summaryrefslogtreecommitdiffstats
path: root/fs/btrfs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/btrfs')
-rw-r--r--fs/btrfs/backref.c138
-rw-r--r--fs/btrfs/backref.h1
-rw-r--r--fs/btrfs/block-group.c2
-rw-r--r--fs/btrfs/compression.c14
-rw-r--r--fs/btrfs/ctree.c52
-rw-r--r--fs/btrfs/ctree.h16
-rw-r--r--fs/btrfs/disk-io.c14
-rw-r--r--fs/btrfs/disk-io.h2
-rw-r--r--fs/btrfs/export.c2
-rw-r--r--fs/btrfs/export.h2
-rw-r--r--fs/btrfs/extent-io-tree.c15
-rw-r--r--fs/btrfs/extent-tree.c25
-rw-r--r--fs/btrfs/file.c29
-rw-r--r--fs/btrfs/inode.c16
-rw-r--r--fs/btrfs/ioctl.c27
-rw-r--r--fs/btrfs/qgroup.c9
-rw-r--r--fs/btrfs/raid56.c18
-rw-r--r--fs/btrfs/scrub.c9
-rw-r--r--fs/btrfs/send.c53
-rw-r--r--fs/btrfs/send.h5
-rw-r--r--fs/btrfs/super.c16
-rw-r--r--fs/btrfs/sysfs.c7
-rw-r--r--fs/btrfs/tests/btrfs-tests.c2
-rw-r--r--fs/btrfs/tests/qgroup-tests.c36
-rw-r--r--fs/btrfs/tree-log.c59
-rw-r--r--fs/btrfs/volumes.c51
-rw-r--r--fs/btrfs/volumes.h4
-rw-r--r--fs/btrfs/zoned.c49
-rw-r--r--fs/btrfs/zoned.h11
29 files changed, 502 insertions, 182 deletions
diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
index dce3a16996b9..18374a6d05bd 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -138,6 +138,7 @@ struct share_check {
u64 root_objectid;
u64 inum;
int share_count;
+ bool have_delayed_delete_refs;
};
static inline int extent_is_shared(struct share_check *sc)
@@ -288,8 +289,10 @@ static void prelim_release(struct preftree *preftree)
struct prelim_ref *ref, *next_ref;
rbtree_postorder_for_each_entry_safe(ref, next_ref,
- &preftree->root.rb_root, rbnode)
+ &preftree->root.rb_root, rbnode) {
+ free_inode_elem_list(ref->inode_list);
free_pref(ref);
+ }
preftree->root = RB_ROOT_CACHED;
preftree->count = 0;
@@ -647,6 +650,18 @@ unode_aux_to_inode_list(struct ulist_node *node)
return (struct extent_inode_elem *)(uintptr_t)node->aux;
}
+static void free_leaf_list(struct ulist *ulist)
+{
+ struct ulist_node *node;
+ struct ulist_iterator uiter;
+
+ ULIST_ITER_INIT(&uiter);
+ while ((node = ulist_next(ulist, &uiter)))
+ free_inode_elem_list(unode_aux_to_inode_list(node));
+
+ ulist_free(ulist);
+}
+
/*
* We maintain three separate rbtrees: one for direct refs, one for
* indirect refs which have a key, and one for indirect refs which do not
@@ -761,7 +776,11 @@ static int resolve_indirect_refs(struct btrfs_fs_info *fs_info,
cond_resched();
}
out:
- ulist_free(parents);
+ /*
+ * We may have inode lists attached to refs in the parents ulist, so we
+ * must free them before freeing the ulist and its refs.
+ */
+ free_leaf_list(parents);
return ret;
}
@@ -820,16 +839,11 @@ static int add_delayed_refs(const struct btrfs_fs_info *fs_info,
struct preftrees *preftrees, struct share_check *sc)
{
struct btrfs_delayed_ref_node *node;
- struct btrfs_delayed_extent_op *extent_op = head->extent_op;
struct btrfs_key key;
- struct btrfs_key tmp_op_key;
struct rb_node *n;
int count;
int ret = 0;
- if (extent_op && extent_op->update_key)
- btrfs_disk_key_to_cpu(&tmp_op_key, &extent_op->key);
-
spin_lock(&head->lock);
for (n = rb_first_cached(&head->ref_tree); n; n = rb_next(n)) {
node = rb_entry(n, struct btrfs_delayed_ref_node,
@@ -855,10 +869,16 @@ static int add_delayed_refs(const struct btrfs_fs_info *fs_info,
case BTRFS_TREE_BLOCK_REF_KEY: {
/* NORMAL INDIRECT METADATA backref */
struct btrfs_delayed_tree_ref *ref;
+ struct btrfs_key *key_ptr = NULL;
+
+ if (head->extent_op && head->extent_op->update_key) {
+ btrfs_disk_key_to_cpu(&key, &head->extent_op->key);
+ key_ptr = &key;
+ }
ref = btrfs_delayed_node_to_tree_ref(node);
ret = add_indirect_ref(fs_info, preftrees, ref->root,
- &tmp_op_key, ref->level + 1,
+ key_ptr, ref->level + 1,
node->bytenr, count, sc,
GFP_ATOMIC);
break;
@@ -884,13 +904,22 @@ static int add_delayed_refs(const struct btrfs_fs_info *fs_info,
key.offset = ref->offset;
/*
- * Found a inum that doesn't match our known inum, we
- * know it's shared.
+ * If we have a share check context and a reference for
+ * another inode, we can't exit immediately. This is
+ * because even if this is a BTRFS_ADD_DELAYED_REF
+ * reference we may find next a BTRFS_DROP_DELAYED_REF
+ * which cancels out this ADD reference.
+ *
+ * If this is a DROP reference and there was no previous
+ * ADD reference, then we need to signal that when we
+ * process references from the extent tree (through
+ * add_inline_refs() and add_keyed_refs()), we should
+ * not exit early if we find a reference for another
+ * inode, because one of the delayed DROP references
+ * may cancel that reference in the extent tree.
*/
- if (sc && sc->inum && ref->objectid != sc->inum) {
- ret = BACKREF_FOUND_SHARED;
- goto out;
- }
+ if (sc && count < 0)
+ sc->have_delayed_delete_refs = true;
ret = add_indirect_ref(fs_info, preftrees, ref->root,
&key, 0, node->bytenr, count, sc,
@@ -920,7 +949,7 @@ static int add_delayed_refs(const struct btrfs_fs_info *fs_info,
}
if (!ret)
ret = extent_is_shared(sc);
-out:
+
spin_unlock(&head->lock);
return ret;
}
@@ -1023,7 +1052,8 @@ static int add_inline_refs(const struct btrfs_fs_info *fs_info,
key.type = BTRFS_EXTENT_DATA_KEY;
key.offset = btrfs_extent_data_ref_offset(leaf, dref);
- if (sc && sc->inum && key.objectid != sc->inum) {
+ if (sc && sc->inum && key.objectid != sc->inum &&
+ !sc->have_delayed_delete_refs) {
ret = BACKREF_FOUND_SHARED;
break;
}
@@ -1033,6 +1063,7 @@ static int add_inline_refs(const struct btrfs_fs_info *fs_info,
ret = add_indirect_ref(fs_info, preftrees, root,
&key, 0, bytenr, count,
sc, GFP_NOFS);
+
break;
}
default:
@@ -1122,7 +1153,8 @@ static int add_keyed_refs(struct btrfs_root *extent_root,
key.type = BTRFS_EXTENT_DATA_KEY;
key.offset = btrfs_extent_data_ref_offset(leaf, dref);
- if (sc && sc->inum && key.objectid != sc->inum) {
+ if (sc && sc->inum && key.objectid != sc->inum &&
+ !sc->have_delayed_delete_refs) {
ret = BACKREF_FOUND_SHARED;
break;
}
@@ -1354,6 +1386,12 @@ again:
if (ret < 0)
goto out;
ref->inode_list = eie;
+ /*
+ * We transferred the list ownership to the ref,
+ * so set to NULL to avoid a double free in case
+ * an error happens after this.
