Merge branch 'for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs

Pull vfs fixes from Al Viro:
 "Switch mnt_hash to hlist, turning the races between __lookup_mnt() and
  hash modifications into false negatives from __lookup_mnt() (instead
  of hangs)"

On the false negatives from __lookup_mnt():
 "The *only* thing we care about is not getting stuck in __lookup_mnt().
  If it misses an entry because something in front of it just got moved
  around, etc, we are fine.  We'll notice that mount_lock mismatch and
  that'll be it"

* 'for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  switch mnt_hash to hlist
  don't bother with propagate_mnt() unless the target is shared
  keep shadowed vfsmounts together
  resizable namespace.c hashes

4 files changed, 134 insertions, 77 deletions

diff --git a/fs/mount.h b/fs/mount.h
index a17458ca6f29..b29e42f05f34 100644
--- a/fs/mount.h
+++ b/fs/mount.h
@@ -19,13 +19,13 @@ struct mnt_pcp {
 };
 
 struct mountpoint {
-        struct list_head m_hash;
+        struct hlist_node m_hash;
         struct dentry *m_dentry;
         int m_count;
 };
 
 struct mount {
-        struct list_head mnt_hash;
+        struct hlist_node mnt_hash;
         struct mount *mnt_parent;
         struct dentry *mnt_mountpoint;
         struct vfsmount mnt;
diff --git a/fs/namespace.c b/fs/namespace.c
index 22e536705c45..2ffc5a2905d4 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -23,11 +23,34 @@
 #include <linux/uaccess.h>
 #include <linux/proc_ns.h>
 #include <linux/magic.h>
+#include <linux/bootmem.h>
 #include "pnode.h"
 #include "internal.h"
 
-#define HASH_SHIFT ilog2(PAGE_SIZE / sizeof(struct list_head))
-#define HASH_SIZE (1UL << HASH_SHIFT)
+static unsigned int m_hash_mask __read_mostly;
+static unsigned int m_hash_shift __read_mostly;
+static unsigned int mp_hash_mask __read_mostly;
+static unsigned int mp_hash_shift __read_mostly;
+
+static __initdata unsigned long mhash_entries;
+static int __init set_mhash_entries(char *str)
+{
+        if (!str)
+                return 0;
+        mhash_entries = simple_strtoul(str, &str, 0);
+        return 1;
+}
+__setup("mhash_entries=", set_mhash_entries);
+
+static __initdata unsigned long mphash_entries;
+static int __init set_mphash_entries(char *str)
+{
+        if (!str)
+                return 0;
+        mphash_entries = simple_strtoul(str, &str, 0);
+        return 1;
+}
+__setup("mphash_entries=", set_mphash_entries);
 
 static int event;
 static DEFINE_IDA(mnt_id_ida);
@@ -36,8 +59,8 @@ static DEFINE_SPINLOCK(mnt_id_lock);
 static int mnt_id_start = 0;
 static int mnt_group_start = 1;
 
-static struct list_head *mount_hashtable __read_mostly;
-static struct list_head *mountpoint_hashtable __read_mostly;
+static struct hlist_head *mount_hashtable __read_mostly;
+static struct hlist_head *mountpoint_hashtable __read_mostly;
 static struct kmem_cache *mnt_cache __read_mostly;
 static DECLARE_RWSEM(namespace_sem);
 
@@ -55,12 +78,19 @@ EXPORT_SYMBOL_GPL(fs_kobj);
  */
 __cacheline_aligned_in_smp DEFINE_SEQLOCK(mount_lock);
 
-static inline unsigned long hash(struct vfsmount *mnt, struct dentry *dentry)
+static inline struct hlist_head *m_hash(struct vfsmount *mnt, struct dentry *dentry)
 {
         unsigned long tmp = ((unsigned long)mnt / L1_CACHE_BYTES);
         tmp += ((unsigned long)dentry / L1_CACHE_BYTES);
-        tmp = tmp + (tmp >> HASH_SHIFT);
-        return tmp & (HASH_SIZE - 1);
+        tmp = tmp + (tmp >> m_hash_shift);
+        return &mount_hashtable[tmp & m_hash_mask];
+}
+
+static inline struct hlist_head *mp_hash(struct dentry *dentry)
+{
+        unsigned long tmp = ((unsigned long)dentry / L1_CACHE_BYTES);
+        tmp = tmp + (tmp >> mp_hash_shift);
+        return &mountpoint_hashtable[tmp & mp_hash_mask];
 }
 
