Date: Mon, 27 Jun 2005 13:12:03 -0700 From: Alfred Perlstein <alfred@freebsd.org> To: Luigi Rizzo <rizzo@icir.org> Cc: cvs-src@FreeBSD.org, Gleb Smirnoff <glebius@FreeBSD.org>, Ruslan Ermilov <ru@FreeBSD.org>, cvs-all@FreeBSD.org, src-committers@FreeBSD.org Subject: Re: cvs commit: src/sys/netinet/libalias alias_ftp.c alias_irc.c alias_local.h alias_proxy.c alias_skinny.c alias_smedia.c alias_util.c Message-ID: <20050627201203.GF40423@elvis.mu.org> In-Reply-To: <20050627083101.B56456@xorpc.icir.org> References: <200506270736.j5R7a3OZ036531@repoman.freebsd.org> <20050627152155.GF93072@ip.net.ua> <20050627083101.B56456@xorpc.icir.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Vendors don't sell performance, they sell features. (you can quote me on that.) * Luigi Rizzo <rizzo@icir.org> [050627 08:31] wrote: > On Mon, Jun 27, 2005 at 06:21:55PM +0300, Ruslan Ermilov wrote: > ... > > Why can't we just say that checksum offloading is incompatible > > with NAT (like with many other things), and do the software > > checksum calculations in libalias? > > actually the more i see it the more i think checksum offloading > is a disgrace rather than a performance boost. > > it needs a lot of special cases throughout the protocol stack > to be supported properly, which constitutes extra overhead > with low-end hardware which does not support the offloading; > > it does not pay on small packets such as acks where you have > to touch the whole packet anyways; > > some hardware needs the checksum engine to be reprogrammed when > changing protocol type (tcp <-> udp) which require extra I/O > cycles on the bus that are expensive; > > some hardware has broken checksum engines; > > cheers > luigi -- - Alfred Perlstein - email: bright@mu.org cell: 408-480-4684
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050627201203.GF40423>