^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) // SPDX-License-Identifier: GPL-2.0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) * Copyright (C) STRATO AG 2012. All rights reserved.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) #include <linux/sched.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) #include <linux/bio.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) #include <linux/slab.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) #include <linux/blkdev.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) #include <linux/kthread.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) #include <linux/math64.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) #include "misc.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) #include "ctree.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) #include "extent_map.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) #include "disk-io.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) #include "transaction.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) #include "print-tree.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) #include "volumes.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) #include "async-thread.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #include "check-integrity.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) #include "rcu-string.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #include "dev-replace.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) #include "sysfs.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) * Device replace overview
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) * [Objective]
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) * To copy all extents (both new and on-disk) from source device to target
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) * device, while still keeping the filesystem read-write.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) * [Method]
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) * There are two main methods involved:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) * - Write duplication
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) * All new writes will be written to both target and source devices, so even
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) * if replace gets canceled, sources device still contans up-to-date data.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) * Location: handle_ops_on_dev_replace() from __btrfs_map_block()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) * Start: btrfs_dev_replace_start()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) * End: btrfs_dev_replace_finishing()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) * Content: Latest data/metadata
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) * - Copy existing extents
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) * This happens by re-using scrub facility, as scrub also iterates through
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) * existing extents from commit root.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) * Location: scrub_write_block_to_dev_replace() from
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) * scrub_block_complete()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) * Content: Data/meta from commit root.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) * Due to the content difference, we need to avoid nocow write when dev-replace
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) * is happening. This is done by marking the block group read-only and waiting
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) * for NOCOW writes.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) * After replace is done, the finishing part is done by swapping the target and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) * source devices.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) * Location: btrfs_dev_replace_update_device_in_mapping_tree() from
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) * btrfs_dev_replace_finishing()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) int scrub_ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) static int btrfs_dev_replace_kthread(void *data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) int btrfs_init_dev_replace(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) struct btrfs_key key;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) struct btrfs_root *dev_root = fs_info->dev_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) struct extent_buffer *eb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) int slot;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) struct btrfs_path *path = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) int item_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) struct btrfs_dev_replace_item *ptr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) u64 src_devid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) path = btrfs_alloc_path();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) if (!path) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) ret = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) key.objectid = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) key.type = BTRFS_DEV_REPLACE_KEY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) key.offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) ret = btrfs_search_slot(NULL, dev_root, &key, path, 0, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) no_valid_dev_replace_entry_found:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) * We don't have a replace item or it's corrupted. If there is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) * a replace target, fail the mount.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) if (btrfs_find_device(fs_info->fs_devices,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) BTRFS_DEV_REPLACE_DEVID, NULL, NULL, false)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) "found replace target device without a valid replace item");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) ret = -EUCLEAN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) dev_replace->replace_state =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) dev_replace->cont_reading_from_srcdev_mode =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) BTRFS_DEV_REPLACE_ITEM_CONT_READING_FROM_SRCDEV_MODE_ALWAYS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) dev_replace->time_started = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) dev_replace->time_stopped = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) atomic64_set(&dev_replace->num_write_errors, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) atomic64_set(&dev_replace->num_uncorrectable_read_errors, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) dev_replace->cursor_left = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) dev_replace->committed_cursor_left = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116) dev_replace->cursor_left_last_write_of_item = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) dev_replace->cursor_right = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) dev_replace->srcdev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) dev_replace->tgtdev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) dev_replace->is_valid = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) dev_replace->item_needs_writeback = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) slot = path->slots[0];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) eb = path->nodes[0];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) item_size = btrfs_item_size_nr(eb, slot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) ptr = btrfs_item_ptr(eb, slot, struct btrfs_dev_replace_item);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) if (item_size != sizeof(struct btrfs_dev_replace_item)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) "dev_replace entry found has unexpected size, ignore entry");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) goto no_valid_dev_replace_entry_found;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) src_devid = btrfs_dev_replace_src_devid(eb, ptr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) dev_replace->cont_reading_from_srcdev_mode =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) btrfs_dev_replace_cont_reading_from_srcdev_mode(eb, ptr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) dev_replace->replace_state = btrfs_dev_replace_replace_state(eb, ptr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) dev_replace->time_started = btrfs_dev_replace_time_started(eb, ptr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) dev_replace->time_stopped =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) btrfs_dev_replace_time_stopped(eb, ptr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) atomic64_set(&dev_replace->num_write_errors,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) btrfs_dev_replace_num_write_errors(eb, ptr));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) atomic64_set(&dev_replace->num_uncorrectable_read_errors,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) btrfs_dev_replace_num_uncorrectable_read_errors(eb, ptr));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) dev_replace->cursor_left = btrfs_dev_replace_cursor_left(eb, ptr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) dev_replace->committed_cursor_left = dev_replace->cursor_left;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) dev_replace->cursor_left_last_write_of_item = dev_replace->cursor_left;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) dev_replace->cursor_right = btrfs_dev_replace_cursor_right(eb, ptr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) dev_replace->is_valid = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) dev_replace->item_needs_writeback = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) switch (dev_replace->replace_state) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) * We don't have an active replace item but if there is a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) * replace target, fail the mount.