From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from waste.org ([66.93.16.53]:38346 "EHLO waste.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932293AbXHVXss (ORCPT ); Wed, 22 Aug 2007 19:48:48 -0400 Date: Wed, 22 Aug 2007 18:50:11 -0500 From: Matt Mackall Subject: Re: [RFC][PATCH] introduce TASK_SIZE_OF() for all arches Message-ID: <20070822235011.GP30556@waste.org> References: <20070821193835.258D7DD4@kernel> <31048.1187773615@redhat.com> <1187823350.16177.1326.camel@localhost> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1187823350.16177.1326.camel@localhost> Sender: linux-arch-owner@vger.kernel.org To: Dave Hansen Cc: David Howells , linux-arch@vger.kernel.org List-ID: On Wed, Aug 22, 2007 at 03:55:50PM -0700, Dave Hansen wrote: > On Wed, 2007-08-22 at 10:06 +0100, David Howells wrote: > > Dave Hansen wrote: > > > > > For the /proc//pagemap code[1], we need to able to query how > > > much virtual address space a particular task has. > > > > How does this relate to NOMMU conditions? > > It probably doesn't make any sense to use the /proc//pagemap > feature on NOMMU systems. There are no pagetables to dump, and no > shared pages, right? Nor is there any present/not present to report. VSS == RSS on NOMMU, right? So we could provide pagemap, but it's not clear it'd be very interesting. kpagemap might be interesting, simply because of the flags though. -- Mathematics is the supreme nostalgia of our time.