^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) 2007 Oracle. 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/fs.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) #include <linux/blkdev.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) #include <linux/radix-tree.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) #include <linux/writeback.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) #include <linux/workqueue.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) #include <linux/kthread.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) #include <linux/slab.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) #include <linux/migrate.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) #include <linux/ratelimit.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) #include <linux/uuid.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) #include <linux/semaphore.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) #include <linux/error-injection.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) #include <linux/crc32c.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) #include <linux/sched/mm.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #include <asm/unaligned.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) #include <crypto/hash.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #include "ctree.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) #include "disk-io.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) #include "transaction.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) #include "btrfs_inode.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) #include "volumes.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) #include "print-tree.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) #include "locking.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) #include "tree-log.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) #include "free-space-cache.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) #include "free-space-tree.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) #include "inode-map.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) #include "check-integrity.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) #include "rcu-string.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) #include "dev-replace.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) #include "raid56.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) #include "sysfs.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) #include "qgroup.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) #include "compression.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) #include "tree-checker.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) #include "ref-verify.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) #include "block-group.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) #include "discard.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) #include "space-info.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) #define BTRFS_SUPER_FLAG_SUPP (BTRFS_HEADER_FLAG_WRITTEN |\
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) BTRFS_HEADER_FLAG_RELOC |\
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) BTRFS_SUPER_FLAG_ERROR |\
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) BTRFS_SUPER_FLAG_SEEDING |\
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) BTRFS_SUPER_FLAG_METADUMP |\
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) BTRFS_SUPER_FLAG_METADUMP_V2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) static void end_workqueue_fn(struct btrfs_work *work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) static void btrfs_destroy_ordered_extents(struct btrfs_root *root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) struct btrfs_fs_info *fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) static void btrfs_destroy_delalloc_inodes(struct btrfs_root *root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) static int btrfs_destroy_marked_extents(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) struct extent_io_tree *dirty_pages,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) int mark);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) static int btrfs_destroy_pinned_extent(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) struct extent_io_tree *pinned_extents);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) static int btrfs_cleanup_transaction(struct btrfs_fs_info *fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) static void btrfs_error_commit_super(struct btrfs_fs_info *fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) * btrfs_end_io_wq structs are used to do processing in task context when an IO
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) * is complete. This is used during reads to verify checksums, and it is used
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) * by writes to insert metadata for new file extents after IO is complete.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) struct btrfs_end_io_wq {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) struct bio *bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) bio_end_io_t *end_io;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) void *private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) struct btrfs_fs_info *info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) blk_status_t status;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) enum btrfs_wq_endio_type metadata;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) struct btrfs_work work;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) static struct kmem_cache *btrfs_end_io_wq_cache;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) int __init btrfs_end_io_wq_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) btrfs_end_io_wq_cache = kmem_cache_create("btrfs_end_io_wq",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) sizeof(struct btrfs_end_io_wq),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) SLAB_MEM_SPREAD,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) if (!btrfs_end_io_wq_cache)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) void __cold btrfs_end_io_wq_exit(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) kmem_cache_destroy(btrfs_end_io_wq_cache);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) static void btrfs_free_csum_hash(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) if (fs_info->csum_shash)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) crypto_free_shash(fs_info->csum_shash);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) * async submit bios are used to offload expensive checksumming
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) * onto the worker threads. They checksum file and metadata bios
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) * just before they are sent down the IO stack.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) struct async_submit_bio {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) void *private_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) struct bio *bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) extent_submit_bio_start_t *submit_bio_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) int mirror_num;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) * bio_offset is optional, can be used if the pages in the bio
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) * can't tell us where in the file the bio should go
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) u64 bio_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) struct btrfs_work work;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) blk_status_t status;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) * Lockdep class keys for extent_buffer->lock's in this root. For a given
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) * eb, the lockdep key is determined by the btrfs_root it belongs to and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) * the level the eb occupies in the tree.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) * Different roots are used for different purposes and may nest inside each
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) * other and they require separate keysets. As lockdep keys should be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) * static, assign keysets according to the purpose of the root as indicated
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) * by btrfs_root->root_key.objectid. This ensures that all special purpose
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) * roots have separate keysets.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) * Lock-nesting across peer nodes is always done with the immediate parent
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) * node locked thus preventing deadlock. As lockdep doesn't know this, use
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) * subclass to avoid triggering lockdep warning in such cases.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) * The key is set by the readpage_end_io_hook after the buffer has passed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) * csum validation but before the pages are unlocked. It is also set by
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) * btrfs_init_new_buffer on freshly allocated blocks.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) * We also add a check to make sure the highest level of the tree is the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) * same as our lockdep setup here. If BTRFS_MAX_LEVEL changes, this code
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) * needs update as well.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) #ifdef CONFIG_DEBUG_LOCK_ALLOC
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) # if BTRFS_MAX_LEVEL != 8
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) # error
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) # endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) static struct btrfs_lockdep_keyset {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) u64 id; /* root objectid */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) const char *name_stem; /* lock name stem */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) char names[BTRFS_MAX_LEVEL + 1][20];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) struct lock_class_key keys[BTRFS_MAX_LEVEL + 1];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) } btrfs_lockdep_keysets[] = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) { .id = BTRFS_ROOT_TREE_OBJECTID, .name_stem = "root" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) { .id = BTRFS_EXTENT_TREE_OBJECTID, .name_stem = "extent" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) { .id = BTRFS_CHUNK_TREE_OBJECTID, .name_stem = "chunk" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) { .id = BTRFS_DEV_TREE_OBJECTID, .name_stem = "dev" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) { .id = BTRFS_FS_TREE_OBJECTID, .name_stem = "fs" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) { .id = BTRFS_CSUM_TREE_OBJECTID, .name_stem = "csum" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) { .id = BTRFS_QUOTA_TREE_OBJECTID, .name_stem = "quota" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) { .id = BTRFS_TREE_LOG_OBJECTID, .name_stem = "log" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) { .id = BTRFS_TREE_RELOC_OBJECTID, .name_stem = "treloc" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) { .id = BTRFS_DATA_RELOC_TREE_OBJECTID, .name_stem = "dreloc" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) { .id = BTRFS_UUID_TREE_OBJECTID, .name_stem = "uuid" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) { .id = BTRFS_FREE_SPACE_TREE_OBJECTID, .name_stem = "free-space" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) { .id = 0, .name_stem = "tree" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) void __init btrfs_init_lockdep(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) int i, j;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) /* initialize lockdep class names */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) for (i = 0; i < ARRAY_SIZE(btrfs_lockdep_keysets); i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) struct btrfs_lockdep_keyset *ks = &btrfs_lockdep_keysets[i];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) for (j = 0; j < ARRAY_SIZE(ks->names); j++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) snprintf(ks->names[j], sizeof(ks->names[j]),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) "btrfs-%s-%02d", ks->name_stem, j);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) void btrfs_set_buffer_lockdep_class(u64 objectid, struct extent_buffer *eb,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) int level)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) struct btrfs_lockdep_keyset *ks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) BUG_ON(level >= ARRAY_SIZE(ks->keys));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) /* find the matching keyset, id 0 is the default entry */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) for (ks = btrfs_lockdep_keysets; ks->id; ks++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) if (ks->id == objectid)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) lockdep_set_class_and_name(&eb->lock,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) &ks->keys[level], ks->names[level]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) * Compute the csum of a btree block and store the result to provided buffer.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) static void csum_tree_block(struct extent_buffer *buf, u8 *result)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) struct btrfs_fs_info *fs_info = buf->fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) const int num_pages = fs_info->nodesize >> PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) SHASH_DESC_ON_STACK(shash, fs_info->csum_shash);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) char *kaddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) shash->tfm = fs_info->csum_shash;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) crypto_shash_init(shash);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) kaddr = page_address(buf->pages[0]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) crypto_shash_update(shash, kaddr + BTRFS_CSUM_SIZE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) PAGE_SIZE - BTRFS_CSUM_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) for (i = 1; i < num_pages; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) kaddr = page_address(buf->pages[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) crypto_shash_update(shash, kaddr, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) memset(result, 0, BTRFS_CSUM_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) crypto_shash_final(shash, result);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) }
^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) * we can't consider a given block up to date unless the transid of the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) * block matches the transid in the parent node's pointer. This is how we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) * detect blocks that either didn't get written at all or got written
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) * in the wrong place.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) static int verify_parent_transid(struct extent_io_tree *io_tree,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) struct extent_buffer *eb, u64 parent_transid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) int atomic)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) struct extent_state *cached_state = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) bool need_lock = (current->journal_info == BTRFS_SEND_TRANS_STUB);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) if (!parent_transid || btrfs_header_generation(eb) == parent_transid)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) if (atomic)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) return -EAGAIN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) if (need_lock) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) btrfs_tree_read_lock(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) btrfs_set_lock_blocking_read(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) lock_extent_bits(io_tree, eb->start, eb->start + eb->len - 1,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) &cached_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) if (extent_buffer_uptodate(eb) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) btrfs_header_generation(eb) == parent_transid) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) btrfs_err_rl(eb->fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) "parent transid verify failed on %llu wanted %llu found %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) eb->start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) parent_transid, btrfs_header_generation(eb));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) ret = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) * Things reading via commit roots that don't have normal protection,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) * like send, can have a really old block in cache that may point at a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) * block that has been freed and re-allocated. So don't clear uptodate
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) * if we find an eb that is under IO (dirty/writeback) because we could
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) * end up reading in the stale data and then writing it back out and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) * making everybody very sad.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) if (!extent_buffer_under_io(eb))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) clear_extent_buffer_uptodate(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) unlock_extent_cached(io_tree, eb->start, eb->start + eb->len - 1,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) &cached_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) if (need_lock)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) btrfs_tree_read_unlock_blocking(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) static bool btrfs_supported_super_csum(u16 csum_type)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) switch (csum_type) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) case BTRFS_CSUM_TYPE_CRC32:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) case BTRFS_CSUM_TYPE_XXHASH:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) case BTRFS_CSUM_TYPE_SHA256:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) case BTRFS_CSUM_TYPE_BLAKE2:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) * Return 0 if the superblock checksum type matches the checksum value of that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) * algorithm. Pass the raw disk superblock data.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) static int btrfs_check_super_csum(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) char *raw_disk_sb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) struct btrfs_super_block *disk_sb =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) (struct btrfs_super_block *)raw_disk_sb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) char result[BTRFS_CSUM_SIZE];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) SHASH_DESC_ON_STACK(shash, fs_info->csum_shash);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) shash->tfm = fs_info->csum_shash;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) * The super_block structure does not span the whole
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) * BTRFS_SUPER_INFO_SIZE range, we expect that the unused space is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) * filled with zeros and is included in the checksum.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) crypto_shash_digest(shash, raw_disk_sb + BTRFS_CSUM_SIZE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE, result);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) if (memcmp(disk_sb->csum, result, btrfs_super_csum_size(disk_sb)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) int btrfs_verify_level_key(struct extent_buffer *eb, int level,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) struct btrfs_key *first_key, u64 parent_transid)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) struct btrfs_fs_info *fs_info = eb->fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) int found_level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) struct btrfs_key found_key;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) found_level = btrfs_header_level(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) if (found_level != level) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) WARN(IS_ENABLED(CONFIG_BTRFS_DEBUG),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) KERN_ERR "BTRFS: tree level check failed\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) "tree level mismatch detected, bytenr=%llu level expected=%u has=%u",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) eb->start, level, found_level);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) return -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) if (!first_key)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) * For live tree block (new tree blocks in current transaction),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) * we need proper lock context to avoid race, which is impossible here.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) * So we only checks tree blocks which is read from disk, whose
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353) * generation <= fs_info->last_trans_committed.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) if (btrfs_header_generation(eb) > fs_info->last_trans_committed)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) /* We have @first_key, so this @eb must have at least one item */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) if (btrfs_header_nritems(eb) == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) "invalid tree nritems, bytenr=%llu nritems=0 expect >0",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) eb->start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) WARN_ON(IS_ENABLED(CONFIG_BTRFS_DEBUG));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) return -EUCLEAN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) if (found_level)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) btrfs_node_key_to_cpu(eb, &found_key, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) btrfs_item_key_to_cpu(eb, &found_key, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) ret = btrfs_comp_cpu_keys(first_key, &found_key);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) WARN(IS_ENABLED(CONFIG_BTRFS_DEBUG),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) KERN_ERR "BTRFS: tree first key check failed\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) "tree first key mismatch detected, bytenr=%llu parent_transid=%llu key expected=(%llu,%u,%llu) has=(%llu,%u,%llu)",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) eb->start, parent_transid, first_key->objectid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) first_key->type, first_key->offset,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) found_key.objectid, found_key.type,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) found_key.offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) * helper to read a given tree block, doing retries as required when
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) * the checksums don't match and we have alternate mirrors to try.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) * @parent_transid: expected transid, skip check if 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) * @level: expected level, mandatory check
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) * @first_key: expected key of first slot, skip check if NULL
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) static int btree_read_extent_buffer_pages(struct extent_buffer *eb,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) u64 parent_transid, int level,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396) struct btrfs_key *first_key)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) struct btrfs_fs_info *fs_info = eb->fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399) struct extent_io_tree *io_tree;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) int failed = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) int num_copies = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) int mirror_num = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) int failed_mirror = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) io_tree = &BTRFS_I(fs_info->btree_inode)->io_tree;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) while (1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) clear_bit(EXTENT_BUFFER_CORRUPT, &eb->bflags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) ret = read_extent_buffer_pages(eb, WAIT_COMPLETE, mirror_num);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) if (!ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) if (verify_parent_transid(io_tree, eb,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) parent_transid, 0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) ret = -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) else if (btrfs_verify_level_key(eb, level,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) first_key, parent_transid))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) ret = -EUCLEAN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) num_copies = btrfs_num_copies(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) eb->start, eb->len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) if (num_copies == 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) if (!failed_mirror) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) failed = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) failed_mirror = eb->read_mirror;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) mirror_num++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) if (mirror_num == failed_mirror)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) mirror_num++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) if (mirror_num > num_copies)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) if (failed && !ret && failed_mirror)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) btrfs_repair_eb_io_failure(eb, failed_mirror);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) * checksum a dirty tree block before IO. This has extra checks to make sure
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) * we only fill in the checksum field in the first page of a multi-page block
^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 csum_dirty_buffer(struct btrfs_fs_info *fs_info, struct page *page)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) u64 start = page_offset(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) u64 found_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) u8 result[BTRFS_CSUM_SIZE];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) u16 csum_size = btrfs_super_csum_size(fs_info->super_copy);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) struct extent_buffer *eb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459) eb = (struct extent_buffer *)page->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460) if (page != eb->pages[0])
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) found_start = btrfs_header_bytenr(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) * Please do not consolidate these warnings into a single if.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) * It is useful to know what went wrong.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) if (WARN_ON(found_start != start))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) return -EUCLEAN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) if (WARN_ON(!PageUptodate(page)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471) return -EUCLEAN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) ASSERT(memcmp_extent_buffer(eb, fs_info->fs_devices->metadata_uuid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) offsetof(struct btrfs_header, fsid),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475) BTRFS_FSID_SIZE) == 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477) csum_tree_block(eb, result);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) if (btrfs_header_level(eb))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) ret = btrfs_check_node(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482) ret = btrfs_check_leaf_full(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) if (ret < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) btrfs_print_tree(eb, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487) "block=%llu write time tree block corruption detected",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) eb->start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) WARN_ON(IS_ENABLED(CONFIG_BTRFS_DEBUG));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) write_extent_buffer(eb, result, 0, csum_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) static int check_tree_block_fsid(struct extent_buffer *eb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) struct btrfs_fs_info *fs_info = eb->fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) struct btrfs_fs_devices *fs_devices = fs_info->fs_devices, *seed_devs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501) u8 fsid[BTRFS_FSID_SIZE];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502) u8 *metadata_uuid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504) read_extent_buffer(eb, fsid, offsetof(struct btrfs_header, fsid),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505) BTRFS_FSID_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507) * Checking the incompat flag is only valid for the current fs. For
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508) * seed devices it's forbidden to have their uuid changed so reading
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509) * ->fsid in this case is fine
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) if (btrfs_fs_incompat(fs_info, METADATA_UUID))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) metadata_uuid = fs_devices->metadata_uuid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514) metadata_uuid = fs_devices->fsid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516) if (!memcmp(fsid, metadata_uuid, BTRFS_FSID_SIZE))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) list_for_each_entry(seed_devs, &fs_devices->seed_list, seed_list)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520) if (!memcmp(fsid, seed_devs->fsid, BTRFS_FSID_SIZE))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) int btrfs_validate_metadata_buffer(struct btrfs_io_bio *io_bio, u64 phy_offset,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527) struct page *page, u64 start, u64 end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) int mirror)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) u64 found_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) int found_level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532) struct extent_buffer *eb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) struct btrfs_fs_info *fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) u16 csum_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536) u8 result[BTRFS_CSUM_SIZE];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) int reads_done;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) if (!page->private)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542) eb = (struct extent_buffer *)page->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) fs_info = eb->fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) csum_size = btrfs_super_csum_size(fs_info->super_copy);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546) /* the pending IO might have been the only thing that kept this buffer
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) * in memory. Make sure we have a ref for all this other checks
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) atomic_inc(&eb->refs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551) reads_done = atomic_dec_and_test(&eb->io_pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) if (!reads_done)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553) goto err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) eb->read_mirror = mirror;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556) if (test_bit(EXTENT_BUFFER_READ_ERR, &eb->bflags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557) ret = -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) goto err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561) found_start = btrfs_header_bytenr(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) if (found_start != eb->start) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563) btrfs_err_rl(fs_info, "bad tree block start, want %llu have %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) eb->start, found_start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565) ret = -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566) goto err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568) if (check_tree_block_fsid(eb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) btrfs_err_rl(fs_info, "bad fsid on block %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570) eb->start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571) ret = -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) goto err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574) found_level = btrfs_header_level(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575) if (found_level >= BTRFS_MAX_LEVEL) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576) btrfs_err(fs_info, "bad tree block level %d on %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 577) (int)btrfs_header_level(eb), eb->start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 578) ret = -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 579) goto err;
^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) btrfs_set_buffer_lockdep_class(btrfs_header_owner(eb),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583) eb, found_level);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585) csum_tree_block(eb, result);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587) if (memcmp_extent_buffer(eb, result, 0, csum_size)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588) u8 val[BTRFS_CSUM_SIZE] = { 0 };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590) read_extent_buffer(eb, &val, 0, csum_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591) btrfs_warn_rl(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592) "%s checksum verify failed on %llu wanted " CSUM_FMT " found " CSUM_FMT " level %d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593) fs_info->sb->s_id, eb->start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594) CSUM_FMT_VALUE(csum_size, val),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595) CSUM_FMT_VALUE(csum_size, result),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 596) btrfs_header_level(eb));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 597) ret = -EUCLEAN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 598) goto err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 599) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 600)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 601) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 602) * If this is a leaf block and it is corrupt, set the corrupt bit so
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 603) * that we don't try and read the other copies of this block, just
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 604) * return -EIO.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 605) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 606) if (found_level == 0 && btrfs_check_leaf_full(eb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 607) set_bit(EXTENT_BUFFER_CORRUPT, &eb->bflags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 608) ret = -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 609) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 610)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 611) if (found_level > 0 && btrfs_check_node(eb))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 612) ret = -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 613)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 614) if (!ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 615) set_extent_buffer_uptodate(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 616) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 617) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 618) "block=%llu read time tree block corruption detected",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 619) eb->start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 620) err:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 621) if (reads_done &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 622) test_and_clear_bit(EXTENT_BUFFER_READAHEAD, &eb->bflags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 623) btree_readahead_hook(eb, ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 624)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 625) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 626) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 627) * our io error hook is going to dec the io pages
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 628) * again, we have to make sure it has something
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 629) * to decrement
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 630) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 631) atomic_inc(&eb->io_pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 632) clear_extent_buffer_uptodate(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 633) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 634) free_extent_buffer(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 635) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 636) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 637) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 638)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 639) static void end_workqueue_bio(struct bio *bio)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 640) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 641) struct btrfs_end_io_wq *end_io_wq = bio->bi_private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 642) struct btrfs_fs_info *fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 643) struct btrfs_workqueue *wq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 644)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 645) fs_info = end_io_wq->info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 646) end_io_wq->status = bio->bi_status;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 647)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 648) if (bio_op(bio) == REQ_OP_WRITE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 649) if (end_io_wq->metadata == BTRFS_WQ_ENDIO_METADATA)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 650) wq = fs_info->endio_meta_write_workers;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 651) else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_FREE_SPACE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 652) wq = fs_info->endio_freespace_worker;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 653) else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 654) wq = fs_info->endio_raid56_workers;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 655) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 656) wq = fs_info->endio_write_workers;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 657) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 658) if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 659) wq = fs_info->endio_raid56_workers;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 660) else if (end_io_wq->metadata)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 661) wq = fs_info->endio_meta_workers;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 662) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 663) wq = fs_info->endio_workers;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 664) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 665)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 666) btrfs_init_work(&end_io_wq->work, end_workqueue_fn, NULL, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 667) btrfs_queue_work(wq, &end_io_wq->work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 668) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 669)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 670) blk_status_t btrfs_bio_wq_end_io(struct btrfs_fs_info *info, struct bio *bio,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 671) enum btrfs_wq_endio_type metadata)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 672) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 673) struct btrfs_end_io_wq *end_io_wq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 674)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 675) end_io_wq = kmem_cache_alloc(btrfs_end_io_wq_cache, GFP_NOFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 676) if (!end_io_wq)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 677) return BLK_STS_RESOURCE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 678)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 679) end_io_wq->private = bio->bi_private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 680) end_io_wq->end_io = bio->bi_end_io;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 681) end_io_wq->info = info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 682) end_io_wq->status = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 683) end_io_wq->bio = bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 684) end_io_wq->metadata = metadata;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 685)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 686) bio->bi_private = end_io_wq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 687) bio->bi_end_io = end_workqueue_bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 688) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 689) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 690)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 691) static void run_one_async_start(struct btrfs_work *work)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 692) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 693) struct async_submit_bio *async;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 694) blk_status_t ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 695)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 696) async = container_of(work, struct async_submit_bio, work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 697) ret = async->submit_bio_start(async->private_data, async->bio,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 698) async->bio_offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 699) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 700) async->status = ret;
^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) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 704) * In order to insert checksums into the metadata in large chunks, we wait
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 705) * until bio submission time. All the pages in the bio are checksummed and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 706) * sums are attached onto the ordered extent record.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 707) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 708) * At IO completion time the csums attached on the ordered extent record are
