From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Thu, 15 Feb 2007 19:50:45 -0800 (PST) From: Christoph Lameter Subject: Re: [RFC] Remove unswappable anonymous pages off the LRU In-Reply-To: <20070215194258.a354f428.akpm@linux-foundation.org> Message-ID: References: <20070215171355.67c7e8b4.akpm@linux-foundation.org> <45D50B79.5080002@mbligh.org> <20070215174957.f1fb8711.akpm@linux-foundation.org> <20070215184800.e2820947.akpm@linux-foundation.org> <20070215191858.1a864874.akpm@linux-foundation.org> <20070215194258.a354f428.akpm@linux-foundation.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-linux-mm@kvack.org Return-Path: To: Andrew Morton Cc: Martin Bligh , linux-mm@kvack.org, Nick Piggin , Peter Zijlstra , KAMEZAWA Hiroyuki , Rik van Riel List-ID: On Thu, 15 Feb 2007, Andrew Morton wrote: > > Maybe we could somehow splite up page->flags into 4 separate bytes? > > Updating one byte would not endanger the other bytes in the other > > sets? > > yipes. I'm not sure that'd work? Are all arches able to do atomic ops on bytes? > compare-and-swap-in-a-loop could be used, I guess. With the obvious problem.. Yucks. There seems to be no easy solution. > I do think that those two swsusp flags are low-hanging-fruit. It'd be > trivial to vmalloc a bitmap or use a radix-tree-holding-longs, but I have a > vague feeling that there were subtle issues with that. Still, Something > Needs To Be Done. I tinkered with some similar radical ideas lately. Maybe a bit vector could be used instead? For 1G of memory we would need 2^(30 - PAGE_SHIFT / 8 = 2^(30-12-3) = 2^15 = 32k bytes of a bitmap. Seems to be reasonable? -- 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