* 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"
-static int close_all_devices(struct btrfs_fs_info *fs_info);
+/* 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)
-static int check_tree_block(struct btrfs_root *root, struct extent_buffer *buf)
+/* Calculate max possible nritems for a leaf/node */
+static u32 max_nritems(u8 level, u32 nodesize)
{
- struct btrfs_fs_devices *fs_devices;
- int ret = 1;
+ if (level == 0)
+ return ((nodesize - sizeof(struct btrfs_header)) /
+ sizeof(struct btrfs_item));
+ return ((nodesize - sizeof(struct btrfs_header)) /
+ sizeof(struct btrfs_key_ptr));
+}
- 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;
- }
+static int check_tree_block(struct btrfs_fs_info *fs_info,
+ struct extent_buffer *buf)
+{
- fs_devices = root->fs_info->fs_devices;
+ struct btrfs_fs_devices *fs_devices;
+ u32 nodesize = btrfs_super_nodesize(fs_info->super_copy);
+ int ret = BTRFS_BAD_FSID;
+
+ 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),
+ nodesize))
+ return BTRFS_BAD_NRITEMS;
+
+ /* Only leaf can be empty */
+ if (btrfs_header_nritems(buf) == 0 &&
+ btrfs_header_level(buf) != 0)
+ return BTRFS_BAD_NRITEMS;
+
+ fs_devices = fs_info->fs_devices;
while (fs_devices) {
- if (!memcmp_extent_buffer(buf, fs_devices->fsid,
- (unsigned long)btrfs_header_fsid(buf),
+ if (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_fs_info *fs_info,
+ 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(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);
void btrfs_csum_final(u32 crc, char *result)
{
- *(__le32 *)result = ~cpu_to_le32(crc);
+ put_unaligned_le32(~crc, result);
}
-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;
+ char result[BTRFS_CSUM_SIZE];
u32 len;
u32 crc = ~(u32)0;
- result = malloc(csum_size * sizeof(char));
- if (!result)
- return 1;
-
len = buf->len - BTRFS_CSUM_SIZE;
crc = crc32c(crc, buf->data + BTRFS_CSUM_SIZE, len);
btrfs_csum_final(crc, result);
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));
- free(result);
+ if (!silent)
+ printk("checksum verify failed on %llu found %08X wanted %08X\n",
+ (unsigned long long)buf->start,
+ *((u32 *)result),
+ *((u32*)(char *)buf->data));
return 1;
}
} else {
write_extent_buffer(buf, result, 0, csum_size);
}
- free(result);
return 0;
}
-int csum_tree_block(struct btrfs_root *root, struct extent_buffer *buf,
- int verify)
+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);
+}
+
+static int csum_tree_block_fs_info(struct btrfs_fs_info *fs_info,
+ struct extent_buffer *buf, int verify)
{
u16 csum_size =
- btrfs_super_csum_size(root->fs_info->super_copy);
+ btrfs_super_csum_size(fs_info->super_copy);
+ if (verify && fs_info->suppress_check_block_errors)
+ return verify_tree_block_csum_silent(buf, csum_size);
return csum_tree_block_size(buf, csum_size, verify);
}
+int csum_tree_block(struct btrfs_root *root, struct extent_buffer *buf,
+ int verify)
+{
+ return csum_tree_block_fs_info(root->fs_info, buf, verify);
+}
+
struct extent_buffer *btrfs_find_tree_block(struct btrfs_root *root,
u64 bytenr, u32 blocksize)
{
bytenr, blocksize);
}
-struct extent_buffer *btrfs_find_create_tree_block(struct btrfs_root *root,
- u64 bytenr, u32 blocksize)
+struct extent_buffer* btrfs_find_create_tree_block(
+ struct btrfs_fs_info *fs_info, u64 bytenr, u32 blocksize)
{
- return alloc_extent_buffer(&root->fs_info->extent_cache, bytenr,
- blocksize);
+ return alloc_extent_buffer(&fs_info->extent_cache, bytenr, 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;
}
read_len = bytes_left;
device = NULL;
- if (!info->on_restoring) {
+ 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);
}
device = multi->stripes[0].dev;
- if (device->fd == 0) {
+ if (device->fd <= 0) {
kfree(multi);
return -EIO;
}
return 0;
}
-struct extent_buffer *read_tree_block(struct btrfs_root *root, u64 bytenr,
- u32 blocksize, u64 parent_transid)
+struct extent_buffer* read_tree_block_fs_info(
+ struct btrfs_fs_info *fs_info, u64 bytenr, u32 blocksize,
+ u64 parent_transid)
{
int ret;
struct extent_buffer *eb;
int num_copies;
int ignore = 0;
- eb = btrfs_find_create_tree_block(root, bytenr, blocksize);
+ eb = btrfs_find_create_tree_block(fs_info, 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 &&
+ ret = read_whole_eb(fs_info, eb, mirror_num);
+ if (ret == 0 && csum_tree_block_fs_info(fs_info, eb, 1) == 0 &&
