Messages in this thread | | | Date | Sun, 6 Sep 2009 11:17:01 -0700 | From | Greg KH <> | Subject | Re: [patch 41/48] ocfs2: Initialize the cluster were writing to in a non-sparse extend |
| |
On Sun, Sep 06, 2009 at 01:32:50AM -0700, Joel Becker wrote: > On Fri, Sep 04, 2009 at 02:00:40PM -0700, Joel Becker wrote: > > On Fri, Sep 04, 2009 at 01:07:53PM -0700, Greg KH wrote: > > > 2.6.27-stable review patch. If anyone has any objections, please let us know. > > > > > > ------------------ > > > From: Sunil Mushran <sunil.mushran@oracle.com> > > > > > > commit e7432675f8ca868a4af365759a8d4c3779a3d922 upstream. > > > > > > In a non-sparse extend, we correctly allocate (and zero) the clusters between > > > the old_i_size and pos, but we don't zero the portions of the cluster we're > > > writing to outside of pos<->len. > > > > > > It handles clustersize > pagesize and blocksize < pagesize. > > > > > > [Cleaned up by Joel Becker.] > > > > Greg, et al, > > Yesterday we found a bug in this patch. The patch still fixes > > the original problem, but we introduced an error in certain > > page_mkwrite() corner cases. We have one more validation test and then > > I'm sending the fix to Linus. > > So we really want both this commit and the followup fix in > > -stable. You'll be CC'd when I send Linus the pull request. > > Ok, the fix is upstream, commit > 8379e7c46cc48f51197dd663fc6676f47f2a1e71. It applies on top of this > patch, and is needed for both stable trees.
Ok, I'll queue it up.
thanks,
greg k-h
| |