Commit 8929ecfa authored by Yan, Zheng's avatar Yan, Zheng Committed by Chris Mason
Browse files

Btrfs: Introduce global metadata reservation



Reserve metadata space for extent tree, checksum tree and root tree
Signed-off-by: default avatarYan Zheng <zheng.yan@oracle.com>
Signed-off-by: default avatarChris Mason <chris.mason@oracle.com>
parent 0ca1f7ce
......@@ -683,21 +683,15 @@ struct btrfs_space_info {
u64 bytes_reserved; /* total bytes the allocator has reserved for
current allocations */
u64 bytes_readonly; /* total bytes that are read only */
u64 bytes_super; /* total bytes reserved for the super blocks */
u64 bytes_root; /* the number of bytes needed to commit a
transaction */
u64 bytes_may_use; /* number of bytes that may be used for
delalloc/allocations */
u64 bytes_delalloc; /* number of bytes currently reserved for
delayed allocation */
u64 disk_used; /* total bytes used on disk */
int full; /* indicates that we cannot allocate any more
chunks for this space */
int force_alloc; /* set if we need to force a chunk alloc for
this space */
int force_delalloc; /* make people start doing filemap_flush until
we're under a threshold */
struct list_head list;
......
......@@ -1463,10 +1463,6 @@ static int cleaner_kthread(void *arg)
struct btrfs_root *root = arg;
do {
smp_mb();
if (root->fs_info->closing)
break;
vfs_check_frozen(root->fs_info->sb, SB_FREEZE_WRITE);
if (!(root->fs_info->sb->s_flags & MS_RDONLY) &&
......@@ -1479,11 +1475,9 @@ static int cleaner_kthread(void *arg)
if (freezing(current)) {
refrigerator();
} else {
smp_mb();
if (root->fs_info->closing)
break;
set_current_state(TASK_INTERRUPTIBLE);
schedule();
if (!kthread_should_stop())
schedule();
__set_current_state(TASK_RUNNING);
}
} while (!kthread_should_stop());
......@@ -1495,36 +1489,40 @@ static int transaction_kthread(void *arg)
struct btrfs_root *root = arg;
struct btrfs_trans_handle *trans;
struct btrfs_transaction *cur;
u64 transid;
unsigned long now;
unsigned long delay;
int ret;
do {
smp_mb();
if (root->fs_info->closing)
break;
delay = HZ * 30;
vfs_check_frozen(root->fs_info->sb, SB_FREEZE_WRITE);
mutex_lock(&root->fs_info->transaction_kthread_mutex);
mutex_lock(&root->fs_info->trans_mutex);
spin_lock(&root->fs_info->new_trans_lock);
cur = root->fs_info->running_transaction;
if (!cur) {
mutex_unlock(&root->fs_info->trans_mutex);
spin_unlock(&root->fs_info->new_trans_lock);
goto sleep;
}
now = get_seconds();
if (now < cur->start_time || now - cur->start_time < 30) {
mutex_unlock(&root->fs_info->trans_mutex);
if (!cur->blocked &&
(now < cur->start_time || now - cur->start_time < 30)) {
spin_unlock(&root->fs_info->new_trans_lock);
delay = HZ * 5;
goto sleep;
}
mutex_unlock(&root->fs_info->trans_mutex);
trans = btrfs_join_transaction(root, 1);
ret = btrfs_commit_transaction(trans, root);
transid = cur->transid;
spin_unlock(&root->fs_info->new_trans_lock);
trans = btrfs_join_transaction(root, 1);
if (transid == trans->transid) {
ret = btrfs_commit_transaction(trans, root);
BUG_ON(ret);
} else {
btrfs_end_transaction(trans, root);
}
sleep:
wake_up_process(root->fs_info->cleaner_kthread);
mutex_unlock(&root->fs_info->transaction_kthread_mutex);
......@@ -1532,10 +1530,10 @@ sleep:
if (freezing(current)) {
refrigerator();
} else {
if (root->fs_info->closing)
break;
set_current_state(TASK_INTERRUPTIBLE);
schedule_timeout(delay);
if (!kthread_should_stop() &&
!btrfs_transaction_blocked(root->fs_info))
schedule_timeout(delay);
__set_current_state(TASK_RUNNING);
}
} while (!kthread_should_stop());
......@@ -1917,17 +1915,18 @@ struct btrfs_root *open_ctree(struct super_block *sb,
csum_root->track_dirty = 1;
fs_info->generation = generation;
fs_info->last_trans_committed = generation;
fs_info->data_alloc_profile = (u64)-1;
fs_info->metadata_alloc_profile = (u64)-1;
fs_info->system_alloc_profile = fs_info->metadata_alloc_profile;
ret = btrfs_read_block_groups(extent_root);
if (ret) {
printk(KERN_ERR "Failed to read block groups: %d\n", ret);
goto fail_block_groups;
}
fs_info->generation = generation;
fs_info->last_trans_committed = generation;
fs_info->data_alloc_profile = (u64)-1;
fs_info->metadata_alloc_profile = (u64)-1;
fs_info->system_alloc_profile = fs_info->metadata_alloc_profile;
fs_info->cleaner_kthread = kthread_run(cleaner_kthread, tree_root,
"btrfs-cleaner");
if (IS_ERR(fs_info->cleaner_kthread))
......@@ -2430,15 +2429,15 @@ int close_ctree(struct btrfs_root *root)
fs_info->closing = 1;
smp_mb();
kthread_stop(root->fs_info->transaction_kthread);
kthread_stop(root->fs_info->cleaner_kthread);
if (!(fs_info->sb->s_flags & MS_RDONLY)) {
ret = btrfs_commit_super(root);
if (ret)
printk(KERN_ERR "btrfs: commit super ret %d\n", ret);
}
kthread_stop(root->fs_info->transaction_kthread);
kthread_stop(root->fs_info->cleaner_kthread);
fs_info->closing = 2;
smp_mb();
......
......@@ -2895,10 +2895,9 @@ int btrfs_check_data_free_space(struct inode *inode, u64 bytes)
again:
/* make sure we have enough space to handle the data first */
spin_lock(&data_sinfo->lock);
used = data_sinfo->bytes_used + data_sinfo->bytes_delalloc +
data_sinfo->bytes_reserved + data_sinfo->bytes_pinned +
data_sinfo->bytes_readonly + data_sinfo->bytes_may_use +
data_sinfo->bytes_super;
used = data_sinfo->bytes_used + data_sinfo->bytes_reserved +
data_sinfo->bytes_pinned + data_sinfo->bytes_readonly +
data_sinfo->bytes_may_use;
if (used + bytes > data_sinfo->total_bytes) {
struct btrfs_trans_handle *trans;
......@@ -2922,7 +2921,7 @@ alloc:
bytes + 2 * 1024 * 1024,
alloc_target, 0);
btrfs_end_transaction(trans, root);
if (ret)
if (ret < 0)
return ret;
if (!data_sinfo) {
......@@ -2945,11 +2944,10 @@ alloc:
goto again;
}
printk(KERN_ERR "no space left, need %llu, %llu delalloc bytes"
", %llu bytes_used, %llu bytes_reserved, "
"%llu bytes_pinned, %llu bytes_readonly, %llu may use "
"%llu total\n", (unsigned long long)bytes,
(unsigned long long)data_sinfo->bytes_delalloc,
printk(KERN_ERR "no space left, need %llu, %llu bytes_used, "
"%llu bytes_reserved, " "%llu bytes_pinned, "
"%llu bytes_readonly, %llu may use %llu total\n",
(unsigned long long)bytes,
(unsigned long long)data_sinfo->bytes_used,
(unsigned long long)data_sinfo->bytes_reserved,
(unsigned long long)data_sinfo->bytes_pinned,
......@@ -3464,6 +3462,91 @@ void btrfs_block_rsv_release(struct btrfs_root *root,
block_rsv_release_bytes(block_rsv, global_rsv, num_bytes);
}
/*
* helper to calculate size of global block reservation.
* the desired value is sum of space used by extent tree,
* checksum tree and root tree
*/
static u64 calc_global_metadata_size(struct btrfs_fs_info *fs_info)
{
struct btrfs_space_info *sinfo;
u64 num_bytes;
u64 meta_used;
u64 data_used;
int csum_size = btrfs_super_csum_size(&fs_info->super_copy);
#if 0
/*
* per tree used space accounting can be inaccuracy, so we
* can't rely on it.
*/
spin_lock(&fs_info->extent_root->accounting_lock);
num_bytes = btrfs_root_used(&fs_info->extent_root->root_item);
spin_unlock(&fs_info->extent_root->accounting_lock);
spin_lock(&fs_info->csum_root->accounting_lock);
num_bytes += btrfs_root_used(&fs_info->csum_root->root_item);
spin_unlock(&fs_info->csum_root->accounting_lock);
spin_lock(&fs_info->tree_root->accounting_lock);
num_bytes += btrfs_root_used(&fs_info->tree_root->root_item);
spin_unlock(&fs_info->tree_root->accounting_lock);
#endif
sinfo = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_DATA);
spin_lock(&sinfo->lock);
data_used = sinfo->bytes_used;
spin_unlock(&sinfo->lock);
sinfo = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA);
spin_lock(&sinfo->lock);
meta_used = sinfo->bytes_used;
spin_unlock(&sinfo->lock);
num_bytes = (data_used >> fs_info->sb->s_blocksize_bits) *
csum_size * 2;
num_bytes += div64_u64(data_used + meta_used, 50);
if (num_bytes * 3 > meta_used)
num_bytes = div64_u64(meta_used, 3);
return ALIGN(num_bytes, fs_info->extent_root->leafsize << 10);
}
static void update_global_block_rsv(struct btrfs_fs_info *fs_info)
{
struct btrfs_block_rsv *block_rsv = &fs_info->global_block_rsv;
struct btrfs_space_info *sinfo = block_rsv->space_info;
u64 num_bytes;
num_bytes = calc_global_metadata_size(fs_info);
spin_lock(&block_rsv->lock);
spin_lock(&sinfo->lock);
block_rsv->size = num_bytes;
num_bytes = sinfo->bytes_used + sinfo->bytes_pinned +
sinfo->bytes_reserved + sinfo->bytes_readonly;
if (sinfo->total_bytes > num_bytes) {
num_bytes = sinfo->total_bytes - num_bytes;
block_rsv->reserved += num_bytes;
sinfo->bytes_reserved += num_bytes;
}
if (block_rsv->reserved >= block_rsv->size) {
num_bytes = block_rsv->reserved - block_rsv->size;
sinfo->bytes_reserved -= num_bytes;
block_rsv->reserved = block_rsv->size;
block_rsv->full = 1;
}
#if 0
printk(KERN_INFO"global block rsv size %llu reserved %llu\n",
block_rsv->size, block_rsv->reserved);
#endif
spin_unlock(&sinfo->lock);
spin_unlock(&block_rsv->lock);
}
static void init_global_block_rsv(struct btrfs_fs_info *fs_info)
{
struct btrfs_space_info *space_info;
......@@ -3473,11 +3556,36 @@ static void init_global_block_rsv(struct btrfs_fs_info *fs_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->extent_root->block_rsv = &fs_info->global_block_rsv;
fs_info->csum_root->block_rsv = &fs_info->global_block_rsv;
fs_info->dev_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);
}
static void release_global_block_rsv(struct btrfs_fs_info *fs_info)
{
block_rsv_release_bytes(&fs_info->global_block_rsv, NULL, (u64)-1);
WARN_ON(fs_info->delalloc_block_rsv.size > 0);
WARN_ON(fs_info->delalloc_block_rsv.reserved > 0);
WARN_ON(fs_info->trans_block_rsv.size > 0);
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);
}
static u64 calc_trans_metadata_size(struct btrfs_root *root, int num_items)
......@@ -3826,6 +3934,8 @@ int btrfs_prepare_extent_commit(struct btrfs_trans_handle *trans,
fs_info->pinned_extents = &fs_info->freed_extents[0];
up_write(&fs_info->extent_commit_sem);
update_global_block_rsv(fs_info);
return 0;
}
......@@ -4818,19 +4928,16 @@ static void dump_space_info(struct btrfs_space_info *info, u64 bytes,
printk(KERN_INFO "space_info has %llu free, is %sfull\n",
(unsigned long long)(info->total_bytes - info->bytes_used -
info->bytes_pinned - info->bytes_reserved -
info->bytes_super),
info->bytes_readonly),
(info->full) ? "" : "not ");
printk(KERN_INFO "space_info total=%llu, pinned=%llu, delalloc=%llu,"
" may_use=%llu, used=%llu, root=%llu, super=%llu, reserved=%llu"
"\n",
printk(KERN_INFO "space_info total=%llu, used=%llu, pinned=%llu, "
"reserved=%llu, may_use=%llu, readonly=%llu\n",
(unsigned long long)info->total_bytes,
(unsigned long long)info->bytes_used,
(unsigned long long)info->bytes_pinned,
(unsigned long long)info->bytes_delalloc,
(unsigned long long)info->bytes_reserved,
(unsigned long long)info->bytes_may_use,
(unsigned long long)info->bytes_used,
(unsigned long long)info->bytes_root,
(unsigned long long)info->bytes_super,
(unsigned long long)info->bytes_reserved);
(unsigned long long)info->bytes_readonly);
spin_unlock(&info->lock);
if (!dump_block_groups)
......@@ -7727,6 +7834,8 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
*/
synchronize_rcu();
release_global_block_rsv(info);
while(!list_empty(&info->space_info)) {
space_info = list_entry(info->space_info.next,
struct btrfs_space_info,
......
......@@ -4060,7 +4060,7 @@ int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc)
struct btrfs_trans_handle *trans;
int ret = 0;
if (root->fs_info->btree_inode == inode)
if (BTRFS_I(inode)->dummy_inode)
return 0;
if (wbc->sync_mode == WB_SYNC_ALL) {
......@@ -4081,10 +4081,19 @@ void btrfs_dirty_inode(struct inode *inode)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_trans_handle *trans;
int ret;
if (BTRFS_I(inode)->dummy_inode)
return;
trans = btrfs_join_transaction(root, 1);
btrfs_set_trans_block_group(trans, inode);
btrfs_update_inode(trans, root, inode);
ret = btrfs_update_inode(trans, root, inode);
if (ret)
printk(KERN_ERR"btrfs: fail to dirty inode %lu error %d\n",
inode->i_ino, ret);
btrfs_end_transaction(trans, root);
}
......
......@@ -1341,8 +1341,10 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
ret = -EPERM;
goto out;
}
btrfs_defrag_root(root, 0);
btrfs_defrag_root(root->fs_info->extent_root, 0);
ret = btrfs_defrag_root(root, 0);
if (ret)
goto out;
ret = btrfs_defrag_root(root->fs_info->extent_root, 0);
break;
case S_IFREG:
if (!(file->f_mode & FMODE_WRITE)) {
......@@ -1372,9 +1374,11 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
/* the rest are all set to zero by kzalloc */
range->len = (u64)-1;
}
btrfs_defrag_file(file, range);
ret = btrfs_defrag_file(file, range);
kfree(range);
break;
default:
ret = -EINVAL;
}
out:
mnt_drop_write(file->f_path.mnt);
......
......@@ -321,10 +321,36 @@ void btrfs_throttle(struct btrfs_root *root)
mutex_unlock(&root->fs_info->trans_mutex);
}
static int should_end_transaction(struct btrfs_trans_handle *trans,
struct btrfs_root *root)
{
int ret;
ret = btrfs_block_rsv_check(trans, root,
&root->fs_info->global_block_rsv, 0, 5);
return ret ? 1 : 0;
}
int btrfs_should_end_transaction(struct btrfs_trans_handle *trans,
struct btrfs_root *root)
{
struct btrfs_transaction *cur_trans = trans->transaction;
int updates;
if (cur_trans->blocked || cur_trans->delayed_refs.flushing)
return 1;
updates = trans->delayed_ref_updates;
trans->delayed_ref_updates = 0;
if (updates)
btrfs_run_delayed_refs(trans, root, updates);
return should_end_transaction(trans, root);
}
static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
struct btrfs_root *root, int throttle)
{
struct btrfs_transaction *cur_trans;
struct btrfs_transaction *cur_trans = trans->transaction;
struct btrfs_fs_info *info = root->fs_info;
int count = 0;
......@@ -350,9 +376,19 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
btrfs_trans_release_metadata(trans, root);
if (!root->fs_info->open_ioctl_trans &&
should_end_transaction(trans, root))
trans->transaction->blocked = 1;
if (cur_trans->blocked && !cur_trans->in_commit) {
if (throttle)
return btrfs_commit_transaction(trans, root);
else
wake_up_process(info->transaction_kthread);
}
mutex_lock(&info->trans_mutex);
cur_trans = info->running_transaction;
WARN_ON(cur_trans != trans->transaction);
WARN_ON(cur_trans != info->running_transaction);
WARN_ON(cur_trans->num_writers < 1);
cur_trans->num_writers--;
......@@ -664,30 +700,30 @@ static noinline int commit_fs_roots(struct btrfs_trans_handle *trans,
int btrfs_defrag_root(struct btrfs_root *root, int cacheonly)
{
struct btrfs_fs_info *info = root->fs_info;
int ret;
struct btrfs_trans_handle *trans;
int ret;
unsigned long nr;
smp_mb();
if (root->defrag_running)
if (xchg(&root->defrag_running, 1))
return 0;
trans = btrfs_start_transaction(root, 1);
while (1) {
root->defrag_running = 1;
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans))
return PTR_ERR(trans);
ret = btrfs_defrag_leaves(trans, root, cacheonly);
nr = trans->blocks_used;
btrfs_end_transaction(trans, root);
btrfs_btree_balance_dirty(info->tree_root, nr);
cond_resched();
trans = btrfs_start_transaction(root, 1);
if (root->fs_info->closing || ret != -EAGAIN)
break;
}
root->defrag_running = 0;
smp_mb();
btrfs_end_transaction(trans, root);
return 0;
return ret;
}
#if 0
......@@ -924,6 +960,16 @@ int btrfs_transaction_in_commit(struct btrfs_fs_info *info)
return ret;
}
int btrfs_transaction_blocked(struct btrfs_fs_info *info)
{
int ret = 0;
spin_lock(&info->new_trans_lock);
if (info->running_transaction)
ret = info->running_transaction->blocked;
spin_unlock(&info->new_trans_lock);
return ret;
}
int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
struct btrfs_root *root)
{
......
......@@ -106,6 +106,8 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
struct btrfs_root *root);
int btrfs_end_transaction_throttle(struct btrfs_trans_handle *trans,
struct btrfs_root *root);
int btrfs_should_end_transaction(struct btrfs_trans_handle *trans,
struct btrfs_root *root);
void btrfs_throttle(struct btrfs_root *root);
int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,
struct btrfs_root *root);
......@@ -115,5 +117,6 @@ int btrfs_write_marked_extents(struct btrfs_root *root,
struct extent_io_tree *dirty_pages, int mark);
int btrfs_wait_marked_extents(struct btrfs_root *root,
struct extent_io_tree *dirty_pages, int mark);
int btrfs_transaction_blocked(struct btrfs_fs_info *info);
int btrfs_transaction_in_commit(struct btrfs_fs_info *info);
#endif
......@@ -117,13 +117,14 @@ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
path->nodes[1], 0,
cache_only, &last_ret,
&root->defrag_progress);
WARN_ON(ret && ret != -EAGAIN);
if (ret) {
WARN_ON(ret == -EAGAIN);
goto out;
}
if (next_key_ret == 0) {
memcpy(&root->defrag_progress, &key, sizeof(key));
ret = -EAGAIN;
}
btrfs_release_path(root, path);
out:
if (path)
btrfs_free_path(path);
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment