* Boston, MA 021110-1307, USA.
*/
-#define _XOPEN_SOURCE 600
-#define __USE_XOPEN2K
-#define _GNU_SOURCE 1
#include <stdio.h>
#include <stdlib.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <unistd.h>
+#include <uuid/uuid.h>
#include "kerncompat.h"
#include "radix-tree.h"
#include "ctree.h"
#include "crc32c.h"
#include "utils.h"
#include "print-tree.h"
+#include "rbtree-utils.h"
+
+/* specified errno for check_tree_block */
+#define BTRFS_BAD_BYTENR (-1)
+#define BTRFS_BAD_FSID (-2)
+#define BTRFS_BAD_LEVEL (-3)
+#define BTRFS_BAD_NRITEMS (-4)
+
+/* Calculate max possible nritems for a leaf/node */
+static u32 max_nritems(u8 level, u32 nodesize)
+{
+
+ if (level == 0)
+ return ((nodesize - sizeof(struct btrfs_header)) /
+ sizeof(struct btrfs_item));
+ return ((nodesize - sizeof(struct btrfs_header)) /
+ sizeof(struct btrfs_key_ptr));
+}
static int check_tree_block(struct btrfs_root *root, struct extent_buffer *buf)
{
struct btrfs_fs_devices *fs_devices;
- int ret = 1;
+ int ret = BTRFS_BAD_FSID;
- if (buf->start != btrfs_header_bytenr(buf)) {
- printk("Check tree block failed, want=%Lu, have=%Lu\n",
- buf->start, btrfs_header_bytenr(buf));
- return ret;
- }
+ if (buf->start != btrfs_header_bytenr(buf))
+ return BTRFS_BAD_BYTENR;
+ if (btrfs_header_level(buf) >= BTRFS_MAX_LEVEL)
+ return BTRFS_BAD_LEVEL;
+ if (btrfs_header_nritems(buf) > max_nritems(btrfs_header_level(buf),
+ root->nodesize))
+ return BTRFS_BAD_NRITEMS;
fs_devices = root->fs_info->fs_devices;
while (fs_devices) {
- if (!memcmp_extent_buffer(buf, fs_devices->fsid,
- (unsigned long)btrfs_header_fsid(buf),
+ if (root->fs_info->ignore_fsid_mismatch ||
+ !memcmp_extent_buffer(buf, fs_devices->fsid,
+ btrfs_header_fsid(),
BTRFS_FSID_SIZE)) {
ret = 0;
break;
return ret;
}
+static void print_tree_block_error(struct btrfs_root *root,
+ struct extent_buffer *eb,
+ int err)
+{
+ char fs_uuid[BTRFS_UUID_UNPARSED_SIZE] = {'\0'};
+ char found_uuid[BTRFS_UUID_UNPARSED_SIZE] = {'\0'};
+ u8 buf[BTRFS_UUID_SIZE];
+
+ switch (err) {
+ case BTRFS_BAD_FSID:
+ read_extent_buffer(eb, buf, btrfs_header_fsid(),
+ BTRFS_UUID_SIZE);
+ uuid_unparse(buf, found_uuid);
+ uuid_unparse(root->fs_info->fsid, fs_uuid);
+ fprintf(stderr, "fsid mismatch, want=%s, have=%s\n",
+ fs_uuid, found_uuid);
+ break;
+ case BTRFS_BAD_BYTENR:
+ fprintf(stderr, "bytenr mismatch, want=%llu, have=%llu\n",
+ eb->start, btrfs_header_bytenr(eb));
+ break;
+ case BTRFS_BAD_LEVEL:
+ fprintf(stderr, "bad level, %u > %u\n",
+ btrfs_header_level(eb), BTRFS_MAX_LEVEL);
+ break;
+ case BTRFS_BAD_NRITEMS:
+ fprintf(stderr, "invalid nr_items: %u\n",
+ btrfs_header_nritems(eb));
+ break;
+ }
+}
+
u32 btrfs_csum_data(struct btrfs_root *root, char *data, u32 seed, size_t len)
{
return crc32c(seed, data, len);
*(__le32 *)result = ~cpu_to_le32(crc);
}
-int csum_tree_block_size(struct extent_buffer *buf, u16 csum_size,
- int verify)
+static int __csum_tree_block_size(struct extent_buffer *buf, u16 csum_size,
+ int verify, int silent)
{
char *result;
u32 len;
if (verify) {
if (memcmp_extent_buffer(buf, result, 0, csum_size)) {
- printk("checksum verify failed on %llu found %08X "
- "wanted %08X\n", (unsigned long long)buf->start,
- *((u32 *)result), *((u32*)(char *)buf->data));
+ if (!silent)
+ printk("checksum verify failed on %llu found %08X wanted %08X\n",
+ (unsigned long long)buf->start,
+ *((u32 *)result),
+ *((u32*)(char *)buf->data));
free(result);
return 1;
}
return 0;
}
+int csum_tree_block_size(struct extent_buffer *buf, u16 csum_size, int verify)
+{
+ return __csum_tree_block_size(buf, csum_size, verify, 0);
+}
+
+int verify_tree_block_csum_silent(struct extent_buffer *buf, u16 csum_size)
+{
+ return __csum_tree_block_size(buf, csum_size, 1, 1);
+}
+
int csum_tree_block(struct btrfs_root *root, struct extent_buffer *buf,
