From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chris Mason Subject: Re: [PATCH] Use do_div() instead of native 64-bit division in btrfs_ordered_sum_size() Date: Wed, 23 Jul 2008 20:19:15 -0400 Message-ID: <1216858755.6932.154.camel@think.oraclecorp.com> References: <1216697528.18980.34.camel@shinybook.infradead.org> <20080723111309.GA19988@infradead.org> <1216819706.6932.146.camel@think.oraclecorp.com> <1216847224.3019.137.camel@shinybook.infradead.org> Mime-Version: 1.0 Content-Type: text/plain Cc: Christoph Hellwig , linux-btrfs@vger.kernel.org To: David Woodhouse Return-path: In-Reply-To: <1216847224.3019.137.camel@shinybook.infradead.org> List-ID: On Wed, 2008-07-23 at 17:07 -0400, David Woodhouse wrote: > On Wed, 2008-07-23 at 09:28 -0400, Chris Mason wrote: > > On Wed, 2008-07-23 at 07:13 -0400, Christoph Hellwig wrote: > > > Chris, can you please put this patch in? Without it btrfs can't be > > > loaded on 32bit platforms. > > > > > > > I've pushed out a slightly different fix. The ordered extents are > > based on ram writeback, and so an unsigned long is enough. > > Does the job for me, although we still need the > s/BUG_ON(spin_trylock(&tree->lock))/assert_spin_locked(&tree->lock)/ > patch I sent a few days ago. > > Now I can actually build the module, load it and avoid it hitting a > BUG_ON() when it first tries to mount, I can perhaps move on to doing > something more useful with it... :) ;) My goal tonight is to figure out why I'm leaving locked data pages everywhere. Once that is done I'll be able to finally integrate the pending patches. -chris