From: lsorense@csclub.uwaterloo.ca (Lennart Sorensen) Subject: Re: [GIT PULL] Ext3 latency fixes Date: Mon, 6 Apr 2009 17:50:34 -0400 Message-ID: <20090406215034.GD3797@csclub.uwaterloo.ca> References: <1238742067-30814-1-git-send-email-tytso@mit.edu> <20090404135719.GA9812@mit.edu> <20090404151649.GE5178@kernel.dk> <20090404173412.GF5178@kernel.dk> <20090404201820.GA18067@elte.hu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Jens Axboe , Linus Torvalds , Theodore Tso , Linux Kernel Developers List , Ext4 Developers List To: Ingo Molnar Return-path: Received: from caffeine.csclub.uwaterloo.ca ([129.97.134.17]:54020 "EHLO caffeine.csclub.uwaterloo.ca" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759781AbZDFVug (ORCPT ); Mon, 6 Apr 2009 17:50:36 -0400 Content-Disposition: inline In-Reply-To: <20090404201820.GA18067@elte.hu> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Sat, Apr 04, 2009 at 10:18:20PM +0200, Ingo Molnar wrote: > btw., just to insert some hard data: usability studies place the > acceptance threshold for delays to around 300 milliseconds. > > If the computer does not 'respond' (in a real or a fake, visible or > audible way) to their input within 0.3 seconds they get annoyed > emotionally. You mean like hitting 'skip commercial' on the remote, and 2 minutes later when the commercials are almost over, then it happens. Yeah users get a bit peeved at that. Fortunately that particularly behaviour is somewhat rare. 5 to 10 seconds is more common and sometimes it works fine. > It does not matter how complex it is for the kernel to solve this > problem, it does not matter how far away and difficult to access the > data is. If we are not absolutely latency centric in Linux, if we > spuriously delay apps that do supposedly simple-looking things the > user _will_ get annoyed and _will_ pick another OS. I would have to get _very_ annoyed before that happened. I would have to find a less annoying OS to switch to as well. > All things considered it is certainly a combined kernel and app > problem space to get there, but not being completely aware of its > importance in the kernel kills any chance of us ever having a > long-term solution. Certainly my problems are likely to a large extent mythtv's fault. -- Len Sorensen