+ check_tree_block(fs_info, 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,
+ &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(fs_info, eb)) {
+ if (!fs_info->suppress_check_block_errors)
+ print_tree_block_error(fs_info, eb,
+ check_tree_block(fs_info, eb));
+ } else {
+ if (!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,
+ num_copies = btrfs_num_copies(&fs_info->mapping_tree,
eb->start, eb->len);
if (num_copies == 1) {
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;
+ return ERR_PTR(ret);
}
-int write_tree_block(struct btrfs_trans_handle *trans, struct btrfs_root *root,
+int read_extent_data(struct btrfs_root *root, char *data,
+ u64 logical, u64 *len, int mirror)
+{
+ u64 offset = 0;
+ struct btrfs_multi_bio *multi = NULL;
+ struct btrfs_fs_info *info = root->fs_info;
+ struct btrfs_device *device;
+ int ret = 0;
+ u64 max_len = *len;
+
+ ret = btrfs_map_block(&info->mapping_tree, READ, logical, len,
+ &multi, mirror, NULL);
+ if (ret) {
+ fprintf(stderr, "Couldn't map the block %llu\n",
+ logical + offset);
+ goto err;
+ }
+ device = multi->stripes[0].dev;
+
+ if (device->fd <= 0)
+ goto err;
+ if (*len > max_len)
+ *len = max_len;
+
+ ret = pread64(device->fd, data, *len, multi->stripes[0].physical);
+ if (ret != *len)
+ ret = -EIO;
+ else
+ ret = 0;
+err:
+ kfree(multi);
+ return ret;
+}
+
+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,
ret = write_extent_to_disk(eb);
BUG_ON(ret);
}
+ kfree(raid_map);
kfree(multi);
return 0;
}
+int write_tree_block(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root,
+ struct extent_buffer *eb)
+{
+ if (check_tree_block(root->fs_info, eb)) {
+ print_tree_block_error(root->fs_info, eb,
+ check_tree_block(root->fs_info, 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;
+ if (root == root->fs_info->chunk_root)
+ goto commit_tree;
free_extent_buffer(root->commit_root);
root->commit_root = NULL;
if (ret)
return ret;
- blocksize = btrfs_level_size(root, btrfs_root_level(&root->root_item));
+ blocksize = root->nodesize;
generation = btrfs_root_generation(&root->root_item);
root->node = read_tree_block(root, btrfs_root_bytenr(&root->root_item),
blocksize, generation);
return 0;
}
- blocksize = btrfs_level_size(tree_root,
- btrfs_super_log_root_level(disk_super));
+ blocksize = tree_root->nodesize;
__setup_root(tree_root->nodesize, tree_root->leafsize,
tree_root->sectorsize, tree_root->stripesize,
return 0;
}
-
-int btrfs_free_fs_root(struct btrfs_fs_info *fs_info,
- struct btrfs_root *root)
+int btrfs_free_fs_root(struct btrfs_root *root)
{
if (root->node)
free_extent_buffer(root->node);
return 0;
}
-static int free_fs_roots(struct btrfs_fs_info *fs_info)
+static void __free_fs_root(struct rb_node *node)
{
- struct cache_extent *cache;
struct btrfs_root *root;
- while (1) {
- cache = find_first_cache_extent(&fs_info->fs_root_cache, 0);
- if (!cache)
- break;
- root = container_of(cache, struct btrfs_root, cache);
- remove_cache_extent(&fs_info->fs_root_cache, cache);
- btrfs_free_fs_root(fs_info, root);
- }
- return 0;
+ root = container_of(node, struct btrfs_root, rb_node);
+ btrfs_free_fs_root(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)
{
u32 blocksize;
int ret = 0;
- root = malloc(sizeof(*root));
+ root = calloc(1, sizeof(*root));
if (!root)
return ERR_PTR(-ENOMEM);
- memset(root, 0, sizeof(*root));
if (location->offset == (u64)-1) {
ret = find_and_setup_root(tree_root, fs_info,
location->objectid, root);
memcpy(&root->root_key, location, sizeof(*location));
ret = 0;
out:
- btrfs_release_path(root, path);
btrfs_free_path(path);
if (ret) {
free(root);
return ERR_PTR(ret);
}
generation = btrfs_root_generation(&root->root_item);
- blocksize = btrfs_level_size(root, btrfs_root_level(&root->root_item));
+ blocksize = root->nodesize;
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_existing_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;
}
-static struct btrfs_fs_info *__open_ctree_fd(int fp, const char *path,
- u64 sb_bytenr,
- u64 root_tree_bytenr, int writes,
- int partial, int restore)
+void btrfs_free_fs_info(struct btrfs_fs_info *fs_info)
{
- u32 sectorsize;
- u32 nodesize;
- u32 leafsize;
- u32 blocksize;
- u32 stripesize;
- u64 generation;
- struct btrfs_key key;
- struct btrfs_root *tree_root = malloc(sizeof(struct btrfs_root));