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 709) * inserted into the tree.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 710) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 711) static void run_one_async_done(struct btrfs_work *work)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 712) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 713) struct async_submit_bio *async;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 714) struct inode *inode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 715) blk_status_t ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 716)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 717) async = container_of(work, struct async_submit_bio, work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 718) inode = async->private_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 719)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 720) /* If an error occurred we just want to clean up the bio and move on */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 721) if (async->status) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 722) async->bio->bi_status = async->status;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 723) bio_endio(async->bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 724) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 725) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 726)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 727) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 728) * All of the bios that pass through here are from async helpers.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 729) * Use REQ_CGROUP_PUNT to issue them from the owning cgroup's context.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 730) * This changes nothing when cgroups aren't in use.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 731) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 732) async->bio->bi_opf |= REQ_CGROUP_PUNT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 733) ret = btrfs_map_bio(btrfs_sb(inode->i_sb), async->bio, async->mirror_num);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 734) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 735) async->bio->bi_status = ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 736) bio_endio(async->bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 737) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 738) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 739)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 740) static void run_one_async_free(struct btrfs_work *work)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 741) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 742) struct async_submit_bio *async;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 743)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 744) async = container_of(work, struct async_submit_bio, work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 745) kfree(async);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 746) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 747)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 748) blk_status_t btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 749) int mirror_num, unsigned long bio_flags,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 750) u64 bio_offset, void *private_data,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 751) extent_submit_bio_start_t *submit_bio_start)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 752) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 753) struct async_submit_bio *async;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 754)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 755) async = kmalloc(sizeof(*async), GFP_NOFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 756) if (!async)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 757) return BLK_STS_RESOURCE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 758)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 759) async->private_data = private_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 760) async->bio = bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 761) async->mirror_num = mirror_num;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 762) async->submit_bio_start = submit_bio_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 763)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 764) btrfs_init_work(&async->work, run_one_async_start, run_one_async_done,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 765) run_one_async_free);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 766)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 767) async->bio_offset = bio_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 768)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 769) async->status = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 770)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 771) if (op_is_sync(bio->bi_opf))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 772) btrfs_set_work_high_priority(&async->work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 773)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 774) btrfs_queue_work(fs_info->workers, &async->work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 775) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 776) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 777)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 778) static blk_status_t btree_csum_one_bio(struct bio *bio)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 779) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 780) struct bio_vec *bvec;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 781) struct btrfs_root *root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 782) int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 783) struct bvec_iter_all iter_all;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 784)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 785) ASSERT(!bio_flagged(bio, BIO_CLONED));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 786) bio_for_each_segment_all(bvec, bio, iter_all) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 787) root = BTRFS_I(bvec->bv_page->mapping->host)->root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 788) ret = csum_dirty_buffer(root->fs_info, bvec->bv_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 789) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 790) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 791) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 792)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 793) return errno_to_blk_status(ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 794) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 795)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 796) static blk_status_t btree_submit_bio_start(void *private_data, struct bio *bio,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 797) u64 bio_offset)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 798) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 799) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 800) * when we're called for a write, we're already in the async
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 801) * submission context. Just jump into btrfs_map_bio
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 802) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 803) return btree_csum_one_bio(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 804) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 805)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 806) static int check_async_write(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 807) struct btrfs_inode *bi)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 808) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 809) if (atomic_read(&bi->sync_writers))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 810) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 811) if (test_bit(BTRFS_FS_CSUM_IMPL_FAST, &fs_info->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 812) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 813) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 814) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 815)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 816) blk_status_t btrfs_submit_metadata_bio(struct inode *inode, struct bio *bio,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 817) int mirror_num, unsigned long bio_flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 818) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 819) struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 820) int async = check_async_write(fs_info, BTRFS_I(inode));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 821) blk_status_t ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 822)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 823) if (bio_op(bio) != REQ_OP_WRITE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 824) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 825) * called for a read, do the setup so that checksum validation
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 826) * can happen in the async kernel threads
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 827) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 828) ret = btrfs_bio_wq_end_io(fs_info, bio,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 829) BTRFS_WQ_ENDIO_METADATA);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 830) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 831) goto out_w_error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 832) ret = btrfs_map_bio(fs_info, bio, mirror_num);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 833) } else if (!async) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 834) ret = btree_csum_one_bio(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 835) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 836) goto out_w_error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 837) ret = btrfs_map_bio(fs_info, bio, mirror_num);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 838) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 839) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 840) * kthread helpers are used to submit writes so that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 841) * checksumming can happen in parallel across all CPUs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 842) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 843) ret = btrfs_wq_submit_bio(fs_info, bio, mirror_num, 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 844) 0, inode, btree_submit_bio_start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 845) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 846)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 847) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 848) goto out_w_error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 849) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 850)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 851) out_w_error:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 852) bio->bi_status = ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 853) bio_endio(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 854) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 855) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 856)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 857) #ifdef CONFIG_MIGRATION
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 858) static int btree_migratepage(struct address_space *mapping,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 859) struct page *newpage, struct page *page,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 860) enum migrate_mode mode)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 861) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 862) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 863) * we can't safely write a btree page from here,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 864) * we haven't done the locking hook
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 865) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 866) if (PageDirty(page))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 867) return -EAGAIN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 868) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 869) * Buffers may be managed in a filesystem specific way.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 870) * We must have no buffers or drop them.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 871) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 872) if (page_has_private(page) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 873) !try_to_release_page(page, GFP_KERNEL))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 874) return -EAGAIN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 875) return migrate_page(mapping, newpage, page, mode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 876) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 877) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 878)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 879)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 880) static int btree_writepages(struct address_space *mapping,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 881) struct writeback_control *wbc)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 882) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 883) struct btrfs_fs_info *fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 884) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 885)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 886) if (wbc->sync_mode == WB_SYNC_NONE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 887)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 888) if (wbc->for_kupdate)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 889) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 890)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 891) fs_info = BTRFS_I(mapping->host)->root->fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 892) /* this is a bit racy, but that's ok */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 893) ret = __percpu_counter_compare(&fs_info->dirty_metadata_bytes,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 894) BTRFS_DIRTY_METADATA_THRESH,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 895) fs_info->dirty_metadata_batch);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 896) if (ret < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 897) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 898) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 899) return btree_write_cache_pages(mapping, wbc);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 900) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 901)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 902) static int btree_releasepage(struct page *page, gfp_t gfp_flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 903) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 904) if (PageWriteback(page) || PageDirty(page))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 905) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 906)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 907) return try_release_extent_buffer(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 908) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 909)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 910) static void btree_invalidatepage(struct page *page, unsigned int offset,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 911) unsigned int length)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 912) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 913) struct extent_io_tree *tree;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 914) tree = &BTRFS_I(page->mapping->host)->io_tree;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 915) extent_invalidatepage(tree, page, offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 916) btree_releasepage(page, GFP_NOFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 917) if (PagePrivate(page)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 918) btrfs_warn(BTRFS_I(page->mapping->host)->root->fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 919) "page private not zero on page %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 920) (unsigned long long)page_offset(page));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 921) detach_page_private(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 922) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 923) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 924)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 925) static int btree_set_page_dirty(struct page *page)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 926) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 927) #ifdef DEBUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 928) struct extent_buffer *eb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 929)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 930) BUG_ON(!PagePrivate(page));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 931) eb = (struct extent_buffer *)page->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 932) BUG_ON(!eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 933) BUG_ON(!test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 934) BUG_ON(!atomic_read(&eb->refs));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 935) btrfs_assert_tree_locked(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 936) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 937) return __set_page_dirty_nobuffers(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 938) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 939)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 940) static const struct address_space_operations btree_aops = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 941) .writepages = btree_writepages,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 942) .releasepage = btree_releasepage,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 943) .invalidatepage = btree_invalidatepage,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 944) #ifdef CONFIG_MIGRATION
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 945) .migratepage = btree_migratepage,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 946) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 947) .set_page_dirty = btree_set_page_dirty,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 948) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 949)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 950) void readahead_tree_block(struct btrfs_fs_info *fs_info, u64 bytenr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 951) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 952) struct extent_buffer *buf = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 953) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 954)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 955) buf = btrfs_find_create_tree_block(fs_info, bytenr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 956) if (IS_ERR(buf))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 957) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 958)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 959) ret = read_extent_buffer_pages(buf, WAIT_NONE, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 960) if (ret < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 961) free_extent_buffer_stale(buf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 962) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 963) free_extent_buffer(buf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 964) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 965)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 966) struct extent_buffer *btrfs_find_create_tree_block(
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 967) struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 968) u64 bytenr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 969) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 970) if (btrfs_is_testing(fs_info))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 971) return alloc_test_extent_buffer(fs_info, bytenr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 972) return alloc_extent_buffer(fs_info, bytenr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 973) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 974)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 975) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 976) * Read tree block at logical address @bytenr and do variant basic but critical
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 977) * verification.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 978) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 979) * @parent_transid: expected transid of this tree block, skip check if 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 980) * @level: expected level, mandatory check
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 981) * @first_key: expected key in slot 0, skip check if NULL
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 982) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 983) struct extent_buffer *read_tree_block(struct btrfs_fs_info *fs_info, u64 bytenr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 984) u64 parent_transid, int level,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 985) struct btrfs_key *first_key)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 986) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 987) struct extent_buffer *buf = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 988) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 989)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 990) buf = btrfs_find_create_tree_block(fs_info, bytenr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 991) if (IS_ERR(buf))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 992) return buf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 993)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 994) ret = btree_read_extent_buffer_pages(buf, parent_transid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 995) level, first_key);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 996) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 997) free_extent_buffer_stale(buf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 998) return ERR_PTR(ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 999) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1000) return buf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1001)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1002) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1003)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1004) void btrfs_clean_tree_block(struct extent_buffer *buf)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1005) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1006) struct btrfs_fs_info *fs_info = buf->fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1007) if (btrfs_header_generation(buf) ==
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1008) fs_info->running_transaction->transid) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1009) btrfs_assert_tree_locked(buf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1010)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1011) if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &buf->bflags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1012) percpu_counter_add_batch(&fs_info->dirty_metadata_bytes,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1013) -buf->len,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1014) fs_info->dirty_metadata_batch);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1015) /* ugh, clear_extent_buffer_dirty needs to lock the page */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1016) btrfs_set_lock_blocking_write(buf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1017) clear_extent_buffer_dirty(buf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1018) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1019) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1020) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1021)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1022) static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1023) u64 objectid)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1024) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1025) bool dummy = test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1026) root->fs_info = fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1027) root->node = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1028) root->commit_root = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1029) root->state = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1030) root->orphan_cleanup_state = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1031)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1032) root->last_trans = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1033) root->highest_objectid = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1034) root->nr_delalloc_inodes = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1035) root->nr_ordered_extents = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1036) root->inode_tree = RB_ROOT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1037) INIT_RADIX_TREE(&root->delayed_nodes_tree, GFP_ATOMIC);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1038) root->block_rsv = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1039)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1040) INIT_LIST_HEAD(&root->dirty_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1041) INIT_LIST_HEAD(&root->root_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1042) INIT_LIST_HEAD(&root->delalloc_inodes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1043) INIT_LIST_HEAD(&root->delalloc_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1044) INIT_LIST_HEAD(&root->ordered_extents);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1045) INIT_LIST_HEAD(&root->ordered_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1046) INIT_LIST_HEAD(&root->reloc_dirty_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1047) INIT_LIST_HEAD(&root->logged_list[0]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1048) INIT_LIST_HEAD(&root->logged_list[1]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1049) spin_lock_init(&root->inode_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1050) spin_lock_init(&root->delalloc_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1051) spin_lock_init(&root->ordered_extent_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1052) spin_lock_init(&root->accounting_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1053) spin_lock_init(&root->log_extents_lock[0]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1054) spin_lock_init(&root->log_extents_lock[1]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1055) spin_lock_init(&root->qgroup_meta_rsv_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1056) mutex_init(&root->objectid_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1057) mutex_init(&root->log_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1058) mutex_init(&root->ordered_extent_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1059) mutex_init(&root->delalloc_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1060) init_waitqueue_head(&root->qgroup_flush_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1061) init_waitqueue_head(&root->log_writer_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1062) init_waitqueue_head(&root->log_commit_wait[0]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1063) init_waitqueue_head(&root->log_commit_wait[1]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1064) INIT_LIST_HEAD(&root->log_ctxs[0]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1065) INIT_LIST_HEAD(&root->log_ctxs[1]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1066) atomic_set(&root->log_commit[0], 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1067) atomic_set(&root->log_commit[1], 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1068) atomic_set(&root->log_writers, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1069) atomic_set(&root->log_batch, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1070) refcount_set(&root->refs, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1071) atomic_set(&root->snapshot_force_cow, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1072) atomic_set(&root->nr_swapfiles, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1073) root->log_transid = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1074) root->log_transid_committed = -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1075) root->last_log_commit = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1076) if (!dummy) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1077) extent_io_tree_init(fs_info, &root->dirty_log_pages,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1078) IO_TREE_ROOT_DIRTY_LOG_PAGES, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1079) extent_io_tree_init(fs_info, &root->log_csum_range,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1080) IO_TREE_LOG_CSUM_RANGE, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1081) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1082)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1083) memset(&root->root_key, 0, sizeof(root->root_key));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1084) memset(&root->root_item, 0, sizeof(root->root_item));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1085) memset(&root->defrag_progress, 0, sizeof(root->defrag_progress));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1086) root->root_key.objectid = objectid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1087) root->anon_dev = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1088)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1089) spin_lock_init(&root->root_item_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1090) btrfs_qgroup_init_swapped_blocks(&root->swapped_blocks);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1091) #ifdef CONFIG_BTRFS_DEBUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1092) INIT_LIST_HEAD(&root->leak_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1093) spin_lock(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1094) list_add_tail(&root->leak_list, &fs_info->allocated_roots);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1095) spin_unlock(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1096) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1097) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1098)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1099) static struct btrfs_root *btrfs_alloc_root(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1100) u64 objectid, gfp_t flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1101) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1102) struct btrfs_root *root = kzalloc(sizeof(*root), flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1103) if (root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1104) __setup_root(root, fs_info, objectid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1105) return root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1106) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1107)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1108) #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1109) /* Should only be used by the testing infrastructure */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1110) struct btrfs_root *btrfs_alloc_dummy_root(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1111) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1112) struct btrfs_root *root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1113)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1114) if (!fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1115) return ERR_PTR(-EINVAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1117) root = btrfs_alloc_root(fs_info, BTRFS_ROOT_TREE_OBJECTID, GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1118) if (!root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1119) return ERR_PTR(-ENOMEM);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1120)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1121) /* We don't use the stripesize in selftest, set it as sectorsize */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1122) root->alloc_bytenr = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1123)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1124) return root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1125) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1126) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1127)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1128) struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1129) u64 objectid)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1130) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1131) struct btrfs_fs_info *fs_info = trans->fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1132) struct extent_buffer *leaf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1133) struct btrfs_root *tree_root = fs_info->tree_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1134) struct btrfs_root *root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1135) struct btrfs_key key;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1136) unsigned int nofs_flag;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1137) int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1138)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1139) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1140) * We're holding a transaction handle, so use a NOFS memory allocation
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1141) * context to avoid deadlock if reclaim happens.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1142) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1143) nofs_flag = memalloc_nofs_save();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1144) root = btrfs_alloc_root(fs_info, objectid, GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1145) memalloc_nofs_restore(nofs_flag);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1146) if (!root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1147) return ERR_PTR(-ENOMEM);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1148)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1149) root->root_key.objectid = objectid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1150) root->root_key.type = BTRFS_ROOT_ITEM_KEY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1151) root->root_key.offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1152)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1153) leaf = btrfs_alloc_tree_block(trans, root, 0, objectid, NULL, 0, 0, 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1154) BTRFS_NESTING_NORMAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1155) if (IS_ERR(leaf)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1156) ret = PTR_ERR(leaf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1157) leaf = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1158) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1159) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1160)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1161) root->node = leaf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1162) btrfs_mark_buffer_dirty(leaf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1163)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1164) root->commit_root = btrfs_root_node(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1165) set_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1166)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1167) root->root_item.flags = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1168) root->root_item.byte_limit = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1169) btrfs_set_root_bytenr(&root->root_item, leaf->start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1170) btrfs_set_root_generation(&root->root_item, trans->transid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1171) btrfs_set_root_level(&root->root_item, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1172) btrfs_set_root_refs(&root->root_item, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1173) btrfs_set_root_used(&root->root_item, leaf->len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1174) btrfs_set_root_last_snapshot(&root->root_item, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1175) btrfs_set_root_dirid(&root->root_item, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1176) if (is_fstree(objectid))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1177) generate_random_guid(root->root_item.uuid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1178) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1179) export_guid(root->root_item.uuid, &guid_null);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1180) root->root_item.drop_level = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1181)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1182) key.objectid = objectid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1183) key.type = BTRFS_ROOT_ITEM_KEY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1184) key.offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1185) ret = btrfs_insert_root(trans, tree_root, &key, &root->root_item);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1186) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1187) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1188)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1189) btrfs_tree_unlock(leaf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1190)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1191) return root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1192)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1193) fail:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1194) if (leaf)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1195) btrfs_tree_unlock(leaf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1196) btrfs_put_root(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1197)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1198) return ERR_PTR(ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1199) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1200)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1201) static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1202) struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1203) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1204) struct btrfs_root *root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1205) struct extent_buffer *leaf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1206)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1207) root = btrfs_alloc_root(fs_info, BTRFS_TREE_LOG_OBJECTID, GFP_NOFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1208) if (!root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1209) return ERR_PTR(-ENOMEM);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1210)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1211) root->root_key.objectid = BTRFS_TREE_LOG_OBJECTID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1212) root->root_key.type = BTRFS_ROOT_ITEM_KEY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1213) root->root_key.offset = BTRFS_TREE_LOG_OBJECTID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1214)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1215) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1216) * DON'T set SHAREABLE bit for log trees.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1217) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1218) * Log trees are not exposed to user space thus can't be snapshotted,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1219) * and they go away before a real commit is actually done.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1220) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1221) * They do store pointers to file data extents, and those reference
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1222) * counts still get updated (along with back refs to the log tree).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1223) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1224)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1225) leaf = btrfs_alloc_tree_block(trans, root, 0, BTRFS_TREE_LOG_OBJECTID,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1226) NULL, 0, 0, 0, BTRFS_NESTING_NORMAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1227) if (IS_ERR(leaf)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1228) btrfs_put_root(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1229) return ERR_CAST(leaf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1230) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1231)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1232) root->node = leaf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1233)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1234) btrfs_mark_buffer_dirty(root->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1235) btrfs_tree_unlock(root->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1236) return root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1237) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1238)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1239) int btrfs_init_log_root_tree(struct btrfs_trans_handle *trans,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1240) struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1241) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1242) struct btrfs_root *log_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1243)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1244) log_root = alloc_log_tree(trans, fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1245) if (IS_ERR(log_root))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1246) return PTR_ERR(log_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1247) WARN_ON(fs_info->log_root_tree);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1248) fs_info->log_root_tree = log_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1249) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1250) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1251)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1252) int btrfs_add_log_tree(struct btrfs_trans_handle *trans,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1253) struct btrfs_root *root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1254) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1255) struct btrfs_fs_info *fs_info = root->fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1256) struct btrfs_root *log_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1257) struct btrfs_inode_item *inode_item;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1258)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1259) log_root = alloc_log_tree(trans, fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1260) if (IS_ERR(log_root))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1261) return PTR_ERR(log_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1262)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1263) log_root->last_trans = trans->transid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1264) log_root->root_key.offset = root->root_key.objectid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1265)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1266) inode_item = &log_root->root_item.inode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1267) btrfs_set_stack_inode_generation(inode_item, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1268) btrfs_set_stack_inode_size(inode_item, 3);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1269) btrfs_set_stack_inode_nlink(inode_item, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1270) btrfs_set_stack_inode_nbytes(inode_item,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1271) fs_info->nodesize);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1272) btrfs_set_stack_inode_mode(inode_item, S_IFDIR | 0755);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1273)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1274) btrfs_set_root_node(&log_root->root_item, log_root->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1275)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1276) WARN_ON(root->log_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1277) root->log_root = log_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1278) root->log_transid = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1279) root->log_transid_committed = -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1280) root->last_log_commit = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1281) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1282) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1283)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1284) static struct btrfs_root *read_tree_root_path(struct btrfs_root *tree_root,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1285) struct btrfs_path *path,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1286) struct btrfs_key *key)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1287) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1288) struct btrfs_root *root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1289) struct btrfs_fs_info *fs_info = tree_root->fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1290) u64 generation;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1291) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1292) int level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1293)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1294) root = btrfs_alloc_root(fs_info, key->objectid, GFP_NOFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1295) if (!root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1296) return ERR_PTR(-ENOMEM);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1297)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1298) ret = btrfs_find_root(tree_root, key, path,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1299) &root->root_item, &root->root_key);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1300) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1301) if (ret > 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1302) ret = -ENOENT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1303) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1304) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1305)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1306) generation = btrfs_root_generation(&root->root_item);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1307) level = btrfs_root_level(&root->root_item);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1308) root->node = read_tree_block(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1309) btrfs_root_bytenr(&root->root_item),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1310) generation, level, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1311) if (IS_ERR(root->node)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1312) ret = PTR_ERR(root->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1313) root->node = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1314) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1315) } else if (!btrfs_buffer_uptodate(root->node, generation, 0)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1316) ret = -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1317) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1318) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1319) root->commit_root = btrfs_root_node(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1320) return root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1321) fail:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1322) btrfs_put_root(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1323) return ERR_PTR(ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1324) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1325)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1326) struct btrfs_root *btrfs_read_tree_root(struct btrfs_root *tree_root,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1327) struct btrfs_key *key)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1328) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1329) struct btrfs_root *root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1330) struct btrfs_path *path;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1331)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1332) path = btrfs_alloc_path();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1333) if (!path)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1334) return ERR_PTR(-ENOMEM);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1335) root = read_tree_root_path(tree_root, path, key);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1336) btrfs_free_path(path);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1337)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1338) return root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1339) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1340)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1341) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1342) * Initialize subvolume root in-memory structure
