From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jens Axboe Subject: Re: [patch] btrfs: fix inode rbtree corruption Date: Wed, 19 Aug 2009 10:59:07 +0200 Message-ID: <20090819085906.GV12579@kernel.dk> References: <20090818164542.GB30325@wotan.suse.de> <3d0408630908181156l16ccbc92p529f38cf622949cb@mail.gmail.com> <20090818211910.GR12579@kernel.dk> <20090819084530.GD25721@wotan.suse.de> <20090819084658.GT12579@kernel.dk> <20090819085208.GF25721@wotan.suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: "Yan, Zheng " , Chris Mason , linux-btrfs@vger.kernel.org To: Nick Piggin Return-path: In-Reply-To: <20090819085208.GF25721@wotan.suse.de> List-ID: On Wed, Aug 19 2009, Nick Piggin wrote: > On Wed, Aug 19, 2009 at 10:46:59AM +0200, Jens Axboe wrote: > > On Wed, Aug 19 2009, Nick Piggin wrote: > > > See my other reply. It *can* work with key aliases, but this particular > > > code does not. > > > > > > It is pretty easy obviously to put in duplicates because the rbtree > > > code doesn't know about keys, but if we do this then it looks like > > > it might cause the search code to miss some valid inodes and instead > > > return freeing inodes -- so you'd also have to look at that and update > > > it which is why I didn't go down this route.. > > > > Mine was just a generic statement, I didn't read the btrfs code (hence > > my comment about potential lookup bug, if you allow aliases you have to > > be careful). > > Ah ok. Well yeah in this case btrfs is definitely wrong in the way it > tried to insert aliases. I looked at the actual problem now and I agree, it cannot work that way. I don't know if Linus is planning another -rc, we should probably get this upstream sooner rather than later. Chris is away this week, so if we can get Yan to agree on this patch as well, I'll submit it. -- Jens Axboe