aboutsummaryrefslogtreecommitdiff
path: root/fs/btrfs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/btrfs')
-rw-r--r--fs/btrfs/compression.c1
-rw-r--r--fs/btrfs/dev-replace.c35
-rw-r--r--fs/btrfs/extent-tree.c67
-rw-r--r--fs/btrfs/file.c16
-rw-r--r--fs/btrfs/inode.c14
-rw-r--r--fs/btrfs/ioctl.c36
-rw-r--r--fs/btrfs/props.c30
-rw-r--r--fs/btrfs/qgroup.c8
-rw-r--r--fs/btrfs/reada.c5
-rw-r--r--fs/btrfs/relocation.c27
-rw-r--r--fs/btrfs/root-tree.c4
-rw-r--r--fs/btrfs/send.c52
-rw-r--r--fs/btrfs/sysfs.c25
-rw-r--r--fs/btrfs/tree-checker.c49
-rw-r--r--fs/btrfs/tree-log.c21
-rw-r--r--fs/btrfs/xattr.c6
-rw-r--r--fs/btrfs/zstd.c20
17 files changed, 270 insertions, 146 deletions
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index daf7908d1e35..84dd4a8980c5 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -1008,6 +1008,7 @@ int btrfs_compress_pages(unsigned int type_level, struct address_space *mapping,
struct list_head *workspace;
int ret;
+ level = btrfs_compress_op[type]->set_level(level);
workspace = get_workspace(type, level);
ret = btrfs_compress_op[type]->compress_pages(workspace, mapping,
start, pages,
diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c
index 55c15f31d00d..ee0989c7e3a9 100644
--- a/fs/btrfs/dev-replace.c
+++ b/fs/btrfs/dev-replace.c
@@ -603,17 +603,33 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
}
btrfs_wait_ordered_roots(fs_info, U64_MAX, 0, (u64)-1);
- trans = btrfs_start_transaction(root, 0);
- if (IS_ERR(trans)) {
- mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
- return PTR_ERR(trans);
+ /*
+ * We have to use this loop approach because at this point src_device
+ * has to be available for transaction commit to complete, yet new
+ * chunks shouldn't be allocated on the device.
+ */
+ while (1) {
+ trans = btrfs_start_transaction(root, 0);
+ if (IS_ERR(trans)) {
+ mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
+ return PTR_ERR(trans);
+ }
+ ret = btrfs_commit_transaction(trans);
+ WARN_ON(ret);
+
+ /* Prevent write_all_supers() during the finishing procedure */
+ mutex_lock(&fs_info->fs_devices->device_list_mutex);
+ /* Prevent new chunks being allocated on the source device */
+ mutex_lock(&fs_info->chunk_mutex);
+
+ if (!list_empty(&src_device->post_commit_list)) {
+ mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+ mutex_unlock(&fs_info->chunk_mutex);
+ } else {
+ break;
+ }
}
- ret = btrfs_commit_transaction(trans);
- WARN_ON(ret);
- /* keep away write_all_supers() during the finishing procedure */
- mutex_lock(&fs_info->fs_devices->device_list_mutex);
- mutex_lock(&fs_info->chunk_mutex);
down_write(&dev_replace->rwsem);
dev_replace->replace_state =
scrub_ret ? BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED
@@ -662,7 +678,6 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
btrfs_device_set_disk_total_bytes(tgt_device,
src_device->disk_total_bytes);
btrfs_device_set_bytes_used(tgt_device, src_device->bytes_used);
- ASSERT(list_empty(&src_device->post_commit_list));
tgt_device->commit_total_bytes = src_device->commit_total_bytes;
tgt_device->commit_bytes_used = src_device->bytes_used;
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index f79e477a378e..5faf057f6f37 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -757,12 +757,14 @@ static struct btrfs_space_info *__find_space_info(struct btrfs_fs_info *info,
}
static void add_pinned_bytes(struct btrfs_fs_info *fs_info,
- struct btrfs_ref *ref)
+ struct btrfs_ref *ref, int sign)
{
struct btrfs_space_info *space_info;
- s64 num_bytes = -ref->len;
+ s64 num_bytes;
u64 flags;
+ ASSERT(sign == 1 || sign == -1);
+ num_bytes = sign * ref->len;
if (ref->type == BTRFS_REF_METADATA) {
if (ref->tree_ref.root == BTRFS_CHUNK_TREE_OBJECTID)
flags = BTRFS_BLOCK_GROUP_SYSTEM;
@@ -2063,7 +2065,7 @@ int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
btrfs_ref_tree_mod(fs_info, generic_ref);
if (ret == 0 && old_ref_mod < 0 && new_ref_mod >= 0)
- add_pinned_bytes(fs_info, generic_ref);
