summaryrefslogtreecommitdiff
path: root/fs/btrfs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/btrfs')
-rw-r--r--fs/btrfs/backref.c41
-rw-r--r--fs/btrfs/backref.h3
-rw-r--r--fs/btrfs/compression.c33
-rw-r--r--fs/btrfs/compression.h4
-rw-r--r--fs/btrfs/ctree.c34
-rw-r--r--fs/btrfs/ctree.h10
-rw-r--r--fs/btrfs/delayed-inode.c8
-rw-r--r--fs/btrfs/disk-io.c21
-rw-r--r--fs/btrfs/extent-tree.c202
-rw-r--r--fs/btrfs/free-space-cache.c20
-rw-r--r--fs/btrfs/inode.c16
-rw-r--r--fs/btrfs/ioctl.c2
-rw-r--r--fs/btrfs/lzo.c15
-rw-r--r--fs/btrfs/scrub.c12
-rw-r--r--fs/btrfs/transaction.c23
-rw-r--r--fs/btrfs/transaction.h2
-rw-r--r--fs/btrfs/tree-log.c11
-rw-r--r--fs/btrfs/volumes.c2
-rw-r--r--fs/btrfs/zlib.c20
19 files changed, 257 insertions, 222 deletions
diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
index 2d3e32ebfd15..f55721ff9385 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -1246,25 +1246,6 @@ int btrfs_check_shared(struct btrfs_trans_handle *trans,
return ret;
}
-/*
- * this makes the path point to (inum INODE_ITEM ioff)
- */
-int inode_item_info(u64 inum, u64 ioff, struct btrfs_root *fs_root,
- struct btrfs_path *path)
-{
- struct btrfs_key key;
- return btrfs_find_item(fs_root, path, inum, ioff,
- BTRFS_INODE_ITEM_KEY, &key);
-}
-
-static int inode_ref_info(u64 inum, u64 ioff, struct btrfs_root *fs_root,
- struct btrfs_path *path,
- struct btrfs_key *found_key)
-{
- return btrfs_find_item(fs_root, path, inum, ioff,
- BTRFS_INODE_REF_KEY, found_key);
-}
-
int btrfs_find_one_extref(struct btrfs_root *root, u64 inode_objectid,
u64 start_off, struct btrfs_path *path,
struct btrfs_inode_extref **ret_extref,
@@ -1374,7 +1355,8 @@ char *btrfs_ref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path,
btrfs_tree_read_unlock_blocking(eb);
free_extent_buffer(eb);
}
- ret = inode_ref_info(parent, 0, fs_root, path, &found_key);
+ ret = btrfs_find_item(fs_root, path, parent, 0,
+ BTRFS_INODE_REF_KEY, &found_key);
if (ret > 0)
ret = -ENOENT;
if (ret)
@@ -1552,7 +1534,6 @@ int tree_backref_for_extent(unsigned long *ptr, struct extent_buffer *eb,
{
int ret;
int type;
- struct btrfs_tree_block_info *info;
struct btrfs_extent_inline_ref *eiref;
if (*ptr == (unsigned long)-1)
@@ -1573,9 +1554,17 @@ int tree_backref_for_extent(unsigned long *ptr, struct extent_buffer *eb,
}
/* we can treat both ref types equally here */
- info = (struct btrfs_tree_block_info *)(ei + 1);
*out_root = btrfs_extent_inline_ref_offset(eb, eiref);
- *out_level = btrfs_tree_block_level(eb, info);
+
+ if (key->type == BTRFS_EXTENT_ITEM_KEY) {
+ struct btrfs_tree_block_info *info;
+
+ info = (struct btrfs_tree_block_info *)(ei + 1);
+ *out_level = btrfs_tree_block_level(eb, info);
+ } else {
+ ASSERT(key->type == BTRFS_METADATA_ITEM_KEY);
+ *out_level = (u8)key->offset;
+ }
if (ret == 1)
*ptr = (unsigned long)-1;
@@ -1720,8 +1709,10 @@ static int iterate_inode_refs(u64 inum, struct btrfs_root *fs_root,
struct btrfs_key found_key;
while (!ret) {
- ret = inode_ref_info(inum, parent ? parent+1 : 0, fs_root, path,
- &found_key);
+ ret = btrfs_find_item(fs_root, path, inum,
+ parent ? parent + 1 : 0, BTRFS_INODE_REF_KEY,
+ &found_key);
+
if (ret < 0)
break;
if (ret) {
diff --git a/fs/btrfs/backref.h b/fs/btrfs/backref.h
index 2a1ac6bfc724..9c41fbac3009 100644
--- a/fs/btrfs/backref.h
+++ b/fs/btrfs/backref.h
@@ -32,9 +32,6 @@ struct inode_fs_paths {
typedef int (iterate_extent_inodes_t)(u64 inum, u64 offset, u64 root,
void *ctx);
-int inode_item_info(u64 inum, u64 ioff, struct btrfs_root *fs_root,
- struct btrfs_path *path);
-
int extent_from_logical(struct btrfs_fs_info *fs_info, u64 logical,
struct btrfs_path *path, struct btrfs_key *found_key,
u64 *flags);
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index 1bf411bc28fd..e9df8862012c 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -1017,8 +1017,6 @@ int btrfs_decompress_buf2page(char *buf, unsigned long buf_start,
bytes = min(bytes, working_bytes);
kaddr = kmap_atomic(page_out);
memcpy(kaddr + *pg_offset, buf + buf_offset, bytes);
- if (*pg_index == (vcnt - 1) && *pg_offset == 0)
- memset(kaddr + bytes, 0, PAGE_CACHE_SIZE - bytes);
kunmap_atomic(kaddr);
flush_dcache_page(page_out);
@@ -1060,3 +1058,34 @@ int btrfs_decompress_buf2page(char *buf, unsigned long buf_start,
return 1;
}
+
+/*
+ * When uncompressing data, we need to make sure and zero any parts of
+ * the biovec that were not filled in by the decompression code. pg_index
+ * and pg_offset indicate the last page and the last offset of that page
+ * that have been filled in. This will zero everything remaining in the
+ * biovec.
