* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (114 commits)
Btrfs: check for a null fs root when writing to the backup root log
Btrfs: fix race during transaction joins
Btrfs: fix a potential btrfs_bio leak on scrub fixups
Btrfs: rename btrfs_bio multi -> bbio for consistency
Btrfs: stop leaking btrfs_bios on readahead
Btrfs: stop the readahead threads on failed mount
Btrfs: fix extent_buffer leak in the metadata IO error handling
Btrfs: fix the new inspection ioctls for 32 bit compat
Btrfs: fix delayed insertion reservation
Btrfs: ClearPageError during writepage and clean_tree_block
Btrfs: be smarter about committing the transaction in reserve_metadata_bytes
Btrfs: make a delayed_block_rsv for the delayed item insertion
Btrfs: add a log of past tree roots
btrfs: separate superblock items out of fs_info
Btrfs: use the global reserve when truncating the free space cache inode
Btrfs: release metadata from global reserve if we have to fallback for unlink
Btrfs: make sure to flush queued bios if write_cache_pages waits
Btrfs: fix extent pinning bugs in the tree log
Btrfs: make sure btrfs_remove_free_space doesn't leak EAGAIN
Btrfs: don't wait as long for more batches during SSD log commit
...
return 0;
src_rsv = trans->block_rsv;
- dst_rsv = &root->fs_info->global_block_rsv;
+ dst_rsv = &root->fs_info->delayed_block_rsv;
num_bytes = btrfs_calc_trans_metadata_size(root, 1);
ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
if (!item->bytes_reserved)
return;
- rsv = &root->fs_info->global_block_rsv;
+ rsv = &root->fs_info->delayed_block_rsv;
btrfs_block_rsv_release(root, rsv,
item->bytes_reserved);
}
u64 num_bytes;
int ret;
- if (!trans->bytes_reserved)
- return 0;
-
src_rsv = trans->block_rsv;
- dst_rsv = &root->fs_info->global_block_rsv;
+ dst_rsv = &root->fs_info->delayed_block_rsv;
num_bytes = btrfs_calc_trans_metadata_size(root, 1);
+
+ /*
+ * btrfs_dirty_inode will update the inode under btrfs_join_transaction
+ * which doesn't reserve space for speed. This is a problem since we
+ * still need to reserve space for this update, so try to reserve the
+ * space.
+ *
+ * Now if src_rsv == delalloc_block_rsv we'll let it just steal since
+ * we're accounted for.
+ */
+ if (!trans->bytes_reserved &&
+ src_rsv != &root->fs_info->delalloc_block_rsv) {
+ ret = btrfs_block_rsv_add_noflush(root, dst_rsv, num_bytes);
+ /*
+ * Since we're under a transaction reserve_metadata_bytes could
+ * try to commit the transaction which will make it return
+ * EAGAIN to make us stop the transaction we have, so return
+ * ENOSPC instead so that btrfs_dirty_inode knows what to do.
+ */
+ if (ret == -EAGAIN)
+ ret = -ENOSPC;
+ if (!ret)
+ node->bytes_reserved = num_bytes;
+ return ret;
+ }
+
ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
if (!ret)
node->bytes_reserved = num_bytes;
if (!node->bytes_reserved)
return;
- rsv = &root->fs_info->global_block_rsv;
+ rsv = &root->fs_info->delayed_block_rsv;
btrfs_block_rsv_release(root, rsv,
node->bytes_reserved);
node->bytes_reserved = 0;
path->leave_spinning = 1;
block_rsv = trans->block_rsv;
- trans->block_rsv = &root->fs_info->global_block_rsv;
+ trans->block_rsv = &root->fs_info->delayed_block_rsv;
delayed_root = btrfs_get_delayed_root(root);
path->leave_spinning = 1;
block_rsv = trans->block_rsv;
- trans->block_rsv = &node->root->fs_info->global_block_rsv;
+ trans->block_rsv = &node->root->fs_info->delayed_block_rsv;
ret = btrfs_insert_delayed_items(trans, path, node->root, node);
if (!ret)
goto free_path;
block_rsv = trans->block_rsv;
- trans->block_rsv = &root->fs_info->global_block_rsv;
+ trans->block_rsv = &root->fs_info->delayed_block_rsv;
ret = btrfs_insert_delayed_items(trans, path, root, delayed_node);
if (!ret)
inode->i_gid = btrfs_stack_inode_gid(inode_item);
btrfs_i_size_write(inode, btrfs_stack_inode_size(inode_item));
inode->i_mode = btrfs_stack_inode_mode(inode_item);
- inode->i_nlink = btrfs_stack_inode_nlink(inode_item);
+ set_nlink(inode, btrfs_stack_inode_nlink(inode_item));
inode_set_bytes(inode, btrfs_stack_inode_nbytes(inode_item));
BTRFS_I(inode)->generation = btrfs_stack_inode_generation(inode_item);
BTRFS_I(inode)->sequence = btrfs_stack_inode_sequence(inode_item);
}
ret = btrfs_delayed_inode_reserve_metadata(trans, root, delayed_node);
- /*
- * we must reserve enough space when we start a new transaction,
- * so reserving metadata failure is impossible
- */
- BUG_ON(ret);
+ if (ret)
+ goto release_node;
fill_stack_inode_item(trans, &delayed_node->inode_item, inode);
delayed_node->inode_dirty = 1;
static int csum_tree_block(struct btrfs_root *root, struct extent_buffer *buf,
int verify)
{
- u16 csum_size =
- btrfs_super_csum_size(&root->fs_info->super_copy);
+ u16 csum_size = btrfs_super_csum_size(root->fs_info->super_copy);
char *result = NULL;
unsigned long len;
unsigned long cur_len;
clear_bit(EXTENT_BUFFER_CORRUPT, &eb->bflags);
io_tree = &BTRFS_I(root->fs_info->btree_inode)->io_tree;
while (1) {
- ret = read_extent_buffer_pages(io_tree, eb, start, 1,
+ ret = read_extent_buffer_pages(io_tree, eb, start,
+ WAIT_COMPLETE,
btree_get_extent, mirror_num);
if (!ret &&
!verify_parent_transid(io_tree, eb, parent_transid))
end = min_t(u64, eb->len, PAGE_CACHE_SIZE);
end = eb->start + end - 1;
err:
+ if (test_bit(EXTENT_BUFFER_READAHEAD, &eb->bflags)) {
+ clear_bit(EXTENT_BUFFER_READAHEAD, &eb->bflags);
+ btree_readahead_hook(root, eb, eb->start, ret);
+ }
+
free_extent_buffer(eb);
out:
return ret;
}
+ static int btree_io_failed_hook(struct bio *failed_bio,
+ struct page *page, u64 start, u64 end,
+ u64 mirror_num, struct extent_state *state)
+ {
+ struct extent_io_tree *tree;
+ unsigned long len;
+ struct extent_buffer *eb;
+ struct btrfs_root *root = BTRFS_I(page->mapping->host)->root;
+
+ tree = &BTRFS_I(page->mapping->host)->io_tree;
+ if (page->private == EXTENT_PAGE_PRIVATE)
+ goto out;
+ if (!page->private)
+ goto out;
+
+ len = page->private >> 2;
+ WARN_ON(len == 0);
+
+ eb = alloc_extent_buffer(tree, start, len, page);
+ if (eb == NULL)
+ goto out;
+
+ if (test_bit(EXTENT_BUFFER_READAHEAD, &eb->bflags)) {
+ clear_bit(EXTENT_BUFFER_READAHEAD, &eb->bflags);
+ btree_readahead_hook(root, eb, eb->start, -EIO);
+ }
+ free_extent_buffer(eb);
+
+ out:
+ return -EIO; /* we fixed nothing */
+ }
+
static void end_workqueue_bio(struct bio *bio, int err)
{
struct end_io_wq *end_io_wq = bio->bi_private;
{
struct extent_io_tree *tree;
tree = &BTRFS_I(page->mapping->host)->io_tree;
- return extent_read_full_page(tree, page, btree_get_extent);
+ return extent_read_full_page(tree, page, btree_get_extent, 0);
}
static int btree_releasepage(struct page *page, gfp_t gfp_flags)
if (!buf)
return 0;
read_extent_buffer_pages(&BTRFS_I(btree_inode)->io_tree,
- buf, 0, 0, btree_get_extent, 0);
+ buf, 0, WAIT_NONE, btree_get_extent, 0);
free_extent_buffer(buf);
return ret;
}
+ int reada_tree_block_flagged(struct btrfs_root *root, u64 bytenr, u32 blocksize,
+ int mirror_num, struct extent_buffer **eb)
+ {
+ struct extent_buffer *buf = NULL;
+ struct inode *btree_inode = root->fs_info->btree_inode;
+ struct extent_io_tree *io_tree = &BTRFS_I(btree_inode)->io_tree;
+ int ret;
+
+ buf = btrfs_find_create_tree_block(root, bytenr, blocksize);
+ if (!buf)
+ return 0;
+
+ set_bit(EXTENT_BUFFER_READAHEAD, &buf->bflags);
+
+ ret = read_extent_buffer_pages(io_tree, buf, 0, WAIT_PAGE_LOCK,
+ btree_get_extent, mirror_num);
+ if (ret) {
+ free_extent_buffer(buf);
+ return ret;
+ }
+
+ if (test_bit(EXTENT_BUFFER_CORRUPT, &buf->bflags)) {
+ free_extent_buffer(buf);
+ return -EIO;
+ } else if (extent_buffer_uptodate(io_tree, buf, NULL)) {
+ *eb = buf;
+ } else {
+ free_extent_buffer(buf);
+ }
+ return 0;
+ }
+
struct extent_buffer *btrfs_find_tree_block(struct btrfs_root *root,
u64 bytenr, u32 blocksize)
{
generation = btrfs_root_generation(&root->root_item);
blocksize = btrfs_level_size(root, btrfs_root_level(&root->root_item));
+ root->commit_root = NULL;
root->node = read_tree_block(root, btrfs_root_bytenr(&root->root_item),
blocksize, generation);
if (!root->node || !btrfs_buffer_uptodate(root->node, generation)) {
free_extent_buffer(root->node);
+ root->node = NULL;
return -EIO;
}
root->commit_root = btrfs_root_node(root);
return 0;
}
+ /*
+ * this will find the highest generation in the array of
+ * root backups. The index of the highest array is returned,
+ * or -1 if we can't find anything.
+ *
+ * We check to make sure the array is valid by comparing the
+ * generation of the latest root in the array with the generation
+ * in the super block. If they don't match we pitch it.
+ */
+ static int find_newest_super_backup(struct btrfs_fs_info *info, u64 newest_gen)
+ {
+ u64 cur;
+ int newest_index = -1;
+ struct btrfs_root_backup *root_backup;
+ int i;
+
+ for (i = 0; i < BTRFS_NUM_BACKUP_ROOTS; i++) {
+ root_backup = info->super_copy->super_roots + i;
+ cur = btrfs_backup_tree_root_gen(root_backup);
+ if (cur == newest_gen)
+ newest_index = i;
+ }
+
+ /* check to see if we actually wrapped around */
+ if (newest_index == BTRFS_NUM_BACKUP_ROOTS - 1) {
+ root_backup = info->super_copy->super_roots;
+ cur = btrfs_backup_tree_root_gen(root_backup);
+ if (cur == newest_gen)
+ newest_index = 0;
+ }
+ return newest_index;
+ }
+
+
+ /*
+ * find the oldest backup so we know where to store new entries
+ * in the backup array. This will set the backup_root_index
+ * field in the fs_info struct
+ */
+ static void find_oldest_super_backup(struct btrfs_fs_info *info,
+ u64 newest_gen)
+ {
+ int newest_index = -1;
+
+ newest_index = find_newest_super_backup(info, newest_gen);
+ /* if there was garbage in there, just move along */
+ if (newest_index == -1) {
+ info->backup_root_index = 0;
+ } else {
+ info->backup_root_index = (newest_index + 1) % BTRFS_NUM_BACKUP_ROOTS;
+ }
+ }
+
+ /*
+ * copy all the root pointers into the super backup array.
+ * this will bump the backup pointer by one when it is
+ * done
+ */
+ static void backup_super_roots(struct btrfs_fs_info *info)
+ {
+ int next_backup;
+ struct btrfs_root_backup *root_backup;
+ int last_backup;
+
+ next_backup = info->backup_root_index;
+ last_backup = (next_backup + BTRFS_NUM_BACKUP_ROOTS - 1) %
+ BTRFS_NUM_BACKUP_ROOTS;
+
+ /*
+ * just overwrite the last backup if we're at the same generation
+ * this happens only at umount
+ */
+ root_backup = info->super_for_commit->super_roots + last_backup;
+ if (btrfs_backup_tree_root_gen(root_backup) ==
+ btrfs_header_generation(info->tree_root->node))
+ next_backup = last_backup;
+
+ root_backup = info->super_for_commit->super_roots + next_backup;
+
+ /*
+ * make sure all of our padding and empty slots get zero filled
+ * regardless of which ones we use today
+ */
+ memset(root_backup, 0, sizeof(*root_backup));
+
+ info->backup_root_index = (next_backup + 1) % BTRFS_NUM_BACKUP_ROOTS;
+
+ btrfs_set_backup_tree_root(root_backup, info->tree_root->node->start);
+ btrfs_set_backup_tree_root_gen(root_backup,
+ btrfs_header_generation(info->tree_root->node));
+
+ btrfs_set_backup_tree_root_level(root_backup,
+ btrfs_header_level(info->tree_root->node));
+
+ btrfs_set_backup_chunk_root(root_backup, info->chunk_root->node->start);
+ btrfs_set_backup_chunk_root_gen(root_backup,
+ btrfs_header_generation(info->chunk_root->node));
+ btrfs_set_backup_chunk_root_level(root_backup,
+ btrfs_header_level(info->chunk_root->node));
+
+ btrfs_set_backup_extent_root(root_backup, info->extent_root->node->start);
+ btrfs_set_backup_extent_root_gen(root_backup,
+ btrfs_header_generation(info->extent_root->node));
+ btrfs_set_backup_extent_root_level(root_backup,
+ btrfs_header_level(info->extent_root->node));
+
+ /*
+ * we might commit during log recovery, which happens before we set
+ * the fs_root. Make sure it is valid before we fill it in.