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) if (btrfs_find_device(fs_info->fs_devices,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) BTRFS_DEV_REPLACE_DEVID, NULL, NULL, false)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) "replace devid present without an active replace item");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) ret = -EUCLEAN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) dev_replace->srcdev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) dev_replace->tgtdev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) dev_replace->srcdev = btrfs_find_device(fs_info->fs_devices,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) src_devid, NULL, NULL, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) dev_replace->tgtdev = btrfs_find_device(fs_info->fs_devices,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) BTRFS_DEV_REPLACE_DEVID,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) NULL, NULL, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) * allow 'btrfs dev replace_cancel' if src/tgt device is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) * missing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) if (!dev_replace->srcdev &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) !btrfs_test_opt(fs_info, DEGRADED)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) ret = -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) "cannot mount because device replace operation is ongoing and");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) "srcdev (devid %llu) is missing, need to run 'btrfs dev scan'?",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) src_devid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) if (!dev_replace->tgtdev &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) !btrfs_test_opt(fs_info, DEGRADED)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) ret = -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) "cannot mount because device replace operation is ongoing and");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) "tgtdev (devid %llu) is missing, need to run 'btrfs dev scan'?",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) BTRFS_DEV_REPLACE_DEVID);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) if (dev_replace->tgtdev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) if (dev_replace->srcdev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) dev_replace->tgtdev->total_bytes =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) dev_replace->srcdev->total_bytes;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) dev_replace->tgtdev->disk_total_bytes =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) dev_replace->srcdev->disk_total_bytes;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) dev_replace->tgtdev->commit_total_bytes =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) dev_replace->srcdev->commit_total_bytes;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) dev_replace->tgtdev->bytes_used =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) dev_replace->srcdev->bytes_used;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) dev_replace->tgtdev->commit_bytes_used =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) dev_replace->srcdev->commit_bytes_used;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) set_bit(BTRFS_DEV_STATE_REPLACE_TGT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) &dev_replace->tgtdev->dev_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) WARN_ON(fs_info->fs_devices->rw_devices == 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) dev_replace->tgtdev->io_width = fs_info->sectorsize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) dev_replace->tgtdev->io_align = fs_info->sectorsize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) dev_replace->tgtdev->sector_size = fs_info->sectorsize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) dev_replace->tgtdev->fs_info = fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) set_bit(BTRFS_DEV_STATE_IN_FS_METADATA,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) &dev_replace->tgtdev->dev_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) btrfs_free_path(path);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) * Initialize a new device for device replace target from a given source dev
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) * and path.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) * Return 0 and new device in @device_out, otherwise return < 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) const char *device_path,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) struct btrfs_device *srcdev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) struct btrfs_device **device_out)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) struct btrfs_device *device;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) struct block_device *bdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) struct rcu_string *name;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) u64 devid = BTRFS_DEV_REPLACE_DEVID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) *device_out = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) if (srcdev->fs_devices->seeding) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) btrfs_err(fs_info, "the filesystem is a seed filesystem!");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255) bdev = blkdev_get_by_path(device_path, FMODE_WRITE | FMODE_EXCL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) fs_info->bdev_holder);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) if (IS_ERR(bdev)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) btrfs_err(fs_info, "target device %s is invalid!", device_path);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) return PTR_ERR(bdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) sync_blockdev(bdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) list_for_each_entry(device, &fs_info->fs_devices->devices, dev_list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) if (device->bdev == bdev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) "target device is in the filesystem!");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) ret = -EEXIST;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) goto error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) if (i_size_read(bdev->bd_inode) <
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) btrfs_device_get_total_bytes(srcdev)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) "target device is smaller than source device!");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) goto error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) device = btrfs_alloc_device(NULL, &devid, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) if (IS_ERR(device)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) ret = PTR_ERR(device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) goto error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) name = rcu_string_strdup(device_path, GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) if (!name) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) btrfs_free_device(device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) ret = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) goto error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) rcu_assign_pointer(device->name, name);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) set_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) device->generation = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) device->io_width = fs_info->sectorsize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) device->io_align = fs_info->sectorsize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) device->sector_size = fs_info->sectorsize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) device->total_bytes = btrfs_device_get_total_bytes(srcdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) device->disk_total_bytes = btrfs_device_get_disk_total_bytes(srcdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) device->bytes_used = btrfs_device_get_bytes_used(srcdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) device->commit_total_bytes = srcdev->commit_total_bytes;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) device->commit_bytes_used = device->bytes_used;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) device->fs_info = fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) device->bdev = bdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) set_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) device->mode = FMODE_EXCL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) device->dev_stats_valid = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) device->fs_devices = fs_info->fs_devices;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) mutex_lock(&fs_info->fs_devices->device_list_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) list_add(&device->dev_list, &fs_info->fs_devices->devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) fs_info->fs_devices->num_devices++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) fs_info->fs_devices->open_devices++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) mutex_unlock(&fs_info->fs_devices->device_list_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) *device_out = device;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) error:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) blkdev_put(bdev, FMODE_EXCL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) * called from commit_transaction. Writes changed device replace state to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) * disk.