move virtrng_remove to .devexit.text
[linux-2.6.git] / fs / block_dev.c
index 67fe72c..9cf4b92 100644 (file)
 #include <linux/kmod.h>
 #include <linux/major.h>
 #include <linux/smp_lock.h>
+#include <linux/device_cgroup.h>
 #include <linux/highmem.h>
 #include <linux/blkdev.h>
 #include <linux/module.h>
 #include <linux/blkpg.h>
 #include <linux/buffer_head.h>
+#include <linux/pagevec.h>
 #include <linux/writeback.h>
 #include <linux/mpage.h>
 #include <linux/mount.h>
 #include <linux/uio.h>
 #include <linux/namei.h>
 #include <linux/log2.h>
+#include <linux/kmemleak.h>
 #include <asm/uaccess.h>
 #include "internal.h"
 
@@ -31,6 +34,8 @@ struct bdev_inode {
        struct inode vfs_inode;
 };
 
+static const struct address_space_operations def_blk_aops;
+
 static inline struct bdev_inode *BDEV_I(struct inode *inode)
 {
        return container_of(inode, struct bdev_inode, vfs_inode);
@@ -72,7 +77,7 @@ int set_blocksize(struct block_device *bdev, int size)
                return -EINVAL;
 
        /* Size cannot be smaller than the size supported by the device */
-       if (size < bdev_hardsect_size(bdev))
+       if (size < bdev_logical_block_size(bdev))
                return -EINVAL;
 
        /* Don't change the size if it is same as current */
@@ -102,7 +107,7 @@ EXPORT_SYMBOL(sb_set_blocksize);
 
 int sb_min_blocksize(struct super_block *sb, int size)
 {
-       int minsize = bdev_hardsect_size(sb->s_bdev);
+       int minsize = bdev_logical_block_size(sb->s_bdev);
        if (size < minsize)
                size = minsize;
        return sb_set_blocksize(sb, size);
@@ -171,202 +176,163 @@ blkdev_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov,
                                iov, offset, nr_segs, blkdev_get_blocks, NULL);
 }
 
-#if 0
-static void blk_end_aio(struct bio *bio, int error)
+int __sync_blockdev(struct block_device *bdev, int wait)
 {
-       struct kiocb *iocb = bio->bi_private;
-       atomic_t *bio_count = &iocb->ki_bio_count;
-
-       if (bio_data_dir(bio) == READ)
-               bio_check_pages_dirty(bio);
-       else {
-               bio_release_pages(bio);
-               bio_put(bio);
-       }
-
-       /* iocb->ki_nbytes stores error code from LLDD */
-       if (error)
-               iocb->ki_nbytes = -EIO;
-
-       if (atomic_dec_and_test(bio_count)) {
-               if ((long)iocb->ki_nbytes < 0)
-                       aio_complete(iocb, iocb->ki_nbytes, 0);
-               else
-                       aio_complete(iocb, iocb->ki_left, 0);
-       }
-
-       return 0;
+       if (!bdev)
+               return 0;
+       if (!wait)
+               return filemap_flush(bdev->bd_inode->i_mapping);
+       return filemap_write_and_wait(bdev->bd_inode->i_mapping);
 }
 
-#define VEC_SIZE       16
-struct pvec {
-       unsigned short nr;
-       unsigned short idx;
-       struct page *page[VEC_SIZE];
-};
-
-#define PAGES_SPANNED(addr, len)       \
-       (DIV_ROUND_UP((addr) + (len), PAGE_SIZE) - (addr) / PAGE_SIZE);
+/*
+ * Write out and wait upon all the dirty data associated with a block
+ * device via its mapping.  Does not take the superblock lock.
+ */
+int sync_blockdev(struct block_device *bdev)
+{
+       return __sync_blockdev(bdev, 1);
+}
+EXPORT_SYMBOL(sync_blockdev);
 
 /*
- * get page pointer for user addr, we internally cache struct page array for
- * (addr, count) range in pvec to avoid frequent call to get_user_pages.  If
- * internal page list is exhausted, a batch count of up to VEC_SIZE is used
- * to get next set of page struct.
+ * Write out and wait upon all dirty data associated with this
+ * device.   Filesystem data as well as the underlying block
+ * device.  Takes the superblock lock.
  */
-static struct page *blk_get_page(unsigned long addr, size_t count, int rw,
-                                struct pvec *pvec)
-{
-       int ret, nr_pages;
-       if (pvec->idx == pvec->nr) {
-               nr_pages = PAGES_SPANNED(addr, count);
-               nr_pages = min(nr_pages, VEC_SIZE);
-               down_read(&current->mm->mmap_sem);
-               ret = get_user_pages(current, current->mm, addr, nr_pages,
-                                    rw == READ, 0, pvec->page, NULL);
-               up_read(&current->mm->mmap_sem);
-               if (ret < 0)
-                       return ERR_PTR(ret);
-               pvec->nr = ret;
-               pvec->idx = 0;
+int fsync_bdev(struct block_device *bdev)
+{
+       struct super_block *sb = get_super(bdev);
+       if (sb) {
+               int res = sync_filesystem(sb);
+               drop_super(sb);
+               return res;
        }
-       return pvec->page[pvec->idx++];
+       return sync_blockdev(bdev);
 }
+EXPORT_SYMBOL(fsync_bdev);
 
-/* return a page back to pvec array */
-static void blk_unget_page(struct page *page, struct pvec *pvec)
+/**
+ * freeze_bdev  --  lock a filesystem and force it into a consistent state
+ * @bdev:      blockdevice to lock
+ *
+ * If a superblock is found on this device, we take the s_umount semaphore
+ * on it to make sure nobody unmounts until the snapshot creation is done.
+ * The reference counter (bd_fsfreeze_count) guarantees that only the last
+ * unfreeze process can unfreeze the frozen filesystem actually when multiple
+ * freeze requests arrive simultaneously. It counts up in freeze_bdev() and
+ * count down in thaw_bdev(). When it becomes 0, thaw_bdev() will unfreeze
+ * actually.
+ */
+struct super_block *freeze_bdev(struct block_device *bdev)
 {
-       pvec->page[--pvec->idx] = page;
-}
+       struct super_block *sb;
+       int error = 0;
 
-static ssize_t
-blkdev_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov,
-                loff_t pos, unsigned long nr_segs)
-{
-       struct inode *inode = iocb->ki_filp->f_mapping->host;
-       unsigned blkbits = blksize_bits(bdev_hardsect_size(I_BDEV(inode)));
-       unsigned blocksize_mask = (1 << blkbits) - 1;
-       unsigned long seg = 0;  /* iov segment iterator */
-       unsigned long nvec;     /* number of bio vec needed */
-       unsigned long cur_off;  /* offset into current page */
-       unsigned long cur_len;  /* I/O len of current page, up to PAGE_SIZE */
-
-       unsigned long addr;     /* user iovec address */
-       size_t count;           /* user iovec len */
-       size_t nbytes = iocb->ki_nbytes = iocb->ki_left; /* total xfer size */
-       loff_t size;            /* size of block device */
-       struct bio *bio;
-       atomic_t *bio_count = &iocb->ki_bio_count;
-       struct page *page;
-       struct pvec pvec;
-
-       pvec.nr = 0;
-       pvec.idx = 0;
-
-       if (pos & blocksize_mask)
-               return -EINVAL;
+       mutex_lock(&bdev->bd_fsfreeze_mutex);
+       if (++bdev->bd_fsfreeze_count > 1) {
+               /*
+                * We don't even need to grab a reference - the first call
+                * to freeze_bdev grab an active reference and only the last
+                * thaw_bdev drops it.
+                */
+               sb = get_super(bdev);
+               drop_super(sb);
+               mutex_unlock(&bdev->bd_fsfreeze_mutex);
+               return sb;
+       }
 
-       size = i_size_read(inode);
-       if (pos + nbytes > size) {
-               nbytes = size - pos;
-               iocb->ki_left = nbytes;
+       sb = get_active_super(bdev);
+       if (!sb)
+               goto out;
+       if (sb->s_flags & MS_RDONLY) {
+               deactivate_locked_super(sb);
+               mutex_unlock(&bdev->bd_fsfreeze_mutex);
+               return sb;
        }
 
-       /*
-        * check first non-zero iov alignment, the remaining
-        * iov alignment is checked inside bio loop below.
-        */
-       do {
-               addr = (unsigned long) iov[seg].iov_base;
-               count = min(iov[seg].iov_len, nbytes);
-               if (addr & blocksize_mask || count & blocksize_mask)
-                       return -EINVAL;
-       } while (!count && ++seg < nr_segs);
-       atomic_set(bio_count, 1);
-
-       while (nbytes) {
-               /* roughly estimate number of bio vec needed */
-               nvec = (nbytes + PAGE_SIZE - 1) / PAGE_SIZE;
-               nvec = max(nvec, nr_segs - seg);
-               nvec = min(nvec, (unsigned long) BIO_MAX_PAGES);
-
-               /* bio_alloc should not fail with GFP_KERNEL flag */
-               bio = bio_alloc(GFP_KERNEL, nvec);
-               bio->bi_bdev = I_BDEV(inode);
-               bio->bi_end_io = blk_end_aio;
-               bio->bi_private = iocb;
-               bio->bi_sector = pos >> blkbits;
-same_bio:
-               cur_off = addr & ~PAGE_MASK;
-               cur_len = PAGE_SIZE - cur_off;
-               if (count < cur_len)
-                       cur_len = count;
-
-               page = blk_get_page(addr, count, rw, &pvec);
-               if (unlikely(IS_ERR(page)))
-                       goto backout;
-
-               if (bio_add_page(bio, page, cur_len, cur_off)) {
-                       pos += cur_len;
-                       addr += cur_len;
-                       count -= cur_len;
-                       nbytes -= cur_len;
-
-                       if (count)
-                               goto same_bio;
-                       while (++seg < nr_segs) {
-                               addr = (unsigned long) iov[seg].iov_base;
-                               count = iov[seg].iov_len;
-                               if (!count)
-                                       continue;
-                               if (unlikely(addr & blocksize_mask ||
-                                            count & blocksize_mask)) {
-                                       page = ERR_PTR(-EINVAL);
-                                       goto backout;
-                               }
-                               count = min(count, nbytes);
-                               goto same_bio;
-                       }
-               } else {
-                       blk_unget_page(page, &pvec);
-               }
+       sb->s_frozen = SB_FREEZE_WRITE;
+       smp_wmb();
 
