Chinh Nguyen wrote:
> If your interface is Point-to-Point, can you use MSG_DONTROUTE flag with
> sendmsg?
>
I just tried the MSG_DONTROUTE flag and it doesn't seem to make a
difference. And in the linux/net/ipv6/udp.c file I don't see where the
msg->msg_flags is checked for that, or passed to a f
On Thu, 2008-02-21 at 01:55 +0100, Matteo Croce wrote:
> Il Wednesday 07 November 2007 16:13:08 Dan Williams ha scritto:
> > Well, ideally we push the patch upstream and then people can send
> > updates to this list... I think it's really close and I feel like it
> > can be submitted for 2.6.25.
>
Il Wednesday 07 November 2007 16:13:08 Dan Williams ha scritto:
> Well, ideally we push the patch upstream and then people can send
> updates to this list... I think it's really close and I feel like it
> can be submitted for 2.6.25.
>
> Dan
I've seen so many updates in GIT, so I applied all up
I am only using IPv4-mapped address because it is the "recommended" (?)
way to specify an IPv4 destination address when using an IPv6 socket. By
recommendation, I mean this is the way unix networking books/man
pages/etc. describe the standard usage of a dual-stack socket (IPv6
socket supporting
In article <[EMAIL PROTECTED]> (at Wed, 20 Feb 2008 10:19:33 -0500), Chinh
Nguyen <[EMAIL PROTECTED]> says:
> Theoretically, I can specify the source address of the outgoing packet
> on an IPv6 socket by using sendmsg with IPV6_PKTINFO option enabled and
> in6_pktinfo. However, I cannot do this
Hi,
I know that linux networking stack is dual-stack. I can send/receive to
both IPv4 and IPv6 addresses with an IPv6 socket using send, sendto,
recvfrom, etc. For IPv4 addresses, I simply convert to IPv4-mapped IPv6
address when populating sockaddr_in6.
Theoretically, I can specify the sour
If your interface is Point-to-Point, can you use MSG_DONTROUTE flag with
sendmsg?
Chinh
--
http://www.certicom.com
Andrew May wrote:
I would really like to be able to send an IPv6 with a arbitrary IP out a
specific interface for a setup/probe type situation.
So there is no route setup yet, and
On Wed, 20 Feb 2008 14:42:16 +0100
Gergely Magyarosi <[EMAIL PROTECTED]> wrote:
> Hi!
>
> Thanks for the hint.
> It is rather difficult to switch to the newest kernel on this system. It
> uses a lot of custom-made patches, which should be adapted to 2.6.24.
> So this will take quite some time.
>
Hi!
Thanks for the hint.
It is rather difficult to switch to the newest kernel on this system. It
uses a lot of custom-made patches, which should be adapted to 2.6.24.
So this will take quite some time.
Meanwhile, I was wondering if this RST is a result of the local
application closing the so