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) int btrfs_run_dev_replace(struct btrfs_trans_handle *trans)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) struct btrfs_fs_info *fs_info = trans->fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) struct btrfs_root *dev_root = fs_info->dev_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) struct btrfs_path *path;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) struct btrfs_key key;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) struct extent_buffer *eb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) struct btrfs_dev_replace_item *ptr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) down_read(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) if (!dev_replace->is_valid ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) !dev_replace->item_needs_writeback) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) up_read(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) up_read(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353) key.objectid = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) key.type = BTRFS_DEV_REPLACE_KEY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) key.offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) path = btrfs_alloc_path();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) if (!path) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) ret = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) ret = btrfs_search_slot(trans, dev_root, &key, path, -1, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) if (ret < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) "error %d while searching for dev_replace item!",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) if (ret == 0 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) btrfs_item_size_nr(path->nodes[0], path->slots[0]) < sizeof(*ptr)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) * need to delete old one and insert a new one.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) * Since no attempt is made to recover any old state, if the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) * dev_replace state is 'running', the data on the target
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) * drive is lost.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) * It would be possible to recover the state: just make sure
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) * that the beginning of the item is never changed and always
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) * contains all the essential information. Then read this
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) * minimal set of information and use it as a base for the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) * new state.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) ret = btrfs_del_item(trans, dev_root, path);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) if (ret != 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) "delete too small dev_replace item failed %d!",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) ret = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) if (ret == 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) /* need to insert a new item */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) btrfs_release_path(path);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396) ret = btrfs_insert_empty_item(trans, dev_root, path,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) &key, sizeof(*ptr));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) if (ret < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) "insert dev_replace item failed %d!", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) eb = path->nodes[0];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) ptr = btrfs_item_ptr(eb, path->slots[0],
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) struct btrfs_dev_replace_item);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) down_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) if (dev_replace->srcdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) btrfs_set_dev_replace_src_devid(eb, ptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) dev_replace->srcdev->devid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) btrfs_set_dev_replace_src_devid(eb, ptr, (u64)-1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) btrfs_set_dev_replace_cont_reading_from_srcdev_mode(eb, ptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) dev_replace->cont_reading_from_srcdev_mode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) btrfs_set_dev_replace_replace_state(eb, ptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) dev_replace->replace_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) btrfs_set_dev_replace_time_started(eb, ptr, dev_replace->time_started);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) btrfs_set_dev_replace_time_stopped(eb, ptr, dev_replace->time_stopped);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) btrfs_set_dev_replace_num_write_errors(eb, ptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) atomic64_read(&dev_replace->num_write_errors));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) btrfs_set_dev_replace_num_uncorrectable_read_errors(eb, ptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) atomic64_read(&dev_replace->num_uncorrectable_read_errors));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) dev_replace->cursor_left_last_write_of_item =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) dev_replace->cursor_left;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) btrfs_set_dev_replace_cursor_left(eb, ptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) dev_replace->cursor_left_last_write_of_item);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) btrfs_set_dev_replace_cursor_right(eb, ptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) dev_replace->cursor_right);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) dev_replace->item_needs_writeback = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) up_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) btrfs_mark_buffer_dirty(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) btrfs_free_path(path);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) static char* btrfs_dev_name(struct btrfs_device *device)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) if (!device || test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) return "<missing disk>";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) return rcu_str_deref(device->name);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) static int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451) const char *tgtdev_name, u64 srcdevid, const char *srcdev_name,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) int read_src)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) struct btrfs_root *root = fs_info->dev_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) struct btrfs_trans_handle *trans;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) struct btrfs_device *tgt_device = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459) struct btrfs_device *src_device = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) src_device = btrfs_find_device_by_devspec(fs_info, srcdevid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) srcdev_name);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) if (IS_ERR(src_device))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) return PTR_ERR(src_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) if (btrfs_pinned_by_swapfile(fs_info, src_device)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) btrfs_warn_in_rcu(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) "cannot replace device %s (devid %llu) due to active swapfile",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) btrfs_dev_name(src_device), src_device->devid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) return -ETXTBSY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) * Here we commit the transaction to make sure commit_total_bytes