-               /* bio is ready, submit it */
-               if (rw == READ)
-                       bio_set_pages_dirty(bio);
-               atomic_inc(bio_count);
-               submit_bio(rw, bio);
-       }
+       sync_filesystem(sb);
 
-completion:
-       iocb->ki_left -= nbytes;
-       nbytes = iocb->ki_left;
-       iocb->ki_pos += nbytes;
+       sb->s_frozen = SB_FREEZE_TRANS;
+       smp_wmb();
 
-       blk_run_address_space(inode->i_mapping);
-       if (atomic_dec_and_test(bio_count))
-               aio_complete(iocb, nbytes, 0);
+       sync_blockdev(sb->s_bdev);
 
-       return -EIOCBQUEUED;
+       if (sb->s_op->freeze_fs) {
+               error = sb->s_op->freeze_fs(sb);
+               if (error) {
+                       printk(KERN_ERR
+                               "VFS:Filesystem freeze failed\n");
+                       sb->s_frozen = SB_UNFROZEN;
+                       deactivate_locked_super(sb);
+                       bdev->bd_fsfreeze_count--;
+                       mutex_unlock(&bdev->bd_fsfreeze_mutex);
+                       return ERR_PTR(error);
+               }
+       }
+       up_write(&sb->s_umount);
 
-backout:
-       /*
-        * back out nbytes count constructed so far for this bio,
-        * we will throw away current bio.
-        */
-       nbytes += bio->bi_size;
-       bio_release_pages(bio);
-       bio_put(bio);
+ out:
+       sync_blockdev(bdev);
+       mutex_unlock(&bdev->bd_fsfreeze_mutex);
+       return sb;      /* thaw_bdev releases s->s_umount */
+}
+EXPORT_SYMBOL(freeze_bdev);
 
-       /*
-        * if no bio was submmitted, return the error code.
-        * otherwise, proceed with pending I/O completion.
-        */
-       if (atomic_read(bio_count) == 1)
-               return PTR_ERR(page);
-       goto completion;
+/**
+ * thaw_bdev  -- unlock filesystem
+ * @bdev:      blockdevice to unlock
+ * @sb:                associated superblock
+ *
+ * Unlocks the filesystem and marks it writeable again after freeze_bdev().
+ */
+int thaw_bdev(struct block_device *bdev, struct super_block *sb)
+{
+       int error = -EINVAL;
+
+       mutex_lock(&bdev->bd_fsfreeze_mutex);
+       if (!bdev->bd_fsfreeze_count)
+               goto out_unlock;
+
+       error = 0;
+       if (--bdev->bd_fsfreeze_count > 0)
+               goto out_unlock;
+
+       if (!sb)
+               goto out_unlock;
+
+       BUG_ON(sb->s_bdev != bdev);
+       down_write(&sb->s_umount);
+       if (sb->s_flags & MS_RDONLY)
+               goto out_deactivate;
+
+       if (sb->s_op->unfreeze_fs) {
+               error = sb->s_op->unfreeze_fs(sb);
+               if (error) {
+                       printk(KERN_ERR
+                               "VFS:Filesystem thaw failed\n");
+                       sb->s_frozen = SB_FREEZE_TRANS;
+                       bdev->bd_fsfreeze_count++;
+                       mutex_unlock(&bdev->bd_fsfreeze_mutex);
+                       return error;
+               }
+       }
+
+       sb->s_frozen = SB_UNFROZEN;
+       smp_wmb();
+       wake_up(&sb->s_wait_unfrozen);
+
+out_deactivate:
+       if (sb)
+               deactivate_locked_super(sb);
+out_unlock:
+       mutex_unlock(&bdev->bd_fsfreeze_mutex);
+       return 0;
 }
-#endif
+EXPORT_SYMBOL(thaw_bdev);
 
 static int blkdev_writepage(struct page *page, struct writeback_control *wbc)
 {
@@ -461,24 +427,24 @@ static void bdev_destroy_inode(struct inode *inode)
 {
        struct bdev_inode *bdi = BDEV_I(inode);
 
-       bdi->bdev.bd_inode_backing_dev_info = NULL;
        kmem_cache_free(bdev_cachep, bdi);
 }
 
-static void init_once(struct kmem_cache * cachep, void *foo)
+static void init_once(void *foo)
 {
        struct bdev_inode *ei = (struct bdev_inode *) foo;
        struct block_device *bdev = &ei->bdev;
 
        memset(bdev, 0, sizeof(*bdev));
        mutex_init(&bdev->bd_mutex);
-       sema_init(&bdev->bd_mount_sem, 1);
        INIT_LIST_HEAD(&bdev->bd_inodes);
        INIT_LIST_HEAD(&bdev->bd_list);
 #ifdef CONFIG_SYSFS
        INIT_LIST_HEAD(&bdev->bd_holder_list);
 #endif
        inode_init_once(&ei->vfs_inode);
+       /* Initialize mutex for freeze. */
+       mutex_init(&bdev->bd_fsfreeze_mutex);
 }
 
 static inline void __bd_forget(struct inode *inode)
@@ -520,12 +486,13 @@ static struct file_system_type bd_type = {
        .kill_sb        = kill_anon_super,
 };
 
