#include <linux/buffer_head.h>
#include <linux/file.h>
#include <linux/fs.h>
+#include <linux/fsnotify.h>
#include <linux/pagemap.h>
#include <linux/highmem.h>
#include <linux/time.h>
#include <linux/string.h>
#include <linux/smp_lock.h>
#include <linux/backing-dev.h>
+#include <linux/mount.h>
#include <linux/mpage.h>
+#include <linux/namei.h>
#include <linux/swap.h>
#include <linux/writeback.h>
#include <linux/statfs.h>
#include <linux/compat.h>
#include <linux/bit_spinlock.h>
+#include <linux/security.h>
#include <linux/version.h>
#include <linux/xattr.h>
+#include <linux/vmalloc.h>
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
-static noinline int create_subvol(struct btrfs_root *root, char *name,
- int namelen)
+static noinline int create_subvol(struct btrfs_root *root,
+ struct dentry *dentry,
+ char *name, int namelen)
{
struct btrfs_trans_handle *trans;
struct btrfs_key key;
if (ret)
goto fail;
- leaf = btrfs_alloc_free_block(trans, root, root->leafsize,
- objectid, trans->transid, 0, 0,
- 0, 0);
+ leaf = btrfs_alloc_free_block(trans, root, root->leafsize, 0,
+ objectid, trans->transid, 0, 0, 0);
if (IS_ERR(leaf)) {
ret = PTR_ERR(leaf);
goto fail;
inode_item->generation = cpu_to_le64(1);
inode_item->size = cpu_to_le64(3);
inode_item->nlink = cpu_to_le32(1);
- inode_item->nblocks = cpu_to_le64(1);
+ inode_item->nbytes = cpu_to_le64(root->leafsize);
inode_item->mode = cpu_to_le32(S_IFDIR | 0755);
btrfs_set_root_bytenr(&root_item, leaf->start);
trans = btrfs_start_transaction(new_root, 1);
BUG_ON(!trans);
- ret = btrfs_create_subvol_root(new_root, trans, new_dirid,
+ ret = btrfs_create_subvol_root(new_root, dentry, trans, new_dirid,
BTRFS_I(dir)->block_group);
if (ret)
goto fail;
- /* Invalidate existing dcache entry for new subvolume. */
- btrfs_invalidate_dcache_root(root, name, namelen);
-
fail:
nr = trans->blocks_used;
err = btrfs_commit_transaction(trans, new_root);
return ret;
}
+/* copy of may_create in fs/namei.c() */
+static inline int btrfs_may_create(struct inode *dir, struct dentry *child)
+{
+ if (child->d_inode)
+ return -EEXIST;
+ if (IS_DEADDIR(dir))
+ return -ENOENT;
+ return inode_permission(dir, MAY_WRITE | MAY_EXEC);
+}
+
+/*
+ * Create a new subvolume below @parent. This is largely modeled after
+ * sys_mkdirat and vfs_mkdir, but we only do a single component lookup
+ * inside this filesystem so it's quite a bit simpler.
+ */
+static noinline int btrfs_mksubvol(struct path *parent, char *name,
+ int mode, int namelen)
+{
+ struct dentry *dentry;
+ int error;
+
+ mutex_lock_nested(&parent->dentry->d_inode->i_mutex, I_MUTEX_PARENT);
+
+ dentry = lookup_one_len(name, parent->dentry, namelen);
+ error = PTR_ERR(dentry);
+ if (IS_ERR(dentry))
+ goto out_unlock;
+
+ error = -EEXIST;
+ if (dentry->d_inode)
+ goto out_dput;
+
+ if (!IS_POSIXACL(parent->dentry->d_inode))
+ mode &= ~current->fs->umask;
+ error = mnt_want_write(parent->mnt);
+ if (error)
+ goto out_dput;
+
+ error = btrfs_may_create(parent->dentry->d_inode, dentry);
+ if (error)
+ goto out_drop_write;
+
+ mode &= (S_IRWXUGO|S_ISVTX);
+ error = security_inode_mkdir(parent->dentry->d_inode, dentry, mode);
+ if (error)
+ goto out_drop_write;
+
+ /*
+ * Actually perform the low-level subvolume creation after all
+ * this VFS fuzz.
+ *
+ * Eventually we want to pass in an inode under which we create this
+ * subvolume, but for now all are under the filesystem root.
+ *
+ * Also we should pass on the mode eventually to allow creating new
+ * subvolume with specific mode bits.
+ */
+ error = create_subvol(BTRFS_I(parent->dentry->d_inode)->root, dentry,
+ name, namelen);
+ if (error)
+ goto out_drop_write;
+
+ fsnotify_mkdir(parent->dentry->d_inode, dentry);
+out_drop_write:
+ mnt_drop_write(parent->mnt);
+out_dput:
+ dput(dentry);
+out_unlock:
+ mutex_unlock(&parent->dentry->d_inode->i_mutex);
+ return error;
+}
+
+
int btrfs_defrag_file(struct file *file)
{
struct inode *inode = fdentry(file)->d_inode;
}
set_page_extent_mapped(page);
- set_extent_delalloc(io_tree, page_start,
- page_end, GFP_NOFS);
+ /*
+ * this makes sure page_mkwrite is called on the
+ * page if it is dirtied again later
+ */
+ clear_page_dirty_for_io(page);
+
+ btrfs_set_extent_delalloc(inode, page_start, page_end);
unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
set_page_dirty(page);
ret = -EFAULT;
goto out;
}
+
+ vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
namelen = strlen(vol_args->name);
- if (namelen > BTRFS_VOL_NAME_MAX) {
- ret = -EINVAL;
- goto out;
- }
mutex_lock(&root->fs_info->volume_mutex);
sizestr = vol_args->name;
return ret;
}
-static noinline int btrfs_ioctl_snap_create(struct btrfs_root *root,
+static noinline int btrfs_ioctl_snap_create(struct file *file,
void __user *arg)
{
+ struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
struct btrfs_ioctl_vol_args *vol_args;
struct btrfs_dir_item *di;
struct btrfs_path *path;
goto out;
}
+ vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
namelen = strlen(vol_args->name);
- if (namelen > BTRFS_VOL_NAME_MAX) {
- ret = -EINVAL;
- goto out;
- }
if (strchr(vol_args->name, '/')) {
ret = -EINVAL;
goto out;
goto out;
}
- mutex_lock(&root->fs_info->drop_mutex);
- if (root == root->fs_info->tree_root)
- ret = create_subvol(root, vol_args->name, namelen);
- else
+ if (root == root->fs_info->tree_root) {
+ ret = btrfs_mksubvol(&file->f_path, vol_args->name,
+ file->f_path.dentry->d_inode->i_mode,
+ namelen);
+ } else {
ret = create_snapshot(root, vol_args->name, namelen);
- mutex_unlock(&root->fs_info->drop_mutex);
+ }
+
out:
kfree(vol_args);
return ret;
ret = -EFAULT;
goto out;
}
+ vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
ret = btrfs_init_new_device(root, vol_args->name);
out:
ret = -EFAULT;
goto out;
}
+ vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
ret = btrfs_rm_device(root, vol_args->name);
out:
return ret;
}
-int dup_item_to_inode(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- struct btrfs_path *path,
- struct extent_buffer *leaf,
- int slot,
- struct btrfs_key *key,
- u64 destino)
-{
- char *dup;
- int len = btrfs_item_size_nr(leaf, slot);
- struct btrfs_key ckey = *key;
- int ret = 0;
-
- dup = kmalloc(len, GFP_NOFS);
- if (!dup)
- return -ENOMEM;
-
- read_extent_buffer(leaf, dup, btrfs_item_ptr_offset(leaf, slot), len);
- btrfs_release_path(root, path);
-
- ckey.objectid = destino;
- ret = btrfs_insert_item(trans, root, &ckey, dup, len);
- kfree(dup);
- return ret;
-}
-
long btrfs_ioctl_clone(struct file *file, unsigned long src_fd)
{
struct inode *inode = fdentry(file)->d_inode;
struct file *src_file;
struct inode *src;
struct btrfs_trans_handle *trans;
- int ret;
- u64 pos;
struct btrfs_path *path;
- struct btrfs_key key;
struct extent_buffer *leaf;
+ char *buf;
+ struct btrfs_key key;
u32 nritems;
int slot;
+ int ret;
src_file = fget(src_fd);
if (!src_file)
return -EBADF;
src = src_file->f_dentry->d_inode;
+ ret = -EISDIR;
+ if (S_ISDIR(src->i_mode) || S_ISDIR(inode->i_mode))
+ goto out_fput;
+
ret = -EXDEV;
- if (src->i_sb != inode->i_sb)
+ if (src->i_sb != inode->i_sb || BTRFS_I(src)->root != root)
goto out_fput;
+ ret = -ENOMEM;
+ buf = vmalloc(btrfs_level_size(root, 0));
+ if (!buf)
+ goto out_fput;
+
+ path = btrfs_alloc_path();
+ if (!path) {
+ vfree(buf);
+ goto out_fput;
+ }
+ path->reada = 2;
+
if (inode < src) {
mutex_lock(&inode->i_mutex);
mutex_lock(&src->i_mutex);
/* do any pending delalloc/csum calc on src, one way or
another, and lock file content */
while (1) {
- filemap_write_and_wait(src->i_mapping);
+ struct btrfs_ordered_extent *ordered;
lock_extent(&BTRFS_I(src)->io_tree, 0, (u64)-1, GFP_NOFS);
- if (BTRFS_I(src)->delalloc_bytes == 0)
+ ordered = btrfs_lookup_first_ordered_extent(inode, (u64)-1);
+ if (BTRFS_I(src)->delalloc_bytes == 0 && !ordered)
break;
unlock_extent(&BTRFS_I(src)->io_tree, 0, (u64)-1, GFP_NOFS);
+ if (ordered)
+ btrfs_put_ordered_extent(ordered);
+ btrfs_wait_ordered_range(src, 0, (u64)-1);
}
- trans = btrfs_start_transaction(root, 0);
- path = btrfs_alloc_path();
- if (!path) {
- ret = -ENOMEM;
- goto out;
- }
- key.offset = 0;
- key.type = BTRFS_EXTENT_DATA_KEY;
+ trans = btrfs_start_transaction(root, 1);
+ BUG_ON(!trans);
+
key.objectid = src->i_ino;
- pos = 0;
- path->reada = 2;
+ key.type = BTRFS_EXTENT_DATA_KEY;
+ key.offset = 0;
while (1) {
/*
if (ret < 0)
goto out;
- if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) {
+ nritems = btrfs_header_nritems(path->nodes[0]);
+ if (path->slots[0] >= nritems) {
ret = btrfs_next_leaf(root, path);
if (ret < 0)
goto out;
if (ret > 0)
break;
+ nritems = btrfs_header_nritems(path->nodes[0]);
}
leaf = path->nodes[0];
slot = path->slots[0];
- btrfs_item_key_to_cpu(leaf, &key, slot);
- nritems = btrfs_header_nritems(leaf);
+ btrfs_item_key_to_cpu(leaf, &key, slot);
if (btrfs_key_type(&key) > BTRFS_CSUM_ITEM_KEY ||
key.objectid != src->i_ino)
break;
+ if (btrfs_key_type(&key) == BTRFS_EXTENT_DATA_KEY ||
+ btrfs_key_type(&key) == BTRFS_CSUM_ITEM_KEY) {
+ u32 size;
+ struct btrfs_key new_key;
+
+ size = btrfs_item_size_nr(leaf, slot);
+ read_extent_buffer(leaf, buf,
+ btrfs_item_ptr_offset(leaf, slot),
+ size);
+ btrfs_release_path(root, path);
+
+ memcpy(&new_key, &key, sizeof(new_key));
+ new_key.objectid = inode->i_ino;
+ ret = btrfs_insert_empty_item(trans, root, path,
+ &new_key, size);
+ if (ret)
+ goto out;
+
+ leaf = path->nodes[0];
+ slot = path->slots[0];
+ write_extent_buffer(leaf, buf,
+ btrfs_item_ptr_offset(leaf, slot),
+ size);
+ btrfs_mark_buffer_dirty(leaf);
+ }
+
if (btrfs_key_type(&key) == BTRFS_EXTENT_DATA_KEY) {
struct btrfs_file_extent_item *extent;
int found_type;
- pos = key.offset;
+
extent = btrfs_item_ptr(leaf, slot,
struct btrfs_file_extent_item);
found_type = btrfs_file_extent_type(leaf, extent);
if (found_type == BTRFS_FILE_EXTENT_REG) {
- u64 len = btrfs_file_extent_num_bytes(leaf,
- extent);
u64 ds = btrfs_file_extent_disk_bytenr(leaf,
extent);
u64 dl = btrfs_file_extent_disk_num_bytes(leaf,
extent);
- u64 off = btrfs_file_extent_offset(leaf,
- extent);
- btrfs_insert_file_extent(trans, root,
- inode->i_ino, pos,
- ds, dl, len, off);
/* ds == 0 means there's a hole */
if (ds != 0) {
- btrfs_inc_extent_ref(trans, root,
- ds, dl,
+ ret = btrfs_inc_extent_ref(trans, root,
+ ds, dl, leaf->start,
root->root_key.objectid,
trans->transid,
- inode->i_ino, pos);
+ inode->i_ino);
+ BUG_ON(ret);
}
- pos = key.offset + len;
- } else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
- ret = dup_item_to_inode(trans, root, path,
- leaf, slot, &key,
- inode->i_ino);
- if (ret)
- goto out;
- pos = key.offset + btrfs_item_size_nr(leaf,
- slot);
}
- } else if (btrfs_key_type(&key) == BTRFS_CSUM_ITEM_KEY) {
- ret = dup_item_to_inode(trans, root, path, leaf,
- slot, &key, inode->i_ino);
-
- if (ret)
- goto out;
}
- key.offset++;
btrfs_release_path(root, path);
+ key.offset++;
}
-
ret = 0;
out:
- btrfs_free_path(path);
-
- inode->i_blocks = src->i_blocks;
- i_size_write(inode, src->i_size);
- btrfs_update_inode(trans, root, inode);
-
- unlock_extent(&BTRFS_I(src)->io_tree, 0, (u64)-1, GFP_NOFS);
-
+ btrfs_release_path(root, path);
+ if (ret == 0) {
+ inode->i_mtime = inode->i_ctime = CURRENT_TIME;
+ inode_set_bytes(inode, inode_get_bytes(src));
+ btrfs_i_size_write(inode, src->i_size);
+ BTRFS_I(inode)->flags = BTRFS_I(src)->flags;
+ ret = btrfs_update_inode(trans, root, inode);
+ }
btrfs_end_transaction(trans, root);
-
+ unlock_extent(&BTRFS_I(src)->io_tree, 0, (u64)-1, GFP_NOFS);
+ if (ret)
+ vmtruncate(inode, 0);
out_unlock:
mutex_unlock(&src->i_mutex);
mutex_unlock(&inode->i_mutex);
+ vfree(buf);
+ btrfs_free_path(path);
out_fput:
fput(src_file);
return ret;
ret = -EINPROGRESS;
goto out;
}
- trans = btrfs_start_transaction(root, 0);
+
+ mutex_lock(&root->fs_info->trans_mutex);
+ root->fs_info->open_ioctl_trans++;
+ mutex_unlock(&root->fs_info->trans_mutex);
+
+ trans = btrfs_start_ioctl_transaction(root, 0);
if (trans)
file->private_data = trans;
else
goto out;
}
btrfs_end_transaction(trans, root);
- file->private_data = 0;
+ file->private_data = NULL;
+
+ mutex_lock(&root->fs_info->trans_mutex);
+ root->fs_info->open_ioctl_trans--;
+ mutex_unlock(&root->fs_info->trans_mutex);
+
out:
return ret;
}
switch (cmd) {
case BTRFS_IOC_SNAP_CREATE:
- return btrfs_ioctl_snap_create(root, (void __user *)arg);
+ return btrfs_ioctl_snap_create(file, (void __user *)arg);
case BTRFS_IOC_DEFRAG:
return btrfs_ioctl_defrag(file);
case BTRFS_IOC_RESIZE:
case BTRFS_IOC_TRANS_END:
return btrfs_ioctl_trans_end(file);
case BTRFS_IOC_SYNC:
+ btrfs_start_delalloc_inodes(root);
btrfs_sync_fs(file->f_dentry->d_sb, 1);
return 0;
}
projects / cascardo / linux.git / blobdiff