- struct btrfs_root *extent_root = malloc(sizeof(struct btrfs_root));
- struct btrfs_root *chunk_root = malloc(sizeof(struct btrfs_root));
- struct btrfs_root *dev_root = malloc(sizeof(struct btrfs_root));
- struct btrfs_root *csum_root = malloc(sizeof(struct btrfs_root));
- struct btrfs_fs_info *fs_info = malloc(sizeof(*fs_info));
- int ret;
- struct btrfs_super_block *disk_super;
- struct btrfs_fs_devices *fs_devices = NULL;
- u64 total_devs;
- u64 features;
-
- memset(tree_root, 0, sizeof(struct btrfs_root));
- memset(extent_root, 0, sizeof(struct btrfs_root));
- memset(chunk_root, 0, sizeof(struct btrfs_root));
- memset(dev_root, 0, sizeof(struct btrfs_root));
- memset(csum_root, 0, sizeof(struct btrfs_root));
-
- if (sb_bytenr == 0)
- sb_bytenr = BTRFS_SUPER_INFO_OFFSET;
-
- /* try to drop all the caches */
- if (posix_fadvise(fp, 0, 0, POSIX_FADV_DONTNEED))
- fprintf(stderr, "Warning, could not drop caches\n");
-
- ret = btrfs_scan_one_device(fp, path, &fs_devices,
- &total_devs, sb_bytenr);
+ free(fs_info->tree_root);
+ free(fs_info->extent_root);
+ free(fs_info->chunk_root);
+ free(fs_info->dev_root);
+ free(fs_info->csum_root);
+ free(fs_info->quota_root);
+ free(fs_info->free_space_root);
+ free(fs_info->super_copy);
+ free(fs_info->log_root_tree);
+ free(fs_info);
+}
- if (ret) {
- fprintf(stderr, "No valid Btrfs found on %s\n", path);
- goto out;
- }
+struct btrfs_fs_info *btrfs_new_fs_info(int writable, u64 sb_bytenr)
+{
+ struct btrfs_fs_info *fs_info;
- if (total_devs != 1) {
- ret = btrfs_scan_for_fsid(fs_devices, total_devs, 1);
- if (ret)
- goto out;
- }
+ fs_info = calloc(1, sizeof(struct btrfs_fs_info));
+ if (!fs_info)
+ return NULL;
- memset(fs_info, 0, sizeof(*fs_info));
+ fs_info->tree_root = calloc(1, sizeof(struct btrfs_root));
+ fs_info->extent_root = calloc(1, sizeof(struct btrfs_root));
+ fs_info->chunk_root = calloc(1, sizeof(struct btrfs_root));
+ fs_info->dev_root = calloc(1, sizeof(struct btrfs_root));
+ fs_info->csum_root = calloc(1, sizeof(struct btrfs_root));
+ fs_info->quota_root = calloc(1, sizeof(struct btrfs_root));
+ fs_info->free_space_root = calloc(1, sizeof(struct btrfs_root));
fs_info->super_copy = calloc(1, BTRFS_SUPER_INFO_SIZE);
- fs_info->tree_root = tree_root;
- fs_info->extent_root = extent_root;
- fs_info->chunk_root = chunk_root;
- fs_info->dev_root = dev_root;
- fs_info->csum_root = csum_root;
- if (!writes)
- fs_info->readonly = 1;
- if (restore)
- fs_info->on_restoring = 1;
+ if (!fs_info->tree_root || !fs_info->extent_root ||
+ !fs_info->chunk_root || !fs_info->dev_root ||
+ !fs_info->csum_root || !fs_info->quota_root ||
+ !fs_info->free_space_root || !fs_info->super_copy)
+ goto free_all;
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);
- cache_tree_init(&fs_info->fs_root_cache);
+ fs_info->excluded_extents = NULL;
+ fs_info->fs_root_tree = RB_ROOT;
cache_tree_init(&fs_info->mapping_tree.cache_tree);
mutex_init(&fs_info->fs_mutex);
- fs_info->fs_devices = fs_devices;
INIT_LIST_HEAD(&fs_info->dirty_cowonly_roots);
INIT_LIST_HEAD(&fs_info->space_info);
+ INIT_LIST_HEAD(&fs_info->recow_ebs);
- __setup_root(4096, 4096, 4096, 4096, tree_root,
- fs_info, BTRFS_ROOT_TREE_OBJECTID);
-
- if (writes)
- ret = btrfs_open_devices(fs_devices, O_RDWR);
- else
- ret = btrfs_open_devices(fs_devices, O_RDONLY);
- if (ret)
- goto out_cleanup;
+ if (!writable)
+ fs_info->readonly = 1;
fs_info->super_bytenr = sb_bytenr;
- disk_super = fs_info->super_copy;
- ret = btrfs_read_dev_super(fs_devices->latest_bdev,
- disk_super, sb_bytenr);
- if (ret) {
- printk("No valid btrfs found\n");
- goto out_devices;
- }
-
- memcpy(fs_info->fsid, &disk_super->fsid, BTRFS_FSID_SIZE);
+ fs_info->data_alloc_profile = (u64)-1;
+ fs_info->metadata_alloc_profile = (u64)-1;
+ fs_info->system_alloc_profile = fs_info->metadata_alloc_profile;
+ return fs_info;
+free_all:
+ btrfs_free_fs_info(fs_info);
+ return NULL;
+}
+int btrfs_check_fs_compatibility(struct btrfs_super_block *sb, int writable)
+{
+ u64 features;
- features = btrfs_super_incompat_flags(disk_super) &
+ features = btrfs_super_incompat_flags(sb) &
~BTRFS_FEATURE_INCOMPAT_SUPP;
if (features) {
printk("couldn't open because of unsupported "
"option features (%Lx).\n",
(unsigned long long)features);
- goto out_devices;
+ return -ENOTSUP;
}
- features = btrfs_super_incompat_flags(disk_super);