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1343) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1344) * @anon_dev: anonymous device to attach to the root, if zero, allocate new
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1345) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1346) static int btrfs_init_fs_root(struct btrfs_root *root, dev_t anon_dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1347) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1348) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1349) unsigned int nofs_flag;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1350)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1351) root->free_ino_ctl = kzalloc(sizeof(*root->free_ino_ctl), GFP_NOFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1352) root->free_ino_pinned = kzalloc(sizeof(*root->free_ino_pinned),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1353) GFP_NOFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1354) if (!root->free_ino_pinned || !root->free_ino_ctl) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1355) ret = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1356) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1357) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1358)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1359) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1360) * We might be called under a transaction (e.g. indirect backref
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1361) * resolution) which could deadlock if it triggers memory reclaim
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1362) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1363) nofs_flag = memalloc_nofs_save();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1364) ret = btrfs_drew_lock_init(&root->snapshot_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1365) memalloc_nofs_restore(nofs_flag);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1366) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1367) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1368)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1369) if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1370) root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1371) set_bit(BTRFS_ROOT_SHAREABLE, &root->state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1372) btrfs_check_and_init_root_item(&root->root_item);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1373) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1374)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1375) btrfs_init_free_ino_ctl(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1376) spin_lock_init(&root->ino_cache_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1377) init_waitqueue_head(&root->ino_cache_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1378)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1379) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1380) * Don't assign anonymous block device to roots that are not exposed to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1381) * userspace, the id pool is limited to 1M
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1382) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1383) if (is_fstree(root->root_key.objectid) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1384) btrfs_root_refs(&root->root_item) > 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1385) if (!anon_dev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1386) ret = get_anon_bdev(&root->anon_dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1387) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1388) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1389) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1390) root->anon_dev = anon_dev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1391) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1392) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1393)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1394) mutex_lock(&root->objectid_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1395) ret = btrfs_find_highest_objectid(root,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1396) &root->highest_objectid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1397) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1398) mutex_unlock(&root->objectid_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1399) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1400) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1401)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1402) ASSERT(root->highest_objectid <= BTRFS_LAST_FREE_OBJECTID);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1403)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1404) mutex_unlock(&root->objectid_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1405)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1406) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1407) fail:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1408) /* The caller is responsible to call btrfs_free_fs_root */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1409) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1410) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1411)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1412) static struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1413) u64 root_id)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1414) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1415) struct btrfs_root *root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1416)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1417) spin_lock(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1418) root = radix_tree_lookup(&fs_info->fs_roots_radix,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1419) (unsigned long)root_id);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1420) if (root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1421) root = btrfs_grab_root(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1422) spin_unlock(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1423) return root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1424) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1425)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1426) static struct btrfs_root *btrfs_get_global_root(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1427) u64 objectid)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1428) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1429) if (objectid == BTRFS_ROOT_TREE_OBJECTID)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1430) return btrfs_grab_root(fs_info->tree_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1431) if (objectid == BTRFS_EXTENT_TREE_OBJECTID)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1432) return btrfs_grab_root(fs_info->extent_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1433) if (objectid == BTRFS_CHUNK_TREE_OBJECTID)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1434) return btrfs_grab_root(fs_info->chunk_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1435) if (objectid == BTRFS_DEV_TREE_OBJECTID)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1436) return btrfs_grab_root(fs_info->dev_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1437) if (objectid == BTRFS_CSUM_TREE_OBJECTID)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1438) return btrfs_grab_root(fs_info->csum_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1439) if (objectid == BTRFS_QUOTA_TREE_OBJECTID)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1440) return btrfs_grab_root(fs_info->quota_root) ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1441) fs_info->quota_root : ERR_PTR(-ENOENT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1442) if (objectid == BTRFS_UUID_TREE_OBJECTID)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1443) return btrfs_grab_root(fs_info->uuid_root) ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1444) fs_info->uuid_root : ERR_PTR(-ENOENT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1445) if (objectid == BTRFS_FREE_SPACE_TREE_OBJECTID)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1446) return btrfs_grab_root(fs_info->free_space_root) ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1447) fs_info->free_space_root : ERR_PTR(-ENOENT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1448) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1449) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1450)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1451) int btrfs_insert_fs_root(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1452) struct btrfs_root *root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1453) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1454) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1455)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1456) ret = radix_tree_preload(GFP_NOFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1457) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1458) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1459)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1460) spin_lock(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1461) ret = radix_tree_insert(&fs_info->fs_roots_radix,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1462) (unsigned long)root->root_key.objectid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1463) root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1464) if (ret == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1465) btrfs_grab_root(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1466) set_bit(BTRFS_ROOT_IN_RADIX, &root->state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1467) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1468) spin_unlock(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1469) radix_tree_preload_end();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1470)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1471) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1472) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1473)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1474) void btrfs_check_leaked_roots(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1475) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1476) #ifdef CONFIG_BTRFS_DEBUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1477) struct btrfs_root *root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1478)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1479) while (!list_empty(&fs_info->allocated_roots)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1480) char buf[BTRFS_ROOT_NAME_BUF_LEN];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1481)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1482) root = list_first_entry(&fs_info->allocated_roots,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1483) struct btrfs_root, leak_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1484) btrfs_err(fs_info, "leaked root %s refcount %d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1485) btrfs_root_name(&root->root_key, buf),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1486) refcount_read(&root->refs));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1487) while (refcount_read(&root->refs) > 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1488) btrfs_put_root(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1489) btrfs_put_root(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1490) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1491) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1492) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1493)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1494) void btrfs_free_fs_info(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1495) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1496) percpu_counter_destroy(&fs_info->dirty_metadata_bytes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1497) percpu_counter_destroy(&fs_info->delalloc_bytes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1498) percpu_counter_destroy(&fs_info->dio_bytes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1499) percpu_counter_destroy(&fs_info->dev_replace.bio_counter);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1500) btrfs_free_csum_hash(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1501) btrfs_free_stripe_hash_table(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1502) btrfs_free_ref_cache(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1503) kfree(fs_info->balance_ctl);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1504) kfree(fs_info->delayed_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1505) btrfs_put_root(fs_info->extent_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1506) btrfs_put_root(fs_info->tree_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1507) btrfs_put_root(fs_info->chunk_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1508) btrfs_put_root(fs_info->dev_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1509) btrfs_put_root(fs_info->csum_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1510) btrfs_put_root(fs_info->quota_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1511) btrfs_put_root(fs_info->uuid_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1512) btrfs_put_root(fs_info->free_space_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1513) btrfs_put_root(fs_info->fs_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1514) btrfs_put_root(fs_info->data_reloc_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1515) btrfs_check_leaked_roots(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1516) btrfs_extent_buffer_leak_debug_check(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1517) kfree(fs_info->super_copy);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1518) kfree(fs_info->super_for_commit);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1519) kvfree(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1520) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1521)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1522)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1523) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1524) * Get an in-memory reference of a root structure.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1525) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1526) * For essential trees like root/extent tree, we grab it from fs_info directly.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1527) * For subvolume trees, we check the cached filesystem roots first. If not
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1528) * found, then read it from disk and add it to cached fs roots.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1529) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1530) * Caller should release the root by calling btrfs_put_root() after the usage.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1531) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1532) * NOTE: Reloc and log trees can't be read by this function as they share the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1533) * same root objectid.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1534) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1535) * @objectid: root id
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1536) * @anon_dev: preallocated anonymous block device number for new roots,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1537) * pass 0 for new allocation.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1538) * @check_ref: whether to check root item references, If true, return -ENOENT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1539) * for orphan roots
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1540) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1541) static struct btrfs_root *btrfs_get_root_ref(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1542) u64 objectid, dev_t anon_dev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1543) bool check_ref)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1544) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1545) struct btrfs_root *root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1546) struct btrfs_path *path;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1547) struct btrfs_key key;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1548) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1549)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1550) root = btrfs_get_global_root(fs_info, objectid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1551) if (root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1552) return root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1553) again:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1554) root = btrfs_lookup_fs_root(fs_info, objectid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1555) if (root) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1556) /* Shouldn't get preallocated anon_dev for cached roots */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1557) ASSERT(!anon_dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1558) if (check_ref && btrfs_root_refs(&root->root_item) == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1559) btrfs_put_root(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1560) return ERR_PTR(-ENOENT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1561) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1562) return root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1563) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1564)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1565) key.objectid = objectid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1566) key.type = BTRFS_ROOT_ITEM_KEY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1567) key.offset = (u64)-1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1568) root = btrfs_read_tree_root(fs_info->tree_root, &key);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1569) if (IS_ERR(root))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1570) return root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1571)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1572) if (check_ref && btrfs_root_refs(&root->root_item) == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1573) ret = -ENOENT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1574) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1575) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1576)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1577) ret = btrfs_init_fs_root(root, anon_dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1578) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1579) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1580)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1581) path = btrfs_alloc_path();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1582) if (!path) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1583) ret = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1584) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1585) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1586) key.objectid = BTRFS_ORPHAN_OBJECTID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1587) key.type = BTRFS_ORPHAN_ITEM_KEY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1588) key.offset = objectid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1589)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1590) ret = btrfs_search_slot(NULL, fs_info->tree_root, &key, path, 0, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1591) btrfs_free_path(path);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1592) if (ret < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1593) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1594) if (ret == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1595) set_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1596)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1597) ret = btrfs_insert_fs_root(fs_info, root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1598) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1599) btrfs_put_root(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1600) if (ret == -EEXIST)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1601) goto again;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1602) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1603) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1604) return root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1605) fail:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1606) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1607) * If our caller provided us an anonymous device, then it's his
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1608) * responsability to free it in case we fail. So we have to set our
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1609) * root's anon_dev to 0 to avoid a double free, once by btrfs_put_root()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1610) * and once again by our caller.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1611) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1612) if (anon_dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1613) root->anon_dev = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1614) btrfs_put_root(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1615) return ERR_PTR(ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1616) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1617)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1618) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1619) * Get in-memory reference of a root structure
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1620) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1621) * @objectid: tree objectid
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1622) * @check_ref: if set, verify that the tree exists and the item has at least
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1623) * one reference
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1624) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1625) struct btrfs_root *btrfs_get_fs_root(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1626) u64 objectid, bool check_ref)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1627) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1628) return btrfs_get_root_ref(fs_info, objectid, 0, check_ref);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1629) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1630)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1631) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1632) * Get in-memory reference of a root structure, created as new, optionally pass
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1633) * the anonymous block device id
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1634) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1635) * @objectid: tree objectid
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1636) * @anon_dev: if zero, allocate a new anonymous block device or use the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1637) * parameter value
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1638) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1639) struct btrfs_root *btrfs_get_new_fs_root(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1640) u64 objectid, dev_t anon_dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1641) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1642) return btrfs_get_root_ref(fs_info, objectid, anon_dev, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1643) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1644)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1645) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1646) * btrfs_get_fs_root_commit_root - return a root for the given objectid
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1647) * @fs_info: the fs_info
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1648) * @objectid: the objectid we need to lookup
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1649) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1650) * This is exclusively used for backref walking, and exists specifically because
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1651) * of how qgroups does lookups. Qgroups will do a backref lookup at delayed ref
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1652) * creation time, which means we may have to read the tree_root in order to look
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1653) * up a fs root that is not in memory. If the root is not in memory we will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1654) * read the tree root commit root and look up the fs root from there. This is a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1655) * temporary root, it will not be inserted into the radix tree as it doesn't
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1656) * have the most uptodate information, it'll simply be discarded once the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1657) * backref code is finished using the root.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1658) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1659) struct btrfs_root *btrfs_get_fs_root_commit_root(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1660) struct btrfs_path *path,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1661) u64 objectid)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1662) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1663) struct btrfs_root *root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1664) struct btrfs_key key;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1665)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1666) ASSERT(path->search_commit_root && path->skip_locking);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1667)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1668) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1669) * This can return -ENOENT if we ask for a root that doesn't exist, but
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1670) * since this is called via the backref walking code we won't be looking
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1671) * up a root that doesn't exist, unless there's corruption. So if root
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1672) * != NULL just return it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1673) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1674) root = btrfs_get_global_root(fs_info, objectid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1675) if (root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1676) return root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1677)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1678) root = btrfs_lookup_fs_root(fs_info, objectid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1679) if (root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1680) return root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1681)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1682) key.objectid = objectid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1683) key.type = BTRFS_ROOT_ITEM_KEY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1684) key.offset = (u64)-1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1685) root = read_tree_root_path(fs_info->tree_root, path, &key);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1686) btrfs_release_path(path);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1687)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1688) return root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1689) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1690)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1691) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1692) * called by the kthread helper functions to finally call the bio end_io
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1693) * functions. This is where read checksum verification actually happens
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1694) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1695) static void end_workqueue_fn(struct btrfs_work *work)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1696) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1697) struct bio *bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1698) struct btrfs_end_io_wq *end_io_wq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1699)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1700) end_io_wq = container_of(work, struct btrfs_end_io_wq, work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1701) bio = end_io_wq->bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1702)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1703) bio->bi_status = end_io_wq->status;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1704) bio->bi_private = end_io_wq->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1705) bio->bi_end_io = end_io_wq->end_io;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1706) bio_endio(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1707) kmem_cache_free(btrfs_end_io_wq_cache, end_io_wq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1708) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1709)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1710) static int cleaner_kthread(void *arg)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1711) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1712) struct btrfs_root *root = arg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1713) struct btrfs_fs_info *fs_info = root->fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1714) int again;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1715)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1716) while (1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1717) again = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1718)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1719) set_bit(BTRFS_FS_CLEANER_RUNNING, &fs_info->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1720)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1721) /* Make the cleaner go to sleep early. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1722) if (btrfs_need_cleaner_sleep(fs_info))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1723) goto sleep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1724)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1725) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1726) * Do not do anything if we might cause open_ctree() to block
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1727) * before we have finished mounting the filesystem.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1728) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1729) if (!test_bit(BTRFS_FS_OPEN, &fs_info->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1730) goto sleep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1731)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1732) if (!mutex_trylock(&fs_info->cleaner_mutex))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1733) goto sleep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1734)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1735) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1736) * Avoid the problem that we change the status of the fs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1737) * during the above check and trylock.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1738) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1739) if (btrfs_need_cleaner_sleep(fs_info)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1740) mutex_unlock(&fs_info->cleaner_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1741) goto sleep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1742) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1743)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1744) btrfs_run_delayed_iputs(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1745)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1746) again = btrfs_clean_one_deleted_snapshot(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1747) mutex_unlock(&fs_info->cleaner_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1748)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1749) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1750) * The defragger has dealt with the R/O remount and umount,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1751) * needn't do anything special here.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1752) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1753) btrfs_run_defrag_inodes(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1754)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1755) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1756) * Acquires fs_info->delete_unused_bgs_mutex to avoid racing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1757) * with relocation (btrfs_relocate_chunk) and relocation
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1758) * acquires fs_info->cleaner_mutex (btrfs_relocate_block_group)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1759) * after acquiring fs_info->delete_unused_bgs_mutex. So we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1760) * can't hold, nor need to, fs_info->cleaner_mutex when deleting
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1761) * unused block groups.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1762) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1763) btrfs_delete_unused_bgs(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1764) sleep:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1765) clear_bit(BTRFS_FS_CLEANER_RUNNING, &fs_info->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1766) if (kthread_should_park())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1767) kthread_parkme();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1768) if (kthread_should_stop())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1769) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1770) if (!again) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1771) set_current_state(TASK_INTERRUPTIBLE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1772) schedule();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1773) __set_current_state(TASK_RUNNING);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1774) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1775) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1776) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1777)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1778) static int transaction_kthread(void *arg)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1779) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1780) struct btrfs_root *root = arg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1781) struct btrfs_fs_info *fs_info = root->fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1782) struct btrfs_trans_handle *trans;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1783) struct btrfs_transaction *cur;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1784) u64 transid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1785) time64_t now;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1786) unsigned long delay;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1787) bool cannot_commit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1788)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1789) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1790) cannot_commit = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1791) delay = HZ * fs_info->commit_interval;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1792) mutex_lock(&fs_info->transaction_kthread_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1793)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1794) spin_lock(&fs_info->trans_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1795) cur = fs_info->running_transaction;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1796) if (!cur) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1797) spin_unlock(&fs_info->trans_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1798) goto sleep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1799) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1800)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1801) now = ktime_get_seconds();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1802) if (cur->state < TRANS_STATE_COMMIT_START &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1803) (now < cur->start_time ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1804) now - cur->start_time < fs_info->commit_interval)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1805) spin_unlock(&fs_info->trans_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1806) delay = HZ * 5;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1807) goto sleep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1808) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1809) transid = cur->transid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1810) spin_unlock(&fs_info->trans_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1811)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1812) /* If the file system is aborted, this will always fail. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1813) trans = btrfs_attach_transaction(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1814) if (IS_ERR(trans)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1815) if (PTR_ERR(trans) != -ENOENT)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1816) cannot_commit = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1817) goto sleep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1818) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1819) if (transid == trans->transid) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1820) btrfs_commit_transaction(trans);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1821) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1822) btrfs_end_transaction(trans);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1823) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1824) sleep:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1825) wake_up_process(fs_info->cleaner_kthread);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1826) mutex_unlock(&fs_info->transaction_kthread_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1827)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1828) if (unlikely(test_bit(BTRFS_FS_STATE_ERROR,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1829) &fs_info->fs_state)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1830) btrfs_cleanup_transaction(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1831) if (!kthread_should_stop() &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1832) (!btrfs_transaction_blocked(fs_info) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1833) cannot_commit))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1834) schedule_timeout_interruptible(delay);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1835) } while (!kthread_should_stop());
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1836) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1837) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1838)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1839) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1840) * This will find the highest generation in the array of root backups. The
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1841) * index of the highest array is returned, or -EINVAL if we can't find
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1842) * anything.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1843) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1844) * We check to make sure the array is valid by comparing the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1845) * generation of the latest root in the array with the generation
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1846) * in the super block. If they don't match we pitch it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1847) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1848) static int find_newest_super_backup(struct btrfs_fs_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1849) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1850) const u64 newest_gen = btrfs_super_generation(info->super_copy);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1851) u64 cur;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1852) struct btrfs_root_backup *root_backup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1853) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1854)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1855) for (i = 0; i < BTRFS_NUM_BACKUP_ROOTS; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1856) root_backup = info->super_copy->super_roots + i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1857) cur = btrfs_backup_tree_root_gen(root_backup);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1858) if (cur == newest_gen)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1859) return i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1860) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1861)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1862) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1863) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1864)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1865) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1866) * copy all the root pointers into the super backup array.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1867) * this will bump the backup pointer by one when it is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1868) * done
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1869) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1870) static void backup_super_roots(struct btrfs_fs_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1871) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1872) const int next_backup = info->backup_root_index;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1873) struct btrfs_root_backup *root_backup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1874)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1875) root_backup = info->super_for_commit->super_roots + next_backup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1876)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1877) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1878) * make sure all of our padding and empty slots get zero filled