+ */
+void btrfs_clear_biovec_end(struct bio_vec *bvec, int vcnt,
+ unsigned long pg_index,
+ unsigned long pg_offset)
+{
+ while (pg_index < vcnt) {
+ struct page *page = bvec[pg_index].bv_page;
+ unsigned long off = bvec[pg_index].bv_offset;
+ unsigned long len = bvec[pg_index].bv_len;
+
+ if (pg_offset < off)
+ pg_offset = off;
+ if (pg_offset < off + len) {
+ unsigned long bytes = off + len - pg_offset;
+ char *kaddr;
+
+ kaddr = kmap_atomic(page);
+ memset(kaddr + pg_offset, 0, bytes);
+ kunmap_atomic(kaddr);
+ }
+ pg_index++;
+ pg_offset = 0;
+ }
+}
diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h
index 0c803b4fbf93..d181f70caae0 100644
--- a/fs/btrfs/compression.h
+++ b/fs/btrfs/compression.h
@@ -45,7 +45,9 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start,
unsigned long nr_pages);
int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
int mirror_num, unsigned long bio_flags);
-
+void btrfs_clear_biovec_end(struct bio_vec *bvec, int vcnt,
+ unsigned long pg_index,
+ unsigned long pg_offset);
struct btrfs_compress_op {
struct list_head *(*alloc_workspace)(void);
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 276d4187cbf0..6b2ec9089229 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -213,11 +213,19 @@ static struct extent_buffer *btrfs_read_lock_root_node(struct btrfs_root *root)
*/
static void add_root_to_dirty_list(struct btrfs_root *root)
{
+ if (test_bit(BTRFS_ROOT_DIRTY, &root->state) ||
+ !test_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state))
+ return;
+
spin_lock(&root->fs_info->trans_lock);
- if (test_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state) &&
- list_empty(&root->dirty_list)) {
- list_add(&root->dirty_list,
- &root->fs_info->dirty_cowonly_roots);
+ if (!test_and_set_bit(BTRFS_ROOT_DIRTY, &root->state)) {
+ /* Want the extent tree to be the last on the list */
+ if (root->objectid == BTRFS_EXTENT_TREE_OBJECTID)
+ list_move_tail(&root->dirty_list,
+ &root->fs_info->dirty_cowonly_roots);
+ else
+ list_move(&root->dirty_list,
+ &root->fs_info->dirty_cowonly_roots);
}
spin_unlock(&root->fs_info->trans_lock);
}
@@ -2606,32 +2614,24 @@ static int key_search(struct extent_buffer *b, struct btrfs_key *key,
return 0;
}
-int btrfs_find_item(struct btrfs_root *fs_root, struct btrfs_path *found_path,
+int btrfs_find_item(struct btrfs_root *fs_root, struct btrfs_path *path,
u64 iobjectid, u64 ioff, u8 key_type,
struct btrfs_key *found_key)
{
int ret;
struct btrfs_key key;
struct extent_buffer *eb;
- struct btrfs_path *path;
+
+ ASSERT(path);
+ ASSERT(found_key);
key.type = key_type;
key.objectid = iobjectid;
key.offset = ioff;
- if (found_path == NULL) {
- path = btrfs_alloc_path();
- if (!path)
- return -ENOMEM;
- } else
- path = found_path;
-
ret = btrfs_search_slot(NULL, fs_root, &key, path, 0, 0);
- if ((ret < 0) || (found_key == NULL)) {
- if (path != found_path)
- btrfs_free_path(path);
+ if (ret < 0)
return ret;
- }
eb = path->nodes[0];
if (ret && path->slots[0] >= btrfs_header_nritems(eb)) {
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index e6fbbd74b716..0b4683f560c8 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -1238,7 +1238,6 @@ enum btrfs_disk_cache_state {
BTRFS_DC_ERROR = 1,
BTRFS_DC_CLEAR = 2,
BTRFS_DC_SETUP = 3,
- BTRFS_DC_NEED_WRITE = 4,
};
struct btrfs_caching_control {
@@ -1276,7 +1275,6 @@ struct btrfs_block_group_cache {
unsigned long full_stripe_len;
unsigned int ro:1;
- unsigned int dirty:1;
unsigned int iref:1;
unsigned int has_caching_ctl:1;
unsigned int removed:1;
@@ -1314,6 +1312,9 @@ struct btrfs_block_group_cache {
struct list_head ro_list;