+ add_pinned_bytes(fs_info, generic_ref, -1);
return ret;
}
@@ -3882,8 +3884,7 @@ static int create_space_info(struct btrfs_fs_info *info, u64 flags)
info->space_info_kobj, "%s",
alloc_name(space_info->flags));
if (ret) {
- percpu_counter_destroy(&space_info->total_bytes_pinned);
- kfree(space_info);
+ kobject_put(&space_info->kobj);
return ret;
}
@@ -7190,7 +7191,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
}
out:
if (pin)
- add_pinned_bytes(fs_info, &generic_ref);
+ add_pinned_bytes(fs_info, &generic_ref, 1);
if (last_ref) {
/*
@@ -7238,7 +7239,7 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_ref *ref)
btrfs_ref_tree_mod(fs_info, ref);
if (ret == 0 && old_ref_mod >= 0 && new_ref_mod < 0)
- add_pinned_bytes(fs_info, ref);
+ add_pinned_bytes(fs_info, ref, 1);
return ret;
}
@@ -10830,17 +10831,6 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
remove_em = (atomic_read(&block_group->trimming) == 0);
spin_unlock(&block_group->lock);
- if (remove_em) {
- struct extent_map_tree *em_tree;
-
- em_tree = &fs_info->mapping_tree.map_tree;
- write_lock(&em_tree->lock);
- remove_extent_mapping(em_tree, em);
- write_unlock(&em_tree->lock);
- /* once for the tree */
- free_extent_map(em);
- }
-
mutex_unlock(&fs_info->chunk_mutex);
ret = remove_block_group_free_space(trans, block_group);
@@ -10857,6 +10847,19 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
goto out;
ret = btrfs_del_item(trans, root, path);
+ if (ret)
+ goto out;
+
+ if (remove_em) {
+ struct extent_map_tree *em_tree;
+
+ em_tree = &fs_info->mapping_tree.map_tree;
+ write_lock(&em_tree->lock);
+ remove_extent_mapping(em_tree, em);
+ write_unlock(&em_tree->lock);
+ /* once for the tree */
+ free_extent_map(em);
+ }
out:
if (remove_rsv)
btrfs_delayed_refs_rsv_release(fs_info, 1);
@@ -11136,13 +11139,11 @@ int btrfs_error_unpin_extent_range(struct btrfs_fs_info *fs_info,
* it while performing the free space search since we have already
* held back allocations.
*/
-static int btrfs_trim_free_extents(struct btrfs_device *device,
- struct fstrim_range *range, u64 *trimmed)
+static int btrfs_trim_free_extents(struct btrfs_device *device, u64 *trimmed)
{
- u64 start, len = 0, end = 0;
+ u64 start = SZ_1M, len = 0, end = 0;
int ret;
- start = max_t(u64, range->start, SZ_1M);
*trimmed = 0;
/* Discard not supported = nothing to do. */
@@ -11185,22 +11186,6 @@ static int btrfs_trim_free_extents(struct btrfs_device *device,
break;
}
- /* Keep going until we satisfy minlen or reach end of space */
- if (len < range->minlen) {
- mutex_unlock(&fs_info->chunk_mutex);
- start += len;
- continue;
- }
-
- /* If we are out of the passed range break */
- if (start > range->start + range->len - 1) {
- mutex_unlock(&fs_info->chunk_mutex);
- break;
- }
-
- start = max(range->start, start);
- len = min(range->len, len);
-
ret = btrfs_issue_discard(device->bdev, start, len,
&bytes);
if (!ret)
@@ -11215,10 +11200,6 @@ static int btrfs_trim_free_extents(struct btrfs_device *device,
start += len;
*trimmed += bytes;
- /* We've trimmed enough */
- if (*trimmed >= range->len)
- break;
-
if (fatal_signal_pending(current)) {
ret = -ERESTARTSYS;
break;
@@ -11302,7 +11283,7 @@ int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range)
mutex_lock(&fs_info->fs_devices->device_list_mutex);
devices = &fs_info->fs_devices->devices;
list_for_each_entry(device, devices, dev_list) {
- ret = btrfs_trim_free_extents(device, range, &group_trimmed);
+ ret = btrfs_trim_free_extents(device, &group_trimmed);
if (ret) {
dev_failed++;
dev_ret = ret;
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 7e85dca0e6f2..89f5be2bfb43 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -2068,6 +2068,18 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
u64 len;
/*
+ * If the inode needs a full sync, make sure we use a full range to
+ * avoid log tree corruption, due to hole detection racing with ordered
+ * extent completion for adjacent ranges, and assertion failures during
+ * hole detection.