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1879) * regardless of which ones we use today
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1880) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1881) memset(root_backup, 0, sizeof(*root_backup));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1882)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1883) info->backup_root_index = (next_backup + 1) % BTRFS_NUM_BACKUP_ROOTS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1884)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1885) btrfs_set_backup_tree_root(root_backup, info->tree_root->node->start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1886) btrfs_set_backup_tree_root_gen(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1887) btrfs_header_generation(info->tree_root->node));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1888)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1889) btrfs_set_backup_tree_root_level(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1890) btrfs_header_level(info->tree_root->node));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1891)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1892) btrfs_set_backup_chunk_root(root_backup, info->chunk_root->node->start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1893) btrfs_set_backup_chunk_root_gen(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1894) btrfs_header_generation(info->chunk_root->node));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1895) btrfs_set_backup_chunk_root_level(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1896) btrfs_header_level(info->chunk_root->node));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1897)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1898) btrfs_set_backup_extent_root(root_backup, info->extent_root->node->start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1899) btrfs_set_backup_extent_root_gen(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1900) btrfs_header_generation(info->extent_root->node));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1901) btrfs_set_backup_extent_root_level(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1902) btrfs_header_level(info->extent_root->node));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1903)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1904) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1905) * we might commit during log recovery, which happens before we set
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1906) * the fs_root. Make sure it is valid before we fill it in.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1907) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1908) if (info->fs_root && info->fs_root->node) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1909) btrfs_set_backup_fs_root(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1910) info->fs_root->node->start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1911) btrfs_set_backup_fs_root_gen(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1912) btrfs_header_generation(info->fs_root->node));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1913) btrfs_set_backup_fs_root_level(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1914) btrfs_header_level(info->fs_root->node));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1915) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1916)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1917) btrfs_set_backup_dev_root(root_backup, info->dev_root->node->start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1918) btrfs_set_backup_dev_root_gen(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1919) btrfs_header_generation(info->dev_root->node));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1920) btrfs_set_backup_dev_root_level(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1921) btrfs_header_level(info->dev_root->node));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1922)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1923) btrfs_set_backup_csum_root(root_backup, info->csum_root->node->start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1924) btrfs_set_backup_csum_root_gen(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1925) btrfs_header_generation(info->csum_root->node));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1926) btrfs_set_backup_csum_root_level(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1927) btrfs_header_level(info->csum_root->node));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1928)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1929) btrfs_set_backup_total_bytes(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1930) btrfs_super_total_bytes(info->super_copy));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1931) btrfs_set_backup_bytes_used(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1932) btrfs_super_bytes_used(info->super_copy));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1933) btrfs_set_backup_num_devices(root_backup,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1934) btrfs_super_num_devices(info->super_copy));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1935)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1936) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1937) * if we don't copy this out to the super_copy, it won't get remembered
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1938) * for the next commit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1939) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1940) memcpy(&info->super_copy->super_roots,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1941) &info->super_for_commit->super_roots,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1942) sizeof(*root_backup) * BTRFS_NUM_BACKUP_ROOTS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1943) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1944)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1945) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1946) * read_backup_root - Reads a backup root based on the passed priority. Prio 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1947) * is the newest, prio 1/2/3 are 2nd newest/3rd newest/4th (oldest) backup roots
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1948) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1949) * fs_info - filesystem whose backup roots need to be read
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1950) * priority - priority of backup root required
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1951) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1952) * Returns backup root index on success and -EINVAL otherwise.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1953) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1954) static int read_backup_root(struct btrfs_fs_info *fs_info, u8 priority)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1955) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1956) int backup_index = find_newest_super_backup(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1957) struct btrfs_super_block *super = fs_info->super_copy;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1958) struct btrfs_root_backup *root_backup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1959)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1960) if (priority < BTRFS_NUM_BACKUP_ROOTS && backup_index >= 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1961) if (priority == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1962) return backup_index;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1963)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1964) backup_index = backup_index + BTRFS_NUM_BACKUP_ROOTS - priority;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1965) backup_index %= BTRFS_NUM_BACKUP_ROOTS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1966) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1967) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1968) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1969)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1970) root_backup = super->super_roots + backup_index;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1971)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1972) btrfs_set_super_generation(super,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1973) btrfs_backup_tree_root_gen(root_backup));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1974) btrfs_set_super_root(super, btrfs_backup_tree_root(root_backup));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1975) btrfs_set_super_root_level(super,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1976) btrfs_backup_tree_root_level(root_backup));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1977) btrfs_set_super_bytes_used(super, btrfs_backup_bytes_used(root_backup));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1978)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1979) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1980) * Fixme: the total bytes and num_devices need to match or we should
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1981) * need a fsck
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1982) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1983) btrfs_set_super_total_bytes(super, btrfs_backup_total_bytes(root_backup));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1984) btrfs_set_super_num_devices(super, btrfs_backup_num_devices(root_backup));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1985)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1986) return backup_index;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1987) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1988)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1989) /* helper to cleanup workers */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1990) static void btrfs_stop_all_workers(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1991) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1992) btrfs_destroy_workqueue(fs_info->fixup_workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1993) btrfs_destroy_workqueue(fs_info->delalloc_workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1994) btrfs_destroy_workqueue(fs_info->workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1995) btrfs_destroy_workqueue(fs_info->endio_workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1996) btrfs_destroy_workqueue(fs_info->endio_raid56_workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1997) btrfs_destroy_workqueue(fs_info->rmw_workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1998) btrfs_destroy_workqueue(fs_info->endio_write_workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1999) btrfs_destroy_workqueue(fs_info->endio_freespace_worker);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2000) btrfs_destroy_workqueue(fs_info->delayed_workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2001) btrfs_destroy_workqueue(fs_info->caching_workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2002) btrfs_destroy_workqueue(fs_info->readahead_workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2003) btrfs_destroy_workqueue(fs_info->flush_workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2004) btrfs_destroy_workqueue(fs_info->qgroup_rescan_workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2005) if (fs_info->discard_ctl.discard_workers)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2006) destroy_workqueue(fs_info->discard_ctl.discard_workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2007) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2008) * Now that all other work queues are destroyed, we can safely destroy
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2009) * the queues used for metadata I/O, since tasks from those other work
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2010) * queues can do metadata I/O operations.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2011) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2012) btrfs_destroy_workqueue(fs_info->endio_meta_workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2013) btrfs_destroy_workqueue(fs_info->endio_meta_write_workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2014) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2015)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2016) static void free_root_extent_buffers(struct btrfs_root *root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2017) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2018) if (root) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2019) free_extent_buffer(root->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2020) free_extent_buffer(root->commit_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2021) root->node = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2022) root->commit_root = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2023) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2024) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2025)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2026) /* helper to cleanup tree roots */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2027) static void free_root_pointers(struct btrfs_fs_info *info, bool free_chunk_root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2028) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2029) free_root_extent_buffers(info->tree_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2030)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2031) free_root_extent_buffers(info->dev_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2032) free_root_extent_buffers(info->extent_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2033) free_root_extent_buffers(info->csum_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2034) free_root_extent_buffers(info->quota_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2035) free_root_extent_buffers(info->uuid_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2036) free_root_extent_buffers(info->fs_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2037) free_root_extent_buffers(info->data_reloc_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2038) if (free_chunk_root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2039) free_root_extent_buffers(info->chunk_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2040) free_root_extent_buffers(info->free_space_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2041) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2042)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2043) void btrfs_put_root(struct btrfs_root *root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2044) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2045) if (!root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2046) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2047)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2048) if (refcount_dec_and_test(&root->refs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2049) WARN_ON(!RB_EMPTY_ROOT(&root->inode_tree));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2050) WARN_ON(test_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2051) if (root->anon_dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2052) free_anon_bdev(root->anon_dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2053) btrfs_drew_lock_destroy(&root->snapshot_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2054) free_root_extent_buffers(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2055) kfree(root->free_ino_ctl);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2056) kfree(root->free_ino_pinned);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2057) #ifdef CONFIG_BTRFS_DEBUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2058) spin_lock(&root->fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2059) list_del_init(&root->leak_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2060) spin_unlock(&root->fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2061) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2062) kfree(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2063) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2064) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2065)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2066) void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2067) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2068) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2069) struct btrfs_root *gang[8];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2070) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2071)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2072) while (!list_empty(&fs_info->dead_roots)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2073) gang[0] = list_entry(fs_info->dead_roots.next,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2074) struct btrfs_root, root_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2075) list_del(&gang[0]->root_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2076)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2077) if (test_bit(BTRFS_ROOT_IN_RADIX, &gang[0]->state))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2078) btrfs_drop_and_free_fs_root(fs_info, gang[0]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2079) btrfs_put_root(gang[0]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2080) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2081)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2082) while (1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2083) ret = radix_tree_gang_lookup(&fs_info->fs_roots_radix,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2084) (void **)gang, 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2085) ARRAY_SIZE(gang));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2086) if (!ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2087) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2088) for (i = 0; i < ret; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2089) btrfs_drop_and_free_fs_root(fs_info, gang[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2090) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2091) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2092)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2093) static void btrfs_init_scrub(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2094) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2095) mutex_init(&fs_info->scrub_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2096) atomic_set(&fs_info->scrubs_running, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2097) atomic_set(&fs_info->scrub_pause_req, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2098) atomic_set(&fs_info->scrubs_paused, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2099) atomic_set(&fs_info->scrub_cancel_req, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2100) init_waitqueue_head(&fs_info->scrub_pause_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2101) refcount_set(&fs_info->scrub_workers_refcnt, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2102) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2103)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2104) static void btrfs_init_balance(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2105) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2106) spin_lock_init(&fs_info->balance_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2107) mutex_init(&fs_info->balance_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2108) atomic_set(&fs_info->balance_pause_req, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2109) atomic_set(&fs_info->balance_cancel_req, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2110) fs_info->balance_ctl = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2111) init_waitqueue_head(&fs_info->balance_wait_q);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2112) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2113)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2114) static void btrfs_init_btree_inode(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2115) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2116) struct inode *inode = fs_info->btree_inode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2117)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2118) inode->i_ino = BTRFS_BTREE_INODE_OBJECTID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2119) set_nlink(inode, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2120) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2121) * we set the i_size on the btree inode to the max possible int.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2122) * the real end of the address space is determined by all of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2123) * the devices in the system
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2124) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2125) inode->i_size = OFFSET_MAX;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2126) inode->i_mapping->a_ops = &btree_aops;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2127)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2128) RB_CLEAR_NODE(&BTRFS_I(inode)->rb_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2129) extent_io_tree_init(fs_info, &BTRFS_I(inode)->io_tree,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2130) IO_TREE_BTREE_INODE_IO, inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2131) BTRFS_I(inode)->io_tree.track_uptodate = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2132) extent_map_tree_init(&BTRFS_I(inode)->extent_tree);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2133)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2134) BTRFS_I(inode)->root = btrfs_grab_root(fs_info->tree_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2135) memset(&BTRFS_I(inode)->location, 0, sizeof(struct btrfs_key));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2136) set_bit(BTRFS_INODE_DUMMY, &BTRFS_I(inode)->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2137) btrfs_insert_inode_hash(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2138) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2139)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2140) static void btrfs_init_dev_replace_locks(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2141) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2142) mutex_init(&fs_info->dev_replace.lock_finishing_cancel_unmount);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2143) init_rwsem(&fs_info->dev_replace.rwsem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2144) init_waitqueue_head(&fs_info->dev_replace.replace_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2145) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2146)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2147) static void btrfs_init_qgroup(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2148) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2149) spin_lock_init(&fs_info->qgroup_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2150) mutex_init(&fs_info->qgroup_ioctl_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2151) fs_info->qgroup_tree = RB_ROOT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2152) INIT_LIST_HEAD(&fs_info->dirty_qgroups);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2153) fs_info->qgroup_seq = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2154) fs_info->qgroup_ulist = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2155) fs_info->qgroup_rescan_running = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2156) mutex_init(&fs_info->qgroup_rescan_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2157) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2158)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2159) static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2160) struct btrfs_fs_devices *fs_devices)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2161) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2162) u32 max_active = fs_info->thread_pool_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2163) unsigned int flags = WQ_MEM_RECLAIM | WQ_FREEZABLE | WQ_UNBOUND;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2164)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2165) fs_info->workers =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2166) btrfs_alloc_workqueue(fs_info, "worker",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2167) flags | WQ_HIGHPRI, max_active, 16);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2168)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2169) fs_info->delalloc_workers =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2170) btrfs_alloc_workqueue(fs_info, "delalloc",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2171) flags, max_active, 2);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2172)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2173) fs_info->flush_workers =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2174) btrfs_alloc_workqueue(fs_info, "flush_delalloc",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2175) flags, max_active, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2176)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2177) fs_info->caching_workers =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2178) btrfs_alloc_workqueue(fs_info, "cache", flags, max_active, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2179)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2180) fs_info->fixup_workers =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2181) btrfs_alloc_workqueue(fs_info, "fixup", flags, 1, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2182)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2183) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2184) * endios are largely parallel and should have a very
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2185) * low idle thresh
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2186) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2187) fs_info->endio_workers =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2188) btrfs_alloc_workqueue(fs_info, "endio", flags, max_active, 4);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2189) fs_info->endio_meta_workers =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2190) btrfs_alloc_workqueue(fs_info, "endio-meta", flags,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2191) max_active, 4);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2192) fs_info->endio_meta_write_workers =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2193) btrfs_alloc_workqueue(fs_info, "endio-meta-write", flags,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2194) max_active, 2);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2195) fs_info->endio_raid56_workers =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2196) btrfs_alloc_workqueue(fs_info, "endio-raid56", flags,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2197) max_active, 4);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2198) fs_info->rmw_workers =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2199) btrfs_alloc_workqueue(fs_info, "rmw", flags, max_active, 2);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2200) fs_info->endio_write_workers =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2201) btrfs_alloc_workqueue(fs_info, "endio-write", flags,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2202) max_active, 2);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2203) fs_info->endio_freespace_worker =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2204) btrfs_alloc_workqueue(fs_info, "freespace-write", flags,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2205) max_active, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2206) fs_info->delayed_workers =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2207) btrfs_alloc_workqueue(fs_info, "delayed-meta", flags,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2208) max_active, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2209) fs_info->readahead_workers =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2210) btrfs_alloc_workqueue(fs_info, "readahead", flags,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2211) max_active, 2);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2212) fs_info->qgroup_rescan_workers =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2213) btrfs_alloc_workqueue(fs_info, "qgroup-rescan", flags, 1, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2214) fs_info->discard_ctl.discard_workers =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2215) alloc_workqueue("btrfs_discard", WQ_UNBOUND | WQ_FREEZABLE, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2216)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2217) if (!(fs_info->workers && fs_info->delalloc_workers &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2218) fs_info->flush_workers &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2219) fs_info->endio_workers && fs_info->endio_meta_workers &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2220) fs_info->endio_meta_write_workers &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2221) fs_info->endio_write_workers && fs_info->endio_raid56_workers &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2222) fs_info->endio_freespace_worker && fs_info->rmw_workers &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2223) fs_info->caching_workers && fs_info->readahead_workers &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2224) fs_info->fixup_workers && fs_info->delayed_workers &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2225) fs_info->qgroup_rescan_workers &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2226) fs_info->discard_ctl.discard_workers)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2227) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2228) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2229)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2230) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2231) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2232)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2233) static int btrfs_init_csum_hash(struct btrfs_fs_info *fs_info, u16 csum_type)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2234) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2235) struct crypto_shash *csum_shash;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2236) const char *csum_driver = btrfs_super_csum_driver(csum_type);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2237)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2238) csum_shash = crypto_alloc_shash(csum_driver, 0, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2239)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2240) if (IS_ERR(csum_shash)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2241) btrfs_err(fs_info, "error allocating %s hash for checksum",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2242) csum_driver);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2243) return PTR_ERR(csum_shash);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2244) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2245)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2246) fs_info->csum_shash = csum_shash;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2247)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2248) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2249) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2250)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2251) static int btrfs_replay_log(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2252) struct btrfs_fs_devices *fs_devices)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2253) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2254) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2255) struct btrfs_root *log_tree_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2256) struct btrfs_super_block *disk_super = fs_info->super_copy;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2257) u64 bytenr = btrfs_super_log_root(disk_super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2258) int level = btrfs_super_log_root_level(disk_super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2259)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2260) if (fs_devices->rw_devices == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2261) btrfs_warn(fs_info, "log replay required on RO media");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2262) return -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2263) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2264)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2265) log_tree_root = btrfs_alloc_root(fs_info, BTRFS_TREE_LOG_OBJECTID,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2266) GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2267) if (!log_tree_root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2268) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2269)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2270) log_tree_root->node = read_tree_block(fs_info, bytenr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2271) fs_info->generation + 1,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2272) level, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2273) if (IS_ERR(log_tree_root->node)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2274) btrfs_warn(fs_info, "failed to read log tree");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2275) ret = PTR_ERR(log_tree_root->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2276) log_tree_root->node = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2277) btrfs_put_root(log_tree_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2278) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2279) } else if (!extent_buffer_uptodate(log_tree_root->node)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2280) btrfs_err(fs_info, "failed to read log tree");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2281) btrfs_put_root(log_tree_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2282) return -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2283) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2284) /* returns with log_tree_root freed on success */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2285) ret = btrfs_recover_log_trees(log_tree_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2286) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2287) btrfs_handle_fs_error(fs_info, ret,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2288) "Failed to recover log tree");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2289) btrfs_put_root(log_tree_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2290) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2291) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2292)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2293) if (sb_rdonly(fs_info->sb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2294) ret = btrfs_commit_super(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2295) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2296) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2297) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2298)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2299) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2300) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2301)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2302) static int btrfs_read_roots(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2303) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2304) struct btrfs_root *tree_root = fs_info->tree_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2305) struct btrfs_root *root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2306) struct btrfs_key location;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2307) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2308)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2309) BUG_ON(!fs_info->tree_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2310)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2311) location.objectid = BTRFS_EXTENT_TREE_OBJECTID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2312) location.type = BTRFS_ROOT_ITEM_KEY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2313) location.offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2314)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2315) root = btrfs_read_tree_root(tree_root, &location);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2316) if (IS_ERR(root)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2317) ret = PTR_ERR(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2318) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2319) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2320) set_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2321) fs_info->extent_root = root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2322)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2323) location.objectid = BTRFS_DEV_TREE_OBJECTID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2324) root = btrfs_read_tree_root(tree_root, &location);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2325) if (IS_ERR(root)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2326) ret = PTR_ERR(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2327) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2328) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2329) set_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2330) fs_info->dev_root = root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2331) btrfs_init_devices_late(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2332)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2333) location.objectid = BTRFS_CSUM_TREE_OBJECTID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2334) root = btrfs_read_tree_root(tree_root, &location);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2335) if (IS_ERR(root)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2336) ret = PTR_ERR(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2337) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2338) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2339) set_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2340) fs_info->csum_root = root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2341)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2342) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2343) * This tree can share blocks with some other fs tree during relocation
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2344) * and we need a proper setup by btrfs_get_fs_root
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2345) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2346) root = btrfs_get_fs_root(tree_root->fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2347) BTRFS_DATA_RELOC_TREE_OBJECTID, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2348) if (IS_ERR(root)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2349) ret = PTR_ERR(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2350) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2351) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2352) set_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2353) fs_info->data_reloc_root = root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2354)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2355) location.objectid = BTRFS_QUOTA_TREE_OBJECTID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2356) root = btrfs_read_tree_root(tree_root, &location);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2357) if (!IS_ERR(root)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2358) set_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2359) set_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2360) fs_info->quota_root = root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2361) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2362)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2363) location.objectid = BTRFS_UUID_TREE_OBJECTID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2364) root = btrfs_read_tree_root(tree_root, &location);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2365) if (IS_ERR(root)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2366) ret = PTR_ERR(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2367) if (ret != -ENOENT)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2368) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2369) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2370) set_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2371) fs_info->uuid_root = root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2372) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2373)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2374) if (btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2375) location.objectid = BTRFS_FREE_SPACE_TREE_OBJECTID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2376) root = btrfs_read_tree_root(tree_root, &location);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2377) if (IS_ERR(root)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2378) ret = PTR_ERR(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2379) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2380) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2381) set_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2382) fs_info->free_space_root = root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2383) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2384)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2385) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2386) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2387) btrfs_warn(fs_info, "failed to read root (objectid=%llu): %d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2388) location.objectid, ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2389) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2390) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2391)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2392) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2393) * Real super block validation
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2394) * NOTE: super csum type and incompat features will not be checked here.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2395) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2396) * @sb: super block to check
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2397) * @mirror_num: the super block number to check its bytenr:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2398) * 0 the primary (1st) sb
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2399) * 1, 2 2nd and 3rd backup copy
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2400) * -1 skip bytenr check
