From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752845AbZKMGE1 (ORCPT ); Fri, 13 Nov 2009 01:04:27 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751754AbZKMGEW (ORCPT ); Fri, 13 Nov 2009 01:04:22 -0500 Received: from terminus.zytor.com ([198.137.202.10]:32928 "EHLO terminus.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751073AbZKMGEV (ORCPT ); Fri, 13 Nov 2009 01:04:21 -0500 Message-ID: <4AFCF6D6.6040607@zytor.com> Date: Thu, 12 Nov 2009 22:04:06 -0800 From: "H. Peter Anvin" User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.1) Gecko/20090814 Fedora/3.0-2.6.b3.fc11 Thunderbird/3.0b3 MIME-Version: 1.0 To: "Ma, Ling" CC: Pavel Machek , "mingo@elte.hu" , "tglx@linutronix.de" , "linux-kernel@vger.kernel.org" Subject: Re: [PATCH RFC] [X86] performance improvement for memcpy_64.S by fast string. References: <1257500482-16182-1-git-send-email-ling.ma@intel.com> <4AF457E0.4040107@zytor.com> <4AF4784C.5090800@zytor.com> <8FED46E8A9CA574792FC7AACAC38FE7714FCF772C9@PDSMSX501.ccr.corp.intel.com> <20091112121604.GC1394@ucw.cz> <8FED46E8A9CA574792FC7AACAC38FE7714FEB070EE@PDSMSX501.ccr.corp.intel.com> In-Reply-To: <8FED46E8A9CA574792FC7AACAC38FE7714FEB070EE@PDSMSX501.ccr.corp.intel.com> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 11/12/2009 09:33 PM, Ma, Ling wrote: >> Well, so you are running cache hot and it is only a win on huge >> copies... how common are those? >> > Hi Pavel Machek > Yes, we intend to introduce movsq for huge hot size(over 1024bytes) > and avoid regression for less 1024bytes. I guess you suggest using > prefetch instruction for cold data (if I was wrong please correct me). > memcpy don't know whether data has been in cache or not, > so only when copy size is over (first level 1 cache)/2 and lower > (last level cache)/2 , prefetch will get benefit. Currently first > level cache size of most cpus is around 32KB, so it is useful for prefetch > when copy size is over 16KB, but as H. Peter Anvin mentioned in last email, > over 16KB copy in kernel is rare. > What it sounds to me is that for Nehalem, we want to use memcpy_c for >= 1024 bytes and the old code for < 1024 bytes; for Core2 it might be the exact opposite. Either way, whatever we do should use the appropriate static replacement mechanism. -hpa -- H. Peter Anvin, Intel Open Source Technology Center I work for Intel. I don't speak on their behalf.