Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755576Ab2HPRXR (ORCPT ); Thu, 16 Aug 2012 13:23:17 -0400 Received: from bar.sig21.net ([80.81.252.164]:42690 "EHLO bar.sig21.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753137Ab2HPRXP (ORCPT ); Thu, 16 Aug 2012 13:23:15 -0400 Date: Thu, 16 Aug 2012 19:22:51 +0200 From: Johannes Stezenbach To: Jeff Garzik Cc: Andi Kleen , "Markus F.X.J. Oberhumer" , linux-kernel@vger.kernel.org, chris.mason@fusionio.com, linux-btrfs@vger.kernel.org, Nitin Gupta , Richard Purdie , richard -rw- weinberger , linux-arm-kernel@lists.infradead.org Subject: Re: [GIT PULL] Update LZO compression Message-ID: <20120816172251.GA25640@sig21.net> References: <50299142.2030504@oberhumer.com> <20120814123937.GA14756@sig21.net> <502B8FE3.7080501@oberhumer.com> <20120815144539.GA8300@sig21.net> <502C92C3.7090701@oberhumer.com> <502D100D.2030609@pobox.com> <20120816162051.GK11413@one.firstfloor.org> <502D2471.407@pobox.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <502D2471.407@pobox.com> User-Agent: Mutt/1.5.21 (2010-09-15) X-Spam-21-Score: -2.9 (--) X-Spam-21-Report: No, score=-2.9 required=8.0 tests=ALL_TRUSTED=-1,BAYES_00=-1.9 autolearn=ham Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1263 Lines: 29 On Thu, Aug 16, 2012 at 12:48:49PM -0400, Jeff Garzik wrote: > On 08/16/2012 12:20 PM, Andi Kleen wrote: > >>If you think a little bit, I bet you could come up with a solution that > >>operates at cacheline-aligned granularity, something that would be _even > >>faster_ than simply fixing the code to do aligned accesses. > > > >Cache aligned compression is unlikely to compress anything at all. > >Compression algorithms are usually by definition unaligned. > > Sure it's a bitstream, but that does not imply the impossibility of > reading data in in an word-aligned manner. > > Maybe cache-aligned is ambitious, because of resultant code bloat, > but machine-int-aligned is doable and reasonable. Well, I for one would be content if the old and new lzo versions could be merged based on CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS. (Assuming that the slowdown on ARM is due to unaligned access, since the old version also uses get/put_unaligned, is the new version actually using more unaligned accesses?) Johannes -- 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/