* @root - the root we're allocating for
* @block_rsv - the block_rsv we're allocating for
* @orig_bytes - the number of bytes we want
- * @flush - wether or not we can flush to make our reservation
+ * @flush - whether or not we can flush to make our reservation
*
* This will reserve orgi_bytes number of bytes from the space info associated
* with the block_rsv. If there is not enough space it will make an attempt to
* We make the other tasks wait for the flush only when we can flush
* all things.
*/
- if (ret && flush == BTRFS_RESERVE_FLUSH_ALL) {
+ if (ret && flush != BTRFS_RESERVE_NO_FLUSH) {
flushing = true;
space_info->flush = 1;
}
int empty_cluster = 2 * 1024 * 1024;
struct btrfs_space_info *space_info;
int loop = 0;
- int index = 0;
+ int index = __get_raid_index(data);
int alloc_type = (data & BTRFS_BLOCK_GROUP_DATA) ?
RESERVE_ALLOC_NO_ACCOUNT : RESERVE_ALLOC;
bool found_uncached_bg = false;
&wc->flags[level]);
if (ret < 0) {
btrfs_tree_unlock_rw(eb, path->locks[level]);
+ path->locks[level] = 0;
return ret;
}
BUG_ON(wc->refs[level] == 0);
if (wc->refs[level] == 1) {
btrfs_tree_unlock_rw(eb, path->locks[level]);
+ path->locks[level] = 0;
return 1;
}
}
if (test_bit(EXTENT_FLAG_COMPRESSED, &prev->flags))
return 0;
+ if (test_bit(EXTENT_FLAG_LOGGING, &prev->flags) ||
+ test_bit(EXTENT_FLAG_LOGGING, &next->flags))
+ return 0;
+
if (extent_map_end(prev) == next->start &&
prev->flags == next->flags &&
prev->bdev == next->bdev &&
}
/**
- * unpint_extent_cache - unpin an extent from the cache
+ * unpin_extent_cache - unpin an extent from the cache
* @tree: tree to unpin the extent in
* @start: logical offset in the file
* @len: length of the extent
* @gen: generation that this extent has been modified in
- * @prealloc: if this is set we need to clear the prealloc flag
*
* Called after an extent has been written to disk properly. Set the generation
* to the generation that actually added the file item to the inode so we know
if (!em)
goto out;
- list_move(&em->list, &tree->modified_extents);
+ if (!test_bit(EXTENT_FLAG_LOGGING, &em->flags))
+ list_move(&em->list, &tree->modified_extents);
em->generation = gen;
clear_bit(EXTENT_FLAG_PINNED, &em->flags);
em->mod_start = em->start;
}
+ void clear_em_logging(struct extent_map_tree *tree, struct extent_map *em)
+ {
+ clear_bit(EXTENT_FLAG_LOGGING, &em->flags);
+ try_merge_map(tree, em);
+ }
+
/**
* add_extent_mapping - add new extent map to the extent tree
* @tree: tree to insert new map in
cond_resched();
- balance_dirty_pages_ratelimited_nr(inode->i_mapping,
- dirty_pages);
+ balance_dirty_pages_ratelimited(inode->i_mapping);
if (dirty_pages < (root->leafsize >> PAGE_CACHE_SHIFT) + 1)
btrfs_btree_balance_dirty(root);
return ret;
}
-static int find_desired_extent(struct inode *inode, loff_t *offset, int origin)
+static int find_desired_extent(struct inode *inode, loff_t *offset, int whence)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
struct extent_map *em;
if (lockend <= lockstart)
lockend = lockstart + root->sectorsize;
+ lockend--;
len = lockend - lockstart + 1;
len = max_t(u64, len, root->sectorsize);
* before the position we want in case there is outstanding delalloc
* going on here.