-static struct vfsmount *bd_mnt __read_mostly;
-struct super_block *blockdev_superblock;
+struct super_block *blockdev_superblock __read_mostly;
 
 void __init bdev_cache_init(void)
 {
        int err;
+       struct vfsmount *bd_mnt;
+
        bdev_cachep = kmem_cache_create("bdev_cache", sizeof(struct bdev_inode),
                        0, (SLAB_HWCACHE_ALIGN|SLAB_RECLAIM_ACCOUNT|
                                SLAB_MEM_SPREAD|SLAB_PANIC),
@@ -536,6 +503,11 @@ void __init bdev_cache_init(void)
        bd_mnt = kern_mount(&bd_type);
        if (IS_ERR(bd_mnt))
                panic("Cannot create bdev pseudo-fs");
+       /*
+        * This vfsmount structure is only used to obtain the
+        * blockdev_superblock, so tell kmemleak not to report it.
+        */
+       kmemleak_not_leak(bd_mnt);
        blockdev_superblock = bd_mnt->mnt_sb;   /* For writeback */
 }
 
@@ -567,7 +539,7 @@ struct block_device *bdget(dev_t dev)
        struct block_device *bdev;
        struct inode *inode;
 
-       inode = iget5_locked(bd_mnt->mnt_sb, hash(dev),
+       inode = iget5_locked(blockdev_superblock, hash(dev),
                        bdev_test, bdev_set, &dev);
 
        if (!inode)
@@ -597,6 +569,16 @@ struct block_device *bdget(dev_t dev)
 
 EXPORT_SYMBOL(bdget);
 
+/**
+ * bdgrab -- Grab a reference to an already referenced block device
+ * @bdev:      Block device to grab a reference to.
+ */
+struct block_device *bdgrab(struct block_device *bdev)
+{
+       atomic_inc(&bdev->bd_inode->i_count);
+       return bdev;
+}
+
 long nr_blockdev_pages(void)
 {
        struct block_device *bdev;
@@ -657,7 +639,7 @@ void bd_forget(struct inode *inode)
 
        spin_lock(&bdev_lock);
        if (inode->i_bdev) {
-               if (inode->i_sb != blockdev_superblock)
+               if (!sb_is_blkdev_sb(inode->i_sb))
                        bdev = inode->i_bdev;
                __bd_forget(inode);
        }
@@ -734,22 +716,6 @@ EXPORT_SYMBOL(bd_release);
  *           /sys/block/sda/holders/dm-0 --> /sys/block/dm-0
  */
 
-static struct kobject *bdev_get_kobj(struct block_device *bdev)
-{
-       if (bdev->bd_contains != bdev)
-               return kobject_get(&bdev->bd_part->dev.kobj);
-       else
-               return kobject_get(&bdev->bd_disk->dev.kobj);
-}
-
-static struct kobject *bdev_get_holder(struct block_device *bdev)
-{
-       if (bdev->bd_contains != bdev)
-               return kobject_get(bdev->bd_part->holder_dir);
-       else
-               return kobject_get(bdev->bd_disk->holder_dir);
-}
-
 static int add_symlink(struct kobject *from, struct kobject *to)
 {
        if (!from || !to)
@@ -798,11 +764,11 @@ static int bd_holder_grab_dirs(struct block_device *bdev,
        if (!bo->hdev)
                goto fail_put_sdir;
 
-       bo->sdev = bdev_get_kobj(bdev);
+       bo->sdev = kobject_get(&part_to_dev(bdev->bd_part)->kobj);
        if (!bo->sdev)
                goto fail_put_hdev;
 
-       bo->hdir = bdev_get_holder(bdev);
+       bo->hdir = kobject_get(bdev->bd_part->holder_dir);
        if (!bo->hdir)
                goto fail_put_sdev;
 
@@ -1050,18 +1016,98 @@ EXPORT_SYMBOL_GPL(bd_release_from_disk);
  * to be used for internal purposes.  If you ever need it - reconsider
  * your API.
  */
-struct block_device *open_by_devnum(dev_t dev, unsigned mode)
+struct block_device *open_by_devnum(dev_t dev, fmode_t mode)
 {
        struct block_device *bdev = bdget(dev);
        int err = -ENOMEM;
-       int flags = mode & FMODE_WRITE ? O_RDWR : O_RDONLY;
        if (bdev)
-               err = blkdev_get(bdev, mode, flags);
+               err = blkdev_get(bdev, mode);
        return err ? ERR_PTR(err) : bdev;
 }
 
 EXPORT_SYMBOL(open_by_devnum);
 