atomic_t trimming;
+
+ /* For dirty block groups */
+ struct list_head dirty_list;
};
/* delayed seq elem */
@@ -1775,6 +1776,7 @@ struct btrfs_subvolume_writers {
#define BTRFS_ROOT_DEFRAG_RUNNING 6
#define BTRFS_ROOT_FORCE_COW 7
#define BTRFS_ROOT_MULTI_LOG_TASKS 8
+#define BTRFS_ROOT_DIRTY 9
/*
* in ram representation of the tree. extent_root is used for all allocations
@@ -3481,8 +3483,8 @@ void btrfs_put_block_group_cache(struct btrfs_fs_info *info);
u64 btrfs_account_ro_block_groups_free_space(struct btrfs_space_info *sinfo);
int btrfs_error_unpin_extent_range(struct btrfs_root *root,
u64 start, u64 end);
-int btrfs_error_discard_extent(struct btrfs_root *root, u64 bytenr,
- u64 num_bytes, u64 *actual_bytes);
+int btrfs_discard_extent(struct btrfs_root *root, u64 bytenr,
+ u64 num_bytes, u64 *actual_bytes);
int btrfs_force_chunk_alloc(struct btrfs_trans_handle *trans,
struct btrfs_root *root, u64 type);
int btrfs_trim_fs(struct btrfs_root *root, struct fstrim_range *range);
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index 054577bddaf2..de4e70fb3cbb 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -1857,6 +1857,14 @@ int btrfs_delayed_delete_inode_ref(struct inode *inode)
{
struct btrfs_delayed_node *delayed_node;
+ /*
+ * we don't do delayed inode updates during log recovery because it
+ * leads to enospc problems. This means we also can't do
+ * delayed inode refs
+ */
+ if (BTRFS_I(inode)->root->fs_info->log_root_recovering)
+ return -EAGAIN;
+
delayed_node = btrfs_get_or_create_delayed_node(inode);
if (IS_ERR(delayed_node))
return PTR_ERR(delayed_node);
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 9c204533fd22..65384de10375 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -1629,6 +1629,8 @@ struct btrfs_root *btrfs_get_fs_root(struct btrfs_fs_info *fs_info,
bool check_ref)
{
struct btrfs_root *root;
+ struct btrfs_path *path;
+ struct btrfs_key key;
int ret;
if (location->objectid == BTRFS_ROOT_TREE_OBJECTID)
@@ -1668,8 +1670,17 @@ again:
if (ret)
goto fail;
- ret = btrfs_find_item(fs_info->tree_root, NULL, BTRFS_ORPHAN_OBJECTID,
- location->objectid, BTRFS_ORPHAN_ITEM_KEY, NULL);
+ path = btrfs_alloc_path();
+ if (!path) {
+ ret = -ENOMEM;
+ goto fail;
+ }
+ key.objectid = BTRFS_ORPHAN_OBJECTID;
+ key.type = BTRFS_ORPHAN_ITEM_KEY;
+ key.offset = location->objectid;
+
+ ret = btrfs_search_slot(NULL, fs_info->tree_root, &key, path, 0, 0);
+ btrfs_free_path(path);
if (ret < 0)
goto fail;
if (ret == 0)
@@ -4120,12 +4131,6 @@ again:
if (ret)
break;
- /* opt_discard */
- if (btrfs_test_opt(root, DISCARD))
- ret = btrfs_error_discard_extent(root, start,
- end + 1 - start,
- NULL);
-
clear_extent_dirty(unpin, start, end, GFP_NOFS);
btrfs_error_unpin_extent_range(root, start, end);
cond_resched();
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 8ff31f81d870..1c591d6eae58 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -74,8 +74,9 @@ enum {
RESERVE_ALLOC_NO_ACCOUNT = 2,
};
-static int update_block_group(struct btrfs_root *root,
- u64 bytenr, u64 num_bytes, int alloc);
+static int update_block_group(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root, u64 bytenr,
+ u64 num_bytes, int alloc);
static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
u64 bytenr, u64 num_bytes, u64 parent,
@@ -1889,8 +1890,8 @@ static int btrfs_issue_discard(struct block_device *bdev,
return blkdev_issue_discard(bdev, start >> 9, len >> 9, GFP_NOFS, 0);
}
-static int btrfs_discard_extent(struct btrfs_root *root, u64 bytenr,
- u64 num_bytes, u64 *actual_bytes)
+int btrfs_discard_extent(struct btrfs_root *root, u64 bytenr,