+ */
+ if (test_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
+ &BTRFS_I(inode)->runtime_flags)) {
+ start = 0;
+ end = LLONG_MAX;
+ }
+
+ /*
* The range length can be represented by u64, we have to do the typecasts
* to avoid signed overflow if it's [0, LLONG_MAX] eg. from fsync()
*/
@@ -2554,10 +2566,8 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
ret = btrfs_punch_hole_lock_range(inode, lockstart, lockend,
&cached_state);
- if (ret) {
- inode_unlock(inode);
+ if (ret)
goto out_only_mutex;
- }
path = btrfs_alloc_path();
if (!path) {
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 9aba9660efe5..a2aabdb85226 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -6433,8 +6433,18 @@ int btrfs_add_link(struct btrfs_trans_handle *trans,
btrfs_i_size_write(parent_inode, parent_inode->vfs_inode.i_size +
name_len * 2);
inode_inc_iversion(&parent_inode->vfs_inode);
- parent_inode->vfs_inode.i_mtime = parent_inode->vfs_inode.i_ctime =
- current_time(&parent_inode->vfs_inode);
+ /*
+ * If we are replaying a log tree, we do not want to update the mtime
+ * and ctime of the parent directory with the current time, since the
+ * log replay procedure is responsible for setting them to their correct
+ * values (the ones it had when the fsync was done).
+ */
+ if (!test_bit(BTRFS_FS_LOG_RECOVERING, &root->fs_info->flags)) {
+ struct timespec64 now = current_time(&parent_inode->vfs_inode);
+
+ parent_inode->vfs_inode.i_mtime = now;
+ parent_inode->vfs_inode.i_ctime = now;
+ }
ret = btrfs_update_inode(trans, root, &parent_inode->vfs_inode);
if (ret)
btrfs_abort_transaction(trans, ret);
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 6dafa857bbb9..cfeff1b8dce0 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -187,7 +187,7 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
struct btrfs_inode *binode = BTRFS_I(inode);
struct btrfs_root *root = binode->root;
struct btrfs_trans_handle *trans;
- unsigned int fsflags;
+ unsigned int fsflags, old_fsflags;
int ret;
const char *comp = NULL;
u32 binode_flags = binode->flags;
@@ -212,13 +212,10 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
inode_lock(inode);
fsflags = btrfs_mask_fsflags_for_type(inode, fsflags);
- if ((fsflags ^ btrfs_inode_flags_to_fsflags(binode->flags)) &
- (FS_APPEND_FL | FS_IMMUTABLE_FL)) {
- if (!capable(CAP_LINUX_IMMUTABLE)) {
- ret = -EPERM;
- goto out_unlock;
- }
- }
+ old_fsflags = btrfs_inode_flags_to_fsflags(binode->flags);
+ ret = vfs_ioc_setflags_prepare(inode, old_fsflags, fsflags);
+ if (ret)
+ goto out_unlock;
if (fsflags & FS_SYNC_FL)
binode_flags |= BTRFS_INODE_SYNC;
@@ -312,8 +309,6 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
btrfs_abort_transaction(trans, ret);
goto out_end_trans;
}
- set_bit(BTRFS_INODE_COPY_EVERYTHING,
- &BTRFS_I(inode)->runtime_flags);
} else {
ret = btrfs_set_prop(trans, inode, "btrfs.compression", NULL,
0, 0);
@@ -378,9 +373,7 @@ static int btrfs_ioctl_fsgetxattr(struct file *file, void __user *arg)
struct btrfs_inode *binode = BTRFS_I(file_inode(file));
struct fsxattr fa;
- memset(&fa, 0, sizeof(fa));
- fa.fsx_xflags = btrfs_inode_flags_to_xflags(binode->flags);
-
+ simple_fill_fsxattr(&fa, btrfs_inode_flags_to_xflags(binode->flags));
if (copy_to_user(arg, &fa, sizeof(fa)))
return -EFAULT;
@@ -393,7 +386,7 @@ static int btrfs_ioctl_fssetxattr(struct file *file, void __user *arg)
struct btrfs_inode *binode = BTRFS_I(inode);
