From owner-freebsd-current@FreeBSD.ORG Wed Nov 1 04:53:35 2006 Return-Path: X-Original-To: freebsd-current@freebsd.org Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 74FB316A407 for ; Wed, 1 Nov 2006 04:53:35 +0000 (UTC) (envelope-from sam@errno.com) Received: from ebb.errno.com (ebb.errno.com [69.12.149.25]) by mx1.FreeBSD.org (Postfix) with ESMTP id 24F9643D45 for ; Wed, 1 Nov 2006 04:53:35 +0000 (GMT) (envelope-from sam@errno.com) Received: from [10.0.0.80] ([10.0.0.80]) (authenticated bits=0) by ebb.errno.com (8.13.6/8.12.6) with ESMTP id kA14rYIT041303 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Tue, 31 Oct 2006 20:53:34 -0800 (PST) (envelope-from sam@errno.com) Message-ID: <45482850.4060500@errno.com> Date: Tue, 31 Oct 2006 20:53:36 -0800 From: Sam Leffler Organization: Errno Consulting User-Agent: Thunderbird 1.5.0.7 (Macintosh/20060909) MIME-Version: 1.0 To: Nicolas Blais References: <200610311629.06271.nb_root@videotron.ca> <200611010358.kA13wprx067313@lava.sentex.ca> <200610312326.05311.nb_root@videotron.ca> In-Reply-To: <200610312326.05311.nb_root@videotron.ca> X-Enigmail-Version: 0.94.1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: freebsd-current@freebsd.org, Mike Tancsa Subject: Re: Hifn 7955/7956 crypto accelerator questions X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 01 Nov 2006 04:53:35 -0000 Nicolas Blais wrote: > They do claim 500mbps throughput for the vpn1461 and 250mbps for the > vpn1401. Then again, this remains to be proven :). I believe the 1461 uses a 7955. I was able to get 500+ Mbps for symmetric crypto ops w/ a 64-bit pci hifn reference card that had a 7956 part on it (cryptotest -z). But you need largish block sizes. As Mike noted you won't see single-stream improvements so much as reduced load on the cpu that allows multiple streams to be handled better. Sam