github-mirror/linux
1
0
Fork 0
mirror of https://github.com/torvalds/linux.git synced 2026-11-07 06:17:02 +01:00
Code Issues Actions Packages Projects Releases Wiki Activity Graph

for-7.4-part2-tag

Browse Source 
-----BEGIN PGP SIGNATURE-----
 
 iQJPBAABCgA5FiEE8rQSAMVO+zA4DBdWxWXV+ddtWDsFAmreM8IbFIAAAAAABAAO
 bWFudTIsMi41KzEuMTIsMiwyAAoJEMVl1fnXbVg73T8P/jISHdwm/8FA8zWPg4Ss
 VAgS/JzzSsGv6+iWI9W2M7FUskHFGOmBqkFg+5nL9vFp5Ptxuy+cIa6ttPw9sdAo
 xuve3qi1qfpT5qFtiJGZpd3+D+RWxv+MBmo+oDnOFh+PPLAChPRDwQIaBQTrwFKy
 z2xrmcOIpFvGuamBNMk/GrAv5m0AWsD78+bTa5xRjVEX+IY2ic5nTg6YZhSsPaWx
 cnrPslJ37rcsyTt6pC+r8JEZAMJQ0Jr/MZxusErb5ZVta38urA0QCJlaQa5ofA5l
 uRYe/lWDOw4nylo41QB/KzcaakCtrceEkUbqK5CQ/t5XqYDqjLcbt7zLOJXMVU/u
 3HFvFYFkCDKlxrrxMahnI7J9bskiJBpDrzj/8hmsrk0+cLtwlab/kjVSwbLUycE6
 RKQ8q1rUnEm2dQY6DQxbxMG/S75VI1YaWAQuLSNoG6F50NRp/n/mhXWLpesdFV5m
 xSdm1eYqR7sG9Q/nUwF63g/PPPekk8W8LiZcuawcBhWmVHgzUtqzhxlpJ3/TTT/F
 kvNsRhzcCuZ5/hR+6dO2eTEMfA9oo3NyBRuFsvXlE9YpNAyhuTkhp7/Gncf85BNb
 RgoPxeHIqsjEZcpp4oooio9/lcFynpC828NVNLFkBCSCoeoecq4qx7rzAeyKJ3y8
 6DUnJnUaxtx+xYgHw0WSOoYA
 =0YoH
 -----END PGP SIGNATURE-----

Merge tag 'for-7.4-part2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull more btrfs updates from David Sterba:
 "This removes unused code of free space caching, disabled in 7.3
  (commit 545e560a5b0f: "btrfs: disable v1 space cache").

  It's been the original free space cache feature, superseded by free
  space tree (4.5) and mkfs default since 5.15. The grace period was
  sufficiently long, backward compatibility still exists with a
  performance impact, effectively as if mounted with '-o nospace_cache'"

* tag 'for-7.4-part2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: remove TRANS_JOIN_NOLOCK
  btrfs: stop reading free space inodes from the commit root
  btrfs: stop special-casing free space inodes in the delalloc accounting
  btrfs: remove the free space inode special cases from the COW paths
  btrfs: remove the free space inode ordered extent special cases
  btrfs: remove BTRFS_RESERVE_FLUSH_FREE_SPACE_INODE
  btrfs: remove the free space cache trimming ranges
  btrfs: replace btrfs_set_free_space_cache_v1_active() with a cleanup helper
  btrfs: remove the SPACE_CACHE mount option flag
  btrfs: remove btrfs_disk_cache_state
  btrfs: remove the v1 space cache load path
  btrfs: drop the transaction handle from the prealloc helpers
  btrfs: rename cache_write_mutex to dirty_bgs_update_mutex
  btrfs: remove the v1 space cache write path
  btrfs: remove the free space cache endio workqueue
  btrfs: remove the v1 space cache writeout from the transaction commit
  btrfs: stop enabling the v1 space cache from the on-disk state
  btrfs: update the inode item when removing a free space inode
  btrfs: drop the extent maps of the written part of a failed ordered extent
  btrfs: fix use-after-free on quota enable allocation failure
This commit is contained in:
Linus Torvalds 2026-10-25 16:15:22 -10:00
commit 390febb580
18 changed files with 147 additions and 2196 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