- int verify)
+ int verify)
{
u16 csum_size =
btrfs_super_csum_size(root->fs_info->super_copy);
+ if (verify && root->fs_info->suppress_check_block_errors)
+ return verify_tree_block_csum_silent(buf, csum_size);
return csum_tree_block_size(buf, csum_size, verify);
}
blocksize);
}
-int readahead_tree_block(struct btrfs_root *root, u64 bytenr, u32 blocksize,
- u64 parent_transid)
+void readahead_tree_block(struct btrfs_root *root, u64 bytenr, u32 blocksize,
+ u64 parent_transid)
{
- int ret;
struct extent_buffer *eb;
u64 length;
struct btrfs_multi_bio *multi = NULL;
struct btrfs_device *device;
eb = btrfs_find_tree_block(root, bytenr, blocksize);
- if (eb && btrfs_buffer_uptodate(eb, parent_transid)) {
- free_extent_buffer(eb);
- return 0;
+ if (!(eb && btrfs_buffer_uptodate(eb, parent_transid)) &&
+ !btrfs_map_block(&root->fs_info->mapping_tree, READ,
+ bytenr, &length, &multi, 0, NULL)) {
+ device = multi->stripes[0].dev;
+ device->total_ios++;
+ blocksize = min(blocksize, (u32)(64 * 1024));
+ readahead(device->fd, multi->stripes[0].physical, blocksize);
}
- length = blocksize;
- ret = btrfs_map_block(&root->fs_info->mapping_tree, READ,
- bytenr, &length, &multi, 0, NULL);
- BUG_ON(ret);
- device = multi->stripes[0].dev;
- device->total_ios++;
- blocksize = min(blocksize, (u32)(64 * 1024));
- readahead(device->fd, multi->stripes[0].physical, blocksize);
+ free_extent_buffer(eb);
kfree(multi);
- return 0;
}
static int verify_parent_transid(struct extent_io_tree *io_tree,
(unsigned long long)parent_transid,
(unsigned long long)btrfs_header_generation(eb));
if (ignore) {
+ eb->flags |= EXTENT_BAD_TRANSID;
printk("Ignoring transid failure\n");
return 0;
}
}
-static int read_whole_eb(struct btrfs_fs_info *info, struct extent_buffer *eb, int mirror)
+int read_whole_eb(struct btrfs_fs_info *info, struct extent_buffer *eb, int mirror)
{
unsigned long offset = 0;
struct btrfs_multi_bio *multi = NULL;
while (bytes_left) {
read_len = bytes_left;
- ret = btrfs_map_block(&info->mapping_tree, READ,
- eb->start + offset, &read_len, &multi,
- mirror, NULL);
- if (ret) {
- printk("Couldn't map the block %Lu\n", eb->start + offset);
- kfree(multi);
- return -EIO;
- }
- device = multi->stripes[0].dev;
+ device = NULL;
+
+ if (!info->on_restoring &&
+ eb->start != BTRFS_SUPER_INFO_OFFSET) {
+ ret = btrfs_map_block(&info->mapping_tree, READ,
+ eb->start + offset, &read_len, &multi,
+ mirror, NULL);
+ if (ret) {
+ printk("Couldn't map the block %Lu\n", eb->start + offset);
+ kfree(multi);
+ return -EIO;
+ }
+ device = multi->stripes[0].dev;
+
+ if (device->fd == 0) {
+ kfree(multi);
+ return -EIO;
+ }
- if (device->fd == 0) {
+ eb->fd = device->fd;
+ device->total_ios++;
+ eb->dev_bytenr = multi->stripes[0].physical;
kfree(multi);
- return -EIO;
- }
+ multi = NULL;
+ } else {
+ /* special case for restore metadump */
+ list_for_each_entry(device, &info->fs_devices->devices, dev_list) {
+ if (device->devid == 1)
+ break;
+ }
- eb->fd = device->fd;
- device->total_ios++;
- eb->dev_bytenr = multi->stripes[0].physical;
- kfree(multi);
- multi = NULL;
+ eb->fd = device->fd;
+ eb->dev_bytenr = eb->start;
+ device->total_ios++;
+ }
if (read_len > bytes_left)
read_len = bytes_left;
eb = btrfs_find_create_tree_block(root, bytenr, blocksize);
if (!eb)
- return NULL;
+ return ERR_PTR(-ENOMEM);
if (btrfs_buffer_uptodate(eb, parent_transid))
return eb;
while (1) {
ret = read_whole_eb(root->fs_info, eb, mirror_num);
- if (ret == 0 && check_tree_block(root, eb) == 0 &&
- csum_tree_block(root, eb, 1) == 0 &&
+ if (ret == 0 && csum_tree_block(root, eb, 1) == 0 &&
+ check_tree_block(root, eb) == 0 &&
verify_parent_transid(eb->tree, eb, parent_transid, ignore)
== 0) {
+ if (eb->flags & EXTENT_BAD_TRANSID &&
+ list_empty(&eb->recow)) {
+ list_add_tail(&eb->recow,
