2011-11-24 08:26:23 +08:00
|
|
|
#include <linux/mount.h>
|
2011-12-07 01:21:54 +08:00
|
|
|
#include <linux/seq_file.h>
|
|
|
|
#include <linux/poll.h>
|
2014-11-01 10:56:04 +08:00
|
|
|
#include <linux/ns_common.h>
|
2015-01-11 08:01:08 +08:00
|
|
|
#include <linux/fs_pin.h>
|
2011-12-07 01:21:54 +08:00
|
|
|
|
|
|
|
struct mnt_namespace {
|
|
|
|
atomic_t count;
|
2014-11-01 10:56:04 +08:00
|
|
|
struct ns_common ns;
|
2011-12-07 02:32:36 +08:00
|
|
|
struct mount * root;
|
2011-12-07 01:21:54 +08:00
|
|
|
struct list_head list;
|
2012-07-27 12:08:32 +08:00
|
|
|
struct user_namespace *user_ns;
|
2016-08-09 03:37:37 +08:00
|
|
|
struct ucounts *ucounts;
|
2010-03-08 10:49:36 +08:00
|
|
|
u64 seq; /* Sequence number to prevent loops */
|
2011-12-07 01:21:54 +08:00
|
|
|
wait_queue_head_t poll;
|
2014-02-28 03:40:10 +08:00
|
|
|
u64 event;
|
2016-09-28 13:27:17 +08:00
|
|
|
unsigned int mounts; /* # of mounts in the namespace */
|
|
|
|
unsigned int pending_mounts;
|
2011-12-07 01:21:54 +08:00
|
|
|
};
|
2011-11-24 08:26:23 +08:00
|
|
|
|
2011-11-25 11:53:09 +08:00
|
|
|
struct mnt_pcp {
|
|
|
|
int mnt_count;
|
|
|
|
int mnt_writers;
|
|
|
|
};
|
|
|
|
|
2013-03-15 22:53:28 +08:00
|
|
|
struct mountpoint {
|
2014-03-01 02:46:44 +08:00
|
|
|
struct hlist_node m_hash;
|
2013-03-15 22:53:28 +08:00
|
|
|
struct dentry *m_dentry;
|
2013-09-23 10:37:01 +08:00
|
|
|
struct hlist_head m_list;
|
2013-03-15 22:53:28 +08:00
|
|
|
int m_count;
|
|
|
|
};
|
|
|
|
|
2011-11-24 01:14:10 +08:00
|
|
|
struct mount {
|
2014-03-21 09:10:51 +08:00
|
|
|
struct hlist_node mnt_hash;
|
2011-11-25 11:19:58 +08:00
|
|
|
struct mount *mnt_parent;
|
2011-11-25 11:25:07 +08:00
|
|
|
struct dentry *mnt_mountpoint;
|
2011-11-24 01:14:10 +08:00
|
|
|
struct vfsmount mnt;
|
2014-08-09 01:08:20 +08:00
|
|
|
union {
|
|
|
|
struct rcu_head mnt_rcu;
|
|
|
|
struct llist_node mnt_llist;
|
|
|
|
};
|
2011-11-25 11:53:09 +08:00
|
|
|
#ifdef CONFIG_SMP
|
|
|
|
struct mnt_pcp __percpu *mnt_pcp;
|
|
|
|
#else
|
|
|
|
int mnt_count;
|
|
|
|
int mnt_writers;
|
|
|
|
#endif
|
2011-11-25 12:24:33 +08:00
|
|
|
struct list_head mnt_mounts; /* list of children, anchored here */
|
|
|
|
struct list_head mnt_child; /* and going through their mnt_child */
|
2011-11-21 19:11:30 +08:00
|
|
|
struct list_head mnt_instance; /* mount instance on sb->s_mounts */
|
2011-11-25 15:25:17 +08:00
|
|
|
const char *mnt_devname; /* Name of device e.g. /dev/dsk/hda1 */
|
2011-11-25 15:19:55 +08:00
|
|
|
struct list_head mnt_list;
|
2011-11-25 13:22:05 +08:00
|
|
|
struct list_head mnt_expire; /* link in fs-specific expiry list */
|
|
|
|
struct list_head mnt_share; /* circular list of shared mounts */
|
|
|
|
struct list_head mnt_slave_list;/* list of slave mounts */
|
|
|
|
struct list_head mnt_slave; /* slave list entry */
|
2011-11-25 13:10:28 +08:00
|
|
|
struct mount *mnt_master; /* slave is on master->mnt_slave_list */
|
2011-11-25 13:46:35 +08:00
|
|
|
struct mnt_namespace *mnt_ns; /* containing namespace */
|
2013-03-15 22:53:28 +08:00
|
|
|
struct mountpoint *mnt_mp; /* where is it mounted */
|
2013-09-23 10:37:01 +08:00
|
|
|
struct hlist_node mnt_mp_list; /* list mounts with the same mountpoint */
|
mnt: In umount propagation reparent in a separate pass
It was observed that in some pathlogical cases that the current code
does not unmount everything it should. After investigation it
was determined that the issue is that mnt_change_mntpoint can
can change which mounts are available to be unmounted during mount
propagation which is wrong.