497
fs/btrfs/block-group.c
View File

@ -904,22 +904,6 @@ static noinline void caching_thread(struct btrfs_work *work)
down_read(&fs_info->commit_root_sem);
load_block_group_size_class(caching_ctl);
if (btrfs_test_opt(fs_info, SPACE_CACHE)) {
ret = load_free_space_cache(block_group);
if (ret == 1) {
ret = 0;
goto done;
}
/*
* We failed to load the space cache, set ourselves to
* CACHE_STARTED and carry on.
*/
spin_lock(&block_group->lock);
block_group->cached = BTRFS_CACHE_STARTED;
spin_unlock(&block_group->lock);
wake_up(&caching_ctl->wait);
}
/*
* If we are in the transaction that populated the free space tree we
@ -933,7 +917,7 @@ static noinline void caching_thread(struct btrfs_work *work)
ret = btrfs_load_free_space_tree(caching_ctl);
else
ret = load_extent_tree_free(caching_ctl);
done:
spin_lock(&block_group->lock);
block_group->caching_ctl = NULL;
block_group->cached = ret ? BTRFS_CACHE_ERROR : BTRFS_CACHE_FINISHED;
@ -1194,36 +1178,21 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
goto out;
}
/*
* get the inode first so any iput calls done for the io_list
* aren't the final iput (no unlinks allowed now)
*/
inode = lookup_free_space_inode(block_group, path);
mutex_lock(&trans->transaction->cache_write_mutex);
/*
* Make sure our free space cache IO is done before removing the
* free space inode
* Do not delete the block group item while
* btrfs_start_dirty_block_groups() is updating it.
*/
mutex_lock(&trans->transaction->dirty_bgs_update_mutex);
spin_lock(&trans->transaction->dirty_bgs_lock);
if (!list_empty(&block_group->io_list)) {
list_del_init(&block_group->io_list);
WARN_ON(!IS_ERR(inode) && inode != block_group->io_ctl.inode);
spin_unlock(&trans->transaction->dirty_bgs_lock);
btrfs_wait_cache_io(trans, block_group, path);
btrfs_put_block_group(block_group);
spin_lock(&trans->transaction->dirty_bgs_lock);
}
if (!list_empty(&block_group->dirty_list)) {
list_del_init(&block_group->dirty_list);
remove_rsv = true;
btrfs_put_block_group(block_group);
}
spin_unlock(&trans->transaction->dirty_bgs_lock);
mutex_unlock(&trans->transaction->cache_write_mutex);
mutex_unlock(&trans->transaction->dirty_bgs_update_mutex);
ret = btrfs_remove_free_space_inode(trans, inode, block_group);
if (unlikely(ret)) {
@ -1287,7 +1256,6 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
spin_lock(&trans->transaction->dirty_bgs_lock);
WARN_ON(!list_empty(&block_group->dirty_list));
WARN_ON(!list_empty(&block_group->io_list));
spin_unlock(&trans->transaction->dirty_bgs_lock);
btrfs_remove_free_space_cache(block_group);
@ -1397,12 +1365,14 @@ struct btrfs_trans_handle *btrfs_start_trans_remove_block_group(
ASSERT(map->start == chunk_offset);
/*
* We need to reserve 3 + N units from the metadata space info in order
* We need to reserve 4 + N units from the metadata space info in order
* to remove a block group (done at btrfs_remove_chunk() and at
* btrfs_remove_block_group()), which are used for:
*
* 1 unit for adding the free space inode's orphan (located in the tree
* of tree roots).
* 1 unit for updating the free space inode's item (located in the tree
* of tree roots).
* 1 unit for deleting the block group item (located in the extent
* tree).
* 1 unit for deleting the free space item (located in tree of tree
@ -1415,7 +1385,7 @@ struct btrfs_trans_handle *btrfs_start_trans_remove_block_group(
* more device items and remove one chunk item), but this is done at
* btrfs_remove_chunk() through a call to check_system_chunk().
*/
num_items = 3 + map->num_stripes;
num_items = 4 + map->num_stripes;
btrfs_free_chunk_map(map);
return btrfs_start_transaction_fallback_global_rsv(root, num_items);
@ -2471,7 +2441,6 @@ static struct btrfs_block_group *btrfs_create_block_group(
INIT_LIST_HEAD(&cache->ro_list);
INIT_LIST_HEAD(&cache->discard_list);
INIT_LIST_HEAD(&cache->dirty_list);
INIT_LIST_HEAD(&cache->io_list);
INIT_LIST_HEAD(&cache->active_bg_list);
btrfs_init_free_space_ctl(cache, cache->free_space_ctl);
atomic_set(&cache->frozen, 0);
@ -2527,8 +2496,7 @@ static int check_chunk_block_group_mappings(struct btrfs_fs_info *fs_info)
static int read_one_block_group(struct btrfs_fs_info *info,
struct btrfs_block_group_item_v2 *bgi,
const struct btrfs_key *key,
bool need_clear)
const struct btrfs_key *key)
{
struct btrfs_block_group *cache;
const bool mixed = btrfs_fs_incompat(info, MIXED_GROUPS);
@ -2554,20 +2522,6 @@ static int read_one_block_group(struct btrfs_fs_info *info,
btrfs_set_free_space_tree_thresholds(cache);
if (need_clear) {
/*
* When we mount with old space cache, we need to
* set BTRFS_DC_CLEAR and set dirty flag.
*
* a) Setting 'BTRFS_DC_CLEAR' makes sure that we
* truncate the old free space cache inode and
* setup a new one.
* b) Setting 'dirty flag' makes sure that we flush
* the new space cache info onto disk.
*/
if (btrfs_test_opt(info, SPACE_CACHE))
cache->disk_cache_state = BTRFS_DC_CLEAR;
}
if (!mixed && ((cache->flags & BTRFS_BLOCK_GROUP_METADATA) &&
(cache->flags & BTRFS_BLOCK_GROUP_DATA))) {
btrfs_err(info,
@ -2708,8 +2662,6 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
struct btrfs_block_group *cache;
struct btrfs_space_info *space_info;
struct btrfs_key key;
bool need_clear = false;
u64 cache_gen;
/*
* Either no extent root (with ibadroots rescue option) or we have
@ -2730,13 +2682,6 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
if (!path)
return -ENOMEM;
cache_gen = btrfs_super_cache_generation(info->super_copy);
if (btrfs_test_opt(info, SPACE_CACHE) &&
btrfs_super_generation(info->super_copy) != cache_gen)
need_clear = true;
if (btrfs_test_opt(info, CLEAR_CACHE))
need_clear = true;
while (1) {
struct btrfs_block_group_item_v2 bgi;
struct extent_buffer *leaf;
@ -2765,7 +2710,7 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
btrfs_item_key_to_cpu(leaf, &key, slot);
btrfs_release_path(path);
ret = read_one_block_group(info, &bgi, &key, need_clear);
ret = read_one_block_group(info, &bgi, &key);
if (ret < 0)
goto error;
key.objectid += key.offset;
@ -3413,207 +3358,16 @@ static int update_block_group_item(struct btrfs_trans_handle *trans,
}
static void cache_save_setup(struct btrfs_block_group *block_group,
struct btrfs_trans_handle *trans,
struct btrfs_path *path)
{
struct btrfs_fs_info *fs_info = block_group->fs_info;
struct inode *inode = NULL;
struct extent_changeset *data_reserved = NULL;
u64 alloc_hint = 0;
int dcs = BTRFS_DC_ERROR;
u64 cache_size = 0;
int retries = 0;
int ret = 0;
if (!btrfs_test_opt(fs_info, SPACE_CACHE))
return;
/*
* If this block group is smaller than 100 megs don't bother caching the
* block group.
*/
if (block_group->length < (100 * SZ_1M)) {
spin_lock(&block_group->lock);
block_group->disk_cache_state = BTRFS_DC_WRITTEN;
spin_unlock(&block_group->lock);
return;
}
if (TRANS_ABORTED(trans))
return;
again:
inode = lookup_free_space_inode(block_group, path);
if (IS_ERR(inode) && PTR_ERR(inode) != -ENOENT) {
ret = PTR_ERR(inode);
btrfs_release_path(path);
goto out;
}
if (IS_ERR(inode)) {
if (retries) {
ret = PTR_ERR(inode);
btrfs_err(fs_info,
"failed to lookup free space inode after creation for block group %llu: %d",
block_group->start, ret);
goto out_free;
}
retries++;
if (block_group->ro)
goto out_free;
ret = create_free_space_inode(trans, block_group, path);
if (ret)
goto out_free;
goto again;
}
/*
* 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
* time.
*/
BTRFS_I(inode)->generation = 0;
ret = btrfs_update_inode(trans, BTRFS_I(inode));
if (unlikely(ret)) {
/*
* So theoretically we could recover from this, simply set the
* super cache generation to 0 so we know to invalidate the
* cache, but then we'd have to keep track of the block groups
* that fail this way so we know we _have_ to reset this cache
* before the next commit or risk reading stale cache. So to
* limit our exposure to horrible edge cases lets just abort the
* transaction, this only happens in really bad situations
* anyway.
*/
btrfs_abort_transaction(trans, ret);
goto out_put;
}
/* 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;
}
if (i_size_read(inode) > 0) {
ret = btrfs_check_trunc_cache_free_space(fs_info,
&fs_info->global_block_rsv);
if (ret)
goto out_put;
ret = btrfs_truncate_free_space_cache(trans, NULL, inode);
if (ret)
goto out_put;
}
spin_lock(&block_group->lock);
if (block_group->cached != BTRFS_CACHE_FINISHED ||
!btrfs_test_opt(fs_info, SPACE_CACHE)) {
/*
* don't bother trying to write stuff out _if_
* a) we're not cached,
* b) we're with nospace_cache mount option,
* c) we're with v2 space_cache (FREE_SPACE_TREE).
*/
dcs = BTRFS_DC_WRITTEN;
spin_unlock(&block_group->lock);
goto out_put;
}
spin_unlock(&block_group->lock);
/*
* We hit an ENOSPC when setting up the cache in this transaction, just
* skip doing the setup, we've already cleared the cache so we're safe.
*/
if (test_bit(BTRFS_TRANS_CACHE_ENOSPC, &trans->transaction->flags))
goto out_put;
/*
* Try to preallocate enough space based on how big the block group is.
* Keep in mind this has to include any pinned space which could end up
* taking up quite a bit since it's not folded into the other space
* cache.
*/
cache_size = div_u64(block_group->length, SZ_256M);
if (!cache_size)
cache_size = 1;
cache_size *= 16;
cache_size *= fs_info->sectorsize;
ret = btrfs_check_data_free_space(BTRFS_I(inode), &data_reserved, 0,
cache_size, false);
if (ret)
goto out_put;
ret = btrfs_prealloc_file_range_trans(inode, trans, 0, 0, cache_size,
cache_size, cache_size,
&alloc_hint);
/*
* Our cache requires contiguous chunks so that we don't modify a bunch
* of metadata or split extents when writing the cache out, which means
* we can enospc if we are heavily fragmented in addition to just normal
* out of space conditions. So if we hit this just skip setting up any
* other block groups for this transaction, maybe we'll unpin enough
* space the next time around.
*/
if (!ret)
dcs = BTRFS_DC_SETUP;
else if (ret == -ENOSPC)
set_bit(BTRFS_TRANS_CACHE_ENOSPC, &trans->transaction->flags);
out_put:
iput(inode);
out_free:
btrfs_release_path(path);
out:
spin_lock(&block_group->lock);
if (!ret && dcs == BTRFS_DC_SETUP)
block_group->cache_generation = trans->transid;
block_group->disk_cache_state = dcs;
spin_unlock(&block_group->lock);
extent_changeset_free(data_reserved);
}
int btrfs_setup_space_cache(struct btrfs_trans_handle *trans)
{
struct btrfs_fs_info *fs_info = trans->fs_info;
struct btrfs_block_group *cache, *tmp;
struct btrfs_transaction *cur_trans = trans->transaction;
BTRFS_PATH_AUTO_FREE(path);
if (list_empty(&cur_trans->dirty_bgs) ||
!btrfs_test_opt(fs_info, SPACE_CACHE))
return 0;
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
/* Could add new block groups, use _safe just in case */
list_for_each_entry_safe(cache, tmp, &cur_trans->dirty_bgs,
dirty_list) {
if (cache->disk_cache_state == BTRFS_DC_CLEAR)
cache_save_setup(cache, trans, path);
}
return 0;
}
/*
* Transaction commit does final block group cache writeback during a critical
* Transaction commit does the final block group item updates during a critical
* section where nothing is allowed to change the FS. This is required in
* order for the cache to actually match the block group, but can introduce a
* order for the items to actually match the block groups, but can introduce a
* lot of latency into the commit.
*
* So, btrfs_start_dirty_block_groups is here to kick off block group cache IO.
* There's a chance we'll have to redo some of it if the block group changes
* again during the commit, but it greatly reduces the commit latency by
* getting rid of the easy block groups while we're still allowing others to
* So, btrfs_start_dirty_block_groups is here to update the block group items
* early. There's a chance we'll have to redo some of it if the block group
* changes again during the commit, but it greatly reduces the commit latency
* by getting rid of the easy block groups while we're still allowing others to
* join the commit.
*/
int btrfs_start_dirty_block_groups(struct btrfs_trans_handle *trans)
@ -3622,10 +3376,8 @@ int btrfs_start_dirty_block_groups(struct btrfs_trans_handle *trans)
struct btrfs_block_group *cache;
struct btrfs_transaction *cur_trans = trans->transaction;
int ret = 0;
int should_put;
BTRFS_PATH_AUTO_FREE(path);
LIST_HEAD(dirty);
struct list_head *io = &cur_trans->io_bgs;
int loops = 0;
spin_lock(&cur_trans->dirty_bgs_lock);
@ -3649,33 +3401,18 @@ int btrfs_start_dirty_block_groups(struct btrfs_trans_handle *trans)
}
/*
* cache_write_mutex is here only to save us from balance or automatic
* removal of empty block groups deleting this block group while we are
* writing out the cache
* dirty_bgs_update_mutex is here only to save us from balance or
* automatic removal of empty block groups deleting this block group
* while we are updating its item
*/
mutex_lock(&trans->transaction->cache_write_mutex);
mutex_lock(&trans->transaction->dirty_bgs_update_mutex);
while (!list_empty(&dirty)) {
bool drop_reserve = true;
cache = list_first_entry(&dirty, struct btrfs_block_group,
dirty_list);
/*
* This can happen if something re-dirties a block group that
* is already under IO. Just wait for it to finish and then do
* it all again
*/
if (!list_empty(&cache->io_list)) {
list_del_init(&cache->io_list);
btrfs_wait_cache_io(trans, cache, path);
btrfs_put_block_group(cache);
}
/*
* btrfs_wait_cache_io uses the cache->dirty_list to decide if
* it should update the cache_state. Don't delete until after
* we wait.
*
* Since we're not running in the commit critical section
* we need the dirty_bgs_lock to protect from update_block_group
*/
@ -3683,72 +3420,39 @@ int btrfs_start_dirty_block_groups(struct btrfs_trans_handle *trans)
list_del_init(&cache->dirty_list);
spin_unlock(&cur_trans->dirty_bgs_lock);
should_put = 1;
cache_save_setup(cache, trans, path);
if (cache->disk_cache_state == BTRFS_DC_SETUP) {
cache->io_ctl.inode = NULL;
ret = btrfs_write_out_cache(trans, cache, path);
if (ret == 0 && cache->io_ctl.inode) {
should_put = 0;
/*
* The cache_write_mutex is protecting the
* io_list, also refer to the definition of
* btrfs_transaction::io_bgs for more details
*/
list_add_tail(&cache->io_list, io);
} else {
/*
* If we failed to write the cache, the
* generation will be bad and life goes on
*/
ret = 0;
}
}
if (!ret) {
ret = update_block_group_item(trans, path, cache);
/*
* Our block group might still be attached to the list
* of new block groups in the transaction handle of some
* other task (struct btrfs_trans_handle->new_bgs). This
* means its block group item isn't yet in the extent
* tree. If this happens ignore the error, as we will
* try again later in the critical section of the
* transaction commit.
*/
if (ret == -ENOENT) {
ret = 0;
spin_lock(&cur_trans->dirty_bgs_lock);
if (list_empty(&cache->dirty_list)) {
list_add_tail(&cache->dirty_list,
&cur_trans->dirty_bgs);
btrfs_get_block_group(cache);
drop_reserve = false;
}
spin_unlock(&cur_trans->dirty_bgs_lock);
} else if (ret) {
btrfs_abort_transaction(trans, ret);
ret = update_block_group_item(trans, path, cache);
/*
* Our block group might still be attached to the list of new
* block groups in the transaction handle of some other task
* (struct btrfs_trans_handle->new_bgs). This means its block
* group item isn't yet in the extent tree. If this happens
* ignore the error, as we will try again later in the critical
* section of the transaction commit.
*/
if (ret == -ENOENT) {
ret = 0;
spin_lock(&cur_trans->dirty_bgs_lock);
if (list_empty(&cache->dirty_list)) {
list_add_tail(&cache->dirty_list,
&cur_trans->dirty_bgs);
btrfs_get_block_group(cache);
drop_reserve = false;
}
spin_unlock(&cur_trans->dirty_bgs_lock);
} else if (ret) {
btrfs_abort_transaction(trans, ret);
}
/* If it's not on the io list, we need to put the block group */
if (should_put)
btrfs_put_block_group(cache);
btrfs_put_block_group(cache);
if (drop_reserve)
btrfs_dec_delayed_refs_rsv_bg_updates(fs_info);
/*
* Avoid blocking other tasks for too long. It might even save
* us from writing caches for block groups that are going to be
* removed.
*/
mutex_unlock(&trans->transaction->cache_write_mutex);
/* Avoid blocking other tasks for too long. */
mutex_unlock(&trans->transaction->dirty_bgs_update_mutex);
if (ret)
goto out;
mutex_lock(&trans->transaction->cache_write_mutex);
mutex_lock(&trans->transaction->dirty_bgs_update_mutex);
}
mutex_unlock(&trans->transaction->cache_write_mutex);
mutex_unlock(&trans->transaction->dirty_bgs_update_mutex);
/*
* Go through delayed refs for all the stuff we've just kicked off
@ -3762,7 +3466,7 @@ int btrfs_start_dirty_block_groups(struct btrfs_trans_handle *trans)
list_splice_init(&cur_trans->dirty_bgs, &dirty);
/*
* dirty_bgs_lock protects us from concurrent block group
* deletes too (not just cache_write_mutex).
* deletes too (not just dirty_bgs_update_mutex).
*/
if (!list_empty(&dirty)) {
spin_unlock(&cur_trans->dirty_bgs_lock);
@ -3787,121 +3491,34 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans)
struct btrfs_block_group *cache;
struct btrfs_transaction *cur_trans = trans->transaction;
int ret = 0;
int should_put;
BTRFS_PATH_AUTO_FREE(path);
struct list_head *io = &cur_trans->io_bgs;
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
/*
* Even though we are in the critical section of the transaction commit,
* we can still have concurrent tasks adding elements to this
* transaction's list of dirty block groups. These tasks correspond to
* endio free space workers started when writeback finishes for a
* space cache, which run inode.c:btrfs_finish_ordered_io(), and can
* allocate new block groups as a result of COWing nodes of the root
* tree when updating the free space inode. The writeback for the space
* caches is triggered by an earlier call to
* btrfs_start_dirty_block_groups() and iterations of the following
* loop.
* Also we want to do the cache_save_setup first and then run the
* delayed refs to make sure we have the best chance at doing this all
* in one shot.
*/
spin_lock(&cur_trans->dirty_bgs_lock);
while (!list_empty(&cur_trans->dirty_bgs)) {
cache = list_first_entry(&cur_trans->dirty_bgs,
struct btrfs_block_group,
dirty_list);
/*
* This can happen if cache_save_setup re-dirties a block group
* that is already under IO. Just wait for it to finish and
* then do it all again
*/
if (!list_empty(&cache->io_list)) {
spin_unlock(&cur_trans->dirty_bgs_lock);
list_del_init(&cache->io_list);
btrfs_wait_cache_io(trans, cache, path);
btrfs_put_block_group(cache);
spin_lock(&cur_trans->dirty_bgs_lock);
}
/*
* Don't remove from the dirty list until after we've waited on
* any pending IO
*/
list_del_init(&cache->dirty_list);
spin_unlock(&cur_trans->dirty_bgs_lock);
should_put = 1;
cache_save_setup(cache, trans, path);
if (!ret)
ret = btrfs_run_delayed_refs(trans, U64_MAX);
if (!ret && cache->disk_cache_state == BTRFS_DC_SETUP) {
cache->io_ctl.inode = NULL;
ret = btrfs_write_out_cache(trans, cache, path);
if (ret == 0 && cache->io_ctl.inode) {
should_put = 0;
list_add_tail(&cache->io_list, io);
} else {
/*
* If we failed to write the cache, the
* generation will be bad and life goes on
*/
ret = 0;
}
}
if (!ret) {
ret = update_block_group_item(trans, path, cache);
/*
* One of the free space endio workers might have
* created a new block group while updating a free space
* cache's inode (at inode.c:btrfs_finish_ordered_io())
* and hasn't released its transaction handle yet, in
* which case the new block group is still attached to
* its transaction handle and its creation has not
* finished yet (no block group item in the extent tree
* yet, etc). If this is the case, wait for all free
* space endio workers to finish and retry. This is a
* very rare case so no need for a more efficient and
* complex approach.
*/
if (ret == -ENOENT) {
wait_event(cur_trans->writer_wait,
atomic_read(&cur_trans->num_writers) == 1);
ret = update_block_group_item(trans, path, cache);
if (ret)
btrfs_abort_transaction(trans, ret);
} else if (ret) {
if (ret)
btrfs_abort_transaction(trans, ret);
}
}
/* If its not on the io list, we need to put the block group */
if (should_put)
btrfs_put_block_group(cache);
btrfs_put_block_group(cache);
btrfs_dec_delayed_refs_rsv_bg_updates(fs_info);
spin_lock(&cur_trans->dirty_bgs_lock);
}
spin_unlock(&cur_trans->dirty_bgs_lock);
/*
* Refer to the definition of io_bgs member for details why it's safe
* to use it without any locking
*/
while (!list_empty(io)) {
cache = list_first_entry(io, struct btrfs_block_group,
io_list);
list_del_init(&cache->io_list);
btrfs_wait_cache_io(trans, cache, path);
btrfs_put_block_group(cache);
}
return ret;
}
@ -3945,10 +3562,10 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
factor = btrfs_bg_type_to_factor(cache->flags);
/*
* If this block group has free space cache written out, we need to make
* sure to load it if we are removing space. This is because we need
* the unpinning stage to actually add the space back to the block group,
* otherwise we will leak space.
* Make sure the free space of this block group is loaded if we are
* removing space. This is because we need the unpinning stage to
* actually add the space back to the block group, otherwise we will
* leak space.
*/
if (!alloc && !btrfs_block_group_done(cache))
btrfs_cache_block_group(cache, true);
@ -3956,10 +3573,6 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
spin_lock(&space_info->lock);
spin_lock(&cache->lock);
if (btrfs_test_opt(info, SPACE_CACHE) &&
cache->disk_cache_state < BTRFS_DC_CLEAR)
cache->disk_cache_state = BTRFS_DC_CLEAR;
old_val = cache->used;
if (alloc) {
old_val += num_bytes;
@ -4004,7 +3617,7 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
/*
* No longer have used bytes in this block group, queue it for deletion.
* We do this after adding the block group to the dirty list to avoid
* races between cleaner kthread and space cache writeout.
* races between the cleaner kthread and the dirty block group writeout.
*/
if (!alloc && old_val == 0) {
if (!btrfs_test_opt(info, DISCARD_ASYNC))
@ -4683,7 +4296,6 @@ void btrfs_put_block_group_cache(struct btrfs_fs_info *info)
block_group->inode = NULL;
spin_unlock(&block_group->lock);
ASSERT(block_group->io_ctl.inode == NULL);
iput(&inode->vfs_inode);
} else {
spin_unlock(&block_group->lock);
@ -4820,7 +4432,6 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
btrfs_remove_free_space_cache(block_group);
ASSERT(block_group->cached != BTRFS_CACHE_STARTED);
ASSERT(list_empty(&block_group->dirty_list));
ASSERT(list_empty(&block_group->io_list));
ASSERT(list_empty(&block_group->bg_list));
ASSERT(refcount_read(&block_group->refs) == 1);
ASSERT(block_group->swap_extents == 0);

14
fs/btrfs/block-group.h
View File

@ -20,13 +20,6 @@ struct btrfs_fs_info;
struct btrfs_inode;
struct btrfs_trans_handle;
enum btrfs_disk_cache_state {
BTRFS_DC_WRITTEN,
BTRFS_DC_ERROR,
BTRFS_DC_CLEAR,
BTRFS_DC_SETUP,
};
enum btrfs_block_group_size_class {
/* Unset */
BTRFS_BG_SZ_NONE,
@ -131,7 +124,6 @@ struct btrfs_block_group {
u64 delalloc_bytes;
u64 bytes_super;
u64 flags;
u64 cache_generation;
u64 global_root_id;
u64 remap_bytes;
u32 identity_remap_count;
@ -171,8 +163,6 @@ struct btrfs_block_group {
unsigned long full_stripe_len;
unsigned long runtime_flags;
enum btrfs_disk_cache_state disk_cache_state;
/* Cache tracking stuff */
enum btrfs_caching_type cached;
struct btrfs_caching_control *caching_ctl;
@ -228,9 +218,6 @@ struct btrfs_block_group {
/* For dirty block groups */
struct list_head dirty_list;
struct list_head io_list;
struct btrfs_io_ctl io_ctl;
/*
* Incremented when doing extent allocations and holding a read lock
@ -368,7 +355,6 @@ int btrfs_inc_block_group_ro(struct btrfs_block_group *cache,
void btrfs_dec_block_group_ro(struct btrfs_block_group *cache);
int btrfs_start_dirty_block_groups(struct btrfs_trans_handle *trans);
int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans);
int btrfs_setup_space_cache(struct btrfs_trans_handle *trans);
int btrfs_update_block_group(struct btrfs_trans_handle *trans,
u64 bytenr, u64 num_bytes, bool alloc);
int btrfs_add_reserved_bytes(struct btrfs_block_group *cache,

6
fs/btrfs/btrfs_inode.h
View File

@ -409,8 +409,6 @@ static inline void btrfs_mod_outstanding_extents(struct btrfs_inode *inode,
{
lockdep_assert_held(&inode->lock);
inode->outstanding_extents += mod;
if (btrfs_is_free_space_inode(inode))
return;
trace_btrfs_inode_mod_outstanding_extents(inode->root, btrfs_ino(inode),
mod, inode->outstanding_extents);
}
@ -598,10 +596,6 @@ int btrfs_wait_on_delayed_iputs(struct btrfs_fs_info *fs_info);
int btrfs_prealloc_file_range(struct inode *inode, int mode,
u64 start, u64 num_bytes, u64 min_size,
loff_t actual_len, u64 *alloc_hint);
int btrfs_prealloc_file_range_trans(struct inode *inode,
struct btrfs_trans_handle *trans, int mode,
u64 start, u64 num_bytes, u64 min_size,
loff_t actual_len, u64 *alloc_hint);
int btrfs_run_delalloc_range(struct btrfs_inode *inode, struct folio *locked_folio,
u64 start, u64 end, struct writeback_control *wbc);
void btrfs_queue_writepage_fixup(struct btrfs_inode *inode, struct folio *folio);

13
fs/btrfs/delalloc-space.c
View File

@ -132,9 +132,7 @@ int btrfs_alloc_data_chunk_ondemand(const struct btrfs_inode *inode, u64 bytes)
/* Make sure bytes are sectorsize aligned */
bytes = ALIGN(bytes, fs_info->sectorsize);
if (btrfs_is_free_space_inode(inode))
flush = BTRFS_RESERVE_FLUSH_FREE_SPACE_INODE;
else if (btrfs_is_zoned(fs_info) && btrfs_is_data_reloc_root(root))
if (btrfs_is_zoned(fs_info) && btrfs_is_data_reloc_root(root))
flush = BTRFS_RESERVE_FLUSH_ZONED_RELOCATION;
return btrfs_reserve_data_bytes(data_sinfo_for_inode(inode), bytes, flush);
@ -155,8 +153,6 @@ int btrfs_check_data_free_space(struct btrfs_inode *inode,
if (noflush)
flush = BTRFS_RESERVE_NO_FLUSH;
else if (btrfs_is_free_space_inode(inode))
flush = BTRFS_RESERVE_FLUSH_FREE_SPACE_INODE;
ret = btrfs_reserve_data_bytes(data_sinfo_for_inode(inode), len, flush);
if (ret < 0)
@ -326,15 +322,10 @@ int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes,
int ret = 0;
/*
* If we are a free space inode we need to not flush since we will be in
* the middle of a transaction commit. We also don't need the delalloc
* mutex since we won't race with anybody. We need this mostly to make
* lockdep shut its filthy mouth.
*
* If we have a transaction open (can happen if we call truncate_block
* from truncate), then we need FLUSH_LIMIT so we don't deadlock.
*/
if (noflush || btrfs_is_free_space_inode(inode)) {
if (noflush) {
flush = BTRFS_RESERVE_NO_FLUSH;
} else {
if (current->journal_info)

81
fs/btrfs/disk-io.c
View File

@ -1780,7 +1780,6 @@ static void btrfs_stop_all_workers(struct btrfs_fs_info *fs_info)
if (fs_info->rmw_workers)
destroy_workqueue(fs_info->rmw_workers);
btrfs_destroy_workqueue(fs_info->endio_write_workers);
btrfs_destroy_workqueue(fs_info->endio_freespace_worker);
btrfs_destroy_workqueue(fs_info->delayed_workers);
btrfs_destroy_workqueue(fs_info->caching_workers);
btrfs_destroy_workqueue(fs_info->flush_workers);
@ -1991,9 +1990,6 @@ static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info)
fs_info->endio_write_workers =
btrfs_alloc_workqueue(fs_info, "endio-write", flags,
max_active, 2);
fs_info->endio_freespace_worker =
btrfs_alloc_workqueue(fs_info, "freespace-write", flags,
max_active, 0);
fs_info->delayed_workers =
btrfs_alloc_workqueue(fs_info, "delayed-meta", flags,
max_active, 0);
@ -2006,8 +2002,7 @@ static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info)
if (!(fs_info->workers &&
fs_info->delalloc_workers && fs_info->flush_workers &&
fs_info->endio_workers && fs_info->endio_meta_workers &&
fs_info->endio_write_workers &&
fs_info->endio_freespace_worker && fs_info->rmw_workers &&
fs_info->endio_write_workers && fs_info->rmw_workers &&
fs_info->caching_workers && fs_info->fixup_workers &&
fs_info->delayed_workers && fs_info->qgroup_rescan_workers &&
fs_info->discard_ctl.discard_workers)) {
@ -3062,7 +3057,6 @@ static int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info)
int btrfs_start_pre_rw_mount(struct btrfs_fs_info *fs_info)
{
int ret;
const bool cache_opt = btrfs_test_opt(fs_info, SPACE_CACHE);
bool rebuild_free_space_tree = false;
if (btrfs_test_opt(fs_info, CLEAR_CACHE) &&
@ -3157,8 +3151,8 @@ int btrfs_start_pre_rw_mount(struct btrfs_fs_info *fs_info)
}
}
if (cache_opt != btrfs_free_space_cache_v1_active(fs_info)) {
ret = btrfs_set_free_space_cache_v1_active(fs_info, cache_opt);
if (btrfs_free_space_cache_v1_active(fs_info)) {
ret = btrfs_cleanup_free_space_cache_v1(fs_info);
if (ret)
return ret;
}
@ -3274,20 +3268,6 @@ int btrfs_check_features(struct btrfs_fs_info *fs_info, bool is_rw_mount)
return -EINVAL;
}
/*
* Subpage/bs > ps runtime limitation on v1 cache.
*
* V1 space cache still has some hard coded PAGE_SIZE usage, while
* we're already defaulting to v2 cache, no need to bother v1 as it's
* going to be deprecated anyway.
*/
if (fs_info->sectorsize != PAGE_SIZE && btrfs_test_opt(fs_info, SPACE_CACHE)) {
btrfs_warn(fs_info,
"v1 space cache is not supported for page size %lu with sectorsize %u",
PAGE_SIZE, fs_info->sectorsize);
return -EINVAL;
}
/* This can be called by remount, we need to protect the super block. */
spin_lock(&fs_info->super_lock);
btrfs_set_super_incompat_flags(disk_super, incompat);
@ -4441,9 +4421,8 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info)
* to finish an ordered extent - end_bbio_compressed_write()
* calls btrfs_finish_ordered_extent() which in turns does a call to
* btrfs_queue_ordered_fn(), and that queues the ordered extent
* completion either in the endio_write_workers work queue or in the
* fs_info->endio_freespace_worker work queue. We flush those queues
* below, so before we flush them we must flush this queue for the
* completion in the endio_write_workers work queue. We flush that
* queue below, so before we flush it we must flush this queue for the
* workers of compressed writes.
*/
flush_workqueue(fs_info->endio_workers);
@ -4469,8 +4448,6 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info)
* btrfs_finish_ordered_io() when we are unmounting).
*/
btrfs_flush_workqueue(fs_info->endio_write_workers);
/* Ordered extents for free space inodes. */
btrfs_flush_workqueue(fs_info->endio_freespace_worker);
/*
* Run delayed iputs in case an async reclaim worker is waiting for them
* to be run as mentioned above.
@ -4859,26 +4836,6 @@ static void btrfs_destroy_pinned_extent(struct btrfs_fs_info *fs_info,
}
}
static void btrfs_cleanup_bg_io(struct btrfs_block_group *cache)
{
struct inode *inode;
inode = cache->io_ctl.inode;
if (inode) {
unsigned int nofs_flag;
nofs_flag = memalloc_nofs_save();
invalidate_inode_pages2(inode->i_mapping);
memalloc_nofs_restore(nofs_flag);
BTRFS_I(inode)->generation = 0;
cache->io_ctl.inode = NULL;
iput(inode);
}
ASSERT(cache->io_ctl.pages == NULL);
btrfs_put_block_group(cache);
}
void btrfs_cleanup_dirty_bgs(struct btrfs_transaction *cur_trans,
struct btrfs_fs_info *fs_info)
{
@ -4890,40 +4847,13 @@ void btrfs_cleanup_dirty_bgs(struct btrfs_transaction *cur_trans,
struct btrfs_block_group,
dirty_list);
if (!list_empty(&cache->io_list)) {
spin_unlock(&cur_trans->dirty_bgs_lock);
list_del_init(&cache->io_list);
btrfs_cleanup_bg_io(cache);
spin_lock(&cur_trans->dirty_bgs_lock);
}
list_del_init(&cache->dirty_list);
spin_lock(&cache->lock);
cache->disk_cache_state = BTRFS_DC_ERROR;
spin_unlock(&cache->lock);
spin_unlock(&cur_trans->dirty_bgs_lock);
btrfs_put_block_group(cache);
btrfs_dec_delayed_refs_rsv_bg_updates(fs_info);
spin_lock(&cur_trans->dirty_bgs_lock);
}
spin_unlock(&cur_trans->dirty_bgs_lock);
/*
* Refer to the definition of io_bgs member for details why it's safe
* to use it without any locking
*/
while (!list_empty(&cur_trans->io_bgs)) {
cache = list_first_entry(&cur_trans->io_bgs,
struct btrfs_block_group,
io_list);
list_del_init(&cache->io_list);
spin_lock(&cache->lock);
cache->disk_cache_state = BTRFS_DC_ERROR;
spin_unlock(&cache->lock);
btrfs_cleanup_bg_io(cache);
}
}
static void btrfs_free_all_qgroup_pertrans(struct btrfs_fs_info *fs_info)
@ -4959,7 +4889,6 @@ void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans)
btrfs_cleanup_dirty_bgs(cur_trans, fs_info);
ASSERT(list_empty(&cur_trans->dirty_bgs));
ASSERT(list_empty(&cur_trans->io_bgs));
list_for_each_entry_safe(dev, tmp, &cur_trans->dev_update_list,
post_commit_list) {

11
fs/btrfs/file-item.c
View File

@ -396,17 +396,6 @@ int btrfs_lookup_bio_sums(struct btrfs_bio *bbio)
if (nblocks > fs_info->csums_per_leaf)
path->reada = READA_FORWARD;
/*
* the free space stuff is only read when it hasn't been
* updated in the current transaction. So, we can safely
* read from the commit root and sidestep a nasty deadlock
* between reading the free space cache and updating the csum tree.
*/
if (btrfs_is_free_space_inode(inode)) {
path->search_commit_root = true;
path->skip_locking = true;
}
/*
* If we are searching for a csum of an extent from a past
* transaction, we can search in the commit root and reduce