+ */
+ eie = NULL;
}
ret = ulist_add_merge_ptr(refs, ref->parent,
ref->inode_list,
@@ -1379,6 +1417,14 @@ again:
eie->next = ref->inode_list;
}
eie = NULL;
+ /*
+ * We have transferred the inode list ownership from
+ * this ref to the ref we added to the 'refs' ulist.
+ * So set this ref's inode list to NULL to avoid
+ * use-after-free when our caller uses it or double
+ * frees in case an error happens before we return.
+ */
+ ref->inode_list = NULL;
}
cond_resched();
}
@@ -1395,24 +1441,6 @@ out:
return ret;
}
-static void free_leaf_list(struct ulist *blocks)
-{
- struct ulist_node *node = NULL;
- struct extent_inode_elem *eie;
- struct ulist_iterator uiter;
-
- ULIST_ITER_INIT(&uiter);
- while ((node = ulist_next(blocks, &uiter))) {
- if (!node->aux)
- continue;
- eie = unode_aux_to_inode_list(node);
- free_inode_elem_list(eie);
- node->aux = 0;
- }
-
- ulist_free(blocks);
-}
-
/*
* Finds all leafs with a reference to the specified combination of bytenr and
* offset. key_list_head will point to a list of corresponding keys (caller must
@@ -1522,6 +1550,9 @@ static bool lookup_backref_shared_cache(struct btrfs_backref_shared_cache *cache
{
struct btrfs_backref_shared_cache_entry *entry;
+ if (!cache->use_cache)
+ return false;
+
if (WARN_ON_ONCE(level >= BTRFS_MAX_LEVEL))
return false;
@@ -1557,6 +1588,19 @@ static bool lookup_backref_shared_cache(struct btrfs_backref_shared_cache *cache
return false;
*is_shared = entry->is_shared;
+ /*
+ * If the node at this level is shared, than all nodes below are also
+ * shared. Currently some of the nodes below may be marked as not shared
+ * because we have just switched from one leaf to another, and switched
+ * also other nodes above the leaf and below the current level, so mark
+ * them as shared.
+ */
+ if (*is_shared) {
+ for (int i = 0; i < level; i++) {
+ cache->entries[i].is_shared = true;
+ cache->entries[i].gen = entry->gen;
+ }
+ }
return true;
}
@@ -1573,6 +1617,9 @@ static void store_backref_shared_cache(struct btrfs_backref_shared_cache *cache,
struct btrfs_backref_shared_cache_entry *entry;
u64 gen;
+ if (!cache->use_cache)
+ return;
+
if (WARN_ON_ONCE(level >= BTRFS_MAX_LEVEL))
return;
@@ -1648,6 +1695,7 @@ int btrfs_is_data_extent_shared(struct btrfs_root *root, u64 inum, u64 bytenr,
.root_objectid = root->root_key.objectid,
.inum = inum,
.share_count = 0,
+ .have_delayed_delete_refs = false,
};
int level;
@@ -1669,6 +1717,7 @@ int btrfs_is_data_extent_shared(struct btrfs_root *root, u64 inum, u64 bytenr,
/* -1 means we are in the bytenr of the data extent. */
level = -1;
ULIST_ITER_INIT(&uiter);
+ cache->use_cache = true;
while (1) {
bool is_shared;
bool cached;
@@ -1698,6 +1747,24 @@ int btrfs_is_data_extent_shared(struct btrfs_root *root, u64 inum, u64 bytenr,
extent_gen > btrfs_root_last_snapshot(&root->root_item))
break;
+ /*
+ * If our data extent was not directly shared (without multiple
+ * reference items), than it might have a single reference item
+ * with a count > 1 for the same offset, which means there are 2
+ * (or more) file extent items that point to the data extent -
+ * this happens when a file extent item needs to be split and
+ * then one item gets moved to another leaf due to a b+tree leaf
+ * split when inserting some item. In this case the file extent
+ * items may be located in different leaves and therefore some
+ * of the leaves may be referenced through shared subtrees while
+ * others are not. Since our extent buffer cache only works for
+ * a single path (by far the most common case and simpler to
+ * deal with), we can not use it if we have multiple leaves
+ * (which implies multiple paths).
+ */
+ if (level == -1 && tmp->nnodes > 1)
+ cache->use_cache = false;
+
if (level >= 0)
store_backref_shared_cache(cache, root, bytenr,
level, false);
@@ -1713,6 +1780,7 @@ int btrfs_is_data_extent_shared(struct btrfs_root *root, u64 inum, u64 bytenr,
break;
}
shared.share_count = 0;
+ shared.have_delayed_delete_refs = false;
cond_resched();
}
diff --git a/fs/btrfs/backref.h b/fs/btrfs/backref.h
index 52ae6957b414..8e69584d538d 100644
--- a/fs/btrfs/backref.h
+++ b/fs/btrfs/backref.h
@@ -29,6 +29,7 @@ struct btrfs_backref_shared_cache {
* a given data extent should never exceed the maximum b+tree height.
*/
struct btrfs_backref_shared_cache_entry entries[BTRFS_MAX_LEVEL];
+ bool use_cache;
};
typedef int (iterate_extent_inodes_t)(u64 inum, u64 offset, u64 root,
diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index 32c415cfbdfe..deebc8ddbd93 100644
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -774,10 +774,8 @@ int btrfs_cache_block_group(struct btrfs_block_group *cache, bool wait)
btrfs_queue_work(fs_info->caching_workers, &caching_ctl->work);
out:
- /* REVIEW */
if (wait && caching_ctl)
ret = btrfs_caching_ctl_wait_done(cache, caching_ctl);
- /* wait_event(caching_ctl->wait, space_cache_v1_done(cache)); */
if (caching_ctl)
btrfs_put_caching_control(caching_ctl);
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index f1f051ad3147..e6635fe70067 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -512,7 +512,7 @@ static u64 bio_end_offset(struct bio *bio)
static noinline int add_ra_bio_pages(struct inode *inode,
u64 compressed_end,
struct compressed_bio *cb,
- unsigned long *pflags)
+ int *memstall, unsigned long *pflags)
{
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
unsigned long end_index;
@@ -581,8 +581,10 @@ static noinline int add_ra_bio_pages(struct inode *inode,
continue;
}
- if (PageWorkingset(page))
+ if (!*memstall && PageWorkingset(page)) {
psi_memstall_enter(pflags);
+ *memstall = 1;
+ }
ret = set_page_extent_mapped(page);
if (ret < 0) {
@@ -670,8 +672,8 @@ void btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
u64 em_len;
u64 em_start;
struct extent_map *em;
- /* Initialize to 1 to make skip psi_memstall_leave unless needed */
- unsigned long pflags = 1;
+ unsigned long pflags;
+ int memstall = 0;
blk_status_t ret;
int ret2;
int i;
@@ -727,7 +729,7 @@ void btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
goto fail;
}
- add_ra_bio_pages(inode, em_start + em_len, cb, &pflags);
+ add_ra_bio_pages(inode, em_start + em_len, cb, &memstall, &pflags);
/* include any pages we added in add_ra-bio_pages */
cb->len = bio->bi_iter.bi_size;
@@ -807,7 +809,7 @@ void btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
}
}
- if (!pflags)
+ if (memstall)
psi_memstall_leave(&pflags);
if (refcount_dec_and_test(&cb->pending_ios))
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index b39b339fbf96..dcb510f38dda 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -114,6 +114,22 @@ noinline void btrfs_release_path(struct btrfs_path *p)
}
/*
+ * We want the transaction abort to print stack trace only for errors where the
+ * cause could be a bug, eg. due to ENOSPC, and not for common errors that are
+ * caused by external factors.
+ */
+bool __cold abort_should_print_stack(int errno)
+{
+ switch (errno) {
+ case -EIO:
+ case -EROFS:
+ case -ENOMEM:
+ return false;
+ }
+ return true;
+}
+
+/*
* safely gets a reference on the root node of a tree. A lock
* is not taken, so a concurrent writer may put a different node
* at the root of the tree. See btrfs_lock_root_node for the
@@ -4647,7 +4663,12 @@ int btrfs_next_old_leaf(struct btrfs_root *root, struct btrfs_path *path,
int ret;
int i;
- ASSERT(!path->nowait);
+ /*
+ * The nowait semantics are used only for write paths, where we don't
+ * use the tree mod log and sequence numbers.
+ */
+ if (time_seq)
+ ASSERT(!path->nowait);
nritems = btrfs_header_nritems(path->nodes[0]);
if (nritems == 0)
@@ -4667,7 +4688,14 @@ again:
if (path->need_commit_sem) {
path->need_commit_sem = 0;
need_commit_sem = true;
- down_read(&fs_info->commit_root_sem);
+ if (path->nowait) {
+ if (!down_read_trylock(&fs_info->commit_root_sem)) {
+ ret = -EAGAIN;
+ goto done;
+ }
+ } else {
+ down_read(&fs_info->commit_root_sem);
+ }
}
ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
}
@@ -4743,7 +4771,7 @@ again:
next = c;
ret = read_block_for_search(root, path, &next, level,
slot, &key);
- if (ret == -EAGAIN)
+ if (ret == -EAGAIN && !path->nowait)
goto again;
if (ret < 0) {
@@ -4753,6 +4781,10 @@ again:
if (!path->skip_locking) {
ret = btrfs_try_tree_read_lock(next);
+ if (!ret && path->nowait) {
+ ret = -EAGAIN;
+ goto done;
+ }
if (!ret && time_seq) {
/*
* If we don't get the lock, we may be racing
@@ -4783,7 +4815,7 @@ again:
ret = read_block_for_search(root, path, &next, level,
0, &key);
- if (ret == -EAGAIN)
+ if (ret == -EAGAIN && !path->nowait)
goto again;
if (ret < 0) {
@@ -4791,8 +4823,16 @@ again:
goto done;
}
- if (!path->skip_locking)
- btrfs_tree_read_lock(next);
+ if (!path->skip_locking) {
+ if (path->nowait) {
+ if (!btrfs_try_tree_read_lock(next)) {
+ ret = -EAGAIN;
+ goto done;
+ }
+ } else {
+ btrfs_tree_read_lock(next);
+ }
+ }
}
ret = 0;
done:
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 727595eee973..9e6d48ff4597 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -3462,7 +3462,10 @@ ssize_t btrfs_encoded_read(struct kiocb *iocb, struct iov_iter *iter,
ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from,
const struct btrfs_ioctl_encoded_io_args *encoded);
-ssize_t btrfs_dio_rw(struct kiocb *iocb, struct iov_iter *iter, size_t done_before);
+ssize_t btrfs_dio_read(struct kiocb *iocb, struct iov_iter *iter,
+ size_t done_before);
+struct iomap_dio *btrfs_dio_write(struct kiocb *iocb, struct iov_iter *iter,
+ size_t done_before);
extern const struct dentry_operations btrfs_dentry_operations;
@@ -3793,9 +3796,11 @@ void __btrfs_abort_transaction(struct btrfs_trans_handle *trans,
const char *function,
unsigned int line, int errno, bool first_hit);
+bool __cold abort_should_print_stack(int errno);
+
/*
* Call btrfs_abort_transaction as early as possible when an error condition is
- * detected, that way the exact line number is reported.
+ * detected, that way the exact stack trace is reported for some errors.
*/
#define btrfs_abort_transaction(trans, errno) \
do { \
@@ -3804,10 +3809,11 @@ do { \
if (!test_and_set_bit(BTRFS_FS_STATE_TRANS_ABORTED, \
&((trans)->fs_info->fs_state))) { \
first = true; \
- if ((errno) != -EIO && (errno) != -EROFS) { \
- WARN(1, KERN_DEBUG \
+ if (WARN(abort_should_print_stack(errno), \
+ KERN_DEBUG \
"BTRFS: Transaction aborted (error %d)\n", \
- (errno)); \
+ (errno))) { \
+ /* Stack trace printed. */ \
} else { \
btrfs_debug((trans)->fs_info, \
"Transaction aborted (error %d)", \
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index a2da9313c694..d99bf7c64611 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -166,11 +166,9 @@ static bool btrfs_supported_super_csum(u16 csum_type)
* Return 0 if the superblock checksum type matches the checksum value of that
* algorithm. Pass the raw disk superblock data.
*/
-static int btrfs_check_super_csum(struct btrfs_fs_info *fs_info,
- char *raw_disk_sb)
+int btrfs_check_super_csum(struct btrfs_fs_info *fs_info,
+ const struct btrfs_super_block *disk_sb)
{
- struct btrfs_super_block *disk_sb =
- (struct btrfs_super_block *)raw_disk_sb;
char result[BTRFS_CSUM_SIZE];
SHASH_DESC_ON_STACK(shash, fs_info->csum_shash);
@@ -181,7 +179,7 @@ static int btrfs_check_super_csum(struct btrfs_fs_info *fs_info,
* BTRFS_SUPER_INFO_SIZE range, we expect that the unused space is
* filled with zeros and is included in the checksum.
*/
- crypto_shash_digest(shash, raw_disk_sb + BTRFS_CSUM_SIZE,
+ crypto_shash_digest(shash, (const u8 *)disk_sb + BTRFS_CSUM_SIZE,
BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE, result);
if (memcmp(disk_sb->csum, result, fs_info->csum_size))
@@ -2553,7 +2551,9 @@ static int btrfs_read_roots(struct btrfs_fs_info *fs_info)
fs_info->dev_root = root;
}
/* Initialize fs_info for all devices in any case */
- btrfs_init_devices_late(fs_info);
+ ret = btrfs_init_devices_late(fs_info);
+ if (ret)
+ goto out;
/*
* This tree can share blocks with some other fs tree during relocation
@@ -3479,7 +3479,7 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device
* We want to check superblock checksum, the type is stored inside.
* Pass the whole disk block of size BTRFS_SUPER_INFO_SIZE (4k).
*/
- if (btrfs_check_super_csum(fs_info, (u8 *)disk_super)) {
+ if (btrfs_check_super_csum(fs_info, disk_super)) {
btrfs_err(fs_info, "superblock checksum mismatch");
err = -EINVAL;
btrfs_release_disk_super(disk_super);
diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h
index c67c15d4d20b..9fa923e005a3 100644
--- a/fs/btrfs/disk-io.h
+++ b/fs/btrfs/disk-io.h
@@ -42,6 +42,8 @@ struct extent_buffer *btrfs_find_create_tree_block(
void btrfs_clean_tree_block(struct extent_buffer *buf);
void btrfs_clear_oneshot_options(struct btrfs_fs_info *fs_info);
int btrfs_start_pre_rw_mount(struct btrfs_fs_info *fs_info);
+int btrfs_check_super_csum(struct btrfs_fs_info *fs_info,
+ const struct btrfs_super_block *disk_sb);
int __cold open_ctree(struct super_block *sb,
struct btrfs_fs_devices *fs_devices,
char *options);
diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c
index 1d4c2397d0d6..fab7eb76e53b 100644
--- a/fs/btrfs/export.c
+++ b/fs/btrfs/export.c
@@ -58,7 +58,7 @@ static int btrfs_encode_fh(struct inode *inode, u32 *fh, int *max_len,
}
struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid,
- u64 root_objectid, u32 generation,
+ u64 root_objectid, u64 generation,
int check_generation)
{
struct btrfs_fs_info *fs_info = btrfs_sb(sb);
diff --git a/fs/btrfs/export.h b/fs/btrfs/export.h
index f32f4113c976..5afb7ca42828 100644
--- a/fs/btrfs/export.h
+++ b/fs/btrfs/export.h
@@ -19,7 +19,7 @@ struct btrfs_fid {
} __attribute__ ((packed));
struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid,
- u64 root_objectid, u32 generation,
+ u64 root_objectid, u64 generation,
int check_generation);
struct dentry *btrfs_get_parent(struct dentry *child);
diff --git a/fs/btrfs/extent-io-tree.c b/fs/btrfs/extent-io-tree.c
index 618275af19c4..83cb0378096f 100644
--- a/fs/btrfs/extent-io-tree.c
+++ b/fs/btrfs/extent-io-tree.c
@@ -1641,16 +1641,17 @@ int lock_extent(struct extent_io_tree *tree, u64 start, u64 end,
int err;
u64 failed_start;
- while (1) {
+ err = __set_extent_bit(tree, start, end, EXTENT_LOCKED, &failed_start,
+ cached_state, NULL, GFP_NOFS);
+ while (err == -EEXIST) {
+ if (failed_start != start)
+ clear_extent_bit(tree, start, failed_start - 1,
+ EXTENT_LOCKED, cached_state);
+
+ wait_extent_bit(tree, failed_start, end, EXTENT_LOCKED);
err = __set_extent_bit(tree, start, end, EXTENT_LOCKED,
&failed_start, cached_state, NULL,
GFP_NOFS);
- if (err == -EEXIST) {
- wait_extent_bit(tree, failed_start, end, EXTENT_LOCKED);
- start = failed_start;
- } else
- break;
- WARN_ON(start > end);
}
return err;
}
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index cd2d36580f1a..2801c991814f 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -3295,21 +3295,22 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
}
/*
- * If this is a leaf and there are tree mod log users, we may
- * have recorded mod log operations that point to this leaf.
