From mboxrd@z Thu Jan 1 00:00:00 1970 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <16781.12572.181444.967905@wombat.chubb.wattle.id.au> Date: Sun, 7 Nov 2004 07:16:28 +1100 From: Peter Chubb Subject: Re: removing mm->rss and mm->anon_rss from kernel? In-Reply-To: References: <4189EC67.40601@yahoo.com.au> <418AD329.3000609@yahoo.com.au> <418AE0F0.5050908@yahoo.com.au> <418AE9BB.1000602@yahoo.com.au> <1099622957.29587.101.camel@gaston> <418C55A7.9030100@yahoo.com.au> Sender: owner-linux-mm@kvack.org Return-Path: To: Christoph Lameter Cc: Nick Piggin , Benjamin Herrenschmidt , Hugh Dickins , linux-mm@kvack.org, linux-ia64@kernel.vger.org List-ID: >>>>> "Christoph" == Christoph Lameter writes: Christoph> So I removed all uses of mm->rss and anon_rss from the Christoph> kernel and introduced a bean counter count_vm() that is Christoph> only run when the corresponding /proc file is Christoph> used. count_vm then runs throught the vm and counts all the Christoph> page types. This could also add additional page types to Christoph> our statistics and solve some of the consistency issues. Is this going to scale properly to large machines, which usually have large numbers of active processes? top is already almost unuseably slow on such machines; if all the pagetables have to be scanned to get RSS, it'll probably slow to a halt. -- 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: aart@kvack.org