From 22098763a10d9c1340827fcf6edab66f153b27f0 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 13 Sep 2026 12:27:00 -0700 Subject: Merge tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace Pull tracing fixes from Steven Rostedt: - Don't destroy user event fields when removal fails User event fields are destroyed before the event is removed from visibility. But that can fail leaving the still visible event with no fields. Move the destroying of the fields to after the event is successfully removed from visibility. - Initialize function graph state is fork before calling copy_exec_state() For non-CLONE_VM forks, copy_exec_state() allocates a new task_exec_state. If that allocation fails, ftrace_graph_exit_task() will free the tasks ret_stack pointer. Since that pointer is still using the parent's ret_stack, it mistakenly frees the parent's pointer too. Call ftrace_graph_init() on the task first which will NULL out the new tasks's ret_stack and if the copy fails, it will not free anything. - Remove FGRAPH_MAX_INDEX The macro FGRAPH_MAX_INDEX was added but never used. Remove it. - Save ent_size in function graph printing of nested functions The function graph tracer needs to look at the next event to see if the next event is the return of the current function entry. If it is, it prints a single line: ktime_get(); Otherwise it prints it like a nested function: tick_nohz_irq_exit() { ktime_get(); kcpustat_irq_exit(); } In order to look at the next event, it must save the current event so that it has the information to print from it. It saves the event in the iterator descriptor called "ent". What it doesn't save is the ent_size of the event which is now used to know if the function graph arguments are to be printed. The peek doesn't save the size so the size used happens to be that of the size of the last event that was seen. Save the entry event size in the iterator descriptor so that the correct size is used. - Fix several errors with freeing data in the histogram code The histogram code had a lot of leaked or or incorrect accounting when failures happen. Correct them. - Fix histogram regression of .percent and .graph modifiers Up until 6.3 histogram values could have "percent" or "graph" modifiers that changed how they were printed. But a change that added restricting histograms values from being strings, stack traces and other modifiers inadvertently prevented them from using the percent and graph modifiers, which were legal use cases for values. Put back the percent and graph modifiers. - Fix various typos in the comments - Set the trace_clock before initializing a histogram with clock argument The histogram API allows the user to specific which trace clock to use via a "clock=" string. The histogram is set up first before the clock is checked. If the passed in clock is not valid, it exits without fully fixing up the histogram leaving it on the list and a use-after-free can trigger. Update the clock argument first and if it fails then exit gracefully before the histogram trigger is placed on any lists. - Restore :mod: trailer after parsing in ftrace_set_clr_event The function ftrace_set_clr_event() modifies the parse string and needs to put it back to what was passed in. It searches for ":mod:" via a strsep() but fails to put back the first ':' in the string. Add back the ':' in the passed in string. - Take trace_array reference when opening a tracer options file The options files are dynamically created and some tracers add their own options. When a tracer adds their own list of options, the trace_array holding them has an array to hold the list of options for each tracer. This array increases in size via a krealloc(), and the new entry gets a newly allocated array to hold the options of the new tracer being added. The element in each entry of the tracer's option array holds a pointer back to the trace_array, a pointer to the tracer it is associated to, a pointer to the flags of the option. The issue is that these arrays are freed when the trace_array is freed when its instance it represents is removed from the instances directory. There's a race that an open of one of these options files can happen when the instance is being removed. Add a new helper function to be called by the open function of the options file to iterate all existing trace_arrays under a lock and find the one that has the given option element in one of it's tracer arrays. If found, then update the associated trace_array's reference counter to keep it from being freed. If not found, have the open call return -ENODEV. - Disable interrupts when acquiring the lock in rb_wake_up_waiters() The function rb_wake_up_waiters() assumes it will be called in interrupt context and does not disable irqs when taking cpu_buffer->reader_lock, which can be called in hard interrupt context. The issue is in PREEMPT_RT, this function is called in thread context leaving this lock open to a deadlock. Take the lock with interrupts disabled. - Use rcu_assign_pointer() for tmp_ops filter hash The tmp_ops used in update_ftrace_direct_mod() assigns its filter_hash field directly, but that field is annotated as __rcu and sparse complains. Assign it with rcu_assign_pointer() - Fix use-after-free in enable_trigger_private_data_free() The trace_event_call is accessed through the event_trigger_data's trace_event_file pointer to put the trace_event_call on freeing. The issue is that the trace_event_file data may have been freed already causing a use-after-free. Add a field to the event_trigger_data