struct btrfs_root *root = binode->root;
struct btrfs_trans_handle *trans;
- struct fsxattr fa;
+ struct fsxattr fa, old_fa;
unsigned old_flags;
unsigned old_i_flags;
int ret = 0;
@@ -404,7 +397,6 @@ static int btrfs_ioctl_fssetxattr(struct file *file, void __user *arg)
if (btrfs_root_readonly(root))
return -EROFS;
- memset(&fa, 0, sizeof(fa));
if (copy_from_user(&fa, arg, sizeof(fa)))
return -EFAULT;
@@ -424,13 +416,11 @@ static int btrfs_ioctl_fssetxattr(struct file *file, void __user *arg)
old_flags = binode->flags;
old_i_flags = inode->i_flags;
- /* We need the capabilities to change append-only or immutable inode */
- if (((old_flags & (BTRFS_INODE_APPEND | BTRFS_INODE_IMMUTABLE)) ||
- (fa.fsx_xflags & (FS_XFLAG_APPEND | FS_XFLAG_IMMUTABLE))) &&
- !capable(CAP_LINUX_IMMUTABLE)) {
- ret = -EPERM;
+ simple_fill_fsxattr(&old_fa,
+ btrfs_inode_flags_to_xflags(binode->flags));
+ ret = vfs_ioc_fssetxattr_check(inode, &old_fa, &fa);
+ if (ret)
goto out_unlock;
- }
if (fa.fsx_xflags & FS_XFLAG_SYNC)
binode->flags |= BTRFS_INODE_SYNC;
@@ -2930,8 +2920,10 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file,
inode_lock(inode);
err = btrfs_delete_subvolume(dir, dentry);
inode_unlock(inode);
- if (!err)
+ if (!err) {
+ fsnotify_rmdir(dir, dentry);
d_delete(dentry);
+ }
out_dput:
dput(dentry);
diff --git a/fs/btrfs/props.c b/fs/btrfs/props.c
index ca2716917e37..a9e2e66152ee 100644
--- a/fs/btrfs/props.c
+++ b/fs/btrfs/props.c
@@ -332,6 +332,7 @@ static int inherit_props(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info = root->fs_info;
int ret;
int i;
+ bool need_reserve = false;
if (!test_bit(BTRFS_INODE_HAS_PROPS,
&BTRFS_I(parent)->runtime_flags))
@@ -357,11 +358,20 @@ static int inherit_props(struct btrfs_trans_handle *trans,
if (ret)
continue;
- num_bytes = btrfs_calc_trans_metadata_size(fs_info, 1);
- ret = btrfs_block_rsv_add(root, trans->block_rsv,
- num_bytes, BTRFS_RESERVE_NO_FLUSH);
- if (ret)
- return ret;
+ /*
+ * Currently callers should be reserving 1 item for properties,
+ * since we only have 1 property that we currently support. If
+ * we add more in the future we need to try and reserve more
+ * space for them. But we should also revisit how we do space
+ * reservations if we do add more properties in the future.
+ */
+ if (need_reserve) {
+ num_bytes = btrfs_calc_trans_metadata_size(fs_info, 1);
+ ret = btrfs_block_rsv_add(root, trans->block_rsv,
+ num_bytes, BTRFS_RESERVE_NO_FLUSH);
+ if (ret)
+ return ret;
+ }
ret = btrfs_setxattr(trans, inode, h->xattr_name, value,
strlen(value), 0);
@@ -375,9 +385,13 @@ static int inherit_props(struct btrfs_trans_handle *trans,
&BTRFS_I(inode)->runtime_flags);
}
- btrfs_block_rsv_release(fs_info, trans->block_rsv, num_bytes);
- if (ret)
- return ret;
+ if (need_reserve) {
+ btrfs_block_rsv_release(fs_info, trans->block_rsv,
+ num_bytes);
+ if (ret)
+ return ret;
+ }
+ need_reserve = true;
}
return 0;
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index 2f708f2c4e67..3e6ffbbd8b0a 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -3830,7 +3830,13 @@ int btrfs_qgroup_add_swapped_blocks(struct btrfs_trans_handle *trans,
subvol_slot);
block->last_snapshot = last_snapshot;
block->level = level;
- if (bg->flags & BTRFS_BLOCK_GROUP_DATA)
+
+ /*
+ * If we have bg == NULL, we're called from btrfs_recover_relocation(),
+ * no one else can modify tree blocks thus we qgroup will not change
+ * no matter the value of trace_leaf.