+ u64 num_bytes, u64 *actual_bytes)
{
int ret;
u64 discarded_bytes = 0;
@@ -3139,9 +3140,11 @@ static int write_one_cache_group(struct btrfs_trans_handle *trans,
struct extent_buffer *leaf;
ret = btrfs_search_slot(trans, extent_root, &cache->key, path, 0, 1);
- if (ret < 0)
+ if (ret) {
+ if (ret > 0)
+ ret = -ENOENT;
goto fail;
- BUG_ON(ret); /* Corruption */
+ }
leaf = path->nodes[0];
bi = btrfs_item_ptr_offset(leaf, path->slots[0]);
@@ -3149,11 +3152,9 @@ static int write_one_cache_group(struct btrfs_trans_handle *trans,
btrfs_mark_buffer_dirty(leaf);
btrfs_release_path(path);
fail:
- if (ret) {
+ if (ret)
btrfs_abort_transaction(trans, root, ret);
- return ret;
- }
- return 0;
+ return ret;
}
@@ -3315,120 +3316,42 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans,
struct btrfs_root *root)
{
struct btrfs_block_group_cache *cache;
- int err = 0;
+ struct btrfs_transaction *cur_trans = trans->transaction;
+ int ret = 0;
struct btrfs_path *path;
- u64 last = 0;
+
+ if (list_empty(&cur_trans->dirty_bgs))
+ return 0;
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
-again:
- while (1) {
- cache = btrfs_lookup_first_block_group(root->fs_info, last);
- while (cache) {
- if (cache->disk_cache_state == BTRFS_DC_CLEAR)
- break;
- cache = next_block_group(root, cache);
- }
- if (!cache) {
- if (last == 0)
- break;
- last = 0;
- continue;
- }
- err = cache_save_setup(cache, trans, path);
- last = cache->key.objectid + cache->key.offset;
- btrfs_put_block_group(cache);
- }
-
- while (1) {
- if (last == 0) {
- err = btrfs_run_delayed_refs(trans, root,
- (unsigned long)-1);
- if (err) /* File system offline */
- goto out;
- }
-
- cache = btrfs_lookup_first_block_group(root->fs_info, last);
- while (cache) {
- if (cache->disk_cache_state == BTRFS_DC_CLEAR) {
- btrfs_put_block_group(cache);
- goto again;
- }
-
- if (cache->dirty)
- break;
- cache = next_block_group(root, cache);
- }
- if (!cache) {
- if (last == 0)
- break;
- last = 0;
- continue;
- }
-
- if (cache->disk_cache_state == BTRFS_DC_SETUP)
- cache->disk_cache_state = BTRFS_DC_NEED_WRITE;
- cache->dirty = 0;
- last = cache->key.objectid + cache->key.offset;
-
- err = write_one_cache_group(trans, root, path, cache);
- btrfs_put_block_group(cache);
- if (err) /* File system offline */
- goto out;
- }
-
- while (1) {
- /*
- * I don't think this is needed since we're just marking our
- * preallocated extent as written, but just in case it can't
- * hurt.
- */
- if (last == 0) {
- err = btrfs_run_delayed_refs(trans, root,
- (unsigned long)-1);
- if (err) /* File system offline */
- goto out;
- }
-
- cache = btrfs_lookup_first_block_group(root->fs_info, last);
- while (cache) {
- /*
- * Really this shouldn't happen, but it could if we
- * couldn't write the entire preallocated extent and
- * splitting the extent resulted in a new block.
- */
- if (cache->dirty) {
- btrfs_put_block_group(cache);
- goto again;
- }
- if (cache->disk_cache_state == BTRFS_DC_NEED_WRITE)
- break;
- cache = next_block_group(root, cache);
- }
- if (!cache) {
- if (last == 0)
- break;
- last = 0;
- continue;
- }
-
- err = btrfs_write_out_cache(root, trans, cache, path);
-
- /*
- * If we didn't have an error then the cache state is still
- * NEED_WRITE, so we can set it to WRITTEN.
- */
- if (!err && cache->disk_cache_state == BTRFS_DC_NEED_WRITE)
- cache->disk_cache_state = BTRFS_DC_WRITTEN;
- last = cache->key.objectid + cache->key.offset;
+ /*
+ * We don't need the lock here since we are protected by the transaction
+ * commit. 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.
