vfs: simple_set_mnt() should return void

[safe/jmp/linux-2.6] / fs / super.c

diff --git a/fs/super.c b/fs/super.c
index dd23bf9..e512fab 100644 (file)
--- a/fs/super.c
+++ b/fs/super.c
@@ -38,6 +38,7 @@
 #include <linux/kobject.h>
 #include <linux/mutex.h>
 #include <linux/file.h>
+#include <linux/async.h>
 #include <asm/uaccess.h>
 #include "internal.h"
 
@@ -71,6 +72,7 @@ static struct super_block *alloc_super(struct file_system_type *type)
                INIT_HLIST_HEAD(&s->s_anon);
                INIT_LIST_HEAD(&s->s_inodes);
                INIT_LIST_HEAD(&s->s_dentry_lru);
+               INIT_LIST_HEAD(&s->s_async_list);
                init_rwsem(&s->s_umount);
                mutex_init(&s->s_lock);
                lockdep_set_class(&s->s_umount, &type->s_umount_key);
@@ -80,7 +82,22 @@ static struct super_block *alloc_super(struct file_system_type *type)
                 * lock ordering than usbfs:
                 */
                lockdep_set_class(&s->s_lock, &type->s_lock_key);
-               down_write(&s->s_umount);
+               /*
+                * sget() can have s_umount recursion.
+                *
+                * When it cannot find a suitable sb, it allocates a new
+                * one (this one), and tries again to find a suitable old
+                * one.
+                *
+                * In case that succeeds, it will acquire the s_umount
+                * lock of the old one. Since these are clearly distrinct
+                * locks, and this object isn't exposed yet, there's no
+                * risk of deadlocks.
+                *
+                * Annotate this by putting this lock in a different
+                * subclass.
+                */
+               down_write_nested(&s->s_umount, SINGLE_DEPTH_NESTING);
                s->s_count = S_BIAS;
                atomic_set(&s->s_active, 1);
                mutex_init(&s->s_vfs_rename_mutex);
@@ -289,11 +306,18 @@ void generic_shutdown_super(struct super_block *sb)
 {
        const struct super_operations *sop = sb->s_op;
 
+
        if (sb->s_root) {
                shrink_dcache_for_umount(sb);
                fsync_super(sb);
                lock_super(sb);
                sb->s_flags &= ~MS_ACTIVE;
+
+               /*
+                * wait for asynchronous fs operations to finish before going further
+                */
+               async_synchronize_full_domain(&sb->s_async_list);
+
                /* bad name - it should be evict_inodes() */
                invalidate_inodes(sb);
                lock_kernel();
@@ -347,8 +371,10 @@ retry:
                                continue;
                        if (!grab_super(old))
                                goto retry;
-                       if (s)
+                       if (s) {
+                               up_write(&s->s_umount);
                                destroy_super(s);
+                       }
                        return old;
                }
        }
@@ -363,6 +389,7 @@ retry:
        err = set(s, data);
        if (err) {
                spin_unlock(&sb_lock);
+               up_write(&s->s_umount);
                destroy_super(s);
                return ERR_PTR(err);
        }
@@ -461,6 +488,7 @@ restart:
                sb->s_count++;
                spin_unlock(&sb_lock);
                down_read(&sb->s_umount);
+               async_synchronize_full_domain(&sb->s_async_list);
                if (sb->s_root && (wait || sb->s_dirt))
                        sb->s_op->sync_fs(sb, wait);
                up_read(&sb->s_umount);
@@ -534,7 +562,7 @@ rescan:
        return NULL;
 }
 
-asmlinkage long sys_ustat(unsigned dev, struct ustat __user * ubuf)
+SYSCALL_DEFINE2(ustat, unsigned, dev, struct ustat __user *, ubuf)
 {
         struct super_block *s;
         struct ustat tmp;
@@ -682,7 +710,7 @@ void emergency_remount(void)
  * filesystems which don't use real block-devices.  -- jrs
  */
 