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475) * of all the devices are updated.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477) trans = btrfs_attach_transaction(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478) if (!IS_ERR(trans)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) ret = btrfs_commit_transaction(trans);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482) } else if (PTR_ERR(trans) != -ENOENT) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) return PTR_ERR(trans);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) ret = btrfs_init_dev_replace_tgtdev(fs_info, tgtdev_name,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487) src_device, &tgt_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491) down_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) switch (dev_replace->replace_state) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493) case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495) case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) ASSERT(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) ret = BTRFS_IOCTL_DEV_REPLACE_RESULT_ALREADY_STARTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501) up_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502) goto leave;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505) dev_replace->cont_reading_from_srcdev_mode = read_src;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) dev_replace->srcdev = src_device;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507) dev_replace->tgtdev = tgt_device;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509) btrfs_info_in_rcu(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) "dev_replace from %s (devid %llu) to %s started",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) btrfs_dev_name(src_device),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) src_device->devid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) rcu_str_deref(tgt_device->name));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516) * from now on, the writes to the srcdev are all duplicated to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517) * go to the tgtdev as well (refer to btrfs_map_block()).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) dev_replace->replace_state = BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520) dev_replace->time_started = ktime_get_real_seconds();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521) dev_replace->cursor_left = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) dev_replace->committed_cursor_left = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523) dev_replace->cursor_left_last_write_of_item = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) dev_replace->cursor_right = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525) dev_replace->is_valid = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) dev_replace->item_needs_writeback = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527) atomic64_set(&dev_replace->num_write_errors, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) atomic64_set(&dev_replace->num_uncorrectable_read_errors, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529) up_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) ret = btrfs_sysfs_add_device(tgt_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) btrfs_err(fs_info, "kobj add dev failed %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) btrfs_wait_ordered_roots(fs_info, U64_MAX, 0, (u64)-1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) /* Commit dev_replace state and reserve 1 item for it. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538) trans = btrfs_start_transaction(root, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) if (IS_ERR(trans)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) ret = PTR_ERR(trans);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541) down_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542) dev_replace->replace_state =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) dev_replace->srcdev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545) dev_replace->tgtdev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546) up_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) goto leave;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550) ret = btrfs_commit_transaction(trans);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551) WARN_ON(ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553) /* the disk copy procedure reuses the scrub code */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) ret = btrfs_scrub_dev(fs_info, src_device->devid, 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) btrfs_device_get_total_bytes(src_device),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556) &dev_replace->scrub_progress, 0, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) ret = btrfs_dev_replace_finishing(fs_info, ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) if (ret == -EINPROGRESS)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560) ret = BTRFS_IOCTL_DEV_REPLACE_RESULT_SCRUB_INPROGRESS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) leave:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565) btrfs_destroy_dev_replace_tgtdev(tgt_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) int btrfs_dev_replace_by_ioctl(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570) struct btrfs_ioctl_dev_replace_args *args)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574) switch (args->start.cont_reading_from_srcdev_mode) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575) case BTRFS_IOCTL_DEV_REPLACE_CONT_READING_FROM_SRCDEV_MODE_ALWAYS:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576) case BTRFS_IOCTL_DEV_REPLACE_CONT_READING_FROM_SRCDEV_MODE_AVOID:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 577) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 578) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 579) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 580) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 581)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 582) if ((args->start.srcdevid == 0 && args->start.srcdev_name[0] == '\0') ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583) args->start.tgtdev_name[0] == '\0')
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586) ret = btrfs_dev_replace_start(fs_info, args->start.tgtdev_name,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587) args->start.srcdevid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588) args->start.srcdev_name,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589) args->start.cont_reading_from_srcdev_mode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590) args->result = ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591) /* don't warn if EINPROGRESS, someone else might be running scrub */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592) if (ret == BTRFS_IOCTL_DEV_REPLACE_RESULT_SCRUB_INPROGRESS ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593) ret == BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 596) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 597) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 598)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 599) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 600) * blocked until all in-flight bios operations are finished.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 601) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 602) static void btrfs_rm_dev_replace_blocked(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 603) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 604) set_bit(BTRFS_FS_STATE_DEV_REPLACING, &fs_info->fs_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 605) wait_event(fs_info->dev_replace.replace_wait, !percpu_counter_sum(
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 606) &fs_info->dev_replace.bio_counter));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 607) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 608)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 609) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 610) * we have removed target device, it is safe to allow new bios request.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 611) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 612) static void btrfs_rm_dev_replace_unblocked(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 613) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 614) clear_bit(BTRFS_FS_STATE_DEV_REPLACING, &fs_info->fs_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 615) wake_up(&fs_info->dev_replace.replace_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 616) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 617)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 618) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 619) * When finishing the device replace, before swapping the source device with the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 620) * target device we must update the chunk allocation state in the target device,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 621) * as it is empty because replace works by directly copying the chunks and not
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 622) * through the normal chunk allocation path.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 623) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 624) static int btrfs_set_target_alloc_state(struct btrfs_device *srcdev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 625) struct btrfs_device *tgtdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 626) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 627) struct extent_state *cached_state = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 628) u64 start = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 629) u64 found_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 630) u64 found_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 631) int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 632)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 633) lockdep_assert_held(&srcdev->fs_info->chunk_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 634)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 635) while (!find_first_extent_bit(&srcdev->alloc_state, start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 636) &found_start, &found_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 637) CHUNK_ALLOCATED, &cached_state)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 638) ret = set_extent_bits(&tgtdev->alloc_state, found_start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 639) found_end, CHUNK_ALLOCATED);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 640) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 641) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 642) start = found_end + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 643) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 644)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 645) free_extent_state(cached_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 646) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 647) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 648)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 649) static void btrfs_dev_replace_update_device_in_mapping_tree(