- * So we must make sure no one reuses this leaf's extent before
- * mod log operations are applied to a node, otherwise after
- * rewinding a node using the mod log operations we get an
- * inconsistent btree, as the leaf's extent may now be used as
- * a node or leaf for another different btree.
+ * If there are tree mod log users we may have recorded mod log
+ * operations for this node. If we re-allocate this node we
+ * could replay operations on this node that happened when it
+ * existed in a completely different root. For example if it
+ * was part of root A, then was reallocated to root B, and we
+ * are doing a btrfs_old_search_slot(root b), we could replay
+ * operations that happened when the block was part of root A,
+ * giving us an inconsistent view of the btree.
+ *
* We are safe from races here because at this point no other
* node or root points to this extent buffer, so if after this
- * check a new tree mod log user joins, it will not be able to
- * find a node pointing to this leaf and record operations that
- * point to this leaf.
+ * check a new tree mod log user joins we will not have an
+ * existing log of operations on this node that we have to
+ * contend with.
*/
- if (btrfs_header_level(buf) == 0 &&
- test_bit(BTRFS_FS_TREE_MOD_LOG_USERS, &fs_info->flags))
+ if (test_bit(BTRFS_FS_TREE_MOD_LOG_USERS, &fs_info->flags))
must_pin = true;
if (must_pin || btrfs_is_zoned(fs_info)) {
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 176b432035ae..d01631d47806 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1598,14 +1598,19 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
write_bytes);
else
btrfs_check_nocow_unlock(BTRFS_I(inode));
+
+ if (nowait && ret == -ENOSPC)
+ ret = -EAGAIN;
break;
}
release_bytes = reserve_bytes;
again:
ret = balance_dirty_pages_ratelimited_flags(inode->i_mapping, bdp_flags);
- if (ret)
+ if (ret) {
+ btrfs_delalloc_release_extents(BTRFS_I(inode), reserve_bytes);
break;
+ }
/*
* This is going to setup the pages array with the number of
@@ -1765,6 +1770,7 @@ static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from)
loff_t endbyte;
ssize_t err;
unsigned int ilock_flags = 0;
+ struct iomap_dio *dio;
if (iocb->ki_flags & IOCB_NOWAIT)
ilock_flags |= BTRFS_ILOCK_TRY;
@@ -1825,11 +1831,22 @@ relock:
* So here we disable page faults in the iov_iter and then retry if we
* got -EFAULT, faulting in the pages before the retry.
*/
-again:
from->nofault = true;
- err = btrfs_dio_rw(iocb, from, written);
+ dio = btrfs_dio_write(iocb, from, written);
from->nofault = false;
+ /*
+ * iomap_dio_complete() will call btrfs_sync_file() if we have a dsync
+ * iocb, and that needs to lock the inode. So unlock it before calling
+ * iomap_dio_complete() to avoid a deadlock.
+ */
+ btrfs_inode_unlock(inode, ilock_flags);
+
+ if (IS_ERR_OR_NULL(dio))
+ err = PTR_ERR_OR_ZERO(dio);
+ else
+ err = iomap_dio_complete(dio);
+
/* No increment (+=) because iomap returns a cumulative value. */
if (err > 0)
written = err;
@@ -1855,12 +1872,10 @@ again:
} else {
fault_in_iov_iter_readable(from, left);
prev_left = left;
- goto again;
+ goto relock;
}
}
- btrfs_inode_unlock(inode, ilock_flags);
-
/*
* If 'err' is -ENOTBLK or we have not written all data, then it means
* we must fallback to buffered IO.
@@ -4035,7 +4050,7 @@ again:
*/
pagefault_disable();
to->nofault = true;
- ret = btrfs_dio_rw(iocb, to, read);
+ ret = btrfs_dio_read(iocb, to, read);
to->nofault = false;
pagefault_enable();
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index b0807c59e321..0e516aefbf51 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -7980,7 +7980,7 @@ static void btrfs_submit_direct(const struct iomap_iter *iter,
*/
status = BLK_STS_RESOURCE;
dip->csums = kcalloc(nr_sectors, fs_info->csum_size, GFP_NOFS);
- if (!dip)
+ if (!dip->csums)
goto out_err;
status = btrfs_lookup_bio_sums(inode, dio_bio, dip->csums);
@@ -8078,13 +8078,21 @@ static const struct iomap_dio_ops btrfs_dio_ops = {
.bio_set = &btrfs_dio_bioset,
};
-ssize_t btrfs_dio_rw(struct kiocb *iocb, struct iov_iter *iter, size_t done_before)
+ssize_t btrfs_dio_read(struct kiocb *iocb, struct iov_iter *iter, size_t done_before)
{
struct btrfs_dio_data data;
return iomap_dio_rw(iocb, iter, &btrfs_dio_iomap_ops, &btrfs_dio_ops,
- IOMAP_DIO_PARTIAL | IOMAP_DIO_NOSYNC,
- &data, done_before);
+ IOMAP_DIO_PARTIAL, &data, done_before);
+}
+
+struct iomap_dio *btrfs_dio_write(struct kiocb *iocb, struct iov_iter *iter,
+ size_t done_before)
+{
+ struct btrfs_dio_data data;
+
+ return __iomap_dio_rw(iocb, iter, &btrfs_dio_iomap_ops, &btrfs_dio_ops,
+ IOMAP_DIO_PARTIAL, &data, done_before);
}
static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index d5dd8bed1488..f897be9ec1e9 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -3105,6 +3105,8 @@ static int btrfs_ioctl_get_subvol_info(struct inode *inode, void __user *argp)
}
}
+ btrfs_free_path(path);
+ path = NULL;
if (copy_to_user(argp, subvol_info, sizeof(*subvol_info)))
ret = -EFAULT;
@@ -3194,6 +3196,8 @@ static int btrfs_ioctl_get_subvol_rootref(struct btrfs_root *root,
}
out:
+ btrfs_free_path(path);
+
if (!ret || ret == -EOVERFLOW) {
rootrefs->num_items = found;
/* update min_treeid for next search */
@@ -3205,7 +3209,6 @@ out:
}
kfree(rootrefs);
- btrfs_free_path(path);
return ret;
}
@@ -4231,6 +4234,8 @@ static long btrfs_ioctl_ino_to_path(struct btrfs_root *root, void __user *arg)
ipath->fspath->val[i] = rel_ptr;
}
+ btrfs_free_path(path);
+ path = NULL;
ret = copy_to_user((void __user *)(unsigned long)ipa->fspath,
ipath->fspath, size);
if (ret) {
@@ -4281,21 +4286,20 @@ static long btrfs_ioctl_logical_to_ino(struct btrfs_fs_info *fs_info,
size = min_t(u32, loi->size, SZ_16M);
}
- path = btrfs_alloc_path();
