Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1764406AbXESVwp (ORCPT ); Sat, 19 May 2007 17:52:45 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755621AbXESVwj (ORCPT ); Sat, 19 May 2007 17:52:39 -0400 Received: from 3a.49.1343.static.theplanet.com ([67.19.73.58]:34015 "EHLO pug.o-hand.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755556AbXESVwi (ORCPT ); Sat, 19 May 2007 17:52:38 -0400 Subject: Re: [RFC] LZO1X de/compression support From: Richard Purdie To: "Bill Rugolsky Jr." Cc: Krzysztof Halasa , Nitin Gupta , linux-kernel@vger.kernel.org In-Reply-To: <20070519185514.GA17869@ti88.telemetry-investments.com> References: <4cefeab80705180258g516a6f92w15a49e666dd62b66@mail.gmail.com> <20070519185514.GA17869@ti88.telemetry-investments.com> Content-Type: text/plain Date: Sat, 19 May 2007 22:52:12 +0100 Message-Id: <1179611532.5852.25.camel@localhost.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.6.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1462 Lines: 34 On Sat, 2007-05-19 at 14:55 -0400, Bill Rugolsky Jr. wrote: > On Fri, May 18, 2007 at 11:14:57PM +0200, Krzysztof Halasa wrote: > > I'm certainly missing something but what are the advantages of this > > code (over current gzip etc.), and what will be using it? > > Richard's patchset added it to the crypto library and wired it into > the JFFS2 file system. Basically, LZO has much faster decompression speeds. For jffs2, there was a 40% filesystem read speed improvement with LZO compared to zlib and that resulted in a device booting 10% faster. The drawback was a slight drop in file compression (say 5% although it varied a lot depending on the test data). For lots of uses, that is an acceptable compromise. It appears resier4 is using LZO too. > We recently started using LZO in a userland UDP > proxy to do stateless per-packet payload compression over a WAN link. > With ~1000 octet packets, our particular data stream sees 60% compression > with zlib, and 50% compression with (mini-)LZO, but LZO runs at ~5.6x > the speed of zlib. IIRC, that translates into > 700Mbps on the input > side on a 2GHZ Opteron, without any further tuning. Those figures sound comparable with my experiences... Richard - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/