return ret;
}
-#define in_range(b, first, len) ((b) >= (first) && (b) < (first) + (len))
static int btrfs_issue_discard(struct block_device *bdev, u64 start, u64 len,
u64 *discarded_bytes)
{
return btrfs_check_space_for_delayed_refs(trans->fs_info);
}
-struct async_delayed_refs {
- struct btrfs_root *root;
- u64 transid;
- int count;
- int error;
- int sync;
- struct completion wait;
- struct btrfs_work work;
-};
-
-static inline struct async_delayed_refs *
-to_async_delayed_refs(struct btrfs_work *work)
-{
- return container_of(work, struct async_delayed_refs, work);
-}
-
-static void delayed_ref_async_start(struct btrfs_work *work)
-{
- struct async_delayed_refs *async = to_async_delayed_refs(work);
- struct btrfs_trans_handle *trans;
- struct btrfs_fs_info *fs_info = async->root->fs_info;
- int ret;
-
- /* if the commit is already started, we don't need to wait here */
- if (btrfs_transaction_blocked(fs_info))
- goto done;
-
- trans = btrfs_join_transaction(async->root);
- if (IS_ERR(trans)) {
- async->error = PTR_ERR(trans);
- goto done;
- }
-
- /*
- * trans->sync means that when we call end_transaction, we won't
- * wait on delayed refs
- */
- trans->sync = true;
-
- /* Don't bother flushing if we got into a different transaction */
- if (trans->transid > async->transid)
- goto end;
-
- ret = btrfs_run_delayed_refs(trans, async->count);
- if (ret)
- async->error = ret;
-end:
- ret = btrfs_end_transaction(trans);
- if (ret && !async->error)
- async->error = ret;
-done:
- if (async->sync)
- complete(&async->wait);
- else
- kfree(async);
-}
-
-int btrfs_async_run_delayed_refs(struct btrfs_fs_info *fs_info,
- unsigned long count, u64 transid, int wait)
-{
- struct async_delayed_refs *async;
- int ret;
-
- async = kmalloc(sizeof(*async), GFP_NOFS);
- if (!async)
- return -ENOMEM;
-
- async->root = fs_info->tree_root;
- async->count = count;
- async->error = 0;
- async->transid = transid;
- if (wait)
- async->sync = 1;
- else
- async->sync = 0;
- init_completion(&async->wait);
-
- btrfs_init_work(&async->work, btrfs_extent_refs_helper,
- delayed_ref_async_start, NULL, NULL);
-
- btrfs_queue_work(fs_info->extent_workers, &async->work);
-
- if (wait) {
- wait_for_completion(&async->wait);
- ret = async->error;
- kfree(async);
- return ret;
- }
- return 0;
-}
-
/*
* this starts processing the delayed reference count updates and
* extent insertions we have queued up so far. count can be
}
static int write_one_cache_group(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info,
struct btrfs_path *path,
struct btrfs_block_group_cache *cache)
{
+ struct btrfs_fs_info *fs_info = trans->fs_info;
int ret;
struct btrfs_root *extent_root = fs_info->extent_root;
unsigned long bi;
}
-static struct btrfs_block_group_cache *
-next_block_group(struct btrfs_fs_info *fs_info,
- struct btrfs_block_group_cache *cache)
+static struct btrfs_block_group_cache *next_block_group(
+ struct btrfs_block_group_cache *cache)
{
+ struct btrfs_fs_info *fs_info = cache->fs_info;
struct rb_node *node;
spin_lock(&fs_info->block_group_cache_lock);
if (block_group->ro)
goto out_free;
- ret = create_free_space_inode(fs_info, trans, block_group,
- path);
+ ret = create_free_space_inode(trans, block_group, path);
if (ret)
goto out_free;
goto again;
return ret;
}
-int btrfs_setup_space_cache(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info)
+int btrfs_setup_space_cache(struct btrfs_trans_handle *trans)
{
+ struct btrfs_fs_info *fs_info = trans->fs_info;
struct btrfs_block_group_cache *cache, *tmp;
struct btrfs_transaction *cur_trans = trans->transaction;
struct btrfs_path *path;
if (cache->disk_cache_state == BTRFS_DC_SETUP) {
cache->io_ctl.inode = NULL;
- ret = btrfs_write_out_cache(fs_info, trans,
- cache, path);
+ ret = btrfs_write_out_cache(trans, cache, path);
if (ret == 0 && cache->io_ctl.inode) {
num_started++;
should_put = 0;
}
}
if (!ret) {
- ret = write_one_cache_group(trans, fs_info,
- path, cache);
+ ret = write_one_cache_group(trans, path, cache);
/*
* Our block group might still be attached to the list
* of new block groups in the transaction handle of some
return ret;
}
-int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *fs_info)
+int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans)
{
+ struct btrfs_fs_info *fs_info = trans->fs_info;
struct btrfs_block_group_cache *cache;
struct btrfs_transaction *cur_trans = trans->transaction;
int ret = 0;
if (!ret && cache->disk_cache_state == BTRFS_DC_SETUP) {
cache->io_ctl.inode = NULL;
- ret = btrfs_write_out_cache(fs_info, trans,
- cache, path);
+ ret = btrfs_write_out_cache(trans, cache, path);
if (ret == 0 && cache->io_ctl.inode) {
num_started++;
should_put = 0;
}
}
if (!ret) {
- ret = write_one_cache_group(trans, fs_info,
- path, cache);
+ ret = write_one_cache_group(trans, path, cache);
/*
* One of the free space endio workers might have
* created a new block group while updating a free space
if (ret == -ENOENT) {
wait_event(cur_trans->writer_wait,
atomic_read(&cur_trans->num_writers) == 1);
- ret = write_one_cache_group(trans, fs_info,
- path, cache);
+ ret = write_one_cache_group(trans, path, cache);
}
if (ret)
btrfs_abort_transaction(trans, ret);
}
static int update_block_group(struct btrfs_trans_handle *trans,
- struct btrfs_fs_info *info, u64 bytenr,
- u64 num_bytes, int alloc)
+ u64 bytenr, u64 num_bytes, int alloc)
{
+ struct btrfs_fs_info *info = trans->fs_info;
struct btrfs_block_group_cache *cache = NULL;
u64 total = num_bytes;
u64 old_val;
if (list_empty(&cache->dirty_list)) {
list_add_tail(&cache->dirty_list,
&trans->transaction->dirty_bgs);
- trans->transaction->num_dirty_bgs++;
trans->delayed_ref_updates++;
btrfs_get_block_group(cache);
}
return ret;
}
-int btrfs_exclude_logged_extents(struct btrfs_fs_info *fs_info,
- struct extent_buffer *eb)
+int btrfs_exclude_logged_extents(struct extent_buffer *eb)
{
+ struct btrfs_fs_info *fs_info = eb->fs_info;
struct btrfs_file_extent_item *item;
struct btrfs_key key;
int found_type;
goto out;
}
- ret = update_block_group(trans, info, bytenr, num_bytes, 0);
+ ret = update_block_group(trans, bytenr, num_bytes, 0);
if (ret) {
btrfs_abort_transaction(trans, ret);
goto out;
if (ret)
return ret;
- ret = update_block_group(trans, fs_info, ins->objectid, ins->offset, 1);
+ ret = update_block_group(trans, ins->objectid, ins->offset, 1);
if (ret) { /* -ENOENT, logic error */
btrfs_err(fs_info, "update block group failed for %llu %llu",
ins->objectid, ins->offset);
if (ret)
return ret;
- ret = update_block_group(trans, fs_info, extent_key.objectid,
+ ret = update_block_group(trans, extent_key.objectid,
fs_info->nodesize, 1);
if (ret) { /* -ENOENT, logic error */
btrfs_err(fs_info, "update block group failed for %llu %llu",
btrfs_set_buffer_lockdep_class(root->root_key.objectid, buf, level);
btrfs_tree_lock(buf);
- clean_tree_block(fs_info, buf);
+ btrfs_clean_tree_block(buf);
clear_bit(EXTENT_BUFFER_STALE, &buf->bflags);
btrfs_set_lock_blocking_write(buf);
ret);
}
}
- /* make block locked assertion in clean_tree_block happy */
+ /* make block locked assertion in btrfs_clean_tree_block happy */
if (!path->locks[level] &&
btrfs_header_generation(eb) == trans->transid) {
btrfs_tree_lock(eb);
btrfs_set_lock_blocking_write(eb);
path->locks[level] = BTRFS_WRITE_LOCK_BLOCKING;
}
- clean_tree_block(fs_info, eb);
+ btrfs_clean_tree_block(eb);
}
if (eb == root->node) {
*/
int btrfs_can_relocate(struct btrfs_fs_info *fs_info, u64 bytenr)
{
- struct btrfs_root *root = fs_info->extent_root;
struct btrfs_block_group_cache *block_group;
struct btrfs_space_info *space_info;
struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
struct btrfs_device *device;
- struct btrfs_trans_handle *trans;
u64 min_free;
u64 dev_min = 1;
u64 dev_nr = 0;
min_free = div64_u64(min_free, dev_min);
}
- /* We need to do this so that we can look at pending chunks */
- trans = btrfs_join_transaction(root);
- if (IS_ERR(trans)) {
- ret = PTR_ERR(trans);
- goto out;
- }
-
mutex_lock(&fs_info->chunk_mutex);
list_for_each_entry(device, &fs_devices->alloc_list, dev_alloc_list) {
u64 dev_offset;
*/
if (device->total_bytes > device->bytes_used + min_free &&
!test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state)) {
- ret = find_free_dev_extent(trans, device, min_free,
+ ret = find_free_dev_extent(device, min_free,
&dev_offset, NULL);
if (!ret)
dev_nr++;
"no space to allocate a new chunk for block group %llu",
block_group->key.objectid);
mutex_unlock(&fs_info->chunk_mutex);
- btrfs_end_transaction(trans);
out:
btrfs_put_block_group(block_group);
return ret;
if (block_group->iref)
break;
spin_unlock(&block_group->lock);
- block_group = next_block_group(info, block_group);
+ block_group = next_block_group(block_group);
}
if (!block_group) {
if (last == 0)
struct btrfs_block_group_cache *cache;
int ret;
- btrfs_set_log_full_commit(fs_info, trans);
+ btrfs_set_log_full_commit(trans);
cache = btrfs_create_block_group_cache(fs_info, chunk_offset, size);
if (!cache)
memcpy(&key, &block_group->key, sizeof(key));
mutex_lock(&fs_info->chunk_mutex);
- if (!list_empty(&em->list)) {
- /* We're in the transaction->pending_chunks list. */
- free_extent_map(em);
- }
spin_lock(&block_group->lock);
block_group->removed = 1;
/*
* the transaction commit has completed.
*/
remove_em = (atomic_read(&block_group->trimming) == 0);
- /*
- * Make sure a trimmer task always sees the em in the pinned_chunks list
- * if it sees block_group->removed == 1 (needs to lock block_group->lock
- * before checking block_group->removed).
- */
- if (!remove_em) {
- /*
- * Our em might be in trans->transaction->pending_chunks which
- * is protected by fs_info->chunk_mutex ([lock|unlock]_chunks),
- * and so is the fs_info->pinned_chunks list.
- *
- * So at this point we must be holding the chunk_mutex to avoid
- * any races with chunk allocation (more specifically at
- * volumes.c:contains_pending_extent()), to ensure it always
- * sees the em, either in the pending_chunks list or in the
- * pinned_chunks list.
- */
- list_move_tail(&em->list, &fs_info->pinned_chunks);
- }
spin_unlock(&block_group->lock);
if (remove_em) {
em_tree = &fs_info->mapping_tree.map_tree;
write_lock(&em_tree->lock);
- /*
- * The em might be in the pending_chunks list, so make sure the
- * chunk mutex is locked, since remove_extent_mapping() will
- * delete us from that list.
- */
remove_extent_mapping(em_tree, em);
write_unlock(&em_tree->lock);
/* once for the tree */
* held back allocations.
*/
static int btrfs_trim_free_extents(struct btrfs_device *device,
- u64 minlen, u64 *trimmed)
+ struct fstrim_range *range, u64 *trimmed)
{
- u64 start = 0, len = 0;
+ u64 start, len = 0, end = 0;
int ret;
+ start = max_t(u64, range->start, SZ_1M);
*trimmed = 0;
/* Discard not supported = nothing to do. */
while (1) {
struct btrfs_fs_info *fs_info = device->fs_info;
- struct btrfs_transaction *trans;
u64 bytes;
ret = mutex_lock_interruptible(&fs_info->chunk_mutex);
if (ret)
break;
- ret = down_read_killable(&fs_info->commit_root_sem);
- if (ret) {
+ find_first_clear_extent_bit(&device->alloc_state, start,
+ &start, &end,
+ CHUNK_TRIMMED | CHUNK_ALLOCATED);
+ /*
+ * If find_first_clear_extent_bit find a range that spans the
+ * end of the device it will set end to -1, in this case it's up
+ * to the caller to trim the value to the size of the device.
+ */
+ end = min(end, device->total_bytes - 1);
+ len = end - start + 1;
+
+ /* We didn't find any extents */
+ if (!len) {
mutex_unlock(&fs_info->chunk_mutex);
+ ret = 0;
break;
}
- spin_lock(&fs_info->trans_lock);
- trans = fs_info->running_transaction;
- if (trans)
- refcount_inc(&trans->use_count);
- spin_unlock(&fs_info->trans_lock);
-
- if (!trans)
- up_read(&fs_info->commit_root_sem);
-
- ret = find_free_dev_extent_start(trans, device, minlen, start,
- &start, &len);
- if (trans) {
- up_read(&fs_info->commit_root_sem);
- btrfs_put_transaction(trans);
+ /* Keep going until we satisfy minlen or reach end of space */
+ if (len < range->minlen) {
+ mutex_unlock(&fs_info->chunk_mutex);
+ start += len;
+ continue;
}
- if (ret) {
+ /* If we are out of the passed range break */
+ if (start > range->start + range->len - 1) {
mutex_unlock(&fs_info->chunk_mutex);
- if (ret == -ENOSPC)
- ret = 0;
break;
}
- ret = btrfs_issue_discard(device->bdev, start, len, &bytes);
+ start = max(range->start, start);
+ len = min(range->len, len);
+
+ ret = btrfs_issue_discard(device->bdev, start, len,
+ &bytes);
+ if (!ret)
+ set_extent_bits(&device->alloc_state, start,
+ start + bytes - 1,
+ CHUNK_TRIMMED);
mutex_unlock(&fs_info->chunk_mutex);
if (ret)
start += len;
*trimmed += bytes;
+ /* We've trimmed enough */
+ if (*trimmed >= range->len)
+ break;
+
if (fatal_signal_pending(current)) {
ret = -ERESTARTSYS;
break;
int ret = 0;
cache = btrfs_lookup_first_block_group(fs_info, range->start);
- for (; cache; cache = next_block_group(fs_info, cache)) {
+ for (; cache; cache = next_block_group(cache)) {
if (cache->key.objectid >= (range->start + range->len)) {
btrfs_put_block_group(cache);
break;
mutex_lock(&fs_info->fs_devices->device_list_mutex);
devices = &fs_info->fs_devices->devices;
list_for_each_entry(device, devices, dev_list) {
- ret = btrfs_trim_free_extents(device, range->minlen,
- &group_trimmed);
+ ret = btrfs_trim_free_extents(device, range, &group_trimmed);
if (ret) {
dev_failed++;
dev_ret = ret;
git.samba.org / sfrench / cifs-2.6.git / blobdiff