 /*
@@ -187,7 +217,7 @@ static struct mount *alloc_vfsmnt(const char *name)
                 mnt->mnt_writers = 0;
 #endif
 
-                INIT_LIST_HEAD(&mnt->mnt_hash);
+                INIT_HLIST_NODE(&mnt->mnt_hash);
                 INIT_LIST_HEAD(&mnt->mnt_child);
                 INIT_LIST_HEAD(&mnt->mnt_mounts);
                 INIT_LIST_HEAD(&mnt->mnt_list);
@@ -575,10 +605,10 @@ bool legitimize_mnt(struct vfsmount *bastard, unsigned seq)
  */
 struct mount *__lookup_mnt(struct vfsmount *mnt, struct dentry *dentry)
 {
-        struct list_head *head = mount_hashtable + hash(mnt, dentry);
+        struct hlist_head *head = m_hash(mnt, dentry);
         struct mount *p;
 
-        list_for_each_entry_rcu(p, head, mnt_hash)
+        hlist_for_each_entry_rcu(p, head, mnt_hash)
                 if (&p->mnt_parent->mnt == mnt && p->mnt_mountpoint == dentry)
                         return p;
         return NULL;
@@ -590,13 +620,17 @@ struct mount *__lookup_mnt(struct vfsmount *mnt, struct dentry *dentry)
  */
 struct mount *__lookup_mnt_last(struct vfsmount *mnt, struct dentry *dentry)
 {
-        struct list_head *head = mount_hashtable + hash(mnt, dentry);
-        struct mount *p;
-
-        list_for_each_entry_reverse(p, head, mnt_hash)
-                if (&p->mnt_parent->mnt == mnt && p->mnt_mountpoint == dentry)
-                        return p;
-        return NULL;
+        struct mount *p, *res;
+        res = p = __lookup_mnt(mnt, dentry);
+        if (!p)
+                goto out;
+        hlist_for_each_entry_continue(p, mnt_hash) {
+                if (&p->mnt_parent->mnt != mnt || p->mnt_mountpoint != dentry)
+                        break;
+                res = p;
+        }
+out:
+        return res;
 }
 
 /*
@@ -633,11 +667,11 @@ struct vfsmount *lookup_mnt(struct path *path)
 
 static struct mountpoint *new_mountpoint(struct dentry *dentry)
 {
-        struct list_head *chain = mountpoint_hashtable + hash(NULL, dentry);
+        struct hlist_head *chain = mp_hash(dentry);
         struct mountpoint *mp;
         int ret;
 
-        list_for_each_entry(mp, chain, m_hash) {
+        hlist_for_each_entry(mp, chain, m_hash) {
                 if (mp->m_dentry == dentry) {
                         /* might be worth a WARN_ON() */
                         if (d_unlinked(dentry))
@@ -659,7 +693,7 @@ static struct mountpoint *new_mountpoint(struct dentry *dentry)
 
         mp->m_dentry = dentry;
         mp->m_count = 1;
-        list_add(&mp->m_hash, chain);
+        hlist_add_head(&mp->m_hash, chain);
         return mp;
 }
 
@@ -670,7 +704,7 @@ static void put_mountpoint(struct mountpoint *mp)
                 spin_lock(&dentry->d_lock);
                 dentry->d_flags &= ~DCACHE_MOUNTED;
                 spin_unlock(&dentry->d_lock);
-                list_del(&mp->m_hash);
+                hlist_del(&mp->m_hash);
                 kfree(mp);
         }
 }
@@ -712,7 +746,7 @@ static void detach_mnt(struct mount *mnt, struct path *old_path)
         mnt->mnt_parent = mnt;
         mnt->mnt_mountpoint = mnt->mnt.mnt_root;
         list_del_init(&mnt->mnt_child);
-        list_del_init(&mnt->mnt_hash);
+        hlist_del_init_rcu(&mnt->mnt_hash);
         put_mountpoint(mnt->mnt_mp);
         mnt->mnt_mp = NULL;
 }
@@ -739,15 +773,14 @@ static void attach_mnt(struct mount *mnt,
                         struct mountpoint *mp)
 {
         mnt_set_mountpoint(parent, mp, mnt);
-        list_add_tail(&mnt->mnt_hash, mount_hashtable +
-                        hash(&parent->mnt, mp->m_dentry));
+        hlist_add_head_rcu(&mnt->mnt_hash, m_hash(&parent->mnt, mp->m_dentry));
         list_add_tail(&mnt->mnt_child, &parent->mnt_mounts);
 }
 
 /*
  * vfsmount lock must be held for write
  */
-static void commit_tree(struct mount *mnt)
+static void commit_tree(struct mount *mnt, struct mount *shadows)
 {
         struct mount *parent = mnt->mnt_parent;
         struct mount *m;
@@ -762,8 +795,11 @@ static void commit_tree(struct mount *mnt)
 
         list_splice(&head, n->list.prev);
 
-        list_add_tail(&mnt->mnt_hash, mount_hashtable +
-                                hash(&parent->mnt, mnt->mnt_mountpoint));
+        if (shadows)
+                hlist_add_after_rcu(&shadows->mnt_hash, &mnt->mnt_hash);
+        else
+                hlist_add_head_rcu(&mnt->mnt_hash,
+                                m_hash(&parent->mnt, mnt->mnt_mountpoint));
         list_add_tail(&mnt->mnt_child, &parent->mnt_mounts);
         touch_mnt_namespace(n);
 }
@@ -1153,26 +1189,28 @@ int may_umount(struct vfsmount *mnt)
 
 EXPORT_SYMBOL(may_umount);
 
-static LIST_HEAD(unmounted);    /* protected by namespace_sem */
+static HLIST_HEAD(unmounted);   /* protected by namespace_sem */
 
 static void namespace_unlock(void)
 {
         struct mount *mnt;
-        LIST_HEAD(head);
+        struct hlist_head head = unmounted;
 
-        if (likely(list_empty(&unmounted))) {
+        if (likely(hlist_empty(&head))) {
                 up_write(&namespace_sem);
                 return;
         }
 
-        list_splice_init(&unmounted, &head);
+        head.first->pprev = &head.first;
+        INIT_HLIST_HEAD(&unmounted);
+
         up_write(&namespace_sem);
 
         synchronize_rcu();
 
-        while (!list_empty(&head)) {
-                mnt = list_first_entry(&head, struct mount, mnt_hash);
-                list_del_init(&mnt->mnt_hash);
+        while (!hlist_empty(&head)) {
+                mnt = hlist_entry(head.first, struct mount, mnt_hash);
+                hlist_del_init(&mnt->mnt_hash);
                 if (mnt->mnt_ex_mountpoint.mnt)
                         path_put(&mnt->mnt_ex_mountpoint);
                 mntput(&mnt->mnt);
@@ -1193,16 +1231,19 @@ static inline void namespace_lock(void)
  */
 void umount_tree(struct mount *mnt, int how)
 {
-        LIST_HEAD(tmp_list);
+        HLIST_HEAD(tmp_list);
         struct mount *p;
+        struct mount *last = NULL;
 
-        for (p = mnt; p; p = next_mnt(p, mnt))
-                list_move(&p->mnt_hash, &tmp_list);
+        for (p = mnt; p; p = next_mnt(p, mnt)) {
+                hlist_del_init_rcu(&p->mnt_hash);
+                hlist_add_head(&p->mnt_hash, &tmp_list);
+        }
 
         if (how)
                 propagate_umount(&tmp_list);
 
-        list_for_each_entry(p, &tmp_list, mnt_hash) {
+        hlist_for_each_entry(p, &tmp_list, mnt_hash) {
                 list_del_init(&p->mnt_expire);
                 list_del_init(&p->mnt_list);
                 __touch_mnt_namespace(p->mnt_ns);
@@ -1220,8 +1261,13 @@ void umount_tree(struct mount *mnt, int how)
                         p->mnt_mp = NULL;
                 }
                 change_mnt_propagation(p, MS_PRIVATE);
+                last = p;
+        }
+        if (last) {
+                last->mnt_hash.next = unmounted.first;
+                unmounted.first = tmp_list.first;
+                unmounted.first->pprev = &unmounted.first;
         }
-        list_splice(&tmp_list, &unmounted);
 }
 