+/**
+ * flush_disk - invalidates all buffer-cache entries on a disk
+ *
+ * @bdev:      struct block device to be flushed
+ *
+ * Invalidates all buffer-cache entries on a disk. It should be called
+ * when a disk has been changed -- either by a media change or online
+ * resize.
+ */
+static void flush_disk(struct block_device *bdev)
+{
+       if (__invalidate_device(bdev)) {
+               char name[BDEVNAME_SIZE] = "";
+
+               if (bdev->bd_disk)
+                       disk_name(bdev->bd_disk, 0, name);
+               printk(KERN_WARNING "VFS: busy inodes on changed media or "
+                      "resized disk %s\n", name);
+       }
+
+       if (!bdev->bd_disk)
+               return;
+       if (disk_partitionable(bdev->bd_disk))
+               bdev->bd_invalidated = 1;
+}
+
+/**
+ * check_disk_size_change - checks for disk size change and adjusts bdev size.
+ * @disk: struct gendisk to check
+ * @bdev: struct bdev to adjust.
+ *
+ * This routine checks to see if the bdev size does not match the disk size
+ * and adjusts it if it differs.
+ */
+void check_disk_size_change(struct gendisk *disk, struct block_device *bdev)
+{
+       loff_t disk_size, bdev_size;
+
+       disk_size = (loff_t)get_capacity(disk) << 9;
+       bdev_size = i_size_read(bdev->bd_inode);
+       if (disk_size != bdev_size) {
+               char name[BDEVNAME_SIZE];
+
+               disk_name(disk, 0, name);
+               printk(KERN_INFO
+                      "%s: detected capacity change from %lld to %lld\n",
+                      name, bdev_size, disk_size);
+               i_size_write(bdev->bd_inode, disk_size);
+               flush_disk(bdev);
+       }
+}
+EXPORT_SYMBOL(check_disk_size_change);
+
+/**
+ * revalidate_disk - wrapper for lower-level driver's revalidate_disk call-back
+ * @disk: struct gendisk to be revalidated
+ *
+ * This routine is a wrapper for lower-level driver's revalidate_disk
+ * call-backs.  It is used to do common pre and post operations needed
+ * for all revalidate_disk operations.
+ */
+int revalidate_disk(struct gendisk *disk)
+{
+       struct block_device *bdev;
+       int ret = 0;
+
+       if (disk->fops->revalidate_disk)
+               ret = disk->fops->revalidate_disk(disk);
+
+       bdev = bdget_disk(disk, 0);
+       if (!bdev)
+               return ret;
+
+       mutex_lock(&bdev->bd_mutex);
+       check_disk_size_change(disk, bdev);
+       mutex_unlock(&bdev->bd_mutex);
+       bdput(bdev);
+       return ret;
+}
+EXPORT_SYMBOL(revalidate_disk);
+
 /*
  * This routine checks whether a removable media has been changed,
  * and invalidates all buffer-cache-entries in that case. This
@@ -1074,20 +1120,16 @@ EXPORT_SYMBOL(open_by_devnum);
 int check_disk_change(struct block_device *bdev)
 {
        struct gendisk *disk = bdev->bd_disk;
-       struct block_device_operations * bdops = disk->fops;
+       const struct block_device_operations *bdops = disk->fops;
 
        if (!bdops->media_changed)
                return 0;
        if (!bdops->media_changed(bdev->bd_disk))
                return 0;
 
-       if (__invalidate_device(bdev))
-               printk("VFS: busy inodes on changed media.\n");
-
+       flush_disk(bdev);
        if (bdops->revalidate_disk)
                bdops->revalidate_disk(bdev->bd_disk);
-       if (bdev->bd_disk->minors > 1)
-               bdev->bd_invalidated = 1;
        return 1;
 }
 
@@ -1095,7 +1137,7 @@ EXPORT_SYMBOL(check_disk_change);
 
 void bd_set_size(struct block_device *bdev, loff_t size)
 {
-       unsigned bsize = bdev_hardsect_size(bdev);
+       unsigned bsize = bdev_logical_block_size(bdev);
 
        bdev->bd_inode->i_size = size;
        while (bsize < PAGE_CACHE_SIZE) {
@@ -1108,9 +1150,7 @@ void bd_set_size(struct block_device *bdev, loff_t size)
 }
 EXPORT_SYMBOL(bd_set_size);
 
-static int __blkdev_get(struct block_device *bdev, mode_t mode, unsigned flags,
-                       int for_part);
-static int __blkdev_put(struct block_device *bdev, int for_part);
+static int __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part);
 
 /*
  * bd_mutex locking:
@@ -1119,33 +1159,63 @@ static int __blkdev_put(struct block_device *bdev, int for_part);
  *    mutex_lock_nested(whole->bd_mutex, 1)
  */
 
-static int do_open(struct block_device *bdev, struct file *file, int for_part)
+static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 {
-       struct module *owner = NULL;
        struct gendisk *disk;
-       int ret = -ENXIO;
-       int part;
+       int ret;
+       int partno;
+       int perm = 0;
 
-       file->f_mapping = bdev->bd_inode->i_mapping;
-       lock_kernel();
-       disk = get_gendisk(bdev->bd_dev, &part);
-       if (!disk) {
-               unlock_kernel();
+       if (mode & FMODE_READ)
+               perm |= MAY_READ;
+       if (mode & FMODE_WRITE)
+               perm |= MAY_WRITE;
+       /*
+        * hooks: /n/, see "layering violations".
+        */
+       ret = devcgroup_inode_permission(bdev->bd_inode, perm);
+       if (ret != 0) {
                bdput(bdev);
                return ret;
        }
-       owner = disk->fops->owner;
+
+       lock_kernel();
+ restart:
+
+       ret = -ENXIO;
+       disk = get_gendisk(bdev->bd_dev, &partno);
+       if (!disk)
+               goto out_unlock_kernel;
 
        mutex_lock_nested(&bdev->bd_mutex, for_part);
        if (!bdev->bd_openers) {
                bdev->bd_disk = disk;
                bdev->bd_contains = bdev;
-               if (!part) {
+               if (!partno) {
                        struct backing_dev_info *bdi;
+
+                       ret = -ENXIO;
+                       bdev->bd_part = disk_get_part(disk, partno);
+                       if (!bdev->bd_part)
+                               goto out_clear;
+
                        if (disk->fops->open) {
-                               ret = disk->fops->open(bdev->bd_inode, file);
+                               ret = disk->fops->open(bdev, mode);
+                               if (ret == -ERESTARTSYS) {
+                                       /* Lost a race with 'disk' being
+                                        * deleted, try again.
+                                        * See md.c
+                                        */
+                                       disk_put_part(bdev->bd_part);
+                                       bdev->bd_part = NULL;
+                                       module_put(disk->fops->owner);
+                                       put_disk(disk);
+                                       bdev->bd_disk = NULL;
+                                       mutex_unlock(&bdev->bd_mutex);
+                                       goto restart;
+                               }
                                if (ret)
-                                       goto out_first;
+                                       goto out_clear;
                        }
                        if (!bdev->bd_openers) {
                                bd_set_size(bdev,(loff_t)get_capacity(disk)<<9);
@@ -1157,36 +1227,35 @@ static int do_open(struct block_device *bdev, struct file *file, int for_part)
                        if (bdev->bd_invalidated)
                                rescan_partitions(disk, bdev);
                } else {
-                       struct hd_struct *p;
                        struct block_device *whole;
                        whole = bdget_disk(disk, 0);
                        ret = -ENOMEM;
                        if (!whole)
-                               goto out_first;
+                               goto out_clear;
                        BUG_ON(for_part);
-                       ret = __blkdev_get(whole, file->f_mode, file->f_flags, 1);
+                       ret = __blkdev_get(whole, mode, 1);
                        if (ret)
-                               goto out_first;
+                               goto out_clear;
                        bdev->bd_contains = whole;
-                       p = disk->part[part - 1];
                        bdev->bd_inode->i_data.backing_dev_info =
                           whole->bd_inode->i_data.backing_dev_info;
-                       if (!(disk->flags & GENHD_FL_UP) || !p || !p->nr_sects) {
+                       bdev->bd_part = disk_get_part(disk, partno);
+                       if (!(disk->flags & GENHD_FL_UP) ||
+                           !bdev->bd_part || !bdev->bd_part->nr_sects) {
                                ret = -ENXIO;
-                               goto out_first;
+                               goto out_clear;
                        }
-                       kobject_get(&p->dev.kobj);
-                       bdev->bd_part = p;
-                       bd_set_size(bdev, (loff_t) p->nr_sects << 9);
+                       bd_set_size(bdev, (loff_t)bdev->bd_part->nr_sects << 9);
                }
        } else {
                put_disk(disk);
-               module_put(owner);
+               module_put(disk->fops->owner);
+               disk = NULL;
                if (bdev->bd_contains == bdev) {
                        if (bdev->bd_disk->fops->open) {
-                               ret = bdev->bd_disk->fops->open(bdev->bd_inode, file);
+                               ret = bdev->bd_disk->fops->open(bdev, mode);
                                if (ret)
-                                       goto out;
+                                       goto out_unlock_bdev;
                        }
                        if (bdev->bd_invalidated)
                                rescan_partitions(bdev->bd_disk, bdev);
@@ -1199,44 +1268,30 @@ static int do_open(struct block_device *bdev, struct file *file, int for_part)
        unlock_kernel();
        return 0;
 
-out_first:
+ out_clear:
+       disk_put_part(bdev->bd_part);
        bdev->bd_disk = NULL;
+       bdev->bd_part = NULL;
        bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info;
        if (bdev != bdev->bd_contains)
-               __blkdev_put(bdev->bd_contains, 1);
+               __blkdev_put(bdev->bd_contains, mode, 1);
        bdev->bd_contains = NULL;
-       put_disk(disk);
-       module_put(owner);
-out:
+ out_unlock_bdev:
        mutex_unlock(&bdev->bd_mutex);
+ out_unlock_kernel:
        unlock_kernel();
-       if (ret)
-               bdput(bdev);
-       return ret;
-}
 
-static int __blkdev_get(struct block_device *bdev, mode_t mode, unsigned flags,
-                       int for_part)
-{
-       /*
-        * This crockload is due to bad choice of ->open() type.
-        * It will go away.
-        * For now, block device ->open() routine must _not_
-        * examine anything in 'inode' argument except ->i_rdev.
-        */
-       struct file fake_file = {};
-       struct dentry fake_dentry = {};
-       fake_file.f_mode = mode;
-       fake_file.f_flags = flags;
-       fake_file.f_path.dentry = &fake_dentry;
-       fake_dentry.d_inode = bdev->bd_inode;
+       if (disk)
+               module_put(disk->fops->owner);
+       put_disk(disk);
+       bdput(bdev);
 
-       return do_open(bdev, &fake_file, for_part);
+       return ret;
 }
 
-int blkdev_get(struct block_device *bdev, mode_t mode, unsigned flags)
+int blkdev_get(struct block_device *bdev, fmode_t mode)
 {
-       return __blkdev_get(bdev, mode, flags, 0);
+       return __blkdev_get(bdev, mode, 0);
 }
 EXPORT_SYMBOL(blkdev_get);
 
@@ -1253,28 +1308,39 @@ static int blkdev_open(struct inode * inode, struct file * filp)
         */
        filp->f_flags |= O_LARGEFILE;
 
