From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail172.messagelabs.com (mail172.messagelabs.com [216.82.254.3]) by kanga.kvack.org (Postfix) with ESMTP id 4F9616B00A2 for ; Sun, 10 May 2009 09:44:31 -0400 (EDT) Date: Sun, 10 May 2009 14:45:33 +0100 From: Alan Cox Subject: Re: [PATCH -mm] vmscan: make mapped executable pages the first class citizen Message-ID: <20090510144533.167010a9@lxorguk.ukuu.org.uk> In-Reply-To: <2f11576a0905100236u15d45f7fm32d470776659cfec@mail.gmail.com> References: <20090430181340.6f07421d.akpm@linux-foundation.org> <1241432635.7620.4732.camel@twins> <20090507121101.GB20934@localhost> <20090507151039.GA2413@cmpxchg.org> <20090507134410.0618b308.akpm@linux-foundation.org> <20090508081608.GA25117@localhost> <20090508125859.210a2a25.akpm@linux-foundation.org> <20090508230045.5346bd32@lxorguk.ukuu.org.uk> <2f11576a0905100159m32c36a9ep9fb7cc5604c60b2@mail.gmail.com> <1241946446.6317.42.camel@laptop> <2f11576a0905100236u15d45f7fm32d470776659cfec@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org To: KOSAKI Motohiro Cc: Peter Zijlstra , Andrew Morton , Wu Fengguang , hannes@cmpxchg.org, riel@redhat.com, linux-kernel@vger.kernel.org, tytso@mit.edu, linux-mm@kvack.org, elladan@eskimo.com, npiggin@suse.de, cl@linux-foundation.org, minchan.kim@gmail.com List-ID: On Sun, 10 May 2009 18:36:19 +0900 KOSAKI Motohiro wrote: > I don't oppose this policy. PROT_EXEC seems good viewpoint. I don't think it is that simple Not only can it be abused but some systems such as java have large PROT_EXEC mapped environments, as do many other JIT based languages. Secondly it moves the pressure from the storage volume holding the system binaries and libraries to the swap device which already has to deal with a lot of random (and thus expensive) I/O, as well as the users filestore for mapped objects there - which may even be on a USB thumbdrive. I still think the focus is on the wrong thing. We shouldn't be trying to micro-optimise page replacement guesswork - we should be macro-optimising the resulting I/O performance. My disks each do 50MBytes/second and even with the Gnome developers finest creations that ought to be enough if the rest of the system was working properly. -- 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