From: Justin Maggard Subject: Re: [PATCH] Fix EXT4_FEATURE_RO_COMPAT_HUGE_FILE check Date: Tue, 24 Aug 2010 12:32:12 -0700 Message-ID: References: <4c7325f6.02558f0a.5a24.ffffc694@mx.google.com> <426FFD16-98B3-4427-BDA5-C80CE184781C@dilger.ca> <4C73F627.2000809@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Cc: Andreas Dilger , linux-ext4@vger.kernel.org To: Eric Sandeen Return-path: Received: from mail-vw0-f46.google.com ([209.85.212.46]:33289 "EHLO mail-vw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751710Ab0HXTcN (ORCPT ); Tue, 24 Aug 2010 15:32:13 -0400 Received: by vws3 with SMTP id 3so791540vws.19 for ; Tue, 24 Aug 2010 12:32:13 -0700 (PDT) In-Reply-To: <4C73F627.2000809@redhat.com> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Tue, Aug 24, 2010 at 9:41 AM, Eric Sandeen wrote: > Andreas Dilger wrote: > >> This isn't the first time that we've had problems like this, and it >> won't be the last time unless there is a better API for checking >> these flags. > > *nod* > On a related note (to my original problem)... is the bad i_blocks count fixup code in the check_blocks() function of pass1.c broken for large files, or am I misunderstanding the code? pb.num_blocks and ext2fs_inode_i_blocks() are both >32 bits, but the fixup code appears to always try to shove pb.num_blocks into the 32-bit inode->i_blocks and set inode->osd2.linux2.l_i_blocks_hi = 0. -Justin