From: Rik van Riel Subject: Re: How Inactive may be much greather than cached? Date: Thu, 18 Oct 2007 10:17:15 -0400 Message-ID: <20071018101715.13c92b6e@bree.surriel.com> References: <4716FC04.6070107@sw.ru> <200710181627.27827.nickpiggin@yahoo.com.au> <471707BD.4000203@sw.ru> <200710181727.01128.nickpiggin@yahoo.com.au> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: Vasily Averin , Linux Kernel Mailing List , linux-ext4@vger.kernel.org, devel@openvz.org To: Nick Piggin Return-path: Received: from mx1.redhat.com ([66.187.233.31]:36852 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758194AbXJROSH (ORCPT ); Thu, 18 Oct 2007 10:18:07 -0400 In-Reply-To: <200710181727.01128.nickpiggin@yahoo.com.au> Sender: linux-ext4-owner@vger.kernel.org List-Id: linux-ext4.vger.kernel.org On Thu, 18 Oct 2007 17:27:00 +1000 Nick Piggin wrote: > Some filesystems, including I believe, ext3 with data=ordered, > can leave orphaned pages around after they have been truncated > out of the pagecache. These pages get left on the LRU and vmscan > reclaims them pretty easily. How can the VM recognize those pages? Are they part of the buffer cache, part of the page cache, or different? I think it would make sense to at least try to rotate those pages to the end of the LRU so kswapd can get rid of them quickly. -- "Debugging is twice as hard as writing the code in the first place. Therefore, if you write the code as cleverly as possible, you are, by definition, not smart enough to debug it." - Brian W. Kernighan