+       if (filp->f_flags & O_NDELAY)
+               filp->f_mode |= FMODE_NDELAY;
+       if (filp->f_flags & O_EXCL)
+               filp->f_mode |= FMODE_EXCL;
+       if ((filp->f_flags & O_ACCMODE) == 3)
+               filp->f_mode |= FMODE_WRITE_IOCTL;
+
        bdev = bd_acquire(inode);
        if (bdev == NULL)
                return -ENOMEM;
 
-       res = do_open(bdev, filp, 0);
+       filp->f_mapping = bdev->bd_inode->i_mapping;
+
+       res = blkdev_get(bdev, filp->f_mode);
        if (res)
                return res;
 
-       if (!(filp->f_flags & O_EXCL) )
-               return 0;
+       if (filp->f_mode & FMODE_EXCL) {
+               res = bd_claim(bdev, filp);
+               if (res)
+                       goto out_blkdev_put;
+       }
 
-       if (!(res = bd_claim(bdev, filp)))
-               return 0;
+       return 0;
 
-       blkdev_put(bdev);
+ out_blkdev_put:
+       blkdev_put(bdev, filp->f_mode);
        return res;
 }
 
-static int __blkdev_put(struct block_device *bdev, int for_part)
+static int __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part)
 {
        int ret = 0;
-       struct inode *bd_inode = bdev->bd_inode;
        struct gendisk *disk = bdev->bd_disk;
        struct block_device *victim = NULL;
 
@@ -1289,18 +1355,15 @@ static int __blkdev_put(struct block_device *bdev, int for_part)
        }
        if (bdev->bd_contains == bdev) {
                if (disk->fops->release)
-                       ret = disk->fops->release(bd_inode, NULL);
+                       ret = disk->fops->release(disk, mode);
        }
        if (!bdev->bd_openers) {
                struct module *owner = disk->fops->owner;
 
                put_disk(disk);
                module_put(owner);
-
-               if (bdev->bd_contains != bdev) {
-                       kobject_put(&bdev->bd_part->dev.kobj);
-                       bdev->bd_part = NULL;
-               }
+               disk_put_part(bdev->bd_part);
+               bdev->bd_part = NULL;
                bdev->bd_disk = NULL;
                bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info;
                if (bdev != bdev->bd_contains)
@@ -1311,13 +1374,13 @@ static int __blkdev_put(struct block_device *bdev, int for_part)
        mutex_unlock(&bdev->bd_mutex);
        bdput(bdev);
        if (victim)
-               __blkdev_put(victim, 1);
+               __blkdev_put(victim, mode, 1);
        return ret;
 }
 
-int blkdev_put(struct block_device *bdev)
+int blkdev_put(struct block_device *bdev, fmode_t mode)
 {
-       return __blkdev_put(bdev, 0);
+       return __blkdev_put(bdev, mode, 0);
 }
 EXPORT_SYMBOL(blkdev_put);
 
@@ -1326,21 +1389,75 @@ static int blkdev_close(struct inode * inode, struct file * filp)
        struct block_device *bdev = I_BDEV(filp->f_mapping->host);
        if (bdev->bd_holder == filp)
                bd_release(bdev);
-       return blkdev_put(bdev);
+       return blkdev_put(bdev, filp->f_mode);
 }
 
 static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 {
-       return blkdev_ioctl(file->f_mapping->host, file, cmd, arg);
+       struct block_device *bdev = I_BDEV(file->f_mapping->host);
+       fmode_t mode = file->f_mode;
+
+       /*
+        * O_NDELAY can be altered using fcntl(.., F_SETFL, ..), so we have
+        * to updated it before every ioctl.
+        */
+       if (file->f_flags & O_NDELAY)
+               mode |= FMODE_NDELAY;
+       else
+               mode &= ~FMODE_NDELAY;
+
+       return blkdev_ioctl(bdev, mode, cmd, arg);
+}
+
+/*
+ * Write data to the block device.  Only intended for the block device itself
+ * and the raw driver which basically is a fake block device.
+ *
+ * Does not take i_mutex for the write and thus is not for general purpose
+ * use.
+ */
+ssize_t blkdev_aio_write(struct kiocb *iocb, const struct iovec *iov,
+                        unsigned long nr_segs, loff_t pos)
+{
+       struct file *file = iocb->ki_filp;
+       ssize_t ret;
+
+       BUG_ON(iocb->ki_pos != pos);
+
+       ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos);
+       if (ret > 0 || ret == -EIOCBQUEUED) {
+               ssize_t err;
+
+               err = generic_write_sync(file, pos, ret);
+               if (err < 0 && ret > 0)
+                       ret = err;
+       }
+       return ret;
+}
+EXPORT_SYMBOL_GPL(blkdev_aio_write);
+
+/*
+ * Try to release a page associated with block device when the system
+ * is under memory pressure.
+ */
+static int blkdev_releasepage(struct page *page, gfp_t wait)
+{
+       struct super_block *super = BDEV_I(page->mapping->host)->bdev.bd_super;
+
+       if (super && super->s_op->bdev_try_to_free_page)
+               return super->s_op->bdev_try_to_free_page(super, page, wait);
+
+       return try_to_free_buffers(page);
 }
 
-const struct address_space_operations def_blk_aops = {
+static const struct address_space_operations def_blk_aops = {
        .readpage       = blkdev_readpage,
        .writepage      = blkdev_writepage,
        .sync_page      = block_sync_page,
        .write_begin    = blkdev_write_begin,
        .write_end      = blkdev_write_end,
        .writepages     = generic_writepages,
+       .releasepage    = blkdev_releasepage,
        .direct_IO      = blkdev_direct_IO,
 };
 
