From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dave Chinner Subject: Re: fs: Inode cache scalability V3 Date: Thu, 14 Oct 2010 10:55:52 +1100 Message-ID: <20101013235552.GA4681@dastard> References: <1286928961-15157-1-git-send-email-david@fromorbit.com> <20101013145102.GA12155@infradead.org> <20101013155845.GB22447@infradead.org> <20101013214609.GA24695@infradead.org> <20101013233647.GA18691@infradead.org> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, axboe@kernel.dk To: Christoph Hellwig Return-path: Content-Disposition: inline In-Reply-To: <20101013233647.GA18691@infradead.org> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Wed, Oct 13, 2010 at 07:36:48PM -0400, Christoph Hellwig wrote: > On Wed, Oct 13, 2010 at 05:46:09PM -0400, Christoph Hellwig wrote: > > On Wed, Oct 13, 2010 at 11:58:45AM -0400, Christoph Hellwig wrote: > > >=20 > > > It's 100% reproducible on my kvm VM. The bug is the assert_spin_= locked > > > in redirty_tail. I really can't find a way how we reach it witho= ut > > > d_lock so this really confuses me. > >=20 > > We are for some reason getting a block device inode that is on the > > dirty list of a bdi that it doesn't point to. Still trying to figu= re > > out how exactly that happens. >=20 > It's because __blkdev_put reset the bdi on the mapping, and bdev inod= es > are still special cased to not use s_bdi unlike everybody else. So > we keep switch between different bdis that get locked. >=20 > I wonder what's a good workaround for that. Just flushing out all > dirty state of a block device inode on last close would fix, but we'd > still have all the dragons hidden underneath until we finally sort > out the bdi reference mess. Perhaps for the moment make __blkdev_put() move the inode onto the dirty lists for the default bdi when it switches them=EE=80=80in the mapping? e.g. add a "inode_switch_bdi" helper that is only called in this case? Cheers, Dave. --=20 Dave Chinner david@fromorbit.com