From: Xiang Wang Subject: The flex_bg inode allocator Date: Fri, 17 Jul 2009 20:38:18 -0700 Message-ID: Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit To: ext4 development Return-path: Received: from smtp-out.google.com ([216.239.33.17]:58164 "EHLO smtp-out.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757489AbZGRDiX (ORCPT ); Fri, 17 Jul 2009 23:38:23 -0400 Received: from wpaz21.hot.corp.google.com (wpaz21.hot.corp.google.com [172.24.198.85]) by smtp-out.google.com with ESMTP id n6I3cLre022967 for ; Sat, 18 Jul 2009 04:38:22 +0100 Received: from ewy11 (ewy11.prod.google.com [10.241.103.11]) by wpaz21.hot.corp.google.com with ESMTP id n6I3cJtA020354 for ; Fri, 17 Jul 2009 20:38:19 -0700 Received: by ewy11 with SMTP id 11so1328846ewy.10 for ; Fri, 17 Jul 2009 20:38:19 -0700 (PDT) Sender: linux-ext4-owner@vger.kernel.org List-ID: Hi, Recently I've found out that the flex_bg inode allocator(the find_group_flex function called by ext4_new_inode) is actually not in use unless we specify the "oldalloc" option on mount as well as setting the flex_bg size to be > 1. Currently, the default option on mount is "orlov". We would like to know: 1) What's the current status of the flex_bg inode allocator? Will it be set as a default soon? 2) If not, are there any particular reasons that it is held back? Is it all because of the worse performance numbers shown in the two metrics ("read tree total" and "read compiled tree total") in Compilebench? 3) Are there any ongoing efforts and/or future plans to improve it? Or is there any work in similar directions? Thanks, Xiang