- if (!path) {
- ret = -ENOMEM;
- goto out;
- }
-
inodes = init_data_container(size);
if (IS_ERR(inodes)) {
ret = PTR_ERR(inodes);
- inodes = NULL;
- goto out;
+ goto out_loi;
}
+ path = btrfs_alloc_path();
+ if (!path) {
+ ret = -ENOMEM;
+ goto out;
+ }
ret = iterate_inodes_from_logical(loi->logical, fs_info, path,
inodes, ignore_offset);
+ btrfs_free_path(path);
if (ret == -EINVAL)
ret = -ENOENT;
if (ret < 0)
@@ -4307,7 +4311,6 @@ static long btrfs_ioctl_logical_to_ino(struct btrfs_fs_info *fs_info,
ret = -EFAULT;
out:
- btrfs_free_path(path);
kvfree(inodes);
out_loi:
kfree(loi);
@@ -5283,7 +5286,7 @@ static int btrfs_ioctl_encoded_read(struct file *file, void __user *argp,
goto out_acct;
}
- ret = import_iovec(READ, args.iov, args.iovcnt, ARRAY_SIZE(iovstack),
+ ret = import_iovec(ITER_DEST, args.iov, args.iovcnt, ARRAY_SIZE(iovstack),
&iov, &iter);
if (ret < 0)
goto out_acct;
@@ -5382,7 +5385,7 @@ static int btrfs_ioctl_encoded_write(struct file *file, void __user *argp, bool
if (args.len > args.unencoded_len - args.unencoded_offset)
goto out_acct;
- ret = import_iovec(WRITE, args.iov, args.iovcnt, ARRAY_SIZE(iovstack),
+ ret = import_iovec(ITER_SOURCE, args.iov, args.iovcnt, ARRAY_SIZE(iovstack),
&iov, &iter);
if (ret < 0)
goto out_acct;
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index 9334c3157c22..b74105a10f16 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -2951,14 +2951,7 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans, u64 srcid,
dstgroup->rsv_rfer = inherit->lim.rsv_rfer;
dstgroup->rsv_excl = inherit->lim.rsv_excl;
- ret = update_qgroup_limit_item(trans, dstgroup);
- if (ret) {
- qgroup_mark_inconsistent(fs_info);
- btrfs_info(fs_info,
- "unable to update quota limit for %llu",
- dstgroup->qgroupid);
- goto unlock;
- }
+ qgroup_dirty(fs_info, dstgroup);
}
if (srcid) {
diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c
index f6395e8288d6..82c8e991300e 100644
--- a/fs/btrfs/raid56.c
+++ b/fs/btrfs/raid56.c
@@ -1632,10 +1632,8 @@ static int full_stripe_write(struct btrfs_raid_bio *rbio)
int ret;
ret = alloc_rbio_parity_pages(rbio);
- if (ret) {
- __free_raid_bio(rbio);
+ if (ret)
return ret;
- }
ret = lock_stripe_add(rbio);
if (ret == 0)
@@ -1823,8 +1821,10 @@ void raid56_parity_write(struct bio *bio, struct btrfs_io_context *bioc)
*/
if (rbio_is_full(rbio)) {
ret = full_stripe_write(rbio);
- if (ret)
+ if (ret) {
+ __free_raid_bio(rbio);
goto fail;
+ }
return;
}
@@ -1838,8 +1838,10 @@ void raid56_parity_write(struct bio *bio, struct btrfs_io_context *bioc)
list_add_tail(&rbio->plug_list, &plug->rbio_list);
} else {
ret = __raid56_parity_write(rbio);
- if (ret)
+ if (ret) {
+ __free_raid_bio(rbio);
goto fail;
+ }
}
return;
@@ -2742,8 +2744,10 @@ raid56_alloc_missing_rbio(struct bio *bio, struct btrfs_io_context *bioc)
rbio->faila = find_logical_bio_stripe(rbio, bio);
if (rbio->faila == -1) {
- BUG();
- kfree(rbio);
+ btrfs_warn_rl(fs_info,
+ "can not determine the failed stripe number for full stripe %llu",
+ bioc->raid_map[0]);
+ __free_raid_bio(rbio);
return NULL;
}
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index f260c53829e5..196c4c6ed1ed 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -2672,17 +2672,11 @@ static int scrub_extent(struct scrub_ctx *sctx, struct map_lookup *map,
u8 csum[BTRFS_CSUM_SIZE];
u32 blocksize;
- /*
- * Block size determines how many scrub_block will be allocated. Here
- * we use BTRFS_STRIPE_LEN (64KiB) as default limit, so we won't
- * allocate too many scrub_block, while still won't cause too large
- * bios for large extents.
- */
if (flags & BTRFS_EXTENT_FLAG_DATA) {
if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK)
blocksize = map->stripe_len;
else
- blocksize = BTRFS_STRIPE_LEN;
+ blocksize = sctx->fs_info->sectorsize;
spin_lock(&sctx->stat_lock);
sctx->stat.data_extents_scrubbed++;
sctx->stat.data_bytes_scrubbed += len;
@@ -3917,7 +3911,6 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx,
if (sctx->is_dev_replace && btrfs_is_zoned(fs_info)) {
if (!test_bit(BLOCK_GROUP_FLAG_TO_COPY, &cache->runtime_flags)) {
- spin_unlock(&cache->lock);
btrfs_put_block_group(cache);
goto skip;
}
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index 4ef4167072b8..1c4b693ee4a3 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -348,6 +348,7 @@ static bool proto_cmd_ok(const struct send_ctx *sctx, int cmd)
switch (sctx->proto) {
case 1: return cmd <= BTRFS_SEND_C_MAX_V1;
case 2: return cmd <= BTRFS_SEND_C_MAX_V2;
+ case 3: return cmd <= BTRFS_SEND_C_MAX_V3;
default: return false;
}
}
@@ -5701,6 +5702,7 @@ static int clone_range(struct send_ctx *sctx, struct btrfs_path *dst_path,
u64 ext_len;
u64 clone_len;
u64 clone_data_offset;
+ bool crossed_src_i_size = false;
if (slot >= btrfs_header_nritems(leaf)) {
ret = btrfs_next_leaf(clone_root->root, path);
@@ -5758,8 +5760,10 @@ static int clone_range(struct send_ctx *sctx, struct btrfs_path *dst_path,
if (key.offset >= clone_src_i_size)
break;
- if (key.offset + ext_len > clone_src_i_size)
+ if (key.offset + ext_len > clone_src_i_size) {
ext_len = clone_src_i_size - key.offset;
+ crossed_src_i_size = true;
+ }
clone_data_offset = btrfs_file_extent_offset(leaf, ei);
if (btrfs_file_extent_disk_bytenr(leaf, ei) == disk_byte) {
@@ -5820,6 +5824,25 @@ static int clone_range(struct send_ctx *sctx, struct btrfs_path *dst_path,
ret = send_clone(sctx, offset, clone_len,
clone_root);
}
+ } else if (crossed_src_i_size && clone_len < len) {
+ /*
+ * If we are at i_size of the clone source inode and we
+ * can not clone from it, terminate the loop. This is
+ * to avoid sending two write operations, one with a
+ * length matching clone_len and the final one after
+ * this loop with a length of len - clone_len.
