/* afs_put_super
* Called from unmount to release super_block. */
static void
afs_put_super(struct super_block *sbp)
{
AFS_GLOCK();
AFS_STATCNT(afs_unmount);
afs_globalVFS = 0;
afs_globalVp = 0;
afs_shutdown();
mntput(afs_cacheMnt);
osi_linux_verify_alloced_memory();
#if defined(HAVE_LINUX_BDI_INIT)
bdi_destroy(afs_backing_dev_info);
#endif
kfree(afs_backing_dev_info);
AFS_GUNLOCK();
sbp->s_dev = 0;
module_put(THIS_MODULE);
}
/*
* dentry_open() will have done dput(dentry) and mntput(mnt) if it returns an
* error.
*/
struct file *dentry_open(struct dentry *dentry, struct vfsmount *mnt, int flags,
const struct cred *cred)
{
int error;
struct file *f;
validate_creds(cred);
/* We must always pass in a valid mount pointer. */
BUG_ON(!mnt);
error = -ENFILE;
f = get_empty_filp();
if (f == NULL) {
dput(dentry);
mntput(mnt);
return ERR_PTR(error);
}
f->f_flags = flags;
return __dentry_open(dentry, mnt, f, NULL, cred);
}
/**
* ptm_open_peer - open the peer of a pty
* @master: the open struct file of the ptmx device node
* @tty: the master of the pty being opened
* @flags: the flags for open
*
* Provide a race free way for userspace to open the slave end of a pty
* (where they have the master fd and cannot access or trust the mount
* namespace /dev/pts was mounted inside).
*/
int ptm_open_peer(struct file *master, struct tty_struct *tty, int flags)
{
int fd = -1;
struct file *filp;
int retval = -EINVAL;
struct path path;
if (tty->driver != ptm_driver)
return -EIO;
fd = get_unused_fd_flags(flags);
if (fd < 0) {
retval = fd;
goto err;
}
/* Compute the slave's path */
path.mnt = devpts_mntget(master, tty->driver_data);
if (IS_ERR(path.mnt)) {
retval = PTR_ERR(path.mnt);
goto err_put;
}
path.dentry = tty->link->driver_data;
filp = dentry_open(&path, flags, current_cred());
mntput(path.mnt);
if (IS_ERR(filp)) {
retval = PTR_ERR(filp);
goto err_put;
}
fd_install(fd, filp);
return fd;
err_put:
put_unused_fd(fd);
err:
return retval;
}
static int acct_on(char *name)
{
struct file *file;
struct _file *_file;
int error;
/* Difference from BSD - they don't do O_APPEND */
file = filp_open(name, O_WRONLY|O_APPEND|O_LARGEFILE, 0);
if (IS_ERR(file))
return PTR_ERR(file);
_file = tx_cache_get_file(file);
if (!S_ISREG(d_get_inode(f_get_dentry_ro(_file))->i_mode)) {
filp_close(file, NULL);
return -EACCES;
}
if (!file->f_op->write) {
filp_close(file, NULL);
return -EIO;
}
error = security_acct(file);
if (error) {
filp_close(file, NULL);
return error;
}
spin_lock(&acct_globals.lock);
mnt_pin(_file->f_path.mnt);
acct_file_reopen(file);
spin_unlock(&acct_globals.lock);
mntput(_file->f_path.mnt); /* it's pinned, now give up active reference */
return 0;
}
/* afs_put_super
* Called from unmount to release super_block. */
static void
afs_put_super(struct super_block *sbp)
{
AFS_GLOCK();
AFS_STATCNT(afs_unmount);
afs_globalVFS = 0;
afs_globalVp = 0;
osi_linux_free_inode_pages(); /* invalidate and release remaining AFS inodes. */
afs_shutdown();
mntput(afs_cacheMnt);
osi_linux_verify_alloced_memory();
#if defined(HAVE_LINUX_BDI_INIT)
bdi_destroy(afs_backing_dev_info);
#endif
osi_Free(afs_backing_dev_info, sizeof(struct backing_dev_info));
AFS_GUNLOCK();
sbp->s_dev = 0;
module_put(THIS_MODULE);
}
/* the real guts of fput() - releasing the last reference to file
*/
static void __fput(struct file *file)
{
struct dentry *dentry = file->f_path.dentry;
struct vfsmount *mnt = file->f_path.mnt;
struct inode *inode = dentry->d_inode;
might_sleep();
fsnotify_close(file);
/*
* The function eventpoll_release() should be the first called
* in the file cleanup chain.