+ */
+ while (!list_empty(&cur_trans->dirty_bgs)) {
+ cache = list_first_entry(&cur_trans->dirty_bgs,
+ struct btrfs_block_group_cache,
+ dirty_list);
+ list_del_init(&cache->dirty_list);
+ if (cache->disk_cache_state == BTRFS_DC_CLEAR)
+ cache_save_setup(cache, trans, path);
+ if (!ret)
+ ret = btrfs_run_delayed_refs(trans, root,
+ (unsigned long) -1);
+ if (!ret && cache->disk_cache_state == BTRFS_DC_SETUP)
+ btrfs_write_out_cache(root, trans, cache, path);
+ if (!ret)
+ ret = write_one_cache_group(trans, root, path, cache);
btrfs_put_block_group(cache);
}
-out:
btrfs_free_path(path);
- return err;
+ return ret;
}
int btrfs_extent_readonly(struct btrfs_root *root, u64 bytenr)
@@ -5375,8 +5298,9 @@ void btrfs_delalloc_release_space(struct inode *inode, u64 num_bytes)
btrfs_free_reserved_data_space(inode, num_bytes);
}
-static int update_block_group(struct btrfs_root *root,
- u64 bytenr, u64 num_bytes, int alloc)
+static int update_block_group(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root, u64 bytenr,
+ u64 num_bytes, int alloc)
{
struct btrfs_block_group_cache *cache = NULL;
struct btrfs_fs_info *info = root->fs_info;
@@ -5414,6 +5338,14 @@ static int update_block_group(struct btrfs_root *root,
if (!alloc && cache->cached == BTRFS_CACHE_NO)
cache_block_group(cache, 1);
+ spin_lock(&trans->transaction->dirty_bgs_lock);
+ if (list_empty(&cache->dirty_list)) {
+ list_add_tail(&cache->dirty_list,
+ &trans->transaction->dirty_bgs);
+ btrfs_get_block_group(cache);
+ }
+ spin_unlock(&trans->transaction->dirty_bgs_lock);
+
byte_in_group = bytenr - cache->key.objectid;
WARN_ON(byte_in_group > cache->key.offset);
@@ -5424,7 +5356,6 @@ static int update_block_group(struct btrfs_root *root,
cache->disk_cache_state < BTRFS_DC_CLEAR)
cache->disk_cache_state = BTRFS_DC_CLEAR;
- cache->dirty = 1;
old_val = btrfs_block_group_used(&cache->item);
num_bytes = min(total, cache->key.offset - byte_in_group);
if (alloc) {
@@ -5727,7 +5658,8 @@ void btrfs_prepare_extent_commit(struct btrfs_trans_handle *trans,
update_global_block_rsv(fs_info);
}
-static int unpin_extent_range(struct btrfs_root *root, u64 start, u64 end)
+static int unpin_extent_range(struct btrfs_root *root, u64 start, u64 end,
+ const bool return_free_space)
{
struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_block_group_cache *cache = NULL;
@@ -5751,7 +5683,8 @@ static int unpin_extent_range(struct btrfs_root *root, u64 start, u64 end)
if (start < cache->last_byte_to_unpin) {
len = min(len, cache->last_byte_to_unpin - start);
- btrfs_add_free_space(cache, start, len);
+ if (return_free_space)
+ btrfs_add_free_space(cache, start, len);
}
start += len;
@@ -5815,7 +5748,7 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans,
end + 1 - start, NULL);
clear_extent_dirty(unpin, start, end, GFP_NOFS);
- unpin_extent_range(root, start, end);
+ unpin_extent_range(root, start, end, true);
cond_resched();
}
@@ -6101,7 +6034,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
}
}
- ret = update_block_group(root, bytenr, num_bytes, 0);
+ ret = update_block_group(trans, root, bytenr, num_bytes, 0);
if (ret) {
btrfs_abort_transaction(trans, extent_root, ret);
goto out;
@@ -7061,7 +6994,7 @@ static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans,
if (ret)
return ret;
- ret = update_block_group(root, ins->objectid, ins->offset, 1);
+ ret = update_block_group(trans, root, 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);
@@ -7150,7 +7083,8 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans,
return ret;
}
- ret = update_block_group(root, ins->objectid, root->nodesize, 1);
+ ret = update_block_group(trans, root, ins->objectid, root->nodesize,
+ 1);
if (ret) { /* -ENOENT, logic error */
btrfs_err(fs_info, "update block group failed for %llu %llu",
ins->objectid, ins->offset);
@@ -8871,6 +8805,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
cache_node);
rb_erase(&block_group->cache_node,
&info->block_group_cache_tree);
+ RB_CLEAR_NODE(&block_group->cache_node);
spin_unlock(&info->block_group_cache_lock);
down_write(&block_group->space_info->groups_sem);
@@ -9001,6 +8936,7 @@ btrfs_create_block_group_cache(struct btrfs_root *root, u64 start, u64 size)
INIT_LIST_HEAD(&cache->cluster_list);
INIT_LIST_HEAD(&cache->bg_list);
INIT_LIST_HEAD(&cache->ro_list);
+ INIT_LIST_HEAD(&cache->dirty_list);
btrfs_init_free_space_ctl(cache);
atomic_set(&cache->trimming, 0);
@@ -9064,9 +9000,8 @@ int btrfs_read_block_groups(struct btrfs_root *root)
* b) Setting 'dirty flag' makes sure that we flush
* the new space cache info onto disk.