 static void shrink_submounts(struct mount *mnt);
@@ -1605,24 +1651,23 @@ static int attach_recursive_mnt(struct mount *source_mnt,
                         struct mountpoint *dest_mp,
                         struct path *parent_path)
 {
-        LIST_HEAD(tree_list);
+        HLIST_HEAD(tree_list);
         struct mount *child, *p;
+        struct hlist_node *n;
         int err;
 
         if (IS_MNT_SHARED(dest_mnt)) {
                 err = invent_group_ids(source_mnt, true);
                 if (err)
                         goto out;
-        }
-        err = propagate_mnt(dest_mnt, dest_mp, source_mnt, &tree_list);
-        if (err)
-                goto out_cleanup_ids;
-
-        lock_mount_hash();
-
-        if (IS_MNT_SHARED(dest_mnt)) {
+                err = propagate_mnt(dest_mnt, dest_mp, source_mnt, &tree_list);
+                if (err)
+                        goto out_cleanup_ids;
+                lock_mount_hash();
                 for (p = source_mnt; p; p = next_mnt(p, source_mnt))
                         set_mnt_shared(p);
+        } else {
+                lock_mount_hash();
         }
         if (parent_path) {
                 detach_mnt(source_mnt, parent_path);
@@ -1630,20 +1675,22 @@ static int attach_recursive_mnt(struct mount *source_mnt,
                 touch_mnt_namespace(source_mnt->mnt_ns);
         } else {
                 mnt_set_mountpoint(dest_mnt, dest_mp, source_mnt);
-                commit_tree(source_mnt);
+                commit_tree(source_mnt, NULL);
         }
 
-        list_for_each_entry_safe(child, p, &tree_list, mnt_hash) {
-                list_del_init(&child->mnt_hash);
-                commit_tree(child);
+        hlist_for_each_entry_safe(child, n, &tree_list, mnt_hash) {
+                struct mount *q;
+                hlist_del_init(&child->mnt_hash);
+                q = __lookup_mnt_last(&child->mnt_parent->mnt,
+                                      child->mnt_mountpoint);
+                commit_tree(child, q);
         }
         unlock_mount_hash();
 
         return 0;
 
  out_cleanup_ids:
-        if (IS_MNT_SHARED(dest_mnt))
-                cleanup_group_ids(source_mnt, NULL);
+        cleanup_group_ids(source_mnt, NULL);
  out:
         return err;
 }
@@ -2777,18 +2824,24 @@ void __init mnt_init(void)
         mnt_cache = kmem_cache_create("mnt_cache", sizeof(struct mount),
                         0, SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL);
 
-        mount_hashtable = (struct list_head *)__get_free_page(GFP_ATOMIC);
-        mountpoint_hashtable = (struct list_head *)__get_free_page(GFP_ATOMIC);
+        mount_hashtable = alloc_large_system_hash("Mount-cache",
+                                sizeof(struct hlist_head),
+                                mhash_entries, 19,
+                                0,
+                                &m_hash_shift, &m_hash_mask, 0, 0);
+        mountpoint_hashtable = alloc_large_system_hash("Mountpoint-cache",
+                                sizeof(struct hlist_head),
+                                mphash_entries, 19,
+                                0,
+                                &mp_hash_shift, &mp_hash_mask, 0, 0);
 
         if (!mount_hashtable || !mountpoint_hashtable)
                 panic("Failed to allocate mount hash table\n");
 
-        printk(KERN_INFO "Mount-cache hash table entries: %lu\n", HASH_SIZE);
-
-        for (u = 0; u < HASH_SIZE; u++)
-                INIT_LIST_HEAD(&mount_hashtable[u]);
-        for (u = 0; u < HASH_SIZE; u++)
-                INIT_LIST_HEAD(&mountpoint_hashtable[u]);
+        for (u = 0; u <= m_hash_mask; u++)
+                INIT_HLIST_HEAD(&mount_hashtable[u]);
+        for (u = 0; u <= mp_hash_mask; u++)
+                INIT_HLIST_HEAD(&mountpoint_hashtable[u]);
 
         kernfs_init();
 