*/
eventpoll_release(file);
locks_remove_flock(file);
if (unlikely(file->f_flags & FASYNC)) {
if (file->f_op && file->f_op->fasync)
file->f_op->fasync(-1, file, 0);
}
if (file->f_op && file->f_op->release)
file->f_op->release(inode, file);
security_file_free(file);
ima_file_free(file);
if (unlikely(S_ISCHR(inode->i_mode) && inode->i_cdev != NULL))
cdev_put(inode->i_cdev);
fops_put(file->f_op);
put_pid(file->f_owner.pid);
file_sb_list_del(file);
if (file->f_mode & FMODE_WRITE)
drop_file_write_access(file);
file->f_path.dentry = NULL;
file->f_path.mnt = NULL;
file_free(file);
dput(dentry);
mntput(mnt);
}
/**
* \<\<public\>\> Releases the root directory and then the mount
* object. VFS takes care of releasing the object in case their
* reference count reaches 0. We need to take care of setting the
* vfs_mnt to NULL.
*
* The last 2 steps don't need to be semaphore protected as we are
* working on copies of both pointers and we wanted to protect the
* access to vfs_mnt only. If anybody tries to get a root from now on
* a new VFS mount will be created. If the root directory has been
* also destroyed, the read_super method takes care of instantiating a
* new super block along with the root directory.
*/
void tcmi_ctlfs_put_root(void)
{
struct vfsmount *cur_vfs_mnt;
struct tcmi_ctlfs_entry *cur_root_dir;
mdbg(INFO3, "sb c_count=%d, s_active=%d", self.sb->s_count,
atomic_read(&self.sb->s_active));
mdbg(INFO3, "mount mnt_count=%d", atomic_read(&self.vfs_mnt->mnt_count));
/* mdbg(INFO3, "root d_count=%d", atomic_read(&self.root_dir->super.dentry->d_count));*/
down(&self.vfs_mnt_sem);
cur_vfs_mnt = self.vfs_mnt;
cur_root_dir = self.root_dir;
/* If we are the last thread accesing the kernel mount */
if (atomic_read(&self.vfs_mnt->mnt_count) == 1)
self.vfs_mnt = NULL;
up(&self.vfs_mnt_sem);
/* no locking needed since working on copies of both
* pointers */
tcmi_ctlfs_entry_put(cur_root_dir);
mntput(cur_vfs_mnt);
}
/*
* free a single branch
*/
static void au_br_do_free(struct au_branch *br)
{
int i;
struct au_wbr *wbr;
struct au_dykey **key;
au_hnotify_fin_br(br);
if (br->br_xino.xi_file)
fput(br->br_xino.xi_file);
mutex_destroy(&br->br_xino.xi_nondir_mtx);
AuDebugOn(atomic_read(&br->br_count));
wbr = br->br_wbr;
if (wbr) {
for (i = 0; i < AuBrWh_Last; i++)
dput(wbr->wbr_wh[i]);
AuDebugOn(atomic_read(&wbr->wbr_wh_running));
AuRwDestroy(&wbr->wbr_wh_rwsem);
}
key = br->br_dykey;
for (i = 0; i < AuBrDynOp; i++, key++)
if (*key)
au_dy_put(*key);
else
break;
/* recursive lock, s_umount of branch's */
lockdep_off();
mntput(br->br_mnt);
lockdep_on();
kfree(wbr);
kfree(br);
}
static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt,
struct file *f,
int (*open)(struct inode *, struct file *),
const struct cred *cred)
{
static const struct file_operations empty_fops = {};
