From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Francois Barre" Subject: Re: Bigendian issue with mdadm Date: Tue, 21 Feb 2006 09:37:33 +0100 Message-ID: References: <17402.21590.564710.372935@cse.unsw.edu.au> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Return-path: In-Reply-To: <17402.21590.564710.372935@cse.unsw.edu.au> Content-Disposition: inline Sender: linux-raid-owner@vger.kernel.org To: linux-raid List-Id: linux-raid.ids 2006/2/21, Neil Brown : > On Monday February 20, francois.barre@gmail.com wrote: > > Hi All, > > > > Please, Help ! > > [...] > > And it seams like md superblock disk format is hostendian, so how > > should I say mdadm to use a endianness ? > > > > Read the man page several times? > > Look for "--update=byteorder" > > You need mdadm-2.0 or later. > > NeilBrown > Ooops... You know what it is : the more you focus on a problem, the less you can see solutions floating around... I had mdadm-1.2 installed, and of course did not think about upgrading... I will test this asap. This will avoid me to code a byte reordering program... Thanks Neil ! BTW : I did not see this feature in the kernel code, and after a quick look at mdadm source, seems to be implemented in super0.c:super0_swap_endian(). Is there any prerequisite regarding the kernel version for this to work ? Thanks again Neil, I am really impressed by your patience answering dummy questions... Regards,