+ */
+ if (bg && bg->flags & BTRFS_BLOCK_GROUP_DATA)
block->trace_leaf = true;
else
block->trace_leaf = false;
diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c
index 10d9589001a9..bb5bd49573b4 100644
--- a/fs/btrfs/reada.c
+++ b/fs/btrfs/reada.c
@@ -747,6 +747,7 @@ static void __reada_start_machine(struct btrfs_fs_info *fs_info)
u64 total = 0;
int i;
+again:
do {
enqueued = 0;
mutex_lock(&fs_devices->device_list_mutex);
@@ -758,6 +759,10 @@ static void __reada_start_machine(struct btrfs_fs_info *fs_info)
mutex_unlock(&fs_devices->device_list_mutex);
total += enqueued;
} while (enqueued && total < 10000);
+ if (fs_devices->seed) {
+ fs_devices = fs_devices->seed;
+ goto again;
+ }
if (enqueued == 0)
return;
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index a459ecddcce4..22a3c69864fa 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -2177,22 +2177,30 @@ static int clean_dirty_subvols(struct reloc_control *rc)
struct btrfs_root *root;
struct btrfs_root *next;
int ret = 0;
+ int ret2;
list_for_each_entry_safe(root, next, &rc->dirty_subvol_roots,
reloc_dirty_list) {
- struct btrfs_root *reloc_root = root->reloc_root;
+ if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID) {
+ /* Merged subvolume, cleanup its reloc root */
+ struct btrfs_root *reloc_root = root->reloc_root;
- clear_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state);
- list_del_init(&root->reloc_dirty_list);
- root->reloc_root = NULL;
- if (reloc_root) {
- int ret2;
+ clear_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state);
+ list_del_init(&root->reloc_dirty_list);
+ root->reloc_root = NULL;
+ if (reloc_root) {
- ret2 = btrfs_drop_snapshot(reloc_root, NULL, 0, 1);
+ ret2 = btrfs_drop_snapshot(reloc_root, NULL, 0, 1);
+ if (ret2 < 0 && !ret)
+ ret = ret2;
+ }
+ btrfs_put_fs_root(root);
+ } else {
+ /* Orphan reloc tree, just clean it up */
+ ret2 = btrfs_drop_snapshot(root, NULL, 0, 1);
if (ret2 < 0 && !ret)
ret = ret2;
}
- btrfs_put_fs_root(root);
}
return ret;
}
@@ -2480,6 +2488,9 @@ again:
}
} else {
list_del_init(&reloc_root->root_list);
+ /* Don't forget to queue this reloc root for cleanup */
+ list_add_tail(&reloc_root->reloc_dirty_list,
+ &rc->dirty_subvol_roots);
}
}
diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c
index 1b9a5d0de139..22124122728c 100644
--- a/fs/btrfs/root-tree.c
+++ b/fs/btrfs/root-tree.c
@@ -132,10 +132,8 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root
return -ENOMEM;
ret = btrfs_search_slot(trans, root, key, path, 0, 1);
- if (ret < 0) {
- btrfs_abort_transaction(trans, ret);
+ if (ret < 0)
goto out;
- }
if (ret > 0) {
btrfs_crit(fs_info,
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index dd38dfe174df..f7fe4770f0e5 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -4999,6 +4999,12 @@ static int send_hole(struct send_ctx *sctx, u64 end)
if (offset >= sctx->cur_inode_size)
return 0;
+ /*
+ * Don't go beyond the inode's i_size due to prealloc extents that start
+ * after the i_size.
+ */
+ end = min_t(u64, end, sctx->cur_inode_size);
+
if (sctx->flags & BTRFS_SEND_FLAG_NO_FILE_DATA)
return send_update_extent(sctx, offset, end - offset);
@@ -5218,10 +5224,50 @@ static int clone_range(struct send_ctx *sctx,
clone_len = min_t(u64, ext_len, len);
if (btrfs_file_extent_disk_bytenr(leaf, ei) == disk_byte &&
- clone_data_offset == data_offset)
- ret = send_clone(sctx, offset, clone_len, clone_root);
- else
+ clone_data_offset == data_offset) {
+ const u64 src_end = clone_root->offset + clone_len;
+ const u64 sectorsize = SZ_64K;
+
+ /*
+ * We can't clone the last block, when its size is not
+ * sector size aligned, into the middle of a file. If we
+ * do so, the receiver will get a failure (-EINVAL) when
+ * trying to clone or will silently corrupt the data in
+ * the destination file if it's on a kernel without the
+ * fix introduced by commit ac765f83f1397646
+ * ("Btrfs: fix data corruption due to cloning of eof
+ * block).
+ *
+ * So issue a clone of the aligned down range plus a
+ * regular write for the eof block, if we hit that case.
+ *
+ * Also, we use the maximum possible sector size, 64K,
+ * because we don't know what's the sector size of the
+ * filesystem that receives the stream, so we have to
+ * assume the largest possible sector size.