@@ -1351,7 +1468,7 @@ const struct file_operations def_blk_fops = {
        .read           = do_sync_read,
        .write          = do_sync_write,
        .aio_read       = generic_file_aio_read,
-       .aio_write      = generic_file_aio_write_nolock,
+       .aio_write      = blkdev_aio_write,
        .mmap           = generic_file_mmap,
        .fsync          = block_fsync,
        .unlocked_ioctl = block_ioctl,
@@ -1367,7 +1484,7 @@ int ioctl_by_bdev(struct block_device *bdev, unsigned cmd, unsigned long arg)
        int res;
        mm_segment_t old_fs = get_fs();
        set_fs(KERNEL_DS);
-       res = blkdev_ioctl(bdev->bd_inode, NULL, cmd, arg);
+       res = blkdev_ioctl(bdev, 0, cmd, arg);
        set_fs(old_fs);
        return res;
 }
@@ -1376,73 +1493,70 @@ EXPORT_SYMBOL(ioctl_by_bdev);
 
 /**
  * lookup_bdev  - lookup a struct block_device by name
+ * @pathname:  special file representing the block device
  *
- * @path:      special file representing the block device
- *
- * Get a reference to the blockdevice at @path in the current
+ * Get a reference to the blockdevice at @pathname in the current
  * namespace if possible and return it.  Return ERR_PTR(error)
  * otherwise.
  */
-struct block_device *lookup_bdev(const char *path)
+struct block_device *lookup_bdev(const char *pathname)
 {
        struct block_device *bdev;
        struct inode *inode;
-       struct nameidata nd;
+       struct path path;
        int error;
 
-       if (!path || !*path)
+       if (!pathname || !*pathname)
                return ERR_PTR(-EINVAL);
 
-       error = path_lookup(path, LOOKUP_FOLLOW, &nd);
+       error = kern_path(pathname, LOOKUP_FOLLOW, &path);
        if (error)
                return ERR_PTR(error);
 
-       inode = nd.path.dentry->d_inode;
+       inode = path.dentry->d_inode;
        error = -ENOTBLK;
        if (!S_ISBLK(inode->i_mode))
                goto fail;
        error = -EACCES;
-       if (nd.path.mnt->mnt_flags & MNT_NODEV)
+       if (path.mnt->mnt_flags & MNT_NODEV)
                goto fail;
        error = -ENOMEM;
        bdev = bd_acquire(inode);
        if (!bdev)
                goto fail;
 out:
-       path_put(&nd.path);
+       path_put(&path);
        return bdev;
 fail:
        bdev = ERR_PTR(error);
        goto out;
 }
+EXPORT_SYMBOL(lookup_bdev);
 
 /**
- * open_bdev_excl  -  open a block device by name and set it up for use
+ * open_bdev_exclusive  -  open a block device by name and set it up for use
  *
  * @path:      special file representing the block device
- * @flags:     %MS_RDONLY for opening read-only
+ * @mode:      FMODE_... combination to pass be used
  * @holder:    owner for exclusion
  *
  * Open the blockdevice described by the special file at @path, claim it
  * for the @holder.
  */
-struct block_device *open_bdev_excl(const char *path, int flags, void *holder)
+struct block_device *open_bdev_exclusive(const char *path, fmode_t mode, void *holder)
 {
        struct block_device *bdev;
-       mode_t mode = FMODE_READ;
        int error = 0;
 
        bdev = lookup_bdev(path);
        if (IS_ERR(bdev))
                return bdev;
 
-       if (!(flags & MS_RDONLY))
-               mode |= FMODE_WRITE;
-       error = blkdev_get(bdev, mode, 0);
+       error = blkdev_get(bdev, mode);
        if (error)
                return ERR_PTR(error);
        error = -EACCES;
-       if (!(flags & MS_RDONLY) && bdev_read_only(bdev))
+       if ((mode & FMODE_WRITE) && bdev_read_only(bdev))
                goto blkdev_put;
        error = bd_claim(bdev, holder);
        if (error)
@@ -1451,26 +1565,27 @@ struct block_device *open_bdev_excl(const char *path, int flags, void *holder)
        return bdev;
        
 blkdev_put:
-       blkdev_put(bdev);
+       blkdev_put(bdev, mode);
        return ERR_PTR(error);
 }
 
-EXPORT_SYMBOL(open_bdev_excl);
+EXPORT_SYMBOL(open_bdev_exclusive);
 
 /**
- * close_bdev_excl  -  release a blockdevice openen by open_bdev_excl()
+ * close_bdev_exclusive  -  close a blockdevice opened by open_bdev_exclusive()
  *
  * @bdev:      blockdevice to close
+ * @mode:      mode, must match that used to open.
  *
- * This is the counterpart to open_bdev_excl().
+ * This is the counterpart to open_bdev_exclusive().
  */
-void close_bdev_excl(struct block_device *bdev)
+void close_bdev_exclusive(struct block_device *bdev, fmode_t mode)
 {
        bd_release(bdev);
-       blkdev_put(bdev);
+       blkdev_put(bdev, mode);
 }
 
-EXPORT_SYMBOL(close_bdev_excl);
+EXPORT_SYMBOL(close_bdev_exclusive);
 
 int __invalidate_device(struct block_device *bdev)
 {