+ */
+ if (info->fs_root && info->fs_root->node) {
+ btrfs_set_backup_fs_root(root_backup,
+ info->fs_root->node->start);
+ btrfs_set_backup_fs_root_gen(root_backup,
+ btrfs_header_generation(info->fs_root->node));
+ btrfs_set_backup_fs_root_level(root_backup,
+ btrfs_header_level(info->fs_root->node));
+ }
+
+ btrfs_set_backup_dev_root(root_backup, info->dev_root->node->start);
+ btrfs_set_backup_dev_root_gen(root_backup,
+ btrfs_header_generation(info->dev_root->node));
+ btrfs_set_backup_dev_root_level(root_backup,
+ btrfs_header_level(info->dev_root->node));
+
+ btrfs_set_backup_csum_root(root_backup, info->csum_root->node->start);
+ btrfs_set_backup_csum_root_gen(root_backup,
+ btrfs_header_generation(info->csum_root->node));
+ btrfs_set_backup_csum_root_level(root_backup,
+ btrfs_header_level(info->csum_root->node));
+
+ btrfs_set_backup_total_bytes(root_backup,
+ btrfs_super_total_bytes(info->super_copy));
+ btrfs_set_backup_bytes_used(root_backup,
+ btrfs_super_bytes_used(info->super_copy));
+ btrfs_set_backup_num_devices(root_backup,
+ btrfs_super_num_devices(info->super_copy));
+
+ /*
+ * if we don't copy this out to the super_copy, it won't get remembered
+ * for the next commit
+ */
+ memcpy(&info->super_copy->super_roots,
+ &info->super_for_commit->super_roots,
+ sizeof(*root_backup) * BTRFS_NUM_BACKUP_ROOTS);
+ }
+
+ /*
+ * this copies info out of the root backup array and back into
+ * the in-memory super block. It is meant to help iterate through
+ * the array, so you send it the number of backups you've already
+ * tried and the last backup index you used.
+ *
+ * this returns -1 when it has tried all the backups
+ */
+ static noinline int next_root_backup(struct btrfs_fs_info *info,
+ struct btrfs_super_block *super,
+ int *num_backups_tried, int *backup_index)
+ {
+ struct btrfs_root_backup *root_backup;
+ int newest = *backup_index;
+
+ if (*num_backups_tried == 0) {
+ u64 gen = btrfs_super_generation(super);
+
+ newest = find_newest_super_backup(info, gen);
+ if (newest == -1)
+ return -1;
+
+ *backup_index = newest;
+ *num_backups_tried = 1;
+ } else if (*num_backups_tried == BTRFS_NUM_BACKUP_ROOTS) {
+ /* we've tried all the backups, all done */
+ return -1;
+ } else {
+ /* jump to the next oldest backup */
+ newest = (*backup_index + BTRFS_NUM_BACKUP_ROOTS - 1) %
+ BTRFS_NUM_BACKUP_ROOTS;
+ *backup_index = newest;
+ *num_backups_tried += 1;
+ }
+ root_backup = super->super_roots + newest;
+
+ btrfs_set_super_generation(super,
+ btrfs_backup_tree_root_gen(root_backup));
+ btrfs_set_super_root(super, btrfs_backup_tree_root(root_backup));
+ btrfs_set_super_root_level(super,
+ btrfs_backup_tree_root_level(root_backup));
+ btrfs_set_super_bytes_used(super, btrfs_backup_bytes_used(root_backup));
+
+ /*
+ * fixme: the total bytes and num_devices need to match or we should
+ * need a fsck
+ */
+ btrfs_set_super_total_bytes(super, btrfs_backup_total_bytes(root_backup));
+ btrfs_set_super_num_devices(super, btrfs_backup_num_devices(root_backup));
+ return 0;
+ }
+
+ /* helper to cleanup tree roots */
+ static void free_root_pointers(struct btrfs_fs_info *info, int chunk_root)
+ {
+ free_extent_buffer(info->tree_root->node);
+ free_extent_buffer(info->tree_root->commit_root);
+ free_extent_buffer(info->dev_root->node);
+ free_extent_buffer(info->dev_root->commit_root);
+ free_extent_buffer(info->extent_root->node);
+ free_extent_buffer(info->extent_root->commit_root);
+ free_extent_buffer(info->csum_root->node);
+ free_extent_buffer(info->csum_root->commit_root);
+
+ info->tree_root->node = NULL;
+ info->tree_root->commit_root = NULL;
+ info->dev_root->node = NULL;
+ info->dev_root->commit_root = NULL;
+ info->extent_root->node = NULL;
+ info->extent_root->commit_root = NULL;
+ info->csum_root->node = NULL;
+ info->csum_root->commit_root = NULL;
+
+ if (chunk_root) {
+ free_extent_buffer(info->chunk_root->node);
+ free_extent_buffer(info->chunk_root->commit_root);
+ info->chunk_root->node = NULL;
+ info->chunk_root->commit_root = NULL;
+ }
+ }
+
+
struct btrfs_root *open_ctree(struct super_block *sb,
struct btrfs_fs_devices *fs_devices,
char *options)
int ret;
int err = -EINVAL;
+ int num_backups_tried = 0;
+ int backup_index = 0;
struct btrfs_super_block *disk_super;
spin_lock_init(&fs_info->fs_roots_radix_lock);
spin_lock_init(&fs_info->delayed_iput_lock);
spin_lock_init(&fs_info->defrag_inodes_lock);
+ spin_lock_init(&fs_info->free_chunk_lock);
mutex_init(&fs_info->reloc_mutex);
init_completion(&fs_info->kobj_unregister);
btrfs_init_block_rsv(&fs_info->trans_block_rsv);
btrfs_init_block_rsv(&fs_info->chunk_block_rsv);
btrfs_init_block_rsv(&fs_info->empty_block_rsv);
- INIT_LIST_HEAD(&fs_info->durable_block_rsv_list);
- mutex_init(&fs_info->durable_block_rsv_mutex);
+ btrfs_init_block_rsv(&fs_info->delayed_block_rsv);
atomic_set(&fs_info->nr_async_submits, 0);
atomic_set(&fs_info->async_delalloc_pages, 0);
atomic_set(&fs_info->async_submit_draining, 0);
fs_info->metadata_ratio = 0;
fs_info->defrag_inodes = RB_ROOT;
fs_info->trans_no_join = 0;
+ fs_info->free_chunk_space = 0;
+
+ /* readahead state */
+ INIT_RADIX_TREE(&fs_info->reada_tree, GFP_NOFS & ~__GFP_WAIT);
+ spin_lock_init(&fs_info->reada_lock);
fs_info->thread_pool_size = min_t(unsigned long,
num_online_cpus() + 2, 8);
sb->s_bdi = &fs_info->bdi;
fs_info->btree_inode->i_ino = BTRFS_BTREE_INODE_OBJECTID;
- fs_info->btree_inode->i_nlink = 1;
+ set_nlink(fs_info->btree_inode, 1);
/*
* we set the i_size on the btree inode to the max possible int.
* the real end of the address space is determined by all of
goto fail_alloc;
}
- memcpy(&fs_info->super_copy, bh->b_data, sizeof(fs_info->super_copy));
- memcpy(&fs_info->super_for_commit, &fs_info->super_copy,
- sizeof(fs_info->super_for_commit));
+ memcpy(fs_info->super_copy, bh->b_data, sizeof(*fs_info->super_copy));
+ memcpy(fs_info->super_for_commit, fs_info->super_copy,
+ sizeof(*fs_info->super_for_commit));
brelse(bh);
- memcpy(fs_info->fsid, fs_info->super_copy.fsid, BTRFS_FSID_SIZE);
+ memcpy(fs_info->fsid, fs_info->super_copy->fsid, BTRFS_FSID_SIZE);
- disk_super = &fs_info->super_copy;
+ disk_super = fs_info->super_copy;
if (!btrfs_super_root(disk_super))
goto fail_alloc;
btrfs_check_super_valid(fs_info, sb->s_flags & MS_RDONLY);
+ /*
+ * run through our array of backup supers and setup
+ * our ring pointer to the oldest one
+ */
+ generation = btrfs_super_generation(disk_super);
+ find_oldest_super_backup(fs_info, generation);
+
/*
* In the long term, we'll store the compression type in the super
* block, and it'll be used for per file compression control.
btrfs_init_workers(&fs_info->delayed_workers, "delayed-meta",
fs_info->thread_pool_size,
&fs_info->generic_worker);
+ btrfs_init_workers(&fs_info->readahead_workers, "readahead",
+ fs_info->thread_pool_size,
+ &fs_info->generic_worker);
/*
* endios are largely parallel and should have a very
fs_info->endio_write_workers.idle_thresh = 2;
fs_info->endio_meta_write_workers.idle_thresh = 2;
+ fs_info->readahead_workers.idle_thresh = 2;
btrfs_start_workers(&fs_info->workers, 1);
btrfs_start_workers(&fs_info->generic_worker, 1);
btrfs_start_workers(&fs_info->endio_freespace_worker, 1);
btrfs_start_workers(&fs_info->delayed_workers, 1);
btrfs_start_workers(&fs_info->caching_workers, 1);
+ btrfs_start_workers(&fs_info->readahead_workers, 1);
fs_info->bdi.ra_pages *= btrfs_super_num_devices(disk_super);
fs_info->bdi.ra_pages = max(fs_info->bdi.ra_pages,
if (!test_bit(EXTENT_BUFFER_UPTODATE, &chunk_root->node->bflags)) {
printk(KERN_WARNING "btrfs: failed to read chunk root on %s\n",
sb->s_id);
- goto fail_chunk_root;
+ goto fail_tree_roots;
}
btrfs_set_root_node(&chunk_root->root_item, chunk_root->node);
chunk_root->commit_root = btrfs_root_node(chunk_root);
if (ret) {
printk(KERN_WARNING "btrfs: failed to read chunk tree on %s\n",
sb->s_id);
- goto fail_chunk_root;
+ goto fail_tree_roots;
}
btrfs_close_extra_devices(fs_devices);
+ retry_root_backup:
blocksize = btrfs_level_size(tree_root,
btrfs_super_root_level(disk_super));
generation = btrfs_super_generation(disk_super);
tree_root->node = read_tree_block(tree_root,
btrfs_super_root(disk_super),
blocksize, generation);
- if (!tree_root->node)
- goto fail_chunk_root;
- if (!test_bit(EXTENT_BUFFER_UPTODATE, &tree_root->node->bflags)) {
+ if (!tree_root->node ||
+ !test_bit(EXTENT_BUFFER_UPTODATE, &tree_root->node->bflags)) {
printk(KERN_WARNING "btrfs: failed to read tree root on %s\n",
sb->s_id);
- goto fail_tree_root;
+
+ goto recovery_tree_root;
}
+
btrfs_set_root_node(&tree_root->root_item, tree_root->node);
tree_root->commit_root = btrfs_root_node(tree_root);
ret = find_and_setup_root(tree_root, fs_info,
BTRFS_EXTENT_TREE_OBJECTID, extent_root);
if (ret)
- goto fail_tree_root;
+ goto recovery_tree_root;
extent_root->track_dirty = 1;
ret = find_and_setup_root(tree_root, fs_info,
BTRFS_DEV_TREE_OBJECTID, dev_root);
if (ret)
- goto fail_extent_root;
+ goto recovery_tree_root;
dev_root->track_dirty = 1;
ret = find_and_setup_root(tree_root, fs_info,
BTRFS_CSUM_TREE_OBJECTID, csum_root);
if (ret)
- goto fail_dev_root;
+ goto recovery_tree_root;
csum_root->track_dirty = 1;
fail_block_groups:
btrfs_free_block_groups(fs_info);
- free_extent_buffer(csum_root->node);
- free_extent_buffer(csum_root->commit_root);
- fail_dev_root:
- free_extent_buffer(dev_root->node);
- free_extent_buffer(dev_root->commit_root);
- fail_extent_root:
- free_extent_buffer(extent_root->node);
- free_extent_buffer(extent_root->commit_root);
- fail_tree_root:
- free_extent_buffer(tree_root->node);
- free_extent_buffer(tree_root->commit_root);
- fail_chunk_root:
- free_extent_buffer(chunk_root->node);
- free_extent_buffer(chunk_root->commit_root);
+
+ fail_tree_roots:
+ free_root_pointers(fs_info, 1);
+
fail_sb_buffer:
btrfs_stop_workers(&fs_info->generic_worker);
+ btrfs_stop_workers(&fs_info->readahead_workers);
btrfs_stop_workers(&fs_info->fixup_workers);
btrfs_stop_workers(&fs_info->delalloc_workers);
btrfs_stop_workers(&fs_info->workers);
btrfs_stop_workers(&fs_info->delayed_workers);
btrfs_stop_workers(&fs_info->caching_workers);
fail_alloc:
- kfree(fs_info->delayed_root);
fail_iput:
invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
iput(fs_info->btree_inode);
fail_srcu:
cleanup_srcu_struct(&fs_info->subvol_srcu);
fail:
- kfree(extent_root);
- kfree(tree_root);
- kfree(fs_info);
- kfree(chunk_root);
- kfree(dev_root);
- kfree(csum_root);
+ free_fs_info(fs_info);
return ERR_PTR(err);
+
+ recovery_tree_root:
+
+ if (!btrfs_test_opt(tree_root, RECOVERY))
+ goto fail_tree_roots;
+
+ free_root_pointers(fs_info, 0);
+
+ /* don't use the log in recovery mode, it won't be valid */
+ btrfs_set_super_log_root(disk_super, 0);
+
+ /* we can't trust the free space cache either */
+ btrfs_set_opt(fs_info->mount_opt, CLEAR_CACHE);
+
+ ret = next_root_backup(fs_info, fs_info->super_copy,
+ &num_backups_tried, &backup_index);
+ if (ret == -1)
+ goto fail_block_groups;
+ goto retry_root_backup;
}
static void btrfs_end_buffer_write_sync(struct buffer_head *bh, int uptodate)
int total_errors = 0;
u64 flags;
- max_errors = btrfs_super_num_devices(&root->fs_info->super_copy) - 1;
+ max_errors = btrfs_super_num_devices(root->fs_info->super_copy) - 1;
do_barriers = !btrfs_test_opt(root, NOBARRIER);
+ backup_super_roots(root->fs_info);
- sb = &root->fs_info->super_for_commit;
+ sb = root->fs_info->super_for_commit;
dev_item = &sb->dev_item;
mutex_lock(&root->fs_info->fs_devices->device_list_mutex);
/* clear out the rbtree of defraggable inodes */
btrfs_run_defrag_inodes(root->fs_info);
- btrfs_put_block_group_cache(fs_info);
-
/*
* Here come 2 situations when btrfs is broken to flip readonly:
*
printk(KERN_ERR "btrfs: commit super ret %d\n", ret);
}
+ btrfs_put_block_group_cache(fs_info);
+
kthread_stop(root->fs_info->transaction_kthread);
kthread_stop(root->fs_info->cleaner_kthread);
del_fs_roots(fs_info);
iput(fs_info->btree_inode);
- kfree(fs_info->delayed_root);
btrfs_stop_workers(&fs_info->generic_worker);
btrfs_stop_workers(&fs_info->fixup_workers);
btrfs_stop_workers(&fs_info->submit_workers);
btrfs_stop_workers(&fs_info->delayed_workers);
btrfs_stop_workers(&fs_info->caching_workers);
+ btrfs_stop_workers(&fs_info->readahead_workers);
btrfs_close_devices(fs_info->fs_devices);
btrfs_mapping_tree_free(&fs_info->mapping_tree);
bdi_destroy(&fs_info->bdi);
cleanup_srcu_struct(&fs_info->subvol_srcu);
- kfree(fs_info->extent_root);
- kfree(fs_info->tree_root);
- kfree(fs_info->chunk_root);
- kfree(fs_info->dev_root);
- kfree(fs_info->csum_root);
- kfree(fs_info);
+ free_fs_info(fs_info);
return 0;
}
return ret;
}
- int btree_lock_page_hook(struct page *page)
+ static int btree_lock_page_hook(struct page *page, void *data,
+ void (*flush_fn)(void *))
{
struct inode *inode = page->mapping->host;
struct btrfs_root *root = BTRFS_I(inode)->root;
if (!eb)
goto out;
- btrfs_tree_lock(eb);
+ if (!btrfs_try_tree_write_lock(eb)) {
+ flush_fn(data);
+ btrfs_tree_lock(eb);
+ }
btrfs_set_header_flag(eb, BTRFS_HEADER_FLAG_WRITTEN);
if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)) {
btrfs_tree_unlock(eb);
free_extent_buffer(eb);
out:
- lock_page(page);
+ if (!trylock_page(page)) {
+ flush_fn(data);
+ lock_page(page);
+ }
return 0;
}
static struct extent_io_ops btree_extent_io_ops = {
.write_cache_pages_lock_hook = btree_lock_page_hook,
.readpage_end_io_hook = btree_readpage_end_io_hook,
+ .readpage_io_failed_hook = btree_io_failed_hook,
.submit_bio_hook = btree_submit_bio_hook,
/* note we're sharing with inode.c for the merge bio hook */
.merge_bio_hook = btrfs_merge_bio_hook,
#include <linux/rcupdate.h>
#include <linux/kthread.h>
#include <linux/slab.h>
+ #include <linux/ratelimit.h>
#include "compat.h"
#include "hash.h"
#include "ctree.h"
CHUNK_ALLOC_LIMITED = 2,
};
+ /*
+ * Control how reservations are dealt with.