+ *
+ * When using encoded writes (BTRFS_SEND_FLAG_COMPRESSED
+ * was passed to the send ioctl), this helps avoid
+ * sending an encoded write for an offset that is not
+ * sector size aligned, in case the i_size of the source
+ * inode is not sector size aligned. That will make the
+ * receiver fallback to decompression of the data and
+ * writing it using regular buffered IO, therefore while
+ * not incorrect, it's not optimal due decompression and
+ * possible re-compression at the receiver.
+ */
+ break;
} else {
ret = send_extent_data(sctx, dst_path, offset,
clone_len);
@@ -6469,7 +6492,9 @@ static int finish_inode_if_needed(struct send_ctx *sctx, int at_end)
if (ret < 0)
goto out;
}
- if (sctx->cur_inode_needs_verity) {
+
+ if (proto_cmd_ok(sctx, BTRFS_SEND_C_ENABLE_VERITY)
+ && sctx->cur_inode_needs_verity) {
ret = process_verity(sctx);
if (ret < 0)
goto out;
@@ -6665,17 +6690,19 @@ static int changed_inode(struct send_ctx *sctx,
/*
* First, process the inode as if it was deleted.
*/
- sctx->cur_inode_gen = right_gen;
- sctx->cur_inode_new = false;
- sctx->cur_inode_deleted = true;
- sctx->cur_inode_size = btrfs_inode_size(
- sctx->right_path->nodes[0], right_ii);
- sctx->cur_inode_mode = btrfs_inode_mode(
- sctx->right_path->nodes[0], right_ii);
- ret = process_all_refs(sctx,
- BTRFS_COMPARE_TREE_DELETED);
- if (ret < 0)
- goto out;
+ if (old_nlinks > 0) {
+ sctx->cur_inode_gen = right_gen;
+ sctx->cur_inode_new = false;
+ sctx->cur_inode_deleted = true;
+ sctx->cur_inode_size = btrfs_inode_size(
+ sctx->right_path->nodes[0], right_ii);
+ sctx->cur_inode_mode = btrfs_inode_mode(
+ sctx->right_path->nodes[0], right_ii);
+ ret = process_all_refs(sctx,
+ BTRFS_COMPARE_TREE_DELETED);
+ if (ret < 0)
+ goto out;
+ }
/*
* Now process the inode as if it was new.
diff --git a/fs/btrfs/send.h b/fs/btrfs/send.h
index 0a4537775e0c..f7585cfa7e52 100644
--- a/fs/btrfs/send.h
+++ b/fs/btrfs/send.h
@@ -10,7 +10,12 @@
#include <linux/types.h>
#define BTRFS_SEND_STREAM_MAGIC "btrfs-stream"
+/* Conditional support for the upcoming protocol version. */
+#ifdef CONFIG_BTRFS_DEBUG
+#define BTRFS_SEND_STREAM_VERSION 3
+#else
#define BTRFS_SEND_STREAM_VERSION 2
+#endif
/*
* In send stream v1, no command is larger than 64K. In send stream v2, no limit
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 9be4fd2db0f4..5942b9384088 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -2555,6 +2555,7 @@ static int check_dev_super(struct btrfs_device *dev)
{
struct btrfs_fs_info *fs_info = dev->fs_info;
struct btrfs_super_block *sb;
+ u16 csum_type;
int ret = 0;
/* This should be called with fs still frozen. */
@@ -2569,6 +2570,21 @@ static int check_dev_super(struct btrfs_device *dev)
if (IS_ERR(sb))
return PTR_ERR(sb);
+ /* Verify the checksum. */
+ csum_type = btrfs_super_csum_type(sb);
+ if (csum_type != btrfs_super_csum_type(fs_info->super_copy)) {
+ btrfs_err(fs_info, "csum type changed, has %u expect %u",
+ csum_type, btrfs_super_csum_type(fs_info->super_copy));
+ ret = -EUCLEAN;
+ goto out;
+ }
+
+ if (btrfs_check_super_csum(fs_info, sb)) {
+ btrfs_err(fs_info, "csum for on-disk super block no longer matches");
+ ret = -EUCLEAN;
+ goto out;
+ }
+
/* Btrfs_validate_super() includes fsid check against super->fsid. */
ret = btrfs_validate_super(fs_info, sb, 0);
if (ret < 0)
diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
index 699b54b3acaa..74fef1f49c35 100644
--- a/fs/btrfs/sysfs.c
+++ b/fs/btrfs/sysfs.c
@@ -2321,8 +2321,11 @@ int __init btrfs_init_sysfs(void)
#ifdef CONFIG_BTRFS_DEBUG
ret = sysfs_create_group(&btrfs_kset->kobj, &btrfs_debug_feature_attr_group);
- if (ret)
- goto out2;
+ if (ret) {
+ sysfs_unmerge_group(&btrfs_kset->kobj,
+ &btrfs_static_feature_attr_group);
+ goto out_remove_group;
+ }
#endif
return 0;
diff --git a/fs/btrfs/tests/btrfs-tests.c b/fs/btrfs/tests/btrfs-tests.c
index 9c478fa256f6..d43cb5242fec 100644
--- a/fs/btrfs/tests/btrfs-tests.c
+++ b/fs/btrfs/tests/btrfs-tests.c
@@ -200,7 +200,7 @@ void btrfs_free_dummy_fs_info(struct btrfs_fs_info *fs_info)
void btrfs_free_dummy_root(struct btrfs_root *root)
{
- if (!root)
+ if (IS_ERR_OR_NULL(root))
return;
/* Will be freed by btrfs_free_fs_roots */
if (WARN_ON(test_bit(BTRFS_ROOT_IN_RADIX, &root->state)))
diff --git a/fs/btrfs/tests/qgroup-tests.c b/fs/btrfs/tests/qgroup-tests.c
index eee1e4459541..63676ea19f29 100644
--- a/fs/btrfs/tests/qgroup-tests.c
+++ b/fs/btrfs/tests/qgroup-tests.c
@@ -225,20 +225,20 @@ static int test_no_shared_qgroup(struct btrfs_root *root,
*/
ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &old_roots, false);
if (ret) {
- ulist_free(old_roots);
test_err("couldn't find old roots: %d", ret);
return ret;
}
ret = insert_normal_tree_ref(root, nodesize, nodesize, 0,
BTRFS_FS_TREE_OBJECTID);
- if (ret)
+ if (ret) {
+ ulist_free(old_roots);
return ret;
+ }
ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &new_roots, false);
if (ret) {
ulist_free(old_roots);
- ulist_free(new_roots);
test_err("couldn't find old roots: %d", ret);
return ret;
}
@@ -250,29 +250,31 @@ static int test_no_shared_qgroup(struct btrfs_root *root,
return ret;
}
+ /* btrfs_qgroup_account_extent() always frees the ulists passed to it. */
+ old_roots = NULL;
+ new_roots = NULL;
+
if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID,
nodesize, nodesize)) {
test_err("qgroup counts didn't match expected values");
return -EINVAL;
}
- old_roots = NULL;
- new_roots = NULL;
ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &old_roots, false);
if (ret) {
- ulist_free(old_roots);
test_err("couldn't find old roots: %d", ret);
return ret;
}
ret = remove_extent_item(root, nodesize, nodesize);
- if (ret)
+ if (ret) {
+ ulist_free(old_roots);
return -EINVAL;
+ }
ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &new_roots, false);
if (ret) {
ulist_free(old_roots);
- ulist_free(new_roots);
test_err("couldn't find old roots: %d", ret);
return ret;
}
@@ -322,20 +324,20 @@ static int test_multiple_refs(struct btrfs_root *root,
ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &old_roots, false);
if (ret) {
- ulist_free(old_roots);
test_err("couldn't find old roots: %d", ret);
return ret;
}
ret = insert_normal_tree_ref(root, nodesize, nodesize, 0,
BTRFS_FS_TREE_OBJECTID);
- if (ret)
+ if (ret) {
+ ulist_free(old_roots);
return ret;
+ }
ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &new_roots, false);
if (ret) {
ulist_free(old_roots);
- ulist_free(new_roots);
test_err("couldn't find old roots: %d", ret);
return ret;
}
@@ -355,20 +357,20 @@ static int test_multiple_refs(struct btrfs_root *root,
ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &old_roots, false);
if (ret) {
- ulist_free(old_roots);
test_err("couldn't find old roots: %d", ret);
return ret;
}
ret = add_tree_ref(root, nodesize, nodesize, 0,
BTRFS_FIRST_FREE_OBJECTID);
- if (ret)
+ if (ret) {
+ ulist_free(old_roots);
return ret;
+ }
ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &new_roots, false);
if (ret) {
ulist_free(old_roots);
- ulist_free(new_roots);
test_err("couldn't find old roots: %d", ret);
return ret;
}
@@ -394,20 +396,20 @@ static int test_multiple_refs(struct btrfs_root *root,
ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &old_roots, false);
if (ret) {
- ulist_free(old_roots);
test_err("couldn't find old roots: %d", ret);
return ret;
}
ret = remove_extent_ref(root, nodesize, nodesize, 0,
BTRFS_FIRST_FREE_OBJECTID);
- if (ret)
+ if (ret) {
+ ulist_free(old_roots);
return ret;
+ }
ret = btrfs_find_all_roots(&trans, fs_info, nodesize, 0, &new_roots, false);
if (ret) {
ulist_free(old_roots);
- ulist_free(new_roots);
test_err("couldn't find old roots: %d", ret);
return ret;
}
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 813986e38258..c3cf3dabe0b1 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -3694,15 +3694,29 @@ static int process_dir_items_leaf(struct btrfs_trans_handle *trans,
u64 *last_old_dentry_offset)
{
struct btrfs_root *log = inode->root->log_root;
- struct extent_buffer *src = path->nodes[0];
- const int nritems = btrfs_header_nritems(src);
+ struct extent_buffer *src;
+ const int nritems = btrfs_header_nritems(path->nodes[0]);
const u64 ino = btrfs_ino(inode);
bool last_found = false;
int batch_start = 0;
int batch_size = 0;
int i;
- for (i = path->slots[0]; i < nritems; i++) {
+ /*
+ * We need to clone the leaf, release the read lock on it, and use the
+ * clone before modifying the log tree. See the comment at copy_items()
+ * about why we need to do this.