-static DEFINE_IDR(unnamed_dev_idr);
+static DEFINE_IDA(unnamed_dev_ida);
 static DEFINE_SPINLOCK(unnamed_dev_lock);/* protects the above */
 
 int set_anon_super(struct super_block *s, void *data)
@@ -691,10 +719,10 @@ int set_anon_super(struct super_block *s, void *data)
        int error;
 
  retry:
-       if (idr_pre_get(&unnamed_dev_idr, GFP_ATOMIC) == 0)
+       if (ida_pre_get(&unnamed_dev_ida, GFP_ATOMIC) == 0)
                return -ENOMEM;
        spin_lock(&unnamed_dev_lock);
-       error = idr_get_new(&unnamed_dev_idr, NULL, &dev);
+       error = ida_get_new(&unnamed_dev_ida, &dev);
        spin_unlock(&unnamed_dev_lock);
        if (error == -EAGAIN)
                /* We raced and lost with another CPU. */
@@ -704,7 +732,7 @@ int set_anon_super(struct super_block *s, void *data)
 
        if ((dev & MAX_ID_MASK) == (1 << MINORBITS)) {
                spin_lock(&unnamed_dev_lock);
-               idr_remove(&unnamed_dev_idr, dev);
+               ida_remove(&unnamed_dev_ida, dev);
                spin_unlock(&unnamed_dev_lock);
                return -EMFILE;
        }
@@ -720,7 +748,7 @@ void kill_anon_super(struct super_block *sb)
 
        generic_shutdown_super(sb);
        spin_lock(&unnamed_dev_lock);
-       idr_remove(&unnamed_dev_idr, slot);
+       ida_remove(&unnamed_dev_ida, slot);
        spin_unlock(&unnamed_dev_lock);
 }
 
@@ -755,9 +783,13 @@ int get_sb_bdev(struct file_system_type *fs_type,
 {
        struct block_device *bdev;
        struct super_block *s;
+       fmode_t mode = FMODE_READ;
        int error = 0;
 
-       bdev = open_bdev_excl(dev_name, flags, fs_type);
+       if (!(flags & MS_RDONLY))
+               mode |= FMODE_WRITE;
+
+       bdev = open_bdev_exclusive(dev_name, mode, fs_type);
        if (IS_ERR(bdev))
                return PTR_ERR(bdev);
 
@@ -780,11 +812,12 @@ int get_sb_bdev(struct file_system_type *fs_type,
                        goto error_bdev;
                }
 
-               close_bdev_excl(bdev);
+               close_bdev_exclusive(bdev, mode);
        } else {
                char b[BDEVNAME_SIZE];
 
                s->s_flags = flags;
+               s->s_mode = mode;
                strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id));
                sb_set_blocksize(s, block_size(bdev));
                error = fill_super(s, data, flags & MS_SILENT ? 1 : 0);
@@ -795,14 +828,16 @@ int get_sb_bdev(struct file_system_type *fs_type,
                }
 
                s->s_flags |= MS_ACTIVE;
+               bdev->bd_super = s;
        }
 
-       return simple_set_mnt(mnt, s);
+       simple_set_mnt(mnt, s);
+       return 0;
 
 error_s:
        error = PTR_ERR(s);
 error_bdev:
-       close_bdev_excl(bdev);
+       close_bdev_exclusive(bdev, mode);
 error:
        return error;
 }
@@ -812,10 +847,12 @@ EXPORT_SYMBOL(get_sb_bdev);
 void kill_block_super(struct super_block *sb)
 {
        struct block_device *bdev = sb->s_bdev;
+       fmode_t mode = sb->s_mode;
 
+       bdev->bd_super = 0;
        generic_shutdown_super(sb);
        sync_blockdev(bdev);
-       close_bdev_excl(bdev);
+       close_bdev_exclusive(bdev, mode);
 }
 
 EXPORT_SYMBOL(kill_block_super);
@@ -841,7 +878,8 @@ int get_sb_nodev(struct file_system_type *fs_type,
                return error;
        }
        s->s_flags |= MS_ACTIVE;
-       return simple_set_mnt(mnt, s);
+       simple_set_mnt(mnt, s);
+       return 0;
 }
 
 EXPORT_SYMBOL(get_sb_nodev);
@@ -873,7 +911,8 @@ int get_sb_single(struct file_system_type *fs_type,
                s->s_flags |= MS_ACTIVE;
        }
        do_remount_sb(s, flags, data, 0);
-       return simple_set_mnt(mnt, s);
+       simple_set_mnt(mnt, s);
+       return 0;
 }
 
 EXPORT_SYMBOL(get_sb_single);
@@ -908,7 +947,7 @@ vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void
                goto out_free_secdata;
        BUG_ON(!mnt->mnt_sb);
 
-       error = security_sb_kern_mount(mnt->mnt_sb, secdata);
+       error = security_sb_kern_mount(mnt->mnt_sb, flags, secdata);
        if (error)
                goto out_sb;