+ *
+ * RESERVE_FREE - freeing a reservation.
+ * RESERVE_ALLOC - allocating space and we need to update bytes_may_use for
+ * ENOSPC accounting
+ * RESERVE_ALLOC_NO_ACCOUNT - allocating space and we should not update
+ * bytes_may_use as the ENOSPC accounting is done elsewhere
+ */
+ enum {
+ RESERVE_FREE = 0,
+ RESERVE_ALLOC = 1,
+ RESERVE_ALLOC_NO_ACCOUNT = 2,
+ };
+
static int update_block_group(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
u64 bytenr, u64 num_bytes, int alloc);
struct btrfs_key *key);
static void dump_space_info(struct btrfs_space_info *info, u64 bytes,
int dump_block_groups);
+ static int btrfs_update_reserved_bytes(struct btrfs_block_group_cache *cache,
+ u64 num_bytes, int reserve);
static noinline int
block_group_cache_done(struct btrfs_block_group_cache *cache)
if (atomic_dec_and_test(&cache->count)) {
WARN_ON(cache->pinned > 0);
WARN_ON(cache->reserved > 0);
- WARN_ON(cache->reserved_pinned > 0);
kfree(cache->free_space_ctl);
kfree(cache);
}
* we likely hold important locks.
*/
if (trans && (!trans->transaction->in_commit) &&
- (root && root != root->fs_info->tree_root)) {
+ (root && root != root->fs_info->tree_root) &&
+ btrfs_test_opt(root, SPACE_CACHE)) {
spin_lock(&cache->lock);
if (cache->cached != BTRFS_CACHE_NO) {
spin_unlock(&cache->lock);
{
int ret;
u64 discarded_bytes = 0;
- struct btrfs_multi_bio *multi = NULL;
+ struct btrfs_bio *bbio = NULL;
/* Tell the block device(s) that the sectors can be discarded */
ret = btrfs_map_block(&root->fs_info->mapping_tree, REQ_DISCARD,
- bytenr, &num_bytes, &multi, 0);
+ bytenr, &num_bytes, &bbio, 0);
if (!ret) {
- struct btrfs_bio_stripe *stripe = multi->stripes;
+ struct btrfs_bio_stripe *stripe = bbio->stripes;
int i;
- for (i = 0; i < multi->num_stripes; i++, stripe++) {
+ for (i = 0; i < bbio->num_stripes; i++, stripe++) {
if (!stripe->dev->can_discard)
continue;
*/
ret = 0;
}
- kfree(multi);
+ kfree(bbio);
}
if (actual_bytes)
goto again;
}
+ /* We've already setup this transaction, go ahead and exit */
+ if (block_group->cache_generation == trans->transid &&
+ i_size_read(inode)) {
+ dcs = BTRFS_DC_SETUP;
+ goto out_put;
+ }
+
/*
* We want to set the generation to 0, that way if anything goes wrong
* from here on out we know not to trust this cache when we load up next
if (!ret)
dcs = BTRFS_DC_SETUP;
btrfs_free_reserved_data_space(inode, num_pages);
+
out_put:
iput(inode);
out_free:
btrfs_release_path(path);
out:
spin_lock(&block_group->lock);
+ if (!ret)
+ block_group->cache_generation = trans->transid;
block_group->disk_cache_state = dcs;
spin_unlock(&block_group->lock);
return -ENOSPC;
}
data_sinfo->bytes_may_use += bytes;
- BTRFS_I(inode)->reserved_bytes += bytes;
spin_unlock(&data_sinfo->lock);
return 0;
}
/*
- * called when we are clearing an delalloc extent from the
- * inode's io_tree or there was an error for whatever reason
- * after calling btrfs_check_data_free_space
+ * Called if we need to clear a data reservation for this inode.
*/
void btrfs_free_reserved_data_space(struct inode *inode, u64 bytes)
{
data_sinfo = BTRFS_I(inode)->space_info;
spin_lock(&data_sinfo->lock);
data_sinfo->bytes_may_use -= bytes;
- BTRFS_I(inode)->reserved_bytes -= bytes;
spin_unlock(&data_sinfo->lock);
}
struct btrfs_space_info *sinfo, u64 alloc_bytes,
int force)
{
+ struct btrfs_block_rsv *global_rsv = &root->fs_info->global_block_rsv;
u64 num_bytes = sinfo->total_bytes - sinfo->bytes_readonly;
u64 num_allocated = sinfo->bytes_used + sinfo->bytes_reserved;
u64 thresh;
if (force == CHUNK_ALLOC_FORCE)
return 1;
+ /*
+ * We need to take into account the global rsv because for all intents
+ * and purposes it's used space. Don't worry about locking the
+ * global_rsv, it doesn't change except when the transaction commits.
+ */
+ num_allocated += global_rsv->size;
+
/*
* in limited mode, we want to have some free space up to
* about 1% of the FS size.
*/
if (force == CHUNK_ALLOC_LIMITED) {
- thresh = btrfs_super_total_bytes(&root->fs_info->super_copy);
+ thresh = btrfs_super_total_bytes(root->fs_info->super_copy);
thresh = max_t(u64, 64 * 1024 * 1024,
div_factor_fine(thresh, 1));
if (num_allocated + alloc_bytes < div_factor(num_bytes, 8))
return 0;
- thresh = btrfs_super_total_bytes(&root->fs_info->super_copy);
+ thresh = btrfs_super_total_bytes(root->fs_info->super_copy);
/* 256MB or 5% of the FS */
thresh = max_t(u64, 256 * 1024 * 1024, div_factor_fine(thresh, 5));
/*
* shrink metadata reservation for delalloc
*/
- static int shrink_delalloc(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, u64 to_reclaim, int sync)
+ static int shrink_delalloc(struct btrfs_root *root, u64 to_reclaim,
+ bool wait_ordered)
{
struct btrfs_block_rsv *block_rsv;
struct btrfs_space_info *space_info;
+ struct btrfs_trans_handle *trans;
u64 reserved;
u64 max_reclaim;
u64 reclaimed = 0;
long time_left;
- int nr_pages = (2 * 1024 * 1024) >> PAGE_CACHE_SHIFT;
+ unsigned long nr_pages = (2 * 1024 * 1024) >> PAGE_CACHE_SHIFT;
int loops = 0;
unsigned long progress;
+ trans = (struct btrfs_trans_handle *)current->journal_info;
block_rsv = &root->fs_info->delalloc_block_rsv;
space_info = block_rsv->space_info;
smp_mb();
- reserved = space_info->bytes_reserved;
+ reserved = space_info->bytes_may_use;
progress = space_info->reservation_progress;
if (reserved == 0)
}
max_reclaim = min(reserved, to_reclaim);
-
+ nr_pages = max_t(unsigned long, nr_pages,
+ max_reclaim >> PAGE_CACHE_SHIFT);
while (loops < 1024) {
/* have the flusher threads jump in and do some IO */
smp_mb();
nr_pages = min_t(unsigned long, nr_pages,
root->fs_info->delalloc_bytes >> PAGE_CACHE_SHIFT);
- writeback_inodes_sb_nr_if_idle(root->fs_info->sb, nr_pages);
+ writeback_inodes_sb_nr_if_idle(root->fs_info->sb, nr_pages,
+ WB_REASON_FS_FREE_SPACE);
spin_lock(&space_info->lock);
- if (reserved > space_info->bytes_reserved)
- reclaimed += reserved - space_info->bytes_reserved;
- reserved = space_info->bytes_reserved;
+ if (reserved > space_info->bytes_may_use)
+ reclaimed += reserved - space_info->bytes_may_use;
+ reserved = space_info->bytes_may_use;
spin_unlock(&space_info->lock);
loops++;
if (trans && trans->transaction->blocked)
return -EAGAIN;
- time_left = schedule_timeout_interruptible(1);
+ if (wait_ordered && !trans) {
+ btrfs_wait_ordered_extents(root, 0, 0);
+ } else {
+ time_left = schedule_timeout_interruptible(1);
- /* We were interrupted, exit */
- if (time_left)
- break;
+ /* We were interrupted, exit */
+ if (time_left)
+ break;
+ }
/* we've kicked the IO a few times, if anything has been freed,
* exit. There is no sense in looping here for a long time
}
}
- if (reclaimed >= to_reclaim && !trans)
- btrfs_wait_ordered_extents(root, 0, 0);
+
return reclaimed >= to_reclaim;
}
- /*
- * Retries tells us how many times we've called reserve_metadata_bytes. The
- * idea is if this is the first call (retries == 0) then we will add to our
- * reserved count if we can't make the allocation in order to hold our place
- * while we go and try and free up space. That way for retries > 1 we don't try
- * and add space, we just check to see if the amount of unused space is >= the
- * total space, meaning that our reservation is valid.
+ /**
+ * maybe_commit_transaction - possibly commit the transaction if its ok to
+ * @root - the root we're allocating for
+ * @bytes - the number of bytes we want to reserve
+ * @force - force the commit
*
- * However if we don't intend to retry this reservation, pass -1 as retries so
- * that it short circuits this logic.
+ * This will check to make sure that committing the transaction will actually
+ * get us somewhere and then commit the transaction if it does. Otherwise it
+ * will return -ENOSPC.
*/
- static int reserve_metadata_bytes(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
+ static int may_commit_transaction(struct btrfs_root *root,
+ struct btrfs_space_info *space_info,
+ u64 bytes, int force)
+ {
+ struct btrfs_block_rsv *delayed_rsv = &root->fs_info->delayed_block_rsv;
+ struct btrfs_trans_handle *trans;
+
+ trans = (struct btrfs_trans_handle *)current->journal_info;
+ if (trans)
+ return -EAGAIN;
+
+ if (force)
+ goto commit;
+
+ /* See if there is enough pinned space to make this reservation */
+ spin_lock(&space_info->lock);
+ if (space_info->bytes_pinned >= bytes) {
+ spin_unlock(&space_info->lock);
+ goto commit;
+ }
+ spin_unlock(&space_info->lock);
+
+ /*
+ * See if there is some space in the delayed insertion reservation for
+ * this reservation.