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2401) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2402) static int validate_super(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2403) struct btrfs_super_block *sb, int mirror_num)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2404) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2405) u64 nodesize = btrfs_super_nodesize(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2406) u64 sectorsize = btrfs_super_sectorsize(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2407) int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2408)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2409) if (btrfs_super_magic(sb) != BTRFS_MAGIC) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2410) btrfs_err(fs_info, "no valid FS found");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2411) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2412) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2413) if (btrfs_super_flags(sb) & ~BTRFS_SUPER_FLAG_SUPP) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2414) btrfs_err(fs_info, "unrecognized or unsupported super flag: %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2415) btrfs_super_flags(sb) & ~BTRFS_SUPER_FLAG_SUPP);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2416) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2417) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2418) if (btrfs_super_root_level(sb) >= BTRFS_MAX_LEVEL) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2419) btrfs_err(fs_info, "tree_root level too big: %d >= %d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2420) btrfs_super_root_level(sb), BTRFS_MAX_LEVEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2421) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2422) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2423) if (btrfs_super_chunk_root_level(sb) >= BTRFS_MAX_LEVEL) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2424) btrfs_err(fs_info, "chunk_root level too big: %d >= %d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2425) btrfs_super_chunk_root_level(sb), BTRFS_MAX_LEVEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2426) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2427) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2428) if (btrfs_super_log_root_level(sb) >= BTRFS_MAX_LEVEL) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2429) btrfs_err(fs_info, "log_root level too big: %d >= %d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2430) btrfs_super_log_root_level(sb), BTRFS_MAX_LEVEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2431) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2432) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2433)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2434) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2435) * Check sectorsize and nodesize first, other check will need it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2436) * Check all possible sectorsize(4K, 8K, 16K, 32K, 64K) here.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2437) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2438) if (!is_power_of_2(sectorsize) || sectorsize < 4096 ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2439) sectorsize > BTRFS_MAX_METADATA_BLOCKSIZE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2440) btrfs_err(fs_info, "invalid sectorsize %llu", sectorsize);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2441) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2442) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2443) /* Only PAGE SIZE is supported yet */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2444) if (sectorsize != PAGE_SIZE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2445) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2446) "sectorsize %llu not supported yet, only support %lu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2447) sectorsize, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2448) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2449) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2450) if (!is_power_of_2(nodesize) || nodesize < sectorsize ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2451) nodesize > BTRFS_MAX_METADATA_BLOCKSIZE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2452) btrfs_err(fs_info, "invalid nodesize %llu", nodesize);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2453) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2454) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2455) if (nodesize != le32_to_cpu(sb->__unused_leafsize)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2456) btrfs_err(fs_info, "invalid leafsize %u, should be %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2457) le32_to_cpu(sb->__unused_leafsize), nodesize);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2458) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2459) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2460)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2461) /* Root alignment check */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2462) if (!IS_ALIGNED(btrfs_super_root(sb), sectorsize)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2463) btrfs_warn(fs_info, "tree_root block unaligned: %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2464) btrfs_super_root(sb));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2465) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2466) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2467) if (!IS_ALIGNED(btrfs_super_chunk_root(sb), sectorsize)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2468) btrfs_warn(fs_info, "chunk_root block unaligned: %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2469) btrfs_super_chunk_root(sb));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2470) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2471) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2472) if (!IS_ALIGNED(btrfs_super_log_root(sb), sectorsize)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2473) btrfs_warn(fs_info, "log_root block unaligned: %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2474) btrfs_super_log_root(sb));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2475) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2476) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2477)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2478) if (memcmp(fs_info->fs_devices->fsid, fs_info->super_copy->fsid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2479) BTRFS_FSID_SIZE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2480) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2481) "superblock fsid doesn't match fsid of fs_devices: %pU != %pU",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2482) fs_info->super_copy->fsid, fs_info->fs_devices->fsid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2483) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2484) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2485)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2486) if (btrfs_fs_incompat(fs_info, METADATA_UUID) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2487) memcmp(fs_info->fs_devices->metadata_uuid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2488) fs_info->super_copy->metadata_uuid, BTRFS_FSID_SIZE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2489) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2490) "superblock metadata_uuid doesn't match metadata uuid of fs_devices: %pU != %pU",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2491) fs_info->super_copy->metadata_uuid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2492) fs_info->fs_devices->metadata_uuid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2493) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2494) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2495)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2496) if (memcmp(fs_info->fs_devices->metadata_uuid, sb->dev_item.fsid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2497) BTRFS_FSID_SIZE) != 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2498) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2499) "dev_item UUID does not match metadata fsid: %pU != %pU",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2500) fs_info->fs_devices->metadata_uuid, sb->dev_item.fsid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2501) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2502) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2503)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2504) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2505) * Hint to catch really bogus numbers, bitflips or so, more exact checks are
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2506) * done later
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2507) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2508) if (btrfs_super_bytes_used(sb) < 6 * btrfs_super_nodesize(sb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2509) btrfs_err(fs_info, "bytes_used is too small %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2510) btrfs_super_bytes_used(sb));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2511) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2512) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2513) if (!is_power_of_2(btrfs_super_stripesize(sb))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2514) btrfs_err(fs_info, "invalid stripesize %u",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2515) btrfs_super_stripesize(sb));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2516) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2517) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2518) if (btrfs_super_num_devices(sb) > (1UL << 31))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2519) btrfs_warn(fs_info, "suspicious number of devices: %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2520) btrfs_super_num_devices(sb));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2521) if (btrfs_super_num_devices(sb) == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2522) btrfs_err(fs_info, "number of devices is 0");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2523) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2524) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2525)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2526) if (mirror_num >= 0 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2527) btrfs_super_bytenr(sb) != btrfs_sb_offset(mirror_num)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2528) btrfs_err(fs_info, "super offset mismatch %llu != %u",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2529) btrfs_super_bytenr(sb), BTRFS_SUPER_INFO_OFFSET);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2530) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2531) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2532)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2533) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2534) * Obvious sys_chunk_array corruptions, it must hold at least one key
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2535) * and one chunk
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2536) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2537) if (btrfs_super_sys_array_size(sb) > BTRFS_SYSTEM_CHUNK_ARRAY_SIZE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2538) btrfs_err(fs_info, "system chunk array too big %u > %u",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2539) btrfs_super_sys_array_size(sb),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2540) BTRFS_SYSTEM_CHUNK_ARRAY_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2541) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2542) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2543) if (btrfs_super_sys_array_size(sb) < sizeof(struct btrfs_disk_key)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2544) + sizeof(struct btrfs_chunk)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2545) btrfs_err(fs_info, "system chunk array too small %u < %zu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2546) btrfs_super_sys_array_size(sb),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2547) sizeof(struct btrfs_disk_key)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2548) + sizeof(struct btrfs_chunk));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2549) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2550) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2551)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2552) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2553) * The generation is a global counter, we'll trust it more than the others
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2554) * but it's still possible that it's the one that's wrong.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2555) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2556) if (btrfs_super_generation(sb) < btrfs_super_chunk_root_generation(sb))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2557) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2558) "suspicious: generation < chunk_root_generation: %llu < %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2559) btrfs_super_generation(sb),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2560) btrfs_super_chunk_root_generation(sb));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2561) if (btrfs_super_generation(sb) < btrfs_super_cache_generation(sb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2562) && btrfs_super_cache_generation(sb) != (u64)-1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2563) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2564) "suspicious: generation < cache_generation: %llu < %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2565) btrfs_super_generation(sb),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2566) btrfs_super_cache_generation(sb));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2567)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2568) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2569) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2570)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2571) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2572) * Validation of super block at mount time.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2573) * Some checks already done early at mount time, like csum type and incompat
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2574) * flags will be skipped.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2575) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2576) static int btrfs_validate_mount_super(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2577) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2578) return validate_super(fs_info, fs_info->super_copy, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2579) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2580)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2581) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2582) * Validation of super block at write time.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2583) * Some checks like bytenr check will be skipped as their values will be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2584) * overwritten soon.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2585) * Extra checks like csum type and incompat flags will be done here.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2586) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2587) static int btrfs_validate_write_super(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2588) struct btrfs_super_block *sb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2589) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2590) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2591)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2592) ret = validate_super(fs_info, sb, -1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2593) if (ret < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2594) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2595) if (!btrfs_supported_super_csum(btrfs_super_csum_type(sb))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2596) ret = -EUCLEAN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2597) btrfs_err(fs_info, "invalid csum type, has %u want %u",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2598) btrfs_super_csum_type(sb), BTRFS_CSUM_TYPE_CRC32);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2599) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2600) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2601) if (btrfs_super_incompat_flags(sb) & ~BTRFS_FEATURE_INCOMPAT_SUPP) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2602) ret = -EUCLEAN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2603) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2604) "invalid incompat flags, has 0x%llx valid mask 0x%llx",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2605) btrfs_super_incompat_flags(sb),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2606) (unsigned long long)BTRFS_FEATURE_INCOMPAT_SUPP);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2607) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2608) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2609) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2610) if (ret < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2611) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2612) "super block corruption detected before writing it to disk");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2613) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2614) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2615)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2616) static int __cold init_tree_roots(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2617) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2618) int backup_index = find_newest_super_backup(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2619) struct btrfs_super_block *sb = fs_info->super_copy;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2620) struct btrfs_root *tree_root = fs_info->tree_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2621) bool handle_error = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2622) int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2623) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2624)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2625) for (i = 0; i < BTRFS_NUM_BACKUP_ROOTS; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2626) u64 generation;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2627) int level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2628)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2629) if (handle_error) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2630) if (!IS_ERR(tree_root->node))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2631) free_extent_buffer(tree_root->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2632) tree_root->node = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2633)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2634) if (!btrfs_test_opt(fs_info, USEBACKUPROOT))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2635) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2636)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2637) free_root_pointers(fs_info, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2638)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2639) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2640) * Don't use the log in recovery mode, it won't be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2641) * valid
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2642) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2643) btrfs_set_super_log_root(sb, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2644)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2645) /* We can't trust the free space cache either */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2646) btrfs_set_opt(fs_info->mount_opt, CLEAR_CACHE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2647)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2648) ret = read_backup_root(fs_info, i);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2649) backup_index = ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2650) if (ret < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2651) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2652) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2653) generation = btrfs_super_generation(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2654) level = btrfs_super_root_level(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2655) tree_root->node = read_tree_block(fs_info, btrfs_super_root(sb),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2656) generation, level, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2657) if (IS_ERR(tree_root->node)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2658) handle_error = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2659) ret = PTR_ERR(tree_root->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2660) tree_root->node = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2661) btrfs_warn(fs_info, "couldn't read tree root");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2662) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2663)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2664) } else if (!extent_buffer_uptodate(tree_root->node)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2665) handle_error = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2666) ret = -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2667) btrfs_warn(fs_info, "error while reading tree root");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2668) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2669) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2670)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2671) btrfs_set_root_node(&tree_root->root_item, tree_root->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2672) tree_root->commit_root = btrfs_root_node(tree_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2673) btrfs_set_root_refs(&tree_root->root_item, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2674)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2675) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2676) * No need to hold btrfs_root::objectid_mutex since the fs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2677) * hasn't been fully initialised and we are the only user
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2678) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2679) ret = btrfs_find_highest_objectid(tree_root,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2680) &tree_root->highest_objectid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2681) if (ret < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2682) handle_error = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2683) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2684) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2685)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2686) ASSERT(tree_root->highest_objectid <= BTRFS_LAST_FREE_OBJECTID);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2687)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2688) ret = btrfs_read_roots(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2689) if (ret < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2690) handle_error = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2691) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2692) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2693)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2694) /* All successful */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2695) fs_info->generation = generation;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2696) fs_info->last_trans_committed = generation;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2697)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2698) /* Always begin writing backup roots after the one being used */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2699) if (backup_index < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2700) fs_info->backup_root_index = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2701) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2702) fs_info->backup_root_index = backup_index + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2703) fs_info->backup_root_index %= BTRFS_NUM_BACKUP_ROOTS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2704) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2705) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2706) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2707)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2708) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2709) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2710)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2711) void btrfs_init_fs_info(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2712) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2713) INIT_RADIX_TREE(&fs_info->fs_roots_radix, GFP_ATOMIC);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2714) INIT_RADIX_TREE(&fs_info->buffer_radix, GFP_ATOMIC);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2715) INIT_LIST_HEAD(&fs_info->trans_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2716) INIT_LIST_HEAD(&fs_info->dead_roots);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2717) INIT_LIST_HEAD(&fs_info->delayed_iputs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2718) INIT_LIST_HEAD(&fs_info->delalloc_roots);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2719) INIT_LIST_HEAD(&fs_info->caching_block_groups);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2720) spin_lock_init(&fs_info->delalloc_root_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2721) spin_lock_init(&fs_info->trans_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2722) spin_lock_init(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2723) spin_lock_init(&fs_info->delayed_iput_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2724) spin_lock_init(&fs_info->defrag_inodes_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2725) spin_lock_init(&fs_info->super_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2726) spin_lock_init(&fs_info->buffer_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2727) spin_lock_init(&fs_info->unused_bgs_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2728) rwlock_init(&fs_info->tree_mod_log_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2729) mutex_init(&fs_info->unused_bg_unpin_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2730) mutex_init(&fs_info->delete_unused_bgs_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2731) mutex_init(&fs_info->reloc_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2732) mutex_init(&fs_info->delalloc_root_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2733) seqlock_init(&fs_info->profiles_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2734)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2735) INIT_LIST_HEAD(&fs_info->dirty_cowonly_roots);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2736) INIT_LIST_HEAD(&fs_info->space_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2737) INIT_LIST_HEAD(&fs_info->tree_mod_seq_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2738) INIT_LIST_HEAD(&fs_info->unused_bgs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2739) #ifdef CONFIG_BTRFS_DEBUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2740) INIT_LIST_HEAD(&fs_info->allocated_roots);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2741) INIT_LIST_HEAD(&fs_info->allocated_ebs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2742) spin_lock_init(&fs_info->eb_leak_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2743) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2744) extent_map_tree_init(&fs_info->mapping_tree);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2745) btrfs_init_block_rsv(&fs_info->global_block_rsv,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2746) BTRFS_BLOCK_RSV_GLOBAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2747) btrfs_init_block_rsv(&fs_info->trans_block_rsv, BTRFS_BLOCK_RSV_TRANS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2748) btrfs_init_block_rsv(&fs_info->chunk_block_rsv, BTRFS_BLOCK_RSV_CHUNK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2749) btrfs_init_block_rsv(&fs_info->empty_block_rsv, BTRFS_BLOCK_RSV_EMPTY);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2750) btrfs_init_block_rsv(&fs_info->delayed_block_rsv,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2751) BTRFS_BLOCK_RSV_DELOPS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2752) btrfs_init_block_rsv(&fs_info->delayed_refs_rsv,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2753) BTRFS_BLOCK_RSV_DELREFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2754)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2755) atomic_set(&fs_info->async_delalloc_pages, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2756) atomic_set(&fs_info->defrag_running, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2757) atomic_set(&fs_info->reada_works_cnt, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2758) atomic_set(&fs_info->nr_delayed_iputs, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2759) atomic64_set(&fs_info->tree_mod_seq, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2760) fs_info->max_inline = BTRFS_DEFAULT_MAX_INLINE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2761) fs_info->metadata_ratio = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2762) fs_info->defrag_inodes = RB_ROOT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2763) atomic64_set(&fs_info->free_chunk_space, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2764) fs_info->tree_mod_log = RB_ROOT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2765) fs_info->commit_interval = BTRFS_DEFAULT_COMMIT_INTERVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2766) fs_info->avg_delayed_ref_runtime = NSEC_PER_SEC >> 6; /* div by 64 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2767) /* readahead state */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2768) INIT_RADIX_TREE(&fs_info->reada_tree, GFP_NOFS & ~__GFP_DIRECT_RECLAIM);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2769) spin_lock_init(&fs_info->reada_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2770) btrfs_init_ref_verify(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2771)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2772) fs_info->thread_pool_size = min_t(unsigned long,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2773) num_online_cpus() + 2, 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2774)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2775) INIT_LIST_HEAD(&fs_info->ordered_roots);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2776) spin_lock_init(&fs_info->ordered_root_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2777)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2778) btrfs_init_scrub(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2779) #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2780) fs_info->check_integrity_print_mask = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2781) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2782) btrfs_init_balance(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2783) btrfs_init_async_reclaim_work(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2784)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2785) spin_lock_init(&fs_info->block_group_cache_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2786) fs_info->block_group_cache_tree = RB_ROOT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2787) fs_info->first_logical_byte = (u64)-1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2788)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2789) extent_io_tree_init(fs_info, &fs_info->excluded_extents,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2790) IO_TREE_FS_EXCLUDED_EXTENTS, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2791) set_bit(BTRFS_FS_BARRIER, &fs_info->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2792)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2793) mutex_init(&fs_info->ordered_operations_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2794) mutex_init(&fs_info->tree_log_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2795) mutex_init(&fs_info->chunk_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2796) mutex_init(&fs_info->transaction_kthread_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2797) mutex_init(&fs_info->cleaner_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2798) mutex_init(&fs_info->ro_block_group_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2799) init_rwsem(&fs_info->commit_root_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2800) init_rwsem(&fs_info->cleanup_work_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2801) init_rwsem(&fs_info->subvol_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2802) sema_init(&fs_info->uuid_tree_rescan_sem, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2803)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2804) btrfs_init_dev_replace_locks(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2805) btrfs_init_qgroup(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2806) btrfs_discard_init(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2807)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2808) btrfs_init_free_cluster(&fs_info->meta_alloc_cluster);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2809) btrfs_init_free_cluster(&fs_info->data_alloc_cluster);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2810)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2811) init_waitqueue_head(&fs_info->transaction_throttle);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2812) init_waitqueue_head(&fs_info->transaction_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2813) init_waitqueue_head(&fs_info->transaction_blocked_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2814) init_waitqueue_head(&fs_info->async_submit_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2815) init_waitqueue_head(&fs_info->delayed_iputs_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2816)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2817) /* Usable values until the real ones are cached from the superblock */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2818) fs_info->nodesize = 4096;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2819) fs_info->sectorsize = 4096;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2820) fs_info->stripesize = 4096;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2821)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2822) spin_lock_init(&fs_info->swapfile_pins_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2823) fs_info->swapfile_pins = RB_ROOT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2824)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2825) fs_info->send_in_progress = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2826) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2827)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2828) static int init_mount_fs_info(struct btrfs_fs_info *fs_info, struct super_block *sb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2829) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2830) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2831)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2832) fs_info->sb = sb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2833) sb->s_blocksize = BTRFS_BDEV_BLOCKSIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2834) sb->s_blocksize_bits = blksize_bits(BTRFS_BDEV_BLOCKSIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2835)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2836) ret = percpu_counter_init(&fs_info->dio_bytes, 0, GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2837) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2838) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2839)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2840) ret = percpu_counter_init(&fs_info->dirty_metadata_bytes, 0, GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2841) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2842) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2843)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2844) fs_info->dirty_metadata_batch = PAGE_SIZE *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2845) (1 + ilog2(nr_cpu_ids));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2846)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2847) ret = percpu_counter_init(&fs_info->delalloc_bytes, 0, GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2848) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2849) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2850)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2851) ret = percpu_counter_init(&fs_info->dev_replace.bio_counter, 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2852) GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2853) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2854) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2855)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2856) fs_info->delayed_root = kmalloc(sizeof(struct btrfs_delayed_root),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2857) GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2858) if (!fs_info->delayed_root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2859) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2860) btrfs_init_delayed_root(fs_info->delayed_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2861)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2862) return btrfs_alloc_stripe_hash_table(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2863) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2864)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2865) static int btrfs_uuid_rescan_kthread(void *data)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2866) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2867) struct btrfs_fs_info *fs_info = (struct btrfs_fs_info *)data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2868) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2869)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2870) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2871) * 1st step is to iterate through the existing UUID tree and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2872) * to delete all entries that contain outdated data.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2873) * 2nd step is to add all missing entries to the UUID tree.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2874) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2875) ret = btrfs_uuid_tree_iterate(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2876) if (ret < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2877) if (ret != -EINTR)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2878) btrfs_warn(fs_info, "iterating uuid_tree failed %d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2879) ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2880) up(&fs_info->uuid_tree_rescan_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2881) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2882) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2883) return btrfs_uuid_scan_kthread(data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2884) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2885)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2886) static int btrfs_check_uuid_tree(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2887) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2888) struct task_struct *task;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2889)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2890) down(&fs_info->uuid_tree_rescan_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2891) task = kthread_run(btrfs_uuid_rescan_kthread, fs_info, "btrfs-uuid");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2892) if (IS_ERR(task)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2893) /* fs_info->update_uuid_tree_gen remains 0 in all error case */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2894) btrfs_warn(fs_info, "failed to start uuid_rescan task");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2895) up(&fs_info->uuid_tree_rescan_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2896) return PTR_ERR(task);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2897) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2898)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2899) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2900) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2901)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2902) int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_devices,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2903) char *options)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2904) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2905) u32 sectorsize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2906) u32 nodesize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2907) u32 stripesize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2908) u64 generation;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2909) u64 features;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2910) u16 csum_type;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2911) struct btrfs_super_block *disk_super;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2912) struct btrfs_fs_info *fs_info = btrfs_sb(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2913) struct btrfs_root *tree_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2914) struct btrfs_root *chunk_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2915) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2916) int err = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2917) int clear_free_space_tree = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2918) int level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2919)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2920) ret = init_mount_fs_info(fs_info, sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2921) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2922) err = ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2923) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2924) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2925)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2926) /* These need to be init'ed before we start creating inodes and such. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2927) tree_root = btrfs_alloc_root(fs_info, BTRFS_ROOT_TREE_OBJECTID,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2928) GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2929) fs_info->tree_root = tree_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2930) chunk_root = btrfs_alloc_root(fs_info, BTRFS_CHUNK_TREE_OBJECTID,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2931) GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2932) fs_info->chunk_root = chunk_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2933) if (!tree_root || !chunk_root) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2934) err = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2935) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2936) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2937)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2938) fs_info->btree_inode = new_inode(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2939) if (!fs_info->btree_inode) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2940) err = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2941) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2942) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2943) mapping_set_gfp_mask(fs_info->btree_inode->i_mapping, GFP_NOFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2944) btrfs_init_btree_inode(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2945)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2946) invalidate_bdev(fs_devices->latest_bdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2947)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2948) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2949) * Read super block and check the signature bytes only
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2950) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2951) disk_super = btrfs_read_dev_super(fs_devices->latest_bdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2952) if (IS_ERR(disk_super)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2953) err = PTR_ERR(disk_super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2954) goto fail_alloc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2955) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2956)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2957) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2958) * Verify the type first, if that or the checksum value are
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2959) * corrupted, we'll find out
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2960) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2961) csum_type = btrfs_super_csum_type(disk_super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2962) if (!btrfs_supported_super_csum(csum_type)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2963) btrfs_err(fs_info, "unsupported checksum algorithm: %u",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2964) csum_type);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2965) err = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2966) btrfs_release_disk_super(disk_super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2967) goto fail_alloc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2968) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2969)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2970) ret = btrfs_init_csum_hash(fs_info, csum_type);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2971) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2972) err = ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2973) btrfs_release_disk_super(disk_super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2974) goto fail_alloc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2975) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2976)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2977) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2978) * We want to check superblock checksum, the type is stored inside.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2979) * Pass the whole disk block of size BTRFS_SUPER_INFO_SIZE (4k).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2980) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2981) if (btrfs_check_super_csum(fs_info, (u8 *)disk_super)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2982) btrfs_err(fs_info, "superblock checksum mismatch");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2983) err = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2984) btrfs_release_disk_super(disk_super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2985) goto fail_alloc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2986) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2987)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2988) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2989) * super_copy is zeroed at allocation time and we never touch the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2990) * following bytes up to INFO_SIZE, the checksum is calculated from
