From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932203AbWFMRLL (ORCPT ); Tue, 13 Jun 2006 13:11:11 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932149AbWFMRLL (ORCPT ); Tue, 13 Jun 2006 13:11:11 -0400 Received: from smtp-out.google.com ([216.239.45.12]:48146 "EHLO smtp-out.google.com") by vger.kernel.org with ESMTP id S932203AbWFMRLJ (ORCPT ); Tue, 13 Jun 2006 13:11:09 -0400 DomainKey-Signature: a=rsa-sha1; s=beta; d=google.com; c=nofws; q=dns; h=received:subject:from:reply-to:to:cc:in-reply-to:references: content-type:organization:date:message-id:mime-version:x-mailer:content-transfer-encoding; b=ID6lOYhgiZh5SyWrEbKfhrRLhqAB94tLDem/P5Lu0YGK6OYk24duTbQsoZHZJAscg glgXlbn/d7qX/ZnR9LQyw== Subject: Re: [PATCH]: Adding a counter in vma to indicate the number of physical_pages_backing it From: Rohit Seth Reply-To: rohitseth@google.com To: Andi Kleen Cc: Albert Cahalan , linux-kernel@vger.kernel.org, akpm@osdl.org, Linux-mm@kvack.org, arjan@infradead.org, jengelh@linux01.gwdg.de In-Reply-To: <200606130756.52669.ak@suse.de> References: <787b0d920606122253o4f1a9e18x1ca49c3ce005696f@mail.gmail.com> <200606130756.52669.ak@suse.de> Content-Type: text/plain Organization: Google Inc Date: Tue, 13 Jun 2006 10:10:36 -0700 Message-Id: <1150218637.9576.73.camel@galaxy.corp.google.com> Mime-Version: 1.0 X-Mailer: Evolution 2.2.1.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 2006-06-13 at 07:56 +0200, Andi Kleen wrote: > On Tuesday 13 June 2006 07:53, Albert Cahalan wrote: > > Quoting two different people: > > > > > BTW, what is smaps used for (who uses it), anyway? > > ... > > > smaps is only a debugging kludge anyways and it's > > > not a good idea to we bloat core data structures for it. > > > > I'd be using it in procps for the pmap command if it > > were not so horribly nasty. I may eventually get around > > to using it, but maybe it's just too gross to tolerate. > > I agree it's pretty ugly. > > But pmap I would consider a debugging kludge too - it should > work when someone needs it, but it doesn't need to be particularly > fast. > Providing useful information about memory consumption is hardly debugging kludge. Quite unfortunately the rss part of the col is filled with dashes at present. This vma based counter will allow Albert to print useful information w/o having to traverse the whole set of page tables. -rohit