+ */
+ src = btrfs_clone_extent_buffer(path->nodes[0]);
+ if (!src)
+ return -ENOMEM;
+
+ i = path->slots[0];
+ btrfs_release_path(path);
+ path->nodes[0] = src;
+ path->slots[0] = i;
+
+ for (; i < nritems; i++) {
struct btrfs_dir_item *di;
struct btrfs_key key;
int ret;
@@ -4303,7 +4317,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
{
struct btrfs_root *log = inode->root->log_root;
struct btrfs_file_extent_item *extent;
- struct extent_buffer *src = src_path->nodes[0];
+ struct extent_buffer *src;
int ret = 0;
struct btrfs_key *ins_keys;
u32 *ins_sizes;
@@ -4314,6 +4328,43 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
const bool skip_csum = (inode->flags & BTRFS_INODE_NODATASUM);
const u64 i_size = i_size_read(&inode->vfs_inode);
+ /*
+ * To keep lockdep happy and avoid deadlocks, clone the source leaf and
+ * use the clone. This is because otherwise we would be changing the log
+ * tree, to insert items from the subvolume tree or insert csum items,
+ * while holding a read lock on a leaf from the subvolume tree, which
+ * creates a nasty lock dependency when COWing log tree nodes/leaves:
+ *
+ * 1) Modifying the log tree triggers an extent buffer allocation while
+ * holding a write lock on a parent extent buffer from the log tree.
+ * Allocating the pages for an extent buffer, or the extent buffer
+ * struct, can trigger inode eviction and finally the inode eviction
+ * will trigger a release/remove of a delayed node, which requires
+ * taking the delayed node's mutex;
+ *
+ * 2) Allocating a metadata extent for a log tree can trigger the async
+ * reclaim thread and make us wait for it to release enough space and
+ * unblock our reservation ticket. The reclaim thread can start
+ * flushing delayed items, and that in turn results in the need to
+ * lock delayed node mutexes and in the need to write lock extent
+ * buffers of a subvolume tree - all this while holding a write lock
+ * on the parent extent buffer in the log tree.
+ *
+ * So one task in scenario 1) running in parallel with another task in
+ * scenario 2) could lead to a deadlock, one wanting to lock a delayed
+ * node mutex while having a read lock on a leaf from the subvolume,
+ * while the other is holding the delayed node's mutex and wants to
+ * write lock the same subvolume leaf for flushing delayed items.
+ */
+ src = btrfs_clone_extent_buffer(src_path->nodes[0]);
+ if (!src)
+ return -ENOMEM;
+
+ i = src_path->slots[0];
+ btrfs_release_path(src_path);
+ src_path->nodes[0] = src;
+ src_path->slots[0] = i;
+
ins_data = kmalloc(nr * sizeof(struct btrfs_key) +
nr * sizeof(u32), GFP_NOFS);
if (!ins_data)
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 94ba46d57920..635f45f1a2ef 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -1011,6 +1011,18 @@ static struct btrfs_fs_devices *clone_fs_devices(struct btrfs_fs_devices *orig)
rcu_assign_pointer(device->name, name);
}
+ if (orig_dev->zone_info) {
+ struct btrfs_zoned_device_info *zone_info;
+
+ zone_info = btrfs_clone_dev_zone_info(orig_dev);
+ if (!zone_info) {
+ btrfs_free_device(device);
+ ret = -ENOMEM;
+ goto error;
+ }
+ device->zone_info = zone_info;
+ }
+
list_add(&device->dev_list, &fs_devices->devices);
device->fs_devices = fs_devices;
fs_devices->num_devices++;
@@ -6918,18 +6930,18 @@ static bool dev_args_match_fs_devices(const struct btrfs_dev_lookup_args *args,
static bool dev_args_match_device(const struct btrfs_dev_lookup_args *args,
const struct btrfs_device *device)
{
- ASSERT((args->devid != (u64)-1) || args->missing);
+ if (args->missing) {
+ if (test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state) &&
+ !device->bdev)
+ return true;
+ return false;
+ }
- if ((args->devid != (u64)-1) && device->devid != args->devid)
+ if (device->devid != args->devid)
return false;
if (args->uuid && memcmp(device->uuid, args->uuid, BTRFS_UUID_SIZE) != 0)
return false;
- if (!args->missing)
- return true;
- if (test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state) &&
- !device->bdev)
- return true;
- return false;
+ return true;
}
/*
@@ -7142,6 +7154,7 @@ static int read_one_chunk(struct btrfs_key *key, struct extent_buffer *leaf,
u64 devid;
u64 type;
u8 uuid[BTRFS_UUID_SIZE];
+ int index;
int num_stripes;
int ret;
int i;
@@ -7149,6 +7162,7 @@ static int read_one_chunk(struct btrfs_key *key, struct extent_buffer *leaf,
logical = key->offset;
length = btrfs_chunk_length(leaf, chunk);
type = btrfs_chunk_type(leaf, chunk);
+ index = btrfs_bg_flags_to_raid_index(type);
num_stripes = btrfs_chunk_num_stripes(leaf, chunk);
#if BITS_PER_LONG == 32
@@ -7202,7 +7216,15 @@ static int read_one_chunk(struct btrfs_key *key, struct extent_buffer *leaf,
map->io_align = btrfs_chunk_io_align(leaf, chunk);
map->stripe_len = btrfs_chunk_stripe_len(leaf, chunk);
map->type = type;
- map->sub_stripes = btrfs_chunk_sub_stripes(leaf, chunk);
+ /*
+ * We can't use the sub_stripes value, as for profiles other than
+ * RAID10, they may have 0 as sub_stripes for filesystems created by
+ * older mkfs (<v5.4).
+ * In that case, it can cause divide-by-zero errors later.
+ * Since currently sub_stripes is fixed for each profile, let's
+ * use the trusted value instead.