that points directly to the trace_event_call so that it can decrement its reference directly without needing to go through the trace_event_file. - Fix accounting of buffer data remote headers trace_buffer_desc_size() and trace_remote_alloc_buffer() undercount the number of pages is needed for the asked for size as it doesn't take into account the meta data on each page. Add a helper function to do the calculation properly and use that in these functions. - Catch nr_page_va overflow in ring_buffer_desc sizing The number of pages per remote ring buffer is capped by ring_buffer_desc::nr_page_va (32 bits). A buffer_size large enough to overflow that field would silently allocate a descriptor smaller than what was asked for. - Do not resize the subbuf order if any per_cpu buffer is disabled The mmapping of ring buffers disables resizing the subbuffers, but it is done per-cpu whereas the subbuf size change is done for all the per_cpu buffers under the buffer->mutex. It could change the size of some while the mapping is happening on others. Have the resize of the subbuf order check all the per_cpu buffers under the lock to see if any of them is disabled before starting and causing an inconsistency between buffers that are being mapped. * tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: (25 commits) ring-buffer: Check resize_disabled before publishing the new subbuf order tracing/remotes: Catch nr_page_va overflow in ring_buffer_desc sizing tracing/remotes: Account for ring buffer page header in size calculation tracing: Don't dereference trace_event_file in deferred trigger free ftrace: Use rcu_assign_pointer() for tmp_ops filter hash ring-buffer: Acquire the lock with irqsave in rb_wake_up_waiters() tracing: Take trace_array reference when opening a tracer options file tracing: Fix ring_buffer_read_page_size() kernel-doc tracing: Restore :mod: trailer after parsing in ftrace_set_clr_event() tracing: Fix memory corruption from a "STACKTRACE" histogram key tracing: Fix memory corruption from the histogram stacktrace modifier tracing: Undo the registration when enabling the histogram trigger fails tracing: Take the reference before publishing the named histogram trigger tracing: Set the trace clock before registering the histogram trigger tracing: Fix typo "preceeded" in comment tracing: Fix typo "availabe" in comment tracing: Let histogram values keep the percent and graph modifiers tracing: Keep the entry count when the histogram stats allocation fails tracing: Free histogram the field rejected for a bad modifier tracing: Free histogram the var ref when its initialization fails ... --- fs/btrfs/dev-replace.c | 1372 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 1372 insertions(+) create mode 100644 fs/btrfs/dev-replace.c (limited to 'fs/btrfs/dev-replace.c') diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c new file mode 100644 index 000000000..af1b89802 --- /dev/null +++ b/fs/btrfs/dev-replace.c @@ -0,0 +1,1372 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) STRATO AG 2012. All rights reserved. + */ + +#include +#include +#include +#include +#include +#include +#include "misc.h" +#include "ctree.h" +#include "disk-io.h" +#include "transaction.h" +#include "volumes.h" +#include "async-thread.h" +#include "dev-replace.h" +#include "sysfs.h" +#include "zoned.h" +#include "block-group.h" +#include "fs.h" +#include "accessors.h" +#include "scrub.h" + +/* + * Device replace overview + * + * [Objective] + * To copy all extents (both new and on-disk) from source device to target + * device, while still keeping the filesystem read-write. + * + * [Method] + * There are two main methods involved: + * + * - Write duplication + * + * All new writes will be written to both target and source devices, so even + * if replace gets canceled, sources device still contains up-to-date data. + * + * Location: handle_ops_on_dev_replace() from btrfs_map_block() + * Start: btrfs_dev_replace_start() + * End: btrfs_dev_replace_finishing() + * Content: Latest data/metadata + * + * - Copy existing extents + * + * This happens by reusing scrub facility, as scrub also iterates through + * existing extents from commit root. + * + * Location: scrub_write_block_to_dev_replace() from + * scrub_block_complete() + * Content: Data/meta from commit root. + * + * Due to the content difference, we need to avoid nocow write when dev-replace + * is happening. This is done by marking the block group read-only and waiting + * for NOCOW writes. + * + * After replace is done, the finishing part is done by swapping the target and + * source devices. + * + * Location: btrfs_dev_replace_update_device_in_mapping_tree() from + * btrfs_dev_replace_finishing() + */ + +static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info, + int scrub_ret); +static int btrfs_dev_replace_kthread(void *data); + +int btrfs_init_dev_replace(struct btrfs_fs_info *fs_info) +{ + struct btrfs_dev_lookup_args args = { .devid = BTRFS_DEV_REPLACE_DEVID }; + struct btrfs_key key; + struct btrfs_root *dev_root = fs_info->dev_root; + struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; + struct extent_buffer *eb; + int slot; + int ret = 0; + BTRFS_PATH_AUTO_FREE(path); + int item_size; + struct btrfs_dev_replace_item *ptr; + u64 src_devid; + + if (!dev_root) + return 0; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + key.objectid = 0; + key.type = BTRFS_DEV_REPLACE_KEY; + key.offset = 0; + ret = btrfs_search_slot(NULL, dev_root, &key, path, 0, 0); + if (ret) { +no_valid_dev_replace_entry_found: + /* + * We don't have a replace item or it's corrupted. If there is + * a replace target, fail the mount. + */ + if (unlikely(btrfs_find_device(fs_info->fs_devices, &args))) { + btrfs_err(fs_info, + "found replace target device without a valid replace item"); + return -EUCLEAN; + } + dev_replace->replace_state = + BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED; + dev_replace->cont_reading_from_srcdev_mode = + BTRFS_DEV_REPLACE_ITEM_CONT_READING_FROM_SRCDEV_MODE_ALWAYS; + dev_replace->time_started = 0; + dev_replace->time_stopped = 0; + atomic64_set(&dev_replace->num_write_errors, 0); + atomic64_set(&dev_replace->num_uncorrectable_read_errors, 0); + 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->srcdev = NULL; + dev_replace->tgtdev = NULL; + dev_replace->is_valid = 0; + dev_replace->item_needs_writeback = 0; + return 0; + } + slot = path->slots[0]; + eb = path->nodes[0]; + item_size = btrfs_item_size(eb, slot); + ptr = btrfs_item_ptr(eb, slot, struct btrfs_dev_replace_item); + + if (item_size != sizeof(struct btrfs_dev_replace_item)) { + btrfs_warn(fs_info, + "dev_replace entry found has unexpected size, ignore entry"); + goto no_valid_dev_replace_entry_found; + } + + src_devid = btrfs_dev_replace_src_devid(eb, ptr); + dev_replace->cont_reading_from_srcdev_mode = + btrfs_dev_replace_cont_reading_from_srcdev_mode(eb, ptr); + dev_replace->replace_state = btrfs_dev_replace_replace_state(eb, ptr); + dev_replace->time_started = btrfs_dev_replace_time_started(eb, ptr); + dev_replace->time_stopped = + btrfs_dev_replace_time_stopped(eb, ptr); + atomic64_set(&dev_replace->num_write_errors, + btrfs_dev_replace_num_write_errors(eb, ptr)); + atomic64_set(&dev_replace->num_uncorrectable_read_errors, + btrfs_dev_replace_num_uncorrectable_read_errors(eb, ptr)); + dev_replace->cursor_left = btrfs_dev_replace_cursor_left(eb, ptr); + dev_replace->committed_cursor_left = dev_replace->cursor_left; + dev_replace->cursor_left_last_write_of_item = dev_replace->cursor_left; + dev_replace->cursor_right = btrfs_dev_replace_cursor_right(eb, ptr); + dev_replace->is_valid = 1; + + dev_replace->item_needs_writeback = 0; + 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: + /* + * We don't have an active replace item but if there is a + * replace target, fail the mount. + */ + if (unlikely(btrfs_find_device(fs_info->fs_devices, &args))) { + btrfs_err(fs_info, +"replace without active item, run 'device scan --forget' on the target device"); + ret = -EUCLEAN; + } else { + dev_replace->srcdev = NULL; + dev_replace->tgtdev = NULL; + } + break; + case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED: + case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED: + dev_replace->tgtdev = btrfs_find_device(fs_info->fs_devices, &args); + args.devid = src_devid; + dev_replace->srcdev = btrfs_find_device(fs_info->fs_devices, &args); + + /* + * allow 'btrfs dev replace_cancel' if src/tgt device is + * missing + */ + if (unlikely(!dev_replace->srcdev && !btrfs_test_opt(fs_info, DEGRADED))) { + ret = -EIO; + btrfs_warn(fs_info, + "cannot mount because device replace operation is ongoing and"); + btrfs_warn(fs_info, + "srcdev (devid %llu) is missing, need to run 'btrfs dev scan'?", + src_devid); + } + if (unlikely(!dev_replace->tgtdev && !btrfs_test_opt(fs_info, DEGRADED))) { + ret = -EIO; + btrfs_warn(fs_info, + "cannot mount because device replace operation is ongoing and"); + btrfs_warn(fs_info, + "tgtdev (devid %llu) is missing, need to run 'btrfs dev scan'?", + BTRFS_DEV_REPLACE_DEVID); + } + if (dev_replace->tgtdev) { + if (dev_replace->srcdev) { + dev_replace->tgtdev->total_bytes = + dev_replace->srcdev->total_bytes; + dev_replace->tgtdev->disk_total_bytes = + dev_replace->srcdev->disk_total_bytes; + dev_replace->tgtdev->commit_total_bytes = + dev_replace->srcdev->commit_total_bytes; + dev_replace->tgtdev->bytes_used = + dev_replace->srcdev->bytes_used; + dev_replace->tgtdev->commit_bytes_used = + dev_replace->srcdev->commit_bytes_used; + } + set_bit(BTRFS_DEV_STATE_REPLACE_TGT, + &dev_replace->tgtdev->dev_state); + + WARN_ON(fs_info->fs_devices->rw_devices == 0); + dev_replace->tgtdev->io_width = fs_info->sectorsize; + dev_replace->tgtdev->io_align = fs_info->sectorsize; + dev_replace->tgtdev->sector_size = fs_info->sectorsize; + dev_replace->tgtdev->fs_info = fs_info; + set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, + &dev_replace->tgtdev->dev_state); + } + break; + } + + 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_fs_devices *fs_devices = fs_info->fs_devices; + struct btrfs_device *device; + struct file *bdev_file; + struct block_device *bdev; + u64 devid = BTRFS_DEV_REPLACE_DEVID; + int ret = 0; + + *device_out = NULL; + if (srcdev->fs_devices->seeding) { + btrfs_err(fs_info, "the filesystem is a seed filesystem!"); + return -EINVAL; + } + + /* Unfreezable for the whole replace; see btrfs_dev_replace_start(). */ + bdev_file = btrfs_open_device_deny_freeze(device_path, fs_info->sb); + if (IS_ERR(bdev_file)) { + btrfs_err(fs_info, "target device %s is invalid!", device_path); + return PTR_ERR(bdev_file); + } + bdev = file_bdev(bdev_file); + + if (!btrfs_check_device_zone_type(fs_info, bdev)) { + btrfs_err(fs_info, + "dev-replace: zoned type of target device mismatch with filesystem"); + ret = -EINVAL; + goto error; + } + + sync_blockdev(bdev); + + list_for_each_entry(device, &fs_devices->devices, dev_list) { + if (device->bdev == bdev) { + btrfs_err(fs_info, + "target device is in the filesystem!"); + ret = -EEXIST; + goto