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 650) struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 651) struct btrfs_device *srcdev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 652) struct btrfs_device *tgtdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 653) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 654) struct extent_map_tree *em_tree = &fs_info->mapping_tree;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 655) struct extent_map *em;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 656) struct map_lookup *map;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 657) u64 start = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 658) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 659)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 660) write_lock(&em_tree->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 661) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 662) em = lookup_extent_mapping(em_tree, start, (u64)-1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 663) if (!em)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 664) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 665) map = em->map_lookup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 666) for (i = 0; i < map->num_stripes; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 667) if (srcdev == map->stripes[i].dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 668) map->stripes[i].dev = tgtdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 669) start = em->start + em->len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 670) free_extent_map(em);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 671) } while (start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 672) write_unlock(&em_tree->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 673) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 674)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 675) static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 676) int scrub_ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 677) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 678) struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 679) struct btrfs_device *tgt_device;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 680) struct btrfs_device *src_device;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 681) struct btrfs_root *root = fs_info->tree_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 682) u8 uuid_tmp[BTRFS_UUID_SIZE];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 683) struct btrfs_trans_handle *trans;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 684) int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 685)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 686) /* don't allow cancel or unmount to disturb the finishing procedure */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 687) mutex_lock(&dev_replace->lock_finishing_cancel_unmount);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 688)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 689) down_read(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 690) /* was the operation canceled, or is it finished? */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 691) if (dev_replace->replace_state !=
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 692) BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 693) up_read(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 694) mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 695) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 696) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 697)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 698) tgt_device = dev_replace->tgtdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 699) src_device = dev_replace->srcdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 700) up_read(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 701)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 702) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 703) * flush all outstanding I/O and inode extent mappings before the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 704) * copy operation is declared as being finished
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 705) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 706) ret = btrfs_start_delalloc_roots(fs_info, U64_MAX, false);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 707) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 708) mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 709) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 710) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 711) btrfs_wait_ordered_roots(fs_info, U64_MAX, 0, (u64)-1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 712)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 713) if (!scrub_ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 714) btrfs_reada_remove_dev(src_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 715)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 716) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 717) * We have to use this loop approach because at this point src_device
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 718) * has to be available for transaction commit to complete, yet new
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 719) * chunks shouldn't be allocated on the device.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 720) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 721) while (1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 722) trans = btrfs_start_transaction(root, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 723) if (IS_ERR(trans)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 724) btrfs_reada_undo_remove_dev(src_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 725) mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 726) return PTR_ERR(trans);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 727) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 728) ret = btrfs_commit_transaction(trans);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 729) WARN_ON(ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 730)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 731) /* Prevent write_all_supers() during the finishing procedure */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 732) mutex_lock(&fs_info->fs_devices->device_list_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 733) /* Prevent new chunks being allocated on the source device */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 734) mutex_lock(&fs_info->chunk_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 735)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 736) if (!list_empty(&src_device->post_commit_list)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 737) mutex_unlock(&fs_info->fs_devices->device_list_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 738) mutex_unlock(&fs_info->chunk_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 739) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 740) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 741) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 742) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 743)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 744) down_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 745) dev_replace->replace_state =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 746) scrub_ret ? BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 747) : BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 748) dev_replace->tgtdev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 749) dev_replace->srcdev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 750) dev_replace->time_stopped = ktime_get_real_seconds();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 751) dev_replace->item_needs_writeback = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 752)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 753) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 754) * Update allocation state in the new device and replace the old device
