On Tue, 2016-05-03 at 13:49 +0200, Andreas Gruenbacher wrote: > On Tue, May 3, 2016 at 4:40 AM, Mimi Zohar <zohar@xxxxxxxxxxxxxxxxxx> wrote: > > Hi Andreas, > > > > On Tue, 2016-05-03 at 00:45 +0200, Andreas Gruenbacher wrote: > >> Hi all, > >> > >> what are your thoughts on this patch set? It applies on top of the > >> work.xattr branch [*], converts the remaining filesystems over to xattr > >> handlers, and replaces the getxattr, setxattr, and removexattr inode > >> operations. The only way to implement getxattr, setxattr, and > >> removexattr with this approach is through xattr handlers. > > > > The patch description should provide the motivation/reason for the > > change (eg. performance, locking). Up to now these xattr functions > > required the caller to take the i_mutex. Is the i_mutex still required? > > Yes, the documentation needs updating. The locking rules are still the same. I meant to say, this cover letter needs to provide the motivation/reason for the change(s). From my perspective, the main motivation would either be "locking" or keeping the file data and metadata in sync. If these aren't the motivation/reasons for the change, please provide an explanation. thanks, Mimi -- To unsubscribe from this list: send the line "unsubscribe linux-unionfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