+ features = btrfs_super_incompat_flags(sb);
if (!(features & BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF)) {
features |= BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF;
- btrfs_set_super_incompat_flags(disk_super, features);
+ btrfs_set_super_incompat_flags(sb, features);
}
- features = btrfs_super_compat_ro_flags(disk_super) &
+ features = btrfs_super_compat_ro_flags(sb) &
~BTRFS_FEATURE_COMPAT_RO_SUPP;
- if (writes && features) {
+ if (writable && features) {
printk("couldn't open RDWR because of unsupported "
"option features (%Lx).\n",
(unsigned long long)features);
- goto out_devices;
+ return -ENOTSUP;
}
+ return 0;
+}
- nodesize = btrfs_super_nodesize(disk_super);
- leafsize = btrfs_super_leafsize(disk_super);
- sectorsize = btrfs_super_sectorsize(disk_super);
- stripesize = btrfs_super_stripesize(disk_super);
- tree_root->nodesize = nodesize;
- tree_root->leafsize = leafsize;
- tree_root->sectorsize = sectorsize;
- tree_root->stripesize = stripesize;
+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;
- ret = btrfs_read_sys_array(tree_root);
- if (ret)
- goto out_devices;
- blocksize = btrfs_level_size(tree_root,
- btrfs_super_chunk_root_level(disk_super));
- generation = btrfs_super_chunk_root_generation(disk_super);
+ 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;
+}
- __setup_root(nodesize, leafsize, sectorsize, stripesize,
- chunk_root, fs_info, BTRFS_CHUNK_TREE_OBJECTID);
+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 nodesize = btrfs_super_nodesize(sb);
+ int ret;
- chunk_root->node = read_tree_block(chunk_root,
- btrfs_super_chunk_root(disk_super),
- blocksize, generation);
- if (!extent_buffer_uptodate(chunk_root->node)) {
- printk("Couldn't read chunk root\n");
- goto out_devices;
+ 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(fs_info, 0, nodesize);
+ if (!info_root->node)
+ return -ENOMEM;
+ clear_extent_buffer_uptodate(NULL, info_root->node);
}
- read_extent_buffer(chunk_root->node, fs_info->chunk_tree_uuid,
- (unsigned long)btrfs_header_chunk_tree_uuid(chunk_root->node),
- BTRFS_UUID_SIZE);
+ return 0;
+}
- if (!(btrfs_super_flags(disk_super) & BTRFS_SUPER_FLAG_METADUMP)) {
- ret = btrfs_read_chunk_tree(chunk_root);
- if (ret) {
- printk("Couldn't read chunk tree\n");
- goto out_chunk;
- }
- }
+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;
+ struct btrfs_key key;
+ u32 sectorsize;
+ u32 nodesize;
+ u32 leafsize;
+ u32 stripesize;
+ u64 generation;
+ u32 blocksize;
+ int ret;
- blocksize = btrfs_level_size(tree_root,
- btrfs_super_root_level(disk_super));
- generation = btrfs_super_generation(disk_super);
+ nodesize = btrfs_super_nodesize(sb);
+ leafsize = btrfs_super_leafsize(sb);
+ sectorsize = btrfs_super_sectorsize(sb);
+ stripesize = btrfs_super_stripesize(sb);
- if (!root_tree_bytenr)
- root_tree_bytenr = btrfs_super_root(disk_super);
- tree_root->node = read_tree_block(tree_root,
- root_tree_bytenr,
- blocksize, generation);
- if (!extent_buffer_uptodate(tree_root->node)) {
- printk("Couldn't read tree root\n");
- goto out_failed;
+ root = fs_info->tree_root;
+ __setup_root(nodesize, leafsize, sectorsize, stripesize,
+ root, fs_info, BTRFS_ROOT_TREE_OBJECTID);
+ blocksize = root->nodesize;
+ generation = btrfs_super_generation(sb);
+
+ 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);
}
- ret = find_and_setup_root(tree_root, fs_info,
- BTRFS_EXTENT_TREE_OBJECTID, extent_root);
- if (ret) {
- printk("Couldn't setup extent tree\n");
- goto out_failed;
+
+ root->node = read_tree_block(root, root_tree_bytenr, blocksize,
+ generation);
+ if (!extent_buffer_uptodate(root->node)) {
+ fprintf(stderr, "Couldn't read tree root\n");
+ return -EIO;
}
- extent_root->track_dirty = 1;
- ret = find_and_setup_root(tree_root, fs_info,
- BTRFS_DEV_TREE_OBJECTID, dev_root);
+ 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);
if (ret) {
printk("Couldn't setup device tree\n");
- goto out_failed;
+ return -EIO;
}
- dev_root->track_dirty = 1;
+ fs_info->dev_root->track_dirty = 1;
- ret = find_and_setup_root(tree_root, fs_info,
- BTRFS_CSUM_TREE_OBJECTID, csum_root);
- if (ret) {
- printk("Couldn't setup csum tree\n");
- if (!partial)
- goto out_failed;
+ 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;
+
+ if (btrfs_fs_compat_ro(fs_info, BTRFS_FEATURE_COMPAT_RO_FREE_SPACE_TREE)) {
