static int btrfs_dev_replace_kthread(void *data)
{
struct btrfs_fs_info *fs_info = data;
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
u64 progress;
int ret;
progress = btrfs_dev_replace_progress(fs_info);
progress = div_u64(progress, 10);
btrfs_info_in_rcu(fs_info,
"continuing dev_replace from %s (devid %llu) to target %s @%u%%",
btrfs_dev_name(dev_replace->srcdev),
dev_replace->srcdev->devid,
btrfs_dev_name(dev_replace->tgtdev),
(unsigned int)progress);
ret = btrfs_scrub_dev(fs_info, dev_replace->srcdev->devid,
dev_replace->committed_cursor_left,
btrfs_device_get_total_bytes(dev_replace->srcdev),
&dev_replace->scrub_progress, 0, 1);
ret = btrfs_dev_replace_finishing(fs_info, ret);
WARN_ON(ret && ret != -ECANCELED);
clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags);
return 0;
}
void btrfs_dev_replace_status(struct btrfs_fs_info *fs_info,
struct btrfs_ioctl_dev_replace_args *args)
{
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
struct btrfs_device *srcdev;
btrfs_dev_replace_lock(dev_replace);
/* even if !dev_replace_is_valid, the values are good enough for
* the replace_status ioctl */
args->result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR;
args->status.replace_state = dev_replace->replace_state;
args->status.time_started = dev_replace->time_started;
args->status.time_stopped = dev_replace->time_stopped;
args->status.num_write_errors =
atomic64_read(&dev_replace->num_write_errors);
args->status.num_uncorrectable_read_errors =
atomic64_read(&dev_replace->num_uncorrectable_read_errors);
switch (dev_replace->replace_state) {
case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
args->status.progress_1000 = 0;
break;
case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
args->status.progress_1000 = 1000;
break;
case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
srcdev = dev_replace->srcdev;
args->status.progress_1000 = div64_u64(dev_replace->cursor_left,
div64_u64(btrfs_device_get_total_bytes(srcdev), 1000));
break;
}
btrfs_dev_replace_unlock(dev_replace);
}
static int btrfs_dev_replace_continue_on_mount(struct btrfs_fs_info *fs_info)
{
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
int ret;
ret = btrfs_scrub_dev(fs_info, dev_replace->srcdev->devid,
dev_replace->committed_cursor_left,
btrfs_device_get_total_bytes(dev_replace->srcdev),
&dev_replace->scrub_progress, 0, 1);
ret = btrfs_dev_replace_finishing(fs_info, ret);
WARN_ON(ret);
return 0;
}
/*
* Read progress of device replace status according to the state and last
* stored position. The value format is the same as for
* btrfs_dev_replace::progress_1000
*/
static u64 btrfs_dev_replace_progress(struct btrfs_fs_info *fs_info)
{
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
u64 ret = 0;
switch (dev_replace->replace_state) {
case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
ret = 0;
break;
case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
ret = 1000;
break;
case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
ret = div64_u64(dev_replace->cursor_left,
div_u64(btrfs_device_get_total_bytes(
dev_replace->srcdev), 1000));
break;
}
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,
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;
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;
}
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;
}
