On Wed 22-07-20 15:58:42, Amir Goldstein wrote:
> Since commit ecf13b5f8fd6 ("fsnotify: send event with parent/name info
> to sb/mount/non-dir marks") the flag FS_EVENT_ON_CHILD has a meaning in
> mask of a mark on a non-dir inode. It means that group is interested
> in the name of the file with events.
>
> Since inotify is only intereseted in names of children of a watching
> parent, do not sete FS_EVENT_ON_CHILD flag for marks on non-dir.
>
> Signed-off-by: Amir Goldstein <amir73il@xxxxxxxxx>
I've placed this commit in the series before "fsnotify: send event with
parent/name info to sb/mount/non-dir marks" and updated changelog
accordingly.
Honza
> ---
> fs/notify/inotify/inotify_user.c | 14 ++++++++------
> 1 file changed, 8 insertions(+), 6 deletions(-)
>
> diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c
> index 5385d5817dd9..186722ba3894 100644
> --- a/fs/notify/inotify/inotify_user.c
> +++ b/fs/notify/inotify/inotify_user.c
> @@ -75,15 +75,17 @@ struct ctl_table inotify_table[] = {
> };
> #endif /* CONFIG_SYSCTL */
>
> -static inline __u32 inotify_arg_to_mask(u32 arg)
> +static inline __u32 inotify_arg_to_mask(struct inode *inode, u32 arg)
> {
> __u32 mask;
>
> /*
> - * everything should accept their own ignored, cares about children,
> - * and should receive events when the inode is unmounted
> + * Everything should accept their own ignored and should receive events
> + * when the inode is unmounted. All directories care about children.
> */
> - mask = (FS_IN_IGNORED | FS_EVENT_ON_CHILD | FS_UNMOUNT);
> + mask = (FS_IN_IGNORED | FS_UNMOUNT);
> + if (S_ISDIR(inode->i_mode))
> + mask |= FS_EVENT_ON_CHILD;
>
> /* mask off the flags used to open the fd */
> mask |= (arg & (IN_ALL_EVENTS | IN_ONESHOT | IN_EXCL_UNLINK));
> @@ -512,7 +514,7 @@ static int inotify_update_existing_watch(struct fsnotify_group *group,
> int create = (arg & IN_MASK_CREATE);
> int ret;
>
> - mask = inotify_arg_to_mask(arg);
> + mask = inotify_arg_to_mask(inode, arg);
>
> fsn_mark = fsnotify_find_mark(&inode->i_fsnotify_marks, group);
> if (!fsn_mark)
> @@ -565,7 +567,7 @@ static int inotify_new_watch(struct fsnotify_group *group,
> struct idr *idr = &group->inotify_data.idr;
> spinlock_t *idr_lock = &group->inotify_data.idr_lock;
>
> - mask = inotify_arg_to_mask(arg);
> + mask = inotify_arg_to_mask(inode, arg);
>
> tmp_i_mark = kmem_cache_alloc(inotify_inode_mark_cachep, GFP_KERNEL);
> if (unlikely(!tmp_i_mark))
> --
> 2.17.1
>
--
Jan Kara <jack@xxxxxxxx>
SUSE Labs, CR
