From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail138.messagelabs.com (mail138.messagelabs.com [216.82.249.35]) by kanga.kvack.org (Postfix) with ESMTP id A9F1E6B0047 for ; Wed, 24 Feb 2010 16:13:25 -0500 (EST) Date: Wed, 24 Feb 2010 13:12:20 -0800 From: Andrew Morton Subject: Re: [patch 36/36] khugepaged Message-Id: <20100224131220.396216af.akpm@linux-foundation.org> In-Reply-To: <4B8592BB.1040007@redhat.com> References: <20100221141009.581909647@redhat.com> <20100221141758.658303189@redhat.com> <20100224121111.232602ba.akpm@linux-foundation.org> <4B858BFC.8020801@redhat.com> <20100224125253.2edb4571.akpm@linux-foundation.org> <4B8592BB.1040007@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org To: Rik van Riel Cc: aarcange@redhat.com, linux-mm@kvack.org, Marcelo Tosatti , Adam Litke , Avi Kivity , Izik Eidus , Hugh Dickins , Nick Piggin , Mel Gorman , Dave Hansen , Benjamin Herrenschmidt , Ingo Molnar , Mike Travis , KAMEZAWA Hiroyuki , Christoph Lameter , Chris Wright , bpicco@redhat.com, KOSAKI Motohiro , Balbir Singh , Arnd Bergmann List-ID: On Wed, 24 Feb 2010 15:57:31 -0500 Rik van Riel wrote: > On 02/24/2010 03:52 PM, Andrew Morton wrote: > > On Wed, 24 Feb 2010 15:28:44 -0500 Rik van Riel wrote: > > > >>> Generally it seems like a bad idea to do this sort of thing > >>> asynchronously. Because it reduces repeatability across runs and > >>> across machines - system behaviour becomes more dependent on the size > >>> of the machine and the amount of activity in unrelated jobs? > >> > >> Isn't system performance already dependent on the size of > >> the machine and the amount of activity in unrelated jobs? > > > > I said "repeatability". > > > >> Using hugepages is a performance enhancement only and > >> otherwise transparent to userspace. > > > > And it's bad that a job run will take a varying amount of CPU time due > > to unrelated activity. Yes, that can already happen, but it's > > undesirable and it's undesirable to worsen things. > > > > If this work could be done synchronously then runtimes become more > > consistent, which is a good thing. > > Only if it means run times become shorter... > That of course would be a problem to be traded off against the advantage. One would need to quantify these things to make that call. I asked a question and all I'm getting in reply is flippancy and unsubstantiated assertions. It may have been a bad question, but they're certainly bad answers :( -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org