From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail138.messagelabs.com (mail138.messagelabs.com [216.82.249.35]) by kanga.kvack.org (Postfix) with SMTP id E05668D0030 for ; Fri, 29 Oct 2010 11:34:01 -0400 (EDT) Received: by ywl5 with SMTP id 5so1612508ywl.14 for ; Fri, 29 Oct 2010 08:33:57 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <20101029145212.GA21205@thunk.org> References: <20101028090002.GA12446@elte.hu> <20101028133036.GA30565@elte.hu> <20101028170132.GY27796@think> <20101029145212.GA21205@thunk.org> Date: Fri, 29 Oct 2010 21:33:56 +0600 Message-ID: Subject: Re: 2.6.36 io bring the system to its knees From: Aidar Kultayev Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Sender: owner-linux-mm@kvack.org To: Ted Ts'o , Pekka Enberg , Chris Mason , Ingo Molnar , Aidar Kultayev , linux-kernel@vger.kernel.org, linux-mm@kvack.org, Linus Torvalds , Andrew Morton , Jens Axboe , Peter Zijlstra , Nick Piggin , Arjan van de Ven , Thomas Gleixner List-ID: puling the git now - I will try whatever you throw at me. On Fri, Oct 29, 2010 at 8:52 PM, Ted Ts'o wrote: > On Thu, Oct 28, 2010 at 08:57:49PM +0300, Pekka Enberg wrote: >> Don't we need to call ext4_should_writeback_data() before we drop the >> lock? It pokes at ->i_mode which needs ->i_mutex AFAICT. > > No, it should be fine. =A0It's not like a file is going to change from > being a regular file to a directory or vice versa. =A0:-) > > From a quick inspection it looks OK, but I haven't had the time to > look more closely to be 100% sure, and of course I haven't run it > through a battery of regression tests. =A0For normal usage it should be > fine though. > > Aidar, if you'd be willing to try it with this patch applied, and with > the file system mounted data=3Dwriteback, and then let me know what the > latencytop reports, that would be useful. =A0I'm fairly sure that fixing > llseek() probably won't make that much difference, since it will > probably spread things out to other places, but it would be good to > make the experiment. > > We will probably also need to use the uninitialized bit for protecting > data from showing up after a crash for extent-based files, and turning > on data=3Dwriteback is a good way to simulate that. =A0(Sorry, no way > we're going to make a change like that this merge cycle, but that > might be something we could do for 2.6.38.) =A0But I am curious to see > what are the next things that come up as being problematic after that. > > Thanks, > > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0- Ted > -- 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