Users] Multicast Loopback on DAD

2006-10-26 Thread Vlad Yasevich 
Suresh wrote:
> Hi,
> 
> I have doubt on the Multicast Loopback handling on DAD.
> Does Linux-IPv6 confirm "9.  APPENDIX A: LOOPBACK SUPPRESSION &
> DUPLICATE ADDRESS DETECTION" of RFC2462 ? According to the following
> code, DAD will not work where two interfaces are using the same
> link-layer address, since it simply discards the packet from other node
> also. Am I right?
> 
> 
>  if (dev->type == ARPHRD_IEEE802_TR) {
>  unsigned char *sadr = skb->mac.raw;
>  if (((sadr[8] ^ dev->dev_addr[0]) &
> 0x7f) == 0 &&
>  sadr[9] == dev->dev_addr[1] &&
>  sadr[10] == dev->dev_addr[2] &&
>  sadr[11] == dev->dev_addr[3] &&
>  sadr[12] == dev->dev_addr[4] &&
>  sadr[13] == dev->dev_addr[5]) {
>  /* looped-back to us */
>  goto out;
>  }

This will only be evaluated on Token Ring devices.

-vlad
___
Users mailing list
Users@ipv6.org
https://lists.ipv6.org/mailman/listinfo/users

[IPv6 Users] Multicast Loopback on DAD

2006-10-26 Thread Suresh 
Hi,

I have doubt on the Multicast Loopback handling on DAD.
Does Linux-IPv6 confirm "9.  APPENDIX A: LOOPBACK SUPPRESSION &
DUPLICATE ADDRESS DETECTION" of RFC2462 ? According to the following
code, DAD will not work where two interfaces are using the same
link-layer address, since it simply discards the packet from other node
also. Am I right?


 if (dev->type == ARPHRD_IEEE802_TR) {
 unsigned char *sadr = skb->mac.raw;
 if (((sadr[8] ^ dev->dev_addr[0]) &
0x7f) == 0 &&
 sadr[9] == dev->dev_addr[1] &&
 sadr[10] == dev->dev_addr[2] &&
 sadr[11] == dev->dev_addr[3] &&
 sadr[12] == dev->dev_addr[4] &&
 sadr[13] == dev->dev_addr[5]) {
 /* looped-back to us */
 goto out;
 }
 }


with regards,
 Suresh

___
Users mailing list
Users@ipv6.org
https://lists.ipv6.org/mailman/listinfo/users