From owner-freebsd-hackers Sun Jul 6 20:58:43 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id UAA21832 for hackers-outgoing; Sun, 6 Jul 1997 20:58:43 -0700 (PDT) Received: from alpha.xerox.com (alpha.Xerox.COM [13.1.64.93]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id UAA21827 for ; Sun, 6 Jul 1997 20:58:41 -0700 (PDT) Received: from crevenia.parc.xerox.com ([13.2.116.11]) by alpha.xerox.com with SMTP id <15380(1)>; Sun, 6 Jul 1997 20:58:09 PDT Received: from localhost by crevenia.parc.xerox.com with SMTP id <177512>; Sun, 6 Jul 1997 20:57:58 -0700 To: dg@root.com cc: Archie Cobbs , freebsd-hackers@freebsd.org Subject: Re: VJ compression and MAX_HDR In-reply-to: Your message of "Sun, 06 Jul 97 17:53:37 PDT." <199707070053.RAA18621@implode.root.com> Date: Sun, 6 Jul 1997 20:57:57 PDT From: Bill Fenner Message-Id: <97Jul6.205758pdt.177512@crevenia.parc.xerox.com> Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk David Greenman wrote: > I think the case of IP+TCP header > MLEN is more than rare...I think it >never happens. Even with all options, the IP header doesn't exceed 64 bytes. Both IP and TCP headers have maximum lengths of 60 bytes. A maximum size IP + max size TCP header, however, is 120 bytes, which > MLEN. sl_compress_tcp() requires the IP and TCP headers to be contiguous; the only way to require that without requiring them to be shorter than MLEN is to require that sl_compress_tcp() be passed a cluster mbuf. Failing that, the code that checks TCP options (maybe just the last BCMP in the check-if-we-should-send-this-one-compressed) needs to learn about mbuf chains. Bill