1354
fs/btrfs/free-space-cache.c
View File

File diff suppressed because it is too large Load Diff

29
fs/btrfs/free-space-cache.h
View File

@ -14,7 +14,6 @@
#include "fs.h"
struct inode;
struct page;
struct btrfs_fs_info;
struct btrfs_path;
struct btrfs_trans_handle;
@ -84,44 +83,18 @@ struct btrfs_free_space_ctl {
s32 discardable_extents[BTRFS_STAT_NR_ENTRIES];
s64 discardable_bytes[BTRFS_STAT_NR_ENTRIES];
struct btrfs_block_group *block_group;
struct mutex cache_writeout_mutex;
struct list_head trimming_ranges;
};
struct btrfs_io_ctl {
void *cur, *orig;
struct page *page;
struct page **pages;
struct btrfs_fs_info *fs_info;
struct inode *inode;
unsigned long size;
int index;
int num_pages;
int entries;
int bitmaps;
};
int __init btrfs_free_space_init(void);
void __cold btrfs_free_space_exit(void);
struct inode *lookup_free_space_inode(struct btrfs_block_group *block_group,
struct btrfs_path *path);
int create_free_space_inode(struct btrfs_trans_handle *trans,
struct btrfs_block_group *block_group,
struct btrfs_path *path);
int btrfs_remove_free_space_inode(struct btrfs_trans_handle *trans,
struct inode *inode,
struct btrfs_block_group *block_group);
int btrfs_truncate_free_space_cache(struct btrfs_trans_handle *trans,
struct btrfs_block_group *block_group,
struct inode *inode);
int load_free_space_cache(struct btrfs_block_group *block_group);
int btrfs_wait_cache_io(struct btrfs_trans_handle *trans,
struct btrfs_block_group *block_group,
struct btrfs_path *path);
int btrfs_write_out_cache(struct btrfs_trans_handle *trans,
struct btrfs_block_group *block_group,
struct btrfs_path *path);
void btrfs_init_free_space_ctl(struct btrfs_block_group *block_group,
struct btrfs_free_space_ctl *ctl);
@ -161,7 +134,7 @@ int btrfs_trim_block_group_bitmaps(struct btrfs_block_group *block_group,
void btrfs_trim_fully_remapped_block_group(struct btrfs_block_group *bg);
bool btrfs_free_space_cache_v1_active(struct btrfs_fs_info *fs_info);
int btrfs_set_free_space_cache_v1_active(struct btrfs_fs_info *fs_info, bool active);
int btrfs_cleanup_free_space_cache_v1(struct btrfs_fs_info *fs_info);
/* Support functions for running our sanity tests */
#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
bool btrfs_use_bitmap(struct btrfs_free_space_ctl *ctl,

2
fs/btrfs/fs.h
View File

@ -259,7 +259,6 @@ enum {
BTRFS_MOUNT_NOSSD = (1ULL << 9),
BTRFS_MOUNT_DISCARD_SYNC = (1ULL << 10),
BTRFS_MOUNT_FORCE_COMPRESS = (1ULL << 11),
BTRFS_MOUNT_SPACE_CACHE = (1ULL << 12),
BTRFS_MOUNT_CLEAR_CACHE = (1ULL << 13),
BTRFS_MOUNT_USER_SUBVOL_RM_ALLOWED = (1ULL << 14),
BTRFS_MOUNT_ENOSPC_DEBUG = (1ULL << 15),
@ -712,7 +711,6 @@ struct btrfs_fs_info {
struct workqueue_struct *endio_meta_workers;
struct workqueue_struct *rmw_workers;
struct btrfs_workqueue *endio_write_workers;
struct btrfs_workqueue *endio_freespace_worker;
struct btrfs_workqueue *caching_workers;
struct workqueue_struct *fixup_workers;

155
fs/btrfs/inode.c
View File

@ -1369,11 +1369,6 @@ static noinline int cow_file_range(struct btrfs_inode *inode,
goto out_unlock;
}
if (btrfs_is_free_space_inode(inode)) {
ret = -EINVAL;
goto out_unlock;
}
num_bytes = ALIGN(end - start + 1, blocksize);
num_bytes = max(blocksize, num_bytes);
ASSERT(num_bytes <= btrfs_super_total_bytes(fs_info->super_copy));
@ -1683,7 +1678,6 @@ static int fallback_to_cow(struct btrfs_inode *inode,
struct folio *locked_folio, const u64 start,
const u64 end)
{
const bool is_space_ino = btrfs_is_free_space_inode(inode);
const bool is_reloc_ino = btrfs_is_data_reloc_root(inode->root);
const u64 range_bytes = end + 1 - start;
struct extent_io_tree *io_tree = &inode->io_tree;
@ -1716,23 +1710,22 @@ static int fallback_to_cow(struct btrfs_inode *inode,
* extent_clear_unlock_delalloc()) the bytes_may_use counter of the
* data space info, which we incremented in the step above.
*
* If we need to fallback to cow and the inode corresponds to a free
* space cache inode or an inode of the data relocation tree, we must
* also increment bytes_may_use of the data space_info for the same
* reason. Space caches and relocated data extents always get a prealloc
* extent for them, however scrub or balance may have set the block
* group that contains that extent to RO mode and therefore force COW
* when starting writeback.
* If we need to fallback to cow and the inode is in the data relocation
* tree, we must also increment bytes_may_use of the data space_info for
* the same reason. Relocated data extents always get a prealloc extent,
* however scrub or balance may have set the block group that contains
* that extent to RO mode and therefore force COW when starting
* writeback.
*/
btrfs_lock_extent(io_tree, start, end, &cached_state);
count = btrfs_count_range_bits(io_tree, &range_start, end, range_bytes,
EXTENT_NORESERVE, false, NULL);
if (count > 0 || is_space_ino || is_reloc_ino) {
if (count > 0 || is_reloc_ino) {
u64 bytes = count;
struct btrfs_fs_info *fs_info = inode->root->fs_info;
struct btrfs_space_info *sinfo = fs_info->data_sinfo;
if (is_space_ino || is_reloc_ino)
if (is_reloc_ino)
bytes = range_bytes;
spin_lock(&sinfo->lock);
@ -1797,7 +1790,6 @@ static int can_nocow_file_extent(struct btrfs_path *path,
struct btrfs_inode *inode,
struct can_nocow_file_extent_args *args)
{
const bool is_freespace_inode = btrfs_is_free_space_inode(inode);
struct extent_buffer *leaf = path->nodes[0];
struct btrfs_root *root = inode->root;
struct btrfs_file_extent_item *fi;
@ -1810,8 +1802,7 @@ static int can_nocow_file_extent(struct btrfs_path *path,
bool nowait = path->nowait;
/* If there are pending snapshots for this root, we must do COW. */
if (args->writeback_path && !is_freespace_inode &&
atomic_read(&root->snapshot_force_cow))
if (args->writeback_path && atomic_read(&root->snapshot_force_cow))
goto out;
fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item);
@ -1860,7 +1851,6 @@ static int can_nocow_file_extent(struct btrfs_path *path,
ret = btrfs_cross_ref_exist(inode, key->offset - args->file_extent.offset,
args->file_extent.disk_bytenr, path);
WARN_ON_ONCE(ret > 0 && is_freespace_inode);
if (ret != 0)
goto out;
@ -1895,7 +1885,6 @@ static int can_nocow_file_extent(struct btrfs_path *path,
ret = btrfs_lookup_csums_list(csum_root, io_start,
io_start + args->file_extent.num_bytes - 1,
NULL, nowait);
WARN_ON_ONCE(ret > 0 && is_freespace_inode);
if (ret != 0)
goto out;
@ -2643,7 +2632,7 @@ void btrfs_set_delalloc_extent(struct btrfs_inode *inode, struct extent_state *s
* and are therefore protected against concurrent calls of this
* function and btrfs_clear_delalloc_extent().
*/
if (!btrfs_is_free_space_inode(inode) && prev_delalloc_bytes == 0)
if (prev_delalloc_bytes == 0)
btrfs_add_delalloc_inode(inode);
}
@ -2687,13 +2676,7 @@ void btrfs_clear_delalloc_extent(struct btrfs_inode *inode,
btrfs_mod_outstanding_extents(inode, -num_extents);
spin_unlock(&inode->lock);
/*
* We don't reserve metadata space for space cache inodes so we
* don't need to call delalloc_release_metadata if there is an
* error.
*/
if (bits & EXTENT_CLEAR_META_RESV &&
root != fs_info->tree_root)
if (bits & EXTENT_CLEAR_META_RESV)
btrfs_delalloc_release_metadata(inode, len, true);
/* For sanity tests. */
@ -2701,7 +2684,6 @@ void btrfs_clear_delalloc_extent(struct btrfs_inode *inode,
return;
if (!btrfs_is_data_reloc_root(root) &&
!btrfs_is_free_space_inode(inode) &&
!(state->state & EXTENT_NORESERVE) &&
(bits & EXTENT_CLEAR_DATA_RESV))
btrfs_free_reserved_data_space_noquota(inode, len);
@ -2719,7 +2701,7 @@ void btrfs_clear_delalloc_extent(struct btrfs_inode *inode,
* and are therefore protected against concurrent calls of this
* function and btrfs_set_delalloc_extent().
*/
if (!btrfs_is_free_space_inode(inode) && new_delalloc_bytes == 0) {
if (new_delalloc_bytes == 0) {
spin_lock(&root->delalloc_lock);
btrfs_del_delalloc_inode(inode);
spin_unlock(&root->delalloc_lock);
@ -3221,7 +3203,7 @@ int btrfs_finish_one_ordered(struct btrfs_ordered_extent *ordered_extent)
int compress_type = 0;
int ret = 0;
u64 logical_len = ordered_extent->num_bytes;
bool freespace_inode;
u64 unwritten_start;
bool truncated = false;
bool clear_reserved_extent = true;
unsigned int clear_bits = 0;
@ -3238,9 +3220,7 @@ int btrfs_finish_one_ordered(struct btrfs_ordered_extent *ordered_extent)
if (!test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags))
clear_bits |= EXTENT_DEFRAG;
freespace_inode = btrfs_is_free_space_inode(inode);
if (!freespace_inode)
btrfs_lockdep_acquire(fs_info, btrfs_ordered_extent);
btrfs_lockdep_acquire(fs_info, btrfs_ordered_extent);
if (unlikely(test_bit(BTRFS_ORDERED_IOERR, &ordered_extent->flags))) {
ret = -EIO;
@ -3275,10 +3255,7 @@ int btrfs_finish_one_ordered(struct btrfs_ordered_extent *ordered_extent)
&cached_state);
}
if (freespace_inode)
trans = btrfs_join_transaction_spacecache(root);
else
trans = btrfs_join_transaction(root);
trans = btrfs_join_transaction(root);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
trans = NULL;
@ -3390,27 +3367,13 @@ int btrfs_finish_one_ordered(struct btrfs_ordered_extent *ordered_extent)
/*
* Drop extent maps for the part of the extent we didn't write.
*
* We have an exception here for the free_space_inode, this is
* because when we do btrfs_get_extent() on the free space inode
* we will search the commit root. If this is a new block group
* we won't find anything, and we will trip over the assert in
* writepage where we do ASSERT(em->block_start !=
* EXTENT_MAP_HOLE).
*
* Theoretically we could also skip this for any NOCOW extent as
* we don't mess with the extent map tree in the NOCOW case, but
* for now simply skip this if we are the free space inode.
* On error the written part is freed below as well, so drop
* its maps too.
*/
if (!btrfs_is_free_space_inode(inode)) {
u64 unwritten_start = start;
if (truncated)
unwritten_start += logical_len;
btrfs_drop_extent_map_range(inode, unwritten_start,
end, false);
}
unwritten_start = start;
if (truncated && !ret)
unwritten_start += logical_len;
btrfs_drop_extent_map_range(inode, unwritten_start, end, false);
/*
* If the ordered extent had an IOERR or something else went
@ -7269,16 +7232,6 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
/* Chances are we'll be called again, so go ahead and do readahead */
path->reada = READA_FORWARD;
/*
* The same explanation in load_free_space_cache applies here as well,
* we only read when we're loading the free space cache, and at that
* point the commit_root has everything we need.
*/
if (btrfs_is_free_space_inode(inode)) {
path->search_commit_root = true;
path->skip_locking = true;
}
ret = btrfs_lookup_file_extent(NULL, root, path, objectid, start, 0);
if (ret < 0) {
goto out;
@ -8135,7 +8088,6 @@ void btrfs_destroy_inode(struct inode *vfs_inode)
struct btrfs_ordered_extent *ordered;
struct btrfs_inode *inode = BTRFS_I(vfs_inode);
struct btrfs_root *root = inode->root;
bool freespace_inode;
WARN_ON(!hlist_empty(&vfs_inode->i_dentry));
WARN_ON(vfs_inode->i_data.nrpages);
@ -8158,12 +8110,6 @@ void btrfs_destroy_inode(struct inode *vfs_inode)
if (!root)
return;
/*
* If this is a free space inode do not take the ordered extents lockdep
* map.
*/
freespace_inode = btrfs_is_free_space_inode(inode);
while (1) {
ordered = btrfs_lookup_first_ordered_extent(inode, (u64)-1);
if (!ordered)
@ -8173,8 +8119,7 @@ void btrfs_destroy_inode(struct inode *vfs_inode)
"found ordered extent %llu %llu on inode cleanup",
ordered->file_offset, ordered->num_bytes);
if (!freespace_inode)
btrfs_lockdep_acquire(root->fs_info, btrfs_ordered_extent);
btrfs_lockdep_acquire(root->fs_info, btrfs_ordered_extent);
btrfs_remove_ordered_extent(ordered);
btrfs_put_ordered_extent(ordered);
@ -9153,14 +9098,13 @@ static int btrfs_symlink(const struct mnt_idmap *idmap, struct inode *dir,
}
static struct btrfs_trans_handle *insert_prealloc_file_extent(
struct btrfs_trans_handle *trans_in,
struct btrfs_inode *inode,
struct btrfs_key *ins,
u64 file_offset)
{
struct btrfs_file_extent_item stack_fi;
struct btrfs_replace_extent_info extent_info;
struct btrfs_trans_handle *trans = trans_in;
struct btrfs_trans_handle *trans;
struct btrfs_path *path;
u64 start = ins->objectid;
u64 len = ins->offset;
@ -9181,15 +9125,6 @@ static struct btrfs_trans_handle *insert_prealloc_file_extent(
if (ret < 0)
return ERR_PTR(ret);
if (trans) {
ret = insert_reserved_file_extent(trans, inode,
file_offset, &stack_fi,
true, qgroup_released);
if (ret)
goto free_qgroup;
return trans;
}
extent_info.disk_offset = start;
extent_info.disk_len = len;
extent_info.data_offset = 0;
@ -9229,12 +9164,12 @@ static struct btrfs_trans_handle *insert_prealloc_file_extent(
return ERR_PTR(ret);
}
static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
u64 start, u64 num_bytes, u64 min_size,
loff_t actual_len, u64 *alloc_hint,
struct btrfs_trans_handle *trans)
int btrfs_prealloc_file_range(struct inode *inode, int mode,
u64 start, u64 num_bytes, u64 min_size,
loff_t actual_len, u64 *alloc_hint)
{
struct btrfs_fs_info *fs_info = inode_to_fs_info(inode);
struct btrfs_trans_handle *trans;
struct extent_map *em;
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_key ins;
@ -9244,11 +9179,8 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
u64 cur_bytes;
u64 last_alloc = (u64)-1;
int ret = 0;
bool own_trans = true;
u64 end = start + num_bytes - 1;
if (trans)
own_trans = false;
while (num_bytes > 0) {
cur_bytes = min_t(u64, num_bytes, SZ_256M);
cur_bytes = max(cur_bytes, min_size);
@ -9274,8 +9206,8 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
clear_offset += ins.offset;
last_alloc = ins.offset;
trans = insert_prealloc_file_extent(trans, BTRFS_I(inode),
&ins, cur_offset);
trans = insert_prealloc_file_extent(BTRFS_I(inode), &ins,
cur_offset);
/*
* Now that we inserted the prealloc extent we can finally
* decrement the number of reservations in the block group.
@ -9347,8 +9279,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
range_start, range_end - range_start);
if (ret) {
btrfs_abort_transaction(trans, ret);
if (own_trans)
btrfs_end_transaction(trans);
btrfs_end_transaction(trans);
break;
}
@ -9360,15 +9291,11 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
if (unlikely(ret)) {
btrfs_abort_transaction(trans, ret);
if (own_trans)
btrfs_end_transaction(trans);
btrfs_end_transaction(trans);
break;
}
if (own_trans) {
btrfs_end_transaction(trans);
trans = NULL;
}
btrfs_end_transaction(trans);
}
if (clear_offset < end)
btrfs_free_reserved_data_space(BTRFS_I(inode), NULL, clear_offset,
@ -9376,24 +9303,6 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
return ret;
}
int btrfs_prealloc_file_range(struct inode *inode, int mode,
u64 start, u64 num_bytes, u64 min_size,
loff_t actual_len, u64 *alloc_hint)
{
return __btrfs_prealloc_file_range(inode, mode, start, num_bytes,
min_size, actual_len, alloc_hint,
NULL);
}
int btrfs_prealloc_file_range_trans(struct inode *inode,
struct btrfs_trans_handle *trans, int mode,
u64 start, u64 num_bytes, u64 min_size,
loff_t actual_len, u64 *alloc_hint)
{
return __btrfs_prealloc_file_range(inode, mode, start, num_bytes,
min_size, actual_len, alloc_hint, trans);
}
/*
* NOTE: in case you are adding MAY_EXEC check for directories:
* we are marking them with IOP_FASTPERM_MAY_EXEC, allowing path lookup to

27
fs/btrfs/ordered-data.c
View File

@ -417,13 +417,10 @@ static bool can_finish_ordered_extent(struct btrfs_ordered_extent *ordered,
static void btrfs_queue_ordered_fn(struct btrfs_ordered_extent *ordered)
{
struct btrfs_inode *inode = ordered->inode;
struct btrfs_fs_info *fs_info = inode->root->fs_info;
struct btrfs_workqueue *wq = btrfs_is_free_space_inode(inode) ?
fs_info->endio_freespace_worker : fs_info->endio_write_workers;
struct btrfs_fs_info *fs_info = ordered->inode->root->fs_info;
btrfs_init_work(&ordered->work, finish_ordered_fn, NULL);
btrfs_queue_work(wq, &ordered->work);
btrfs_queue_work(fs_info->endio_write_workers, &ordered->work);
}
void btrfs_finish_ordered_extent(struct btrfs_ordered_extent *ordered,
@ -657,13 +654,6 @@ void btrfs_remove_ordered_extent(struct btrfs_ordered_extent *entry)
struct btrfs_fs_info *fs_info = root->fs_info;
struct rb_node *node;
bool pending;
bool freespace_inode;
/*
* If this is a free space inode the thread has not acquired the ordered
* extents lockdep map.
*/
freespace_inode = btrfs_is_free_space_inode(btrfs_inode);
btrfs_lockdep_acquire(fs_info, btrfs_trans_pending_ordered);
/* This is paired with alloc_ordered_extent(). */
@ -738,8 +728,7 @@ void btrfs_remove_ordered_extent(struct btrfs_ordered_extent *entry)
}
spin_unlock(&root->ordered_extent_lock);
wake_up(&entry->wait);
if (!freespace_inode)
btrfs_lockdep_release(fs_info, btrfs_ordered_extent);
btrfs_lockdep_release(fs_info, btrfs_ordered_extent);
}
static void btrfs_run_ordered_extent_work(struct btrfs_work *work)
@ -870,16 +859,9 @@ void btrfs_start_ordered_extent_nowriteback(struct btrfs_ordered_extent *entry,
u64 start = entry->file_offset;
u64 end = start + entry->num_bytes - 1;
struct btrfs_inode *inode = entry->inode;
bool freespace_inode;
trace_btrfs_ordered_extent_start(inode, entry);
/*
* If this is a free space inode do not take the ordered extents lockdep
* map.
*/
freespace_inode = btrfs_is_free_space_inode(inode);
/*
* pages in the range can be dirty, clean or writeback. We
* start IO on any dirty ones so the wait doesn't stall waiting
@ -899,8 +881,7 @@ void btrfs_start_ordered_extent_nowriteback(struct btrfs_ordered_extent *entry,
}
}
if (!freespace_inode)
btrfs_might_wait_for_event(inode->root->fs_info, btrfs_ordered_extent);
btrfs_might_wait_for_event(inode->root->fs_info, btrfs_ordered_extent);
wait_event(entry->wait, test_bit(BTRFS_ORDERED_COMPLETE, &entry->flags));
}

2
fs/btrfs/relocation.c
View File

@ -3356,7 +3356,7 @@ static int delete_block_group_cache(struct btrfs_block_group *block_group,
goto out;
}
ret = btrfs_truncate_free_space_cache(trans, block_group, inode);
ret = btrfs_truncate_free_space_cache(trans, inode);
btrfs_end_transaction(trans);
btrfs_btree_balance_dirty(fs_info);

2
fs/btrfs/space-info.c
View File

@ -1704,7 +1704,6 @@ static int handle_reserve_ticket(struct btrfs_space_info *space_info,
evict_flush_states,
ARRAY_SIZE(evict_flush_states));
break;
case BTRFS_RESERVE_FLUSH_FREE_SPACE_INODE:
case BTRFS_RESERVE_FLUSH_ZONED_RELOCATION:
priority_reclaim_data_space(space_info, ticket);
break;
@ -1968,7 +1967,6 @@ int btrfs_reserve_data_bytes(struct btrfs_space_info *space_info, u64 bytes,
int ret;
ASSERT(flush == BTRFS_RESERVE_FLUSH_DATA ||
flush == BTRFS_RESERVE_FLUSH_FREE_SPACE_INODE ||
flush == BTRFS_RESERVE_FLUSH_ZONED_RELOCATION ||
flush == BTRFS_RESERVE_NO_FLUSH, "flush=%d", flush);
ASSERT(!current->journal_info || flush != BTRFS_RESERVE_FLUSH_DATA,

4
fs/btrfs/space-info.h
View File

@ -66,7 +66,6 @@ enum btrfs_reserve_flush_enum {
* Can be interrupted by a fatal signal.
*/
BTRFS_RESERVE_FLUSH_DATA,
BTRFS_RESERVE_FLUSH_FREE_SPACE_INODE,
BTRFS_RESERVE_FLUSH_ALL,
/*
@ -82,9 +81,6 @@ enum btrfs_reserve_flush_enum {
* priority flushing for this, because otherwise we can deadlock on
* waiting for a ticket, that cannot be granted, because we cannot do
* any allocations.
*
* Apart from being specific to zoned relocation, it is equal to
* BTRFS_FLUSH_FREE_SPACE_INODE.
*/
BTRFS_RESERVE_FLUSH_ZONED_RELOCATION,

48
fs/btrfs/super.c
View File

@ -524,7 +524,6 @@ static int btrfs_parse_param(struct fs_context *fc, struct fs_parameter *param)
btrfs_warn(NULL,
"v1 space cache is deprecated, falling back to no space cache");
btrfs_set_opt(ctx->mount_opt, NOSPACECACHE);
btrfs_clear_opt(ctx->mount_opt, SPACE_CACHE);
btrfs_clear_opt(ctx->mount_opt, FREE_SPACE_TREE);
break;
case Opt_space_cache_version:
@ -533,11 +532,9 @@ static int btrfs_parse_param(struct fs_context *fc, struct fs_parameter *param)
btrfs_warn(NULL,
"v1 space cache is deprecated, falling back to no space cache");
btrfs_set_opt(ctx->mount_opt, NOSPACECACHE);
btrfs_clear_opt(ctx->mount_opt, SPACE_CACHE);
btrfs_clear_opt(ctx->mount_opt, FREE_SPACE_TREE);
break;
case Opt_space_cache_v2:
btrfs_clear_opt(ctx->mount_opt, SPACE_CACHE);
btrfs_set_opt(ctx->mount_opt, FREE_SPACE_TREE);
break;
default:
@ -723,13 +720,6 @@ bool btrfs_check_options(const struct btrfs_fs_info *info,
if (btrfs_check_mountopts_zoned(info, mount_opt))
ret = false;
if (!test_bit(BTRFS_FS_STATE_REMOUNTING, &info->fs_state)) {
if (btrfs_raw_test_opt(*mount_opt, SPACE_CACHE)) {
btrfs_warn(info,
"space cache v1 is being deprecated and will be removed in a future release, please use -o space_cache=v2");
}
}
return ret;
}
@ -747,14 +737,6 @@ bool btrfs_check_options(const struct btrfs_fs_info *info,
*/
void btrfs_set_free_space_cache_settings(struct btrfs_fs_info *fs_info)
{
if (fs_info->sectorsize != PAGE_SIZE && btrfs_test_opt(fs_info, SPACE_CACHE)) {
btrfs_info(fs_info,
"forcing free space tree for sector size %u with page size %lu",
fs_info->sectorsize, PAGE_SIZE);
btrfs_clear_opt(fs_info->mount_opt, SPACE_CACHE);
btrfs_set_opt(fs_info->mount_opt, FREE_SPACE_TREE);
}
/*
* At this point our mount options are populated, so we only mess with
* these settings if we don't have any settings already.
@ -769,20 +751,17 @@ void btrfs_set_free_space_cache_settings(struct btrfs_fs_info *fs_info)
return;
}
if (btrfs_test_opt(fs_info, SPACE_CACHE))
return;
if (btrfs_test_opt(fs_info, NOSPACECACHE))
return;
/*
* At this point we don't have explicit options set by the user, set
* them ourselves based on the state of the file system.
* them ourselves based on the state of the file system. An existing
* v1 space cache is no longer used and gets cleaned up once the
* filesystem is mounted read-write.
*/
if (btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE))
btrfs_set_opt(fs_info->mount_opt, FREE_SPACE_TREE);
else if (btrfs_free_space_cache_v1_active(fs_info))
btrfs_set_opt(fs_info->mount_opt, SPACE_CACHE);
}
static void set_device_specific_options(struct btrfs_fs_info *fs_info)
@ -1128,9 +1107,7 @@ static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry)
seq_puts(seq, ",discard=async");
if (!(info->sb->s_flags & SB_POSIXACL))
seq_puts(seq, ",noacl");
if (btrfs_free_space_cache_v1_active(info))
seq_puts(seq, ",space_cache");
else if (btrfs_fs_compat_ro(info, FREE_SPACE_TREE))
if (btrfs_fs_compat_ro(info, FREE_SPACE_TREE))
seq_puts(seq, ",space_cache=v2");
else
seq_puts(seq, ",nospace_cache");
@ -1264,7 +1241,6 @@ static void btrfs_resize_thread_pool(struct btrfs_fs_info *fs_info,
workqueue_set_max_active(fs_info->endio_workers, new_pool_size);
workqueue_set_max_active(fs_info->endio_meta_workers, new_pool_size);
btrfs_workqueue_set_max(fs_info->endio_write_workers, new_pool_size);
btrfs_workqueue_set_max(fs_info->endio_freespace_worker, new_pool_size);
btrfs_workqueue_set_max(fs_info->delayed_workers, new_pool_size);
}
@ -1285,8 +1261,6 @@ static inline void btrfs_remount_begin(struct btrfs_fs_info *fs_info,
static inline void btrfs_remount_cleanup(struct btrfs_fs_info *fs_info,
unsigned long long old_opts)
{
const bool cache_opt = btrfs_test_opt(fs_info, SPACE_CACHE);
/*
* We need to cleanup all defraggable inodes if the autodefragment is
* close or the filesystem is read only.
@ -1303,10 +1277,6 @@ static inline void btrfs_remount_cleanup(struct btrfs_fs_info *fs_info,
else if (btrfs_raw_test_opt(old_opts, DISCARD_ASYNC) &&
!btrfs_test_opt(fs_info, DISCARD_ASYNC))
btrfs_discard_cleanup(fs_info);
/* If we toggled space cache */
if (cache_opt != btrfs_free_space_cache_v1_active(fs_info))
btrfs_set_free_space_cache_v1_active(fs_info, cache_opt);
}
static int btrfs_remount_rw(struct btrfs_fs_info *fs_info)
@ -1471,7 +1441,6 @@ static void btrfs_emit_options(struct btrfs_fs_info *info,
btrfs_info_if_set(info, old, DISCARD_SYNC, "turning on sync discard");
btrfs_info_if_set(info, old, DISCARD_ASYNC, "turning on async discard");
btrfs_info_if_set(info, old, FREE_SPACE_TREE, "enabling free space tree");
btrfs_info_if_set(info, old, SPACE_CACHE, "enabling disk space caching");
btrfs_info_if_set(info, old, CLEAR_CACHE, "force clearing of disk cache");
btrfs_info_if_set(info, old, AUTO_DEFRAG, "enabling auto defrag");
btrfs_info_if_set(info, old, FRAGMENT_DATA, "fragmenting data");
@ -1489,7 +1458,6 @@ static void btrfs_emit_options(struct btrfs_fs_info *info,
btrfs_info_if_unset(info, old, SSD_SPREAD, "not using spread ssd allocation scheme");
btrfs_info_if_unset(info, old, NOBARRIER, "turning on barriers");
btrfs_info_if_unset(info, old, NOTREELOG, "enabling tree log");
btrfs_info_if_unset(info, old, SPACE_CACHE, "disabling disk space caching");
btrfs_info_if_unset(info, old, FREE_SPACE_TREE, "disabling free space tree");
btrfs_info_if_unset(info, old, AUTO_DEFRAG, "disabling auto defrag");
btrfs_info_if_unset(info, old, COMPRESS, "use no compression");
@ -1554,14 +1522,10 @@ static int btrfs_reconfigure(struct fs_context *fc)
btrfs_warn(fs_info,
"remount supports changing free space tree only from RO to RW");
/* Make sure free space cache options match the state on disk. */
if (btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE)) {
if (btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE))
btrfs_set_opt(fs_info->mount_opt, FREE_SPACE_TREE);
btrfs_clear_opt(fs_info->mount_opt, SPACE_CACHE);
}
if (btrfs_free_space_cache_v1_active(fs_info)) {
else
btrfs_clear_opt(fs_info->mount_opt, FREE_SPACE_TREE);
btrfs_set_opt(fs_info->mount_opt, SPACE_CACHE);
}
}
ret = 0;

62
fs/btrfs/transaction.c
View File

@ -42,7 +42,6 @@ enum {
ENUM_BIT(TRANS_START),
ENUM_BIT(TRANS_ATTACH),
ENUM_BIT(TRANS_JOIN),
ENUM_BIT(TRANS_JOIN_NOLOCK),
#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
ENUM_BIT(TRANS_DUMMY),
#endif
@ -141,17 +140,14 @@ static const unsigned int btrfs_blocked_trans_types[TRANS_STATE_MAX] = {
[TRANS_STATE_UNBLOCKED] = (TRANS_START |
TRANS_ATTACH |
TRANS_JOIN |
TRANS_JOIN_NOLOCK |
TRANS_JOIN_NOSTART),
[TRANS_STATE_SUPER_COMMITTED] = (TRANS_START |
TRANS_ATTACH |
TRANS_JOIN |
TRANS_JOIN_NOLOCK |
TRANS_JOIN_NOSTART),
[TRANS_STATE_COMPLETED] = (TRANS_START |
TRANS_ATTACH |
TRANS_JOIN |
TRANS_JOIN_NOLOCK |
TRANS_JOIN_NOSTART),
};
@ -342,12 +338,6 @@ static noinline int join_transaction(struct btrfs_fs_info *fs_info,
if (type == TRANS_ATTACH || type == TRANS_JOIN_NOSTART)
return -ENOENT;
/*
* JOIN_NOLOCK only happens during the transaction commit, so
* it is impossible that ->running_transaction is NULL
*/
BUG_ON(type == TRANS_JOIN_NOLOCK);
cur_trans = kmalloc_obj(*cur_trans, GFP_NOFS);
if (!cur_trans)
return -ENOMEM;
@ -411,9 +401,8 @@ static noinline int join_transaction(struct btrfs_fs_info *fs_info,
INIT_LIST_HEAD(&cur_trans->dev_update_list);
INIT_LIST_HEAD(&cur_trans->switch_commits);
INIT_LIST_HEAD(&cur_trans->dirty_bgs);
INIT_LIST_HEAD(&cur_trans->io_bgs);
INIT_LIST_HEAD(&cur_trans->dropped_roots);
mutex_init(&cur_trans->cache_write_mutex);
mutex_init(&cur_trans->dirty_bgs_update_mutex);
spin_lock_init(&cur_trans->dirty_bgs_lock);
INIT_LIST_HEAD(&cur_trans->deleted_bgs);
spin_lock_init(&cur_trans->dropped_roots_lock);
@ -751,14 +740,8 @@ start_transaction(struct btrfs_root *root, unsigned int num_items,
/*
* Only TRANS_START and TRANS_JOIN require sb intwrite lock.
*
* If we are JOIN_NOLOCK we're already committing a transaction and
* waiting on this guy, so we don't need to do the sb_start_intwrite
* because we're already holding a ref. We need this because we could
* have raced in and did an fsync() on a file which can kick a commit
* and then we deadlock with somebody doing a freeze.
*
* If we are ATTACH, it means we just want to catch the current
* transaction and commit it, so we needn't do sb_start_intwrite().
* transaction and commit it, so we needn't do sb_start_intwrite().
*/
if (type & TRANS_SB_INTWRITER_MASK)
sb_start_intwrite(fs_info->sb);
@ -896,12 +879,6 @@ struct btrfs_trans_handle *btrfs_join_transaction(struct btrfs_root *root)
true);
}
struct btrfs_trans_handle *btrfs_join_transaction_spacecache(struct btrfs_root *root)
{
return start_transaction(root, 0, TRANS_JOIN_NOLOCK,
BTRFS_RESERVE_NO_FLUSH, true);
}
/*
* Similar to regular join but it never starts a transaction when none is
* running or when there's a running one at a state >= TRANS_STATE_UNBLOCKED.
@ -1404,7 +1381,6 @@ static noinline int commit_cowonly_roots(struct btrfs_trans_handle *trans)
{
struct btrfs_fs_info *fs_info = trans->fs_info;
struct list_head *dirty_bgs = &trans->transaction->dirty_bgs;
struct list_head *io_bgs = &trans->transaction->io_bgs;
struct extent_buffer *eb;
int ret;
@ -1434,10 +1410,6 @@ static noinline int commit_cowonly_roots(struct btrfs_trans_handle *trans)
if (ret)
return ret;
ret = btrfs_setup_space_cache(trans);
if (ret)
return ret;
again:
while (!list_empty(&fs_info->dirty_cowonly_roots)) {
struct btrfs_root *root;
@ -1458,7 +1430,7 @@ static noinline int commit_cowonly_roots(struct btrfs_trans_handle *trans)
if (ret)
return ret;
while (!list_empty(dirty_bgs) || !list_empty(io_bgs)) {
while (!list_empty(dirty_bgs)) {
ret = btrfs_write_dirty_block_groups(trans);
if (ret)
return ret;
@ -2036,9 +2008,7 @@ static void update_super_roots(struct btrfs_fs_info *fs_info)
super->root = root_item->bytenr;
super->generation = root_item->generation;
super->root_level = root_item->level;
if (btrfs_test_opt(fs_info, SPACE_CACHE))
super->cache_generation = root_item->generation;
else if (test_bit(BTRFS_FS_CLEANUP_SPACE_CACHE_V1, &fs_info->flags))
if (test_bit(BTRFS_FS_CLEANUP_SPACE_CACHE_V1, &fs_info->flags))
super->cache_generation = 0;
if (test_bit(BTRFS_FS_UPDATE_UUID_TREE_GEN, &fs_info->flags))
super->uuid_tree_generation = root_item->generation;
@ -2320,18 +2290,16 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
if (!test_bit(BTRFS_TRANS_DIRTY_BG_RUN, &cur_trans->flags)) {
bool run_it = false;
/* this mutex is also taken before trying to set
* block groups readonly. We need to make sure
* that nobody has set a block group readonly
* after a extents from that block group have been
* allocated for cache files. btrfs_set_block_group_ro
* will wait for the transaction to commit if it
* finds BTRFS_TRANS_DIRTY_BG_RUN set.
/*
* This mutex is also taken before trying to set block groups
* readonly. btrfs_inc_block_group_ro() will wait for the
* transaction to commit if it finds BTRFS_TRANS_DIRTY_BG_RUN
* set.
*
* The BTRFS_TRANS_DIRTY_BG_RUN flag is also used to make sure
* only one process starts all the block group IO. It wouldn't
* hurt to have more than one go through, but there's no
* real advantage to it either.
* only one process starts all the block group item updates. It
* wouldn't hurt to have more than one go through, but there's
* no real advantage to it either.
*/
mutex_lock(&fs_info->ro_block_group_mutex);
if (!test_and_set_bit(BTRFS_TRANS_DIRTY_BG_RUN,
@ -2565,10 +2533,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
if (unlikely(ret))
goto unlock_reloc;
/*
* The tasks which save the space cache and inode cache may also
* update ->aborted, check it.
*/
/* Other tasks may also have updated ->aborted, check it. */
if (TRANS_ABORTED(cur_trans)) {
ret = cur_trans->aborted;
goto unlock_reloc;
@ -2589,7 +2554,6 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
switch_commit_roots(trans);
ASSERT(list_empty(&cur_trans->dirty_bgs));
ASSERT(list_empty(&cur_trans->io_bgs));
update_super_roots(fs_info);
btrfs_set_super_log_root(fs_info->super_copy, 0);

27
fs/btrfs/transaction.h
View File

@ -47,7 +47,6 @@ enum btrfs_trans_state {
#define BTRFS_TRANS_HAVE_FREE_BGS 0
#define BTRFS_TRANS_DIRTY_BG_RUN 1
#define BTRFS_TRANS_CACHE_ENOSPC 2
struct btrfs_transaction {
u64 transid;
@ -78,32 +77,15 @@ struct btrfs_transaction {
struct list_head dev_update_list;
struct list_head switch_commits;
struct list_head dirty_bgs;
/*
* There is no explicit lock which protects io_bgs, rather its
* consistency is implied by the fact that all the sites which modify
* it do so under some form of transaction critical section, namely:
*
* - btrfs_start_dirty_block_groups - This function can only ever be
* run by one of the transaction committers. Refer to
* BTRFS_TRANS_DIRTY_BG_RUN usage in btrfs_commit_transaction
*
* - btrfs_write_dirty_blockgroups - this is called by
* commit_cowonly_roots from transaction critical section
* (TRANS_STATE_COMMIT_DOING)
*
* - btrfs_cleanup_dirty_bgs - called on transaction abort
*/
struct list_head io_bgs;
struct list_head dropped_roots;
struct extent_io_tree pinned_extents;
/*
* we need to make sure block group deletion doesn't race with
* free space cache writeout. This mutex keeps them from stomping
* on each other
* We need to make sure block group deletion doesn't race with the
* dirty block group item updates done outside the commit critical
* section. This mutex keeps them from stomping on each other.
*/
struct mutex cache_write_mutex;
struct mutex dirty_bgs_update_mutex;
spinlock_t dirty_bgs_lock;
/* Protected by spin lock fs_info->unused_bgs_lock. */
struct list_head deleted_bgs;
@ -305,7 +287,6 @@ struct btrfs_trans_handle *btrfs_start_transaction_fallback_global_rsv(
struct btrfs_root *root,
unsigned int num_items);
struct btrfs_trans_handle *btrfs_join_transaction(struct btrfs_root *root);
struct btrfs_trans_handle *btrfs_join_transaction_spacecache(struct btrfs_root *root);
struct btrfs_trans_handle *btrfs_join_transaction_nostart(struct btrfs_root *root);
struct btrfs_trans_handle *btrfs_attach_transaction(struct btrfs_root *root);
struct btrfs_trans_handle *btrfs_attach_transaction_barrier(

9
fs/btrfs/zoned.c
View File

@ -804,15 +804,6 @@ int btrfs_check_mountopts_zoned(const struct btrfs_fs_info *info,
if (!btrfs_is_zoned(info))
return 0;
/*
* Space cache writing is not COWed. Disable that to avoid write errors
* in sequential zones.
*/
if (btrfs_raw_test_opt(*mount_opt, SPACE_CACHE)) {
btrfs_err(info, "zoned: space cache v1 is not supported");
return -EINVAL;
}
if (btrfs_raw_test_opt(*mount_opt, NODATACOW)) {
btrfs_err(info, "zoned: NODATACOW not supported");
return -EINVAL;