1693 lines
41 KiB
C
1693 lines
41 KiB
C
/*
|
|
* Copyright (C) 2018 Facebook
|
|
*
|
|
* This file is part of libbtrfsutil.
|
|
*
|
|
* libbtrfsutil is free software: you can redistribute it and/or modify
|
|
* it under the terms of the GNU Lesser General Public License as published by
|
|
* the Free Software Foundation, either version 3 of the License, or
|
|
* (at your option) any later version.
|
|
*
|
|
* libbtrfsutil is distributed in the hope that it will be useful,
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
* GNU Lesser General Public License for more details.
|
|
*
|
|
* You should have received a copy of the GNU Lesser General Public License
|
|
* along with libbtrfsutil. If not, see <http://www.gnu.org/licenses/>.
|
|
*/
|
|
|
|
#include <errno.h>
|
|
#include <fcntl.h>
|
|
#include <stdlib.h>
|
|
#include <string.h>
|
|
#include <unistd.h>
|
|
#include <sys/ioctl.h>
|
|
#include <sys/stat.h>
|
|
#include <sys/types.h>
|
|
#include <sys/vfs.h>
|
|
#include <linux/magic.h>
|
|
#include "stubs.h"
|
|
|
|
#include "btrfsutil_internal.h"
|
|
|
|
static bool is_root(void)
|
|
{
|
|
return geteuid() == 0;
|
|
}
|
|
|
|
/*
|
|
* This intentionally duplicates btrfs_util_is_subvolume_fd() instead of opening
|
|
* a file descriptor and calling it, because fstat() and fstatfs() don't accept
|
|
* file descriptors opened with O_PATH on old kernels (before v3.6 and before
|
|
* v3.12, respectively), but stat() and statfs() can be called on a path that
|
|
* the user doesn't have read or write permissions to.
|
|
*/
|
|
PUBLIC enum btrfs_util_error btrfs_util_is_subvolume(const char *path)
|
|
{
|
|
struct statfs sfs;
|
|
struct stat st;
|
|
int ret;
|
|
|
|
ret = statfs(path, &sfs);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_STATFS_FAILED;
|
|
|
|
if (sfs.f_type != BTRFS_SUPER_MAGIC) {
|
|
errno = EINVAL;
|
|
return BTRFS_UTIL_ERROR_NOT_BTRFS;
|
|
}
|
|
|
|
ret = stat(path, &st);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_STAT_FAILED;
|
|
|
|
if (st.st_ino != BTRFS_FIRST_FREE_OBJECTID || !S_ISDIR(st.st_mode)) {
|
|
errno = EINVAL;
|
|
return BTRFS_UTIL_ERROR_NOT_SUBVOLUME;
|
|
}
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_is_subvolume_fd(int fd)
|
|
{
|
|
struct statfs sfs;
|
|
struct stat st;
|
|
int ret;
|
|
|
|
ret = fstatfs(fd, &sfs);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_STATFS_FAILED;
|
|
|
|
if (sfs.f_type != BTRFS_SUPER_MAGIC) {
|
|
errno = EINVAL;
|
|
return BTRFS_UTIL_ERROR_NOT_BTRFS;
|
|
}
|
|
|
|
ret = fstat(fd, &st);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_STAT_FAILED;
|
|
|
|
if (st.st_ino != BTRFS_FIRST_FREE_OBJECTID || !S_ISDIR(st.st_mode)) {
|
|
errno = EINVAL;
|
|
return BTRFS_UTIL_ERROR_NOT_SUBVOLUME;
|
|
}
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_subvolume_id(const char *path,
|
|
uint64_t *id_ret)
|
|
{
|
|
enum btrfs_util_error err;
|
|
int fd;
|
|
|
|
fd = open(path, O_RDONLY);
|
|
if (fd == -1)
|
|
return BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
|
|
err = btrfs_util_subvolume_id_fd(fd, id_ret);
|
|
SAVE_ERRNO_AND_CLOSE(fd);
|
|
return err;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_subvolume_id_fd(int fd,
|
|
uint64_t *id_ret)
|
|
{
|
|
struct btrfs_ioctl_ino_lookup_args args = {
|
|
.treeid = 0,
|
|
.objectid = BTRFS_FIRST_FREE_OBJECTID,
|
|
};
|
|
int ret;
|
|
|
|
ret = ioctl(fd, BTRFS_IOC_INO_LOOKUP, &args);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_INO_LOOKUP_FAILED;
|
|
|
|
*id_ret = args.treeid;
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_subvolume_path(const char *path,
|
|
uint64_t id,
|
|
char **path_ret)
|
|
{
|
|
enum btrfs_util_error err;
|
|
int fd;
|
|
|
|
fd = open(path, O_RDONLY);
|
|
if (fd == -1)
|
|
return BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
|
|
err = btrfs_util_subvolume_path_fd(fd, id, path_ret);
|
|
SAVE_ERRNO_AND_CLOSE(fd);
|
|
return err;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_subvolume_path_fd(int fd, uint64_t id,
|
|
char **path_ret)
|
|
{
|
|
char *path, *p;
|
|
size_t capacity = 4096;
|
|
|
|
if (id == 0) {
|
|
enum btrfs_util_error err;
|
|
|
|
err = btrfs_util_is_subvolume_fd(fd);
|
|
if (err)
|
|
return err;
|
|
|
|
err = btrfs_util_subvolume_id_fd(fd, &id);
|
|
if (err)
|
|
return err;
|
|
}
|
|
|
|
path = malloc(capacity);
|
|
if (!path)
|
|
return BTRFS_UTIL_ERROR_NO_MEMORY;
|
|
p = path + capacity - 1;
|
|
p[0] = '\0';
|
|
|
|
while (id != BTRFS_FS_TREE_OBJECTID) {
|
|
struct btrfs_ioctl_search_args search = {
|
|
.key = {
|
|
.tree_id = BTRFS_ROOT_TREE_OBJECTID,
|
|
.min_objectid = id,
|
|
.max_objectid = id,
|
|
.min_type = BTRFS_ROOT_BACKREF_KEY,
|
|
.max_type = BTRFS_ROOT_BACKREF_KEY,
|
|
.min_offset = 0,
|
|
.max_offset = UINT64_MAX,
|
|
.min_transid = 0,
|
|
.max_transid = UINT64_MAX,
|
|
.nr_items = 1,
|
|
},
|
|
};
|
|
struct btrfs_ioctl_ino_lookup_args lookup;
|
|
const struct btrfs_ioctl_search_header *header;
|
|
const struct btrfs_root_ref *ref;
|
|
const char *name;
|
|
uint16_t name_len;
|
|
size_t lookup_len;
|
|
size_t total_len;
|
|
int ret;
|
|
|
|
ret = ioctl(fd, BTRFS_IOC_TREE_SEARCH, &search);
|
|
if (ret == -1) {
|
|
free(path);
|
|
return BTRFS_UTIL_ERROR_SEARCH_FAILED;
|
|
}
|
|
|
|
if (search.key.nr_items == 0) {
|
|
free(path);
|
|
errno = ENOENT;
|
|
return BTRFS_UTIL_ERROR_SUBVOLUME_NOT_FOUND;
|
|
}
|
|
|
|
header = (struct btrfs_ioctl_search_header *)search.buf;
|
|
ref = (struct btrfs_root_ref *)(header + 1);
|
|
name = (char *)(ref + 1);
|
|
name_len = le16_to_cpu(ref->name_len);
|
|
|
|
id = header->offset;
|
|
|
|
lookup.treeid = id;
|
|
lookup.objectid = le64_to_cpu(ref->dirid);
|
|
ret = ioctl(fd, BTRFS_IOC_INO_LOOKUP, &lookup);
|
|
if (ret == -1) {
|
|
free(path);
|
|
return BTRFS_UTIL_ERROR_SEARCH_FAILED;
|
|
}
|
|
lookup_len = strlen(lookup.name);
|
|
|
|
total_len = name_len + lookup_len + (id != BTRFS_FS_TREE_OBJECTID);
|
|
if (p - total_len < path) {
|
|
char *new_path, *new_p;
|
|
size_t new_capacity = capacity * 2;
|
|
|
|
new_path = malloc(new_capacity);
|
|
if (!new_path) {
|
|
free(path);
|
|
return BTRFS_UTIL_ERROR_NO_MEMORY;
|
|
}
|
|
new_p = new_path + new_capacity - (path + capacity - p);
|
|
memcpy(new_p, p, path + capacity - p);
|
|
free(path);
|
|
path = new_path;
|
|
p = new_p;
|
|
capacity = new_capacity;
|
|
}
|
|
p -= name_len;
|
|
memcpy(p, name, name_len);
|
|
p -= lookup_len;
|
|
memcpy(p, lookup.name, lookup_len);
|
|
if (id != BTRFS_FS_TREE_OBJECTID)
|
|
*--p = '/';
|
|
}
|
|
|
|
if (p != path)
|
|
memmove(path, p, path + capacity - p);
|
|
|
|
*path_ret = path;
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
static void copy_timespec(struct timespec *timespec,
|
|
const struct btrfs_timespec *btrfs_timespec)
|
|
{
|
|
timespec->tv_sec = le64_to_cpu(btrfs_timespec->sec);
|
|
timespec->tv_nsec = le32_to_cpu(btrfs_timespec->nsec);
|
|
}
|
|
|
|
static void copy_root_item(struct btrfs_util_subvolume_info *subvol,
|
|
const struct btrfs_root_item *root)
|
|
{
|
|
subvol->flags = le64_to_cpu(root->flags);
|
|
memcpy(subvol->uuid, root->uuid, sizeof(subvol->uuid));
|
|
memcpy(subvol->parent_uuid, root->parent_uuid,
|
|
sizeof(subvol->parent_uuid));
|
|
memcpy(subvol->received_uuid, root->received_uuid,
|
|
sizeof(subvol->received_uuid));
|
|
subvol->generation = le64_to_cpu(root->generation);
|
|
subvol->ctransid = le64_to_cpu(root->ctransid);
|
|
subvol->otransid = le64_to_cpu(root->otransid);
|
|
subvol->stransid = le64_to_cpu(root->stransid);
|
|
subvol->rtransid = le64_to_cpu(root->rtransid);
|
|
copy_timespec(&subvol->ctime, &root->ctime);
|
|
copy_timespec(&subvol->otime, &root->otime);
|
|
copy_timespec(&subvol->stime, &root->stime);
|
|
copy_timespec(&subvol->rtime, &root->rtime);
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_subvolume_info(const char *path,
|
|
uint64_t id,
|
|
struct btrfs_util_subvolume_info *subvol)
|
|
{
|
|
enum btrfs_util_error err;
|
|
int fd;
|
|
|
|
fd = open(path, O_RDONLY);
|
|
if (fd == -1)
|
|
return BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
|
|
err = btrfs_util_subvolume_info_fd(fd, id, subvol);
|
|
SAVE_ERRNO_AND_CLOSE(fd);
|
|
return err;
|
|
}
|
|
|
|
static enum btrfs_util_error get_subvolume_info_privileged(int fd, uint64_t id,
|
|
struct btrfs_util_subvolume_info *subvol)
|
|
{
|
|
struct btrfs_ioctl_search_args search = {
|
|
.key = {
|
|
.tree_id = BTRFS_ROOT_TREE_OBJECTID,
|
|
.min_type = BTRFS_ROOT_ITEM_KEY,
|
|
.max_type = BTRFS_ROOT_BACKREF_KEY,
|
|
.min_offset = 0,
|
|
.max_offset = UINT64_MAX,
|
|
.min_transid = 0,
|
|
.max_transid = UINT64_MAX,
|
|
.nr_items = 0,
|
|
},
|
|
};
|
|
size_t items_pos = 0, buf_off = 0;
|
|
bool need_root_item = true, need_root_backref = true;
|
|
int ret;
|
|
|
|
search.key.min_objectid = search.key.max_objectid = id;
|
|
|
|
if (subvol) {
|
|
subvol->id = id;
|
|
subvol->parent_id = 0;
|
|
subvol->dir_id = 0;
|
|
if (id == BTRFS_FS_TREE_OBJECTID)
|
|
need_root_backref = false;
|
|
} else {
|
|
/*
|
|
* We only need the backref for filling in the subvolume info.
|
|
*/
|
|
need_root_backref = false;
|
|
}
|
|
|
|
/* Don't bother searching for the backref if we don't need it. */
|
|
if (!need_root_backref)
|
|
search.key.max_type = BTRFS_ROOT_ITEM_KEY;
|
|
|
|
while (need_root_item || need_root_backref) {
|
|
const struct btrfs_ioctl_search_header *header;
|
|
|
|
if (items_pos >= search.key.nr_items) {
|
|
search.key.nr_items = 4096;
|
|
ret = ioctl(fd, BTRFS_IOC_TREE_SEARCH, &search);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_SEARCH_FAILED;
|
|
items_pos = 0;
|
|
buf_off = 0;
|
|
|
|
if (search.key.nr_items == 0) {
|
|
if (need_root_item) {
|
|
errno = ENOENT;
|
|
return BTRFS_UTIL_ERROR_SUBVOLUME_NOT_FOUND;
|
|
} else {
|
|
break;
|
|
}
|
|
}
|
|
}
|
|
|
|
header = (struct btrfs_ioctl_search_header *)(search.buf + buf_off);
|
|
if (header->type == BTRFS_ROOT_ITEM_KEY) {
|
|
if (subvol) {
|
|
const struct btrfs_root_item *root;
|
|
|
|
root = (const struct btrfs_root_item *)(header + 1);
|
|
copy_root_item(subvol, root);
|
|
}
|
|
need_root_item = false;
|
|
search.key.min_type = BTRFS_ROOT_BACKREF_KEY;
|
|
} else if (header->type == BTRFS_ROOT_BACKREF_KEY) {
|
|
if (subvol) {
|
|
const struct btrfs_root_ref *ref;
|
|
|
|
ref = (const struct btrfs_root_ref *)(header + 1);
|
|
subvol->parent_id = header->offset;
|
|
subvol->dir_id = le64_to_cpu(ref->dirid);
|
|
}
|
|
need_root_backref = false;
|
|
search.key.min_type = UINT32_MAX;
|
|
}
|
|
|
|
items_pos++;
|
|
buf_off += sizeof(*header) + header->len;
|
|
}
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
static enum btrfs_util_error get_subvolume_info_unprivileged(int fd,
|
|
struct btrfs_util_subvolume_info *subvol)
|
|
{
|
|
struct btrfs_ioctl_get_subvol_info_args info;
|
|
int ret;
|
|
|
|
ret = ioctl(fd, BTRFS_IOC_GET_SUBVOL_INFO, &info);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_GET_SUBVOL_INFO_FAILED;
|
|
|
|
subvol->id = info.treeid;
|
|
subvol->parent_id = info.parent_id;
|
|
subvol->dir_id = info.dirid;
|
|
subvol->flags = info.flags;
|
|
subvol->generation = info.generation;
|
|
|
|
memcpy(subvol->uuid, info.uuid, sizeof(subvol->uuid));
|
|
memcpy(subvol->parent_uuid, info.parent_uuid,
|
|
sizeof(subvol->parent_uuid));
|
|
memcpy(subvol->received_uuid, info.received_uuid,
|
|
sizeof(subvol->received_uuid));
|
|
|
|
subvol->ctransid = info.ctransid;
|
|
subvol->otransid = info.otransid;
|
|
subvol->stransid = info.stransid;
|
|
subvol->rtransid = info.rtransid;
|
|
|
|
subvol->ctime.tv_sec = info.ctime.sec;
|
|
subvol->ctime.tv_nsec = info.ctime.nsec;
|
|
subvol->otime.tv_sec = info.otime.sec;
|
|
subvol->otime.tv_nsec = info.otime.nsec;
|
|
subvol->stime.tv_sec = info.stime.sec;
|
|
subvol->stime.tv_nsec = info.stime.nsec;
|
|
subvol->rtime.tv_sec = info.rtime.sec;
|
|
subvol->rtime.tv_nsec = info.rtime.nsec;
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_subvolume_info_fd(int fd, uint64_t id,
|
|
struct btrfs_util_subvolume_info *subvol)
|
|
{
|
|
enum btrfs_util_error err;
|
|
|
|
if (id == 0) {
|
|
err = btrfs_util_is_subvolume_fd(fd);
|
|
if (err)
|
|
return err;
|
|
|
|
if (!is_root())
|
|
return get_subvolume_info_unprivileged(fd, subvol);
|
|
|
|
err = btrfs_util_subvolume_id_fd(fd, &id);
|
|
if (err)
|
|
return err;
|
|
}
|
|
|
|
if ((id < BTRFS_FIRST_FREE_OBJECTID && id != BTRFS_FS_TREE_OBJECTID) ||
|
|
id > BTRFS_LAST_FREE_OBJECTID) {
|
|
errno = ENOENT;
|
|
return BTRFS_UTIL_ERROR_SUBVOLUME_NOT_FOUND;
|
|
}
|
|
|
|
return get_subvolume_info_privileged(fd, id, subvol);
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_get_subvolume_read_only_fd(int fd,
|
|
bool *read_only_ret)
|
|
{
|
|
uint64_t flags;
|
|
int ret;
|
|
|
|
ret = ioctl(fd, BTRFS_IOC_SUBVOL_GETFLAGS, &flags);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_SUBVOL_GETFLAGS_FAILED;
|
|
|
|
*read_only_ret = flags & BTRFS_SUBVOL_RDONLY;
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_get_subvolume_read_only(const char *path,
|
|
bool *ret)
|
|
{
|
|
enum btrfs_util_error err;
|
|
int fd;
|
|
|
|
fd = open(path, O_RDONLY);
|
|
if (fd == -1)
|
|
return BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
|
|
err = btrfs_util_get_subvolume_read_only_fd(fd, ret);
|
|
SAVE_ERRNO_AND_CLOSE(fd);
|
|
return err;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_set_subvolume_read_only(const char *path,
|
|
bool read_only)
|
|
{
|
|
enum btrfs_util_error err;
|
|
int fd;
|
|
|
|
fd = open(path, O_RDONLY);
|
|
if (fd == -1)
|
|
return BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
|
|
err = btrfs_util_set_subvolume_read_only_fd(fd, read_only);
|
|
SAVE_ERRNO_AND_CLOSE(fd);
|
|
return err;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_set_subvolume_read_only_fd(int fd,
|
|
bool read_only)
|
|
{
|
|
uint64_t flags;
|
|
int ret;
|
|
|
|
ret = ioctl(fd, BTRFS_IOC_SUBVOL_GETFLAGS, &flags);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_SUBVOL_GETFLAGS_FAILED;
|
|
|
|
if (read_only)
|
|
flags |= BTRFS_SUBVOL_RDONLY;
|
|
else
|
|
flags &= ~BTRFS_SUBVOL_RDONLY;
|
|
|
|
ret = ioctl(fd, BTRFS_IOC_SUBVOL_SETFLAGS, &flags);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_SUBVOL_SETFLAGS_FAILED;
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_get_default_subvolume(const char *path,
|
|
uint64_t *id_ret)
|
|
{
|
|
enum btrfs_util_error err;
|
|
int fd;
|
|
|
|
fd = open(path, O_RDONLY);
|
|
if (fd == -1)
|
|
return BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
|
|
err = btrfs_util_get_default_subvolume_fd(fd, id_ret);
|
|
SAVE_ERRNO_AND_CLOSE(fd);
|
|
return err;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_get_default_subvolume_fd(int fd,
|
|
uint64_t *id_ret)
|
|
{
|
|
struct btrfs_ioctl_search_args search = {
|
|
.key = {
|
|
.tree_id = BTRFS_ROOT_TREE_OBJECTID,
|
|
.min_objectid = BTRFS_ROOT_TREE_DIR_OBJECTID,
|
|
.max_objectid = BTRFS_ROOT_TREE_DIR_OBJECTID,
|
|
.min_type = BTRFS_DIR_ITEM_KEY,
|
|
.max_type = BTRFS_DIR_ITEM_KEY,
|
|
.min_offset = 0,
|
|
.max_offset = UINT64_MAX,
|
|
.min_transid = 0,
|
|
.max_transid = UINT64_MAX,
|
|
.nr_items = 0,
|
|
},
|
|
};
|
|
size_t items_pos = 0, buf_off = 0;
|
|
int ret;
|
|
|
|
for (;;) {
|
|
const struct btrfs_ioctl_search_header *header;
|
|
|
|
if (items_pos >= search.key.nr_items) {
|
|
search.key.nr_items = 4096;
|
|
ret = ioctl(fd, BTRFS_IOC_TREE_SEARCH, &search);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_SEARCH_FAILED;
|
|
items_pos = 0;
|
|
buf_off = 0;
|
|
|
|
if (search.key.nr_items == 0) {
|
|
errno = ENOENT;
|
|
return BTRFS_UTIL_ERROR_SUBVOLUME_NOT_FOUND;
|
|
}
|
|
}
|
|
|
|
header = (struct btrfs_ioctl_search_header *)(search.buf + buf_off);
|
|
if (header->type == BTRFS_DIR_ITEM_KEY) {
|
|
const struct btrfs_dir_item *dir;
|
|
const char *name;
|
|
uint16_t name_len;
|
|
|
|
dir = (struct btrfs_dir_item *)(header + 1);
|
|
name = (const char *)(dir + 1);
|
|
name_len = le16_to_cpu(dir->name_len);
|
|
if (strncmp(name, "default", name_len) == 0) {
|
|
*id_ret = le64_to_cpu(dir->location.objectid);
|
|
break;
|
|
}
|
|
}
|
|
|
|
items_pos++;
|
|
buf_off += sizeof(*header) + header->len;
|
|
search.key.min_offset = header->offset + 1;
|
|
}
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_set_default_subvolume(const char *path,
|
|
uint64_t id)
|
|
{
|
|
enum btrfs_util_error err;
|
|
int fd;
|
|
|
|
fd = open(path, O_RDONLY);
|
|
if (fd == -1)
|
|
return BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
|
|
err = btrfs_util_set_default_subvolume_fd(fd, id);
|
|
SAVE_ERRNO_AND_CLOSE(fd);
|
|
return err;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_set_default_subvolume_fd(int fd,
|
|
uint64_t id)
|
|
{
|
|
enum btrfs_util_error err;
|
|
int ret;
|
|
|
|
if (id == 0) {
|
|
err = btrfs_util_is_subvolume_fd(fd);
|
|
if (err)
|
|
return err;
|
|
|
|
err = btrfs_util_subvolume_id_fd(fd, &id);
|
|
if (err)
|
|
return err;
|
|
}
|
|
|
|
ret = ioctl(fd, BTRFS_IOC_DEFAULT_SUBVOL, &id);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_DEFAULT_SUBVOL_FAILED;
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
static enum btrfs_util_error openat_parent_and_name(int dirfd, const char *path,
|
|
char *name, size_t name_len,
|
|
int *fd)
|
|
{
|
|
char *tmp_path, *slash, *dirname, *basename;
|
|
size_t len;
|
|
|
|
/* Ignore trailing slashes. */
|
|
len = strlen(path);
|
|
while (len > 1 && path[len - 1] == '/')
|
|
len--;
|
|
|
|
tmp_path = malloc(len + 1);
|
|
if (!tmp_path)
|
|
return BTRFS_UTIL_ERROR_NO_MEMORY;
|
|
memcpy(tmp_path, path, len);
|
|
tmp_path[len] = '\0';
|
|
|
|
slash = memrchr(tmp_path, '/', len);
|
|
if (slash == tmp_path) {
|
|
dirname = "/";
|
|
basename = tmp_path + 1;
|
|
} else if (slash) {
|
|
*slash = '\0';
|
|
dirname = tmp_path;
|
|
basename = slash + 1;
|
|
} else {
|
|
dirname = ".";
|
|
basename = tmp_path;
|
|
}
|
|
|
|
len = strlen(basename);
|
|
if (len >= name_len) {
|
|
free(tmp_path);
|
|
errno = ENAMETOOLONG;
|
|
return BTRFS_UTIL_ERROR_INVALID_ARGUMENT;
|
|
}
|
|
memcpy(name, basename, len);
|
|
name[len] = '\0';
|
|
|
|
*fd = openat(dirfd, dirname, O_RDONLY | O_DIRECTORY);
|
|
if (*fd == -1) {
|
|
free(tmp_path);
|
|
return BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
}
|
|
|
|
free(tmp_path);
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_create_subvolume(const char *path,
|
|
int flags,
|
|
uint64_t *async_transid,
|
|
struct btrfs_util_qgroup_inherit *qgroup_inherit)
|
|
{
|
|
char name[BTRFS_SUBVOL_NAME_MAX + 1];
|
|
enum btrfs_util_error err;
|
|
int parent_fd;
|
|
|
|
err = openat_parent_and_name(AT_FDCWD, path, name, sizeof(name),
|
|
&parent_fd);
|
|
if (err)
|
|
return err;
|
|
|
|
err = btrfs_util_create_subvolume_fd(parent_fd, name, flags,
|
|
async_transid, qgroup_inherit);
|
|
SAVE_ERRNO_AND_CLOSE(parent_fd);
|
|
return err;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_create_subvolume_fd(int parent_fd,
|
|
const char *name,
|
|
int flags,
|
|
uint64_t *async_transid,
|
|
struct btrfs_util_qgroup_inherit *qgroup_inherit)
|
|
{
|
|
struct btrfs_ioctl_vol_args_v2 args = {};
|
|
size_t len;
|
|
int ret;
|
|
|
|
if (flags) {
|
|
errno = EINVAL;
|
|
return BTRFS_UTIL_ERROR_INVALID_ARGUMENT;
|
|
}
|
|
|
|
if (async_transid)
|
|
args.flags |= BTRFS_SUBVOL_CREATE_ASYNC;
|
|
if (qgroup_inherit) {
|
|
args.flags |= BTRFS_SUBVOL_QGROUP_INHERIT;
|
|
args.qgroup_inherit = (struct btrfs_qgroup_inherit *)qgroup_inherit;
|
|
args.size = (sizeof(*args.qgroup_inherit) +
|
|
args.qgroup_inherit->num_qgroups *
|
|
sizeof(args.qgroup_inherit->qgroups[0]));
|
|
}
|
|
|
|
len = strlen(name);
|
|
if (len >= sizeof(args.name)) {
|
|
errno = ENAMETOOLONG;
|
|
return BTRFS_UTIL_ERROR_INVALID_ARGUMENT;
|
|
}
|
|
memcpy(args.name, name, len);
|
|
args.name[len] = '\0';
|
|
|
|
ret = ioctl(parent_fd, BTRFS_IOC_SUBVOL_CREATE_V2, &args);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_SUBVOL_CREATE_FAILED;
|
|
|
|
if (async_transid)
|
|
*async_transid = args.transid;
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
#define BTRFS_UTIL_SUBVOLUME_ITERATOR_CLOSE_FD (1 << 30)
|
|
|
|
struct search_stack_entry {
|
|
union {
|
|
/* Used for subvolume_iterator_next_tree_search(). */
|
|
struct {
|
|
struct btrfs_ioctl_search_args search;
|
|
size_t buf_off;
|
|
};
|
|
/* Used for subvolume_iterator_next_unprivileged(). */
|
|
struct {
|
|
uint64_t id;
|
|
struct btrfs_ioctl_get_subvol_rootref_args rootref_args;
|
|
};
|
|
};
|
|
/* Used for both. */
|
|
size_t items_pos;
|
|
size_t path_len;
|
|
};
|
|
|
|
struct btrfs_util_subvolume_iterator {
|
|
bool use_tree_search;
|
|
int fd;
|
|
/* cur_fd is only used for subvolume_iterator_next_unprivileged(). */
|
|
int cur_fd;
|
|
int flags;
|
|
|
|
struct search_stack_entry *search_stack;
|
|
size_t search_stack_len;
|
|
size_t search_stack_capacity;
|
|
|
|
char *cur_path;
|
|
size_t cur_path_capacity;
|
|
};
|
|
|
|
static struct search_stack_entry *top_search_stack_entry(struct btrfs_util_subvolume_iterator *iter)
|
|
{
|
|
return &iter->search_stack[iter->search_stack_len - 1];
|
|
}
|
|
|
|
/*
|
|
* Check that a path that we opened is the subvolume which we expect. It may not
|
|
* be if there is another filesystem mounted over a parent directory or the
|
|
* subvolume itself.
|
|
*/
|
|
static enum btrfs_util_error check_expected_subvolume(int fd, int parent_fd,
|
|
uint64_t tree_id)
|
|
{
|
|
struct btrfs_ioctl_fs_info_args parent_fs_info, fs_info;
|
|
enum btrfs_util_error err;
|
|
uint64_t id;
|
|
int ret;
|
|
|
|
/* Make sure it's a subvolume. */
|
|
err = btrfs_util_is_subvolume_fd(fd);
|
|
if (err == BTRFS_UTIL_ERROR_NOT_BTRFS ||
|
|
err == BTRFS_UTIL_ERROR_NOT_SUBVOLUME) {
|
|
errno = ENOENT;
|
|
return BTRFS_UTIL_ERROR_SUBVOLUME_NOT_FOUND;
|
|
} else if (err) {
|
|
return err;
|
|
}
|
|
|
|
/* Make sure it's on the same filesystem. */
|
|
ret = ioctl(parent_fd, BTRFS_IOC_FS_INFO, &parent_fs_info);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_FS_INFO_FAILED;
|
|
ret = ioctl(fd, BTRFS_IOC_FS_INFO, &fs_info);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_FS_INFO_FAILED;
|
|
if (memcmp(parent_fs_info.fsid, fs_info.fsid, sizeof(fs_info.fsid)) != 0) {
|
|
errno = ENOENT;
|
|
return BTRFS_UTIL_ERROR_SUBVOLUME_NOT_FOUND;
|
|
}
|
|
|
|
/* Make sure it's the subvolume that we expected. */
|
|
err = btrfs_util_subvolume_id_fd(fd, &id);
|
|
if (err)
|
|
return err;
|
|
if (id != tree_id) {
|
|
errno = ENOENT;
|
|
return BTRFS_UTIL_ERROR_SUBVOLUME_NOT_FOUND;
|
|
}
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
static enum btrfs_util_error append_to_search_stack(struct btrfs_util_subvolume_iterator *iter,
|
|
uint64_t tree_id,
|
|
size_t path_len)
|
|
{
|
|
struct search_stack_entry *entry;
|
|
|
|
if (iter->search_stack_len >= iter->search_stack_capacity) {
|
|
size_t new_capacity = iter->search_stack_capacity * 2;
|
|
struct search_stack_entry *new_search_stack;
|
|
|
|
new_search_stack = reallocarray(iter->search_stack,
|
|
new_capacity,
|
|
sizeof(*iter->search_stack));
|
|
if (!new_search_stack)
|
|
return BTRFS_UTIL_ERROR_NO_MEMORY;
|
|
|
|
iter->search_stack_capacity = new_capacity;
|
|
iter->search_stack = new_search_stack;
|
|
}
|
|
|
|
entry = &iter->search_stack[iter->search_stack_len];
|
|
|
|
memset(entry, 0, sizeof(*entry));
|
|
entry->path_len = path_len;
|
|
if (iter->use_tree_search) {
|
|
entry->search.key.tree_id = BTRFS_ROOT_TREE_OBJECTID;
|
|
entry->search.key.min_objectid = tree_id;
|
|
entry->search.key.max_objectid = tree_id;
|
|
entry->search.key.min_type = BTRFS_ROOT_REF_KEY;
|
|
entry->search.key.max_type = BTRFS_ROOT_REF_KEY;
|
|
entry->search.key.min_offset = 0;
|
|
entry->search.key.max_offset = UINT64_MAX;
|
|
entry->search.key.min_transid = 0;
|
|
entry->search.key.max_transid = UINT64_MAX;
|
|
entry->search.key.nr_items = 0;
|
|
} else {
|
|
entry->id = tree_id;
|
|
|
|
if (iter->search_stack_len) {
|
|
struct search_stack_entry *top;
|
|
enum btrfs_util_error err;
|
|
char *path;
|
|
int fd;
|
|
|
|
top = top_search_stack_entry(iter);
|
|
path = &iter->cur_path[top->path_len];
|
|
if (*path == '/')
|
|
path++;
|
|
fd = openat(iter->cur_fd, path, O_RDONLY);
|
|
if (fd == -1)
|
|
return BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
|
|
err = check_expected_subvolume(fd, iter->cur_fd,
|
|
tree_id);
|
|
if (err) {
|
|
close(fd);
|
|
return err;
|
|
}
|
|
|
|
close(iter->cur_fd);
|
|
iter->cur_fd = fd;
|
|
}
|
|
}
|
|
|
|
iter->search_stack_len++;
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
static enum btrfs_util_error pop_search_stack(struct btrfs_util_subvolume_iterator *iter)
|
|
{
|
|
struct search_stack_entry *top, *parent;
|
|
int fd, parent_fd;
|
|
size_t i;
|
|
|
|
if (iter->use_tree_search || iter->search_stack_len == 1) {
|
|
iter->search_stack_len--;
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
top = top_search_stack_entry(iter);
|
|
iter->search_stack_len--;
|
|
parent = top_search_stack_entry(iter);
|
|
|
|
fd = iter->cur_fd;
|
|
for (i = parent->path_len; i < top->path_len; i++) {
|
|
if (i == 0 || iter->cur_path[i] == '/') {
|
|
parent_fd = openat(fd, "..", O_RDONLY);
|
|
if (fd != iter->cur_fd)
|
|
SAVE_ERRNO_AND_CLOSE(fd);
|
|
if (parent_fd == -1)
|
|
return BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
fd = parent_fd;
|
|
}
|
|
}
|
|
if (iter->cur_fd != iter->fd)
|
|
close(iter->cur_fd);
|
|
iter->cur_fd = fd;
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_create_subvolume_iterator(const char *path,
|
|
uint64_t top,
|
|
int flags,
|
|
struct btrfs_util_subvolume_iterator **ret)
|
|
{
|
|
enum btrfs_util_error err;
|
|
int fd;
|
|
|
|
fd = open(path, O_RDONLY);
|
|
if (fd == -1)
|
|
return BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
|
|
err = btrfs_util_create_subvolume_iterator_fd(fd, top, flags, ret);
|
|
if (err)
|
|
SAVE_ERRNO_AND_CLOSE(fd);
|
|
else
|
|
(*ret)->flags |= BTRFS_UTIL_SUBVOLUME_ITERATOR_CLOSE_FD;
|
|
|
|
return err;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_create_subvolume_iterator_fd(int fd,
|
|
uint64_t top,
|
|
int flags,
|
|
struct btrfs_util_subvolume_iterator **ret)
|
|
{
|
|
struct btrfs_util_subvolume_iterator *iter;
|
|
enum btrfs_util_error err;
|
|
bool use_tree_search;
|
|
|
|
if (flags & ~BTRFS_UTIL_SUBVOLUME_ITERATOR_MASK) {
|
|
errno = EINVAL;
|
|
return BTRFS_UTIL_ERROR_INVALID_ARGUMENT;
|
|
}
|
|
|
|
use_tree_search = top != 0 || is_root();
|
|
if (top == 0) {
|
|
err = btrfs_util_is_subvolume_fd(fd);
|
|
if (err)
|
|
return err;
|
|
|
|
err = btrfs_util_subvolume_id_fd(fd, &top);
|
|
if (err)
|
|
return err;
|
|
}
|
|
|
|
iter = malloc(sizeof(*iter));
|
|
if (!iter)
|
|
return BTRFS_UTIL_ERROR_NO_MEMORY;
|
|
|
|
iter->fd = fd;
|
|
iter->cur_fd = fd;
|
|
iter->flags = flags;
|
|
iter->use_tree_search = use_tree_search;
|
|
|
|
iter->search_stack_len = 0;
|
|
iter->search_stack_capacity = 4;
|
|
iter->search_stack = malloc(sizeof(*iter->search_stack) *
|
|
iter->search_stack_capacity);
|
|
if (!iter->search_stack) {
|
|
err = BTRFS_UTIL_ERROR_NO_MEMORY;
|
|
goto out_iter;
|
|
}
|
|
|
|
iter->cur_path_capacity = 256;
|
|
iter->cur_path = malloc(iter->cur_path_capacity);
|
|
if (!iter->cur_path) {
|
|
err = BTRFS_UTIL_ERROR_NO_MEMORY;
|
|
goto out_search_stack;
|
|
}
|
|
|
|
err = append_to_search_stack(iter, top, 0);
|
|
if (err)
|
|
goto out_cur_path;
|
|
|
|
*ret = iter;
|
|
|
|
return BTRFS_UTIL_OK;
|
|
|
|
out_cur_path:
|
|
free(iter->cur_path);
|
|
out_search_stack:
|
|
free(iter->search_stack);
|
|
out_iter:
|
|
free(iter);
|
|
return err;
|
|
}
|
|
|
|
static enum btrfs_util_error snapshot_subvolume_children(int fd, int parent_fd,
|
|
const char *name,
|
|
uint64_t *async_transid)
|
|
{
|
|
struct btrfs_util_subvolume_iterator *iter;
|
|
enum btrfs_util_error err;
|
|
int dstfd;
|
|
|
|
dstfd = openat(parent_fd, name, O_RDONLY);
|
|
if (dstfd == -1)
|
|
return BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
|
|
err = btrfs_util_create_subvolume_iterator_fd(fd, 0, 0, &iter);
|
|
if (err)
|
|
goto out;
|
|
|
|
for (;;) {
|
|
char child_name[BTRFS_SUBVOL_NAME_MAX + 1];
|
|
char *child_path;
|
|
int child_fd, new_parent_fd;
|
|
uint64_t tmp_transid;
|
|
|
|
err = btrfs_util_subvolume_iterator_next(iter, &child_path,
|
|
NULL);
|
|
if (err) {
|
|
if (err == BTRFS_UTIL_ERROR_STOP_ITERATION)
|
|
err = BTRFS_UTIL_OK;
|
|
break;
|
|
}
|
|
|
|
/* Remove the placeholder directory. */
|
|
if (unlinkat(dstfd, child_path, AT_REMOVEDIR) == -1) {
|
|
free(child_path);
|
|
err = BTRFS_UTIL_ERROR_RMDIR_FAILED;
|
|
break;
|
|
}
|
|
|
|
child_fd = openat(fd, child_path, O_RDONLY);
|
|
if (child_fd == -1) {
|
|
free(child_path);
|
|
err = BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
break;
|
|
}
|
|
|
|
err = openat_parent_and_name(dstfd, child_path, child_name,
|
|
sizeof(child_name),
|
|
&new_parent_fd);
|
|
free(child_path);
|
|
if (err) {
|
|
SAVE_ERRNO_AND_CLOSE(child_fd);
|
|
break;
|
|
}
|
|
|
|
err = btrfs_util_create_snapshot_fd2(child_fd, new_parent_fd,
|
|
child_name, 0,
|
|
async_transid ? &tmp_transid : NULL,
|
|
NULL);
|
|
SAVE_ERRNO_AND_CLOSE(child_fd);
|
|
SAVE_ERRNO_AND_CLOSE(new_parent_fd);
|
|
if (err)
|
|
break;
|
|
if (async_transid && tmp_transid > *async_transid)
|
|
*async_transid = tmp_transid;
|
|
}
|
|
|
|
btrfs_util_destroy_subvolume_iterator(iter);
|
|
out:
|
|
SAVE_ERRNO_AND_CLOSE(dstfd);
|
|
return err;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_create_snapshot(const char *source,
|
|
const char *path,
|
|
int flags,
|
|
uint64_t *async_transid,
|
|
struct btrfs_util_qgroup_inherit *qgroup_inherit)
|
|
{
|
|
enum btrfs_util_error err;
|
|
int fd;
|
|
|
|
fd = open(source, O_RDONLY);
|
|
if (fd == -1)
|
|
return BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
|
|
err = btrfs_util_create_snapshot_fd(fd, path, flags, async_transid,
|
|
qgroup_inherit);
|
|
SAVE_ERRNO_AND_CLOSE(fd);
|
|
return err;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_create_snapshot_fd(int fd,
|
|
const char *path,
|
|
int flags,
|
|
uint64_t *async_transid,
|
|
struct btrfs_util_qgroup_inherit *qgroup_inherit)
|
|
{
|
|
char name[BTRFS_SUBVOL_NAME_MAX + 1];
|
|
enum btrfs_util_error err;
|
|
int parent_fd;
|
|
|
|
err = openat_parent_and_name(AT_FDCWD, path, name, sizeof(name),
|
|
&parent_fd);
|
|
if (err)
|
|
return err;
|
|
|
|
err = btrfs_util_create_snapshot_fd2(fd, parent_fd, name, flags,
|
|
async_transid, qgroup_inherit);
|
|
SAVE_ERRNO_AND_CLOSE(parent_fd);
|
|
return err;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_create_snapshot_fd2(int fd,
|
|
int parent_fd,
|
|
const char *name,
|
|
int flags,
|
|
uint64_t *async_transid,
|
|
struct btrfs_util_qgroup_inherit *qgroup_inherit)
|
|
{
|
|
struct btrfs_ioctl_vol_args_v2 args = {.fd = fd};
|
|
enum btrfs_util_error err;
|
|
size_t len;
|
|
int ret;
|
|
|
|
if ((flags & ~BTRFS_UTIL_CREATE_SNAPSHOT_MASK) ||
|
|
((flags & BTRFS_UTIL_CREATE_SNAPSHOT_READ_ONLY) &&
|
|
(flags & BTRFS_UTIL_CREATE_SNAPSHOT_RECURSIVE))) {
|
|
errno = EINVAL;
|
|
return BTRFS_UTIL_ERROR_INVALID_ARGUMENT;
|
|
}
|
|
|
|
if (flags & BTRFS_UTIL_CREATE_SNAPSHOT_READ_ONLY)
|
|
args.flags |= BTRFS_SUBVOL_RDONLY;
|
|
if (async_transid)
|
|
args.flags |= BTRFS_SUBVOL_CREATE_ASYNC;
|
|
if (qgroup_inherit) {
|
|
args.flags |= BTRFS_SUBVOL_QGROUP_INHERIT;
|
|
args.qgroup_inherit = (struct btrfs_qgroup_inherit *)qgroup_inherit;
|
|
args.size = (sizeof(*args.qgroup_inherit) +
|
|
args.qgroup_inherit->num_qgroups *
|
|
sizeof(args.qgroup_inherit->qgroups[0]));
|
|
}
|
|
|
|
len = strlen(name);
|
|
if (len >= sizeof(args.name)) {
|
|
errno = ENAMETOOLONG;
|
|
return BTRFS_UTIL_ERROR_INVALID_ARGUMENT;
|
|
}
|
|
memcpy(args.name, name, len);
|
|
args.name[len] = '\0';
|
|
|
|
ret = ioctl(parent_fd, BTRFS_IOC_SNAP_CREATE_V2, &args);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_SUBVOL_CREATE_FAILED;
|
|
|
|
if (async_transid)
|
|
*async_transid = args.transid;
|
|
|
|
if (flags & BTRFS_UTIL_CREATE_SNAPSHOT_RECURSIVE) {
|
|
err = snapshot_subvolume_children(fd, parent_fd, name,
|
|
async_transid);
|
|
if (err)
|
|
return err;
|
|
}
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
static enum btrfs_util_error delete_subvolume_children(int parent_fd,
|
|
const char *name)
|
|
{
|
|
struct btrfs_util_subvolume_iterator *iter;
|
|
enum btrfs_util_error err;
|
|
int fd;
|
|
|
|
fd = openat(parent_fd, name, O_RDONLY);
|
|
if (fd == -1)
|
|
return BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
|
|
err = btrfs_util_create_subvolume_iterator_fd(fd, 0,
|
|
BTRFS_UTIL_SUBVOLUME_ITERATOR_POST_ORDER,
|
|
&iter);
|
|
if (err)
|
|
goto out;
|
|
|
|
for (;;) {
|
|
char child_name[BTRFS_PATH_NAME_MAX + 1];
|
|
char *child_path;
|
|
int child_parent_fd;
|
|
|
|
err = btrfs_util_subvolume_iterator_next(iter, &child_path,
|
|
NULL);
|
|
if (err) {
|
|
if (err == BTRFS_UTIL_ERROR_STOP_ITERATION)
|
|
err = BTRFS_UTIL_OK;
|
|
break;
|
|
}
|
|
|
|
err = openat_parent_and_name(fd, child_path, child_name,
|
|
sizeof(child_name),
|
|
&child_parent_fd);
|
|
free(child_path);
|
|
if (err)
|
|
break;
|
|
|
|
err = btrfs_util_delete_subvolume_fd(child_parent_fd,
|
|
child_name, 0);
|
|
SAVE_ERRNO_AND_CLOSE(child_parent_fd);
|
|
if (err)
|
|
break;
|
|
}
|
|
|
|
btrfs_util_destroy_subvolume_iterator(iter);
|
|
out:
|
|
SAVE_ERRNO_AND_CLOSE(fd);
|
|
return err;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_delete_subvolume(const char *path,
|
|
int flags)
|
|
{
|
|
char name[BTRFS_PATH_NAME_MAX + 1];
|
|
enum btrfs_util_error err;
|
|
int parent_fd;
|
|
|
|
err = openat_parent_and_name(AT_FDCWD, path, name, sizeof(name),
|
|
&parent_fd);
|
|
if (err)
|
|
return err;
|
|
|
|
err = btrfs_util_delete_subvolume_fd(parent_fd, name, flags);
|
|
SAVE_ERRNO_AND_CLOSE(parent_fd);
|
|
return err;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_delete_subvolume_fd(int parent_fd,
|
|
const char *name,
|
|
int flags)
|
|
{
|
|
struct btrfs_ioctl_vol_args args = {};
|
|
enum btrfs_util_error err;
|
|
size_t len;
|
|
int ret;
|
|
|
|
if (flags & ~BTRFS_UTIL_DELETE_SUBVOLUME_MASK) {
|
|
errno = EINVAL;
|
|
return BTRFS_UTIL_ERROR_INVALID_ARGUMENT;
|
|
}
|
|
|
|
if (flags & BTRFS_UTIL_DELETE_SUBVOLUME_RECURSIVE) {
|
|
err = delete_subvolume_children(parent_fd, name);
|
|
if (err)
|
|
return err;
|
|
}
|
|
|
|
len = strlen(name);
|
|
if (len >= sizeof(args.name)) {
|
|
errno = ENAMETOOLONG;
|
|
return BTRFS_UTIL_ERROR_INVALID_ARGUMENT;
|
|
}
|
|
memcpy(args.name, name, len);
|
|
args.name[len] = '\0';
|
|
|
|
ret = ioctl(parent_fd, BTRFS_IOC_SNAP_DESTROY, &args);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_SNAP_DESTROY_FAILED;
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
PUBLIC void btrfs_util_destroy_subvolume_iterator(struct btrfs_util_subvolume_iterator *iter)
|
|
{
|
|
if (iter) {
|
|
free(iter->cur_path);
|
|
free(iter->search_stack);
|
|
if (iter->cur_fd != iter->fd)
|
|
SAVE_ERRNO_AND_CLOSE(iter->cur_fd);
|
|
if (iter->flags & BTRFS_UTIL_SUBVOLUME_ITERATOR_CLOSE_FD)
|
|
SAVE_ERRNO_AND_CLOSE(iter->fd);
|
|
free(iter);
|
|
}
|
|
}
|
|
|
|
PUBLIC int btrfs_util_subvolume_iterator_fd(const struct btrfs_util_subvolume_iterator *iter)
|
|
{
|
|
return iter->fd;
|
|
}
|
|
|
|
static enum btrfs_util_error build_subvol_path(struct btrfs_util_subvolume_iterator *iter,
|
|
const char *name, size_t name_len,
|
|
const char *dir, size_t dir_len,
|
|
size_t *path_len_ret)
|
|
{
|
|
struct search_stack_entry *top = top_search_stack_entry(iter);
|
|
size_t path_len;
|
|
char *p;
|
|
|
|
path_len = top->path_len;
|
|
/*
|
|
* We need a joining slash if we have a current path and a subdirectory.
|
|
*/
|
|
if (top->path_len && dir_len)
|
|
path_len++;
|
|
path_len += dir_len;
|
|
/*
|
|
* We need another joining slash if we have a current path and a name,
|
|
* but not if we have a subdirectory, because the lookup ioctl includes
|
|
* a trailing slash.
|
|
*/
|
|
if (top->path_len && !dir_len && name_len)
|
|
path_len++;
|
|
path_len += name_len;
|
|
|
|
/* We need one extra character for the NUL terminator. */
|
|
if (path_len + 1 > iter->cur_path_capacity) {
|
|
char *tmp = realloc(iter->cur_path, path_len + 1);
|
|
|
|
if (!tmp)
|
|
return BTRFS_UTIL_ERROR_NO_MEMORY;
|
|
iter->cur_path = tmp;
|
|
iter->cur_path_capacity = path_len + 1;
|
|
}
|
|
|
|
p = iter->cur_path + top->path_len;
|
|
if (top->path_len && dir_len)
|
|
*p++ = '/';
|
|
memcpy(p, dir, dir_len);
|
|
p += dir_len;
|
|
if (top->path_len && !dir_len && name_len)
|
|
*p++ = '/';
|
|
memcpy(p, name, name_len);
|
|
p += name_len;
|
|
*p = '\0';
|
|
|
|
*path_len_ret = path_len;
|
|
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
static enum btrfs_util_error build_subvol_path_privileged(struct btrfs_util_subvolume_iterator *iter,
|
|
const struct btrfs_ioctl_search_header *header,
|
|
const struct btrfs_root_ref *ref,
|
|
const char *name,
|
|
size_t *path_len_ret)
|
|
{
|
|
struct btrfs_ioctl_ino_lookup_args lookup = {
|
|
.treeid = header->objectid,
|
|
.objectid = le64_to_cpu(ref->dirid),
|
|
};
|
|
int ret;
|
|
|
|
ret = ioctl(iter->fd, BTRFS_IOC_INO_LOOKUP, &lookup);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_INO_LOOKUP_FAILED;
|
|
|
|
return build_subvol_path(iter, name, le16_to_cpu(ref->name_len),
|
|
lookup.name, strlen(lookup.name),
|
|
path_len_ret);
|
|
}
|
|
|
|
static enum btrfs_util_error build_subvol_path_unprivileged(struct btrfs_util_subvolume_iterator *iter,
|
|
uint64_t treeid,
|
|
uint64_t dirid,
|
|
size_t *path_len_ret)
|
|
{
|
|
struct btrfs_ioctl_ino_lookup_user_args args = {
|
|
.treeid = treeid,
|
|
.dirid = dirid,
|
|
};
|
|
int ret;
|
|
|
|
ret = ioctl(iter->cur_fd, BTRFS_IOC_INO_LOOKUP_USER, &args);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_INO_LOOKUP_USER_FAILED;
|
|
|
|
return build_subvol_path(iter, args.name, strlen(args.name),
|
|
args.path, strlen(args.path), path_len_ret);
|
|
}
|
|
|
|
static enum btrfs_util_error subvolume_iterator_next_tree_search(struct btrfs_util_subvolume_iterator *iter,
|
|
char **path_ret,
|
|
uint64_t *id_ret)
|
|
{
|
|
struct search_stack_entry *top;
|
|
const struct btrfs_ioctl_search_header *header;
|
|
const struct btrfs_root_ref *ref;
|
|
const char *name;
|
|
enum btrfs_util_error err;
|
|
size_t path_len;
|
|
int ret;
|
|
|
|
for (;;) {
|
|
for (;;) {
|
|
if (iter->search_stack_len == 0)
|
|
return BTRFS_UTIL_ERROR_STOP_ITERATION;
|
|
|
|
top = top_search_stack_entry(iter);
|
|
if (top->items_pos < top->search.key.nr_items) {
|
|
break;
|
|
} else {
|
|
top->search.key.nr_items = 4096;
|
|
ret = ioctl(iter->fd, BTRFS_IOC_TREE_SEARCH, &top->search);
|
|
if (ret == -1)
|
|
return BTRFS_UTIL_ERROR_SEARCH_FAILED;
|
|
top->items_pos = 0;
|
|
top->buf_off = 0;
|
|
|
|
if (top->search.key.nr_items == 0) {
|
|
/*
|
|
* This never fails for use_tree_search.
|
|
*/
|
|
pop_search_stack(iter);
|
|
if ((iter->flags & BTRFS_UTIL_SUBVOLUME_ITERATOR_POST_ORDER) &&
|
|
iter->search_stack_len)
|
|
goto out;
|
|
}
|
|
}
|
|
}
|
|
|
|
header = (struct btrfs_ioctl_search_header *)(top->search.buf + top->buf_off);
|
|
|
|
top->items_pos++;
|
|
top->buf_off += sizeof(*header) + header->len;
|
|
top->search.key.min_offset = header->offset + 1;
|
|
|
|
/* This shouldn't happen, but handle it just in case. */
|
|
if (header->type != BTRFS_ROOT_REF_KEY)
|
|
continue;
|
|
|
|
ref = (struct btrfs_root_ref *)(header + 1);
|
|
name = (const char *)(ref + 1);
|
|
err = build_subvol_path_privileged(iter, header, ref, name,
|
|
&path_len);
|
|
if (err)
|
|
return err;
|
|
|
|
err = append_to_search_stack(iter, header->offset, path_len);
|
|
if (err)
|
|
return err;
|
|
|
|
if (!(iter->flags & BTRFS_UTIL_SUBVOLUME_ITERATOR_POST_ORDER)) {
|
|
top = top_search_stack_entry(iter);
|
|
goto out;
|
|
}
|
|
}
|
|
|
|
out:
|
|
if (path_ret) {
|
|
*path_ret = malloc(top->path_len + 1);
|
|
if (!*path_ret)
|
|
return BTRFS_UTIL_ERROR_NO_MEMORY;
|
|
memcpy(*path_ret, iter->cur_path, top->path_len);
|
|
(*path_ret)[top->path_len] = '\0';
|
|
}
|
|
if (id_ret)
|
|
*id_ret = top->search.key.min_objectid;
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
static enum btrfs_util_error subvolume_iterator_next_unprivileged(struct btrfs_util_subvolume_iterator *iter,
|
|
char **path_ret,
|
|
uint64_t *id_ret)
|
|
{
|
|
struct search_stack_entry *top;
|
|
uint64_t treeid, dirid;
|
|
enum btrfs_util_error err;
|
|
size_t path_len;
|
|
int ret;
|
|
|
|
for (;;) {
|
|
for (;;) {
|
|
if (iter->search_stack_len == 0)
|
|
return BTRFS_UTIL_ERROR_STOP_ITERATION;
|
|
|
|
top = top_search_stack_entry(iter);
|
|
if (top->items_pos < top->rootref_args.num_items) {
|
|
break;
|
|
} else {
|
|
ret = ioctl(iter->cur_fd,
|
|
BTRFS_IOC_GET_SUBVOL_ROOTREF,
|
|
&top->rootref_args);
|
|
if (ret == -1 && errno != EOVERFLOW)
|
|
return BTRFS_UTIL_ERROR_GET_SUBVOL_ROOTREF_FAILED;
|
|
top->items_pos = 0;
|
|
|
|
if (top->rootref_args.num_items == 0) {
|
|
err = pop_search_stack(iter);
|
|
if (err)
|
|
return err;
|
|
if ((iter->flags & BTRFS_UTIL_SUBVOLUME_ITERATOR_POST_ORDER) &&
|
|
iter->search_stack_len)
|
|
goto out;
|
|
}
|
|
}
|
|
}
|
|
|
|
treeid = top->rootref_args.rootref[top->items_pos].treeid;
|
|
dirid = top->rootref_args.rootref[top->items_pos].dirid;
|
|
top->items_pos++;
|
|
err = build_subvol_path_unprivileged(iter, treeid, dirid,
|
|
&path_len);
|
|
if (err) {
|
|
/* Skip the subvolume if we can't access it. */
|
|
if (errno == EACCES)
|
|
continue;
|
|
return err;
|
|
}
|
|
|
|
err = append_to_search_stack(iter, treeid, path_len);
|
|
if (err) {
|
|
/*
|
|
* Skip the subvolume if it does not exist (which can
|
|
* happen if there is another filesystem mounted over a
|
|
* parent directory) or we don't have permission to
|
|
* access it.
|
|
*/
|
|
if (errno == ENOENT || errno == EACCES)
|
|
continue;
|
|
return err;
|
|
}
|
|
|
|
if (!(iter->flags & BTRFS_UTIL_SUBVOLUME_ITERATOR_POST_ORDER)) {
|
|
top = top_search_stack_entry(iter);
|
|
goto out;
|
|
}
|
|
}
|
|
|
|
out:
|
|
if (path_ret) {
|
|
*path_ret = malloc(top->path_len + 1);
|
|
if (!*path_ret)
|
|
return BTRFS_UTIL_ERROR_NO_MEMORY;
|
|
memcpy(*path_ret, iter->cur_path, top->path_len);
|
|
(*path_ret)[top->path_len] = '\0';
|
|
}
|
|
if (id_ret)
|
|
*id_ret = top->id;
|
|
return BTRFS_UTIL_OK;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_subvolume_iterator_next(struct btrfs_util_subvolume_iterator *iter,
|
|
char **path_ret,
|
|
uint64_t *id_ret)
|
|
{
|
|
if (iter->use_tree_search) {
|
|
return subvolume_iterator_next_tree_search(iter, path_ret,
|
|
id_ret);
|
|
} else {
|
|
return subvolume_iterator_next_unprivileged(iter, path_ret,
|
|
id_ret);
|
|
}
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_subvolume_iterator_next_info(struct btrfs_util_subvolume_iterator *iter,
|
|
char **path_ret,
|
|
struct btrfs_util_subvolume_info *subvol)
|
|
{
|
|
enum btrfs_util_error err;
|
|
uint64_t id;
|
|
|
|
err = btrfs_util_subvolume_iterator_next(iter, path_ret, &id);
|
|
if (err)
|
|
return err;
|
|
|
|
if (iter->use_tree_search)
|
|
return btrfs_util_subvolume_info_fd(iter->fd, id, subvol);
|
|
else
|
|
return btrfs_util_subvolume_info_fd(iter->cur_fd, 0, subvol);
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_deleted_subvolumes(const char *path,
|
|
uint64_t **ids,
|
|
size_t *n)
|
|
{
|
|
enum btrfs_util_error err;
|
|
int fd;
|
|
|
|
fd = open(path, O_RDONLY);
|
|
if (fd == -1)
|
|
return BTRFS_UTIL_ERROR_OPEN_FAILED;
|
|
|
|
err = btrfs_util_deleted_subvolumes_fd(fd, ids, n);
|
|
SAVE_ERRNO_AND_CLOSE(fd);
|
|
return err;
|
|
}
|
|
|
|
PUBLIC enum btrfs_util_error btrfs_util_deleted_subvolumes_fd(int fd,
|
|
uint64_t **ids,
|
|
size_t *n)
|
|
{
|
|
size_t capacity = 0;
|
|
struct btrfs_ioctl_search_args search = {
|
|
.key = {
|
|
.tree_id = BTRFS_ROOT_TREE_OBJECTID,
|
|
.min_objectid = BTRFS_ORPHAN_OBJECTID,
|
|
.max_objectid = BTRFS_ORPHAN_OBJECTID,
|
|
.min_type = BTRFS_ORPHAN_ITEM_KEY,
|
|
.max_type = BTRFS_ORPHAN_ITEM_KEY,
|
|
.min_offset = 0,
|
|
.max_offset = UINT64_MAX,
|
|
.min_transid = 0,
|
|
.max_transid = UINT64_MAX,
|
|
.nr_items = 0,
|
|
},
|
|
};
|
|
enum btrfs_util_error err;
|
|
size_t items_pos = 0, buf_off = 0;
|
|
int ret;
|
|
|
|
*ids = NULL;
|
|
*n = 0;
|
|
for (;;) {
|
|
const struct btrfs_ioctl_search_header *header;
|
|
|
|
if (items_pos >= search.key.nr_items) {
|
|
search.key.nr_items = 4096;
|
|
ret = ioctl(fd, BTRFS_IOC_TREE_SEARCH, &search);
|
|
if (ret == -1) {
|
|
err = BTRFS_UTIL_ERROR_SEARCH_FAILED;
|
|
goto out;
|
|
}
|
|
items_pos = 0;
|
|
buf_off = 0;
|
|
|
|
if (search.key.nr_items == 0)
|
|
break;
|
|
}
|
|
|
|
header = (struct btrfs_ioctl_search_header *)(search.buf + buf_off);
|
|
|
|
/*
|
|
* The orphan item might be for a free space cache inode, so
|
|
* check if there's a matching root item.
|
|
*/
|
|
err = btrfs_util_subvolume_info_fd(fd, header->offset, NULL);
|
|
if (!err) {
|
|
if (*n >= capacity) {
|
|
size_t new_capacity;
|
|
uint64_t *new_ids;
|
|
|
|
new_capacity = capacity ? capacity * 2 : 1;
|
|
new_ids = reallocarray(*ids, new_capacity,
|
|
sizeof(**ids));
|
|
if (!new_ids) {
|
|
err = BTRFS_UTIL_ERROR_NO_MEMORY;
|
|
goto out;
|
|
}
|
|
|
|
*ids = new_ids;
|
|
capacity = new_capacity;
|
|
}
|
|
(*ids)[(*n)++] = header->offset;
|
|
} else if (err != BTRFS_UTIL_ERROR_SUBVOLUME_NOT_FOUND) {
|
|
goto out;
|
|
}
|
|
|
|
items_pos++;
|
|
buf_off += sizeof(*header) + header->len;
|
|
search.key.min_offset = header->offset + 1;
|
|
}
|
|
|
|
err = BTRFS_UTIL_OK;
|
|
out:
|
|
if (err) {
|
|
free(*ids);
|
|
*ids = NULL;
|
|
*n = 0;
|
|
}
|
|
return err;
|
|
}
|