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 755) * with the new one in the mapping tree.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 756) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 757) if (!scrub_ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 758) scrub_ret = btrfs_set_target_alloc_state(src_device, tgt_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 759) if (scrub_ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 760) goto error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 761) btrfs_dev_replace_update_device_in_mapping_tree(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 762) src_device,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 763) tgt_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 764) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 765) if (scrub_ret != -ECANCELED)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 766) btrfs_err_in_rcu(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 767) "btrfs_scrub_dev(%s, %llu, %s) failed %d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 768) btrfs_dev_name(src_device),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 769) src_device->devid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 770) rcu_str_deref(tgt_device->name), scrub_ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 771) error:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 772) up_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 773) mutex_unlock(&fs_info->chunk_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 774) mutex_unlock(&fs_info->fs_devices->device_list_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 775) btrfs_reada_undo_remove_dev(src_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 776) btrfs_rm_dev_replace_blocked(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 777) if (tgt_device)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 778) btrfs_destroy_dev_replace_tgtdev(tgt_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 779) btrfs_rm_dev_replace_unblocked(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 780) mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 781)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 782) return scrub_ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 783) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 784)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 785) btrfs_info_in_rcu(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 786) "dev_replace from %s (devid %llu) to %s finished",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 787) btrfs_dev_name(src_device),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 788) src_device->devid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 789) rcu_str_deref(tgt_device->name));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 790) clear_bit(BTRFS_DEV_STATE_REPLACE_TGT, &tgt_device->dev_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 791) tgt_device->devid = src_device->devid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 792) src_device->devid = BTRFS_DEV_REPLACE_DEVID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 793) memcpy(uuid_tmp, tgt_device->uuid, sizeof(uuid_tmp));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 794) memcpy(tgt_device->uuid, src_device->uuid, sizeof(tgt_device->uuid));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 795) memcpy(src_device->uuid, uuid_tmp, sizeof(src_device->uuid));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 796) btrfs_device_set_total_bytes(tgt_device, src_device->total_bytes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 797) btrfs_device_set_disk_total_bytes(tgt_device,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 798) src_device->disk_total_bytes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 799) btrfs_device_set_bytes_used(tgt_device, src_device->bytes_used);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 800) tgt_device->commit_bytes_used = src_device->bytes_used;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 801)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 802) btrfs_assign_next_active_device(src_device, tgt_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 803)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 804) list_add(&tgt_device->dev_alloc_list, &fs_info->fs_devices->alloc_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 805) fs_info->fs_devices->rw_devices++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 806)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 807) up_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 808) btrfs_rm_dev_replace_blocked(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 809)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 810) btrfs_rm_dev_replace_remove_srcdev(src_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 811)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 812) btrfs_rm_dev_replace_unblocked(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 813)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 814) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 815) * Increment dev_stats_ccnt so that btrfs_run_dev_stats() will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 816) * update on-disk dev stats value during commit transaction
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 817) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 818) atomic_inc(&tgt_device->dev_stats_ccnt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 819)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 820) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 821) * this is again a consistent state where no dev_replace procedure
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 822) * is running, the target device is part of the filesystem, the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 823) * source device is not part of the filesystem anymore and its 1st
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 824) * superblock is scratched out so that it is no longer marked to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 825) * belong to this filesystem.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 826) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 827) mutex_unlock(&fs_info->chunk_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 828) mutex_unlock(&fs_info->fs_devices->device_list_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 829)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 830) /* replace the sysfs entry */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 831) btrfs_sysfs_remove_device(src_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 832) btrfs_sysfs_update_devid(tgt_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 833) if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &src_device->dev_state))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 834) btrfs_scratch_superblocks(fs_info, src_device->bdev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 835) src_device->name->str);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 836)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 837) /* write back the superblocks */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 838) trans = btrfs_start_transaction(root, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 839) if (!IS_ERR(trans))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 840) btrfs_commit_transaction(trans);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 841)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 842) mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 843)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 844) btrfs_rm_dev_replace_free_srcdev(src_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 845)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 846) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 847) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 848)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 849) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 850) * Read progress of device replace status according to the state and last
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 851) * stored position. The value format is the same as for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 852) * btrfs_dev_replace::progress_1000
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 853) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 854) static u64 btrfs_dev_replace_progress(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 855) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 856) struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 857) u64 ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 858)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 859) switch (dev_replace->replace_state) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 860) case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 861) case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 862) ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 863) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 864) case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 865) ret = 1000;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 866) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 867) case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 868) case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 869) ret = div64_u64(dev_replace->cursor_left,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 870) div_u64(btrfs_device_get_total_bytes(