The trivial reproducer is:
$ cat ./pathological.sh
mount -t tmpfs test-base /mnt
cd /mnt
mkdir 1 2 1/1
mount --bind 1 1
mount --make-shared 1
mount --bind 1 2
mount --bind 1/1 1/1
mount --bind 1/1 1/1
echo
grep test-base /proc/self/mountinfo
umount 1/1
echo
grep test-base /proc/self/mountinfo
$ unshare -Urm ./pathological.sh
The expected output looks like:
46 31 0:25 / /mnt rw,relatime - tmpfs test-base rw,uid=1000,gid=1000
47 46 0:25 /1 /mnt/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
48 46 0:25 /1 /mnt/2 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
49 54 0:25 /1/1 /mnt/1/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
50 53 0:25 /1/1 /mnt/2/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
51 49 0:25 /1/1 /mnt/1/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
54 47 0:25 /1/1 /mnt/1/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
53 48 0:25 /1/1 /mnt/2/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
52 50 0:25 /1/1 /mnt/2/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
46 31 0:25 / /mnt rw,relatime - tmpfs test-base rw,uid=1000,gid=1000
47 46 0:25 /1 /mnt/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
48 46 0:25 /1 /mnt/2 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
The output without the fix looks like:
46 31 0:25 / /mnt rw,relatime - tmpfs test-base rw,uid=1000,gid=1000
47 46 0:25 /1 /mnt/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
48 46 0:25 /1 /mnt/2 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
49 54 0:25 /1/1 /mnt/1/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
50 53 0:25 /1/1 /mnt/2/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
51 49 0:25 /1/1 /mnt/1/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
54 47 0:25 /1/1 /mnt/1/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
53 48 0:25 /1/1 /mnt/2/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
52 50 0:25 /1/1 /mnt/2/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
46 31 0:25 / /mnt rw,relatime - tmpfs test-base rw,uid=1000,gid=1000
47 46 0:25 /1 /mnt/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
48 46 0:25 /1 /mnt/2 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
52 48 0:25 /1/1 /mnt/2/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000
That last mount in the output was in the propgation tree to be unmounted but
was missed because the mnt_change_mountpoint changed it's parent before the walk
through the mount propagation tree observed it.
Cc: stable@vger.kernel.org
Fixes: 1064f874abc0 ("mnt: Tuck mounts under others instead of creating shadow/side mounts.")
Acked-by: Andrei Vagin <avagin@virtuozzo.com>
Reviewed-by: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
2017-05-16 03:42:07 +08:00
|
|
|
struct list_head mnt_reparent; /* reparent list entry */
|
2011-11-25 15:35:16 +08:00
|
|
|
#ifdef CONFIG_FSNOTIFY
|
2017-02-01 16:21:58 +08:00
|
|
|
struct fsnotify_mark_connector __rcu *mnt_fsnotify_marks;
|
2011-11-25 15:35:16 +08:00
|
|
|
__u32 mnt_fsnotify_mask;
|
|
|
|
#endif
|
2011-11-25 13:50:41 +08:00
|
|
|
int mnt_id; /* mount identifier */
|
|
|
|
int mnt_group_id; /* peer group identifier */
|
2011-11-25 13:57:42 +08:00
|
|
|
int mnt_expiry_mark; /* true if marked for expiry */
