Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756869Ab1FUUCB (ORCPT ); Tue, 21 Jun 2011 16:02:01 -0400 Received: from mx1.redhat.com ([209.132.183.28]:15009 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753946Ab1FUUB7 (ORCPT ); Tue, 21 Jun 2011 16:01:59 -0400 Message-ID: <4E00F88F.2080603@redhat.com> Date: Tue, 21 Jun 2011 16:01:19 -0400 From: Rik van Riel User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.17) Gecko/20110428 Fedora/3.1.10-1.fc15 Lightning/1.0b3pre Thunderbird/3.1.10 MIME-Version: 1.0 To: Cong Wang CC: Andrea Arcangeli , Vivek Goyal , Mel Gorman , linux-kernel@vger.kernel.org, akpm@linux-foundation.org, Johannes Weiner , KAMEZAWA Hiroyuki , linux-mm@kvack.org Subject: Re: [PATCH 1/3] mm: completely disable THP by transparent_hugepage=never References: <1308587683-2555-1-git-send-email-amwang@redhat.com> <20110620165844.GA9396@suse.de> <4DFF7E3B.1040404@redhat.com> <4DFF7F0A.8090604@redhat.com> <4DFF8106.8090702@redhat.com> <4DFF8327.1090203@redhat.com> <4DFF84BB.3050209@redhat.com> <4DFF8848.2060802@redhat.com> <20110620182558.GF4749@redhat.com> <20110620192117.GG20843@redhat.com> <4E00192E.70901@redhat.com> In-Reply-To: <4E00192E.70901@redhat.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1055 Lines: 26 On 06/21/2011 12:08 AM, Cong Wang wrote: > The thing is that we can save ~10K by adding 3 lines of code as this > patch showed, where else in kernel can you save 10K by 3 lines of code? > (except some kfree() cases, of course) So, again, why not have it? ;) Because we'll end up with hundreds of lines of code, just to save under 1MB of memory. Which ends up not being saved at all, because people will still give their kdump kernel 128MB :) The only really big gain you are likely to get is making sure all the per-cpu memory is not allocated in the kdump kernel (which is booted with 1 cpu). That is a big, multi-MB, optimization that can be implemented in one place. Large savings for a localized change, so you actually have a chance of having the changes accepted upstream. -- All rights reversed -- 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/