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2991) * the whole block of INFO_SIZE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2992) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2993) memcpy(fs_info->super_copy, disk_super, sizeof(*fs_info->super_copy));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2994) btrfs_release_disk_super(disk_super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2995)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2996) disk_super = fs_info->super_copy;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2997)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2998)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2999) features = btrfs_super_flags(disk_super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3000) if (features & BTRFS_SUPER_FLAG_CHANGING_FSID_V2) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3001) features &= ~BTRFS_SUPER_FLAG_CHANGING_FSID_V2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3002) btrfs_set_super_flags(disk_super, features);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3003) btrfs_info(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3004) "found metadata UUID change in progress flag, clearing");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3005) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3006)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3007) memcpy(fs_info->super_for_commit, fs_info->super_copy,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3008) sizeof(*fs_info->super_for_commit));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3009)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3010) ret = btrfs_validate_mount_super(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3011) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3012) btrfs_err(fs_info, "superblock contains fatal errors");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3013) err = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3014) goto fail_alloc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3015) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3016)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3017) if (!btrfs_super_root(disk_super))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3018) goto fail_alloc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3019)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3020) /* check FS state, whether FS is broken. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3021) if (btrfs_super_flags(disk_super) & BTRFS_SUPER_FLAG_ERROR)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3022) set_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3023)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3024) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3025) * In the long term, we'll store the compression type in the super
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3026) * block, and it'll be used for per file compression control.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3027) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3028) fs_info->compress_type = BTRFS_COMPRESS_ZLIB;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3029)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3030) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3031) * Flag our filesystem as having big metadata blocks if they are bigger
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3032) * than the page size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3033) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3034) if (btrfs_super_nodesize(disk_super) > PAGE_SIZE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3035) if (!(features & BTRFS_FEATURE_INCOMPAT_BIG_METADATA))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3036) btrfs_info(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3037) "flagging fs with big metadata feature");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3038) features |= BTRFS_FEATURE_INCOMPAT_BIG_METADATA;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3039) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3040)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3041) /* Set up fs_info before parsing mount options */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3042) nodesize = btrfs_super_nodesize(disk_super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3043) sectorsize = btrfs_super_sectorsize(disk_super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3044) stripesize = sectorsize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3045) fs_info->dirty_metadata_batch = nodesize * (1 + ilog2(nr_cpu_ids));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3046) fs_info->delalloc_batch = sectorsize * 512 * (1 + ilog2(nr_cpu_ids));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3047)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3048) /* Cache block sizes */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3049) fs_info->nodesize = nodesize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3050) fs_info->sectorsize = sectorsize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3051) fs_info->stripesize = stripesize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3052)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3053) ret = btrfs_parse_options(fs_info, options, sb->s_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3054) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3055) err = ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3056) goto fail_alloc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3057) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3058)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3059) features = btrfs_super_incompat_flags(disk_super) &
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3060) ~BTRFS_FEATURE_INCOMPAT_SUPP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3061) if (features) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3062) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3063) "cannot mount because of unsupported optional features (%llx)",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3064) features);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3065) err = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3066) goto fail_alloc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3067) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3068)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3069) features = btrfs_super_incompat_flags(disk_super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3070) features |= BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3071) if (fs_info->compress_type == BTRFS_COMPRESS_LZO)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3072) features |= BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3073) else if (fs_info->compress_type == BTRFS_COMPRESS_ZSTD)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3074) features |= BTRFS_FEATURE_INCOMPAT_COMPRESS_ZSTD;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3075)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3076) if (features & BTRFS_FEATURE_INCOMPAT_SKINNY_METADATA)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3077) btrfs_info(fs_info, "has skinny extents");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3078)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3079) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3080) * mixed block groups end up with duplicate but slightly offset
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3081) * extent buffers for the same range. It leads to corruptions
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3082) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3083) if ((features & BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3084) (sectorsize != nodesize)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3085) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3086) "unequal nodesize/sectorsize (%u != %u) are not allowed for mixed block groups",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3087) nodesize, sectorsize);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3088) goto fail_alloc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3089) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3090)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3091) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3092) * Needn't use the lock because there is no other task which will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3093) * update the flag.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3094) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3095) btrfs_set_super_incompat_flags(disk_super, features);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3096)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3097) features = btrfs_super_compat_ro_flags(disk_super) &
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3098) ~BTRFS_FEATURE_COMPAT_RO_SUPP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3099) if (!sb_rdonly(sb) && features) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3100) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3101) "cannot mount read-write because of unsupported optional features (%llx)",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3102) features);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3103) err = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3104) goto fail_alloc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3105) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3106)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3107) ret = btrfs_init_workqueues(fs_info, fs_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3108) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3109) err = ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3110) goto fail_sb_buffer;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3111) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3112)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3113) sb->s_bdi->ra_pages *= btrfs_super_num_devices(disk_super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3114) sb->s_bdi->ra_pages = max(sb->s_bdi->ra_pages, SZ_4M / PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3115)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3116) sb->s_blocksize = sectorsize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3117) sb->s_blocksize_bits = blksize_bits(sectorsize);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3118) memcpy(&sb->s_uuid, fs_info->fs_devices->fsid, BTRFS_FSID_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3119)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3120) mutex_lock(&fs_info->chunk_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3121) ret = btrfs_read_sys_array(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3122) mutex_unlock(&fs_info->chunk_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3123) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3124) btrfs_err(fs_info, "failed to read the system array: %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3125) goto fail_sb_buffer;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3126) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3127)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3128) generation = btrfs_super_chunk_root_generation(disk_super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3129) level = btrfs_super_chunk_root_level(disk_super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3130)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3131) chunk_root->node = read_tree_block(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3132) btrfs_super_chunk_root(disk_super),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3133) generation, level, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3134) if (IS_ERR(chunk_root->node) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3135) !extent_buffer_uptodate(chunk_root->node)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3136) btrfs_err(fs_info, "failed to read chunk root");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3137) if (!IS_ERR(chunk_root->node))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3138) free_extent_buffer(chunk_root->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3139) chunk_root->node = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3140) goto fail_tree_roots;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3141) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3142) btrfs_set_root_node(&chunk_root->root_item, chunk_root->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3143) chunk_root->commit_root = btrfs_root_node(chunk_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3144)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3145) read_extent_buffer(chunk_root->node, fs_info->chunk_tree_uuid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3146) offsetof(struct btrfs_header, chunk_tree_uuid),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3147) BTRFS_UUID_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3148)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3149) ret = btrfs_read_chunk_tree(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3150) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3151) btrfs_err(fs_info, "failed to read chunk tree: %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3152) goto fail_tree_roots;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3153) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3154)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3155) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3156) * Keep the devid that is marked to be the target device for the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3157) * device replace procedure
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3158) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3159) btrfs_free_extra_devids(fs_devices, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3160)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3161) if (!fs_devices->latest_bdev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3162) btrfs_err(fs_info, "failed to read devices");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3163) goto fail_tree_roots;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3164) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3165)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3166) ret = init_tree_roots(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3167) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3168) goto fail_tree_roots;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3169)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3170) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3171) * If we have a uuid root and we're not being told to rescan we need to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3172) * check the generation here so we can set the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3173) * BTRFS_FS_UPDATE_UUID_TREE_GEN bit. Otherwise we could commit the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3174) * transaction during a balance or the log replay without updating the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3175) * uuid generation, and then if we crash we would rescan the uuid tree,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3176) * even though it was perfectly fine.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3177) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3178) if (fs_info->uuid_root && !btrfs_test_opt(fs_info, RESCAN_UUID_TREE) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3179) fs_info->generation == btrfs_super_uuid_tree_generation(disk_super))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3180) set_bit(BTRFS_FS_UPDATE_UUID_TREE_GEN, &fs_info->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3181)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3182) ret = btrfs_verify_dev_extents(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3183) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3184) btrfs_err(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3185) "failed to verify dev extents against chunks: %d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3186) ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3187) goto fail_block_groups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3188) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3189) ret = btrfs_recover_balance(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3190) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3191) btrfs_err(fs_info, "failed to recover balance: %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3192) goto fail_block_groups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3193) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3194)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3195) ret = btrfs_init_dev_stats(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3196) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3197) btrfs_err(fs_info, "failed to init dev_stats: %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3198) goto fail_block_groups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3199) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3200)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3201) ret = btrfs_init_dev_replace(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3202) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3203) btrfs_err(fs_info, "failed to init dev_replace: %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3204) goto fail_block_groups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3205) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3206)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3207) btrfs_free_extra_devids(fs_devices, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3208)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3209) ret = btrfs_sysfs_add_fsid(fs_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3210) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3211) btrfs_err(fs_info, "failed to init sysfs fsid interface: %d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3212) ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3213) goto fail_block_groups;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3214) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3215)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3216) ret = btrfs_sysfs_add_mounted(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3217) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3218) btrfs_err(fs_info, "failed to init sysfs interface: %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3219) goto fail_fsdev_sysfs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3220) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3221)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3222) ret = btrfs_init_space_info(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3223) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3224) btrfs_err(fs_info, "failed to initialize space info: %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3225) goto fail_sysfs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3226) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3227)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3228) ret = btrfs_read_block_groups(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3229) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3230) btrfs_err(fs_info, "failed to read block groups: %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3231) goto fail_sysfs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3232) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3233)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3234) if (!sb_rdonly(sb) && fs_info->fs_devices->missing_devices &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3235) !btrfs_check_rw_degradable(fs_info, NULL)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3236) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3237) "writable mount is not allowed due to too many missing devices");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3238) goto fail_sysfs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3239) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3240)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3241) fs_info->cleaner_kthread = kthread_run(cleaner_kthread, tree_root,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3242) "btrfs-cleaner");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3243) if (IS_ERR(fs_info->cleaner_kthread))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3244) goto fail_sysfs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3245)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3246) fs_info->transaction_kthread = kthread_run(transaction_kthread,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3247) tree_root,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3248) "btrfs-transaction");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3249) if (IS_ERR(fs_info->transaction_kthread))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3250) goto fail_cleaner;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3251)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3252) if (!btrfs_test_opt(fs_info, NOSSD) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3253) !fs_info->fs_devices->rotating) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3254) btrfs_set_and_info(fs_info, SSD, "enabling ssd optimizations");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3255) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3256)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3257) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3258) * Mount does not set all options immediately, we can do it now and do
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3259) * not have to wait for transaction commit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3260) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3261) btrfs_apply_pending_changes(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3262)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3263) #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3264) if (btrfs_test_opt(fs_info, CHECK_INTEGRITY)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3265) ret = btrfsic_mount(fs_info, fs_devices,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3266) btrfs_test_opt(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3267) CHECK_INTEGRITY_INCLUDING_EXTENT_DATA) ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3268) 1 : 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3269) fs_info->check_integrity_print_mask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3270) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3271) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3272) "failed to initialize integrity check module: %d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3273) ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3274) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3275) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3276) ret = btrfs_read_qgroup_config(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3277) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3278) goto fail_trans_kthread;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3279)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3280) if (btrfs_build_ref_tree(fs_info))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3281) btrfs_err(fs_info, "couldn't build ref tree");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3282)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3283) /* do not make disk changes in broken FS or nologreplay is given */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3284) if (btrfs_super_log_root(disk_super) != 0 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3285) !btrfs_test_opt(fs_info, NOLOGREPLAY)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3286) btrfs_info(fs_info, "start tree-log replay");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3287) ret = btrfs_replay_log(fs_info, fs_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3288) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3289) err = ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3290) goto fail_qgroup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3291) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3292) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3293)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3294) ret = btrfs_find_orphan_roots(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3295) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3296) goto fail_qgroup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3297)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3298) if (!sb_rdonly(sb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3299) ret = btrfs_cleanup_fs_roots(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3300) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3301) goto fail_qgroup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3302)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3303) mutex_lock(&fs_info->cleaner_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3304) ret = btrfs_recover_relocation(tree_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3305) mutex_unlock(&fs_info->cleaner_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3306) if (ret < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3307) btrfs_warn(fs_info, "failed to recover relocation: %d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3308) ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3309) err = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3310) goto fail_qgroup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3311) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3312) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3313)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3314) fs_info->fs_root = btrfs_get_fs_root(fs_info, BTRFS_FS_TREE_OBJECTID, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3315) if (IS_ERR(fs_info->fs_root)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3316) err = PTR_ERR(fs_info->fs_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3317) btrfs_warn(fs_info, "failed to read fs tree: %d", err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3318) fs_info->fs_root = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3319) goto fail_qgroup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3320) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3321)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3322) if (sb_rdonly(sb))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3323) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3324)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3325) if (btrfs_test_opt(fs_info, CLEAR_CACHE) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3326) btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3327) clear_free_space_tree = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3328) } else if (btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3329) !btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE_VALID)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3330) btrfs_warn(fs_info, "free space tree is invalid");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3331) clear_free_space_tree = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3332) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3333)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3334) if (clear_free_space_tree) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3335) btrfs_info(fs_info, "clearing free space tree");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3336) ret = btrfs_clear_free_space_tree(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3337) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3338) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3339) "failed to clear free space tree: %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3340) close_ctree(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3341) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3342) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3343) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3344)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3345) if (btrfs_test_opt(fs_info, FREE_SPACE_TREE) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3346) !btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3347) btrfs_info(fs_info, "creating free space tree");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3348) ret = btrfs_create_free_space_tree(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3349) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3350) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3351) "failed to create free space tree: %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3352) close_ctree(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3353) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3354) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3355) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3356)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3357) down_read(&fs_info->cleanup_work_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3358) if ((ret = btrfs_orphan_cleanup(fs_info->fs_root)) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3359) (ret = btrfs_orphan_cleanup(fs_info->tree_root))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3360) up_read(&fs_info->cleanup_work_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3361) close_ctree(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3362) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3363) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3364) up_read(&fs_info->cleanup_work_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3365)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3366) ret = btrfs_resume_balance_async(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3367) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3368) btrfs_warn(fs_info, "failed to resume balance: %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3369) close_ctree(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3370) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3371) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3372)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3373) ret = btrfs_resume_dev_replace_async(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3374) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3375) btrfs_warn(fs_info, "failed to resume device replace: %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3376) close_ctree(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3377) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3378) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3379)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3380) btrfs_qgroup_rescan_resume(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3381) btrfs_discard_resume(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3382)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3383) if (!fs_info->uuid_root) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3384) btrfs_info(fs_info, "creating UUID tree");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3385) ret = btrfs_create_uuid_tree(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3386) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3387) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3388) "failed to create the UUID tree: %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3389) close_ctree(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3390) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3391) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3392) } else if (btrfs_test_opt(fs_info, RESCAN_UUID_TREE) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3393) fs_info->generation !=
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3394) btrfs_super_uuid_tree_generation(disk_super)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3395) btrfs_info(fs_info, "checking UUID tree");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3396) ret = btrfs_check_uuid_tree(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3397) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3398) btrfs_warn(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3399) "failed to check the UUID tree: %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3400) close_ctree(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3401) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3402) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3403) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3404) set_bit(BTRFS_FS_OPEN, &fs_info->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3405)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3406) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3407) * backuproot only affect mount behavior, and if open_ctree succeeded,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3408) * no need to keep the flag
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3409) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3410) btrfs_clear_opt(fs_info->mount_opt, USEBACKUPROOT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3411)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3412) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3413)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3414) fail_qgroup:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3415) btrfs_free_qgroup_config(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3416) fail_trans_kthread:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3417) kthread_stop(fs_info->transaction_kthread);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3418) btrfs_cleanup_transaction(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3419) btrfs_free_fs_roots(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3420) fail_cleaner:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3421) kthread_stop(fs_info->cleaner_kthread);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3422)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3423) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3424) * make sure we're done with the btree inode before we stop our
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3425) * kthreads
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3426) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3427) filemap_write_and_wait(fs_info->btree_inode->i_mapping);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3428)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3429) fail_sysfs:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3430) btrfs_sysfs_remove_mounted(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3431)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3432) fail_fsdev_sysfs:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3433) btrfs_sysfs_remove_fsid(fs_info->fs_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3434)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3435) fail_block_groups:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3436) btrfs_put_block_group_cache(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3437)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3438) fail_tree_roots:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3439) if (fs_info->data_reloc_root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3440) btrfs_drop_and_free_fs_root(fs_info, fs_info->data_reloc_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3441) free_root_pointers(fs_info, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3442) invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3443)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3444) fail_sb_buffer:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3445) btrfs_stop_all_workers(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3446) btrfs_free_block_groups(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3447) fail_alloc:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3448) btrfs_mapping_tree_free(&fs_info->mapping_tree);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3449)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3450) iput(fs_info->btree_inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3451) fail:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3452) btrfs_close_devices(fs_info->fs_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3453) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3454) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3455) ALLOW_ERROR_INJECTION(open_ctree, ERRNO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3456)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3457) static void btrfs_end_super_write(struct bio *bio)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3458) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3459) struct btrfs_device *device = bio->bi_private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3460) struct bio_vec *bvec;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3461) struct bvec_iter_all iter_all;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3462) struct page *page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3463)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3464) bio_for_each_segment_all(bvec, bio, iter_all) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3465) page = bvec->bv_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3466)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3467) if (bio->bi_status) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3468) btrfs_warn_rl_in_rcu(device->fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3469) "lost page write due to IO error on %s (%d)",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3470) rcu_str_deref(device->name),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3471) blk_status_to_errno(bio->bi_status));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3472) ClearPageUptodate(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3473) SetPageError(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3474) btrfs_dev_stat_inc_and_print(device,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3475) BTRFS_DEV_STAT_WRITE_ERRS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3476) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3477) SetPageUptodate(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3478) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3479)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3480) put_page(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3481) unlock_page(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3482) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3483)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3484) bio_put(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3485) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3486)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3487) struct btrfs_super_block *btrfs_read_dev_one_super(struct block_device *bdev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3488) int copy_num)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3489) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3490) struct btrfs_super_block *super;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3491) struct page *page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3492) u64 bytenr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3493) struct address_space *mapping = bdev->bd_inode->i_mapping;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3494)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3495) bytenr = btrfs_sb_offset(copy_num);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3496) if (bytenr + BTRFS_SUPER_INFO_SIZE >= i_size_read(bdev->bd_inode))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3497) return ERR_PTR(-EINVAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3498)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3499) page = read_cache_page_gfp(mapping, bytenr >> PAGE_SHIFT, GFP_NOFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3500) if (IS_ERR(page))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3501) return ERR_CAST(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3502)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3503) super = page_address(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3504) if (btrfs_super_magic(super) != BTRFS_MAGIC) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3505) btrfs_release_disk_super(super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3506) return ERR_PTR(-ENODATA);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3507) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3508)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3509) if (btrfs_super_bytenr(super) != bytenr) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3510) btrfs_release_disk_super(super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3511) return ERR_PTR(-EINVAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3512) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3513)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3514) return super;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3515) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3516)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3517)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3518) struct btrfs_super_block *btrfs_read_dev_super(struct block_device *bdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3519) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3520) struct btrfs_super_block *super, *latest = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3521) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3522) u64 transid = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3523)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3524) /* we would like to check all the supers, but that would make
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3525) * a btrfs mount succeed after a mkfs from a different FS.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3526) * So, we need to add a special mount option to scan for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3527) * later supers, using BTRFS_SUPER_MIRROR_MAX instead
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3528) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3529) for (i = 0; i < 1; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3530) super = btrfs_read_dev_one_super(bdev, i);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3531) if (IS_ERR(super))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3532) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3533)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3534) if (!latest || btrfs_super_generation(super) > transid) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3535) if (latest)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3536) btrfs_release_disk_super(super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3537)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3538) latest = super;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3539) transid = btrfs_super_generation(super);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3540) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3541) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3542)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3543) return super;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3544) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3545)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3546) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3547) * Write superblock @sb to the @device. Do not wait for completion, all the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3548) * pages we use for writing are locked.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3549) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3550) * Write @max_mirrors copies of the superblock, where 0 means default that fit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3551) * the expected device size at commit time. Note that max_mirrors must be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3552) * same for write and wait phases.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3553) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3554) * Return number of errors when page is not found or submission fails.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3555) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3556) static int write_dev_supers(struct btrfs_device *device,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3557) struct btrfs_super_block *sb, int max_mirrors)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3558) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3559) struct btrfs_fs_info *fs_info = device->fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3560) struct address_space *mapping = device->bdev->bd_inode->i_mapping;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3561) SHASH_DESC_ON_STACK(shash, fs_info->csum_shash);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3562) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3563) int errors = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3564) u64 bytenr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3565)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3566) if (max_mirrors == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3567) max_mirrors = BTRFS_SUPER_MIRROR_MAX;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3568)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3569) shash->tfm = fs_info->csum_shash;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3570)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3571) for (i = 0; i < max_mirrors; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3572) struct page *page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3573) struct bio *bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3574) struct btrfs_super_block *disk_super;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3575)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3576) bytenr = btrfs_sb_offset(i);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3577) if (bytenr + BTRFS_SUPER_INFO_SIZE >=