|
2014-08-07 18:23:41 +08:00
|
|
|
struct hlist_head mnt_pins;
|
2015-01-11 08:01:08 +08:00
|
|
|
struct fs_pin mnt_umount;
|
|
|
|
struct dentry *mnt_ex_mountpoint;
|
2011-11-24 01:14:10 +08:00
|
|
|
};
|
|
|
|
|
2012-06-09 12:59:08 +08:00
|
|
|
#define MNT_NS_INTERNAL ERR_PTR(-EINVAL) /* distinct from any mnt_namespace */
|
|
|
|
|
2011-11-24 01:14:10 +08:00
|
|
|
static inline struct mount *real_mount(struct vfsmount *mnt)
|
|
|
|
{
|
|
|
|
return container_of(mnt, struct mount, mnt);
|
|
|
|
}
|
|
|
|
|
2011-11-25 10:47:05 +08:00
|
|
|
static inline int mnt_has_parent(struct mount *mnt)
|
2011-11-24 08:26:23 +08:00
|
|
|
{
|
2011-11-25 11:19:58 +08:00
|
|
|
return mnt != mnt->mnt_parent;
|
2011-11-24 08:26:23 +08:00
|
|
|
}
|
2011-11-25 07:22:03 +08:00
|
|
|
|
2012-06-09 12:59:08 +08:00
|
|
|
static inline int is_mounted(struct vfsmount *mnt)
|
|
|
|
{
|
|
|
|
/* neither detached nor internal? */
|
2014-01-21 07:26:15 +08:00
|
|
|
return !IS_ERR_OR_NULL(real_mount(mnt)->mnt_ns);
|
2012-06-09 12:59:08 +08:00
|
|
|
}
|
|
|
|
|
2013-10-02 04:11:26 +08:00
|
|
|
extern struct mount *__lookup_mnt(struct vfsmount *, struct dentry *);
|
2011-12-07 01:21:54 +08:00
|
|
|
|
2015-05-08 23:43:53 +08:00
|
|
|
extern int __legitimize_mnt(struct vfsmount *, unsigned);
|
2013-09-30 10:06:07 +08:00
|
|
|
extern bool legitimize_mnt(struct vfsmount *, unsigned);
|
|
|
|
|
2016-11-24 05:03:41 +08:00
|
|
|
static inline bool __path_is_mountpoint(const struct path *path)
|
|
|
|
{
|
|
|
|
struct mount *m = __lookup_mnt(path->mnt, path->dentry);
|
|
|
|
return m && likely(!(m->mnt.mnt_flags & MNT_SYNC_UMOUNT));
|
|
|
|
}
|
|
|
|
|
2013-10-03 16:31:18 +08:00
|
|
|
extern void __detach_mounts(struct dentry *dentry);
|
|
|
|
|
|
|
|
static inline void detach_mounts(struct dentry *dentry)
|
|
|
|
{
|
|
|
|
if (!d_mountpoint(dentry))
|
|
|
|
return;
|
|
|
|
__detach_mounts(dentry);
|
|
|
|
}
|
|
|
|
|
2011-12-07 01:21:54 +08:00
|
|
|
static inline void get_mnt_ns(struct mnt_namespace *ns)
|
|
|
|
{
|
|
|
|
atomic_inc(&ns->count);
|
|
|
|
}
|
|
|
|
|
2013-09-30 10:06:07 +08:00
|
|
|
extern seqlock_t mount_lock;
|
2013-09-29 23:24:49 +08:00
|
|
|
|
|
|
|
static inline void lock_mount_hash(void)
|
|
|
|
{
|
2013-09-30 10:06:07 +08:00
|
|
|
write_seqlock(&mount_lock);
|
2013-09-29 23:24:49 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
static inline void unlock_mount_hash(void)
|
|
|
|
{
|
2013-09-30 10:06:07 +08:00
|
|
|
write_sequnlock(&mount_lock);
|
2013-09-29 23:24:49 +08:00
|
|
|
}
|
|
|
|
|
2011-12-07 01:21:54 +08:00
|
|
|
struct proc_mounts {
|
|
|
|
struct mnt_namespace *ns;
|
|
|
|
struct path root;
|
|
|
|
int (*show)(struct seq_file *, struct vfsmount *);
|
2014-02-28 03:40:10 +08:00
|
|
|
void *cached_mount;
|
|
|
|
u64 cached_event;
|
|
|
|
loff_t cached_index;
|
2011-12-07 01:21:54 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
extern const struct seq_operations mounts_op;
|
2013-10-05 10:15:13 +08:00
|
|
|
|
|
|
|
extern bool __is_local_mountpoint(struct dentry *dentry);
|
|
|
|
static inline bool is_local_mountpoint(struct dentry *dentry)
|
|
|
|
{
|
|
|
|
if (!d_mountpoint(dentry))
|
|
|
|
return false;
|
|
|
|
|
|
|
|
return __is_local_mountpoint(dentry);
|
|
|
|
}
|