+ ret = find_and_setup_root(root, fs_info, BTRFS_FREE_SPACE_TREE_OBJECTID,
+ fs_info->free_space_root);
+ if (ret) {
+ printk("Couldn't read free space tree\n");
+ return -EIO;
+ }
+ fs_info->free_space_root->track_dirty = 1;
}
- csum_root->track_dirty = 1;
- find_and_setup_log_root(tree_root, fs_info, disk_super);
+ ret = find_and_setup_log_root(root, fs_info, sb);
+ if (ret) {
+ printk("Couldn't setup log root tree\n");
+ 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)
- goto out_failed;
-
- fs_info->data_alloc_profile = (u64)-1;
- fs_info->metadata_alloc_profile = (u64)-1;
- fs_info->system_alloc_profile = fs_info->metadata_alloc_profile;
-
- return fs_info;
-
-out_failed:
- if (partial)
- return fs_info;
+ 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->free_space_root)
+ free_extent_buffer(fs_info->free_space_root->node);
+ 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)
free_extent_buffer(fs_info->extent_root->node);
if (fs_info->tree_root)
free_extent_buffer(fs_info->tree_root->node);
-out_chunk:
+ if (fs_info->log_root_tree)
+ free_extent_buffer(fs_info->log_root_tree->node);
if (fs_info->chunk_root)
free_extent_buffer(fs_info->chunk_root->node);
-out_devices:
- close_all_devices(fs_info);
-out_cleanup:
+}
+
+static void free_map_lookup(struct cache_extent *ce)
+{
+ struct map_lookup *map;
+
+ map = container_of(ce, struct map_lookup, ce);
+ kfree(map);
+}
+
+FREE_EXTENT_CACHE_BASED_TREE(mapping_cache, free_map_lookup);
+
+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);
extent_io_tree_cleanup(&fs_info->block_group_cache);
extent_io_tree_cleanup(&fs_info->pinned_extents);
extent_io_tree_cleanup(&fs_info->pending_del);
extent_io_tree_cleanup(&fs_info->extent_ins);
-out:
- free(tree_root);
- free(extent_root);
- free(chunk_root);
- free(dev_root);
- free(csum_root);
- free(fs_info);
- return NULL;
}
-struct btrfs_fs_info *open_ctree_fs_info_restore(const char *filename,
- u64 sb_bytenr, u64 root_tree_bytenr,
- int writes, int partial)
+int btrfs_scan_fs_devices(int fd, const char *path,
+ struct btrfs_fs_devices **fs_devices,
+ u64 sb_bytenr, int super_recover,
+ int skip_devices)
{
- int fp;
- struct btrfs_fs_info *info;
- int flags = O_CREAT | O_RDWR;
- int restore = 1;
+ u64 total_devs;
+ u64 dev_size;
+ off_t seek_ret;
+ int ret;
+ if (!sb_bytenr)
+ sb_bytenr = BTRFS_SUPER_INFO_OFFSET;
- if (!writes)
- flags = O_RDONLY;
+ seek_ret = lseek(fd, 0, SEEK_END);
+ if (seek_ret < 0)
+ return -errno;
- fp = open(filename, flags, 0600);
- if (fp < 0) {
- fprintf (stderr, "Could not open %s\n", filename);
+ 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, sb_bytenr, super_recover);
+ if (ret) {
+ fprintf(stderr, "No valid Btrfs found on %s\n", path);
+ return ret;
+ }
+
+ if (!skip_devices && total_devs != 1) {
+ ret = btrfs_scan_lblkid();
+ if (ret)
+ return ret;
+ }
+ return 0;
+}
+
+int btrfs_setup_chunk_tree_and_device_map(struct btrfs_fs_info *fs_info,
+ u64 chunk_root_bytenr)
+{
+ struct btrfs_super_block *sb = fs_info->super_copy;
+ u32 sectorsize;
+ u32 nodesize;
+ u32 leafsize;
+ u32 blocksize;
+ u32 stripesize;
+ u64 generation;
+ int ret;
+
+ nodesize = btrfs_super_nodesize(sb);
+ leafsize = btrfs_super_leafsize(sb);
+ sectorsize = btrfs_super_sectorsize(sb);
+ stripesize = btrfs_super_stripesize(sb);
+
+ __setup_root(nodesize, leafsize, sectorsize, stripesize,
+ fs_info->chunk_root, fs_info, BTRFS_CHUNK_TREE_OBJECTID);
+
+ ret = btrfs_read_sys_array(fs_info->chunk_root);
+ if (ret)
+ return ret;
+
+ blocksize = fs_info->chunk_root->nodesize;
+ generation = btrfs_super_chunk_root_generation(sb);
+
+ if (chunk_root_bytenr && !IS_ALIGNED(chunk_root_bytenr,
+ btrfs_super_sectorsize(sb))) {
+ warning("chunk_root_bytenr %llu is unaligned to %u, ignore it",
+ chunk_root_bytenr, btrfs_super_sectorsize(sb));
+ chunk_root_bytenr = 0;
+ }
+
+ if (!chunk_root_bytenr)
+ chunk_root_bytenr = btrfs_super_chunk_root(sb);
+ else
+ generation = 0;
+
+ fs_info->chunk_root->node = read_tree_block(fs_info->chunk_root,
+ chunk_root_bytenr,
+ blocksize, generation);
+ if (!extent_buffer_uptodate(fs_info->chunk_root->node)) {
+ if (fs_info->ignore_chunk_tree_error) {
+ warning("cannot read chunk root, continue anyway");
+ fs_info->chunk_root = NULL;
+ return 0;
+ } else {
+ error("cannot read chunk root");
+ return -EIO;
+ }
+ }
+
+ if (!(btrfs_super_flags(sb) & BTRFS_SUPER_FLAG_METADUMP)) {
+ ret = btrfs_read_chunk_tree(fs_info->chunk_root);
+ if (ret) {
+ fprintf(stderr, "Couldn't read chunk tree\n");
+ return ret;
+ }
+ }
+ return 0;
+}
+
+static struct btrfs_fs_info *__open_ctree_fd(int fp, const char *path,
+ u64 sb_bytenr,
+ u64 root_tree_bytenr,
+ u64 chunk_root_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;
+
+ /* try to drop all the caches */
+ if (posix_fadvise(fp, 0, 0, POSIX_FADV_DONTNEED))
+ fprintf(stderr, "Warning, could not drop caches\n");
+
+ 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;
}
- info = __open_ctree_fd(fp, filename, sb_bytenr, root_tree_bytenr,
- writes, partial, restore);
- close(fp);
- return info;
+ 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;
+ if (flags & OPEN_CTREE_IGNORE_CHUNK_TREE_ERROR)
+ fs_info->ignore_chunk_tree_error = 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 (flags & OPEN_CTREE_WRITES)
+ oflags = O_RDWR;
+ else
+ 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;
+ 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,
+ flags & OPEN_CTREE_WRITES);
+ if (ret)
+ goto out_devices;
+
+ ret = btrfs_setup_chunk_tree_and_device_map(fs_info, chunk_root_bytenr);
+ if (ret)
+ goto out_chunk;
+
+ /* Chunk tree root is unable to read, return directly */
+ if (!fs_info->chunk_root)
+ return fs_info;
+
+ eb = fs_info->chunk_root->node;
+ read_extent_buffer(eb, fs_info->chunk_tree_uuid,
+ btrfs_header_chunk_tree_uuid(eb),
+ BTRFS_UUID_SIZE);
+
+ ret = btrfs_setup_all_roots(fs_info, root_tree_bytenr, flags);
+ if (ret && !(flags & __OPEN_CTREE_RETURN_CHUNK_ROOT) &&
+ !fs_info->ignore_chunk_tree_error)
+ goto out_chunk;
+
+ return fs_info;
+
+out_chunk:
+ btrfs_release_all_roots(fs_info);
+ btrfs_cleanup_all_caches(fs_info);
+out_devices:
+ btrfs_close_devices(fs_devices);
+out:
+ btrfs_free_fs_info(fs_info);
+ return NULL;
}
struct btrfs_fs_info *open_ctree_fs_info(const char *filename,
u64 sb_bytenr, u64 root_tree_bytenr,
- int writes, int partial)
+ u64 chunk_root_bytenr,
+ enum btrfs_open_ctree_flags flags)
{
int fp;
+ int ret;
struct btrfs_fs_info *info;
- int flags = O_CREAT | O_RDWR;
+ int oflags = O_RDWR;
+ struct stat st;
- if (!writes)
- flags = O_RDONLY;
+ ret = stat(filename, &st);
+ if (ret < 0) {
+ error("cannot stat '%s': %s", filename, strerror(errno));
+ return NULL;
+ }
+ if (!(((st.st_mode & S_IFMT) == S_IFREG) || ((st.st_mode & S_IFMT) == S_IFBLK))) {
+ error("not a regular file or block device: %s", filename);
+ return NULL;
+ }
- fp = open(filename, flags, 0600);
+ if (!(flags & OPEN_CTREE_WRITES))
+ oflags = O_RDONLY;
+
+ fp = open(filename, oflags);
if (fp < 0) {
- fprintf (stderr, "Could not open %s\n", filename);
+ error("cannot open '%s': %s", filename, strerror(errno));
return NULL;
}
info = __open_ctree_fd(fp, filename, sb_bytenr, root_tree_bytenr,
- writes, partial, 0);
+ chunk_root_bytenr, 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);
+ /* This flags may not return fs_info with any valid root */
+ BUG_ON(flags & OPEN_CTREE_IGNORE_CHUNK_TREE_ERROR);
+ info = open_ctree_fs_info(filename, sb_bytenr, 0, 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, 0);
+
+ /* This flags may not return fs_info with any valid root */
+ BUG_ON(flags & OPEN_CTREE_IGNORE_CHUNK_TREE_ERROR);
+ info = __open_ctree_fd(fp, path, sb_bytenr, 0, 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)
+/*
+ * Check if the super is valid:
+ * - nodesize/sectorsize - minimum, maximum, alignment
+ * - tree block starts - alignment
+ * - number of devices - something sane
+ * - sys array size - maximum
+ */
+static int check_super(struct btrfs_super_block *sb)
+{
+ char result[BTRFS_CSUM_SIZE];
+ u32 crc;
+ u16 csum_type;
+ int csum_size;
+
+ if (btrfs_super_magic(sb) != BTRFS_MAGIC) {
+ error("superblock magic doesn't match");
+ return -EIO;
+ }
+
+ csum_type = btrfs_super_csum_type(sb);
+ if (csum_type >= ARRAY_SIZE(btrfs_csum_sizes)) {
+ error("unsupported checksum algorithm %u\n", csum_type);
+ return -EIO;
+ }
+ csum_size = btrfs_csum_sizes[csum_type];
+
+ crc = ~(u32)0;
+ crc = btrfs_csum_data(NULL, (char *)sb + BTRFS_CSUM_SIZE, crc,
+ BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE);
+ btrfs_csum_final(crc, result);
+
+ if (memcmp(result, sb->csum, csum_size)) {
+ error("superblock checksum mismatch");
+ return -EIO;
+ }
+ if (btrfs_super_root_level(sb) >= BTRFS_MAX_LEVEL) {
+ error("tree_root level too big: %d >= %d",
+ btrfs_super_root_level(sb), BTRFS_MAX_LEVEL);
+ goto error_out;
+ }
+ if (btrfs_super_chunk_root_level(sb) >= BTRFS_MAX_LEVEL) {
+ error("chunk_root level too big: %d >= %d",
+ btrfs_super_chunk_root_level(sb), BTRFS_MAX_LEVEL);
+ goto error_out;
+ }
+ if (btrfs_super_log_root_level(sb) >= BTRFS_MAX_LEVEL) {
+ error("log_root level too big: %d >= %d",
+ btrfs_super_log_root_level(sb), BTRFS_MAX_LEVEL);
+ goto error_out;
+ }
+
+ if (!IS_ALIGNED(btrfs_super_root(sb), 4096)) {
+ error("tree_root block unaligned: %llu", btrfs_super_root(sb));
+ goto error_out;
+ }
+ if (!IS_ALIGNED(btrfs_super_chunk_root(sb), 4096)) {
+ error("chunk_root block unaligned: %llu",
+ btrfs_super_chunk_root(sb));
+ goto error_out;
+ }
+ if (!IS_ALIGNED(btrfs_super_log_root(sb), 4096)) {
+ error("log_root block unaligned: %llu",
+ btrfs_super_log_root(sb));
+ goto error_out;
+ }
+ if (btrfs_super_nodesize(sb) < 4096) {
+ error("nodesize too small: %u < 4096",
+ btrfs_super_nodesize(sb));
+ goto error_out;
+ }
+ if (!IS_ALIGNED(btrfs_super_nodesize(sb), 4096)) {
+ error("nodesize unaligned: %u", btrfs_super_nodesize(sb));
+ goto error_out;
+ }
+ if (btrfs_super_sectorsize(sb) < 4096) {
+ error("sectorsize too small: %u < 4096",
+ btrfs_super_sectorsize(sb));
+ goto error_out;
+ }
+ if (!IS_ALIGNED(btrfs_super_sectorsize(sb), 4096)) {
+ error("sectorsize unaligned: %u", btrfs_super_sectorsize(sb));
+ goto error_out;
+ }
+ if (btrfs_super_total_bytes(sb) == 0) {
+ error("invalid total_bytes 0");
+ goto error_out;
+ }
+ if (btrfs_super_bytes_used(sb) < 6 * btrfs_super_nodesize(sb)) {
+ error("invalid bytes_used %llu", btrfs_super_bytes_used(sb));
+ goto error_out;
+ }
+ if ((btrfs_super_stripesize(sb) != 4096)
+ && (btrfs_super_stripesize(sb) != btrfs_super_sectorsize(sb))) {
+ error("invalid stripesize %u", btrfs_super_stripesize(sb));
+ goto error_out;
+ }
+
+ if (memcmp(sb->fsid, sb->dev_item.fsid, BTRFS_UUID_SIZE) != 0) {
+ char fsid[BTRFS_UUID_UNPARSED_SIZE];
+ char dev_fsid[BTRFS_UUID_UNPARSED_SIZE];
+
+ uuid_unparse(sb->fsid, fsid);
+ uuid_unparse(sb->dev_item.fsid, dev_fsid);
+ error("dev_item UUID does not match fsid: %s != %s",
+ dev_fsid, fsid);
+ goto error_out;
+ }
+
+ /*
+ * Hint to catch really bogus numbers, bitflips or so
+ */
+ if (btrfs_super_num_devices(sb) > (1UL << 31)) {
+ warning("suspicious number of devices: %llu",
+ btrfs_super_num_devices(sb));
+ }
+
+ if (btrfs_super_num_devices(sb) == 0) {
+ error("number of devices is 0");
+ goto error_out;
+ }
+
+ /*
+ * Obvious sys_chunk_array corruptions, it must hold at least one key
+ * and one chunk
+ */
+ if (btrfs_super_sys_array_size(sb) > BTRFS_SYSTEM_CHUNK_ARRAY_SIZE) {
+ error("system chunk array too big %u > %u",
+ btrfs_super_sys_array_size(sb),
+ BTRFS_SYSTEM_CHUNK_ARRAY_SIZE);
+ goto error_out;
+ }
+ if (btrfs_super_sys_array_size(sb) < sizeof(struct btrfs_disk_key)
+ + sizeof(struct btrfs_chunk)) {
+ error("system chunk array too small %u < %lu",
+ btrfs_super_sys_array_size(sb),
+ sizeof(struct btrfs_disk_key) +
+ sizeof(struct btrfs_chunk));
+ goto error_out;
+ }
+
+ return 0;
+
+error_out:
+ error("superblock checksum matches but it has invalid members");
+ return -EIO;
+}
+
+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;
+ char tmp[BTRFS_SUPER_INFO_SIZE];
+ struct btrfs_super_block *buf = (struct btrfs_super_block *)tmp;
int i;
int ret;
+ int max_super = super_recover ? BTRFS_SUPER_MIRROR_MAX : 1;
u64 transid = 0;
u64 bytenr;
if (sb_bytenr != BTRFS_SUPER_INFO_OFFSET) {
- ret = pread64(fd, &buf, sizeof(buf), sb_bytenr);
- if (ret < sizeof(buf))
+ ret = pread64(fd, buf, BTRFS_SUPER_INFO_SIZE, sb_bytenr);
+ if (ret < BTRFS_SUPER_INFO_SIZE)
return -1;
- if (btrfs_super_bytenr(&buf) != sb_bytenr ||
- buf.magic != cpu_to_le64(BTRFS_MAGIC))
+ if (btrfs_super_bytenr(buf) != sb_bytenr)
return -1;
- memcpy(sb, &buf, sizeof(*sb));
+ if (check_super(buf))
+ return -1;
+ memcpy(sb, buf, BTRFS_SUPER_INFO_SIZE);
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))
+ ret = pread64(fd, buf, BTRFS_SUPER_INFO_SIZE, bytenr);
+ if (ret < BTRFS_SUPER_INFO_SIZE)
break;
- if (btrfs_super_bytenr(&buf) != bytenr )
+ if (btrfs_super_bytenr(buf) != bytenr )
continue;
/* if magic is NULL, the device was removed */
- if (buf.magic == 0 && i == 0)
- return -1;
- if (buf.magic != cpu_to_le64(BTRFS_MAGIC))
+ if (btrfs_super_magic(buf) == 0 && i == 0)
+ break;
+ if (check_super(buf))
continue;
if (!fsid_is_initialized) {
- memcpy(fsid, buf.fsid, sizeof(fsid));
+ memcpy(fsid, buf->fsid, sizeof(fsid));
fsid_is_initialized = 1;
- } else if (memcmp(fsid, buf.fsid, sizeof(fsid))) {
+ } else if (memcmp(fsid, buf->fsid, sizeof(fsid))) {
/*
* the superblocks (the original one and
* its backups) contain data of different
continue;
}
- if (btrfs_super_generation(&buf) > transid) {
- memcpy(sb, &buf, sizeof(*sb));
- transid = btrfs_super_generation(&buf);
+ if (btrfs_super_generation(buf) > transid) {
+ memcpy(sb, buf, BTRFS_SUPER_INFO_SIZE);
+ transid = btrfs_super_generation(buf);
}
}
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;
ret = pwrite64(device->fd, root->fs_info->super_copy,
BTRFS_SUPER_INFO_SIZE,
root->fs_info->super_bytenr);
- BUG_ON(ret != BTRFS_SUPER_INFO_SIZE);
+ if (ret != BTRFS_SUPER_INFO_SIZE)
+ goto write_err;
return 0;
}
*/
ret = pwrite64(device->fd, root->fs_info->super_copy,
BTRFS_SUPER_INFO_SIZE, bytenr);
- BUG_ON(ret != BTRFS_SUPER_INFO_SIZE);
+ if (ret != BTRFS_SUPER_INFO_SIZE)
+ goto write_err;
}
return 0;
+
+write_err:
+ if (ret > 0)
+ fprintf(stderr, "WARNING: failed to write all sb data\n");
+ else
+ fprintf(stderr, "WARNING: failed to write sb: %s\n",
+ strerror(errno));
+ return ret;
}
int write_all_supers(struct btrfs_root *root)
return ret;
}
-static int close_all_devices(struct btrfs_fs_info *fs_info)
-{
- struct list_head *list;
- struct btrfs_device *device;
-
- list = &fs_info->fs_devices->devices;
- while (!list_empty(list)) {
- device = list_entry(list->next, struct btrfs_device, dev_list);
- list_del_init(&device->dev_list);
- if (device->fd) {
- fsync(device->fd);
- if (posix_fadvise(device->fd, 0, 0, POSIX_FADV_DONTNEED))
- fprintf(stderr, "Warning, could not drop caches\n");
- }
- close(device->fd);
- kfree(device->name);
- kfree(device->label);
- kfree(device);
- }
- list_del(&fs_info->fs_devices->list);
- kfree(fs_info->fs_devices);
- return 0;
-}
-
-static void free_mapping_cache(struct btrfs_fs_info *fs_info)
-{
- struct cache_tree *cache_tree = &fs_info->mapping_tree.cache_tree;
- struct cache_extent *ce;
- struct map_lookup *map;
-
- while ((ce = find_first_cache_extent(cache_tree, 0))) {
- map = container_of(ce, struct map_lookup, ce);
- remove_cache_extent(cache_tree, ce);
- kfree(map);
- }
-}
-
-int close_ctree(struct btrfs_root *root)
+int close_ctree_fs_info(struct btrfs_fs_info *fs_info)
{
int ret;
struct btrfs_trans_handle *trans;
- struct btrfs_fs_info *fs_info = root->fs_info;
+ struct btrfs_root *root = fs_info->tree_root;
if (fs_info->last_trans_committed !=
fs_info->generation) {
+ BUG_ON(!root);
trans = btrfs_start_transaction(root, 1);
btrfs_commit_transaction(trans, root);
trans = btrfs_start_transaction(root, 1);
}
btrfs_free_block_groups(fs_info);
- free_fs_roots(fs_info);
-
- if (fs_info->extent_root->node)
- free_extent_buffer(fs_info->extent_root->node);
- if (fs_info->tree_root->node)
- free_extent_buffer(fs_info->tree_root->node);
- if (fs_info->chunk_root->node)
- free_extent_buffer(fs_info->chunk_root->node);
- if (fs_info->dev_root->node)
- free_extent_buffer(fs_info->dev_root->node);
- if (fs_info->csum_root->node)
- free_extent_buffer(fs_info->csum_root->node);
-
- if (fs_info->log_root_tree) {
- if (fs_info->log_root_tree->node)
- free_extent_buffer(fs_info->log_root_tree->node);
- free(fs_info->log_root_tree);
- }
-
- close_all_devices(fs_info);
- free_mapping_cache(fs_info);
- extent_io_tree_cleanup(&fs_info->extent_cache);
- extent_io_tree_cleanup(&fs_info->free_space_cache);
- extent_io_tree_cleanup(&fs_info->block_group_cache);
- extent_io_tree_cleanup(&fs_info->pinned_extents);
- extent_io_tree_cleanup(&fs_info->pending_del);
- extent_io_tree_cleanup(&fs_info->extent_ins);
-
- free(fs_info->tree_root);
- free(fs_info->extent_root);
- free(fs_info->chunk_root);
- free(fs_info->dev_root);
- free(fs_info->csum_root);
- free(fs_info);
+ free_fs_roots_tree(&fs_info->fs_root_tree);
+ btrfs_release_all_roots(fs_info);
+ btrfs_close_devices(fs_info->fs_devices);
+ btrfs_cleanup_all_caches(fs_info);
+ btrfs_free_fs_info(fs_info);
return 0;
}