From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751729Ab0JCXzf (ORCPT ); Sun, 3 Oct 2010 19:55:35 -0400 Received: from lo.gmane.org ([80.91.229.12]:54618 "EHLO lo.gmane.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751173Ab0JCXze (ORCPT ); Sun, 3 Oct 2010 19:55:34 -0400 X-Injected-Via-Gmane: http://gmane.org/ To: linux-kernel@vger.kernel.org From: walt Subject: Very bad latency on x86 desktop when a process is swapping: Am I expecting too much? Date: Sun, 03 Oct 2010 16:55:17 -0700 Message-ID: Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Complaints-To: usenet@dough.gmane.org X-Gmane-NNTP-Posting-Host: adsl-69-234-214-85.dsl.irvnca.pacbell.net User-Agent: Mozilla/5.0 (X11; Linux i686; rv:2.0b7pre) Gecko/20101001 Thunderbird/3.3a1pre Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi lkml, This is most likely a corner case for the average linux desktop user, but it's very annoying for me and I'd like to get a feel for what I 'should' be expecting. I track the mozilla-firefox-trunk sources from their hg repository every day, and there is one place in the build where roughly 20-30 large libraries are linked, all in one giant gcc command. All those libraries won't fit in my puny 1GB of RAM, so the machine starts to swap, naturally. During the 2-3 minutes that all those libraries are being linked, the machine is truly unusable for any normal lightweight desktop task like reading email. The swap usage tops out at about 15% of available swap space. I know the problem would go away if install more RAM, but that's not my real question. What should I realistically expect from a desktop workstation under these fairly uncommon conditions? Many thanks.