*/
- if (origin == SEEK_HOLE && start != 0) {
+ if (whence == SEEK_HOLE && start != 0) {
if (start <= root->sectorsize)
em = btrfs_get_extent_fiemap(inode, NULL, 0, 0,
root->sectorsize, 0);
}
}
- if (origin == SEEK_HOLE) {
+ if (whence == SEEK_HOLE) {
*offset = start;
free_extent_map(em);
break;
}
} else {
- if (origin == SEEK_DATA) {
+ if (whence == SEEK_DATA) {
if (em->block_start == EXTENT_MAP_DELALLOC) {
if (start >= inode->i_size) {
free_extent_map(em);
}
}
- *offset = start;
- free_extent_map(em);
- break;
+ if (!test_bit(EXTENT_FLAG_PREALLOC,
+ &em->flags)) {
+ *offset = start;
+ free_extent_map(em);
+ break;
+ }
}
}
return ret;
}
-static loff_t btrfs_file_llseek(struct file *file, loff_t offset, int origin)
+static loff_t btrfs_file_llseek(struct file *file, loff_t offset, int whence)
{
struct inode *inode = file->f_mapping->host;
int ret;
mutex_lock(&inode->i_mutex);
- switch (origin) {
+ switch (whence) {
case SEEK_END:
case SEEK_CUR:
- offset = generic_file_llseek(file, offset, origin);
+ offset = generic_file_llseek(file, offset, whence);
goto out;
case SEEK_DATA:
case SEEK_HOLE:
return -ENXIO;
}
- ret = find_desired_extent(inode, &offset, origin);
+ ret = find_desired_extent(inode, &offset, whence);
if (ret) {
mutex_unlock(&inode->i_mutex);
return ret;
[S_IFLNK >> S_SHIFT] = BTRFS_FT_SYMLINK,
};
- static int btrfs_setsize(struct inode *inode, loff_t newsize);
+ static int btrfs_setsize(struct inode *inode, struct iattr *attr);
static int btrfs_truncate(struct inode *inode);
static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent);
static noinline int cow_file_range(struct inode *inode,
continue;
}
nr_truncate++;
+
+ /* 1 for the orphan item deletion. */
+ trans = btrfs_start_transaction(root, 1);
+ if (IS_ERR(trans)) {
+ ret = PTR_ERR(trans);
+ goto out;
+ }
+ ret = btrfs_orphan_add(trans, inode);
+ btrfs_end_transaction(trans, root);
+ if (ret)
+ goto out;
+
ret = btrfs_truncate(inode);
} else {
nr_unlink++;
block_end - cur_offset, 0);
if (IS_ERR(em)) {
err = PTR_ERR(em);
+ em = NULL;
break;
}
last_byte = min(extent_map_end(em), block_end);
return err;
}
- static int btrfs_setsize(struct inode *inode, loff_t newsize)
+ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_trans_handle *trans;
loff_t oldsize = i_size_read(inode);
+ loff_t newsize = attr->ia_size;
+ int mask = attr->ia_valid;
int ret;
if (newsize == oldsize)
return 0;
+ /*
+ * The regular truncate() case without ATTR_CTIME and ATTR_MTIME is a
+ * special case where we need to update the times despite not having
+ * these flags set. For all other operations the VFS set these flags
+ * explicitly if it wants a timestamp update.
+ */
+ if (newsize != oldsize && (!(mask & (ATTR_CTIME | ATTR_MTIME))))
+ inode->i_ctime = inode->i_mtime = current_fs_time(inode->i_sb);
+
if (newsize > oldsize) {
truncate_pagecache(inode, oldsize, newsize);
ret = btrfs_cont_expand(inode, oldsize, newsize);
set_bit(BTRFS_INODE_ORDERED_DATA_CLOSE,
&BTRFS_I(inode)->runtime_flags);
+ /*
+ * 1 for the orphan item we're going to add
+ * 1 for the orphan item deletion.
+ */
+ trans = btrfs_start_transaction(root, 2);
+ if (IS_ERR(trans))
+ return PTR_ERR(trans);
+
+ /*
+ * We need to do this in case we fail at _any_ point during the
+ * actual truncate. Once we do the truncate_setsize we could
+ * invalidate pages which forces any outstanding ordered io to
+ * be instantly completed which will give us extents that need
+ * to be truncated. If we fail to get an orphan inode down we
+ * could have left over extents that were never meant to live,
+ * so we need to garuntee from this point on that everything
+ * will be consistent.