*/
- cache->disk_cache_state = BTRFS_DC_CLEAR;
if (btrfs_test_opt(root, SPACE_CACHE))
- cache->dirty = 1;
+ cache->disk_cache_state = BTRFS_DC_CLEAR;
}
read_extent_buffer(leaf, &cache->item,
@@ -9129,6 +9064,7 @@ int btrfs_read_block_groups(struct btrfs_root *root)
spin_lock(&info->block_group_cache_lock);
rb_erase(&cache->cache_node,
&info->block_group_cache_tree);
+ RB_CLEAR_NODE(&cache->cache_node);
spin_unlock(&info->block_group_cache_lock);
btrfs_put_block_group(cache);
goto error;
@@ -9270,6 +9206,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
spin_lock(&root->fs_info->block_group_cache_lock);
rb_erase(&cache->cache_node,
&root->fs_info->block_group_cache_tree);
+ RB_CLEAR_NODE(&cache->cache_node);
spin_unlock(&root->fs_info->block_group_cache_lock);
btrfs_put_block_group(cache);
return ret;
@@ -9455,6 +9392,13 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
}
}
+ spin_lock(&trans->transaction->dirty_bgs_lock);
+ if (!list_empty(&block_group->dirty_list)) {
+ list_del_init(&block_group->dirty_list);
+ btrfs_put_block_group(block_group);
+ }
+ spin_unlock(&trans->transaction->dirty_bgs_lock);
+
btrfs_remove_free_space_cache(block_group);
spin_lock(&block_group->space_info->lock);
@@ -9689,13 +9633,7 @@ out:
int btrfs_error_unpin_extent_range(struct btrfs_root *root, u64 start, u64 end)
{
- return unpin_extent_range(root, start, end);
-}
-
-int btrfs_error_discard_extent(struct btrfs_root *root, u64 bytenr,
- u64 num_bytes, u64 *actual_bytes)
-{
- return btrfs_discard_extent(root, bytenr, num_bytes, actual_bytes);
+ return unpin_extent_range(root, start, end, false);
}
int btrfs_trim_fs(struct btrfs_root *root, struct fstrim_range *range)
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index 030847bf7cec..80a3141463e7 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -1243,6 +1243,7 @@ int btrfs_write_out_cache(struct btrfs_root *root,
struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
struct inode *inode;
int ret = 0;
+ enum btrfs_disk_cache_state dcs = BTRFS_DC_WRITTEN;
root = root->fs_info->tree_root;
@@ -1266,9 +1267,7 @@ int btrfs_write_out_cache(struct btrfs_root *root,
ret = __btrfs_write_out_cache(root, inode, ctl, block_group, trans,
path, block_group->key.objectid);
if (ret) {
- spin_lock(&block_group->lock);
- block_group->disk_cache_state = BTRFS_DC_ERROR;
- spin_unlock(&block_group->lock);
+ dcs = BTRFS_DC_ERROR;
ret = 0;
#ifdef DEBUG
btrfs_err(root->fs_info,
@@ -1277,6 +1276,9 @@ int btrfs_write_out_cache(struct btrfs_root *root,
#endif
}
+ spin_lock(&block_group->lock);
+ block_group->disk_cache_state = dcs;
+ spin_unlock(&block_group->lock);
iput(inode);
return ret;
}
@@ -2966,8 +2968,8 @@ static int do_trimming(struct btrfs_block_group_cache *block_group,
spin_unlock(&block_group->lock);
spin_unlock(&space_info->lock);
- ret = btrfs_error_discard_extent(fs_info->extent_root,
- start, bytes, &trimmed);
+ ret = btrfs_discard_extent(fs_info->extent_root,
+ start, bytes, &trimmed);
if (!ret)
*total_trimmed += trimmed;
@@ -3185,16 +3187,18 @@ out:
spin_unlock(&block_group->lock);
+ lock_chunks(block_group->fs_info->chunk_root);
em_tree = &block_group->fs_info->mapping_tree.map_tree;
write_lock(&em_tree->lock);
em = lookup_extent_mapping(em_tree, block_group->key.objectid,
1);
BUG_ON(!em); /* logic error, can't happen */
+ /*
+ * remove_extent_mapping() will delete us from the pinned_chunks
+ * list, which is protected by the chunk mutex.
+ */
remove_extent_mapping(em_tree, em);
write_unlock(&em_tree->lock);
-
- lock_chunks(block_group->fs_info->chunk_root);
- list_del_init(&em->list);
unlock_chunks(block_group->fs_info->chunk_root);
/* once for us and once for the tree */
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 8de23355f6cf..370f23416080 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5009,6 +5009,7 @@ static int fixup_tree_root_location(struct btrfs_root *root,
struct btrfs_root *new_root;
struct btrfs_root_ref *ref;
struct extent_buffer *leaf;
+ struct btrfs_key key;
int ret;
int err = 0;
@@ -5019,9 +5020,12 @@ static int fixup_tree_root_location(struct btrfs_root *root,
}
err = -ENOENT;
- ret = btrfs_find_item(root->fs_info->tree_root, path,
- BTRFS_I(dir)->root->root_key.objectid,
- location->objectid, BTRFS_ROOT_REF_KEY, NULL);
+ key.objectid = BTRFS_I(dir)->root->root_key.objectid;
+ key.type = BTRFS_ROOT_REF_KEY;
+ key.offset = location->objectid;