error; + } + } + + + if (bdev_nr_bytes(bdev) < 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, device_path); + if (IS_ERR(device)) { + ret = PTR_ERR(device); + goto error; + } + + ret = lookup_bdev(device_path, &device->devt); + if (ret) + goto error; + + 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; + device->bdev_file = bdev_file; + set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state); + set_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state); + /* Check the comment in btrfs_init_new_device() for the reason. */ + atomic_inc(&device->dev_stats_ccnt); + device->dev_stats_valid = 1; + set_blocksize(bdev_file, BTRFS_BDEV_BLOCKSIZE); + device->fs_devices = fs_devices; + + ret = btrfs_get_dev_zone_info(device, false); + if (ret) + goto error; + + mutex_lock(&fs_devices->device_list_mutex); + list_add(&device->dev_list, &fs_devices->devices); + fs_devices->num_devices++; + fs_devices->open_devices++; + mutex_unlock(&fs_devices->device_list_mutex); + + *device_out = device; + return 0; + +error: + /* Undo the open-time freeze deny. */ + btrfs_release_device_allow_freeze(bdev_file); + return ret; +} + +/* + * called from commit_transaction. Writes changed device replace state to + * disk. + */ +int btrfs_run_dev_replace(struct btrfs_trans_handle *trans) +{ + struct btrfs_fs_info *fs_info = trans->fs_info; + int ret; + struct btrfs_root *dev_root = fs_info->dev_root; + BTRFS_PATH_AUTO_FREE(path); + struct btrfs_key key; + struct extent_buffer *eb; + struct btrfs_dev_replace_item *ptr; + struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; + + down_read(&dev_replace->rwsem); + if (!dev_replace->is_valid || + !dev_replace->item_needs_writeback) { + up_read(&dev_replace->rwsem); + return 0; + } + up_read(&dev_replace->rwsem); + + key.objectid = 0; + key.type = BTRFS_DEV_REPLACE_KEY; + key.offset = 0; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + ret = btrfs_search_slot(trans, dev_root, &key, path, -1, 1); + if (ret < 0) { + btrfs_warn(fs_info, + "error %d while searching for dev_replace item!", + ret); + return ret; + } + + if (ret == 0 && + btrfs_item_size(path->nodes[0], path->slots[0]) < sizeof(*ptr)) { + /* + * need to delete old one and insert a new one. + * Since no attempt is made to recover any old state, if the + * dev_replace state is 'running', the data on the target + * drive is lost. + * It would be possible to recover the state: just make sure + * that the beginning of the item is never changed and always + * contains all the essential information. Then read this + * minimal set of information and use it as a base for the + * new state. + */ + ret = btrfs_del_item(trans, dev_root, path); + if (ret != 0) { + btrfs_warn(fs_info, + "delete too small dev_replace item failed %d!", + ret); + return ret; + } + ret = 1; + } + + if (ret == 1) { + /* need to insert a new item */ + btrfs_release_path(path); + ret = btrfs_insert_empty_item(trans, dev_root, path, + &key, sizeof(*ptr)); + if (ret < 0) { + btrfs_warn(fs_info, + "insert dev_replace item failed %d!", ret); + return ret; + } + } + + eb = path->nodes[0]; + ptr = btrfs_item_ptr(eb, path->slots[0], + struct btrfs_dev_replace_item); + + down_write(&dev_replace->rwsem); + if (dev_replace->srcdev) + btrfs_set_dev_replace_src_devid(eb, ptr, + dev_replace->srcdev->devid); + else + btrfs_set_dev_replace_src_devid(eb, ptr, (u64)-1); + btrfs_set_dev_replace_cont_reading_from_srcdev_mode(eb, ptr, + dev_replace->cont_reading_from_srcdev_mode); + btrfs_set_dev_replace_replace_state(eb, ptr, + dev_replace->replace_state); + btrfs_set_dev_replace_time_started(eb, ptr, dev_replace->time_started); + btrfs_set_dev_replace_time_stopped(eb, ptr, dev_replace->time_stopped); + btrfs_set_dev_replace_num_write_errors(eb, ptr, + atomic64_read(&dev_replace->num_write_errors)); + btrfs_set_dev_replace_num_uncorrectable_read_errors(eb, ptr, + atomic64_read(&dev_replace->num_uncorrectable_read_errors)); + dev_replace->cursor_left_last_write_of_item = + dev_replace->cursor_left; + btrfs_set_dev_replace_cursor_left(eb, ptr, + dev_replace->cursor_left_last_write_of_item); + btrfs_set_dev_replace_cursor_right(eb, ptr, + dev_replace->cursor_right); + dev_replace->item_needs_writeback = 0; + up_write(&dev_replace->rwsem); + + return ret; +} + +static int mark_block_group_to_copy(struct btrfs_fs_info *fs_info, + struct btrfs_device *src_dev) +{ + struct btrfs_path *path; + struct btrfs_key key; + struct btrfs_key found_key; + struct btrfs_root *root = fs_info->dev_root; + struct btrfs_dev_extent *dev_extent = NULL; + struct btrfs_block_group *cache; + struct btrfs_trans_handle *trans; + int iter_ret = 0; + int ret = 0; + u64 chunk_offset; + + /* Do not use "to_copy" on non zoned filesystem for now */ + if (!btrfs_is_zoned(fs_info)) + return 0; + + mutex_lock(&fs_info->chunk_mutex); + + /* Ensure we don't have pending new block group */ + spin_lock(&fs_info->trans_lock); + while (fs_info->running_transaction && + !list_empty(&fs_info->running_transaction->dev_update_list)) { + spin_unlock(&fs_info->trans_lock); + mutex_unlock(&fs_info->chunk_mutex); + trans = btrfs_attach_transaction(root); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + mutex_lock(&fs_info->chunk_mutex); + if (ret == -ENOENT) { + spin_lock(&fs_info->trans_lock); + continue; + } else { + goto unlock; + } + } + + ret = btrfs_commit_transaction(trans); + mutex_lock(&fs_info->chunk_mutex); + if (ret) + goto unlock; + + spin_lock(&fs_info->trans_lock); + } + spin_unlock(&fs_info->trans_lock); + + path = btrfs_alloc_path(); + if (!path) { + ret = -ENOMEM; + goto unlock; + } + + path->reada = READA_FORWARD; + path->search_commit_root = true; + path->skip_locking = true; + + key.objectid = src_dev->devid; + key.type = BTRFS_DEV_EXTENT_KEY; + key.offset = 0; + + btrfs_for_each_slot(root, &key, &found_key, path, iter_ret) { + struct extent_buffer *leaf = path->nodes[0]; + + if (found_key.objectid != src_dev->devid) + break; + + if (found_key.type != BTRFS_DEV_EXTENT_KEY) + break; + + if (found_key.offset < key.offset) + break; + + dev_extent = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_dev_extent); + + chunk_offset = btrfs_dev_extent_chunk_offset(leaf, dev_extent); + + cache = btrfs_lookup_block_group(fs_info, chunk_offset); + if (!cache) + continue; + + set_bit(BLOCK_GROUP_FLAG_TO_COPY, &cache->runtime_flags); + btrfs_put_block_group(cache); + } + if (iter_ret < 0) + ret = iter_ret; + + btrfs_free_path(path); +unlock: + mutex_unlock(&fs_info->chunk_mutex); + + return ret; +} + +bool btrfs_finish_block_group_to_copy(struct btrfs_device *srcdev, + struct btrfs_block_group *cache, + u64 physical) +{ + struct btrfs_fs_info *fs_info = cache->fs_info; + struct btrfs_chunk_map *map; + u64 chunk_offset = cache->start; + int num_extents, cur_extent; + int i; + + /* Do not use "to_copy" on non zoned filesystem for now */ + if (!btrfs_is_zoned(fs_info)) + return true; + + spin_lock(&cache->lock); + if (test_bit(BLOCK_GROUP_FLAG_REMOVED, &cache->runtime_flags)) { + spin_unlock(&cache->lock); + return true; + } + spin_unlock(&cache->lock); + + map = btrfs_get_chunk_map(fs_info, chunk_offset, 1); + ASSERT(!IS_ERR(map)); + + num_extents = 0; + cur_extent = 0; + for (i = 0; i < map->num_stripes; i++) { + /* We have more device extent to copy */ + if (srcdev != map->stripes[i].dev) + continue; + + num_extents++; + if (physical == map->stripes[i].physical) + cur_extent = i; + } + + btrfs_free_chunk_map(map); + + if (num_extents > 1 && cur_extent < num_extents - 1) { + /* + * Has more stripes on this device. Keep this block group + * readonly until we finish all the stripes. + */ + return false; + } + + /* Last stripe on this device */ + clear_bit(BLOCK_GROUP_FLAG_TO_COPY, &cache->runtime_flags); + + return true; +} + +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; + + 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(fs_info, + "cannot replace device %s (devid %llu) due to active swapfile", + btrfs_dev_name(src_device), src_device->devid); + return -ETXTBSY; + } + + /* + * 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); + } + + ret = btrfs_init_dev_replace_tgtdev(fs_info, tgtdev_name, + src_device, &tgt_device); + if (ret) + return ret; + + /* Deny the source before mark, so every 'leave' unwinds both denied. */ + if (src_device->bdev) { + ret = bdev_deny_freeze(src_device->bdev); + if (ret) { + btrfs_destroy_dev_replace_tgtdev(tgt_device, true); + return ret; + } + } + + ret = mark_block_group_to_copy(fs_info, src_device); + if (ret) + goto leave; + + down_write(&dev_replace->rwsem); + dev_replace->replace_task = current; + 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: + DEBUG_WARN("unexpected STARTED or SUSPENDED dev-replace state"); + ret = BTRFS_IOCTL_DEV_REPLACE_RESULT_ALREADY_STARTED; + up_write(&dev_replace->rwsem); + goto leave; + } + + dev_replace->cont_reading_from_srcdev_mode = read_src; + dev_replace->srcdev = src_device; + dev_replace->tgtdev = tgt_device; + + btrfs_info(fs_info, + "dev_replace from %s (devid %llu) to %s started", + btrfs_dev_name(src_device), + src_device->devid, + btrfs_dev_name(tgt_device)); + + /* + * 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); + + ret = btrfs_sysfs_add_device(tgt_device); + if (ret) + btrfs_err(fs_info, "kobj add dev failed %d", ret); + + btrfs_wait_ordered_roots(fs_info, U64_MAX, NULL); + + /* + * Commit dev_replace state and reserve 1 item for it. + * This is crucial to ensure we won't miss copying extents for new block + * groups that are allocated after we started the device replace, and + * must be done after setting up the device replace state. + */ + trans = btrfs_start_transaction(root, 1); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + down_write(&dev_replace->rwsem); + dev_replace->replace_state = + BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED; + dev_replace->srcdev = NULL; + dev_replace->tgtdev = NULL; + up_write(&dev_replace->rwsem); + 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, false, true); + + ret = btrfs_dev_replace_finishing(fs_info, ret); + if (ret == -EINPROGRESS) + ret = BTRFS_IOCTL_DEV_REPLACE_RESULT_SCRUB_INPROGRESS; + + return ret; + +leave: + if (src_device->bdev) + bdev_allow_freeze(src_device->bdev); + btrfs_destroy_dev_replace_tgtdev(tgt_device, true); + return ret; +} + +static int btrfs_check_replace_dev_names(struct btrfs_ioctl_dev_replace_args *args) +{ + if (args->start.srcdevid == 0) { + if (memchr(args->start.srcdev_name, 0, + sizeof(args->start.srcdev_name)) == NULL) + return -ENAMETOOLONG; + } else { + args->start.srcdev_name[0] = 0; + } + + if (memchr(args->start.tgtdev_name, 0, + sizeof(args->start.tgtdev_name)) == NULL) + return -ENAMETOOLONG; + + return 0; +} + +int btrfs_dev_replace_by_ioctl(struct btrfs_fs_info *fs_info, + struct btrfs_ioctl_dev_replace_args *args) +{ + int ret; + + 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; + } + ret = btrfs_check_replace_dev_names(args); + if (ret < 0) + return ret; + + ret = btrfs_dev_replace_start(fs_info, args->start.tgtdev_name, + args->start.srcdevid, + args->start.srcdev_name, + args->start.cont_reading_from_srcdev_mode); + args->result = ret; + /* don't warn if EINPROGRESS, someone else might be running scrub */ + if (ret == BTRFS_IOCTL_DEV_REPLACE_RESULT_SCRUB_INPROGRESS || + ret == BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR) + return 0; + + return ret; +} + +/* + * blocked until all in-flight bios operations are finished. + */ +static void btrfs_rm_dev_replace_blocked(struct btrfs_fs_info *fs_info) +{ + set_bit(BTRFS_FS_STATE_DEV_REPLACING, &fs_info->fs_state); + wait_event(fs_info->dev_replace.replace_wait, !percpu_counter_sum( + &fs_info->dev_replace.bio_counter)); +} + +/* + * we have removed target device, it is safe to allow new bios request. + */ +static void btrfs_rm_dev_replace_unblocked(struct btrfs_fs_info *fs_info) +{ + clear_bit(BTRFS_FS_STATE_DEV_REPLACING, &fs_info->fs_state); + wake_up(&fs_info->dev_replace.replace_wait); +} + +/* + * When finishing the device replace, before swapping the source device with the + * target device we must update the chunk allocation state in the target device, + * as it is empty because replace works by directly copying the chunks and not + * through the normal chunk allocation path. + */ +static int btrfs_set_target_alloc_state(struct btrfs_device *srcdev, + struct btrfs_device *tgtdev) +{ + struct extent_state *cached_state = NULL; + u64 start = 0; + u64 found_start; + u64 found_end; + int ret = 0; + + lockdep_assert_held(&srcdev->fs_info->chunk_mutex); + + while (btrfs_find_first_extent_bit(&srcdev->alloc_state, start, + &found_start, &found_end, + CHUNK_ALLOCATED, &cached_state)) { + ret = btrfs_set_extent_bit(&tgtdev->alloc_state, found_start, + found_end, CHUNK_ALLOCATED, NULL); + if (ret) + break; + start = found_end + 1; + } + + btrfs_free_extent_state(cached_state); + return ret; +} + +static void btrfs_dev_replace_update_device_in_mapping_tree( + struct btrfs_fs_info *fs_info, + struct btrfs_device *srcdev, + struct btrfs_device *tgtdev) +{ + struct rb_node *node; + + /* + * The chunk mutex must be held so that no new chunks can be created + * while we are updating existing chunks. This guarantees we don't miss + * any new chunk that gets created for a range that falls before the + * range of the last chunk we processed. + */ + lockdep_assert_held(&fs_info->chunk_mutex); + + write_lock(&fs_info->mapping_tree_lock); + node = rb_first_cached(&fs_info->mapping_tree); + while (node) { + struct rb_node *next = rb_next(node); + struct btrfs_chunk_map *map; + u64 next_start; + + map = rb_entry(node, struct btrfs_chunk_map, rb_node); + next_start = map->start + map->chunk_len; + + for (int i = 0; i < map->num_stripes; i++) + if (srcdev == map->stripes[i].dev) + map->stripes[i].dev = tgtdev; + + if (cond_resched_rwlock_write(&fs_info->mapping_tree_lock)) { + map = btrfs_find_chunk_map_nolock(fs_info, next_start, U64_MAX); + if (!map) + break; + node = &map->rb_node; + /* + * Drop the lookup reference since we are holding the + * lock in write mode and no one can remove the chunk + * map from the tree and drop its tree reference. + */ + btrfs_free_chunk_map(map); + } else { + node = next; + } + } + write_unlock(&fs_info->mapping_tree_lock); +} + +static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info, + int scrub_ret) +{ + struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; + struct btrfs_fs_devices *fs_devices = fs_info->fs_devices; + struct btrfs_device *tgt_device; + struct btrfs_device *src_device; + struct btrfs_root *root = fs_info->tree_root; + u8 uuid_tmp[BTRFS_UUID_SIZE]; + struct btrfs_trans_handle *trans; + int ret = 0; + + /* don't allow cancel or unmount to disturb the finishing procedure */ + mutex_lock(&dev_replace->lock_finishing_cancel_unmount); + + down_read(&dev_replace->rwsem); + /* was the operation canceled, or is it finished? */ + if (dev_replace->replace_state != + BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED) { + up_read(&dev_replace->rwsem); + mutex_unlock(&dev_replace->lock_finishing_cancel_unmount); + return 0; + } + + tgt_device = dev_replace->tgtdev; + src_device = dev_replace->srcdev; + up_read(&dev_replace->rwsem); + + /* + * flush all outstanding I/O and inode extent mappings before the + * copy operation is declared as being finished + */ + ret = btrfs_start_delalloc_roots(fs_info, LONG_MAX, false); + if (ret) { + /* Stays started/resumable; keep both denied. */ + mutex_unlock(&dev_replace->lock_finishing_cancel_unmount); + return ret; + } + btrfs_wait_ordered_roots(fs_info, U64_MAX, NULL); + + /* + * 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)) { + /* Stays started/resumable; keep both denied. */ + 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_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_devices->device_list_mutex); + mutex_unlock(&fs_info->chunk_mutex); + } else { + break; + } + } + + down_write(&dev_replace->rwsem); + dev_replace->replace_state = + scrub_ret ? BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED + : BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED; + dev_replace->tgtdev = NULL; + dev_replace->srcdev = NULL; + dev_replace->time_stopped = ktime_get_real_seconds(); + dev_replace->item_needs_writeback = 1; + + /* + * Update allocation state in the new device and replace the old device + * with the new one in the mapping tree. + */ + if (!scrub_ret) { + scrub_ret = btrfs_set_target_alloc_state(src_device, tgt_device); + if (scrub_ret) + goto error; + btrfs_dev_replace_update_device_in_mapping_tree(fs_info, + src_device, + tgt_device); + } else { + if (scrub_ret != -ECANCELED) + btrfs_err(fs_info, + "btrfs_scrub_dev(%s, %llu, %s) failed %d", + btrfs_dev_name(src_device), + src_device->devid, + btrfs_dev_name(tgt_device), scrub_ret); +error: + up_write(&dev_replace->rwsem); + mutex_unlock(&fs_info->chunk_mutex); + mutex_unlock(&fs_devices->device_list_mutex); + btrfs_rm_dev_replace_blocked(fs_info); + if (tgt_device) + btrfs_destroy_dev_replace_tgtdev(tgt_device, true); + /* The source stays a member; re-allow freezing it. */ + if (src_device->bdev) + bdev_allow_freeze(src_device->bdev); + btrfs_rm_dev_replace_unblocked(fs_info); + mutex_unlock(&dev_replace->lock_finishing_cancel_unmount); + + return scrub_ret; + } + + btrfs_info(fs_info, + "dev_replace from %s (devid %llu) to %s finished", + btrfs_dev_name(src_device), + src_device->devid, + btrfs_dev_name(tgt_device)); + clear_bit(BTRFS_DEV_STATE_REPLACE_TGT, &tgt_device->dev_state); + tgt_device->devid = src_device->devid; + src_device->devid = BTRFS_DEV_REPLACE_DEVID; + memcpy(uuid_tmp, tgt_device->uuid, sizeof(uuid_tmp)); + memcpy(tgt_device->uuid, src_device->uuid, sizeof(tgt_device->uuid)); + memcpy(src_device->uuid, uuid_tmp, sizeof(src_device->uuid)); + btrfs_device_set_total_bytes(tgt_device, src_device->total_bytes); + btrfs_device_set_disk_total_bytes(tgt_device, + src_device->disk_total_bytes); + btrfs_device_set_bytes_used(tgt_device, src_device->bytes_used); + tgt_device->commit_bytes_used = src_device->bytes_used; + + btrfs_assign_next_active_device(src_device, tgt_device); + + list_add(&tgt_device->dev_alloc_list, &fs_devices->alloc_list); + fs_devices->rw_devices++; + + dev_replace->replace_task = NULL; + up_write(&dev_replace->rwsem); + btrfs_rm_dev_replace_blocked(fs_info); + + btrfs_rm_dev_replace_remove_srcdev(src_device); + + btrfs_rm_dev_replace_unblocked(fs_info); + + /* + * Increment dev_stats_ccnt so that btrfs_run_dev_stats() will + * update on-disk dev stats value during commit transaction + */ + atomic_inc(&tgt_device->dev_stats_ccnt); + + /* + * this is again a consistent state where no dev_replace procedure + * is running, the target device is part of the filesystem, the + * source device is not part of the filesystem anymore and its 1st + * superblock is scratched out so that it is no longer marked to + * belong to this filesystem. + */ + mutex_unlock(&fs_info->chunk_mutex); + mutex_unlock(&fs_devices->device_list_mutex); + + /* replace the sysfs entry */ + btrfs_sysfs_remove_device(src_device); + btrfs_sysfs_update_devid(tgt_device); + if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &src_device->dev_state)) + btrfs_scratch_superblocks(fs_info, src_device); + + /* write back the superblocks */ + trans = btrfs_start_transaction(root, 0); + if (!IS_ERR(trans)) { + /* + * Ignore any error here, if we failed to remove the DEV_STATS + * item for devid 0, it's not a big deal. We have other ways + * to address it. + */ + btrfs_remove_dev_stat_item(trans, BTRFS_DEV_REPLACE_DEVID); + btrfs_commit_transaction(trans); + } + + mutex_unlock(&dev_replace->lock_finishing_cancel_unmount); + + /* The target is now a member; the source is freed (allow + release). */ + bdev_allow_freeze(tgt_device->bdev); + btrfs_rm_dev_replace_free_srcdev(src_device); + + 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; +} + +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; + + down_read(&dev_replace->rwsem); + /* 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); + args->status.progress_1000 = btrfs_dev_replace_progress(fs_info); + up_read(&dev_replace->rwsem); +} + +int btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info) +{ + struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; + struct btrfs_device *tgt_device = NULL; + struct btrfs_device *src_device = NULL; + struct btrfs_trans_handle *trans; + struct btrfs_root *root = fs_info->tree_root; + int result; + int ret; + + if (sb_rdonly(fs_info->sb)) + return -EROFS; + + mutex_lock(&dev_replace->lock_finishing_cancel_unmount); + 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: + result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NOT_STARTED; + up_write(&dev_replace->rwsem); + break; + case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED: + tgt_device = dev_replace->tgtdev; + src_device = dev_replace->srcdev; + up_write(&dev_replace->rwsem); + ret = btrfs_scrub_cancel(fs_info); + if (ret < 0) { + result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NOT_STARTED; + } else { + result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR; + /* + * btrfs_dev_replace_finishing() will handle the + * cleanup part + */ + btrfs_info(fs_info, + "dev_replace from %s (devid %llu) to %s canceled", + btrfs_dev_name(src_device), src_device->devid, + btrfs_dev_name(tgt_device)); + } + break; + case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED: + /* + * Scrub doing the replace isn't running so we need to do the + * cleanup step of btrfs_dev_replace_finishing() here + */ + result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR; + tgt_device = dev_replace->tgtdev; + src_device = dev_replace->srcdev; + dev_replace->tgtdev = NULL; + dev_replace->srcdev = NULL; + dev_replace->replace_state = + BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED; + dev_replace->time_stopped = ktime_get_real_seconds(); + dev_replace->item_needs_writeback = 1; + + up_write(&dev_replace->rwsem); + + /* Scrub for replace must not be running in suspended state */ + btrfs_scrub_cancel(fs_info); + + 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); + + btrfs_info(fs_info, + "suspended dev_replace from %s (devid %llu) to %s canceled", + btrfs_dev_name(src_device), src_device->devid, + btrfs_dev_name(tgt_device)); + + /* A suspended replace never re-denied freezing; do not allow. */ + if (tgt_device) + btrfs_destroy_dev_replace_tgtdev(tgt_device, false); + break; + default: + up_write(&dev_replace->rwsem); + result = -EINVAL; + } + + mutex_unlock(&dev_replace->lock_finishing_cancel_unmount); + return result; +} + +void btrfs_dev_replace_suspend_for_unmount(struct btrfs_fs_info *fs_info) +{ + struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; + + mutex_lock(&dev_replace->lock_finishing_cancel_unmount); + 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: + case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED: + break; + case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED: + dev_replace->replace_state = + BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED; + dev_replace->time_stopped = ktime_get_real_seconds(); + dev_replace->item_needs_writeback = 1; + btrfs_info(fs_info, "suspending dev_replace for unmount"); + /* Reopened freezable next mount; resume re-denies. */ + if (dev_replace->srcdev && dev_replace->srcdev->bdev) + bdev_allow_freeze(dev_replace->srcdev->bdev); + if (dev_replace->tgtdev && dev_replace->tgtdev->bdev) + bdev_allow_freeze(dev_replace->tgtdev->bdev); + break; + } + + up_write(&dev_replace->rwsem); + mutex_unlock(&dev_replace->lock_finishing_cancel_unmount); +} + +/* resume dev_replace procedure that was interrupted by unmount */ +int btrfs_resume_dev_replace_async(struct btrfs_fs_info *fs_info) +{ + struct task_struct *task; + struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; + int ret = 0; + + 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: + up_write(&dev_replace->rwsem); + return 0; + case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED: + break; + case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED: + dev_replace->replace_state = + BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED; + break; + } + if (!dev_replace->tgtdev || !dev_replace->tgtdev->bdev) { + btrfs_info(fs_info, + "cannot continue dev_replace, tgtdev is missing"); + btrfs_info(fs_info, + "you may cancel the operation after 'mount -o degraded'"); + dev_replace->replace_state = + BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED; + up_write(&dev_replace->rwsem); + return 0; + } + up_write(&dev_replace->rwsem); + + /* + * This could collide with a paused balance, but the exclusive op logic + * should never allow both to start and pause. We don't want to allow + * dev-replace to start anyway. + */ + if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_DEV_REPLACE)) { + down_write(&dev_replace->rwsem); + dev_replace->replace_state = + BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED; + up_write(&dev_replace->rwsem); + btrfs_info(fs_info, + "cannot resume dev-replace, other exclusive operation running"); + return 0; + } + + /* Re-deny for the resumed replace; stay suspended if frozen now. */ + if (dev_replace->srcdev->bdev && + bdev_deny_freeze(dev_replace->srcdev->bdev)) + goto suspend; + if (bdev_deny_freeze(dev_replace->tgtdev->bdev)) { + if (dev_replace->srcdev->bdev) + bdev_allow_freeze(dev_replace->srcdev->bdev); + goto suspend; + } + + task = kthread_run(btrfs_dev_replace_kthread, fs_info, "btrfs-devrepl"); + if (IS_ERR(task)) { + bdev_allow_freeze(dev_replace->tgtdev->bdev); + if (dev_replace->srcdev->bdev) + bdev_allow_freeze(dev_replace->srcdev->bdev); + /* Undo the deny and suspend, but still fail the mount. */ + ret = PTR_ERR(task); + goto suspend; + } + return 0; + +suspend: + btrfs_exclop_finish(fs_info); + down_write(&dev_replace->rwsem); + dev_replace->replace_state = BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED; + up_write(&dev_replace->rwsem); + return ret; +} + +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(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, false, true); + ret = btrfs_dev_replace_finishing(fs_info, ret); + WARN_ON(ret && ret != -ECANCELED); + + btrfs_exclop_finish(fs_info); + return 0; +} + +bool __pure btrfs_dev_replace_is_ongoing(struct btrfs_dev_replace *dev_replace) +{ + if (!dev_replace->is_valid) + return false; + + 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: + return false; + case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED: + case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED: + /* + * return true even if tgtdev is missing (this is + * something that can happen if the dev_replace + * procedure is suspended by an umount and then + * the tgtdev is missing (or "btrfs dev scan") was + * not called and the filesystem is remounted + * in degraded state. This does not stop the + * dev_replace procedure. It needs to be canceled + * manually if the cancellation is wanted. + */ + break; + } + return true; +} + +void btrfs_bio_counter_sub(struct btrfs_fs_info *fs_info, s64 amount) +{ + percpu_counter_sub(&fs_info->dev_replace.bio_counter, amount); + cond_wake_up_nomb(&fs_info->dev_replace.replace_wait); +} + +void btrfs_bio_counter_inc_blocked(struct btrfs_fs_info *fs_info) +{ + while (1) { + percpu_counter_inc(&fs_info->dev_replace.bio_counter); + if (likely(!test_bit(BTRFS_FS_STATE_DEV_REPLACING, + &fs_info->fs_state))) + break; + + btrfs_bio_counter_dec(fs_info); + wait_event(fs_info->dev_replace.replace_wait, + !test_bit(BTRFS_FS_STATE_DEV_REPLACING, + &fs_info->fs_state)); + } +} -- cgit v1.3.1