From: Andreas Dilger Subject: Re: [PATCH] block: Add Kconfig help which notes that ext4 needs CONFIG_LSF Date: Thu, 08 Jan 2009 04:39:15 -0700 Message-ID: <20090108113915.GL13721@webber.adilger.int> References: <1231262490-15594-1-git-send-email-tytso@mit.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7BIT Cc: Ext4 Developers List , Jens Axboe To: "Theodore Ts'o" Return-path: Received: from sca-es-mail-1.Sun.COM ([192.18.43.132]:63542 "EHLO sca-es-mail-1.sun.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752418AbZAHLjS (ORCPT ); Thu, 8 Jan 2009 06:39:18 -0500 Received: from fe-sfbay-09.sun.com ([192.18.43.129]) by sca-es-mail-1.sun.com (8.13.7+Sun/8.12.9) with ESMTP id n08BdHQI000208 for ; Thu, 8 Jan 2009 03:39:17 -0800 (PST) Received: from conversion-daemon.fe-sfbay-09.sun.com by fe-sfbay-09.sun.com (Sun Java System Messaging Server 6.2-8.04 (built Feb 28 2007)) id <0KD500H01HKQJE00@fe-sfbay-09.sun.com> (original mail from adilger@sun.com) for linux-ext4@vger.kernel.org; Thu, 08 Jan 2009 03:39:17 -0800 (PST) In-reply-to: <1231262490-15594-1-git-send-email-tytso@mit.edu> Content-disposition: inline Sender: linux-ext4-owner@vger.kernel.org List-ID: On Jan 06, 2009 12:21 -0500, Theodore Ts'o wrote: > @@ -64,6 +64,12 @@ config LSF > + The ext4 filesystem requires that this feature be enabled in > + order to support filesystems that have the huge_file feature > + enabled. Otherwise, it will refuse to mount any filesystems > + that use the huge_file feature, which is enabled by default > + by mke2fs.ext4. Wouldn't it make sense to just enable this by default if ext4 is enabled? On a related note, I wonder if we need to re-examine the use of the COMPAT flags in ext4. We went from a usage of "kernel enables flag when feature is used" (e.g. LARGE_FILE) to "admin must set flag in order to use feature" (e.g HUGE_FILE). The former system was recording the state of currently-used features in the filesystem, and maximized portability of the filesystem if the feature was NOT used. The newer system makes a filesystem somewhat less portable even if the feature is not being used at all (e.g. files > 2TB in size). I wonder if we should split the s_feature_*compat fields in the superblock to "present feature" and "allowed feature" so that e.g. HUGE_FILE count be put into "allowed feature" and would only appear in "present feature" if a file > 2TB in size arrives. That ensures the filesystem remains maximally compatible, while allowing the admin to select which features are permissible. Cheers, Andreas -- Andreas Dilger Sr. Staff Engineer, Lustre Group Sun Microsystems of Canada, Inc.