+ */
+ ret = btrfs_orphan_add(trans, inode);
+ btrfs_end_transaction(trans, root);
+ if (ret)
+ return ret;
+
/* we don't support swapfiles, so vmtruncate shouldn't fail */
truncate_setsize(inode, newsize);
ret = btrfs_truncate(inode);
+ if (ret && inode->i_nlink)
+ btrfs_orphan_del(NULL, inode);
}
return ret;
return err;
if (S_ISREG(inode->i_mode) && (attr->ia_valid & ATTR_SIZE)) {
- err = btrfs_setsize(inode, attr->ia_size);
+ err = btrfs_setsize(inode, attr);
if (err)
return err;
}
if (dentry->d_name.len > BTRFS_NAME_LEN)
return ERR_PTR(-ENAMETOOLONG);
- if (unlikely(d_need_lookup(dentry))) {
- memcpy(&location, dentry->d_fsdata, sizeof(struct btrfs_key));
- kfree(dentry->d_fsdata);
- dentry->d_fsdata = NULL;
- /* This thing is hashed, drop it for now */
- d_drop(dentry);
- } else {
- ret = btrfs_inode_by_name(dir, dentry, &location);
- }
-
+ ret = btrfs_inode_by_name(dir, dentry, &location);
if (ret < 0)
return ERR_PTR(ret);
struct dentry *ret;
ret = d_splice_alias(btrfs_lookup_dentry(dir, dentry), dentry);
- if (unlikely(d_need_lookup(dentry))) {
- spin_lock(&dentry->d_lock);
- dentry->d_flags &= ~DCACHE_NEED_LOOKUP;
- spin_unlock(&dentry->d_lock);
- }
return ret;
}
return em;
if (em) {
/*
- * if our em maps to a hole, there might
- * actually be delalloc bytes behind it
+ * if our em maps to
+ * - a hole or
+ * - a pre-alloc extent,
+ * there might actually be delalloc bytes behind it.
*/
- if (em->block_start != EXTENT_MAP_HOLE)
+ if (em->block_start != EXTENT_MAP_HOLE &&
+ !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
return em;
else
hole_em = em;
*/
em->block_start = hole_em->block_start;
em->block_len = hole_len;
+ if (test_bit(EXTENT_FLAG_PREALLOC, &hole_em->flags))
+ set_bit(EXTENT_FLAG_PREALLOC, &em->flags);
} else {
em->start = range_start;
em->len = found;
/*
* 1 for the truncate slack space
- * 1 for the orphan item we're going to add
- * 1 for the orphan item deletion
* 1 for updating the inode.
*/
- trans = btrfs_start_transaction(root, 4);
+ trans = btrfs_start_transaction(root, 2);
if (IS_ERR(trans)) {
err = PTR_ERR(trans);
goto out;
min_size);
BUG_ON(ret);
- ret = btrfs_orphan_add(trans, inode);
- if (ret) {
- btrfs_end_transaction(trans, root);
- goto out;
- }
-
/*
* setattr is responsible for setting the ordered_data_close flag,
* but that is only tested during the last file release. That
ret = btrfs_orphan_del(trans, inode);
if (ret)
err = ret;
- } else if (ret && inode->i_nlink > 0) {
- /*
- * Failed to do the truncate, remove us from the in memory
- * orphan list.
- */
- ret = btrfs_orphan_del(NULL, inode);
}
if (trans) {
*/
int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput)
{
- struct list_head *head = &root->fs_info->delalloc_inodes;
struct btrfs_inode *binode;
struct inode *inode;
struct btrfs_delalloc_work *work, *next;
struct list_head works;
+ struct list_head splice;
int ret = 0;
if (root->fs_info->sb->s_flags & MS_RDONLY)
return -EROFS;
INIT_LIST_HEAD(&works);
-
+ INIT_LIST_HEAD(&splice);
+ again:
spin_lock(&root->fs_info->delalloc_lock);
- while (!list_empty(head)) {
- binode = list_entry(head->next, struct btrfs_inode,
+ list_splice_init(&root->fs_info->delalloc_inodes, &splice);
+ while (!list_empty(&splice)) {
+ binode = list_entry(splice.next, struct btrfs_inode,
delalloc_inodes);
+
+ list_del_init(&binode->delalloc_inodes);
+
inode = igrab(&binode->vfs_inode);
if (!inode)
- list_del_init(&binode->delalloc_inodes);
+ continue;
+
+ list_add_tail(&binode->delalloc_inodes,
+ &root->fs_info->delalloc_inodes);
spin_unlock(&root->fs_info->delalloc_lock);
- if (inode) {
- work = btrfs_alloc_delalloc_work(inode, 0, delay_iput);
- if (!work) {
- ret = -ENOMEM;
- goto out;
- }
- list_add_tail(&work->list, &works);
- btrfs_queue_worker(&root->fs_info->flush_workers,
