Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261221AbVAWE5v (ORCPT ); Sat, 22 Jan 2005 23:57:51 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261224AbVAWE5v (ORCPT ); Sat, 22 Jan 2005 23:57:51 -0500 Received: from smtpout.mac.com ([17.250.248.47]:19402 "EHLO smtpout.mac.com") by vger.kernel.org with ESMTP id S261221AbVAWE5u (ORCPT ); Sat, 22 Jan 2005 23:57:50 -0500 In-Reply-To: References: <20050122203326.402087000@blunzn.suse.de> <20050122203618.962749000@blunzn.suse.de> Mime-Version: 1.0 (Apple Message framework v619) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: <5ADB1D78-6CFB-11D9-86B4-000D9352858E@mac.com> Content-Transfer-Encoding: 7bit Cc: Neil Brown , linux-kernel@vger.kernel.org, Buck Huppmann , Trond Myklebust , Andreas Gruenbacher , "Andries E. Brouwer" , Andrew Morton , Olaf Kirch From: Felipe Alfaro Solana Subject: Re: [patch 1/13] Qsort Date: Sun, 23 Jan 2005 05:58:00 +0100 To: Andi Kleen X-Mailer: Apple Mail (2.619) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 793 Lines: 20 On 23 Jan 2005, at 03:39, Andi Kleen wrote: > Felipe Alfaro Solana writes: >> >> AFAIK, XOR is quite expensive on IA32 when compared to simple MOV >> operatings. Also, since the original patch uses 3 MOVs to perform the >> swapping, and your version uses 3 XOR operations, I don't see any >> gains. > > Both are one cycle latency for register<->register on all x86 cores > I've looked at. What makes you think differently? I thought XOR was more expensie. Anyways, I still don't see any advantage in replacing 3 MOVs with 3 XORs. - 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/