From: johngibson1541@yahoo.com
Date: Fri Jun 22 2007 - 21:43:41 ART
I know IPv6 routers will not fragment a packet.
This is the advantage according most.
So, the fragment header must be create by the
end host sending the packet.
Have you ever thought about why would the end
host make a packet that it needs to fragment
by itself ? If the end host knows the packet
is too big, why not just make 2 or more
packets rather than making 1 just to be
broken up by itself ?
Maybe the BSD API let the programming software
engineer user make as big an IPv6 packet as
he likes. Then the IPv6 stack in BSD model will
fragment it. Maybe thats just too much work
for the programmer to save those CPU cycles.
So, maybe this IPv6 fragment header is designed
intending to ease the programmer's job??
This archive was generated by hypermail 2.1.4 : Sun Jul 01 2007 - 17:24:51 ART