static int init_dedup_info(struct btrfs_fs_info *fs_info, u16 type, u16 backend,
u64 blocksize, u64 limit)
{
struct btrfs_dedup_info *dedup_info;
int ret;
fs_info->dedup_info = kzalloc(sizeof(*dedup_info), GFP_NOFS);
if (!fs_info->dedup_info)
return -ENOMEM;
dedup_info = fs_info->dedup_info;
dedup_info->hash_type = type;
dedup_info->backend = backend;
dedup_info->blocksize = blocksize;
dedup_info->limit_nr = limit;
/* Only support SHA256 yet */
dedup_info->dedup_driver = crypto_alloc_shash("sha256", 0, 0);
if (IS_ERR(dedup_info->dedup_driver)) {
btrfs_err(fs_info, "failed to init sha256 driver");
ret = PTR_ERR(dedup_info->dedup_driver);
kfree(fs_info->dedup_info);
fs_info->dedup_info = NULL;
return ret;
}
dedup_info->hash_root = RB_ROOT;
dedup_info->bytenr_root = RB_ROOT;
dedup_info->current_nr = 0;
INIT_LIST_HEAD(&dedup_info->lru_list);
mutex_init(&dedup_info->lock);
return 0;
}
int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info,
struct btrfs_block_group_cache *block_group,
struct btrfs_path *path)
{
struct btrfs_root *root = fs_info->free_space_root;
struct btrfs_free_space_info *info;
struct btrfs_key key, found_key;
struct extent_buffer *leaf;
unsigned long *bitmap;
char *bitmap_cursor;
u64 start, end;
u64 bitmap_range, i;
u32 bitmap_size, flags, expected_extent_count;
u32 extent_count = 0;
int done = 0, nr;
int ret;
bitmap_size = free_space_bitmap_size(block_group->key.offset,
block_group->sectorsize);
bitmap = alloc_bitmap(bitmap_size);
if (!bitmap) {
ret = -ENOMEM;
goto out;
}
start = block_group->key.objectid;
end = block_group->key.objectid + block_group->key.offset;
key.objectid = end - 1;
key.type = (u8)-1;
key.offset = (u64)-1;
while (!done) {
ret = btrfs_search_prev_slot(trans, root, &key, path, -1, 1);
if (ret)
goto out;
leaf = path->nodes[0];
nr = 0;
path->slots[0]++;
while (path->slots[0] > 0) {
btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0] - 1);
if (found_key.type == BTRFS_FREE_SPACE_INFO_KEY) {
ASSERT(found_key.objectid == block_group->key.objectid);
ASSERT(found_key.offset == block_group->key.offset);
done = 1;
break;
} else if (found_key.type == BTRFS_FREE_SPACE_EXTENT_KEY) {
u64 first, last;
ASSERT(found_key.objectid >= start);
ASSERT(found_key.objectid < end);
ASSERT(found_key.objectid + found_key.offset <= end);
first = div_u64(found_key.objectid - start,
block_group->sectorsize);
last = div_u64(found_key.objectid + found_key.offset - start,
block_group->sectorsize);
bitmap_set(bitmap, first, last - first);
extent_count++;
nr++;
path->slots[0]--;
} else {
ASSERT(0);
}
}
ret = btrfs_del_items(trans, root, path, path->slots[0], nr);
if (ret)
goto out;
btrfs_release_path(path);
}
info = search_free_space_info(trans, fs_info, block_group, path, 1);
if (IS_ERR(info)) {
ret = PTR_ERR(info);
goto out;
}
leaf = path->nodes[0];
flags = btrfs_free_space_flags(leaf, info);
flags |= BTRFS_FREE_SPACE_USING_BITMAPS;
btrfs_set_free_space_flags(leaf, info, flags);
expected_extent_count = btrfs_free_space_extent_count(leaf, info);
btrfs_mark_buffer_dirty(leaf);
btrfs_release_path(path);
if (extent_count != expected_extent_count) {