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 871) dev_replace->srcdev), 1000));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 872) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 873) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 874)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 875) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 876) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 877)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 878) void btrfs_dev_replace_status(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 879) struct btrfs_ioctl_dev_replace_args *args)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 880) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 881) struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 882)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 883) down_read(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 884) /* even if !dev_replace_is_valid, the values are good enough for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 885) * the replace_status ioctl */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 886) args->result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 887) args->status.replace_state = dev_replace->replace_state;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 888) args->status.time_started = dev_replace->time_started;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 889) args->status.time_stopped = dev_replace->time_stopped;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 890) args->status.num_write_errors =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 891) atomic64_read(&dev_replace->num_write_errors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 892) args->status.num_uncorrectable_read_errors =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 893) atomic64_read(&dev_replace->num_uncorrectable_read_errors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 894) args->status.progress_1000 = btrfs_dev_replace_progress(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 895) up_read(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 896) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 897)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 898) int btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 899) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 900) struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 901) struct btrfs_device *tgt_device = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 902) struct btrfs_device *src_device = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 903) struct btrfs_trans_handle *trans;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 904) struct btrfs_root *root = fs_info->tree_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 905) int result;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 906) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 907)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 908) if (sb_rdonly(fs_info->sb))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 909) return -EROFS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 910)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 911) mutex_lock(&dev_replace->lock_finishing_cancel_unmount);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 912) down_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 913) switch (dev_replace->replace_state) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 914) case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 915) case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 916) case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 917) result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NOT_STARTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 918) up_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 919) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 920) case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 921) tgt_device = dev_replace->tgtdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 922) src_device = dev_replace->srcdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 923) up_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 924) ret = btrfs_scrub_cancel(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 925) if (ret < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 926) result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NOT_STARTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 927) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 928) result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 929) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 930) * btrfs_dev_replace_finishing() will handle the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 931) * cleanup part
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 932) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 933) btrfs_info_in_rcu(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 934) "dev_replace from %s (devid %llu) to %s canceled",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 935) btrfs_dev_name(src_device), src_device->devid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 936) btrfs_dev_name(tgt_device));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 937) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 938) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 939) case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 940) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 941) * Scrub doing the replace isn't running so we need to do the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 942) * cleanup step of btrfs_dev_replace_finishing() here
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 943) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 944) result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 945) tgt_device = dev_replace->tgtdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 946) src_device = dev_replace->srcdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 947) dev_replace->tgtdev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 948) dev_replace->srcdev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 949) dev_replace->replace_state =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 950) BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 951) dev_replace->time_stopped = ktime_get_real_seconds();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 952) dev_replace->item_needs_writeback = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 953)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 954) up_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 955)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 956) /* Scrub for replace must not be running in suspended state */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 957) ret = btrfs_scrub_cancel(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 958) ASSERT(ret != -ENOTCONN);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 959)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 960) trans = btrfs_start_transaction(root, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 961) if (IS_ERR(trans)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 962) mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 963) return PTR_ERR(trans);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 964) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 965) ret = btrfs_commit_transaction(trans);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 966) WARN_ON(ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 967)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 968) btrfs_info_in_rcu(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 969) "suspended dev_replace from %s (devid %llu) to %s canceled",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 970) btrfs_dev_name(src_device), src_device->devid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 971) btrfs_dev_name(tgt_device));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 972)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 973) if (tgt_device)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 974) btrfs_destroy_dev_replace_tgtdev(tgt_device);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 975) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 976) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 977) up_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 978) result = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 979) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 980)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 981) mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 982) return result;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 983) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 984)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 985) void btrfs_dev_replace_suspend_for_unmount(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 986) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 987) struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 988)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 989) mutex_lock(&dev_replace->lock_finishing_cancel_unmount);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 990) down_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 991)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 992) switch (dev_replace->replace_state) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 993) case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 994) case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 995) case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 996) case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 997) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 998) case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 999) dev_replace->replace_state =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1000) BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1001) dev_replace->time_stopped = ktime_get_real_seconds();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1002) dev_replace->item_needs_writeback = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1003) btrfs_info(fs_info, "suspending dev_replace for unmount");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1004) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1005) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1006)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1007) up_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1008) mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1009) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1010)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1011) /* resume dev_replace procedure that was interrupted by unmount */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1012) int btrfs_resume_dev_replace_async(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1013) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1014) struct task_struct *task;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1015) struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1016)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1017) down_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1018)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1019) switch (dev_replace->replace_state) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1020) case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1021) case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1022) case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1023) up_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1024) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1025) case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1026) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1027) case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1028) dev_replace->replace_state =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1029) BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1030) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1031) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1032) if (!dev_replace->tgtdev || !dev_replace->tgtdev->bdev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1033) btrfs_info(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1034) "cannot continue dev_replace, tgtdev is missing");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1035) btrfs_info(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1036) "you may cancel the operation after 'mount -o degraded'");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1037) dev_replace->replace_state =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1038) BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1039) up_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1040) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1041) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1042) up_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1043)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1044) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1045) * This could collide with a paused balance, but the exclusive op logic
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1046) * should never allow both to start and pause. We don't want to allow
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1047) * dev-replace to start anyway.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1048) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1049) if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_DEV_REPLACE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1050) down_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1051) dev_replace->replace_state =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1052) BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1053) up_write(&dev_replace->rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1054) btrfs_info(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1055) "cannot resume dev-replace, other exclusive operation running");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1056) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1057) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1058)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1059) task = kthread_run(btrfs_dev_replace_kthread, fs_info, "btrfs-devrepl");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1060) return PTR_ERR_OR_ZERO(task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1061) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1062)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1063) static int btrfs_dev_replace_kthread(void *data)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1064) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1065) struct btrfs_fs_info *fs_info = data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1066) struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1067) u64 progress;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1068) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1069)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1070) progress = btrfs_dev_replace_progress(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1071) progress = div_u64(progress, 10);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1072) btrfs_info_in_rcu(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1073) "continuing dev_replace from %s (devid %llu) to target %s @%u%%",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1074) btrfs_dev_name(dev_replace->srcdev),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1075) dev_replace->srcdev->devid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1076) btrfs_dev_name(dev_replace->tgtdev),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1077) (unsigned int)progress);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1078)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1079) ret = btrfs_scrub_dev(fs_info, dev_replace->srcdev->devid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1080) dev_replace->committed_cursor_left,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1081) btrfs_device_get_total_bytes(dev_replace->srcdev),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1082) &dev_replace->scrub_progress, 0, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1083) ret = btrfs_dev_replace_finishing(fs_info, ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1084) WARN_ON(ret && ret != -ECANCELED);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1085)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1086) btrfs_exclop_finish(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1087) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1088) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1089)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1090) int __pure btrfs_dev_replace_is_ongoing(struct btrfs_dev_replace *dev_replace)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1091) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1092) if (!dev_replace->is_valid)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1093) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1094)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1095) switch (dev_replace->replace_state) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1096) case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1097) case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1098) case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1099) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1100) case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1101) case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1102) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1103) * return true even if tgtdev is missing (this is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1104) * something that can happen if the dev_replace
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1105) * procedure is suspended by an umount and then
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1106) * the tgtdev is missing (or "btrfs dev scan") was
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1107) * not called and the filesystem is remounted
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1108) * in degraded state. This does not stop the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1109) * dev_replace procedure. It needs to be canceled
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1110) * manually if the cancellation is wanted.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1111) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1112) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1113) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1114) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1115) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1117) void btrfs_bio_counter_inc_noblocked(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1118) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1119) percpu_counter_inc(&fs_info->dev_replace.bio_counter);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1120) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1121)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1122) void btrfs_bio_counter_sub(struct btrfs_fs_info *fs_info, s64 amount)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1123) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1124) percpu_counter_sub(&fs_info->dev_replace.bio_counter, amount);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1125) cond_wake_up_nomb(&fs_info->dev_replace.replace_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1126) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1127)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1128) void btrfs_bio_counter_inc_blocked(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1129) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1130) while (1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1131) percpu_counter_inc(&fs_info->dev_replace.bio_counter);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1132) if (likely(!test_bit(BTRFS_FS_STATE_DEV_REPLACING,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1133) &fs_info->fs_state)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1134) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1135)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1136) btrfs_bio_counter_dec(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1137) wait_event(fs_info->dev_replace.replace_wait,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1138) !test_bit(BTRFS_FS_STATE_DEV_REPLACING,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1139) &fs_info->fs_state));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1140) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1141) }
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags