Date: Sun, 31 Jan 1999 23:17:27 -0600 (CST) From: Richard Wackerbarth <rkw@dataplex.net> To: Archie Cobbs <archie@whistle.com> Cc: Poul-Henning Kamp <phk@critter.freebsd.dk>, julian@whistle.com, net@FreeBSD.ORG Subject: Re: netgraph... Message-ID: <Pine.BSF.4.05.9901312255030.31837-100000@nomad.dataplex.net> In-Reply-To: <199902010405.UAA25175@bubba.whistle.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 31 Jan 1999, Archie Cobbs wrote: > You can't avoid having the encoding and decoding done in the kernel. Sure you CAN. As you note below > If you wanted it *all* in user space then you have to: > > - Separate the encoding/decoding methods from the rest of > the node methods, or > - Do the complicated ELF sections stuff, or > - Some other complicated thing.. Is it so "complicated" to export a template (eg. format string) in response to a single call to the node? There are a number of advantages in getting the "guts" of the translation out of the kernel. For example, we might wish to do some date/time formatting based on the locale and time zone. IMHO, this is better kept out of the kernel. It should be sufficient for the node to designate that certain bits represent time stored in a particular format. > This is what we want to avoid -- and you were the first one to > argue for doing so! We all agree with that. So we're avoiding this > by having these two new message types. And, using your scheme, how do I compute the time interval between two messages? Do you propose to convert ALL packet fields for EVERY packet in a dump? Even if I care about only one or two fields? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.05.9901312255030.31837-100000>