btrfs_err(fs_info, "incorrect extent count for %llu; counted %u, expected %u",
block_group->key.objectid, extent_count,
expected_extent_count);
ASSERT(0);
ret = -EIO;
goto out;
}
bitmap_cursor = (char *)bitmap;
bitmap_range = block_group->sectorsize * BTRFS_FREE_SPACE_BITMAP_BITS;
i = start;
while (i < end) {
unsigned long ptr;
u64 extent_size;
u32 data_size;
extent_size = min(end - i, bitmap_range);
data_size = free_space_bitmap_size(extent_size,
block_group->sectorsize);
key.objectid = i;
key.type = BTRFS_FREE_SPACE_BITMAP_KEY;
key.offset = extent_size;
ret = btrfs_insert_empty_item(trans, root, path, &key,
data_size);
if (ret)
goto out;
leaf = path->nodes[0];
ptr = btrfs_item_ptr_offset(leaf, path->slots[0]);
write_extent_buffer(leaf, bitmap_cursor, ptr,
data_size);
btrfs_mark_buffer_dirty(leaf);
btrfs_release_path(path);
i += extent_size;
bitmap_cursor += data_size;
}
ret = 0;
out:
kvfree(bitmap);
if (ret)
btrfs_abort_transaction(trans, ret);
return ret;
}
int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info,
struct btrfs_block_group_cache *block_group,
struct btrfs_path *path)
{
struct btrfs_root *root = fs_info->free_space_root;
struct btrfs_free_space_info *info;
struct btrfs_key key, found_key;
struct extent_buffer *leaf;
unsigned long *bitmap;
u64 start, end;
/* Initialize to silence GCC. */
u64 extent_start = 0;
u64 offset;
u32 bitmap_size, flags, expected_extent_count;
int prev_bit = 0, bit, bitnr;
u32 extent_count = 0;
int done = 0, nr;
int ret;
bitmap_size = free_space_bitmap_size(block_group->key.offset,
block_group->sectorsize);
bitmap = alloc_bitmap(bitmap_size);
if (!bitmap) {
ret = -ENOMEM;
goto out;
}
start = block_group->key.objectid;
end = block_group->key.objectid + block_group->key.offset;
key.objectid = end - 1;
key.type = (u8)-1;
key.offset = (u64)-1;
while (!done) {
ret = btrfs_search_prev_slot(trans, root, &key, path, -1, 1);
if (ret)
goto out;
leaf = path->nodes[0];
nr = 0;
path->slots[0]++;
while (path->slots[0] > 0) {
btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0] - 1);
if (found_key.type == BTRFS_FREE_SPACE_INFO_KEY) {
ASSERT(found_key.objectid == block_group->key.objectid);
ASSERT(found_key.offset == block_group->key.offset);
done = 1;
break;
} else if (found_key.type == BTRFS_FREE_SPACE_BITMAP_KEY) {
unsigned long ptr;
char *bitmap_cursor;
u32 bitmap_pos, data_size;
ASSERT(found_key.objectid >= start);
ASSERT(found_key.objectid < end);
ASSERT(found_key.objectid + found_key.offset <= end);
bitmap_pos = div_u64(found_key.objectid - start,
block_group->sectorsize *
BITS_PER_BYTE);
bitmap_cursor = ((char *)bitmap) + bitmap_pos;
data_size = free_space_bitmap_size(found_key.offset,
block_group->sectorsize);
ptr = btrfs_item_ptr_offset(leaf, path->slots[0] - 1);
read_extent_buffer(leaf, bitmap_cursor, ptr,
data_size);
nr++;
path->slots[0]--;
} else {
ASSERT(0);
}
}
ret = btrfs_del_items(trans, root, path, path->slots[0], nr);
if (ret)
goto out;
btrfs_release_path(path);
}
info = search_free_space_info(trans, fs_info, block_group, path, 1);
if (IS_ERR(info)) {
ret = PTR_ERR(info);
goto out;
}
leaf = path->nodes[0];
flags = btrfs_free_space_flags(leaf, info);
flags &= ~BTRFS_FREE_SPACE_USING_BITMAPS;
btrfs_set_free_space_flags(leaf, info, flags);
expected_extent_count = btrfs_free_space_extent_count(leaf, info);
btrfs_mark_buffer_dirty(leaf);
btrfs_release_path(path);
offset = start;
bitnr = 0;
while (offset < end) {
bit = !!test_bit(bitnr, bitmap);
if (prev_bit == 0 && bit == 1) {
extent_start = offset;
} else if (prev_bit == 1 && bit == 0) {
key.objectid = extent_start;
key.type = BTRFS_FREE_SPACE_EXTENT_KEY;
key.offset = offset - extent_start;
ret = btrfs_insert_empty_item(trans, root, path, &key, 0);
if (ret)
goto out;
btrfs_release_path(path);
extent_count++;
}
prev_bit = bit;
offset += block_group->sectorsize;
bitnr++;
}
if (prev_bit == 1) {
key.objectid = extent_start;
key.type = BTRFS_FREE_SPACE_EXTENT_KEY;
key.offset = end - extent_start;
ret = btrfs_insert_empty_item(trans, root, path, &key, 0);
if (ret)
goto out;
btrfs_release_path(path);
extent_count++;
}
if (extent_count != expected_extent_count) {
btrfs_err(fs_info, "incorrect extent count for %llu; counted %u, expected %u",
block_group->key.objectid, extent_count,
expected_extent_count);
ASSERT(0);
ret = -EIO;
goto out;
}
ret = 0;
out:
kvfree(bitmap);
if (ret)
btrfs_abort_transaction(trans, ret);
return ret;
}
static int load_free_space_bitmaps(struct btrfs_caching_control *caching_ctl,
struct btrfs_path *path,
u32 expected_extent_count)
{
struct btrfs_block_group_cache *block_group;
struct btrfs_fs_info *fs_info;
struct btrfs_root *root;
struct btrfs_key key;
int prev_bit = 0, bit;
/* Initialize to silence GCC. */
u64 extent_start = 0;
u64 end, offset;
u64 total_found = 0;
u32 extent_count = 0;
int ret;
block_group = caching_ctl->block_group;
fs_info = block_group->fs_info;
root = fs_info->free_space_root;
end = block_group->key.objectid + block_group->key.offset;
while (1) {
ret = btrfs_next_item(root, path);
if (ret < 0)
goto out;
if (ret)
break;
btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
if (key.type == BTRFS_FREE_SPACE_INFO_KEY)
break;
ASSERT(key.type == BTRFS_FREE_SPACE_BITMAP_KEY);
ASSERT(key.objectid < end && key.objectid + key.offset <= end);
caching_ctl->progress = key.objectid;
offset = key.objectid;
while (offset < key.objectid + key.offset) {
bit = free_space_test_bit(block_group, path, offset);
if (prev_bit == 0 && bit == 1) {
extent_start = offset;
} else if (prev_bit == 1 && bit == 0) {
total_found += add_new_free_space(block_group,
fs_info,
extent_start,
offset);
if (total_found > CACHING_CTL_WAKE_UP) {
total_found = 0;
wake_up(&caching_ctl->wait);
}
extent_count++;
}
prev_bit = bit;
offset += block_group->sectorsize;
}
}
if (prev_bit == 1) {
total_found += add_new_free_space(block_group, fs_info,
extent_start, end);
extent_count++;
}
if (extent_count != expected_extent_count) {
btrfs_err(fs_info, "incorrect extent count for %llu; counted %u, expected %u",
block_group->key.objectid, extent_count,
expected_extent_count);
ASSERT(0);
ret = -EIO;
goto out;
}
caching_ctl->progress = (u64)-1;
ret = 0;
out:
return ret;
}
static int load_free_space_extents(struct btrfs_caching_control *caching_ctl,
struct btrfs_path *path,
u32 expected_extent_count)
{
struct btrfs_block_group_cache *block_group;
struct btrfs_fs_info *fs_info;
struct btrfs_root *root;
struct btrfs_key key;
u64 end;
u64 total_found = 0;
u32 extent_count = 0;
int ret;
block_group = caching_ctl->block_group;
fs_info = block_group->fs_info;
root = fs_info->free_space_root;
end = block_group->key.objectid + block_group->key.offset;
while (1) {
ret = btrfs_next_item(root, path);
if (ret < 0)
goto out;
if (ret)
break;
btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
if (key.type == BTRFS_FREE_SPACE_INFO_KEY)
break;
ASSERT(key.type == BTRFS_FREE_SPACE_EXTENT_KEY);
ASSERT(key.objectid < end && key.objectid + key.offset <= end);
caching_ctl->progress = key.objectid;
total_found += add_new_free_space(block_group, fs_info,
key.objectid,
key.objectid + key.offset);
if (total_found > CACHING_CTL_WAKE_UP) {
total_found = 0;
wake_up(&caching_ctl->wait);
}
extent_count++;
}
if (extent_count != expected_extent_count) {
btrfs_err(fs_info, "incorrect extent count for %llu; counted %u, expected %u",
block_group->key.objectid, extent_count,
expected_extent_count);
ASSERT(0);
ret = -EIO;
goto out;
}
caching_ctl->progress = (u64)-1;
ret = 0;
out:
return ret;
}
static int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
const char *tgtdev_name, u64 srcdevid, const char *srcdev_name,
int read_src)
{
struct btrfs_root *root = fs_info->dev_root;
struct btrfs_trans_handle *trans;
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
int ret;
struct btrfs_device *tgt_device = NULL;
struct btrfs_device *src_device = NULL;
bool need_unlock;
src_device = btrfs_find_device_by_devspec(fs_info, srcdevid,
srcdev_name);
if (IS_ERR(src_device))
return PTR_ERR(src_device);
if (btrfs_pinned_by_swapfile(fs_info, src_device)) {
btrfs_warn_in_rcu(fs_info,
"cannot replace device %s (devid %llu) due to active swapfile",
btrfs_dev_name(src_device), src_device->devid);
return -ETXTBSY;
}
ret = btrfs_init_dev_replace_tgtdev(fs_info, tgtdev_name,
src_device, &tgt_device);
if (ret)
return ret;
/*
* Here we commit the transaction to make sure commit_total_bytes
* of all the devices are updated.
*/
trans = btrfs_attach_transaction(root);
if (!IS_ERR(trans)) {
ret = btrfs_commit_transaction(trans);
if (ret)
return ret;
} else if (PTR_ERR(trans) != -ENOENT) {
return PTR_ERR(trans);
}
need_unlock = true;
down_write(&dev_replace->rwsem);
switch (dev_replace->replace_state) {
case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
break;
case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
ASSERT(0);
ret = BTRFS_IOCTL_DEV_REPLACE_RESULT_ALREADY_STARTED;
goto leave;
}
dev_replace->cont_reading_from_srcdev_mode = read_src;
WARN_ON(!src_device);
dev_replace->srcdev = src_device;
dev_replace->tgtdev = tgt_device;
btrfs_info_in_rcu(fs_info,
"dev_replace from %s (devid %llu) to %s started",
btrfs_dev_name(src_device),
src_device->devid,
rcu_str_deref(tgt_device->name));
/*
* from now on, the writes to the srcdev are all duplicated to
* go to the tgtdev as well (refer to btrfs_map_block()).
*/
dev_replace->replace_state = BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED;
dev_replace->time_started = ktime_get_real_seconds();
dev_replace->cursor_left = 0;
dev_replace->committed_cursor_left = 0;
dev_replace->cursor_left_last_write_of_item = 0;
dev_replace->cursor_right = 0;
dev_replace->is_valid = 1;
dev_replace->item_needs_writeback = 1;
atomic64_set(&dev_replace->num_write_errors, 0);
atomic64_set(&dev_replace->num_uncorrectable_read_errors, 0);
up_write(&dev_replace->rwsem);
need_unlock = false;
ret = btrfs_sysfs_add_device_link(tgt_device->fs_devices, tgt_device);
if (ret)
btrfs_err(fs_info, "kobj add dev failed %d", ret);
btrfs_wait_ordered_roots(fs_info, U64_MAX, 0, (u64)-1);
/* force writing the updated state information to disk */
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
need_unlock = true;
down_write(&dev_replace->rwsem);
dev_replace->replace_state =
BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED;
dev_replace->srcdev = NULL;
dev_replace->tgtdev = NULL;
goto leave;
}
ret = btrfs_commit_transaction(trans);
WARN_ON(ret);
/* the disk copy procedure reuses the scrub code */
ret = btrfs_scrub_dev(fs_info, src_device->devid, 0,
btrfs_device_get_total_bytes(src_device),
&dev_replace->scrub_progress, 0, 1);
ret = btrfs_dev_replace_finishing(fs_info, ret);
if (ret == -EINPROGRESS) {
ret = BTRFS_IOCTL_DEV_REPLACE_RESULT_SCRUB_INPROGRESS;
} else if (ret != -ECANCELED) {
WARN_ON(ret);
}
return ret;
leave:
if (need_unlock)
up_write(&dev_replace->rwsem);
btrfs_destroy_dev_replace_tgtdev(tgt_device);
return ret;
}
/*
* Initialize a new device for device replace target from a given source dev
* and path.
*
* Return 0 and new device in @device_out, otherwise return < 0
*/
static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
const char *device_path,
struct btrfs_device *srcdev,
struct btrfs_device **device_out)
{
struct btrfs_device *device;
struct block_device *bdev;
struct list_head *devices;
struct rcu_string *name;
u64 devid = BTRFS_DEV_REPLACE_DEVID;
int ret = 0;
*device_out = NULL;
if (fs_info->fs_devices->seeding) {
btrfs_err(fs_info, "the filesystem is a seed filesystem!");
return -EINVAL;
}
bdev = blkdev_get_by_path(device_path, FMODE_WRITE | FMODE_EXCL,
fs_info->bdev_holder);
if (IS_ERR(bdev)) {
btrfs_err(fs_info, "target device %s is invalid!", device_path);
return PTR_ERR(bdev);
}
filemap_write_and_wait(bdev->bd_inode->i_mapping);
devices = &fs_info->fs_devices->devices;
list_for_each_entry(device, devices, dev_list) {
if (device->bdev == bdev) {
btrfs_err(fs_info,
"target device is in the filesystem!");
ret = -EEXIST;
goto error;
}
}
if (i_size_read(bdev->bd_inode) <
btrfs_device_get_total_bytes(srcdev)) {
btrfs_err(fs_info,
"target device is smaller than source device!");
ret = -EINVAL;
goto error;
}
device = btrfs_alloc_device(NULL, &devid, NULL);
if (IS_ERR(device)) {
ret = PTR_ERR(device);
goto error;
}
name = rcu_string_strdup(device_path, GFP_KERNEL);
if (!name) {
btrfs_free_device(device);
ret = -ENOMEM;
goto error;
}
rcu_assign_pointer(device->name, name);
mutex_lock(&fs_info->fs_devices->device_list_mutex);
set_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
device->generation = 0;
device->io_width = fs_info->sectorsize;
device->io_align = fs_info->sectorsize;
device->sector_size = fs_info->sectorsize;
device->total_bytes = btrfs_device_get_total_bytes(srcdev);
device->disk_total_bytes = btrfs_device_get_disk_total_bytes(srcdev);
device->bytes_used = btrfs_device_get_bytes_used(srcdev);
device->commit_total_bytes = srcdev->commit_total_bytes;
device->commit_bytes_used = device->bytes_used;
device->fs_info = fs_info;
device->bdev = bdev;
set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state);
set_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state);
device->mode = FMODE_EXCL;
device->dev_stats_valid = 1;
set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE);
device->fs_devices = fs_info->fs_devices;
list_add(&device->dev_list, &fs_info->fs_devices->devices);
fs_info->fs_devices->num_devices++;
fs_info->fs_devices->open_devices++;
mutex_unlock(&fs_info->fs_devices->device_list_mutex);
*device_out = device;
return 0;
error:
blkdev_put(bdev, FMODE_EXCL);
return ret;
}
int btrfs_dev_replace_start(struct btrfs_root *root, char *tgtdev_name,
u64 srcdevid, char *srcdev_name, int read_src)
{
struct btrfs_trans_handle *trans;
struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
int ret;
struct btrfs_device *tgt_device = NULL;
struct btrfs_device *src_device = NULL;
/* the disk copy procedure reuses the scrub code */
mutex_lock(&fs_info->volume_mutex);
ret = btrfs_find_device_by_devspec(root, srcdevid,
srcdev_name, &src_device);
if (ret) {
mutex_unlock(&fs_info->volume_mutex);
return ret;
}
ret = btrfs_init_dev_replace_tgtdev(root, tgtdev_name,
src_device, &tgt_device);
mutex_unlock(&fs_info->volume_mutex);
if (ret)
return ret;
/*
* Here we commit the transaction to make sure commit_total_bytes
* of all the devices are updated.
*/
trans = btrfs_attach_transaction(root);
if (!IS_ERR(trans)) {
ret = btrfs_commit_transaction(trans, root);
if (ret)
return ret;
} else if (PTR_ERR(trans) != -ENOENT) {
return PTR_ERR(trans);
}
btrfs_dev_replace_lock(dev_replace, 1);
switch (dev_replace->replace_state) {
case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
break;
case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
ret = BTRFS_IOCTL_DEV_REPLACE_RESULT_ALREADY_STARTED;
goto leave;
}
dev_replace->cont_reading_from_srcdev_mode = read_src;
WARN_ON(!src_device);
dev_replace->srcdev = src_device;
WARN_ON(!tgt_device);
dev_replace->tgtdev = tgt_device;
btrfs_info_in_rcu(fs_info,
"dev_replace from %s (devid %llu) to %s started",
src_device->missing ? "<missing disk>" :
rcu_str_deref(src_device->name),
src_device->devid,
rcu_str_deref(tgt_device->name));
/*
* from now on, the writes to the srcdev are all duplicated to
* go to the tgtdev as well (refer to btrfs_map_block()).
*/
dev_replace->replace_state = BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED;
dev_replace->time_started = get_seconds();
dev_replace->cursor_left = 0;
dev_replace->committed_cursor_left = 0;
dev_replace->cursor_left_last_write_of_item = 0;
dev_replace->cursor_right = 0;
dev_replace->is_valid = 1;
dev_replace->item_needs_writeback = 1;
atomic64_set(&dev_replace->num_write_errors, 0);
atomic64_set(&dev_replace->num_uncorrectable_read_errors, 0);
btrfs_dev_replace_unlock(dev_replace, 1);
ret = btrfs_sysfs_add_device_link(tgt_device->fs_devices, tgt_device);
if (ret)
btrfs_err(fs_info, "kobj add dev failed %d\n", ret);
btrfs_wait_ordered_roots(root->fs_info, -1, 0, (u64)-1);
/* force writing the updated state information to disk */
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
btrfs_dev_replace_lock(dev_replace, 1);
goto leave;
}
ret = btrfs_commit_transaction(trans, root);
WARN_ON(ret);
/* the disk copy procedure reuses the scrub code */
ret = btrfs_scrub_dev(fs_info, src_device->devid, 0,
btrfs_device_get_total_bytes(src_device),
&dev_replace->scrub_progress, 0, 1);
ret = btrfs_dev_replace_finishing(fs_info, ret);
if (ret == -EINPROGRESS) {
ret = BTRFS_IOCTL_DEV_REPLACE_RESULT_SCRUB_INPROGRESS;
} else {
WARN_ON(ret);
}
return ret;
leave:
dev_replace->srcdev = NULL;
dev_replace->tgtdev = NULL;
btrfs_dev_replace_unlock(dev_replace, 1);
btrfs_destroy_dev_replace_tgtdev(fs_info, tgt_device);
return ret;
}
int btrfs_dev_replace_start(struct btrfs_root *root,
struct btrfs_ioctl_dev_replace_args *args)
{
struct btrfs_trans_handle *trans;
struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
int ret;
struct btrfs_device *tgt_device = NULL;
struct btrfs_device *src_device = NULL;
if (btrfs_fs_incompat(fs_info, RAID56)) {
btrfs_warn(fs_info, "dev_replace cannot yet handle RAID5/RAID6");
return -EINVAL;
}
switch (args->start.cont_reading_from_srcdev_mode) {
case BTRFS_IOCTL_DEV_REPLACE_CONT_READING_FROM_SRCDEV_MODE_ALWAYS:
case BTRFS_IOCTL_DEV_REPLACE_CONT_READING_FROM_SRCDEV_MODE_AVOID:
break;
default:
return -EINVAL;
}
if ((args->start.srcdevid == 0 && args->start.srcdev_name[0] == '\0') ||
args->start.tgtdev_name[0] == '\0')
return -EINVAL;
mutex_lock(&fs_info->volume_mutex);
ret = btrfs_init_dev_replace_tgtdev(root, args->start.tgtdev_name,
&tgt_device);
if (ret) {
btrfs_err(fs_info, "target device %s is invalid!",
args->start.tgtdev_name);
mutex_unlock(&fs_info->volume_mutex);
return -EINVAL;
}
ret = btrfs_dev_replace_find_srcdev(root, args->start.srcdevid,
args->start.srcdev_name,
&src_device);
mutex_unlock(&fs_info->volume_mutex);
if (ret) {
ret = -EINVAL;
goto leave_no_lock;
}
if (tgt_device->total_bytes < src_device->total_bytes) {
btrfs_err(fs_info, "target device is smaller than source device!");
ret = -EINVAL;
goto leave_no_lock;
}
btrfs_dev_replace_lock(dev_replace);
switch (dev_replace->replace_state) {
case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
break;
case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
args->result = BTRFS_IOCTL_DEV_REPLACE_RESULT_ALREADY_STARTED;
goto leave;
}
dev_replace->cont_reading_from_srcdev_mode =
args->start.cont_reading_from_srcdev_mode;
WARN_ON(!src_device);
dev_replace->srcdev = src_device;
WARN_ON(!tgt_device);
dev_replace->tgtdev = tgt_device;
printk_in_rcu(KERN_INFO
"BTRFS: dev_replace from %s (devid %llu) to %s started\n",
src_device->missing ? "<missing disk>" :
rcu_str_deref(src_device->name),
src_device->devid,
rcu_str_deref(tgt_device->name));
tgt_device->total_bytes = src_device->total_bytes;
tgt_device->disk_total_bytes = src_device->disk_total_bytes;
tgt_device->bytes_used = src_device->bytes_used;
/*
* from now on, the writes to the srcdev are all duplicated to
* go to the tgtdev as well (refer to btrfs_map_block()).
*/
dev_replace->replace_state = BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED;
dev_replace->time_started = get_seconds();
dev_replace->cursor_left = 0;
dev_replace->committed_cursor_left = 0;
dev_replace->cursor_left_last_write_of_item = 0;
dev_replace->cursor_right = 0;
dev_replace->is_valid = 1;
dev_replace->item_needs_writeback = 1;
args->result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR;
btrfs_dev_replace_unlock(dev_replace);
btrfs_wait_ordered_roots(root->fs_info, -1);
/* force writing the updated state information to disk */
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
btrfs_dev_replace_lock(dev_replace);
goto leave;
}
ret = btrfs_commit_transaction(trans, root);
WARN_ON(ret);
/* the disk copy procedure reuses the scrub code */
ret = btrfs_scrub_dev(fs_info, src_device->devid, 0,
src_device->total_bytes,
&dev_replace->scrub_progress, 0, 1);
ret = btrfs_dev_replace_finishing(root->fs_info, ret);
WARN_ON(ret);
return 0;
leave:
dev_replace->srcdev = NULL;
dev_replace->tgtdev = NULL;
btrfs_dev_replace_unlock(dev_replace);
leave_no_lock:
if (tgt_device)
btrfs_destroy_dev_replace_tgtdev(fs_info, tgt_device);
return ret;
}
int btrfs_dev_replace_start(struct btrfs_root *root,
struct btrfs_ioctl_dev_replace_args *args)
{
struct btrfs_trans_handle *trans;
struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
int ret;
struct btrfs_device *tgt_device = NULL;
struct btrfs_device *src_device = NULL;
switch (args->start.cont_reading_from_srcdev_mode) {
case BTRFS_IOCTL_DEV_REPLACE_CONT_READING_FROM_SRCDEV_MODE_ALWAYS:
case BTRFS_IOCTL_DEV_REPLACE_CONT_READING_FROM_SRCDEV_MODE_AVOID:
break;
default:
return -EINVAL;
}
if ((args->start.srcdevid == 0 && args->start.srcdev_name[0] == '\0') ||
args->start.tgtdev_name[0] == '\0')
return -EINVAL;
/*
* Here we commit the transaction to make sure commit_total_bytes
* of all the devices are updated.