+ */
+ if (src_end == clone_src_i_size &&
+ !IS_ALIGNED(src_end, sectorsize) &&
+ offset + clone_len < sctx->cur_inode_size) {
+ u64 slen;
+
+ slen = ALIGN_DOWN(src_end - clone_root->offset,
+ sectorsize);
+ if (slen > 0) {
+ ret = send_clone(sctx, offset, slen,
+ clone_root);
+ if (ret < 0)
+ goto out;
+ }
+ ret = send_extent_data(sctx, offset + slen,
+ clone_len - slen);
+ } else {
+ ret = send_clone(sctx, offset, clone_len,
+ clone_root);
+ }
+ } else {
ret = send_extent_data(sctx, offset, clone_len);
+ }
if (ret < 0)
goto out;
diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
index 5a5930e3d32b..c1dfc97893ba 100644
--- a/fs/btrfs/sysfs.c
+++ b/fs/btrfs/sysfs.c
@@ -303,11 +303,12 @@ static ssize_t raid_bytes_show(struct kobject *kobj,
return snprintf(buf, PAGE_SIZE, "%llu\n", val);
}
-static struct attribute *raid_attributes[] = {
+static struct attribute *raid_attrs[] = {
BTRFS_ATTR_PTR(raid, total_bytes),
BTRFS_ATTR_PTR(raid, used_bytes),
NULL
};
+ATTRIBUTE_GROUPS(raid);
static void release_raid_kobj(struct kobject *kobj)
{
@@ -317,7 +318,7 @@ static void release_raid_kobj(struct kobject *kobj)
struct kobj_type btrfs_raid_ktype = {
.sysfs_ops = &kobj_sysfs_ops,
.release = release_raid_kobj,
- .default_attrs = raid_attributes,
+ .default_groups = raid_groups,
};
#define SPACE_INFO_ATTR(field) \
@@ -364,6 +365,7 @@ static struct attribute *space_info_attrs[] = {
BTRFS_ATTR_PTR(space_info, total_bytes_pinned),
NULL,
};
+ATTRIBUTE_GROUPS(space_info);
static void space_info_release(struct kobject *kobj)
{
@@ -375,7 +377,7 @@ static void space_info_release(struct kobject *kobj)
struct kobj_type space_info_ktype = {
.sysfs_ops = &kobj_sysfs_ops,
.release = space_info_release,
- .default_attrs = space_info_attrs,
+ .default_groups = space_info_groups,
};
static const struct attribute *allocation_attrs[] = {
@@ -825,7 +827,12 @@ int btrfs_sysfs_add_fsid(struct btrfs_fs_devices *fs_devs,
fs_devs->fsid_kobj.kset = btrfs_kset;
error = kobject_init_and_add(&fs_devs->fsid_kobj,
&btrfs_ktype, parent, "%pU", fs_devs->fsid);
- return error;
+ if (error) {
+ kobject_put(&fs_devs->fsid_kobj);
+ return error;
+ }
+
+ return 0;
}
int btrfs_sysfs_add_mounted(struct btrfs_fs_info *fs_info)
@@ -905,12 +912,10 @@ void btrfs_sysfs_feature_update(struct btrfs_fs_info *fs_info,
ret = sysfs_create_group(fsid_kobj, &btrfs_feature_attr_group);
}
-static int btrfs_init_debugfs(void)
+static void btrfs_init_debugfs(void)
{
#ifdef CONFIG_DEBUG_FS
btrfs_debugfs_root_dentry = debugfs_create_dir("btrfs", NULL);
- if (!btrfs_debugfs_root_dentry)
- return -ENOMEM;
/*
* Example code, how to export data through debugfs.
@@ -924,7 +929,6 @@ static int btrfs_init_debugfs(void)
#endif
#endif
- return 0;
}
int __init btrfs_init_sysfs(void)
@@ -935,9 +939,7 @@ int __init btrfs_init_sysfs(void)
if (!btrfs_kset)
return -ENOMEM;
- ret = btrfs_init_debugfs();
- if (ret)
- goto out1;
+ btrfs_init_debugfs();
init_feature_attrs();
ret = sysfs_create_group(&btrfs_kset->kobj, &btrfs_feature_attr_group);
@@ -954,7 +956,6 @@ out_remove_group:
sysfs_remove_group(&btrfs_kset->kobj, &btrfs_feature_attr_group);
out2:
debugfs_remove_recursive(btrfs_debugfs_root_dentry);
-out1:
kset_unregister(btrfs_kset);
return ret;
diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c
index 748cd1598255..96fce4bef4e7 100644
--- a/fs/btrfs/tree-checker.c
+++ b/fs/btrfs/tree-checker.c
@@ -107,8 +107,26 @@ static void file_extent_err(const struct extent_buffer *eb, int slot,
(!IS_ALIGNED(btrfs_file_extent_##name((leaf), (fi)), (alignment))); \
})
+static u64 file_extent_end(struct extent_buffer *leaf,
+ struct btrfs_key *key,
+ struct btrfs_file_extent_item *extent)
+{
+ u64 end;
+ u64 len;
+
+ if (btrfs_file_extent_type(leaf, extent) == BTRFS_FILE_EXTENT_INLINE) {
+ len = btrfs_file_extent_ram_bytes(leaf, extent);
+ end = ALIGN(key->offset + len, leaf->fs_info->sectorsize);
+ } else {
+ len = btrfs_file_extent_num_bytes(leaf, extent);
+ end = key->offset + len;
+ }
+ return end;
+}
+
static int check_extent_data_item(struct extent_buffer *leaf,
- struct btrfs_key *key, int slot)
+ struct btrfs_key *key, int slot,
+ struct btrfs_key *prev_key)
{
struct btrfs_fs_info *fs_info = leaf->fs_info;
struct btrfs_file_extent_item *fi;
@@ -188,6 +206,28 @@ static int check_extent_data_item(struct extent_buffer *leaf,
CHECK_FE_ALIGNED(leaf, slot, fi, offset, sectorsize) ||
CHECK_FE_ALIGNED(leaf, slot, fi, num_bytes, sectorsize))
return -EUCLEAN;
+
+ /*
+ * Check that no two consecutive file extent items, in the same leaf,
+ * present ranges that overlap each other.
