In your previous mail you wrote:
I have one more question about the option. What if the size of an
outgoing packet is larger than the MTU of the outgoing interface on
the sending node, when the application set the IPV6_RECVPATHMTU
option? Should the kerenl pass the IPV6_PATHMTU ancillary data with
the MTU of the outgoing interface?
=> I can't see the issue. The kernel doesn't return ancillary data
on sending system calls. It will fragment the packet or return an
EMSGSIZE if fragmentation is disable, and will return the path MTU
(less or equal to the MTU of the outgoing interface) in an ancillary
data on a receiving system call or a getsockopt().
(BTW, our current implementation does not pass the ancillary data on
the sending node. In such a case, it just fragment the packet using
the MTU of the outgoing interface.)
=> IPV6_RECVPATHMTU is not supposed to do something on the sending side,
is it?
[EMAIL PROTECTED]
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page: http://playground.sun.com/ipng
FTP archive: ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------
