From owner-freebsd-security@FreeBSD.ORG Fri Apr 16 15:50:42 2004 Return-Path: Delivered-To: freebsd-security@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 5CCEB16A4CE for ; Fri, 16 Apr 2004 15:50:42 -0700 (PDT) Received: from isber.ucsb.edu (research.isber.ucsb.edu [128.111.147.5]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2D61B43D5C for ; Fri, 16 Apr 2004 15:50:42 -0700 (PDT) (envelope-from randall@ucsb.edu) Received: from ip68-6-116-39.sb.sd.cox.net ([68.6.116.39] helo=home) by isber.ucsb.edu with asmtp (TLSv1:RC4-MD5:128) (Exim 3.36 #2) id 1BEcAL-000Csz-00; Fri, 16 Apr 2004 15:50:33 -0700 Message-ID: <02cf01c42405$39a33450$4102a8c0@home> From: "randall ehren" To: "Jason Stone" , References: <20040408144322.GA83448@bewilderbeast.blackhelicopters.org><20040413181943.GA55219@bewilderbeast.blackhelicopters.org><6.0.3.0.0.20040414230754.07d7cf18@209.112.4.2><6.0.3.0.0.20040415105459.0477f488@209.112.4.2><20040415180518.GA46433@phobos.osem.com> <20040416153835.K45935@walter> Date: Fri, 16 Apr 2004 15:50:32 -0700 Organization: ISBER MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1409 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 X-Scanner: exiscan *1BEcAL-000Csz-00*LgtJ/sr2h6I* (ISBER - Institute for Social, Behavioral, and Economic Research) X-Mailman-Approved-At: Mon, 19 Apr 2004 02:31:39 -0700 Subject: Re: recommended SSL-friendly crypto accelerator X-BeenThere: freebsd-security@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Security issues [members-only posting] List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 16 Apr 2004 22:50:42 -0000 | > For $79, it's cheap enough that I could put a whole stack of them in a | > machine. Can FreeBSD take advantage of multiple cards like that? | | another question is, is there logic, either in the driver or in openssl, | to notice if crypto operations are getting backed up waiting for the | crypto card while the main cpu is idle and, in that case, to start doing | the crypto on the main cpu rather than on the crypto card? | | in other words, if the main cpu is actually way faster than the crypto | card, is it possible that the crypto card could actually _slow_ crypto | operations on that system? | | last time I checked, the stats on the cheap soekris cards were way slower | than the output of "openssl speed" run on my system during normal load.... how would one test this out: just run % openssl speed on a system with the card, and then again without the card? where are you referencing your stats from? -randall