+ */
+ if (slot > 0 &&
+ prev_key->objectid == key->objectid &&
+ prev_key->type == BTRFS_EXTENT_DATA_KEY) {
+ struct btrfs_file_extent_item *prev_fi;
+ u64 prev_end;
+
+ prev_fi = btrfs_item_ptr(leaf, slot - 1,
+ struct btrfs_file_extent_item);
+ prev_end = file_extent_end(leaf, prev_key, prev_fi);
+ if (prev_end > key->offset) {
+ file_extent_err(leaf, slot - 1,
+"file extent end range (%llu) goes beyond start offset (%llu) of the next file extent",
+ prev_end, key->offset);
+ return -EUCLEAN;
+ }
+ }
+
return 0;
}
@@ -774,14 +814,15 @@ static int check_inode_item(struct extent_buffer *leaf,
* Common point to switch the item-specific validation.
*/
static int check_leaf_item(struct extent_buffer *leaf,
- struct btrfs_key *key, int slot)
+ struct btrfs_key *key, int slot,
+ struct btrfs_key *prev_key)
{
int ret = 0;
struct btrfs_chunk *chunk;
switch (key->type) {
case BTRFS_EXTENT_DATA_KEY:
- ret = check_extent_data_item(leaf, key, slot);
+ ret = check_extent_data_item(leaf, key, slot, prev_key);
break;
case BTRFS_EXTENT_CSUM_KEY:
ret = check_csum_item(leaf, key, slot);
@@ -928,7 +969,7 @@ static int check_leaf(struct extent_buffer *leaf, bool check_item_data)
* Check if the item size and content meet other
* criteria
*/
- ret = check_leaf_item(leaf, &key, slot);
+ ret = check_leaf_item(leaf, &key, slot, &prev_key);
if (ret < 0)
return ret;
}
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 6adcd8a2c5c7..3fc8d854d7fb 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -3110,6 +3110,12 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
log->log_transid = root->log_transid;
root->log_start_pid = 0;
/*
+ * Update or create log root item under the root's log_mutex to prevent
+ * races with concurrent log syncs that can lead to failure to update
+ * log root item because it was not created yet.
+ */
+ ret = update_log_root(trans, log);
+ /*
* IO has been started, blocks of the log tree have WRITTEN flag set
* in their headers. new modifications of the log will be written to
* new positions. so it's safe to allow log writers to go in.
@@ -3128,8 +3134,6 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
mutex_unlock(&log_root_tree->log_mutex);
- ret = update_log_root(trans, log);
-
mutex_lock(&log_root_tree->log_mutex);
if (atomic_dec_and_test(&log_root_tree->log_writers)) {
/* atomic_dec_and_test implies a barrier */
@@ -4182,6 +4186,7 @@ fill_holes:
*last_extent, 0,
0, len, 0, len,
0, 0, 0);
+ *last_extent += len;
}
}
}
@@ -5477,7 +5482,6 @@ static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans,
{
int ret = 0;
struct dentry *old_parent = NULL;
- struct btrfs_inode *orig_inode = inode;
/*
* for regular files, if its inode is already on disk, we don't
@@ -5497,16 +5501,6 @@ static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans,
}
while (1) {
- /*
- * If we are logging a directory then we start with our inode,
- * not our parent's inode, so we need to skip setting the
- * logged_trans so that further down in the log code we don't
- * think this inode has already been logged.