diff --git a/fs/pnode.c b/fs/pnode.c
index c7221bb19801..88396df725b4 100644
--- a/fs/pnode.c
+++ b/fs/pnode.c
@@ -220,14 +220,14 @@ static struct mount *get_source(struct mount *dest,
  * @tree_list : list of heads of trees to be attached.
  */
 int propagate_mnt(struct mount *dest_mnt, struct mountpoint *dest_mp,
-                    struct mount *source_mnt, struct list_head *tree_list)
+                    struct mount *source_mnt, struct hlist_head *tree_list)
 {
         struct user_namespace *user_ns = current->nsproxy->mnt_ns->user_ns;
         struct mount *m, *child;
         int ret = 0;
         struct mount *prev_dest_mnt = dest_mnt;
         struct mount *prev_src_mnt  = source_mnt;
-        LIST_HEAD(tmp_list);
+        HLIST_HEAD(tmp_list);
 
         for (m = propagation_next(dest_mnt, dest_mnt); m;
                         m = propagation_next(m, dest_mnt)) {
@@ -246,27 +246,29 @@ int propagate_mnt(struct mount *dest_mnt, struct mountpoint *dest_mp,
                 child = copy_tree(source, source->mnt.mnt_root, type);
                 if (IS_ERR(child)) {
                         ret = PTR_ERR(child);
-                        list_splice(tree_list, tmp_list.prev);
+                        tmp_list = *tree_list;
+                        tmp_list.first->pprev = &tmp_list.first;
+                        INIT_HLIST_HEAD(tree_list);
                         goto out;
                 }
 
                 if (is_subdir(dest_mp->m_dentry, m->mnt.mnt_root)) {
                         mnt_set_mountpoint(m, dest_mp, child);
-                        list_add_tail(&child->mnt_hash, tree_list);
+                        hlist_add_head(&child->mnt_hash, tree_list);
                 } else {
                         /*
                          * This can happen if the parent mount was bind mounted
                          * on some subdirectory of a shared/slave mount.
                          */
-                        list_add_tail(&child->mnt_hash, &tmp_list);
+                        hlist_add_head(&child->mnt_hash, &tmp_list);
                 }
                 prev_dest_mnt = m;
                 prev_src_mnt  = child;
         }
 out:
         lock_mount_hash();
-        while (!list_empty(&tmp_list)) {
-                child = list_first_entry(&tmp_list, struct mount, mnt_hash);
+        while (!hlist_empty(&tmp_list)) {
+                child = hlist_entry(tmp_list.first, struct mount, mnt_hash);
                 umount_tree(child, 0);
         }
         unlock_mount_hash();
@@ -338,8 +340,10 @@ static void __propagate_umount(struct mount *mnt)
                  * umount the child only if the child has no
                  * other children
                  */
-                if (child && list_empty(&child->mnt_mounts))
-                        list_move_tail(&child->mnt_hash, &mnt->mnt_hash);
+                if (child && list_empty(&child->mnt_mounts)) {
+                        hlist_del_init_rcu(&child->mnt_hash);
+                        hlist_add_before_rcu(&child->mnt_hash, &mnt->mnt_hash);
+                }
         }
 }
 
@@ -350,11 +354,11 @@ static void __propagate_umount(struct mount *mnt)
  *
  * vfsmount lock must be held for write
  */
-int propagate_umount(struct list_head *list)
+int propagate_umount(struct hlist_head *list)
 {
         struct mount *mnt;
 
-        list_for_each_entry(mnt, list, mnt_hash)
+        hlist_for_each_entry(mnt, list, mnt_hash)
                 __propagate_umount(mnt);
         return 0;
 }
diff --git a/fs/pnode.h b/fs/pnode.h
index 59e7eda1851e..fc28a27fa892 100644
--- a/fs/pnode.h
+++ b/fs/pnode.h
@@ -36,8 +36,8 @@ static inline void set_mnt_shared(struct mount *mnt)
 
 void change_mnt_propagation(struct mount *, int);
 int propagate_mnt(struct mount *, struct mountpoint *, struct mount *,
-                struct list_head *);
-int propagate_umount(struct list_head *);
+                struct hlist_head *);
+int propagate_umount(struct hlist_head *);
 int propagate_mount_busy(struct mount *, int);
 void mnt_release_group_id(struct mount *);
 int get_dominating_id(struct mount *mnt, const struct path *root);