- &work->work);
+
+ work = btrfs_alloc_delalloc_work(inode, 0, delay_iput);
+ if (unlikely(!work)) {
+ ret = -ENOMEM;
+ goto out;
}
+ list_add_tail(&work->list, &works);
+ btrfs_queue_worker(&root->fs_info->flush_workers,
+ &work->work);
+
cond_resched();
spin_lock(&root->fs_info->delalloc_lock);
}
spin_unlock(&root->fs_info->delalloc_lock);
+ list_for_each_entry_safe(work, next, &works, list) {
+ list_del_init(&work->list);
+ btrfs_wait_and_free_delalloc_work(work);
+ }
+
+ spin_lock(&root->fs_info->delalloc_lock);
+ if (!list_empty(&root->fs_info->delalloc_inodes)) {
+ spin_unlock(&root->fs_info->delalloc_lock);
+ goto again;
+ }
+ spin_unlock(&root->fs_info->delalloc_lock);
+
/* the filemap_flush will queue IO into the worker threads, but
* we have to make sure the IO is actually started and that
* ordered extents get created before we return
atomic_read(&root->fs_info->async_delalloc_pages) == 0));
}
atomic_dec(&root->fs_info->async_submit_draining);
+ return 0;
out:
list_for_each_entry_safe(work, next, &works, list) {
list_del_init(&work->list);
btrfs_wait_and_free_delalloc_work(work);
}
+
+ if (!list_empty_careful(&splice)) {
+ spin_lock(&root->fs_info->delalloc_lock);
+ list_splice_tail(&splice, &root->fs_info->delalloc_inodes);
+ spin_unlock(&root->fs_info->delalloc_lock);
+ }
return ret;
}
}
defrag_count += ret;
- balance_dirty_pages_ratelimited_nr(inode->i_mapping, ret);
+ balance_dirty_pages_ratelimited(inode->i_mapping);
mutex_unlock(&inode->i_mutex);
if (newer_than) {
if (atomic_xchg(&root->fs_info->mutually_exclusive_operation_running,
1)) {
pr_info("btrfs: dev add/delete/balance/replace/resize operation in progress\n");
- return -EINPROGRESS;
+ mnt_drop_write_file(file);
+ return -EINVAL;
}
mutex_lock(&root->fs_info->volume_mutex);
printk(KERN_INFO "btrfs: resizing devid %llu\n",
(unsigned long long)devid);
}
+
device = btrfs_find_device(root->fs_info, devid, NULL, NULL);
if (!device) {
printk(KERN_INFO "btrfs: resizer unable to find device %llu\n",
ret = -EINVAL;
goto out_free;
}
- if (device->fs_devices && device->fs_devices->seeding) {
+
+ if (!device->writeable) {
printk(KERN_INFO "btrfs: resizer unable to apply on "
- "seeding device %llu\n",
+ "readonly device %llu\n",
(unsigned long long)devid);
ret = -EINVAL;
goto out_free;
kfree(vol_args);
out:
mutex_unlock(&root->fs_info->volume_mutex);
- mnt_drop_write_file(file);
atomic_set(&root->fs_info->mutually_exclusive_operation_running, 0);
+ mnt_drop_write_file(file);
return ret;
}
err = inode_permission(inode, MAY_WRITE | MAY_EXEC);
if (err)
goto out_dput;
-
- /* check if subvolume may be deleted by a non-root user */
- err = btrfs_may_delete(dir, dentry, 1);
- if (err)
- goto out_dput;
}
+ /* check if subvolume may be deleted by a user */
+ err = btrfs_may_delete(dir, dentry, 1);
+ if (err)
+ goto out_dput;
+
if (btrfs_ino(inode) != BTRFS_FIRST_FREE_OBJECTID) {
err = -EINVAL;
goto out_dput;
struct btrfs_ioctl_defrag_range_args *range;
int ret;
- if (btrfs_root_readonly(root))
- return -EROFS;
+ ret = mnt_want_write_file(file);
+ if (ret)
+ return ret;
if (atomic_xchg(&root->fs_info->mutually_exclusive_operation_running,
1)) {
pr_info("btrfs: dev add/delete/balance/replace/resize operation in progress\n");
- return -EINPROGRESS;
+ mnt_drop_write_file(file);
+ return -EINVAL;
}
- ret = mnt_want_write_file(file);
- if (ret) {
- atomic_set(&root->fs_info->mutually_exclusive_operation_running,
- 0);
- return ret;
+
+ if (btrfs_root_readonly(root)) {
+ ret = -EROFS;
+ goto out;
}
switch (inode->i_mode & S_IFMT) {
ret = -EINVAL;
}
out:
- mnt_drop_write_file(file);
atomic_set(&root->fs_info->mutually_exclusive_operation_running, 0);
+ mnt_drop_write_file(file);
return ret;
}
if (atomic_xchg(&root->fs_info->mutually_exclusive_operation_running,
1)) {
pr_info("btrfs: dev add/delete/balance/replace/resize operation in progress\n");
- return -EINPROGRESS;
+ return -EINVAL;
}
mutex_lock(&root->fs_info->volume_mutex);
1)) {
pr_info("btrfs: dev add/delete/balance/replace/resize operation in progress\n");
mnt_drop_write_file(file);
- return -EINPROGRESS;
+ return -EINVAL;
}
mutex_lock(&root->fs_info->volume_mutex);
kfree(vol_args);
out:
mutex_unlock(&root->fs_info->volume_mutex);
- mnt_drop_write_file(file);
atomic_set(&root->fs_info->mutually_exclusive_operation_running, 0);
+ mnt_drop_write_file(file);
return ret;
}
struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_ioctl_balance_args *bargs;
struct btrfs_balance_control *bctl;
+ bool need_unlock; /* for mut. excl. ops lock */
int ret;
- int need_to_clear_lock = 0;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
if (ret)
return ret;
- mutex_lock(&fs_info->volume_mutex);
+ again:
+ if (!atomic_xchg(&fs_info->mutually_exclusive_operation_running, 1)) {
+ mutex_lock(&fs_info->volume_mutex);
+ mutex_lock(&fs_info->balance_mutex);
+ need_unlock = true;
+ goto locked;
+ }
+
+ /*
+ * mut. excl. ops lock is locked. Three possibilites:
+ * (1) some other op is running
+ * (2) balance is running
+ * (3) balance is paused -- special case (think resume)
+ */
mutex_lock(&fs_info->balance_mutex);
+ if (fs_info->balance_ctl) {
+ /* this is either (2) or (3) */
+ if (!atomic_read(&fs_info->balance_running)) {
+ mutex_unlock(&fs_info->balance_mutex);
+ if (!mutex_trylock(&fs_info->volume_mutex))
+ goto again;
+ mutex_lock(&fs_info->balance_mutex);
+
+ if (fs_info->balance_ctl &&
+ !atomic_read(&fs_info->balance_running)) {
+ /* this is (3) */
+ need_unlock = false;
+ goto locked;
+ }
+
+ mutex_unlock(&fs_info->balance_mutex);
+ mutex_unlock(&fs_info->volume_mutex);
+ goto again;
+ } else {
+ /* this is (2) */
+ mutex_unlock(&fs_info->balance_mutex);
+ ret = -EINPROGRESS;
+ goto out;
+ }
+ } else {
+ /* this is (1) */
+ mutex_unlock(&fs_info->balance_mutex);
+ pr_info("btrfs: dev add/delete/balance/replace/resize operation in progress\n");
+ ret = -EINVAL;
+ goto out;
+ }
+
+ locked:
+ BUG_ON(!atomic_read(&fs_info->mutually_exclusive_operation_running));
if (arg) {
bargs = memdup_user(arg, sizeof(*bargs));
if (IS_ERR(bargs)) {
ret = PTR_ERR(bargs);
- goto out;
+ goto out_unlock;
}
if (bargs->flags & BTRFS_BALANCE_RESUME) {
bargs = NULL;
}
- if (atomic_xchg(&root->fs_info->mutually_exclusive_operation_running,
- 1)) {
- pr_info("btrfs: dev add/delete/balance/replace/resize operation in progress\n");
+ if (fs_info->balance_ctl) {
ret = -EINPROGRESS;
goto out_bargs;
}
- need_to_clear_lock = 1;
bctl = kzalloc(sizeof(*bctl), GFP_NOFS);
if (!bctl) {
}
do_balance:
- ret = btrfs_balance(bctl, bargs);
/*
- * bctl is freed in __cancel_balance or in free_fs_info if
- * restriper was paused all the way until unmount
+ * Ownership of bctl and mutually_exclusive_operation_running
+ * goes to to btrfs_balance. bctl is freed in __cancel_balance,
+ * or, if restriper was paused all the way until unmount, in
+ * free_fs_info. mutually_exclusive_operation_running is
+ * cleared in __cancel_balance.
*/
+ need_unlock = false;
+
+ ret = btrfs_balance(bctl, bargs);
+
if (arg) {
if (copy_to_user(arg, bargs, sizeof(*bargs)))
ret = -EFAULT;
out_bargs:
kfree(bargs);
- out:
- if (need_to_clear_lock)
- atomic_set(&root->fs_info->mutually_exclusive_operation_running,
- 0);
+ out_unlock:
mutex_unlock(&fs_info->balance_mutex);
mutex_unlock(&fs_info->volume_mutex);
+ if (need_unlock)
+ atomic_set(&fs_info->mutually_exclusive_operation_running, 0);
+ out:
mnt_drop_write_file(file);
return ret;
}
goto drop_write;
}
+ if (!sa->qgroupid) {
+ ret = -EINVAL;
+ goto out;
+ }
+
trans = btrfs_join_transaction(root);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);