From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: with ECARTIS (v1.0.0; list xfs); Wed, 19 Mar 2008 05:56:34 -0700 (PDT) Received: from cuda.sgi.com (cuda2.sgi.com [192.48.168.29]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m2JCu1Nj009737 for ; Wed, 19 Mar 2008 05:56:02 -0700 Received: from sandeen.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id D98246BCDD5 for ; Wed, 19 Mar 2008 05:53:16 -0700 (PDT) Received: from sandeen.net (sandeen.net [209.173.210.139]) by cuda.sgi.com with ESMTP id Fmc4wrz4U8I8OQpZ for ; Wed, 19 Mar 2008 05:53:16 -0700 (PDT) Message-ID: <47E10CBA.9020904@sandeen.net> Date: Wed, 19 Mar 2008 07:53:14 -0500 From: Eric Sandeen MIME-Version: 1.0 Subject: Re: [PATCH] fix dir2 shortform structures on ARM old ABI References: <20080319002307.GA11349@josefsipek.net> <47E0F746.1090508@ventoso.org> In-Reply-To: <47E0F746.1090508@ventoso.org> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: xfs-bounce@oss.sgi.com Errors-to: xfs-bounce@oss.sgi.com List-Id: xfs To: Luca Olivetti Cc: xfs@oss.sgi.com Luca Olivetti wrote: > En/na Josef 'Jeff' Sipek ha escrit: > > >> Thanks for the heads up wrt ARM EABI. > > Well, I have a linkstation lspro, even if it's running now "freelink" > (i.e. debian, oabi, for the linkstation > http://buffalo.nas-central.org/index.php/FreeLink), it's using the > stock, eabi[*], kernel (Linux lspro 2.6.12.6-arm1 #2 Mon Jul 23 22:35:39 > CEST 2007 armv5tejl GNU/Linux) and xfs works just fine. yeah, it should. For my list of on-disk structures, arm eabi structure layout is identical to ia64 - I don't think any heads up is needed. > [*] though I think, but I'm not sure, it was not the real thing but > something that marvell patched in. > > However, any later kernel I tried "breaks" xfs (that's why I originally > subscribed to this list), i.e. I cannot see the contents of some directories > (more details here: > http://buffalo.nas-central.org/forums/viewtopic.php?p=35061#p35061) do you still have that small fs image? I'll take a look. There was also an arm get_unaligned issue which hit xfs in another spot, http://marc.info/?l=git-commits-head&m=120433318323826&w=3 might be worth seeing if you still have trouble with that fix in place (which is slated for 2.6.25) -Eric > The strange thing is that I can mount the failing image on i386, so it > probably has the correct structures on disk. > > Maybe looking at what marvell/buffalo patched in that kernel could give > some insight on the xfs issues with arm. > Bye