+ &root->fs_info->recow_ebs);
+ eb->refs++;
+ }
btrfs_set_buffer_uptodate(eb);
return eb;
}
if (ignore) {
- if (check_tree_block(root, eb))
- printk("read block failed check_tree_block\n");
- else
- printk("Csum didn't match\n");
+ if (check_tree_block(root, eb)) {
+ if (!root->fs_info->suppress_check_block_errors)
+ print_tree_block_error(root, eb,
+ check_tree_block(root, eb));
+ } else {
+ if (!root->fs_info->suppress_check_block_errors)
+ fprintf(stderr, "Csum didn't match\n");
+ }
+ ret = -EIO;
break;
}
num_copies = btrfs_num_copies(&root->fs_info->mapping_tree,
ignore = 1;
continue;
}
- if (btrfs_header_generation(eb) > best_transid) {
+ if (btrfs_header_generation(eb) > best_transid && mirror_num) {
best_transid = btrfs_header_generation(eb);
good_mirror = mirror_num;
}
}
}
free_extent_buffer(eb);
- return NULL;
-}
-
-static int rmw_eb(struct btrfs_fs_info *info,
- struct extent_buffer *eb, struct extent_buffer *orig_eb)
-{
- int ret;
- unsigned long orig_off = 0;
- unsigned long dest_off = 0;
- unsigned long copy_len = eb->len;
-
- ret = read_whole_eb(info, eb, 0);
- if (ret)
- return ret;
-
- if (eb->start + eb->len <= orig_eb->start ||
- eb->start >= orig_eb->start + orig_eb->len)
- return 0;
- /*
- * | ----- orig_eb ------- |
- * | ----- stripe ------- |
- * | ----- orig_eb ------- |
- * | ----- orig_eb ------- |
- */
- if (eb->start > orig_eb->start)
- orig_off = eb->start - orig_eb->start;
- if (orig_eb->start > eb->start)
- dest_off = orig_eb->start - eb->start;
-
- if (copy_len > orig_eb->len - orig_off)
- copy_len = orig_eb->len - orig_off;
- if (copy_len > eb->len - dest_off)
- copy_len = eb->len - dest_off;
-
- memcpy(eb->data + dest_off, orig_eb->data + orig_off, copy_len);
- return 0;
-}
-
-static void split_eb_for_raid56(struct btrfs_fs_info *info,
- struct extent_buffer *orig_eb,
- struct extent_buffer **ebs,
- u64 stripe_len, u64 *raid_map,
- int num_stripes)
-{
- struct extent_buffer *eb;
- u64 start = orig_eb->start;
- u64 this_eb_start;
- int i;
- int ret;
-
- for (i = 0; i < num_stripes; i++) {
- if (raid_map[i] >= BTRFS_RAID5_P_STRIPE)
- break;
-
- eb = malloc(sizeof(struct extent_buffer) + stripe_len);
- if (!eb)
- BUG();
- memset(eb, 0, sizeof(struct extent_buffer) + stripe_len);
-
- eb->start = raid_map[i];
- eb->len = stripe_len;
- eb->refs = 1;
- eb->flags = 0;
- eb->fd = -1;
- eb->dev_bytenr = (u64)-1;
-
- this_eb_start = raid_map[i];
-
- if (start > this_eb_start ||
- start + orig_eb->len < this_eb_start + stripe_len) {
- ret = rmw_eb(info, eb, orig_eb);
- BUG_ON(ret);
- } else {
- memcpy(eb->data, orig_eb->data + eb->start - start, stripe_len);
- }
- ebs[i] = eb;
- }
+ return ERR_PTR(ret);
}
-static int write_raid56_with_parity(struct btrfs_fs_info *info,
- struct extent_buffer *eb,
- struct btrfs_multi_bio *multi,
- u64 stripe_len, u64 *raid_map)
-{
- struct extent_buffer *ebs[multi->num_stripes], *p_eb = NULL, *q_eb = NULL;
- int i;
- int j;
- int ret;
- int alloc_size = eb->len;
-
- if (stripe_len > alloc_size)
- alloc_size = stripe_len;
-
- split_eb_for_raid56(info, eb, ebs, stripe_len, raid_map,
- multi->num_stripes);
-
- for (i = 0; i < multi->num_stripes; i++) {
- struct extent_buffer *new_eb;
- if (raid_map[i] < BTRFS_RAID5_P_STRIPE) {
- ebs[i]->dev_bytenr = multi->stripes[i].physical;
- ebs[i]->fd = multi->stripes[i].dev->fd;
- multi->stripes[i].dev->total_ios++;
- BUG_ON(ebs[i]->start != raid_map[i]);
- continue;
- }
- new_eb = kmalloc(sizeof(*eb) + alloc_size, GFP_NOFS);
- BUG_ON(!new_eb);
- new_eb->dev_bytenr = multi->stripes[i].physical;
- new_eb->fd = multi->stripes[i].dev->fd;
- multi->stripes[i].dev->total_ios++;
- new_eb->len = stripe_len;
-
- if (raid_map[i] == BTRFS_RAID5_P_STRIPE)
- p_eb = new_eb;
- else if (raid_map[i] == BTRFS_RAID6_Q_STRIPE)
- q_eb = new_eb;
- }
- if (q_eb) {