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3578) device->commit_total_bytes)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3579) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3580)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3581) btrfs_set_super_bytenr(sb, bytenr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3582)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3583) crypto_shash_digest(shash, (const char *)sb + BTRFS_CSUM_SIZE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3584) BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3585) sb->csum);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3586)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3587) page = find_or_create_page(mapping, bytenr >> PAGE_SHIFT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3588) GFP_NOFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3589) if (!page) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3590) btrfs_err(device->fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3591) "couldn't get super block page for bytenr %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3592) bytenr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3593) errors++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3594) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3595) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3596)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3597) /* Bump the refcount for wait_dev_supers() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3598) get_page(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3599)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3600) disk_super = page_address(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3601) memcpy(disk_super, sb, BTRFS_SUPER_INFO_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3602)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3603) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3604) * Directly use bios here instead of relying on the page cache
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3605) * to do I/O, so we don't lose the ability to do integrity
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3606) * checking.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3607) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3608) bio = bio_alloc(GFP_NOFS, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3609) bio_set_dev(bio, device->bdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3610) bio->bi_iter.bi_sector = bytenr >> SECTOR_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3611) bio->bi_private = device;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3612) bio->bi_end_io = btrfs_end_super_write;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3613) __bio_add_page(bio, page, BTRFS_SUPER_INFO_SIZE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3614) offset_in_page(bytenr));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3615)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3616) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3617) * We FUA only the first super block. The others we allow to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3618) * go down lazy and there's a short window where the on-disk
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3619) * copies might still contain the older version.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3620) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3621) bio->bi_opf = REQ_OP_WRITE | REQ_SYNC | REQ_META | REQ_PRIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3622) if (i == 0 && !btrfs_test_opt(device->fs_info, NOBARRIER))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3623) bio->bi_opf |= REQ_FUA;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3624)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3625) btrfsic_submit_bio(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3626) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3627) return errors < i ? 0 : -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3628) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3629)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3630) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3631) * Wait for write completion of superblocks done by write_dev_supers,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3632) * @max_mirrors same for write and wait phases.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3633) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3634) * Return number of errors when page is not found or not marked up to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3635) * date.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3636) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3637) static int wait_dev_supers(struct btrfs_device *device, int max_mirrors)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3638) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3639) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3640) int errors = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3641) bool primary_failed = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3642) u64 bytenr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3643)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3644) if (max_mirrors == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3645) max_mirrors = BTRFS_SUPER_MIRROR_MAX;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3646)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3647) for (i = 0; i < max_mirrors; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3648) struct page *page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3649)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3650) bytenr = btrfs_sb_offset(i);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3651) if (bytenr + BTRFS_SUPER_INFO_SIZE >=
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3652) device->commit_total_bytes)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3653) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3654)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3655) page = find_get_page(device->bdev->bd_inode->i_mapping,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3656) bytenr >> PAGE_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3657) if (!page) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3658) errors++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3659) if (i == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3660) primary_failed = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3661) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3662) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3663) /* Page is submitted locked and unlocked once the IO completes */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3664) wait_on_page_locked(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3665) if (PageError(page)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3666) errors++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3667) if (i == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3668) primary_failed = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3669) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3670)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3671) /* Drop our reference */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3672) put_page(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3673)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3674) /* Drop the reference from the writing run */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3675) put_page(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3676) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3677)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3678) /* log error, force error return */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3679) if (primary_failed) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3680) btrfs_err(device->fs_info, "error writing primary super block to device %llu",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3681) device->devid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3682) return -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3683) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3684)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3685) return errors < i ? 0 : -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3686) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3687)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3688) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3689) * endio for the write_dev_flush, this will wake anyone waiting
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3690) * for the barrier when it is done
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3691) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3692) static void btrfs_end_empty_barrier(struct bio *bio)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3693) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3694) complete(bio->bi_private);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3695) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3696)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3697) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3698) * Submit a flush request to the device if it supports it. Error handling is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3699) * done in the waiting counterpart.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3700) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3701) static void write_dev_flush(struct btrfs_device *device)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3702) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3703) struct bio *bio = device->flush_bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3704)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3705) #ifndef CONFIG_BTRFS_FS_CHECK_INTEGRITY
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3706) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3707) * When a disk has write caching disabled, we skip submission of a bio
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3708) * with flush and sync requests before writing the superblock, since
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3709) * it's not needed. However when the integrity checker is enabled, this
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3710) * results in reports that there are metadata blocks referred by a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3711) * superblock that were not properly flushed. So don't skip the bio
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3712) * submission only when the integrity checker is enabled for the sake
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3713) * of simplicity, since this is a debug tool and not meant for use in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3714) * non-debug builds.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3715) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3716) struct request_queue *q = bdev_get_queue(device->bdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3717) if (!test_bit(QUEUE_FLAG_WC, &q->queue_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3718) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3719) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3720)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3721) bio_reset(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3722) bio->bi_end_io = btrfs_end_empty_barrier;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3723) bio_set_dev(bio, device->bdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3724) bio->bi_opf = REQ_OP_WRITE | REQ_SYNC | REQ_PREFLUSH;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3725) init_completion(&device->flush_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3726) bio->bi_private = &device->flush_wait;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3727)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3728) btrfsic_submit_bio(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3729) set_bit(BTRFS_DEV_STATE_FLUSH_SENT, &device->dev_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3730) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3731)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3732) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3733) * If the flush bio has been submitted by write_dev_flush, wait for it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3734) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3735) static blk_status_t wait_dev_flush(struct btrfs_device *device)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3736) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3737) struct bio *bio = device->flush_bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3738)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3739) if (!test_bit(BTRFS_DEV_STATE_FLUSH_SENT, &device->dev_state))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3740) return BLK_STS_OK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3741)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3742) clear_bit(BTRFS_DEV_STATE_FLUSH_SENT, &device->dev_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3743) wait_for_completion_io(&device->flush_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3744)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3745) return bio->bi_status;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3746) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3747)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3748) static int check_barrier_error(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3749) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3750) if (!btrfs_check_rw_degradable(fs_info, NULL))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3751) return -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3752) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3753) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3754)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3755) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3756) * send an empty flush down to each device in parallel,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3757) * then wait for them
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3758) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3759) static int barrier_all_devices(struct btrfs_fs_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3760) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3761) struct list_head *head;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3762) struct btrfs_device *dev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3763) int errors_wait = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3764) blk_status_t ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3765)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3766) lockdep_assert_held(&info->fs_devices->device_list_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3767) /* send down all the barriers */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3768) head = &info->fs_devices->devices;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3769) list_for_each_entry(dev, head, dev_list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3770) if (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3771) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3772) if (!dev->bdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3773) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3774) if (!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &dev->dev_state) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3775) !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3776) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3777)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3778) write_dev_flush(dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3779) dev->last_flush_error = BLK_STS_OK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3780) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3781)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3782) /* wait for all the barriers */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3783) list_for_each_entry(dev, head, dev_list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3784) if (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3785) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3786) if (!dev->bdev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3787) errors_wait++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3788) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3789) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3790) if (!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &dev->dev_state) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3791) !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3792) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3793)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3794) ret = wait_dev_flush(dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3795) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3796) dev->last_flush_error = ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3797) btrfs_dev_stat_inc_and_print(dev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3798) BTRFS_DEV_STAT_FLUSH_ERRS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3799) errors_wait++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3800) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3801) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3802)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3803) if (errors_wait) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3804) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3805) * At some point we need the status of all disks
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3806) * to arrive at the volume status. So error checking
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3807) * is being pushed to a separate loop.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3808) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3809) return check_barrier_error(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3810) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3811) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3812) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3813)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3814) int btrfs_get_num_tolerated_disk_barrier_failures(u64 flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3815) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3816) int raid_type;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3817) int min_tolerated = INT_MAX;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3818)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3819) if ((flags & BTRFS_BLOCK_GROUP_PROFILE_MASK) == 0 ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3820) (flags & BTRFS_AVAIL_ALLOC_BIT_SINGLE))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3821) min_tolerated = min_t(int, min_tolerated,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3822) btrfs_raid_array[BTRFS_RAID_SINGLE].
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3823) tolerated_failures);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3824)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3825) for (raid_type = 0; raid_type < BTRFS_NR_RAID_TYPES; raid_type++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3826) if (raid_type == BTRFS_RAID_SINGLE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3827) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3828) if (!(flags & btrfs_raid_array[raid_type].bg_flag))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3829) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3830) min_tolerated = min_t(int, min_tolerated,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3831) btrfs_raid_array[raid_type].
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3832) tolerated_failures);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3833) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3834)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3835) if (min_tolerated == INT_MAX) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3836) pr_warn("BTRFS: unknown raid flag: %llu", flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3837) min_tolerated = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3838) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3839)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3840) return min_tolerated;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3841) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3842)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3843) int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3844) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3845) struct list_head *head;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3846) struct btrfs_device *dev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3847) struct btrfs_super_block *sb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3848) struct btrfs_dev_item *dev_item;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3849) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3850) int do_barriers;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3851) int max_errors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3852) int total_errors = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3853) u64 flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3854)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3855) do_barriers = !btrfs_test_opt(fs_info, NOBARRIER);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3856)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3857) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3858) * max_mirrors == 0 indicates we're from commit_transaction,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3859) * not from fsync where the tree roots in fs_info have not
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3860) * been consistent on disk.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3861) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3862) if (max_mirrors == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3863) backup_super_roots(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3864)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3865) sb = fs_info->super_for_commit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3866) dev_item = &sb->dev_item;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3867)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3868) mutex_lock(&fs_info->fs_devices->device_list_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3869) head = &fs_info->fs_devices->devices;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3870) max_errors = btrfs_super_num_devices(fs_info->super_copy) - 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3871)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3872) if (do_barriers) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3873) ret = barrier_all_devices(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3874) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3875) mutex_unlock(
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3876) &fs_info->fs_devices->device_list_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3877) btrfs_handle_fs_error(fs_info, ret,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3878) "errors while submitting device barriers.");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3879) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3880) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3881) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3882)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3883) list_for_each_entry(dev, head, dev_list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3884) if (!dev->bdev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3885) total_errors++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3886) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3887) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3888) if (!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &dev->dev_state) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3889) !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3890) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3891)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3892) btrfs_set_stack_device_generation(dev_item, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3893) btrfs_set_stack_device_type(dev_item, dev->type);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3894) btrfs_set_stack_device_id(dev_item, dev->devid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3895) btrfs_set_stack_device_total_bytes(dev_item,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3896) dev->commit_total_bytes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3897) btrfs_set_stack_device_bytes_used(dev_item,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3898) dev->commit_bytes_used);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3899) btrfs_set_stack_device_io_align(dev_item, dev->io_align);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3900) btrfs_set_stack_device_io_width(dev_item, dev->io_width);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3901) btrfs_set_stack_device_sector_size(dev_item, dev->sector_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3902) memcpy(dev_item->uuid, dev->uuid, BTRFS_UUID_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3903) memcpy(dev_item->fsid, dev->fs_devices->metadata_uuid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3904) BTRFS_FSID_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3905)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3906) flags = btrfs_super_flags(sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3907) btrfs_set_super_flags(sb, flags | BTRFS_HEADER_FLAG_WRITTEN);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3908)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3909) ret = btrfs_validate_write_super(fs_info, sb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3910) if (ret < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3911) mutex_unlock(&fs_info->fs_devices->device_list_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3912) btrfs_handle_fs_error(fs_info, -EUCLEAN,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3913) "unexpected superblock corruption detected");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3914) return -EUCLEAN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3915) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3916)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3917) ret = write_dev_supers(dev, sb, max_mirrors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3918) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3919) total_errors++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3920) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3921) if (total_errors > max_errors) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3922) btrfs_err(fs_info, "%d errors while writing supers",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3923) total_errors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3924) mutex_unlock(&fs_info->fs_devices->device_list_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3925)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3926) /* FUA is masked off if unsupported and can't be the reason */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3927) btrfs_handle_fs_error(fs_info, -EIO,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3928) "%d errors while writing supers",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3929) total_errors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3930) return -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3931) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3932)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3933) total_errors = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3934) list_for_each_entry(dev, head, dev_list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3935) if (!dev->bdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3936) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3937) if (!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &dev->dev_state) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3938) !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3939) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3940)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3941) ret = wait_dev_supers(dev, max_mirrors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3942) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3943) total_errors++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3944) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3945) mutex_unlock(&fs_info->fs_devices->device_list_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3946) if (total_errors > max_errors) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3947) btrfs_handle_fs_error(fs_info, -EIO,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3948) "%d errors while writing supers",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3949) total_errors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3950) return -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3951) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3952) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3953) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3954)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3955) /* Drop a fs root from the radix tree and free it. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3956) void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3957) struct btrfs_root *root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3958) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3959) bool drop_ref = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3960)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3961) spin_lock(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3962) radix_tree_delete(&fs_info->fs_roots_radix,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3963) (unsigned long)root->root_key.objectid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3964) if (test_and_clear_bit(BTRFS_ROOT_IN_RADIX, &root->state))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3965) drop_ref = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3966) spin_unlock(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3967)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3968) if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3969) ASSERT(root->log_root == NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3970) if (root->reloc_root) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3971) btrfs_put_root(root->reloc_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3972) root->reloc_root = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3973) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3974) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3975)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3976) if (root->free_ino_pinned)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3977) __btrfs_remove_free_space_cache(root->free_ino_pinned);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3978) if (root->free_ino_ctl)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3979) __btrfs_remove_free_space_cache(root->free_ino_ctl);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3980) if (root->ino_cache_inode) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3981) iput(root->ino_cache_inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3982) root->ino_cache_inode = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3983) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3984) if (drop_ref)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3985) btrfs_put_root(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3986) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3987)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3988) int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3989) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3990) u64 root_objectid = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3991) struct btrfs_root *gang[8];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3992) int i = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3993) int err = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3994) unsigned int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3995)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3996) while (1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3997) spin_lock(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3998) ret = radix_tree_gang_lookup(&fs_info->fs_roots_radix,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3999) (void **)gang, root_objectid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4000) ARRAY_SIZE(gang));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4001) if (!ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4002) spin_unlock(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4003) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4004) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4005) root_objectid = gang[ret - 1]->root_key.objectid + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4006)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4007) for (i = 0; i < ret; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4008) /* Avoid to grab roots in dead_roots */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4009) if (btrfs_root_refs(&gang[i]->root_item) == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4010) gang[i] = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4011) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4012) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4013) /* grab all the search result for later use */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4014) gang[i] = btrfs_grab_root(gang[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4015) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4016) spin_unlock(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4017)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4018) for (i = 0; i < ret; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4019) if (!gang[i])
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4020) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4021) root_objectid = gang[i]->root_key.objectid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4022) err = btrfs_orphan_cleanup(gang[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4023) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4024) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4025) btrfs_put_root(gang[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4026) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4027) root_objectid++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4028) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4029)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4030) /* release the uncleaned roots due to error */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4031) for (; i < ret; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4032) if (gang[i])
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4033) btrfs_put_root(gang[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4034) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4035) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4036) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4037)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4038) int btrfs_commit_super(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4039) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4040) struct btrfs_root *root = fs_info->tree_root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4041) struct btrfs_trans_handle *trans;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4042)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4043) mutex_lock(&fs_info->cleaner_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4044) btrfs_run_delayed_iputs(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4045) mutex_unlock(&fs_info->cleaner_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4046) wake_up_process(fs_info->cleaner_kthread);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4047)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4048) /* wait until ongoing cleanup work done */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4049) down_write(&fs_info->cleanup_work_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4050) up_write(&fs_info->cleanup_work_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4051)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4052) trans = btrfs_join_transaction(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4053) if (IS_ERR(trans))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4054) return PTR_ERR(trans);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4055) return btrfs_commit_transaction(trans);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4056) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4057)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4058) void __cold close_ctree(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4059) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4060) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4061)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4062) set_bit(BTRFS_FS_CLOSING_START, &fs_info->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4063) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4064) * We don't want the cleaner to start new transactions, add more delayed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4065) * iputs, etc. while we're closing. We can't use kthread_stop() yet
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4066) * because that frees the task_struct, and the transaction kthread might
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4067) * still try to wake up the cleaner.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4068) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4069) kthread_park(fs_info->cleaner_kthread);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4070)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4071) /* wait for the qgroup rescan worker to stop */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4072) btrfs_qgroup_wait_for_completion(fs_info, false);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4073)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4074) /* wait for the uuid_scan task to finish */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4075) down(&fs_info->uuid_tree_rescan_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4076) /* avoid complains from lockdep et al., set sem back to initial state */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4077) up(&fs_info->uuid_tree_rescan_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4078)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4079) /* pause restriper - we want to resume on mount */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4080) btrfs_pause_balance(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4081)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4082) btrfs_dev_replace_suspend_for_unmount(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4083)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4084) btrfs_scrub_cancel(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4085)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4086) /* wait for any defraggers to finish */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4087) wait_event(fs_info->transaction_wait,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4088) (atomic_read(&fs_info->defrag_running) == 0));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4089)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4090) /* clear out the rbtree of defraggable inodes */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4091) btrfs_cleanup_defrag_inodes(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4092)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4093) cancel_work_sync(&fs_info->async_reclaim_work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4094) cancel_work_sync(&fs_info->async_data_reclaim_work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4095)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4096) /* Cancel or finish ongoing discard work */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4097) btrfs_discard_cleanup(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4098)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4099) if (!sb_rdonly(fs_info->sb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4100) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4101) * The cleaner kthread is stopped, so do one final pass over
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4102) * unused block groups.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4103) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4104) btrfs_delete_unused_bgs(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4105)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4106) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4107) * There might be existing delayed inode workers still running
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4108) * and holding an empty delayed inode item. We must wait for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4109) * them to complete first because they can create a transaction.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4110) * This happens when someone calls btrfs_balance_delayed_items()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4111) * and then a transaction commit runs the same delayed nodes
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4112) * before any delayed worker has done something with the nodes.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4113) * We must wait for any worker here and not at transaction
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4114) * commit time since that could cause a deadlock.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4115) * This is a very rare case.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4116) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4117) btrfs_flush_workqueue(fs_info->delayed_workers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4118)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4119) ret = btrfs_commit_super(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4120) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4121) btrfs_err(fs_info, "commit super ret %d", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4122) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4123)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4124) if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4125) test_bit(BTRFS_FS_STATE_TRANS_ABORTED, &fs_info->fs_state))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4126) btrfs_error_commit_super(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4127)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4128) kthread_stop(fs_info->transaction_kthread);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4129) kthread_stop(fs_info->cleaner_kthread);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4130)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4131) ASSERT(list_empty(&fs_info->delayed_iputs));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4132) set_bit(BTRFS_FS_CLOSING_DONE, &fs_info->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4133)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4134) if (btrfs_check_quota_leak(fs_info)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4135) WARN_ON(IS_ENABLED(CONFIG_BTRFS_DEBUG));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4136) btrfs_err(fs_info, "qgroup reserved space leaked");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4137) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4138)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4139) btrfs_free_qgroup_config(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4140) ASSERT(list_empty(&fs_info->delalloc_roots));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4141)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4142) if (percpu_counter_sum(&fs_info->delalloc_bytes)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4143) btrfs_info(fs_info, "at unmount delalloc count %lld",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4144) percpu_counter_sum(&fs_info->delalloc_bytes));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4145) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4146)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4147) if (percpu_counter_sum(&fs_info->dio_bytes))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4148) btrfs_info(fs_info, "at unmount dio bytes count %lld",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4149) percpu_counter_sum(&fs_info->dio_bytes));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4150)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4151) btrfs_sysfs_remove_mounted(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4152) btrfs_sysfs_remove_fsid(fs_info->fs_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4153)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4154) btrfs_put_block_group_cache(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4155)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4156) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4157) * we must make sure there is not any read request to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4158) * submit after we stopping all workers.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4159) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4160) invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4161) btrfs_stop_all_workers(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4162)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4163) clear_bit(BTRFS_FS_OPEN, &fs_info->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4164) free_root_pointers(fs_info, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4165) btrfs_free_fs_roots(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4166)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4167) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4168) * We must free the block groups after dropping the fs_roots as we could