+ */
+ map->sub_stripes = btrfs_raid_array[index].sub_stripes;
map->verified_stripes = 0;
em->orig_block_len = btrfs_calc_stripe_length(em);
for (i = 0; i < num_stripes; i++) {
@@ -7734,10 +7756,11 @@ error:
return ret;
}
-void btrfs_init_devices_late(struct btrfs_fs_info *fs_info)
+int btrfs_init_devices_late(struct btrfs_fs_info *fs_info)
{
struct btrfs_fs_devices *fs_devices = fs_info->fs_devices, *seed_devs;
struct btrfs_device *device;
+ int ret = 0;
fs_devices->fs_info = fs_info;
@@ -7746,12 +7769,18 @@ void btrfs_init_devices_late(struct btrfs_fs_info *fs_info)
device->fs_info = fs_info;
list_for_each_entry(seed_devs, &fs_devices->seed_list, seed_list) {
- list_for_each_entry(device, &seed_devs->devices, dev_list)
+ list_for_each_entry(device, &seed_devs->devices, dev_list) {
device->fs_info = fs_info;
+ ret = btrfs_get_dev_zone_info(device, false);
+ if (ret)
+ break;
+ }
seed_devs->fs_info = fs_info;
}
mutex_unlock(&fs_devices->device_list_mutex);
+
+ return ret;
}
static u64 btrfs_dev_stats_value(const struct extent_buffer *eb,
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index 599b9d5af349..099def5613b8 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -395,6 +395,7 @@ typedef void (*btrfs_bio_end_io_t)(struct btrfs_bio *bbio);
*/
struct btrfs_bio {
unsigned int mirror_num;
+ struct bvec_iter iter;
/* for direct I/O */
u64 file_offset;
@@ -403,7 +404,6 @@ struct btrfs_bio {
struct btrfs_device *device;
u8 *csum;
u8 csum_inline[BTRFS_BIO_INLINE_CSUM_SIZE];
- struct bvec_iter iter;
/* End I/O information supplied to btrfs_bio_alloc */
btrfs_bio_end_io_t end_io;
@@ -671,7 +671,7 @@ int find_free_dev_extent(struct btrfs_device *device, u64 num_bytes,
void btrfs_dev_stat_inc_and_print(struct btrfs_device *dev, int index);
int btrfs_get_dev_stats(struct btrfs_fs_info *fs_info,
struct btrfs_ioctl_get_dev_stats *stats);
-void btrfs_init_devices_late(struct btrfs_fs_info *fs_info);
+int btrfs_init_devices_late(struct btrfs_fs_info *fs_info);
int btrfs_init_dev_stats(struct btrfs_fs_info *fs_info);
int btrfs_run_dev_stats(struct btrfs_trans_handle *trans);
void btrfs_rm_dev_replace_remove_srcdev(struct btrfs_device *srcdev);
diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c
index e2d073b08a7d..c9e2b0c85309 100644
--- a/fs/btrfs/zoned.c
+++ b/fs/btrfs/zoned.c
@@ -134,7 +134,8 @@ static int sb_write_pointer(struct block_device *bdev, struct blk_zone *zones,
super[i] = page_address(page[i]);
}
- if (super[0]->generation > super[1]->generation)
+ if (btrfs_super_generation(super[0]) >
+ btrfs_super_generation(super[1]))
sector = zones[1].start;
else
sector = zones[0].start;
@@ -466,7 +467,7 @@ int btrfs_get_dev_zone_info(struct btrfs_device *device, bool populate_cache)
goto out;
}
- zones = kcalloc(BTRFS_REPORT_NR_ZONES, sizeof(struct blk_zone), GFP_KERNEL);
+ zones = kvcalloc(BTRFS_REPORT_NR_ZONES, sizeof(struct blk_zone), GFP_KERNEL);
if (!zones) {
ret = -ENOMEM;
goto out;
@@ -585,7 +586,7 @@ int btrfs_get_dev_zone_info(struct btrfs_device *device, bool populate_cache)
}
- kfree(zones);
+ kvfree(zones);
switch (bdev_zoned_model(bdev)) {
case BLK_ZONED_HM:
@@ -617,7 +618,7 @@ int btrfs_get_dev_zone_info(struct btrfs_device *device, bool populate_cache)
return 0;
out:
- kfree(zones);
+ kvfree(zones);
out_free_zone_info:
btrfs_destroy_dev_zone_info(device);
@@ -639,6 +640,46 @@ void btrfs_destroy_dev_zone_info(struct btrfs_device *device)
device->zone_info = NULL;
}
+struct btrfs_zoned_device_info *btrfs_clone_dev_zone_info(struct btrfs_device *orig_dev)
+{
+ struct btrfs_zoned_device_info *zone_info;
+
+ zone_info = kmemdup(orig_dev->zone_info, sizeof(*zone_info), GFP_KERNEL);
+ if (!zone_info)
+ return NULL;
+
+ zone_info->seq_zones = bitmap_zalloc(zone_info->nr_zones, GFP_KERNEL);
+ if (!zone_info->seq_zones)
+ goto out;
+
+ bitmap_copy(zone_info->seq_zones, orig_dev->zone_info->seq_zones,
+ zone_info->nr_zones);
+
+ zone_info->empty_zones = bitmap_zalloc(zone_info->nr_zones, GFP_KERNEL);
+ if (!zone_info->empty_zones)
+ goto out;
+
+ bitmap_copy(zone_info->empty_zones, orig_dev->zone_info->empty_zones,
+ zone_info->nr_zones);
+
+ zone_info->active_zones = bitmap_zalloc(zone_info->nr_zones, GFP_KERNEL);
+ if (!zone_info->active_zones)
+ goto out;
+
+ bitmap_copy(zone_info->active_zones, orig_dev->zone_info->active_zones,
+ zone_info->nr_zones);
+ zone_info->zone_cache = NULL;
+
+ return zone_info;
+
+out:
+ bitmap_free(zone_info->seq_zones);
+ bitmap_free(zone_info->empty_zones);
+ bitmap_free(zone_info->active_zones);
+ kfree(zone_info);
+ return NULL;
+}
+
int btrfs_get_dev_zone(struct btrfs_device *device, u64 pos,
struct blk_zone *zone)
{
diff --git a/fs/btrfs/zoned.h b/fs/btrfs/zoned.h
index e17462db3a84..8bd16d40b7c6 100644
--- a/fs/btrfs/zoned.h
+++ b/fs/btrfs/zoned.h
@@ -36,6 +36,7 @@ int btrfs_get_dev_zone(struct btrfs_device *device, u64 pos,
int btrfs_get_dev_zone_info_all_devices(struct btrfs_fs_info *fs_info);
int btrfs_get_dev_zone_info(struct btrfs_device *device, bool populate_cache);
void btrfs_destroy_dev_zone_info(struct btrfs_device *device);
+struct btrfs_zoned_device_info *btrfs_clone_dev_zone_info(struct btrfs_device *orig_dev);
int btrfs_check_zoned_mode(struct btrfs_fs_info *fs_info);
int btrfs_check_mountopts_zoned(struct btrfs_fs_info *info);
int btrfs_sb_log_location_bdev(struct block_device *bdev, int mirror, int rw,
@@ -103,6 +104,16 @@ static inline int btrfs_get_dev_zone_info(struct btrfs_device *device,
static inline void btrfs_destroy_dev_zone_info(struct btrfs_device *device) { }
+/*
+ * In case the kernel is compiled without CONFIG_BLK_DEV_ZONED we'll never call
+ * into btrfs_clone_dev_zone_info() so it's safe to return NULL here.
+ */
+static inline struct btrfs_zoned_device_info *btrfs_clone_dev_zone_info(
+ struct btrfs_device *orig_dev)
+{
+ return NULL;
+}
+
static inline int btrfs_check_zoned_mode(const struct btrfs_fs_info *fs_info)
{
if (!btrfs_is_zoned(fs_info))
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-11 06:15:43 +0000