From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965626AbXDBSCS (ORCPT ); Mon, 2 Apr 2007 14:02:18 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S965634AbXDBSCS (ORCPT ); Mon, 2 Apr 2007 14:02:18 -0400 Received: from smtpout.mac.com ([17.250.248.178]:59636 "EHLO smtpout.mac.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S965626AbXDBSCR (ORCPT ); Mon, 2 Apr 2007 14:02:17 -0400 In-Reply-To: <200704021354.32636.gene.heskett@gmail.com> References: <200704021030.29747.gene.heskett@gmail.com> <20070402173544.GA19343@deepthought> <200704021354.32636.gene.heskett@gmail.com> Mime-Version: 1.0 (Apple Message framework v752.2) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: <8EF5AE61-FC1C-4BA2-BB8D-B55221D8846B@mac.com> Cc: linux-kernel@vger.kernel.org Content-Transfer-Encoding: 7bit From: Kyle Moffett Subject: Re: mmap error? Date: Mon, 2 Apr 2007 14:02:11 -0400 To: Gene Heskett X-Mailer: Apple Mail (2.752.2) X-Brightmail-Tracker: AAAAAA== X-Brightmail-scanned: yes Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Apr 02, 2007, at 13:54:28, Gene Heskett wrote: > Ahh, thank you very much. But I have 7.0, supposedly fixed, > sigh... That did fix it, the dvd is burning as I type, again thanks. > > ???? Can I put this into rc.local so its global? I guess I will > just for effects. Or is there an option in a make xconfig to reset > this to a usable sized limit?, because I can see cups also having > problems since I have its render cache set for 127 megs, which > should also trigger this. Well, that's not a limit on how much memory you can map, it's a limit on how many pages you can lock into memory (as in, cannot be swapped out or written back to disk and reused). For low-latency applications (IE: cd burning) you need to have all memory mlocked so that you don't get a low-latency wakup and then promptly lose all latency guarantees by swapping madly, but something like CUPS doesn't have a problem because it doesn't use mlock() or mlockall(). Cheers, Kyle Moffett