From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from thermo.lanl.gov (thermo.lanl.gov [128.165.59.202]) by mailwasher-b.lanl.gov (8.12.11/8.12.11/(ccn-5)) with SMTP id jA4FJfoP024135 for ; Fri, 4 Nov 2005 08:19:41 -0700 Subject: Re: [Lhms-devel] [PATCH 0/7] Fragmentation Avoidance V19 Message-Id: <20051104151937.C88561845E1@thermo.lanl.gov> Date: Fri, 4 Nov 2005 08:19:37 -0700 (MST) From: andy@thermo.lanl.gov (Andy Nelson) Sender: owner-linux-mm@kvack.org Return-Path: To: torvalds@osdl.org Cc: akpm@osdl.org, arjan@infradead.org, arjanv@infradead.org, haveblue@us.ibm.com, kravetz@us.ibm.com, lhms-devel@lists.sourceforge.net, linux-kernel@vger.kernel.org, linux-mm@kvack.org, mbligh@mbligh.org, mel@csn.ul.ie, mingo@elte.hu, nickpiggin@yahoo.com.au List-ID: Nick Piggin wrote: >Mel Gorman wrote: >> On Fri, 4 Nov 2005, Nick Piggin wrote: >> >> Todays massive machiens are tomorrows desktop. Weak comment, I know, but >> it's happened before. >> >Oh I wouldn't bet against it. And if desktops of the future are using >100s of GB then they probably would be happy to use 64K pages as well. Just a note. The data I referenced in my other post that can be found on comp.arch uses 64k pages as the smallest page size in the study. Pages sized 1M and 16M were the other two. As I understand it, only a few arch's have hw support for more than 2 page sizes, but my response is that they will eventually need them. The larger the memory, the larger the possible page size needs to be too. Otherwise you are just pushing out the problem for a few years. Andy -- 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