Subject: Re: Path MTU discovery
To: None <tech-net@NetBSD.ORG>
From: Michael Richardson <>
List: tech-net
Date: 07/14/1997 10:16:13

>>>>> "Kevin" == Kevin M Lahey <> writes:
    Kevin> In message
    Kevin> <>"Michael
    Kevin> C. Ri chardson" writes
    >> Gee, I thought 4.4 did Path MTU discovery, and thus NetBSD did.
    >> My impression, from reading the code is that we almost do, but
    >> don't.

    Kevin> Nope.  We definitely don't support it.  I don't know that
    Kevin> we are that close -- it still seems like a pain to have to
    Kevin> allocate the new host routes as we do the PMTU, clean 'em
    Kevin> up when they expire, etc.  It certainly would be nice to
    Kevin> have, but then I've been saying that for awhile, but I
    Kevin> haven't written any code for it.  :-(

  Does there exist any available PMTU code for 4.4 or 4.3 kernels that
would fit in?

  Alternatively, is there something reasonably available, that might
boot from a floppy that does it? Something with telnet or something on
the floppy? My test packet generation equipment is all NetBSD/i386,
and I need to test some DF bit handling code. I guess I could take a
look for a suitable Linux boot image..

  One question: wouldn't the host routes get allocated by the TCP
code, and referenced in their route cache pointer?

]     The food on Finnair flights is quite good really          | one quark   [
]   Michael Richardson, Sandelman Software Works, Ottawa, ON    | two quark   [
] | red q blue q[
] panic("Just another NetBSD/notebook using, kernel hacking, security guy");  [

Version: 2.6.3ia
Charset: latin1
Comment: Processed by Mailcrypt 3.4, an Emacs/PGP interface