+
+ ret = btrfs_search_slot(NULL, root->fs_info->tree_root, &key, path,
+ 0, 0);
if (ret) {
if (ret < 0)
err = ret;
@@ -5355,7 +5359,7 @@ static struct dentry *btrfs_lookup(struct inode *dir, struct dentry *dentry,
return ERR_CAST(inode);
}
- return d_materialise_unique(dentry, inode);
+ return d_splice_alias(inode, dentry);
}
unsigned char btrfs_filetype_table[] = {
@@ -6255,8 +6259,10 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
out_fail:
btrfs_end_transaction(trans, root);
- if (drop_on_err)
+ if (drop_on_err) {
+ inode_dec_link_count(inode);
iput(inode);
+ }
btrfs_balance_delayed_items(root);
btrfs_btree_balance_dirty(root);
return err;
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index b590e23fa03e..d49fe8a0f6b5 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -5268,7 +5268,7 @@ long btrfs_ioctl(struct file *file, unsigned int
ret = btrfs_start_delalloc_roots(root->fs_info, 0, -1);
if (ret)
return ret;
- ret = btrfs_sync_fs(file->f_dentry->d_sb, 1);
+ ret = btrfs_sync_fs(file_inode(file)->i_sb, 1);
/*
* The transaction thread may want to do more work,
* namely it pokes the cleaner ktread that will start
diff --git a/fs/btrfs/lzo.c b/fs/btrfs/lzo.c
index 78285f30909e..617553cdb7d3 100644
--- a/fs/btrfs/lzo.c
+++ b/fs/btrfs/lzo.c
@@ -373,6 +373,8 @@ cont:
}
done:
kunmap(pages_in[page_in_index]);
+ if (!ret)
+ btrfs_clear_biovec_end(bvec, vcnt, page_out_index, pg_offset);
return ret;
}
@@ -410,10 +412,23 @@ static int lzo_decompress(struct list_head *ws, unsigned char *data_in,
goto out;
}
+ /*
+ * the caller is already checking against PAGE_SIZE, but lets
+ * move this check closer to the memcpy/memset
+ */
+ destlen = min_t(unsigned long, destlen, PAGE_SIZE);
bytes = min_t(unsigned long, destlen, out_len - start_byte);
kaddr = kmap_atomic(dest_page);
memcpy(kaddr, workspace->buf + start_byte, bytes);
+
+ /*
+ * btrfs_getblock is doing a zero on the tail of the page too,
+ * but this will cover anything missing from the decompressed
+ * data.
+ */
+ if (bytes < destlen)
+ memset(kaddr+bytes, 0, destlen-bytes);
kunmap_atomic(kaddr);
out:
return ret;
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index f2bb13a23f86..53575a45f7d1 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -520,6 +520,7 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root,
struct inode_fs_paths *ipath = NULL;
struct btrfs_root *local_root;
struct btrfs_key root_key;
+ struct btrfs_key key;
root_key.objectid = root;
root_key.type = BTRFS_ROOT_ITEM_KEY;
@@ -530,7 +531,14 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root,
goto err;
}
- ret = inode_item_info(inum, 0, local_root, swarn->path);
+ /*
+ * this makes the path point to (inum INODE_ITEM ioff)
+ */
+ key.objectid = inum;
+ key.type = BTRFS_INODE_ITEM_KEY;
+ key.offset = 0;
+
+ ret = btrfs_search_slot(NULL, local_root, &key, swarn->path, 0, 0);
if (ret) {
btrfs_release_path(swarn->path);
goto err;
@@ -2607,9 +2615,9 @@ static int scrub_extent_for_parity(struct scrub_parity *sparity,
ret = scrub_pages_for_parity(sparity, logical, l, physical, dev,
flags, gen, mirror_num,
have_csum ? csum : NULL);
-skip:
if (ret)
return ret;
+skip:
len -= l;
logical += l;
physical += l;
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index a605d4e2f2bc..e0faf803513a 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -248,6 +248,8 @@ loop:
INIT_LIST_HEAD(&cur_trans->pending_chunks);
INIT_LIST_HEAD(&cur_trans->switch_commits);
INIT_LIST_HEAD(&cur_trans->pending_ordered);
+ INIT_LIST_HEAD(&cur_trans->dirty_bgs);
+ spin_lock_init(&cur_trans->dirty_bgs_lock);
list_add_tail(&cur_trans->list, &fs_info->trans_list);
extent_io_tree_init(&cur_trans->dirty_pages,
fs_info->btree_inode->i_mapping);
@@ -1020,6 +1022,7 @@ static int update_cowonly_root(struct btrfs_trans_handle *trans,
u64 old_root_bytenr;
u64 old_root_used;
struct btrfs_root *tree_root = root->fs_info->tree_root;
+ bool extent_root = (root->objectid == BTRFS_EXTENT_TREE_OBJECTID);
old_root_used = btrfs_root_used(&root->root_item);
btrfs_write_dirty_block_groups(trans, root);
@@ -1027,7 +1030,9 @@ static int update_cowonly_root(struct btrfs_trans_handle *trans,
while (1) {
old_root_bytenr = btrfs_root_bytenr(&root->root_item);
if (old_root_bytenr == root->node->start &&
- old_root_used == btrfs_root_used(&root->root_item))
+ old_root_used == btrfs_root_used(&root->root_item) &&
+ (!extent_root ||