- */
- if (inode != orig_inode)
- inode->logged_trans = trans->transid;
- smp_mb();
-
if (btrfs_must_commit_transaction(trans, inode)) {
ret = 1;
break;
@@ -6383,7 +6377,6 @@ void btrfs_record_unlink_dir(struct btrfs_trans_handle *trans,
* if this directory was already logged any new
* names for this file/dir will get recorded
*/
- smp_mb();
if (dir->logged_trans == trans->transid)
return;
diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c
index 78b6ba2029e8..95d9aebff2c4 100644
--- a/fs/btrfs/xattr.c
+++ b/fs/btrfs/xattr.c
@@ -213,6 +213,9 @@ int btrfs_setxattr(struct btrfs_trans_handle *trans, struct inode *inode,
}
out:
btrfs_free_path(path);
+ if (!ret)
+ set_bit(BTRFS_INODE_COPY_EVERYTHING,
+ &BTRFS_I(inode)->runtime_flags);
return ret;
}
@@ -236,7 +239,6 @@ int btrfs_setxattr_trans(struct inode *inode, const char *name,
inode_inc_iversion(inode);
inode->i_ctime = current_time(inode);
- set_bit(BTRFS_INODE_COPY_EVERYTHING, &BTRFS_I(inode)->runtime_flags);
ret = btrfs_update_inode(trans, root, inode);
BUG_ON(ret);
out:
@@ -388,8 +390,6 @@ static int btrfs_xattr_handler_set_prop(const struct xattr_handler *handler,
if (!ret) {
inode_inc_iversion(inode);
inode->i_ctime = current_time(inode);
- set_bit(BTRFS_INODE_COPY_EVERYTHING,
- &BTRFS_I(inode)->runtime_flags);
ret = btrfs_update_inode(trans, root, inode);
BUG_ON(ret);
}
diff --git a/fs/btrfs/zstd.c b/fs/btrfs/zstd.c
index a6ff07cf11d5..3837ca180d52 100644
--- a/fs/btrfs/zstd.c
+++ b/fs/btrfs/zstd.c
@@ -105,10 +105,10 @@ static void zstd_reclaim_timer_fn(struct timer_list *timer)
unsigned long reclaim_threshold = jiffies - ZSTD_BTRFS_RECLAIM_JIFFIES;
struct list_head *pos, *next;
- spin_lock(&wsm.lock);
+ spin_lock_bh(&wsm.lock);
if (list_empty(&wsm.lru_list)) {
- spin_unlock(&wsm.lock);
+ spin_unlock_bh(&wsm.lock);
return;
}
@@ -137,7 +137,7 @@ static void zstd_reclaim_timer_fn(struct timer_list *timer)
if (!list_empty(&wsm.lru_list))
mod_timer(&wsm.timer, jiffies + ZSTD_BTRFS_RECLAIM_JIFFIES);
- spin_unlock(&wsm.lock);
+ spin_unlock_bh(&wsm.lock);
}
/*
@@ -198,7 +198,7 @@ static void zstd_cleanup_workspace_manager(void)
struct workspace *workspace;
int i;
- spin_lock(&wsm.lock);
+ spin_lock_bh(&wsm.lock);
for (i = 0; i < ZSTD_BTRFS_MAX_LEVEL; i++) {
while (!list_empty(&wsm.idle_ws[i])) {
workspace = container_of(wsm.idle_ws[i].next,
@@ -208,7 +208,7 @@ static void zstd_cleanup_workspace_manager(void)
zstd_free_workspace(&workspace->list);
}
}
- spin_unlock(&wsm.lock);
+ spin_unlock_bh(&wsm.lock);
del_timer_sync(&wsm.timer);
}
@@ -230,7 +230,7 @@ static struct list_head *zstd_find_workspace(unsigned int level)
struct workspace *workspace;
int i = level - 1;
- spin_lock(&wsm.lock);
+ spin_lock_bh(&wsm.lock);
for_each_set_bit_from(i, &wsm.active_map, ZSTD_BTRFS_MAX_LEVEL) {
if (!list_empty(&wsm.idle_ws[i])) {
ws = wsm.idle_ws[i].next;
@@ -242,11 +242,11 @@ static struct list_head *zstd_find_workspace(unsigned int level)
list_del(&workspace->lru_list);
if (list_empty(&wsm.idle_ws[i]))
clear_bit(i, &wsm.active_map);
- spin_unlock(&wsm.lock);
+ spin_unlock_bh(&wsm.lock);
return ws;
}
}
- spin_unlock(&wsm.lock);
+ spin_unlock_bh(&wsm.lock);
return NULL;
}
@@ -305,7 +305,7 @@ static void zstd_put_workspace(struct list_head *ws)
{
struct workspace *workspace = list_to_workspace(ws);
- spin_lock(&wsm.lock);
+ spin_lock_bh(&wsm.lock);
/* A node is only taken off the lru if we are the corresponding level */
if (workspace->req_level == workspace->level) {
@@ -325,7 +325,7 @@ static void zstd_put_workspace(struct list_head *ws)
list_add(&workspace->list, &wsm.idle_ws[workspace->level - 1]);
workspace->req_level = 0;
- spin_unlock(&wsm.lock);
+ spin_unlock_bh(&wsm.lock);
if (workspace->level == ZSTD_BTRFS_MAX_LEVEL)
cond_wake_up(&wsm.wait);
Powered by cgit, Gruvboxed by Blaster4385.