Date: Sun, 9 Jun 2002 23:09:37 -0700 From: Luigi Rizzo <rizzo@icir.org> To: Robert Watson <rwatson@freebsd.org> Cc: ipfw@freebsd.org Subject: Re: New ipfw code available Message-ID: <20020609230937.A53454@iguana.icir.org> In-Reply-To: <Pine.NEB.3.96L.1020610005651.89066B-100000@fledge.watson.org>; from rwatson@freebsd.org on Mon, Jun 10, 2002 at 12:57:44AM -0400 References: <20020608201909.A41807@iguana.icir.org> <Pine.NEB.3.96L.1020610005651.89066B-100000@fledge.watson.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Jun 10, 2002 at 12:57:44AM -0400, Robert Watson wrote: > Sounds very cool indeed. However, the usual question when hard-coded-ness > is traded for flexibility is: what's the performance like? Do you have > any performance measurements you could tell us about in the before/after > scenarios? You mention 'faster' as well as 'flexible', which bodes well > :-). i have not run any comparative test yet, that was the point of the posting, find some good soul who was willing to run the new and old code and compare performance :) Anyways in this case (barring stupid bugs in the implementation of course) it is rather obvious that the new architecture must be substantially faster -- the fact is, the old ipfw macroinstruction has to test some 20-25 distinct flags even when there is nothing to be done, all of which is filtered out by the compiler with the new approach. Things would be different if the macroinstructions were executed in hardware, but they were not... cheers luigi To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-ipfw" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020609230937.A53454>