+ */
+ if (space_info != delayed_rsv->space_info)
+ return -ENOSPC;
+
+ spin_lock(&delayed_rsv->lock);
+ if (delayed_rsv->size < bytes) {
+ spin_unlock(&delayed_rsv->lock);
+ return -ENOSPC;
+ }
+ spin_unlock(&delayed_rsv->lock);
+
+ commit:
+ trans = btrfs_join_transaction(root);
+ if (IS_ERR(trans))
+ return -ENOSPC;
+
+ return btrfs_commit_transaction(trans, root);
+ }
+
+ /**
+ * reserve_metadata_bytes - try to reserve bytes from the block_rsv's space
+ * @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
+ *
+ * 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
+ * flush out space to make room. It will do this by flushing delalloc if
+ * possible or committing the transaction. If flush is 0 then no attempts to
+ * regain reservations will be made and this will fail if there is not enough
+ * space already.
+ */
+ static int reserve_metadata_bytes(struct btrfs_root *root,
struct btrfs_block_rsv *block_rsv,
u64 orig_bytes, int flush)
{
struct btrfs_space_info *space_info = block_rsv->space_info;
- u64 unused;
+ u64 used;
u64 num_bytes = orig_bytes;
int retries = 0;
int ret = 0;
bool committed = false;
bool flushing = false;
+ bool wait_ordered = false;
again:
ret = 0;
* deadlock since we are waiting for the flusher to finish, but
* hold the current transaction open.
*/
- if (trans)
+ if (current->journal_info)
return -EAGAIN;
ret = wait_event_interruptible(space_info->wait,
!space_info->flush);
}
ret = -ENOSPC;
- unused = space_info->bytes_used + space_info->bytes_reserved +
- space_info->bytes_pinned + space_info->bytes_readonly +
- space_info->bytes_may_use;
+ used = space_info->bytes_used + space_info->bytes_reserved +
+ space_info->bytes_pinned + space_info->bytes_readonly +
+ space_info->bytes_may_use;
/*
* The idea here is that we've not already over-reserved the block group
* lets start flushing stuff first and then come back and try to make
* our reservation.
*/
- if (unused <= space_info->total_bytes) {
- unused = space_info->total_bytes - unused;
- if (unused >= num_bytes) {
- space_info->bytes_reserved += orig_bytes;
+ if (used <= space_info->total_bytes) {
+ if (used + orig_bytes <= space_info->total_bytes) {
+ space_info->bytes_may_use += orig_bytes;
ret = 0;
} else {
/*
* amount plus the amount of bytes that we need for this
* reservation.
*/
- num_bytes = unused - space_info->total_bytes +
+ wait_ordered = true;
+ num_bytes = used - space_info->total_bytes +
(orig_bytes * (retries + 1));
}
+ if (ret) {
+ u64 profile = btrfs_get_alloc_profile(root, 0);
+ u64 avail;
+
+ /*
+ * If we have a lot of space that's pinned, don't bother doing
+ * the overcommit dance yet and just commit the transaction.
+ */
+ avail = (space_info->total_bytes - space_info->bytes_used) * 8;
+ do_div(avail, 10);
+ if (space_info->bytes_pinned >= avail && flush && !committed) {
+ space_info->flush = 1;
+ flushing = true;
+ spin_unlock(&space_info->lock);
+ ret = may_commit_transaction(root, space_info,
+ orig_bytes, 1);
+ if (ret)
+ goto out;
+ committed = true;
+ goto again;
+ }
+
+ spin_lock(&root->fs_info->free_chunk_lock);
+ avail = root->fs_info->free_chunk_space;
+
+ /*
+ * If we have dup, raid1 or raid10 then only half of the free
+ * space is actually useable.
+ */
+ if (profile & (BTRFS_BLOCK_GROUP_DUP |
+ BTRFS_BLOCK_GROUP_RAID1 |
+ BTRFS_BLOCK_GROUP_RAID10))
+ avail >>= 1;
+
+ /*
+ * If we aren't flushing don't let us overcommit too much, say
+ * 1/8th of the space. If we can flush, let it overcommit up to
+ * 1/2 of the space.
+ */
+ if (flush)
+ avail >>= 3;
+ else
+ avail >>= 1;
+ spin_unlock(&root->fs_info->free_chunk_lock);
+
+ if (used + num_bytes < space_info->total_bytes + avail) {
+ space_info->bytes_may_use += orig_bytes;
+ ret = 0;
+ } else {
+ wait_ordered = true;
+ }
+ }
+
/*
* Couldn't make our reservation, save our place so while we're trying
* to reclaim space we can actually use it instead of somebody else
* We do synchronous shrinking since we don't actually unreserve
* metadata until after the IO is completed.
*/
- ret = shrink_delalloc(trans, root, num_bytes, 1);
+ ret = shrink_delalloc(root, num_bytes, wait_ordered);
if (ret < 0)
goto out;
* so go back around and try again.
*/
if (retries < 2) {
+ wait_ordered = true;
retries++;
goto again;
}
- /*
- * Not enough space to be reclaimed, don't bother committing the
- * transaction.
- */
- spin_lock(&space_info->lock);
- if (space_info->bytes_pinned < orig_bytes)
- ret = -ENOSPC;
- spin_unlock(&space_info->lock);
- if (ret)
- goto out;
-
- ret = -EAGAIN;
- if (trans)
- goto out;
-
ret = -ENOSPC;
if (committed)
goto out;
- trans = btrfs_join_transaction(root);
- if (IS_ERR(trans))
- goto out;
- ret = btrfs_commit_transaction(trans, root);
+ ret = may_commit_transaction(root, space_info, orig_bytes, 0);
if (!ret) {
- trans = NULL;
committed = true;
goto again;
}
static struct btrfs_block_rsv *get_block_rsv(struct btrfs_trans_handle *trans,
struct btrfs_root *root)
{
- struct btrfs_block_rsv *block_rsv;
- if (root->ref_cows)
+ struct btrfs_block_rsv *block_rsv = NULL;
+
+ if (root->ref_cows || root == root->fs_info->csum_root)
block_rsv = trans->block_rsv;
- else
+
+ if (!block_rsv)
block_rsv = root->block_rsv;
if (!block_rsv)
}
if (num_bytes) {
spin_lock(&space_info->lock);
- space_info->bytes_reserved -= num_bytes;
+ space_info->bytes_may_use -= num_bytes;
space_info->reservation_progress++;
spin_unlock(&space_info->lock);
}
{
memset(rsv, 0, sizeof(*rsv));
spin_lock_init(&rsv->lock);
- atomic_set(&rsv->usage, 1);
- rsv->priority = 6;
- INIT_LIST_HEAD(&rsv->list);
}
struct btrfs_block_rsv *btrfs_alloc_block_rsv(struct btrfs_root *root)
void btrfs_free_block_rsv(struct btrfs_root *root,
struct btrfs_block_rsv *rsv)
{
- if (rsv && atomic_dec_and_test(&rsv->usage)) {
- btrfs_block_rsv_release(root, rsv, (u64)-1);
- if (!rsv->durable)
- kfree(rsv);
- }
+ btrfs_block_rsv_release(root, rsv, (u64)-1);
+ kfree(rsv);
}
- /*
- * make the block_rsv struct be able to capture freed space.
- * the captured space will re-add to the the block_rsv struct
- * after transaction commit
- */
- void btrfs_add_durable_block_rsv(struct btrfs_fs_info *fs_info,
- struct btrfs_block_rsv *block_rsv)
+ int btrfs_block_rsv_add(struct btrfs_root *root,
+ struct btrfs_block_rsv *block_rsv,
+ u64 num_bytes)
{
- block_rsv->durable = 1;
- mutex_lock(&fs_info->durable_block_rsv_mutex);
- list_add_tail(&block_rsv->list, &fs_info->durable_block_rsv_list);
- mutex_unlock(&fs_info->durable_block_rsv_mutex);
+ int ret;
+
+ if (num_bytes == 0)
+ return 0;
+
+ ret = reserve_metadata_bytes(root, block_rsv, num_bytes, 1);
+ if (!ret) {
+ block_rsv_add_bytes(block_rsv, num_bytes, 1);
+ return 0;
+ }
+
+ return ret;
}
- int btrfs_block_rsv_add(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- struct btrfs_block_rsv *block_rsv,
- u64 num_bytes)
+ int btrfs_block_rsv_add_noflush(struct btrfs_root *root,
+ struct btrfs_block_rsv *block_rsv,
+ u64 num_bytes)
{
int ret;
if (num_bytes == 0)
return 0;
- ret = reserve_metadata_bytes(trans, root, block_rsv, num_bytes, 1);
+ ret = reserve_metadata_bytes(root, block_rsv, num_bytes, 0);
if (!ret) {
block_rsv_add_bytes(block_rsv, num_bytes, 1);
return 0;
return ret;
}
- int btrfs_block_rsv_check(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- struct btrfs_block_rsv *block_rsv,
- u64 min_reserved, int min_factor)
+ int btrfs_block_rsv_check(struct btrfs_root *root,
+ struct btrfs_block_rsv *block_rsv, int min_factor)
{
u64 num_bytes = 0;
- int commit_trans = 0;
int ret = -ENOSPC;
if (!block_rsv)
return 0;
spin_lock(&block_rsv->lock);
- if (min_factor > 0)
- num_bytes = div_factor(block_rsv->size, min_factor);
- if (min_reserved > num_bytes)
- num_bytes = min_reserved;
+ num_bytes = div_factor(block_rsv->size, min_factor);
+ if (block_rsv->reserved >= num_bytes)
+ ret = 0;
+ spin_unlock(&block_rsv->lock);
- if (block_rsv->reserved >= num_bytes) {
+ return ret;
+ }
+
+ int btrfs_block_rsv_refill(struct btrfs_root *root,
+ struct btrfs_block_rsv *block_rsv,
+ u64 min_reserved)
+ {
+ u64 num_bytes = 0;
+ int ret = -ENOSPC;
+
+ if (!block_rsv)
+ return 0;
+
+ spin_lock(&block_rsv->lock);
+ num_bytes = min_reserved;
+ if (block_rsv->reserved >= num_bytes)
ret = 0;
- } else {
+ else
num_bytes -= block_rsv->reserved;
- if (block_rsv->durable &&
- block_rsv->freed[0] + block_rsv->freed[1] >= num_bytes)
- commit_trans = 1;
- }
spin_unlock(&block_rsv->lock);
+
if (!ret)
return 0;
- if (block_rsv->refill_used) {
- ret = reserve_metadata_bytes(trans, root, block_rsv,
- num_bytes, 0);
- if (!ret) {
- block_rsv_add_bytes(block_rsv, num_bytes, 0);
- return 0;
- }
- }
-
- if (commit_trans) {
- if (trans)
- return -EAGAIN;
- trans = btrfs_join_transaction(root);
- BUG_ON(IS_ERR(trans));
- ret = btrfs_commit_transaction(trans, root);
+ ret = reserve_metadata_bytes(root, block_rsv, num_bytes, 1);
+ if (!ret) {
+ block_rsv_add_bytes(block_rsv, num_bytes, 0);
return 0;
}
- return -ENOSPC;
+ return ret;
}
int btrfs_block_rsv_migrate(struct btrfs_block_rsv *src_rsv,
u64 num_bytes;
u64 meta_used;
u64 data_used;
- int csum_size = btrfs_super_csum_size(&fs_info->super_copy);
+ int csum_size = btrfs_super_csum_size(fs_info->super_copy);
sinfo = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_DATA);
spin_lock(&sinfo->lock);
if (sinfo->total_bytes > num_bytes) {
num_bytes = sinfo->total_bytes - num_bytes;
block_rsv->reserved += num_bytes;
- sinfo->bytes_reserved += num_bytes;
+ sinfo->bytes_may_use += num_bytes;
}
if (block_rsv->reserved >= block_rsv->size) {
num_bytes = block_rsv->reserved - block_rsv->size;
- sinfo->bytes_reserved -= num_bytes;
+ sinfo->bytes_may_use -= num_bytes;
sinfo->reservation_progress++;
block_rsv->reserved = block_rsv->size;
block_rsv->full = 1;
space_info = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_SYSTEM);
fs_info->chunk_block_rsv.space_info = space_info;
- fs_info->chunk_block_rsv.priority = 10;
space_info = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA);
fs_info->global_block_rsv.space_info = space_info;
- fs_info->global_block_rsv.priority = 10;
- fs_info->global_block_rsv.refill_used = 1;
fs_info->delalloc_block_rsv.space_info = space_info;
fs_info->trans_block_rsv.space_info = space_info;
fs_info->empty_block_rsv.space_info = space_info;
- fs_info->empty_block_rsv.priority = 10;
+ fs_info->delayed_block_rsv.space_info = space_info;
fs_info->extent_root->block_rsv = &fs_info->global_block_rsv;
fs_info->csum_root->block_rsv = &fs_info->global_block_rsv;
fs_info->tree_root->block_rsv = &fs_info->global_block_rsv;
fs_info->chunk_root->block_rsv = &fs_info->chunk_block_rsv;
- btrfs_add_durable_block_rsv(fs_info, &fs_info->global_block_rsv);
-
- btrfs_add_durable_block_rsv(fs_info, &fs_info->delalloc_block_rsv);
-
update_global_block_rsv(fs_info);
}
WARN_ON(fs_info->trans_block_rsv.reserved > 0);
WARN_ON(fs_info->chunk_block_rsv.size > 0);
WARN_ON(fs_info->chunk_block_rsv.reserved > 0);
- }
-
- int btrfs_truncate_reserve_metadata(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- struct btrfs_block_rsv *rsv)
- {
- struct btrfs_block_rsv *trans_rsv = &root->fs_info->trans_block_rsv;
- u64 num_bytes;
- int ret;
-
- /*
- * Truncate should be freeing data, but give us 2 items just in case it
- * needs to use some space. We may want to be smarter about this in the
- * future.
- */
- num_bytes = btrfs_calc_trans_metadata_size(root, 2);
-
- /* We already have enough bytes, just return */
- if (rsv->reserved >= num_bytes)
- return 0;
-
- num_bytes -= rsv->reserved;
-
- /*
- * You should have reserved enough space before hand to do this, so this
- * should not fail.
- */
- ret = block_rsv_migrate_bytes(trans_rsv, rsv, num_bytes);
- BUG_ON(ret);
-
- return 0;
+ WARN_ON(fs_info->delayed_block_rsv.size > 0);
+ WARN_ON(fs_info->delayed_block_rsv.reserved > 0);
}
void btrfs_trans_release_metadata(struct btrfs_trans_handle *trans,
if (!trans->bytes_reserved)
return;
- BUG_ON(trans->block_rsv != &root->fs_info->trans_block_rsv);
- btrfs_block_rsv_release(root, trans->block_rsv,
- trans->bytes_reserved);
+ btrfs_block_rsv_release(root, trans->block_rsv, trans->bytes_reserved);
trans->bytes_reserved = 0;
}
return block_rsv_migrate_bytes(src_rsv, dst_rsv, num_bytes);
}
+ /**
+ * drop_outstanding_extent - drop an outstanding extent
+ * @inode: the inode we're dropping the extent for
+ *
+ * This is called when we are freeing up an outstanding extent, either called
+ * after an error or after an extent is written. This will return the number of
+ * reserved extents that need to be freed. This must be called with
+ * BTRFS_I(inode)->lock held.
+ */
static unsigned drop_outstanding_extent(struct inode *inode)
{
unsigned dropped_extents = 0;
- spin_lock(&BTRFS_I(inode)->lock);
BUG_ON(!BTRFS_I(inode)->outstanding_extents);
BTRFS_I(inode)->outstanding_extents--;
*/
if (BTRFS_I(inode)->outstanding_extents >=
BTRFS_I(inode)->reserved_extents)
- goto out;
+ return 0;
dropped_extents = BTRFS_I(inode)->reserved_extents -
BTRFS_I(inode)->outstanding_extents;
BTRFS_I(inode)->reserved_extents -= dropped_extents;
- out:
- spin_unlock(&BTRFS_I(inode)->lock);
return dropped_extents;
}
- static u64 calc_csum_metadata_size(struct inode *inode, u64 num_bytes)
+ /**
+ * calc_csum_metadata_size - return the amount of metada space that must be
+ * reserved/free'd for the given bytes.
+ * @inode: the inode we're manipulating
+ * @num_bytes: the number of bytes in question
+ * @reserve: 1 if we are reserving space, 0 if we are freeing space
+ *
+ * This adjusts the number of csum_bytes in the inode and then returns the
+ * correct amount of metadata that must either be reserved or freed. We
+ * calculate how many checksums we can fit into one leaf and then divide the
+ * number of bytes that will need to be checksumed by this value to figure out
+ * how many checksums will be required. If we are adding bytes then the number
+ * may go up and we will return the number of additional bytes that must be
+ * reserved. If it is going down we will return the number of bytes that must
+ * be freed.
+ *
+ * This must be called with BTRFS_I(inode)->lock held.
+ */
+ static u64 calc_csum_metadata_size(struct inode *inode, u64 num_bytes,
+ int reserve)
{
- return num_bytes >>= 3;
+ struct btrfs_root *root = BTRFS_I(inode)->root;
+ u64 csum_size;
+ int num_csums_per_leaf;
+ int num_csums;
+ int old_csums;
+
+ if (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM &&
+ BTRFS_I(inode)->csum_bytes == 0)
+ return 0;
+
+ old_csums = (int)div64_u64(BTRFS_I(inode)->csum_bytes, root->sectorsize);
+ if (reserve)
+ BTRFS_I(inode)->csum_bytes += num_bytes;
+ else
+ BTRFS_I(inode)->csum_bytes -= num_bytes;
+ csum_size = BTRFS_LEAF_DATA_SIZE(root) - sizeof(struct btrfs_item);
+ num_csums_per_leaf = (int)div64_u64(csum_size,
+ sizeof(struct btrfs_csum_item) +
+ sizeof(struct btrfs_disk_key));
+ num_csums = (int)div64_u64(BTRFS_I(inode)->csum_bytes, root->sectorsize);
+ num_csums = num_csums + num_csums_per_leaf - 1;
+ num_csums = num_csums / num_csums_per_leaf;
+
+ old_csums = old_csums + num_csums_per_leaf - 1;
+ old_csums = old_csums / num_csums_per_leaf;
+
+ /* No change, no need to reserve more */
+ if (old_csums == num_csums)
+ return 0;
+
+ if (reserve)
+ return btrfs_calc_trans_metadata_size(root,
+ num_csums - old_csums);
+
+ return btrfs_calc_trans_metadata_size(root, old_csums - num_csums);
}
int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
struct btrfs_block_rsv *block_rsv = &root->fs_info->delalloc_block_rsv;
u64 to_reserve = 0;
unsigned nr_extents = 0;
+ int flush = 1;
int ret;
- if (btrfs_transaction_in_commit(root->fs_info))
+ if (btrfs_is_free_space_inode(root, inode))
+ flush = 0;
+
+ if (flush && btrfs_transaction_in_commit(root->fs_info))
schedule_timeout(1);
num_bytes = ALIGN(num_bytes, root->sectorsize);
to_reserve = btrfs_calc_trans_metadata_size(root, nr_extents);
}
+ to_reserve += calc_csum_metadata_size(inode, num_bytes, 1);
spin_unlock(&BTRFS_I(inode)->lock);
- to_reserve += calc_csum_metadata_size(inode, num_bytes);
- ret = reserve_metadata_bytes(NULL, root, block_rsv, to_reserve, 1);
+ ret = reserve_metadata_bytes(root, block_rsv, to_reserve, flush);
if (ret) {
+ u64 to_free = 0;
unsigned dropped;
+
+ spin_lock(&BTRFS_I(inode)->lock);
+ dropped = drop_outstanding_extent(inode);
+ to_free = calc_csum_metadata_size(inode, num_bytes, 0);
+ spin_unlock(&BTRFS_I(inode)->lock);
+ to_free += btrfs_calc_trans_metadata_size(root, dropped);
+
/*
- * We don't need the return value since our reservation failed,
- * we just need to clean up our counter.
+ * Somebody could have come in and twiddled with the
+ * reservation, so if we have to free more than we would have
+ * reserved from this reservation go ahead and release those
+ * bytes.
*/
- dropped = drop_outstanding_extent(inode);
- WARN_ON(dropped > 1);
+ to_free -= to_reserve;
+ if (to_free)
+ btrfs_block_rsv_release(root, block_rsv, to_free);
return ret;
}
return 0;
}
+ /**
+ * btrfs_delalloc_release_metadata - release a metadata reservation for an inode
+ * @inode: the inode to release the reservation for
+ * @num_bytes: the number of bytes we're releasing
+ *
+ * This will release the metadata reservation for an inode. This can be called
+ * once we complete IO for a given set of bytes to release their metadata
+ * reservations.
+ */
void btrfs_delalloc_release_metadata(struct inode *inode, u64 num_bytes)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
unsigned dropped;
num_bytes = ALIGN(num_bytes, root->sectorsize);
+ spin_lock(&BTRFS_I(inode)->lock);
dropped = drop_outstanding_extent(inode);
- to_free = calc_csum_metadata_size(inode, num_bytes);
+ to_free = calc_csum_metadata_size(inode, num_bytes, 0);
+ spin_unlock(&BTRFS_I(inode)->lock);
if (dropped > 0)
to_free += btrfs_calc_trans_metadata_size(root, dropped);
to_free);
}
+ /**
+ * btrfs_delalloc_reserve_space - reserve data and metadata space for delalloc
+ * @inode: inode we're writing to
+ * @num_bytes: the number of bytes we want to allocate
+ *
+ * This will do the following things
+ *
+ * o reserve space in the data space info for num_bytes
+ * o reserve space in the metadata space info based on number of outstanding
+ * extents and how much csums will be needed
+ * o add to the inodes ->delalloc_bytes
+ * o add it to the fs_info's delalloc inodes list.
+ *
+ * This will return 0 for success and -ENOSPC if there is no space left.
+ */
int btrfs_delalloc_reserve_space(struct inode *inode, u64 num_bytes)
{
int ret;
return 0;
}
+ /**
+ * btrfs_delalloc_release_space - release data and metadata space for delalloc
+ * @inode: inode we're releasing space for
+ * @num_bytes: the number of bytes we want to free up
+ *
+ * This must be matched with a call to btrfs_delalloc_reserve_space. This is
+ * called in the case that we don't need the metadata AND data reservations
+ * anymore. So if there is an error or we insert an inline extent.
+ *
+ * This function will release the metadata space that was not used and will
+ * decrement ->delalloc_bytes and remove it from the fs_info delalloc_inodes
+ * list if there are no delalloc bytes left.
+ */
void btrfs_delalloc_release_space(struct inode *inode, u64 num_bytes)
{
btrfs_delalloc_release_metadata(inode, num_bytes);
/* block accounting for super block */
spin_lock(&info->delalloc_lock);
- old_val = btrfs_super_bytes_used(&info->super_copy);
+ old_val = btrfs_super_bytes_used(info->super_copy);
if (alloc)
old_val += num_bytes;
else
old_val -= num_bytes;
- btrfs_set_super_bytes_used(&info->super_copy, old_val);
+ btrfs_set_super_bytes_used(info->super_copy, old_val);
spin_unlock(&info->delalloc_lock);
while (total) {
spin_lock(&cache->space_info->lock);
spin_lock(&cache->lock);
- if (btrfs_super_cache_generation(&info->super_copy) != 0 &&
+ if (btrfs_test_opt(root, SPACE_CACHE) &&
cache->disk_cache_state < BTRFS_DC_CLEAR)
cache->disk_cache_state = BTRFS_DC_CLEAR;
btrfs_set_block_group_used(&cache->item, old_val);
cache->reserved -= num_bytes;
cache->space_info->bytes_reserved -= num_bytes;
- cache->space_info->reservation_progress++;
cache->space_info->bytes_used += num_bytes;
cache->space_info->disk_used += num_bytes * factor;
spin_unlock(&cache->lock);
if (reserved) {
cache->reserved -= num_bytes;
cache->space_info->bytes_reserved -= num_bytes;
- cache->space_info->reservation_progress++;
}
spin_unlock(&cache->lock);
spin_unlock(&cache->space_info->lock);
}
/*
- * update size of reserved extents. this function may return -EAGAIN
- * if 'reserve' is true or 'sinfo' is false.
+ * this function must be called within transaction
+ */
+ int btrfs_pin_extent_for_log_replay(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root,
+ u64 bytenr, u64 num_bytes)
+ {
+ struct btrfs_block_group_cache *cache;
+
+ cache = btrfs_lookup_block_group(root->fs_info, bytenr);
+ BUG_ON(!cache);
+
+ /*
+ * pull in the free space cache (if any) so that our pin
+ * removes the free space from the cache. We have load_only set
+ * to one because the slow code to read in the free extents does check
+ * the pinned extents.
+ */
+ cache_block_group(cache, trans, root, 1);
+
+ pin_down_extent(root, cache, bytenr, num_bytes, 0);
+
+ /* remove us from the free space cache (if we're there at all) */
+ btrfs_remove_free_space(cache, bytenr, num_bytes);
+ btrfs_put_block_group(cache);
+ return 0;
+ }
+
+ /**
+ * btrfs_update_reserved_bytes - update the block_group and space info counters
+ * @cache: The cache we are manipulating
+ * @num_bytes: The number of bytes in question
+ * @reserve: One of the reservation enums
+ *
+ * This is called by the allocator when it reserves space, or by somebody who is
+ * freeing space that was never actually used on disk. For example if you
+ * reserve some space for a new leaf in transaction A and before transaction A
+ * commits you free that leaf, you call this with reserve set to 0 in order to
+ * clear the reservation.
+ *
+ * Metadata reservations should be called with RESERVE_ALLOC so we do the proper
+ * ENOSPC accounting. For data we handle the reservation through clearing the
+ * delalloc bits in the io_tree. We have to do this since we could end up
+ * allocating less disk space for the amount of data we have reserved in the
+ * case of compression.
+ *
+ * If this is a reservation and the block group has become read only we cannot
+ * make the reservation and return -EAGAIN, otherwise this function always
+ * succeeds.
*/
- int btrfs_update_reserved_bytes(struct btrfs_block_group_cache *cache,
- u64 num_bytes, int reserve, int sinfo)
+ static int btrfs_update_reserved_bytes(struct btrfs_block_group_cache *cache,
+ u64 num_bytes, int reserve)
{
+ struct btrfs_space_info *space_info = cache->space_info;
int ret = 0;
- if (sinfo) {
- struct btrfs_space_info *space_info = cache->space_info;
- spin_lock(&space_info->lock);
- spin_lock(&cache->lock);
- if (reserve) {
- if (cache->ro) {
- ret = -EAGAIN;
- } else {
- cache->reserved += num_bytes;
- space_info->bytes_reserved += num_bytes;
- }
- } else {
- if (cache->ro)
- space_info->bytes_readonly += num_bytes;
- cache->reserved -= num_bytes;
- space_info->bytes_reserved -= num_bytes;
- space_info->reservation_progress++;
- }
- spin_unlock(&cache->lock);
- spin_unlock(&space_info->lock);
- } else {
- spin_lock(&cache->lock);
+ spin_lock(&space_info->lock);
+ spin_lock(&cache->lock);
+ if (reserve != RESERVE_FREE) {
if (cache->ro) {
ret = -EAGAIN;
} else {
- if (reserve)
- cache->reserved += num_bytes;
- else
- cache->reserved -= num_bytes;
+ cache->reserved += num_bytes;
+ space_info->bytes_reserved += num_bytes;
+ if (reserve == RESERVE_ALLOC) {
+ BUG_ON(space_info->bytes_may_use < num_bytes);
+ space_info->bytes_may_use -= num_bytes;
+ }
}
- spin_unlock(&cache->lock);
+ } else {
+ if (cache->ro)
+ space_info->bytes_readonly += num_bytes;
+ cache->reserved -= num_bytes;
+ space_info->bytes_reserved -= num_bytes;
+ space_info->reservation_progress++;
}
+ spin_unlock(&cache->lock);
+ spin_unlock(&space_info->lock);
return ret;
}
spin_lock(&cache->lock);
cache->pinned -= len;
cache->space_info->bytes_pinned -= len;
- if (cache->ro) {
+ if (cache->ro)
cache->space_info->bytes_readonly += len;
- } else if (cache->reserved_pinned > 0) {
- len = min(len, cache->reserved_pinned);
- cache->reserved_pinned -= len;
- cache->space_info->bytes_reserved += len;
- }
spin_unlock(&cache->lock);
spin_unlock(&cache->space_info->lock);
}
{
struct btrfs_fs_info *fs_info = root->fs_info;
struct extent_io_tree *unpin;
- struct btrfs_block_rsv *block_rsv;
- struct btrfs_block_rsv *next_rsv;
u64 start;
u64 end;
- int idx;
int ret;
if (fs_info->pinned_extents == &fs_info->freed_extents[0])
cond_resched();
}
- mutex_lock(&fs_info->durable_block_rsv_mutex);
- list_for_each_entry_safe(block_rsv, next_rsv,
- &fs_info->durable_block_rsv_list, list) {
-
- idx = trans->transid & 0x1;
- if (block_rsv->freed[idx] > 0) {
- block_rsv_add_bytes(block_rsv,
- block_rsv->freed[idx], 0);
- block_rsv->freed[idx] = 0;
- }
- if (atomic_read(&block_rsv->usage) == 0) {
- btrfs_block_rsv_release(root, block_rsv, (u64)-1);
-
- if (block_rsv->freed[0] == 0 &&
- block_rsv->freed[1] == 0) {
- list_del_init(&block_rsv->list);
- kfree(block_rsv);
- }
- } else {
- btrfs_block_rsv_release(root, block_rsv, 0);
- }
- }
- mutex_unlock(&fs_info->durable_block_rsv_mutex);
-
return 0;
}
struct extent_buffer *buf,
u64 parent, int last_ref)
{
- struct btrfs_block_rsv *block_rsv;
struct btrfs_block_group_cache *cache = NULL;
int ret;
if (!last_ref)
return;
- block_rsv = get_block_rsv(trans, root);
cache = btrfs_lookup_block_group(root->fs_info, buf->start);
- if (block_rsv->space_info != cache->space_info)
- goto out;
if (btrfs_header_generation(buf) == trans->transid) {
if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) {
ret = check_ref_cleanup(trans, root, buf->start);
if (!ret)
- goto pin;
+ goto out;
}
if (btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN)) {
pin_down_extent(root, cache, buf->start, buf->len, 1);
- goto pin;
+ goto out;
}
WARN_ON(test_bit(EXTENT_BUFFER_DIRTY, &buf->bflags));
btrfs_add_free_space(cache, buf->start, buf->len);
- ret = btrfs_update_reserved_bytes(cache, buf->len, 0, 0);
- if (ret == -EAGAIN) {
- /* block group became read-only */
- btrfs_update_reserved_bytes(cache, buf->len, 0, 1);
- goto out;
- }
-
- ret = 1;
- spin_lock(&block_rsv->lock);
- if (block_rsv->reserved < block_rsv->size) {
- block_rsv->reserved += buf->len;
- ret = 0;
- }
- spin_unlock(&block_rsv->lock);
-
- if (ret) {
- spin_lock(&cache->space_info->lock);
- cache->space_info->bytes_reserved -= buf->len;
- cache->space_info->reservation_progress++;
- spin_unlock(&cache->space_info->lock);
- }
- goto out;
- }
- pin:
- if (block_rsv->durable && !cache->ro) {
- ret = 0;
- spin_lock(&cache->lock);
- if (!cache->ro) {
- cache->reserved_pinned += buf->len;
- ret = 1;
- }
- spin_unlock(&cache->lock);
-
- if (ret) {
- spin_lock(&block_rsv->lock);
- block_rsv->freed[trans->transid & 0x1] += buf->len;
- spin_unlock(&block_rsv->lock);
- }
+ btrfs_update_reserved_bytes(cache, buf->len, RESERVE_FREE);
}
out:
/*
int last_ptr_loop = 0;
int loop = 0;
int index = 0;
+ int alloc_type = (data & BTRFS_BLOCK_GROUP_DATA) ?
+ RESERVE_ALLOC_NO_ACCOUNT : RESERVE_ALLOC;
bool found_uncached_bg = false;
bool failed_cluster_refill = false;
bool failed_alloc = false;
bool use_cluster = true;
+ bool have_caching_bg = false;
u64 ideal_cache_percent = 0;
u64 ideal_cache_offset = 0;
}
}
search:
+ have_caching_bg = false;
down_read(&space_info->groups_sem);
list_for_each_entry(block_group, &space_info->block_groups[index],
list) {
failed_alloc = true;
goto have_block_group;
} else if (!offset) {
+ if (!cached)
+ have_caching_bg = true;
goto loop;
}
checks:
search_start - offset);
BUG_ON(offset > search_start);
- ret = btrfs_update_reserved_bytes(block_group, num_bytes, 1,
- (data & BTRFS_BLOCK_GROUP_DATA));
+ ret = btrfs_update_reserved_bytes(block_group, num_bytes,
+ alloc_type);
if (ret == -EAGAIN) {
btrfs_add_free_space(block_group, offset, num_bytes);
goto loop;
}
up_read(&space_info->groups_sem);
+ if (!ins->objectid && loop >= LOOP_CACHING_WAIT && have_caching_bg)
+ goto search;
+
if (!ins->objectid && ++index < BTRFS_NR_RAID_TYPES)
goto search;
int index = 0;
spin_lock(&info->lock);
- printk(KERN_INFO "space_info has %llu free, is %sfull\n",
+ printk(KERN_INFO "space_info %llu has %llu free, is %sfull\n",
+ (unsigned long long)info->flags,
(unsigned long long)(info->total_bytes - info->bytes_used -
info->bytes_pinned - info->bytes_reserved -
info->bytes_readonly),
return ret;
}
- int btrfs_free_reserved_extent(struct btrfs_root *root, u64 start, u64 len)
+ static int __btrfs_free_reserved_extent(struct btrfs_root *root,
+ u64 start, u64 len, int pin)
{
struct btrfs_block_group_cache *cache;
int ret = 0;
if (btrfs_test_opt(root, DISCARD))
ret = btrfs_discard_extent(root, start, len, NULL);
- btrfs_add_free_space(cache, start, len);
- btrfs_update_reserved_bytes(cache, len, 0, 1);
+ if (pin)
+ pin_down_extent(root, cache, start, len, 1);
+ else {
+ btrfs_add_free_space(cache, start, len);
+ btrfs_update_reserved_bytes(cache, len, RESERVE_FREE);
+ }
btrfs_put_block_group(cache);
trace_btrfs_reserved_extent_free(root, start, len);
return ret;
}
+ int btrfs_free_reserved_extent(struct btrfs_root *root,
+ u64 start, u64 len)
+ {
+ return __btrfs_free_reserved_extent(root, start, len, 0);
+ }
+
+ int btrfs_free_and_pin_reserved_extent(struct btrfs_root *root,
+ u64 start, u64 len)
+ {
+ return __btrfs_free_reserved_extent(root, start, len, 1);
+ }
+
static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
u64 parent, u64 root_objectid,
put_caching_control(caching_ctl);
}
- ret = btrfs_update_reserved_bytes(block_group, ins->offset, 1, 1);
+ ret = btrfs_update_reserved_bytes(block_group, ins->offset,
+ RESERVE_ALLOC_NO_ACCOUNT);
BUG_ON(ret);
btrfs_put_block_group(block_group);
ret = alloc_reserved_file_extent(trans, root, 0, root_objectid,
block_rsv = get_block_rsv(trans, root);
if (block_rsv->size == 0) {
- ret = reserve_metadata_bytes(trans, root, block_rsv,
- blocksize, 0);
+ ret = reserve_metadata_bytes(root, block_rsv, blocksize, 0);
/*
* If we couldn't reserve metadata bytes try and use some from
* the global reserve.
if (!ret)
return block_rsv;
if (ret) {
- WARN_ON(1);
- ret = reserve_metadata_bytes(trans, root, block_rsv, blocksize,
- 0);
+ static DEFINE_RATELIMIT_STATE(_rs,
+ DEFAULT_RATELIMIT_INTERVAL,
+ /*DEFAULT_RATELIMIT_BURST*/ 2);
+ if (__ratelimit(&_rs)) {
+ printk(KERN_DEBUG "btrfs: block rsv returned %d\n", ret);
+ WARN_ON(1);
+ }
+ ret = reserve_metadata_bytes(root, block_rsv, blocksize, 0);
if (!ret) {
- spin_lock(&block_rsv->lock);
- block_rsv->size += blocksize;
- spin_unlock(&block_rsv->lock);
return block_rsv;
} else if (ret && block_rsv != global_rsv) {
ret = block_rsv_use_bytes(global_rsv, blocksize);
cache->bytes_super - btrfs_block_group_used(&cache->item);
if (sinfo->bytes_used + sinfo->bytes_reserved + sinfo->bytes_pinned +
- sinfo->bytes_may_use + sinfo->bytes_readonly +
- cache->reserved_pinned + num_bytes + min_allocable_bytes <=
- sinfo->total_bytes) {
+ sinfo->bytes_may_use + sinfo->bytes_readonly + num_bytes +
+ min_allocable_bytes <= sinfo->total_bytes) {
sinfo->bytes_readonly += num_bytes;
- sinfo->bytes_reserved += cache->reserved_pinned;
- cache->reserved_pinned = 0;
cache->ro = 1;
ret = 0;
}
struct btrfs_space_info,
list);
if (space_info->bytes_pinned > 0 ||
- space_info->bytes_reserved > 0) {
+ space_info->bytes_reserved > 0 ||
+ space_info->bytes_may_use > 0) {
WARN_ON(1);
dump_space_info(space_info, 0, 0);
}
return -ENOMEM;
path->reada = 1;
- cache_gen = btrfs_super_cache_generation(&root->fs_info->super_copy);
- if (cache_gen != 0 &&
- btrfs_super_generation(&root->fs_info->super_copy) != cache_gen)
+ cache_gen = btrfs_super_cache_generation(root->fs_info->super_copy);
+ if (btrfs_test_opt(root, SPACE_CACHE) &&
+ btrfs_super_generation(root->fs_info->super_copy) != cache_gen)
need_clear = 1;
if (btrfs_test_opt(root, CLEAR_CACHE))
need_clear = 1;
- if (!btrfs_test_opt(root, SPACE_CACHE) && cache_gen)
- printk(KERN_INFO "btrfs: disk space caching is enabled\n");
while (1) {
ret = find_first_block_group(root, path, &key);
goto out;
}
- inode = lookup_free_space_inode(root, block_group, path);
+ inode = lookup_free_space_inode(tree_root, block_group, path);
if (!IS_ERR(inode)) {
ret = btrfs_orphan_add(trans, inode);
BUG_ON(ret);
spin_unlock(&block_group->lock);
}
/* One for our lookup ref */
- iput(inode);
+ btrfs_add_delayed_iput(inode);
}
key.objectid = BTRFS_FREE_SPACE_OBJECTID;
int mixed = 0;
int ret;
- disk_super = &fs_info->super_copy;
+ disk_super = fs_info->super_copy;
if (!btrfs_super_root(disk_super))
return 1;
int i;
unsigned long index = pos >> PAGE_CACHE_SHIFT;
struct inode *inode = fdentry(file)->d_inode;
+ gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping);
int err = 0;
int faili = 0;
u64 start_pos;
again:
for (i = 0; i < num_pages; i++) {
pages[i] = find_or_create_page(inode->i_mapping, index + i,
- GFP_NOFS);
+ mask);
if (!pages[i]) {
faili = i - 1;
err = -ENOMEM;
goto out;
}
- ret = btrfs_check_data_free_space(inode, alloc_end - alloc_start);
- if (ret)
- goto out;
-
locked_end = alloc_end - 1;
while (1) {
struct btrfs_ordered_extent *ordered;
if (em->block_start == EXTENT_MAP_HOLE ||
(cur_offset >= inode->i_size &&
!test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) {
+
+ /*
+ * Make sure we have enough space before we do the
+ * allocation.
+ */
+ ret = btrfs_check_data_free_space(inode, last_byte -
+ cur_offset);
+ if (ret) {
+ free_extent_map(em);
+ break;
+ }
+
ret = btrfs_prealloc_file_range(inode, mode, cur_offset,
last_byte - cur_offset,
1 << inode->i_blkbits,
offset + len,
&alloc_hint);
+
+ /* Let go of our reservation. */
+ btrfs_free_reserved_data_space(inode, last_byte -
+ cur_offset);
if (ret < 0) {
free_extent_map(em);
break;
}
unlock_extent_cached(&BTRFS_I(inode)->io_tree, alloc_start, locked_end,
&cached_state, GFP_NOFS);
-
- btrfs_free_reserved_data_space(inode, alloc_end - alloc_start);
out:
mutex_unlock(&inode->i_mutex);
return ret;
switch (origin) {
case SEEK_END:
case SEEK_CUR:
- offset = generic_file_llseek_unlocked(file, offset, origin);
+ offset = generic_file_llseek(file, offset, origin);
goto out;
case SEEK_DATA:
case SEEK_HOLE:
#include "btrfs_inode.h"
#include "ioctl.h"
#include "print-tree.h"
- #include "volumes.h"
#include "ordered-data.h"
#include "xattr.h"
#include "tree-log.h"
+ #include "volumes.h"
#include "compression.h"
#include "locking.h"
#include "free-space-cache.h"
(BTRFS_I(inode)->flags & BTRFS_INODE_COMPRESS))) {
WARN_ON(pages);
pages = kzalloc(sizeof(struct page *) * nr_pages, GFP_NOFS);
- BUG_ON(!pages);
+ if (!pages) {
+ /* just bail out to the uncompressed code */
+ goto cont;
+ }
if (BTRFS_I(inode)->force_compress)
compress_type = BTRFS_I(inode)->force_compress;
will_compress = 1;
}
}
+ cont:
if (start == 0) {
trans = btrfs_join_transaction(root);
BUG_ON(IS_ERR(trans));
}
BUG_ON(disk_num_bytes >
- btrfs_super_total_bytes(&root->fs_info->super_copy));
+ btrfs_super_total_bytes(root->fs_info->super_copy));
alloc_hint = get_extent_allocation_hint(inode, start, num_bytes);
btrfs_drop_extent_cache(inode, start, start + num_bytes - 1, 0);
}
ret = 0;
out:
- if (nolock) {
- if (trans)
- btrfs_end_transaction_nolock(trans, root);
- } else {
+ if (root != root->fs_info->tree_root)
btrfs_delalloc_release_metadata(inode, ordered_extent->len);
- if (trans)
+ if (trans) {
+ if (nolock)
+ btrfs_end_transaction_nolock(trans, root);
+ else
btrfs_end_transaction(trans, root);
}
return btrfs_finish_ordered_io(page->mapping->host, start, end);
}
- /*
- * When IO fails, either with EIO or csum verification fails, we
- * try other mirrors that might have a good copy of the data. This
- * io_failure_record is used to record state as we go through all the
- * mirrors. If another mirror has good data, the page is set up to date
- * and things continue. If a good mirror can't be found, the original
- * bio end_io callback is called to indicate things have failed.
- */
- struct io_failure_record {
- struct page *page;
- u64 start;
- u64 len;
- u64 logical;
- unsigned long bio_flags;
- int last_mirror;
- };
-
- static int btrfs_io_failed_hook(struct bio *failed_bio,
- struct page *page, u64 start, u64 end,
- struct extent_state *state)
- {
- struct io_failure_record *failrec = NULL;
- u64 private;
- struct extent_map *em;
- struct inode *inode = page->mapping->host;
- struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;
- struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
- struct bio *bio;
- int num_copies;
- int ret;
- int rw;
- u64 logical;
-
- ret = get_state_private(failure_tree, start, &private);
- if (ret) {
- failrec = kmalloc(sizeof(*failrec), GFP_NOFS);
- if (!failrec)
- return -ENOMEM;
- failrec->start = start;
- failrec->len = end - start + 1;
- failrec->last_mirror = 0;
- failrec->bio_flags = 0;
-
- read_lock(&em_tree->lock);
- em = lookup_extent_mapping(em_tree, start, failrec->len);
- if (em->start > start || em->start + em->len < start) {
- free_extent_map(em);
- em = NULL;
- }
- read_unlock(&em_tree->lock);
-
- if (IS_ERR_OR_NULL(em)) {
- kfree(failrec);
- return -EIO;
- }
- logical = start - em->start;
- logical = em->block_start + logical;
- if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) {
- logical = em->block_start;
- failrec->bio_flags = EXTENT_BIO_COMPRESSED;
- extent_set_compress_type(&failrec->bio_flags,
- em->compress_type);
- }
- failrec->logical = logical;
- free_extent_map(em);
- set_extent_bits(failure_tree, start, end, EXTENT_LOCKED |
- EXTENT_DIRTY, GFP_NOFS);
- set_state_private(failure_tree, start,
- (u64)(unsigned long)failrec);
- } else {
- failrec = (struct io_failure_record *)(unsigned long)private;
- }
- num_copies = btrfs_num_copies(
- &BTRFS_I(inode)->root->fs_info->mapping_tree,
- failrec->logical, failrec->len);
- failrec->last_mirror++;
- if (!state) {
- spin_lock(&BTRFS_I(inode)->io_tree.lock);
- state = find_first_extent_bit_state(&BTRFS_I(inode)->io_tree,
- failrec->start,
- EXTENT_LOCKED);
- if (state && state->start != failrec->start)
- state = NULL;
- spin_unlock(&BTRFS_I(inode)->io_tree.lock);
- }
- if (!state || failrec->last_mirror > num_copies) {
- set_state_private(failure_tree, failrec->start, 0);
- clear_extent_bits(failure_tree, failrec->start,
- failrec->start + failrec->len - 1,
- EXTENT_LOCKED | EXTENT_DIRTY, GFP_NOFS);
- kfree(failrec);
- return -EIO;
- }
- bio = bio_alloc(GFP_NOFS, 1);
- bio->bi_private = state;
- bio->bi_end_io = failed_bio->bi_end_io;
- bio->bi_sector = failrec->logical >> 9;
- bio->bi_bdev = failed_bio->bi_bdev;
- bio->bi_size = 0;
-
- bio_add_page(bio, page, failrec->len, start - page_offset(page));
- if (failed_bio->bi_rw & REQ_WRITE)
- rw = WRITE;
- else
- rw = READ;
-
- ret = BTRFS_I(inode)->io_tree.ops->submit_bio_hook(inode, rw, bio,
- failrec->last_mirror,
- failrec->bio_flags, 0);
- return ret;
- }
-
- /*
- * each time an IO finishes, we do a fast check in the IO failure tree
- * to see if we need to process or clean up an io_failure_record
- */
- static int btrfs_clean_io_failures(struct inode *inode, u64 start)
- {
- u64 private;
- u64 private_failure;
- struct io_failure_record *failure;
- int ret;
-
- private = 0;
- if (count_range_bits(&BTRFS_I(inode)->io_failure_tree, &private,
- (u64)-1, 1, EXTENT_DIRTY, 0)) {
- ret = get_state_private(&BTRFS_I(inode)->io_failure_tree,
- start, &private_failure);
- if (ret == 0) {
- failure = (struct io_failure_record *)(unsigned long)
- private_failure;
- set_state_private(&BTRFS_I(inode)->io_failure_tree,
- failure->start, 0);
- clear_extent_bits(&BTRFS_I(inode)->io_failure_tree,
- failure->start,
- failure->start + failure->len - 1,
- EXTENT_DIRTY | EXTENT_LOCKED,
- GFP_NOFS);
- kfree(failure);
- }
- }
- return 0;
- }
-
/*
* when reads are done, we need to check csums to verify the data is correct
- * if there's a match, we allow the bio to finish. If not, we go through
- * the io_failure_record routines to find good copies
+ * if there's a match, we allow the bio to finish. If not, the code in
+ * extent_io.c will try to find good copies for us.
*/
static int btrfs_readpage_end_io_hook(struct page *page, u64 start, u64 end,
struct extent_state *state)
kunmap_atomic(kaddr, KM_USER0);
good:
- /* if the io failure tree for this inode is non-empty,
- * check to see if we've recovered from a failed IO
- */
- btrfs_clean_io_failures(inode, start);
return 0;
zeroit:
up_read(&root->fs_info->cleanup_work_sem);
}
- /*
- * calculate extra metadata reservation when snapshotting a subvolume
- * contains orphan files.
- */
- void btrfs_orphan_pre_snapshot(struct btrfs_trans_handle *trans,
- struct btrfs_pending_snapshot *pending,
- u64 *bytes_to_reserve)
- {
- struct btrfs_root *root;
- struct btrfs_block_rsv *block_rsv;
- u64 num_bytes;
- int index;
-
- root = pending->root;
- if (!root->orphan_block_rsv || list_empty(&root->orphan_list))
- return;
-
- block_rsv = root->orphan_block_rsv;
-
- /* orphan block reservation for the snapshot */
- num_bytes = block_rsv->size;
-
- /*
- * after the snapshot is created, COWing tree blocks may use more
- * space than it frees. So we should make sure there is enough
- * reserved space.
- */
- index = trans->transid & 0x1;
- if (block_rsv->reserved + block_rsv->freed[index] < block_rsv->size) {
- num_bytes += block_rsv->size -
- (block_rsv->reserved + block_rsv->freed[index]);
- }
-
- *bytes_to_reserve += num_bytes;
- }
-
- void btrfs_orphan_post_snapshot(struct btrfs_trans_handle *trans,
- struct btrfs_pending_snapshot *pending)
- {
- struct btrfs_root *root = pending->root;
- struct btrfs_root *snap = pending->snap;
- struct btrfs_block_rsv *block_rsv;
- u64 num_bytes;
- int index;
- int ret;
-
- if (!root->orphan_block_rsv || list_empty(&root->orphan_list))
- return;
-
- /* refill source subvolume's orphan block reservation */
- block_rsv = root->orphan_block_rsv;
- index = trans->transid & 0x1;
- if (block_rsv->reserved + block_rsv->freed[index] < block_rsv->size) {
- num_bytes = block_rsv->size -
- (block_rsv->reserved + block_rsv->freed[index]);
- ret = btrfs_block_rsv_migrate(&pending->block_rsv,
- root->orphan_block_rsv,
- num_bytes);
- BUG_ON(ret);
- }
-
- /* setup orphan block reservation for the snapshot */
- block_rsv = btrfs_alloc_block_rsv(snap);
- BUG_ON(!block_rsv);
-
- btrfs_add_durable_block_rsv(root->fs_info, block_rsv);
- snap->orphan_block_rsv = block_rsv;
-
- num_bytes = root->orphan_block_rsv->size;
- ret = btrfs_block_rsv_migrate(&pending->block_rsv,
- block_rsv, num_bytes);
- BUG_ON(ret);
-
- #if 0
- /* insert orphan item for the snapshot */
- WARN_ON(!root->orphan_item_inserted);
- ret = btrfs_insert_orphan_item(trans, root->fs_info->tree_root,
- snap->root_key.objectid);
- BUG_ON(ret);
- snap->orphan_item_inserted = 1;
- #endif
- }
-
enum btrfs_orphan_cleanup_state {
ORPHAN_CLEANUP_STARTED = 1,
ORPHAN_CLEANUP_DONE = 2,
}
spin_unlock(&root->orphan_lock);
- if (block_rsv)
- btrfs_add_durable_block_rsv(root->fs_info, block_rsv);
-
/* grab metadata reservation from transaction handle */
if (reserve) {
ret = btrfs_orphan_reserve_metadata(trans, inode);
struct btrfs_key key, found_key;
struct btrfs_trans_handle *trans;
struct inode *inode;
+ u64 last_objectid = 0;
int ret = 0, nr_unlink = 0, nr_truncate = 0;
if (cmpxchg(&root->orphan_cleanup_state, 0, ORPHAN_CLEANUP_STARTED))
* crossing root thing. we store the inode number in the
* offset of the orphan item.
*/
+
+ if (found_key.offset == last_objectid) {
+ printk(KERN_ERR "btrfs: Error removing orphan entry, "
+ "stopping orphan cleanup\n");
+ ret = -EINVAL;
+ goto out;
+ }
+
+ last_objectid = found_key.offset;
+
found_key.objectid = found_key.offset;
found_key.type = BTRFS_INODE_ITEM_KEY;
found_key.offset = 0;
inode = btrfs_iget(root->fs_info->sb, &found_key, root, NULL);
- if (IS_ERR(inode)) {
- ret = PTR_ERR(inode);
+ ret = PTR_RET(inode);
+ if (ret && ret != -ESTALE)
goto out;
- }
/*
- * add this inode to the orphan list so btrfs_orphan_del does
- * the proper thing when we hit it
+ * Inode is already gone but the orphan item is still there,
+ * kill the orphan item.
*/
- spin_lock(&root->orphan_lock);
- list_add(&BTRFS_I(inode)->i_orphan, &root->orphan_list);
- spin_unlock(&root->orphan_lock);
-
- /*
- * if this is a bad inode, means we actually succeeded in
- * removing the inode, but not the orphan record, which means
- * we need to manually delete the orphan since iput will just
- * do a destroy_inode
- */
- if (is_bad_inode(inode)) {
- trans = btrfs_start_transaction(root, 0);
+ if (ret == -ESTALE) {
+ trans = btrfs_start_transaction(root, 1);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
goto out;
}
- btrfs_orphan_del(trans, inode);
+ ret = btrfs_del_orphan_item(trans, root,
+ found_key.objectid);
+ BUG_ON(ret);
btrfs_end_transaction(trans, root);
- iput(inode);
continue;
}
+ /*
+ * add this inode to the orphan list so btrfs_orphan_del does
+ * the proper thing when we hit it
+ */
+ spin_lock(&root->orphan_lock);
+ list_add(&BTRFS_I(inode)->i_orphan, &root->orphan_list);
+ spin_unlock(&root->orphan_lock);
+
/* if we have links, this was a truncate, lets do that */
if (inode->i_nlink) {
if (!S_ISREG(inode->i_mode)) {
inode_item = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_inode_item);
inode->i_mode = btrfs_inode_mode(leaf, inode_item);
- inode->i_nlink = btrfs_inode_nlink(leaf, inode_item);
+ set_nlink(inode, btrfs_inode_nlink(leaf, inode_item));
inode->i_uid = btrfs_inode_uid(leaf, inode_item);
inode->i_gid = btrfs_inode_gid(leaf, inode_item);
btrfs_i_size_write(inode, btrfs_inode_size(leaf, inode_item));
u64 ino = btrfs_ino(inode);
u64 dir_ino = btrfs_ino(dir);
- trans = btrfs_start_transaction(root, 10);
+ /*
+ * 1 for the possible orphan item
+ * 1 for the dir item
+ * 1 for the dir index
+ * 1 for the inode ref
+ * 1 for the inode ref in the tree log
+ * 2 for the dir entries in the log
+ * 1 for the inode
+ */
+ trans = btrfs_start_transaction(root, 8);
if (!IS_ERR(trans) || PTR_ERR(trans) != -ENOSPC)
return trans;
return ERR_PTR(-ENOMEM);
}
- trans = btrfs_start_transaction(root, 0);
+ /* 1 for the orphan item */
+ trans = btrfs_start_transaction(root, 1);
if (IS_ERR(trans)) {
btrfs_free_path(path);
root->fs_info->enospc_unlink = 0;
err = 0;
out:
btrfs_free_path(path);
+ /* Migrate the orphan reservation over */
+ if (!err)
+ err = btrfs_block_rsv_migrate(trans->block_rsv,
+ &root->fs_info->global_block_rsv,
+ trans->bytes_reserved);
+
if (err) {
btrfs_end_transaction(trans, root);
root->fs_info->enospc_unlink = 0;
struct btrfs_root *root)
{
if (trans->block_rsv == &root->fs_info->global_block_rsv) {
+ btrfs_block_rsv_release(root, trans->block_rsv,
+ trans->bytes_reserved);
+ trans->block_rsv = &root->fs_info->trans_block_rsv;
BUG_ON(!root->fs_info->enospc_unlink);
root->fs_info->enospc_unlink = 0;
}
pgoff_t index = from >> PAGE_CACHE_SHIFT;
unsigned offset = from & (PAGE_CACHE_SIZE-1);
struct page *page;
+ gfp_t mask = btrfs_alloc_write_mask(mapping);
int ret = 0;
u64 page_start;
u64 page_end;
ret = -ENOMEM;
again:
- page = find_or_create_page(mapping, index, GFP_NOFS);
+ page = find_or_create_page(mapping, index, mask);
if (!page) {
btrfs_delalloc_release_space(inode, PAGE_CACHE_SIZE);
goto out;
{
struct btrfs_trans_handle *trans;
struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_block_rsv *rsv, *global_rsv;
+ u64 min_size = btrfs_calc_trunc_metadata_size(root, 1);
unsigned long nr;
int ret;
goto no_delete;
}
+ rsv = btrfs_alloc_block_rsv(root);
+ if (!rsv) {
+ btrfs_orphan_del(NULL, inode);
+ goto no_delete;
+ }
+ rsv->size = min_size;
+ global_rsv = &root->fs_info->global_block_rsv;
+
btrfs_i_size_write(inode, 0);
+ /*
+ * This is a bit simpler than btrfs_truncate since
+ *
+ * 1) We've already reserved our space for our orphan item in the
+ * unlink.
+ * 2) We're going to delete the inode item, so we don't need to update
+ * it at all.
+ *
+ * So we just need to reserve some slack space in case we add bytes when
+ * doing the truncate.
+ */
while (1) {
- trans = btrfs_join_transaction(root);
- BUG_ON(IS_ERR(trans));
- trans->block_rsv = root->orphan_block_rsv;
+ ret = btrfs_block_rsv_refill(root, rsv, min_size);
+
+ /*
+ * Try and steal from the global reserve since we will
+ * likely not use this space anyway, we want to try as
+ * hard as possible to get this to work.
+ */
+ if (ret)
+ ret = btrfs_block_rsv_migrate(global_rsv, rsv, min_size);
- ret = btrfs_block_rsv_check(trans, root,
- root->orphan_block_rsv, 0, 5);
if (ret) {
- BUG_ON(ret != -EAGAIN);
- ret = btrfs_commit_transaction(trans, root);
- BUG_ON(ret);
- continue;
+ printk(KERN_WARNING "Could not get space for a "
+ "delete, will truncate on mount %d\n", ret);
+ btrfs_orphan_del(NULL, inode);
+ btrfs_free_block_rsv(root, rsv);
+ goto no_delete;
}
+ trans = btrfs_start_transaction(root, 0);
+ if (IS_ERR(trans)) {
+ btrfs_orphan_del(NULL, inode);
+ btrfs_free_block_rsv(root, rsv);
+ goto no_delete;
+ }
+
+ trans->block_rsv = rsv;
+
ret = btrfs_truncate_inode_items(trans, root, inode, 0, 0);
if (ret != -EAGAIN)
break;
btrfs_end_transaction(trans, root);
trans = NULL;
btrfs_btree_balance_dirty(root, nr);
-
}
+ btrfs_free_block_rsv(root, rsv);
+
if (ret == 0) {
+ trans->block_rsv = root->orphan_block_rsv;
ret = btrfs_orphan_del(trans, inode);
BUG_ON(ret);
}
+ trans->block_rsv = &root->fs_info->trans_block_rsv;
if (!(root == root->fs_info->tree_root ||
root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID))
btrfs_return_ino(root, btrfs_ino(inode));
if (test_bit(BTRFS_ORDERED_NOCOW, &ordered->flags)) {
ret = btrfs_ordered_update_i_size(inode, 0, ordered);
if (!ret)
- ret = btrfs_update_inode(trans, root, inode);
- err = ret;
+ err = btrfs_update_inode(trans, root, inode);
goto out;
}
{
struct extent_io_tree *tree;
tree = &BTRFS_I(page->mapping->host)->io_tree;
- return extent_read_full_page(tree, page, btrfs_get_extent);
+ return extent_read_full_page(tree, page, btrfs_get_extent, 0);
}
static int btrfs_writepage(struct page *page, struct writeback_control *wbc)
struct btrfs_trans_handle *trans;
unsigned long nr;
u64 mask = root->sectorsize - 1;
+ u64 min_size = btrfs_calc_trunc_metadata_size(root, 1);
ret = btrfs_truncate_page(inode->i_mapping, inode->i_size);
if (ret)
rsv = btrfs_alloc_block_rsv(root);
if (!rsv)
return -ENOMEM;
- btrfs_add_durable_block_rsv(root->fs_info, rsv);
+ rsv->size = min_size;
+ /*
+ * 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);
if (IS_ERR(trans)) {
err = PTR_ERR(trans);
goto out;
}
- /*
- * Reserve space for the truncate process. Truncate should be adding
- * space, but if there are snapshots it may end up using space.
- */
- ret = btrfs_truncate_reserve_metadata(trans, root, rsv);
+ /* Migrate the slack space for the truncate to our reserve */
+ ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv, rsv,
+ min_size);
BUG_ON(ret);
ret = btrfs_orphan_add(trans, inode);
goto out;
}
- nr = trans->blocks_used;
- btrfs_end_transaction(trans, root);
- btrfs_btree_balance_dirty(root, nr);
-
- /*
- * Ok so we've already migrated our bytes over for the truncate, so here
- * just reserve the one slot we need for updating the inode.
- */
- trans = btrfs_start_transaction(root, 1);
- if (IS_ERR(trans)) {
- err = PTR_ERR(trans);
- goto out;
- }
- trans->block_rsv = rsv;
-
/*
* setattr is responsible for setting the ordered_data_close flag,
* but that is only tested during the last file release. That
btrfs_add_ordered_operation(trans, root, inode);
while (1) {
+ ret = btrfs_block_rsv_refill(root, rsv, min_size);
+ if (ret) {
+ /*
+ * This can only happen with the original transaction we
+ * started above, every other time we shouldn't have a
+ * transaction started yet.
+ */
+ if (ret == -EAGAIN)
+ goto end_trans;
+ err = ret;
+ break;
+ }
+
if (!trans) {
- trans = btrfs_start_transaction(root, 3);
+ /* Just need the 1 for updating the inode */
+ trans = btrfs_start_transaction(root, 1);
if (IS_ERR(trans)) {
err = PTR_ERR(trans);
goto out;
}
-
- ret = btrfs_truncate_reserve_metadata(trans, root,
- rsv);
- BUG_ON(ret);
-
- trans->block_rsv = rsv;
}
+ trans->block_rsv = rsv;
+
ret = btrfs_truncate_inode_items(trans, root, inode,
inode->i_size,
BTRFS_EXTENT_DATA_KEY);
err = ret;
break;
}
-
+ end_trans:
nr = trans->blocks_used;
btrfs_end_transaction(trans, root);
trans = NULL;
inode->i_op = &btrfs_dir_inode_operations;
inode->i_fop = &btrfs_dir_file_operations;
- inode->i_nlink = 1;
+ set_nlink(inode, 1);
btrfs_i_size_write(inode, 0);
err = btrfs_update_inode(trans, new_root, inode);
ei->last_sub_trans = 0;
ei->logged_trans = 0;
ei->delalloc_bytes = 0;
- ei->reserved_bytes = 0;
ei->disk_i_size = 0;
ei->flags = 0;
+ ei->csum_bytes = 0;
ei->index_cnt = (u64)-1;
ei->last_unlink_trans = 0;
WARN_ON(inode->i_data.nrpages);
WARN_ON(BTRFS_I(inode)->outstanding_extents);
WARN_ON(BTRFS_I(inode)->reserved_extents);
+ WARN_ON(BTRFS_I(inode)->delalloc_bytes);
+ WARN_ON(BTRFS_I(inode)->csum_bytes);
/*
* This can happen where we create an inode, but somebody else also
.readpage_end_io_hook = btrfs_readpage_end_io_hook,
.writepage_end_io_hook = btrfs_writepage_end_io_hook,
.writepage_start_hook = btrfs_writepage_start_hook,
- .readpage_io_failed_hook = btrfs_io_failed_hook,
.set_bit_hook = btrfs_set_bit_hook,
.clear_bit_hook = btrfs_clear_bit_hook,
.merge_extent_hook = btrfs_merge_extent_hook,
struct walk_control *wc, u64 gen)
{
if (wc->pin)
- btrfs_pin_extent(log->fs_info->extent_root,
- eb->start, eb->len, 0);
+ btrfs_pin_extent_for_log_replay(wc->trans,
+ log->fs_info->extent_root,
+ eb->start, eb->len);
if (btrfs_buffer_uptodate(eb, gen)) {
if (wc->write)
}
btrfs_release_path(path);
if (nlink != inode->i_nlink) {
- inode->i_nlink = nlink;
+ set_nlink(inode, nlink);
btrfs_update_inode(trans, root, inode);
}
BTRFS_I(inode)->index_cnt = (u64)-1;
WARN_ON(root_owner !=
BTRFS_TREE_LOG_OBJECTID);
- ret = btrfs_free_reserved_extent(root,
+ ret = btrfs_free_and_pin_reserved_extent(root,
bytenr, blocksize);
BUG_ON(ret);
}
btrfs_tree_unlock(next);
WARN_ON(root_owner != BTRFS_TREE_LOG_OBJECTID);
- ret = btrfs_free_reserved_extent(root,
+ ret = btrfs_free_and_pin_reserved_extent(root,
path->nodes[*level]->start,
path->nodes[*level]->len);
BUG_ON(ret);
WARN_ON(log->root_key.objectid !=
BTRFS_TREE_LOG_OBJECTID);
- ret = btrfs_free_reserved_extent(log, next->start,
+ ret = btrfs_free_and_pin_reserved_extent(log, next->start,
next->len);
BUG_ON(ret);
}
/* wait for previous tree log sync to complete */
if (atomic_read(&root->log_commit[(index1 + 1) % 2]))
wait_log_commit(trans, root, root->log_transid - 1);
-
while (1) {
unsigned long batch = root->log_batch;
- if (root->log_multiple_pids) {
+ /* when we're on an ssd, just kick the log commit out */
+ if (!btrfs_test_opt(root, SSD) && root->log_multiple_pids) {
mutex_unlock(&root->log_mutex);
schedule_timeout_uninterruptible(1);
mutex_lock(&root->log_mutex);
BUG_ON(ret);
btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
- btrfs_set_super_log_root(&root->fs_info->super_for_commit,
+ btrfs_set_super_log_root(root->fs_info->super_for_commit,
log_root_tree->node->start);
- btrfs_set_super_log_root_level(&root->fs_info->super_for_commit,
+ btrfs_set_super_log_root_level(root->fs_info->super_for_commit,
btrfs_header_level(log_root_tree->node));
log_root_tree->log_batch = 0;
again:
ret = btrfs_insert_xattr_item(trans, root, path, btrfs_ino(inode),
name, name_len, value, size);
+ /*
+ * If we're setting an xattr to a new value but the new value is say
+ * exactly BTRFS_MAX_XATTR_SIZE, we could end up with EOVERFLOW getting
+ * back from split_leaf. This is because it thinks we'll be extending
+ * the existing item size, but we're asking for enough space to add the
+ * item itself. So if we get EOVERFLOW just set ret to EEXIST and let
+ * the rest of the function figure it out.
+ */
+ if (ret == -EOVERFLOW)
+ ret = -EEXIST;
+
if (ret == -EEXIST) {
if (flags & XATTR_CREATE)
goto out;
XATTR_REPLACE);
}
-int btrfs_xattr_security_init(struct btrfs_trans_handle *trans,
- struct inode *inode, struct inode *dir,
- const struct qstr *qstr)
+int btrfs_initxattrs(struct inode *inode, const struct xattr *xattr_array,
+ void *fs_info)
{
- int err;
- size_t len;
- void *value;
- char *suffix;
+ const struct xattr *xattr;
+ struct btrfs_trans_handle *trans = fs_info;
char *name;
+ int err = 0;
- err = security_inode_init_security(inode, dir, qstr, &suffix, &value,
- &len);
- if (err) {
- if (err == -EOPNOTSUPP)
- return 0;
- return err;
- }
-
- name = kmalloc(XATTR_SECURITY_PREFIX_LEN + strlen(suffix) + 1,
- GFP_NOFS);
- if (!name) {
- err = -ENOMEM;
- } else {
+ for (xattr = xattr_array; xattr->name != NULL; xattr++) {
+ name = kmalloc(XATTR_SECURITY_PREFIX_LEN +
+ strlen(xattr->name) + 1, GFP_NOFS);
+ if (!name) {
+ err = -ENOMEM;
+ break;
+ }
strcpy(name, XATTR_SECURITY_PREFIX);
- strcpy(name + XATTR_SECURITY_PREFIX_LEN, suffix);
- err = __btrfs_setxattr(trans, inode, name, value, len, 0);
+ strcpy(name + XATTR_SECURITY_PREFIX_LEN, xattr->name);
+ err = __btrfs_setxattr(trans, inode, name,
+ xattr->value, xattr->value_len, 0);
kfree(name);
+ if (err < 0)
+ break;
}
-
- kfree(suffix);
- kfree(value);
return err;
}
+
+int btrfs_xattr_security_init(struct btrfs_trans_handle *trans,
+ struct inode *inode, struct inode *dir,
+ const struct qstr *qstr)
+{
+ return security_inode_init_security(inode, dir, qstr,
+ &btrfs_initxattrs, trans);
+}