- void *pointers[multi->num_stripes];
- ebs[multi->num_stripes - 2] = p_eb;
- ebs[multi->num_stripes - 1] = q_eb;
-
- for (i = 0; i < multi->num_stripes; i++)
- pointers[i] = ebs[i]->data;
-
- raid6_gen_syndrome(multi->num_stripes, stripe_len, pointers);
- } else {
- ebs[multi->num_stripes - 1] = p_eb;
- memcpy(p_eb->data, ebs[0]->data, stripe_len);
- for (j = 1; j < multi->num_stripes - 1; j++) {
- for (i = 0; i < stripe_len; i += sizeof(unsigned long)) {
- *(unsigned long *)(p_eb->data + i) ^=
- *(unsigned long *)(ebs[j]->data + i);
- }
- }
- }
-
- for (i = 0; i < multi->num_stripes; i++) {
- ret = write_extent_to_disk(ebs[i]);
- BUG_ON(ret);
- if (ebs[i] != eb)
- kfree(ebs[i]);
- }
- return 0;
-}
-
-int write_tree_block(struct btrfs_trans_handle *trans, struct btrfs_root *root,
+int write_and_map_eb(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root,
struct extent_buffer *eb)
{
int ret;
u64 *raid_map = NULL;
struct btrfs_multi_bio *multi = NULL;
- if (check_tree_block(root, eb))
- BUG();
- if (!btrfs_buffer_uptodate(eb, trans->transid))
- BUG();
-
- btrfs_set_header_flag(eb, BTRFS_HEADER_FLAG_WRITTEN);
- csum_tree_block(root, eb, 0);
-
dev_nr = 0;
length = eb->len;
ret = btrfs_map_block(&root->fs_info->mapping_tree, WRITE,
return 0;
}
+int write_tree_block(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root,
+ struct extent_buffer *eb)
+{
+ if (check_tree_block(root, eb)) {
+ print_tree_block_error(root, eb, check_tree_block(root, eb));
+ BUG();
+ }
+
+ if (trans && !btrfs_buffer_uptodate(eb, trans->transid))
+ BUG();
+
+ btrfs_set_header_flag(eb, BTRFS_HEADER_FLAG_WRITTEN);
+ csum_tree_block(root, eb, 0);
+
+ return write_and_map_eb(trans, root, eb);
+}
+
int __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
u32 stripesize, struct btrfs_root *root,
struct btrfs_fs_info *fs_info, u64 objectid)
root->last_inode_alloc = 0;
INIT_LIST_HEAD(&root->dirty_list);
+ INIT_LIST_HEAD(&root->orphan_data_extents);
memset(&root->root_key, 0, sizeof(root->root_key));
memset(&root->root_item, 0, sizeof(root->root_item));
root->root_key.objectid = objectid;
if (root->commit_root == root->node)
goto commit_tree;
+ if (root == root->fs_info->tree_root)
+ goto commit_tree;
free_extent_buffer(root->commit_root);
root->commit_root = NULL;
return 0;
}
-
int btrfs_free_fs_root(struct btrfs_root *root)
{
if (root->node)
return 0;
}
-static void __free_fs_root(struct cache_extent *cache)
+static void __free_fs_root(struct rb_node *node)
{
struct btrfs_root *root;
- root = container_of(cache, struct btrfs_root, cache);
+ root = container_of(node, struct btrfs_root, rb_node);
btrfs_free_fs_root(root);
}
-FREE_EXTENT_CACHE_BASED_TREE(fs_roots, __free_fs_root);
+FREE_RB_BASED_TREE(fs_roots, __free_fs_root);
struct btrfs_root *btrfs_read_fs_root_no_cache(struct btrfs_fs_info *fs_info,
struct btrfs_key *location)
memcpy(&root->root_key, location, sizeof(*location));
ret = 0;
out:
- btrfs_release_path(root, path);
btrfs_free_path(path);
if (ret) {
free(root);
blocksize = btrfs_level_size(root, btrfs_root_level(&root->root_item));
root->node = read_tree_block(root, btrfs_root_bytenr(&root->root_item),
blocksize, generation);
- BUG_ON(!root->node);
+ if (!extent_buffer_uptodate(root->node)) {
+ free(root);
+ return ERR_PTR(-EIO);
+ }
insert:
root->ref_cows = 1;
return root;
}
+static int btrfs_fs_roots_compare_objectids(struct rb_node *node,
+ void *data)
+{
+ u64 objectid = *((u64 *)data);
+ struct btrfs_root *root;
+
+ root = rb_entry(node, struct btrfs_root, rb_node);
+ if (objectid > root->objectid)
+ return 1;
+ else if (objectid < root->objectid)
+ return -1;
+ else
+ return 0;
+}
+
+static int btrfs_fs_roots_compare_roots(struct rb_node *node1,
+ struct rb_node *node2)
+{
+ struct btrfs_root *root;
+
+ root = rb_entry(node2, struct btrfs_root, rb_node);
+ return btrfs_fs_roots_compare_objectids(node1, (void *)&root->objectid);
+}
+
struct btrfs_root *btrfs_read_fs_root(struct btrfs_fs_info *fs_info,
struct btrfs_key *location)
{
struct btrfs_root *root;
- struct cache_extent *cache;
+ struct rb_node *node;
int ret;
+ u64 objectid = location->objectid;
if (location->objectid == BTRFS_ROOT_TREE_OBJECTID)
return fs_info->tree_root;
return fs_info->dev_root;
if (location->objectid == BTRFS_CSUM_TREE_OBJECTID)
return fs_info->csum_root;
+ if (location->objectid == BTRFS_QUOTA_TREE_OBJECTID)
+ return fs_info->quota_root;
BUG_ON(location->objectid == BTRFS_TREE_RELOC_OBJECTID ||
location->offset != (u64)-1);
- cache = find_cache_extent(&fs_info->fs_root_cache,
- location->objectid, 1);
- if (cache)
- return container_of(cache, struct btrfs_root, cache);
+ node = rb_search(&fs_info->fs_root_tree, (void *)&objectid,
+ btrfs_fs_roots_compare_objectids, NULL);
+ if (node)
+ return container_of(node, struct btrfs_root, rb_node);
root = btrfs_read_fs_root_no_cache(fs_info, location);
if (IS_ERR(root))
return root;
- root->cache.start = location->objectid;
- root->cache.size = 1;
- ret = insert_cache_extent(&fs_info->fs_root_cache, &root->cache);
+ ret = rb_insert(&fs_info->fs_root_tree, &root->rb_node,
+ btrfs_fs_roots_compare_roots);
BUG_ON(ret);
return root;
}
free(fs_info->chunk_root);
free(fs_info->dev_root);
free(fs_info->csum_root);
+ free(fs_info->quota_root);
free(fs_info->super_copy);
free(fs_info->log_root_tree);
free(fs_info);
fs_info->chunk_root = malloc(sizeof(struct btrfs_root));
fs_info->dev_root = malloc(sizeof(struct btrfs_root));
fs_info->csum_root = malloc(sizeof(struct btrfs_root));
+ fs_info->quota_root = malloc(sizeof(struct btrfs_root));
fs_info->super_copy = malloc(BTRFS_SUPER_INFO_SIZE);
if (!fs_info->tree_root || !fs_info->extent_root ||
!fs_info->chunk_root || !fs_info->dev_root ||
- !fs_info->csum_root || !fs_info->super_copy)
+ !fs_info->csum_root || !fs_info->quota_root ||
+ !fs_info->super_copy)
goto free_all;
memset(fs_info->super_copy, 0, BTRFS_SUPER_INFO_SIZE);
memset(fs_info->chunk_root, 0, sizeof(struct btrfs_root));
memset(fs_info->dev_root, 0, sizeof(struct btrfs_root));
memset(fs_info->csum_root, 0, sizeof(struct btrfs_root));
+ memset(fs_info->quota_root, 0, sizeof(struct btrfs_root));
extent_io_tree_init(&fs_info->extent_cache);
extent_io_tree_init(&fs_info->free_space_cache);
extent_io_tree_init(&fs_info->pinned_extents);
extent_io_tree_init(&fs_info->pending_del);
extent_io_tree_init(&fs_info->extent_ins);
+ fs_info->excluded_extents = NULL;
- cache_tree_init(&fs_info->fs_root_cache);
+ fs_info->fs_root_tree = RB_ROOT;
cache_tree_init(&fs_info->mapping_tree.cache_tree);
mutex_init(&fs_info->fs_mutex);
INIT_LIST_HEAD(&fs_info->dirty_cowonly_roots);
INIT_LIST_HEAD(&fs_info->space_info);
+ INIT_LIST_HEAD(&fs_info->recow_ebs);
if (!writable)
fs_info->readonly = 1;
return 0;
}
-int btrfs_setup_all_roots(struct btrfs_fs_info *fs_info,
- u64 root_tree_bytenr, int partial)
+static int find_best_backup_root(struct btrfs_super_block *super)
+{
+ struct btrfs_root_backup *backup;
+ u64 orig_gen = btrfs_super_generation(super);
+ u64 gen = 0;
+ int best_index = 0;
+ int i;
+
+ for (i = 0; i < BTRFS_NUM_BACKUP_ROOTS; i++) {
+ backup = super->super_roots + i;
+ if (btrfs_backup_tree_root_gen(backup) != orig_gen &&
+ btrfs_backup_tree_root_gen(backup) > gen) {
+ best_index = i;
+ gen = btrfs_backup_tree_root_gen(backup);
+ }
+ }
+ return best_index;
+}
+
+static int setup_root_or_create_block(struct btrfs_fs_info *fs_info,
+ enum btrfs_open_ctree_flags flags,
+ struct btrfs_root *info_root,
+ u64 objectid, char *str)
+{
+ struct btrfs_super_block *sb = fs_info->super_copy;
+ struct btrfs_root *root = fs_info->tree_root;
+ u32 leafsize = btrfs_super_leafsize(sb);
+ int ret;
+
+ ret = find_and_setup_root(root, fs_info, objectid, info_root);
+ if (ret) {
+ printk("Couldn't setup %s tree\n", str);
+ if (!(flags & OPEN_CTREE_PARTIAL))
+ return -EIO;
+ /*
+ * Need a blank node here just so we don't screw up in the
+ * million of places that assume a root has a valid ->node
+ */
+ info_root->node =
+ btrfs_find_create_tree_block(info_root, 0, leafsize);
+ if (!info_root->node)
+ return -ENOMEM;
+ clear_extent_buffer_uptodate(NULL, info_root->node);
+ }
+
+ return 0;
+}
+
+int btrfs_setup_all_roots(struct btrfs_fs_info *fs_info, u64 root_tree_bytenr,
+ enum btrfs_open_ctree_flags flags)
{
struct btrfs_super_block *sb = fs_info->super_copy;
struct btrfs_root *root;
blocksize = btrfs_level_size(root, btrfs_super_root_level(sb));
generation = btrfs_super_generation(sb);
- if (!root_tree_bytenr)
+ if (!root_tree_bytenr && !(flags & OPEN_CTREE_BACKUP_ROOT)) {
root_tree_bytenr = btrfs_super_root(sb);
+ } else if (flags & OPEN_CTREE_BACKUP_ROOT) {
+ struct btrfs_root_backup *backup;
+ int index = find_best_backup_root(sb);
+ if (index >= BTRFS_NUM_BACKUP_ROOTS) {
+ fprintf(stderr, "Invalid backup root number\n");
+ return -EIO;
+ }
+ backup = fs_info->super_copy->super_roots + index;
+ root_tree_bytenr = btrfs_backup_tree_root(backup);
+ generation = btrfs_backup_tree_root_gen(backup);
+ }
+
root->node = read_tree_block(root, root_tree_bytenr, blocksize,
generation);
if (!extent_buffer_uptodate(root->node)) {
return -EIO;
}
- ret = find_and_setup_root(root, fs_info, BTRFS_EXTENT_TREE_OBJECTID,
- fs_info->extent_root);
- if (ret) {
- printk("Couldn't setup extent tree\n");
- return -EIO;
- }
+ ret = setup_root_or_create_block(fs_info, flags, fs_info->extent_root,
+ BTRFS_EXTENT_TREE_OBJECTID, "extent");
+ if (ret)
+ return ret;
fs_info->extent_root->track_dirty = 1;
ret = find_and_setup_root(root, fs_info, BTRFS_DEV_TREE_OBJECTID,
}
fs_info->dev_root->track_dirty = 1;
- ret = find_and_setup_root(root, fs_info, BTRFS_CSUM_TREE_OBJECTID,
- fs_info->csum_root);
- if (ret) {
- printk("Couldn't setup csum tree\n");
- if (!partial)
- return -EIO;
- }
+ ret = setup_root_or_create_block(fs_info, flags, fs_info->csum_root,
+ BTRFS_CSUM_TREE_OBJECTID, "csum");
+ if (ret)
+ return ret;
fs_info->csum_root->track_dirty = 1;
+ ret = find_and_setup_root(root, fs_info, BTRFS_QUOTA_TREE_OBJECTID,
+ fs_info->quota_root);
+ if (ret == 0)
+ fs_info->quota_enabled = 1;
+
ret = find_and_setup_log_root(root, fs_info, sb);
if (ret) {
printk("Couldn't setup log root tree\n");
- return -EIO;
+ if (!(flags & OPEN_CTREE_PARTIAL))
+ return -EIO;
}
fs_info->generation = generation;
fs_info->last_trans_committed = generation;
- btrfs_read_block_groups(fs_info->tree_root);
+ if (extent_buffer_uptodate(fs_info->extent_root->node) &&
+ !(flags & OPEN_CTREE_NO_BLOCK_GROUPS))
+ btrfs_read_block_groups(fs_info->tree_root);
key.objectid = BTRFS_FS_TREE_OBJECTID;
key.type = BTRFS_ROOT_ITEM_KEY;
key.offset = (u64)-1;
fs_info->fs_root = btrfs_read_fs_root(fs_info, &key);
- if (!fs_info->fs_root)
+ if (IS_ERR(fs_info->fs_root))
return -EIO;
return 0;
}
void btrfs_release_all_roots(struct btrfs_fs_info *fs_info)
{
+ if (fs_info->quota_root)
+ free_extent_buffer(fs_info->quota_root->node);
if (fs_info->csum_root)
free_extent_buffer(fs_info->csum_root->node);
if (fs_info->dev_root)
void btrfs_cleanup_all_caches(struct btrfs_fs_info *fs_info)
{
+ while (!list_empty(&fs_info->recow_ebs)) {
+ struct extent_buffer *eb;
+ eb = list_first_entry(&fs_info->recow_ebs,
+ struct extent_buffer, recow);
+ list_del_init(&eb->recow);
+ free_extent_buffer(eb);
+ }
free_mapping_cache_tree(&fs_info->mapping_tree.cache_tree);
extent_io_tree_cleanup(&fs_info->extent_cache);
extent_io_tree_cleanup(&fs_info->free_space_cache);
}
int btrfs_scan_fs_devices(int fd, const char *path,
- struct btrfs_fs_devices **fs_devices)
+ struct btrfs_fs_devices **fs_devices,
+ u64 sb_bytenr, int super_recover,
+ int skip_devices)
{
u64 total_devs;
+ u64 dev_size;
+ off_t seek_ret;
int ret;
+ if (!sb_bytenr)
+ sb_bytenr = BTRFS_SUPER_INFO_OFFSET;
+
+ seek_ret = lseek(fd, 0, SEEK_END);
+ if (seek_ret < 0)
+ return -errno;
+
+ dev_size = seek_ret;
+ lseek(fd, 0, SEEK_SET);
+ if (sb_bytenr > dev_size) {
+ fprintf(stderr, "Superblock bytenr is larger than device size\n");
+ return -EINVAL;
+ }
ret = btrfs_scan_one_device(fd, path, fs_devices,
- &total_devs, BTRFS_SUPER_INFO_OFFSET);
+ &total_devs, sb_bytenr, super_recover);
if (ret) {
fprintf(stderr, "No valid Btrfs found on %s\n", path);
return ret;
}
- if (total_devs != 1) {
- ret = btrfs_scan_for_fsid(*fs_devices, total_devs, 1);
+ if (!skip_devices && total_devs != 1) {
+ ret = btrfs_scan_lblkid();
if (ret)
return ret;
}
fs_info->chunk_root->node = read_tree_block(fs_info->chunk_root,
btrfs_super_chunk_root(sb),
blocksize, generation);
- if (!fs_info->chunk_root->node ||
- !extent_buffer_uptodate(fs_info->chunk_root->node)) {
+ if (!extent_buffer_uptodate(fs_info->chunk_root->node)) {
fprintf(stderr, "Couldn't read chunk root\n");
- return ret;
+ return -EIO;
}
if (!(btrfs_super_flags(sb) & BTRFS_SUPER_FLAG_METADUMP)) {
static struct btrfs_fs_info *__open_ctree_fd(int fp, const char *path,
u64 sb_bytenr,
- u64 root_tree_bytenr, int writes,
- int partial)
+ u64 root_tree_bytenr,
+ enum btrfs_open_ctree_flags flags)
{
struct btrfs_fs_info *fs_info;
struct btrfs_super_block *disk_super;
struct btrfs_fs_devices *fs_devices = NULL;
struct extent_buffer *eb;
int ret;
+ int oflags;
if (sb_bytenr == 0)
sb_bytenr = BTRFS_SUPER_INFO_OFFSET;
if (posix_fadvise(fp, 0, 0, POSIX_FADV_DONTNEED))
fprintf(stderr, "Warning, could not drop caches\n");
- fs_info = btrfs_new_fs_info(writes, sb_bytenr);
+ fs_info = btrfs_new_fs_info(flags & OPEN_CTREE_WRITES, sb_bytenr);
if (!fs_info) {
fprintf(stderr, "Failed to allocate memory for fs_info\n");
return NULL;
}
-
- ret = btrfs_scan_fs_devices(fp, path, &fs_devices);
+ if (flags & OPEN_CTREE_RESTORE)
+ fs_info->on_restoring = 1;
+ if (flags & OPEN_CTREE_SUPPRESS_CHECK_BLOCK_ERRORS)
+ fs_info->suppress_check_block_errors = 1;
+ if (flags & OPEN_CTREE_IGNORE_FSID_MISMATCH)
+ fs_info->ignore_fsid_mismatch = 1;
+
+ ret = btrfs_scan_fs_devices(fp, path, &fs_devices, sb_bytenr,
+ (flags & OPEN_CTREE_RECOVER_SUPER),
+ (flags & OPEN_CTREE_NO_DEVICES));
if (ret)
goto out;
fs_info->fs_devices = fs_devices;
- if (writes)
- ret = btrfs_open_devices(fs_devices, O_RDWR);
+ if (flags & OPEN_CTREE_WRITES)
+ oflags = O_RDWR;
else
- ret = btrfs_open_devices(fs_devices, O_RDONLY);
- if (ret)
- goto out_devices;
+ oflags = O_RDONLY;
+ if (flags & OPEN_CTREE_EXCLUSIVE)
+ oflags |= O_EXCL;
+
+ ret = btrfs_open_devices(fs_devices, oflags);
+ if (ret)
+ goto out;
disk_super = fs_info->super_copy;
- ret = btrfs_read_dev_super(fs_devices->latest_bdev,
- disk_super, sb_bytenr);
+ if (!(flags & OPEN_CTREE_RECOVER_SUPER))
+ ret = btrfs_read_dev_super(fs_devices->latest_bdev,
+ disk_super, sb_bytenr, 1);
+ else
+ ret = btrfs_read_dev_super(fp, disk_super, sb_bytenr, 0);
if (ret) {
printk("No valid btrfs found\n");
goto out_devices;
}
+ if (btrfs_super_flags(disk_super) & BTRFS_SUPER_FLAG_CHANGING_FSID &&
+ !fs_info->ignore_fsid_mismatch) {
+ fprintf(stderr, "ERROR: Filesystem UUID change in progress\n");
+ goto out_devices;
+ }
+
memcpy(fs_info->fsid, &disk_super->fsid, BTRFS_FSID_SIZE);
- ret = btrfs_check_fs_compatibility(fs_info->super_copy, writes);
+ ret = btrfs_check_fs_compatibility(fs_info->super_copy,
+ flags & OPEN_CTREE_WRITES);
if (ret)
goto out_devices;
eb = fs_info->chunk_root->node;
read_extent_buffer(eb, fs_info->chunk_tree_uuid,
- (unsigned long)btrfs_header_chunk_tree_uuid(eb),
+ btrfs_header_chunk_tree_uuid(eb),
BTRFS_UUID_SIZE);
- ret = btrfs_setup_all_roots(fs_info, root_tree_bytenr, partial);
- if (ret)
- goto out_failed;
+ ret = btrfs_setup_all_roots(fs_info, root_tree_bytenr, flags);
+ if (ret && !(flags & __OPEN_CTREE_RETURN_CHUNK_ROOT))
+ goto out_chunk;
return fs_info;
-out_failed:
- if (partial)
- return fs_info;
out_chunk:
btrfs_release_all_roots(fs_info);
btrfs_cleanup_all_caches(fs_info);
struct btrfs_fs_info *open_ctree_fs_info(const char *filename,
u64 sb_bytenr, u64 root_tree_bytenr,
- int writes, int partial)
+ enum btrfs_open_ctree_flags flags)
{
int fp;
struct btrfs_fs_info *info;
- int flags = O_CREAT | O_RDWR;
+ int oflags = O_CREAT | O_RDWR;
- if (!writes)
- flags = O_RDONLY;
+ if (!(flags & OPEN_CTREE_WRITES))
+ oflags = O_RDONLY;
- fp = open(filename, flags, 0600);
+ fp = open(filename, oflags, 0600);
if (fp < 0) {
fprintf (stderr, "Could not open %s\n", filename);
return NULL;
}
info = __open_ctree_fd(fp, filename, sb_bytenr, root_tree_bytenr,
- writes, partial);
+ flags);
close(fp);
return info;
}
-struct btrfs_root *open_ctree(const char *filename, u64 sb_bytenr, int writes)
+struct btrfs_root *open_ctree(const char *filename, u64 sb_bytenr,
+ enum btrfs_open_ctree_flags flags)
{
struct btrfs_fs_info *info;
- info = open_ctree_fs_info(filename, sb_bytenr, 0, writes, 0);
+ info = open_ctree_fs_info(filename, sb_bytenr, 0, flags);
if (!info)
return NULL;
+ if (flags & __OPEN_CTREE_RETURN_CHUNK_ROOT)
+ return info->chunk_root;
return info->fs_root;
}
struct btrfs_root *open_ctree_fd(int fp, const char *path, u64 sb_bytenr,
- int writes)
+ enum btrfs_open_ctree_flags flags)
{
struct btrfs_fs_info *info;
- info = __open_ctree_fd(fp, path, sb_bytenr, 0, writes, 0);
+ info = __open_ctree_fd(fp, path, sb_bytenr, 0, flags);
if (!info)
return NULL;
+ if (flags & __OPEN_CTREE_RETURN_CHUNK_ROOT)
+ return info->chunk_root;
return info->fs_root;
}
-int btrfs_read_dev_super(int fd, struct btrfs_super_block *sb, u64 sb_bytenr)
+int btrfs_read_dev_super(int fd, struct btrfs_super_block *sb, u64 sb_bytenr,
+ int super_recover)
{
u8 fsid[BTRFS_FSID_SIZE];
int fsid_is_initialized = 0;
struct btrfs_super_block buf;
int i;
int ret;
+ int max_super = super_recover ? BTRFS_SUPER_MIRROR_MAX : 1;
u64 transid = 0;
u64 bytenr;
return -1;
if (btrfs_super_bytenr(&buf) != sb_bytenr ||
- buf.magic != cpu_to_le64(BTRFS_MAGIC))
+ btrfs_super_magic(&buf) != BTRFS_MAGIC)
return -1;
memcpy(sb, &buf, sizeof(*sb));
return 0;
}
- for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
+ /*
+ * we would like to check all the supers, but that would make
+ * a btrfs mount succeed after a mkfs from a different FS.
+ * So, we need to add a special mount option to scan for
+ * later supers, using BTRFS_SUPER_MIRROR_MAX instead
+ */
+
+ for (i = 0; i < max_super; i++) {
bytenr = btrfs_sb_offset(i);
ret = pread64(fd, &buf, sizeof(buf), bytenr);
if (ret < sizeof(buf))
if (btrfs_super_bytenr(&buf) != bytenr )
continue;
/* if magic is NULL, the device was removed */
- if (buf.magic == 0 && i == 0)
+ if (btrfs_super_magic(&buf) == 0 && i == 0)
return -1;
- if (buf.magic != cpu_to_le64(BTRFS_MAGIC))
+ if (btrfs_super_magic(&buf) != BTRFS_MAGIC)
continue;
if (!fsid_is_initialized) {
return transid > 0 ? 0 : -1;
}
-int write_dev_supers(struct btrfs_root *root, struct btrfs_super_block *sb,
- struct btrfs_device *device)
+static int write_dev_supers(struct btrfs_root *root,
+ struct btrfs_super_block *sb,
+ struct btrfs_device *device)
{
u64 bytenr;
u32 crc;
}
btrfs_free_block_groups(fs_info);
- free_fs_roots_tree(&fs_info->fs_root_cache);
+ free_fs_roots_tree(&fs_info->fs_root_tree);
btrfs_release_all_roots(fs_info);
btrfs_close_devices(fs_info->fs_devices);
projects / platform / upstream / btrfs-progs.git / blobdiff