+ list_empty(&trans->transaction->dirty_bgs)))
break;
btrfs_set_root_node(&root->root_item, root->node);
@@ -1038,7 +1043,15 @@ static int update_cowonly_root(struct btrfs_trans_handle *trans,
return ret;
old_root_used = btrfs_root_used(&root->root_item);
- ret = btrfs_write_dirty_block_groups(trans, root);
+ if (extent_root) {
+ ret = btrfs_write_dirty_block_groups(trans, root);
+ if (ret)
+ return ret;
+ }
+ ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
+ if (ret)
+ return ret;
+ ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
if (ret)
return ret;
}
@@ -1061,10 +1074,6 @@ static noinline int commit_cowonly_roots(struct btrfs_trans_handle *trans,
struct extent_buffer *eb;
int ret;
- ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
- if (ret)
- return ret;
-
eb = btrfs_lock_root_node(fs_info->tree_root);
ret = btrfs_cow_block(trans, fs_info->tree_root, eb, NULL,
0, &eb);
@@ -1097,6 +1106,7 @@ static noinline int commit_cowonly_roots(struct btrfs_trans_handle *trans,
next = fs_info->dirty_cowonly_roots.next;
list_del_init(next);
root = list_entry(next, struct btrfs_root, dirty_list);
+ clear_bit(BTRFS_ROOT_DIRTY, &root->state);
if (root != fs_info->extent_root)
list_add_tail(&root->dirty_list,
@@ -1983,6 +1993,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
switch_commit_roots(cur_trans, root->fs_info);
assert_qgroups_uptodate(trans);
+ ASSERT(list_empty(&cur_trans->dirty_bgs));
update_super_roots(root);
btrfs_set_super_log_root(root->fs_info->super_copy, 0);
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index 00ed29c4b3f9..3305451451ca 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -58,6 +58,8 @@ struct btrfs_transaction {
struct list_head pending_chunks;
struct list_head pending_ordered;
struct list_head switch_commits;
+ struct list_head dirty_bgs;
+ spinlock_t dirty_bgs_lock;
struct btrfs_delayed_ref_root delayed_refs;
int aborted;
};
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 4a42edc224a8..67e5bf709dca 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -1254,13 +1254,14 @@ out:
}
static int insert_orphan_item(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, u64 offset)
+ struct btrfs_root *root, u64 ino)
{
int ret;
- ret = btrfs_find_item(root, NULL, BTRFS_ORPHAN_OBJECTID,
- offset, BTRFS_ORPHAN_ITEM_KEY, NULL);
- if (ret > 0)
- ret = btrfs_insert_orphan_item(trans, root, offset);
+
+ ret = btrfs_insert_orphan_item(trans, root, ino);
+ if (ret == -EEXIST)
+ ret = 0;
+
return ret;
}
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index f0af9cd0814c..0d9bfebdaa4c 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -1485,7 +1485,7 @@ static void update_dev_time(char *path_name)
struct file *filp;
filp = filp_open(path_name, O_RDWR, 0);
- if (!filp)
+ if (IS_ERR(filp))
return;
file_update_time(filp);
filp_close(filp, NULL);
diff --git a/fs/btrfs/zlib.c b/fs/btrfs/zlib.c
index 759fa4e2de8f..fb22fd8d8fb8 100644
--- a/fs/btrfs/zlib.c
+++ b/fs/btrfs/zlib.c
@@ -299,6 +299,8 @@ done:
zlib_inflateEnd(&workspace->strm);
if (data_in)
kunmap(pages_in[page_in_index]);
+ if (!ret)
+ btrfs_clear_biovec_end(bvec, vcnt, page_out_index, pg_offset);
return ret;
}
@@ -310,10 +312,14 @@ static int zlib_decompress(struct list_head *ws, unsigned char *data_in,
struct workspace *workspace = list_entry(ws, struct workspace, list);
int ret = 0;
int wbits = MAX_WBITS;
- unsigned long bytes_left = destlen;
+ unsigned long bytes_left;
unsigned long total_out = 0;
+ unsigned long pg_offset = 0;
char *kaddr;
+ destlen = min_t(unsigned long, destlen, PAGE_SIZE);
+ bytes_left = destlen;
+
workspace->strm.next_in = data_in;
workspace->strm.avail_in = srclen;
workspace->strm.total_in = 0;
@@ -341,7 +347,6 @@ static int zlib_decompress(struct list_head *ws, unsigned char *data_in,
unsigned long buf_start;
unsigned long buf_offset;
unsigned long bytes;
- unsigned long pg_offset = 0;
ret = zlib_inflate(&workspace->strm, Z_NO_FLUSH);
if (ret != Z_OK && ret != Z_STREAM_END)
@@ -384,6 +389,17 @@ next:
ret = 0;
zlib_inflateEnd(&workspace->strm);
+
+ /*
+ * this should only happen if zlib returned fewer bytes than we
+ * expected. btrfs_get_block is responsible for zeroing from the
+ * end of the inline extent (destlen) to the end of the page
+ */
+ if (pg_offset < destlen) {
+ kaddr = kmap_atomic(dest_page);
+ memset(kaddr + pg_offset, 0, destlen - pg_offset);
+ kunmap_atomic(kaddr);
+ }
return ret;
}