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4169) * have had an IO error and have left over tree log blocks that aren't
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4170) * cleaned up until the fs roots are freed. This makes the block group
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4171) * accounting appear to be wrong because there's pending reserved bytes,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4172) * so make sure we do the block group cleanup afterwards.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4173) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4174) btrfs_free_block_groups(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4175)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4176) iput(fs_info->btree_inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4177)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4178) #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4179) if (btrfs_test_opt(fs_info, CHECK_INTEGRITY))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4180) btrfsic_unmount(fs_info->fs_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4181) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4182)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4183) btrfs_mapping_tree_free(&fs_info->mapping_tree);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4184) btrfs_close_devices(fs_info->fs_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4185) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4186)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4187) int btrfs_buffer_uptodate(struct extent_buffer *buf, u64 parent_transid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4188) int atomic)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4189) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4190) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4191) struct inode *btree_inode = buf->pages[0]->mapping->host;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4192)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4193) ret = extent_buffer_uptodate(buf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4194) if (!ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4195) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4196)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4197) ret = verify_parent_transid(&BTRFS_I(btree_inode)->io_tree, buf,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4198) parent_transid, atomic);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4199) if (ret == -EAGAIN)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4200) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4201) return !ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4202) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4203)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4204) void btrfs_mark_buffer_dirty(struct extent_buffer *buf)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4205) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4206) struct btrfs_fs_info *fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4207) struct btrfs_root *root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4208) u64 transid = btrfs_header_generation(buf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4209) int was_dirty;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4210)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4211) #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4212) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4213) * This is a fast path so only do this check if we have sanity tests
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4214) * enabled. Normal people shouldn't be using unmapped buffers as dirty
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4215) * outside of the sanity tests.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4216) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4217) if (unlikely(test_bit(EXTENT_BUFFER_UNMAPPED, &buf->bflags)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4218) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4219) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4220) root = BTRFS_I(buf->pages[0]->mapping->host)->root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4221) fs_info = root->fs_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4222) btrfs_assert_tree_locked(buf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4223) if (transid != fs_info->generation)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4224) WARN(1, KERN_CRIT "btrfs transid mismatch buffer %llu, found %llu running %llu\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4225) buf->start, transid, fs_info->generation);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4226) was_dirty = set_extent_buffer_dirty(buf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4227) if (!was_dirty)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4228) percpu_counter_add_batch(&fs_info->dirty_metadata_bytes,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4229) buf->len,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4230) fs_info->dirty_metadata_batch);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4231) #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4232) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4233) * Since btrfs_mark_buffer_dirty() can be called with item pointer set
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4234) * but item data not updated.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4235) * So here we should only check item pointers, not item data.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4236) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4237) if (btrfs_header_level(buf) == 0 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4238) btrfs_check_leaf_relaxed(buf)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4239) btrfs_print_leaf(buf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4240) ASSERT(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4241) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4242) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4243) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4244)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4245) static void __btrfs_btree_balance_dirty(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4246) int flush_delayed)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4247) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4248) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4249) * looks as though older kernels can get into trouble with
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4250) * this code, they end up stuck in balance_dirty_pages forever
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4251) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4252) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4253)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4254) if (current->flags & PF_MEMALLOC)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4255) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4256)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4257) if (flush_delayed)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4258) btrfs_balance_delayed_items(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4259)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4260) ret = __percpu_counter_compare(&fs_info->dirty_metadata_bytes,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4261) BTRFS_DIRTY_METADATA_THRESH,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4262) fs_info->dirty_metadata_batch);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4263) if (ret > 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4264) balance_dirty_pages_ratelimited(fs_info->btree_inode->i_mapping);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4265) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4266) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4267)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4268) void btrfs_btree_balance_dirty(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4269) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4270) __btrfs_btree_balance_dirty(fs_info, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4271) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4272)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4273) void btrfs_btree_balance_dirty_nodelay(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4274) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4275) __btrfs_btree_balance_dirty(fs_info, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4276) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4277)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4278) int btrfs_read_buffer(struct extent_buffer *buf, u64 parent_transid, int level,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4279) struct btrfs_key *first_key)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4280) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4281) return btree_read_extent_buffer_pages(buf, parent_transid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4282) level, first_key);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4283) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4284)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4285) static void btrfs_error_commit_super(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4286) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4287) /* cleanup FS via transaction */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4288) btrfs_cleanup_transaction(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4289)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4290) mutex_lock(&fs_info->cleaner_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4291) btrfs_run_delayed_iputs(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4292) mutex_unlock(&fs_info->cleaner_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4293)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4294) down_write(&fs_info->cleanup_work_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4295) up_write(&fs_info->cleanup_work_sem);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4296) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4297)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4298) static void btrfs_drop_all_logs(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4299) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4300) struct btrfs_root *gang[8];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4301) u64 root_objectid = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4302) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4303)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4304) spin_lock(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4305) while ((ret = radix_tree_gang_lookup(&fs_info->fs_roots_radix,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4306) (void **)gang, root_objectid,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4307) ARRAY_SIZE(gang))) != 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4308) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4309)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4310) for (i = 0; i < ret; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4311) gang[i] = btrfs_grab_root(gang[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4312) spin_unlock(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4313)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4314) for (i = 0; i < ret; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4315) if (!gang[i])
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4316) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4317) root_objectid = gang[i]->root_key.objectid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4318) btrfs_free_log(NULL, gang[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4319) btrfs_put_root(gang[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4320) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4321) root_objectid++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4322) spin_lock(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4323) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4324) spin_unlock(&fs_info->fs_roots_radix_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4325) btrfs_free_log_root_tree(NULL, fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4326) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4327)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4328) static void btrfs_destroy_ordered_extents(struct btrfs_root *root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4329) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4330) struct btrfs_ordered_extent *ordered;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4331)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4332) spin_lock(&root->ordered_extent_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4333) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4334) * This will just short circuit the ordered completion stuff which will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4335) * make sure the ordered extent gets properly cleaned up.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4336) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4337) list_for_each_entry(ordered, &root->ordered_extents,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4338) root_extent_list)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4339) set_bit(BTRFS_ORDERED_IOERR, &ordered->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4340) spin_unlock(&root->ordered_extent_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4341) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4342)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4343) static void btrfs_destroy_all_ordered_extents(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4344) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4345) struct btrfs_root *root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4346) struct list_head splice;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4347)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4348) INIT_LIST_HEAD(&splice);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4349)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4350) spin_lock(&fs_info->ordered_root_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4351) list_splice_init(&fs_info->ordered_roots, &splice);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4352) while (!list_empty(&splice)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4353) root = list_first_entry(&splice, struct btrfs_root,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4354) ordered_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4355) list_move_tail(&root->ordered_root,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4356) &fs_info->ordered_roots);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4357)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4358) spin_unlock(&fs_info->ordered_root_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4359) btrfs_destroy_ordered_extents(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4360)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4361) cond_resched();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4362) spin_lock(&fs_info->ordered_root_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4363) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4364) spin_unlock(&fs_info->ordered_root_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4365)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4366) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4367) * We need this here because if we've been flipped read-only we won't
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4368) * get sync() from the umount, so we need to make sure any ordered
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4369) * extents that haven't had their dirty pages IO start writeout yet
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4370) * actually get run and error out properly.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4371) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4372) btrfs_wait_ordered_roots(fs_info, U64_MAX, 0, (u64)-1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4373) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4374)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4375) static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4376) struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4377) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4378) struct rb_node *node;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4379) struct btrfs_delayed_ref_root *delayed_refs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4380) struct btrfs_delayed_ref_node *ref;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4381) int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4382)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4383) delayed_refs = &trans->delayed_refs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4384)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4385) spin_lock(&delayed_refs->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4386) if (atomic_read(&delayed_refs->num_entries) == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4387) spin_unlock(&delayed_refs->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4388) btrfs_debug(fs_info, "delayed_refs has NO entry");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4389) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4390) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4391)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4392) while ((node = rb_first_cached(&delayed_refs->href_root)) != NULL) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4393) struct btrfs_delayed_ref_head *head;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4394) struct rb_node *n;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4395) bool pin_bytes = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4396)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4397) head = rb_entry(node, struct btrfs_delayed_ref_head,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4398) href_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4399) if (btrfs_delayed_ref_lock(delayed_refs, head))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4400) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4401)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4402) spin_lock(&head->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4403) while ((n = rb_first_cached(&head->ref_tree)) != NULL) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4404) ref = rb_entry(n, struct btrfs_delayed_ref_node,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4405) ref_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4406) ref->in_tree = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4407) rb_erase_cached(&ref->ref_node, &head->ref_tree);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4408) RB_CLEAR_NODE(&ref->ref_node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4409) if (!list_empty(&ref->add_list))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4410) list_del(&ref->add_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4411) atomic_dec(&delayed_refs->num_entries);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4412) btrfs_put_delayed_ref(ref);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4413) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4414) if (head->must_insert_reserved)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4415) pin_bytes = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4416) btrfs_free_delayed_extent_op(head->extent_op);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4417) btrfs_delete_ref_head(delayed_refs, head);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4418) spin_unlock(&head->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4419) spin_unlock(&delayed_refs->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4420) mutex_unlock(&head->mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4421)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4422) if (pin_bytes) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4423) struct btrfs_block_group *cache;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4424)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4425) cache = btrfs_lookup_block_group(fs_info, head->bytenr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4426) BUG_ON(!cache);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4427)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4428) spin_lock(&cache->space_info->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4429) spin_lock(&cache->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4430) cache->pinned += head->num_bytes;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4431) btrfs_space_info_update_bytes_pinned(fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4432) cache->space_info, head->num_bytes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4433) cache->reserved -= head->num_bytes;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4434) cache->space_info->bytes_reserved -= head->num_bytes;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4435) spin_unlock(&cache->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4436) spin_unlock(&cache->space_info->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4437) percpu_counter_add_batch(
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4438) &cache->space_info->total_bytes_pinned,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4439) head->num_bytes, BTRFS_TOTAL_BYTES_PINNED_BATCH);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4440)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4441) btrfs_put_block_group(cache);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4442)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4443) btrfs_error_unpin_extent_range(fs_info, head->bytenr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4444) head->bytenr + head->num_bytes - 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4445) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4446) btrfs_cleanup_ref_head_accounting(fs_info, delayed_refs, head);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4447) btrfs_put_delayed_ref_head(head);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4448) cond_resched();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4449) spin_lock(&delayed_refs->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4450) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4451) btrfs_qgroup_destroy_extent_records(trans);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4452)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4453) spin_unlock(&delayed_refs->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4454)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4455) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4456) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4457)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4458) static void btrfs_destroy_delalloc_inodes(struct btrfs_root *root)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4459) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4460) struct btrfs_inode *btrfs_inode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4461) struct list_head splice;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4462)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4463) INIT_LIST_HEAD(&splice);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4464)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4465) spin_lock(&root->delalloc_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4466) list_splice_init(&root->delalloc_inodes, &splice);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4467)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4468) while (!list_empty(&splice)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4469) struct inode *inode = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4470) btrfs_inode = list_first_entry(&splice, struct btrfs_inode,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4471) delalloc_inodes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4472) __btrfs_del_delalloc_inode(root, btrfs_inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4473) spin_unlock(&root->delalloc_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4474)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4475) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4476) * Make sure we get a live inode and that it'll not disappear
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4477) * meanwhile.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4478) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4479) inode = igrab(&btrfs_inode->vfs_inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4480) if (inode) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4481) invalidate_inode_pages2(inode->i_mapping);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4482) iput(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4483) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4484) spin_lock(&root->delalloc_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4485) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4486) spin_unlock(&root->delalloc_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4487) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4488)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4489) static void btrfs_destroy_all_delalloc_inodes(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4490) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4491) struct btrfs_root *root;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4492) struct list_head splice;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4493)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4494) INIT_LIST_HEAD(&splice);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4495)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4496) spin_lock(&fs_info->delalloc_root_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4497) list_splice_init(&fs_info->delalloc_roots, &splice);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4498) while (!list_empty(&splice)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4499) root = list_first_entry(&splice, struct btrfs_root,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4500) delalloc_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4501) root = btrfs_grab_root(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4502) BUG_ON(!root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4503) spin_unlock(&fs_info->delalloc_root_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4504)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4505) btrfs_destroy_delalloc_inodes(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4506) btrfs_put_root(root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4507)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4508) spin_lock(&fs_info->delalloc_root_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4509) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4510) spin_unlock(&fs_info->delalloc_root_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4511) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4512)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4513) static int btrfs_destroy_marked_extents(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4514) struct extent_io_tree *dirty_pages,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4515) int mark)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4516) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4517) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4518) struct extent_buffer *eb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4519) u64 start = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4520) u64 end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4521)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4522) while (1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4523) ret = find_first_extent_bit(dirty_pages, start, &start, &end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4524) mark, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4525) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4526) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4527)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4528) clear_extent_bits(dirty_pages, start, end, mark);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4529) while (start <= end) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4530) eb = find_extent_buffer(fs_info, start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4531) start += fs_info->nodesize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4532) if (!eb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4533) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4534) wait_on_extent_buffer_writeback(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4535)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4536) if (test_and_clear_bit(EXTENT_BUFFER_DIRTY,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4537) &eb->bflags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4538) clear_extent_buffer_dirty(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4539) free_extent_buffer_stale(eb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4540) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4541) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4542)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4543) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4544) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4545)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4546) static int btrfs_destroy_pinned_extent(struct btrfs_fs_info *fs_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4547) struct extent_io_tree *unpin)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4548) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4549) u64 start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4550) u64 end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4551) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4552)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4553) while (1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4554) struct extent_state *cached_state = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4555)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4556) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4557) * The btrfs_finish_extent_commit() may get the same range as
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4558) * ours between find_first_extent_bit and clear_extent_dirty.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4559) * Hence, hold the unused_bg_unpin_mutex to avoid double unpin
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4560) * the same extent range.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4561) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4562) mutex_lock(&fs_info->unused_bg_unpin_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4563) ret = find_first_extent_bit(unpin, 0, &start, &end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4564) EXTENT_DIRTY, &cached_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4565) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4566) mutex_unlock(&fs_info->unused_bg_unpin_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4567) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4568) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4569)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4570) clear_extent_dirty(unpin, start, end, &cached_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4571) free_extent_state(cached_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4572) btrfs_error_unpin_extent_range(fs_info, start, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4573) mutex_unlock(&fs_info->unused_bg_unpin_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4574) cond_resched();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4575) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4576)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4577) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4578) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4579)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4580) static void btrfs_cleanup_bg_io(struct btrfs_block_group *cache)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4581) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4582) struct inode *inode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4583)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4584) inode = cache->io_ctl.inode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4585) if (inode) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4586) invalidate_inode_pages2(inode->i_mapping);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4587) BTRFS_I(inode)->generation = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4588) cache->io_ctl.inode = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4589) iput(inode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4590) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4591) ASSERT(cache->io_ctl.pages == NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4592) btrfs_put_block_group(cache);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4593) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4594)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4595) void btrfs_cleanup_dirty_bgs(struct btrfs_transaction *cur_trans,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4596) struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4597) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4598) struct btrfs_block_group *cache;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4599)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4600) spin_lock(&cur_trans->dirty_bgs_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4601) while (!list_empty(&cur_trans->dirty_bgs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4602) cache = list_first_entry(&cur_trans->dirty_bgs,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4603) struct btrfs_block_group,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4604) dirty_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4605)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4606) if (!list_empty(&cache->io_list)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4607) spin_unlock(&cur_trans->dirty_bgs_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4608) list_del_init(&cache->io_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4609) btrfs_cleanup_bg_io(cache);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4610) spin_lock(&cur_trans->dirty_bgs_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4611) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4612)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4613) list_del_init(&cache->dirty_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4614) spin_lock(&cache->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4615) cache->disk_cache_state = BTRFS_DC_ERROR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4616) spin_unlock(&cache->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4617)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4618) spin_unlock(&cur_trans->dirty_bgs_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4619) btrfs_put_block_group(cache);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4620) btrfs_delayed_refs_rsv_release(fs_info, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4621) spin_lock(&cur_trans->dirty_bgs_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4622) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4623) spin_unlock(&cur_trans->dirty_bgs_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4624)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4625) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4626) * Refer to the definition of io_bgs member for details why it's safe
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4627) * to use it without any locking
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4628) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4629) while (!list_empty(&cur_trans->io_bgs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4630) cache = list_first_entry(&cur_trans->io_bgs,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4631) struct btrfs_block_group,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4632) io_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4633)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4634) list_del_init(&cache->io_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4635) spin_lock(&cache->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4636) cache->disk_cache_state = BTRFS_DC_ERROR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4637) spin_unlock(&cache->lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4638) btrfs_cleanup_bg_io(cache);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4639) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4640) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4641)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4642) void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4643) struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4644) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4645) struct btrfs_device *dev, *tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4646)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4647) btrfs_cleanup_dirty_bgs(cur_trans, fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4648) ASSERT(list_empty(&cur_trans->dirty_bgs));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4649) ASSERT(list_empty(&cur_trans->io_bgs));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4650)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4651) list_for_each_entry_safe(dev, tmp, &cur_trans->dev_update_list,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4652) post_commit_list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4653) list_del_init(&dev->post_commit_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4654) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4655)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4656) btrfs_destroy_delayed_refs(cur_trans, fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4657)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4658) cur_trans->state = TRANS_STATE_COMMIT_START;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4659) wake_up(&fs_info->transaction_blocked_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4660)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4661) cur_trans->state = TRANS_STATE_UNBLOCKED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4662) wake_up(&fs_info->transaction_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4663)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4664) btrfs_destroy_delayed_inodes(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4665)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4666) btrfs_destroy_marked_extents(fs_info, &cur_trans->dirty_pages,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4667) EXTENT_DIRTY);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4668) btrfs_destroy_pinned_extent(fs_info, &cur_trans->pinned_extents);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4669)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4670) cur_trans->state =TRANS_STATE_COMPLETED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4671) wake_up(&cur_trans->commit_wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4672) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4673)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4674) static int btrfs_cleanup_transaction(struct btrfs_fs_info *fs_info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4675) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4676) struct btrfs_transaction *t;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4677)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4678) mutex_lock(&fs_info->transaction_kthread_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4679)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4680) spin_lock(&fs_info->trans_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4681) while (!list_empty(&fs_info->trans_list)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4682) t = list_first_entry(&fs_info->trans_list,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4683) struct btrfs_transaction, list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4684) if (t->state >= TRANS_STATE_COMMIT_START) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4685) refcount_inc(&t->use_count);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4686) spin_unlock(&fs_info->trans_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4687) btrfs_wait_for_commit(fs_info, t->transid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4688) btrfs_put_transaction(t);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4689) spin_lock(&fs_info->trans_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4690) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4691) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4692) if (t == fs_info->running_transaction) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4693) t->state = TRANS_STATE_COMMIT_DOING;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4694) spin_unlock(&fs_info->trans_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4695) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4696) * We wait for 0 num_writers since we don't hold a trans
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4697) * handle open currently for this transaction.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4698) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4699) wait_event(t->writer_wait,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4700) atomic_read(&t->num_writers) == 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4701) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4702) spin_unlock(&fs_info->trans_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4703) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4704) btrfs_cleanup_one_transaction(t, fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4705)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4706) spin_lock(&fs_info->trans_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4707) if (t == fs_info->running_transaction)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4708) fs_info->running_transaction = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4709) list_del_init(&t->list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4710) spin_unlock(&fs_info->trans_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4711)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4712) btrfs_put_transaction(t);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4713) trace_btrfs_transaction_commit(fs_info->tree_root);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4714) spin_lock(&fs_info->trans_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4715) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4716) spin_unlock(&fs_info->trans_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4717) btrfs_destroy_all_ordered_extents(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4718) btrfs_destroy_delayed_inodes(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4719) btrfs_assert_delayed_root_empty(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4720) btrfs_destroy_all_delalloc_inodes(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4721) btrfs_drop_all_logs(fs_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4722) mutex_unlock(&fs_info->transaction_kthread_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4723)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4724) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4725) }
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags