From: Mark Fasheh Date: Fri, 19 Mar 2010 15:01:16 -0700 Subject: [Ocfs2-devel] [PATCH] ocfs2: Clear undo bits when local alloc is freed In-Reply-To: <20100318215314.GA8122@mail.oracle.com> References: <20100312023109.GS11402@wotan.suse.de> <20100318215314.GA8122@mail.oracle.com> Message-ID: <20100319220116.GH11402@wotan.suse.de> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: ocfs2-devel@oss.oracle.com On Thu, Mar 18, 2010 at 02:53:15PM -0700, Joel Becker wrote: > On Thu, Mar 11, 2010 at 06:31:09PM -0800, Mark Fasheh wrote: > > When the local alloc file changes windows, unused bits are freed back to the > > global bitmap. By defnition, those bits can not be in use by any file. Also, > > the local alloc will never have been able to allocate those bits if they > > were part of a previous truncate. Therefore it makes sense that we should > > clear unused local alloc bits in the undo buffer so that they can be used > > immediatly. > > > > Signed-off-by: Mark Fasheh > > --- > > fs/ocfs2/alloc.c | 4 ++-- > > fs/ocfs2/dir.c | 2 +- > > fs/ocfs2/localalloc.c | 2 +- > > fs/ocfs2/refcounttree.c | 2 +- > > fs/ocfs2/suballoc.c | 44 ++++++++++++++++++++++++++++---------------- > > fs/ocfs2/suballoc.h | 5 +++-- > > fs/ocfs2/xattr.c | 2 +- > > 7 files changed, 37 insertions(+), 24 deletions(-) > > > > diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c > > index d17bdc7..c9608a5 100644 > > --- a/fs/ocfs2/alloc.c > > +++ b/fs/ocfs2/alloc.c > > @@ -5921,7 +5921,7 @@ static int ocfs2_replay_truncate_records(struct ocfs2_super *osb, > > > > status = ocfs2_free_clusters(handle, data_alloc_inode, > > data_alloc_bh, start_blk, > > - num_clusters); > > + num_clusters, 0); > > Seeing all these ', 0' arguments appear, I was confused until I > saw the actual ocfs2_free_clusters change later in the patch. That made > me worry I would be re-confused coming back to this code six months from > now. > Here's my shot at leaving ocfs2_free_clusters() and > ocfs2_free_suballoc_bits() intact while providing the same > functionality. Yeah, that patch looks better. Thanks for the cleanup. Acked-by: Mark Fasheh -- Mark Fasheh