*/
trans = btrfs_attach_transaction(root);
if (!IS_ERR(trans)) {
ret = btrfs_commit_transaction(trans, root);
if (ret)
return ret;
} else if (PTR_ERR(trans) != -ENOENT) {
return PTR_ERR(trans);
}
/* the disk copy procedure reuses the scrub code */
mutex_lock(&fs_info->volume_mutex);
ret = btrfs_dev_replace_find_srcdev(root, args->start.srcdevid,
args->start.srcdev_name,
&src_device);
if (ret) {
mutex_unlock(&fs_info->volume_mutex);
return ret;
}
ret = btrfs_init_dev_replace_tgtdev(root, args->start.tgtdev_name,
src_device, &tgt_device);
mutex_unlock(&fs_info->volume_mutex);
if (ret)
return ret;
btrfs_dev_replace_lock(dev_replace);
switch (dev_replace->replace_state) {
case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
break;
case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
args->result = BTRFS_IOCTL_DEV_REPLACE_RESULT_ALREADY_STARTED;
goto leave;
}
dev_replace->cont_reading_from_srcdev_mode =
args->start.cont_reading_from_srcdev_mode;
WARN_ON(!src_device);
dev_replace->srcdev = src_device;
WARN_ON(!tgt_device);
dev_replace->tgtdev = tgt_device;
ret = btrfs_kobj_add_device(tgt_device->fs_devices, tgt_device);
if (ret)
btrfs_err(root->fs_info, "kobj add dev failed %d\n", ret);
printk_in_rcu(KERN_INFO
"BTRFS: dev_replace from %s (devid %llu) to %s started\n",
src_device->missing ? "<missing disk>" :
rcu_str_deref(src_device->name),
src_device->devid,
rcu_str_deref(tgt_device->name));
/*
* from now on, the writes to the srcdev are all duplicated to
* go to the tgtdev as well (refer to btrfs_map_block()).
*/
dev_replace->replace_state = BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED;
dev_replace->time_started = get_seconds();
dev_replace->cursor_left = 0;
dev_replace->committed_cursor_left = 0;
dev_replace->cursor_left_last_write_of_item = 0;
dev_replace->cursor_right = 0;
dev_replace->is_valid = 1;
dev_replace->item_needs_writeback = 1;
args->result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR;
btrfs_dev_replace_unlock(dev_replace);
btrfs_wait_ordered_roots(root->fs_info, -1);
/* force writing the updated state information to disk */
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
btrfs_dev_replace_lock(dev_replace);
goto leave;
}
ret = btrfs_commit_transaction(trans, root);
WARN_ON(ret);
/* the disk copy procedure reuses the scrub code */
ret = btrfs_scrub_dev(fs_info, src_device->devid, 0,
btrfs_device_get_total_bytes(src_device),
&dev_replace->scrub_progress, 0, 1);
ret = btrfs_dev_replace_finishing(root->fs_info, ret);
/* don't warn if EINPROGRESS, someone else might be running scrub */
if (ret == -EINPROGRESS) {
args->result = BTRFS_IOCTL_DEV_REPLACE_RESULT_SCRUB_INPROGRESS;
ret = 0;
} else {
WARN_ON(ret);
}
return ret;
leave:
dev_replace->srcdev = NULL;
dev_replace->tgtdev = NULL;
btrfs_dev_replace_unlock(dev_replace);
btrfs_destroy_dev_replace_tgtdev(fs_info, tgt_device);
return ret;
}
