From: Theodore Ts'o Subject: Re: xfstest-bld generic/018 fails due to e4defrag issue Date: Thu, 10 Apr 2014 09:56:37 -0400 Message-ID: <20140410135637.GC15925@thunk.org> References: <20140409220320.GC15303@thunk.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: "linux-ext4@vger.kernel.org List" To: jon ernst Return-path: Received: from imap.thunk.org ([74.207.234.97]:52517 "EHLO imap.thunk.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S964868AbaDJN4l (ORCPT ); Thu, 10 Apr 2014 09:56:41 -0400 Content-Disposition: inline In-Reply-To: Sender: linux-ext4-owner@vger.kernel.org List-ID: On Thu, Apr 10, 2014 at 12:13:49AM -0400, jon ernst wrote: > > Because bigalloc requires cluster-aware bitfield operations, which > means we need EXT2_FLAG_64BITS. > I see e2image.c creates image always with EXT2_FLAG_64BITS flag. It is > safe to do same thing for e4defrag in my opinion. Please correct me if > I am wrong. Um.... I *think* so. e4defrag is one of the less well tested/maintained parts of e2fsprogs, as well as the kernel-side code which supports e4defrag. I can't think of any reason why there would be any 32-bit dependencies in the kernel side code, although someone should probably do a quick audit of the e4defrag code to make sure it's not using blk_t where it should be using blk64_t, or have other 32-bit dependencies. - Ted