On Tue 18-11-14 16:27:17, Andrew Morton wrote:
> On Wed, 12 Nov 2014 17:07:42 +0100 Jan Kara <jack@xxxxxxx> wrote:
>
> > There's lot of common code in inode and mount marks handling. Factor it
> > out to common helper function.
>
> Confused.
>
> > --- a/fs/notify/inode_mark.c
> > +++ b/fs/notify/inode_mark.c
> > @@ -31,28 +31,13 @@
> > #include "../internal.h"
> >
> > /*
> > - * Recalculate the mask of events relevant to a given inode locked.
> > - */
> > -static void fsnotify_recalc_inode_mask_locked(struct inode *inode)
> > -{
> > - struct fsnotify_mark *mark;
> > - __u32 new_mask = 0;
> > -
> > - assert_spin_locked(&inode->i_lock);
> > -
> > - hlist_for_each_entry(mark, &inode->i_fsnotify_marks, i.i_list)
> > - new_mask |= mark->mask;
> > - inode->i_fsnotify_mask = new_mask;
> > -}
>
> Gone.
>
> > @@ -202,36 +153,9 @@ int fsnotify_add_inode_mark(struct fsnotify_mark *mark,
> > assert_spin_locked(&mark->lock);
> >
> > spin_lock(&inode->i_lock);
> > -
> > - mark->i.inode = inode;
> > -
> > - /* is mark the first mark? */
> > - if (hlist_empty(&inode->i_fsnotify_marks)) {
> > - hlist_add_head_rcu(&mark->i.i_list, &inode->i_fsnotify_marks);
> > - goto out;
> > - }
> > -
> > - /* should mark be in the middle of the current list? */
> > - hlist_for_each_entry(lmark, &inode->i_fsnotify_marks, i.i_list) {
> > - last = lmark;
> > -
> > - if ((lmark->group == group) && !allow_dups) {
> > - ret = -EEXIST;
> > - goto out;
> > - }
> > -
> > - cmp = fsnotify_compare_groups(lmark->group, mark->group);
> > - if (cmp < 0)
> > - continue;
> > -
> > - hlist_add_before_rcu(&mark->i.i_list, &lmark->i.i_list);
> > - goto out;
> > - }
> > -
> > - BUG_ON(last == NULL);
> > - /* mark should be the last entry. last is the current last entry */
> > - hlist_add_behind_rcu(&mark->i.i_list, &last->i.i_list);
> > -out:
> > + mark->inode = inode;
> > + ret = fsnotify_add_mark_list(&inode->i_fsnotify_marks, mark,
> > + allow_dups);
> > fsnotify_recalc_inode_mask_locked(inode);
>
> But not forgotten. The compiler noticed this ;)
>
> I'll drop fsnotify-unify-inode-and-mount-marks-handling.patch and
> fsnotify-remove-destroy_list-from-fsnotify_mark.patch - could you take
> a look please?
>
> I had to stage those two patches after linux-next.patch because someone
> has been fiddling with fs/notify/ in -next, causing conflicts. Perhaps
> that had something to do with it, but I don't see how.
Bah, looks like some stupidity on my side. Not sure how that happened.
I'll fix it up and resend.
Honza
--
Jan Kara <jack@xxxxxxx>
SUSE Labs, CR
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
