From: Nick Piggin Subject: Re: [patch 12/17] vfs: pagecache usage optimization for pagesize!=blocksize Date: Tue, 29 Jul 2008 11:11:42 +1000 Message-ID: <200807291111.42941.nickpiggin@yahoo.com.au> References: <200807282246.m6SMkaHT032267@imap1.linux-foundation.org> <20080728230031.GA22218@infradead.org> <20080728160904.f644fb85.akpm@linux-foundation.org> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Cc: Christoph Hellwig , torvalds@linux-foundation.org, hifumi.hisashi@oss.ntt.co.jp, jack@ucw.cz, linux-ext4@vger.kernel.org To: Andrew Morton Return-path: Received: from smtp112.mail.mud.yahoo.com ([209.191.84.65]:27129 "HELO smtp112.mail.mud.yahoo.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1750819AbYG2BLz (ORCPT ); Mon, 28 Jul 2008 21:11:55 -0400 In-Reply-To: <20080728160904.f644fb85.akpm@linux-foundation.org> Content-Disposition: inline Sender: linux-ext4-owner@vger.kernel.org List-ID: On Tuesday 29 July 2008 09:09, Andrew Morton wrote: > On Mon, 28 Jul 2008 19:00:32 -0400 > > Christoph Hellwig wrote: > > On Mon, Jul 28, 2008 at 03:46:36PM -0700, akpm@linux-foundation.org wrote: > > > From: Hisashi Hifumi > > > > > > When we read some part of a file through pagecache, if there is a > > > pagecache of corresponding index but this page is not uptodate, read IO > > > is issued and this page will be uptodate. > > > > I was under the impression we wanted to do this in a nicer way than > > the hacky method? > > The only description I've seen of "a nicer way" is vague two-word > descriptions of "changing readpage". Or something. No indication of > what those changes are, nor who will implement them nor when. > > That just isn't solid enough to block a change which has significant > performance benefits. Yes I thought it would be nicer to do it by changing readpage, but I said I'm happy to try to consolidate them myself down the track. It is fairly low impact on core code.