From mboxrd@z Thu Jan 1 00:00:00 1970 From: Clay Haapala Subject: Re: Request for review of Linux iSCSI driver version 4.0.0.1 Date: Fri, 21 Nov 2003 12:18:17 -0600 Sender: linux-scsi-owner@vger.kernel.org Message-ID: References: <200310231734.10263.krmurthy@cisco.com> <20031027153932.A16679@infradead.org> <20031121173236.GX30485@parcelfarce.linux.theplanet.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from sj-iport-5.cisco.com ([171.68.10.87]:13404 "EHLO sj-iport-5.cisco.com") by vger.kernel.org with ESMTP id S264405AbTKUSSW (ORCPT ); Fri, 21 Nov 2003 13:18:22 -0500 In-Reply-To: <20031121173236.GX30485@parcelfarce.linux.theplanet.co.uk> (Matthew Wilcox's message of "Fri, 21 Nov 2003 17:32:36 +0000") List-Id: linux-scsi@vger.kernel.org To: Matthew Wilcox Cc: Christoph Hellwig , Krishna Murthy , linux-scsi@vger.kernel.org, davmyers@cisco.com On Fri, 21 Nov 2003, Matthew Wilcox spake thusly: > On Fri, Nov 21, 2003 at 10:42:40AM -0600, Clay Haapala wrote: >> On Mon, 27 Oct 2003, Christoph Hellwig said: >> > >> > highlevel: >> > - kill md5.c and iscsi-crc.c and use common cryptoapi >> > md5 / common crc code instead. >> >> md5: That appears straight-forward, and has been implemented, >> subject to testing. SELECT CRYPTO and SELECT MD5 makes it >> available in the kernel. >> >> crc32: Ah, the iSCSI crc32 uses a different polynomial (stronger) >> than Ethernet and the rest. We argued during the RFC process for >> just using Ethernet's as good enough, but were voted down. So we >> are stuck with a separate implementation in this case. > > OK, but please convert it to live in the crypto/ directory and use > the crypto interfaces. That way if some unrelated project decides > to implement the same algorithm, we don't end up with two copies of > it. I'm happy to do that. Having established that it is not a redundant implementation, though, I would like to have it not be a factor with respect to driver inclusion, as there are bigger fish to fry. That being said, I'll probably jump on it, anyway. Question: Should the implementation use the crypto api or be added to the lib/crc32 code? -- Clay Haapala (chaapala@cisco.com) Cisco Systems SRBU +1 763-398-1056 6450 Wedgwood Rd, Suite 130 Maple Grove MN 55311 PGP: C89240AD Well, looks like hypocrisy is back on the airwaves. C'mon, Rush! Do the crime, do the time, right?