struct inode *inode;
int error;
f->f_mode = OPEN_FMODE(f->f_flags) | FMODE_LSEEK |
FMODE_PREAD | FMODE_PWRITE;
if (unlikely(f->f_flags & O_PATH))
f->f_mode = FMODE_PATH;
inode = dentry->d_inode;
if (f->f_mode & FMODE_WRITE) {
error = __get_file_write_access(inode, mnt);
if (error)
goto cleanup_file;
if (!special_file(inode->i_mode))
file_take_write(f);
}
f->f_mapping = inode->i_mapping;
f->f_path.dentry = dentry;
f->f_path.mnt = mnt;
f->f_pos = 0;
file_sb_list_add(f, inode->i_sb);
if (unlikely(f->f_mode & FMODE_PATH)) {
f->f_op = &empty_fops;
return f;
}
f->f_op = fops_get(inode->i_fop);
error = security_dentry_open(f, cred);
if (error)
goto cleanup_all;
error = break_lease(inode, f->f_flags);
if (error)
goto cleanup_all;
if (!open && f->f_op)
open = f->f_op->open;
if (open) {
error = open(inode, f);
if (error)
goto cleanup_all;
}
if ((f->f_mode & (FMODE_READ | FMODE_WRITE)) == FMODE_READ)
i_readcount_inc(inode);
f->f_flags &= ~(O_CREAT | O_EXCL | O_NOCTTY | O_TRUNC);
file_ra_state_init(&f->f_ra, f->f_mapping->host->i_mapping);
/* NB: we're sure to have correct a_ops only after f_op->open */
if (f->f_flags & O_DIRECT) {
if (!f->f_mapping->a_ops ||
((!f->f_mapping->a_ops->direct_IO) &&
(!f->f_mapping->a_ops->get_xip_mem))) {
fput(f);
f = ERR_PTR(-EINVAL);
}
}
return f;
cleanup_all:
fops_put(f->f_op);
if (f->f_mode & FMODE_WRITE) {
put_write_access(inode);
if (!special_file(inode->i_mode)) {
/*
* We don't consider this a real
* mnt_want/drop_write() pair
* because it all happenend right
* here, so just reset the state.
*/
file_reset_write(f);
mnt_drop_write(mnt);
}
}
file_sb_list_del(f);
f->f_path.dentry = NULL;
f->f_path.mnt = NULL;
cleanup_file:
put_filp(f);
dput(dentry);
mntput(mnt);
return ERR_PTR(error);
}
void pid_ns_release_proc(struct pid_namespace *ns)
{
mntput(ns->proc_mnt);
}
static struct file *__dentry_open(struct _dentry *dentry, struct vfsmount *mnt,
int flags, struct file *f,
int (*open)(struct _inode *, struct file *))
{
struct _inode *inode;
int error;
struct super_block *sb;
struct _file *_f = tx_cache_get_file(f);
_f->f_flags = flags;
_f->f_mode = ((flags+1) & O_ACCMODE) | FMODE_LSEEK |
FMODE_PREAD | FMODE_PWRITE;
inode = d_get_inode(dentry);
sb = inode->i_sb;
if (_f->f_mode & FMODE_WRITE) {
error = get_write_access(parent(inode));
if (error)
goto cleanup_file;
}
f->f_mapping = inode->i_mapping;
_f->f_path.dentry = parent(dentry);
_f->f_path.mnt = mnt;
_f->f_pos = 0;
f->f_op = fops_get(inode->i_fop);
file_move(f, &sb->s_files);
if (!open && f->f_op)
open = f->f_op->open;
if (open) {
error = open(inode, f);
if (error)
goto cleanup_all;
}
_f->f_flags &= ~(O_CREAT | O_EXCL | O_NOCTTY | O_TRUNC);
file_ra_state_init(&f->f_ra,
tx_cache_get_inode(f->f_mapping->host)->i_mapping);
/* NB: we're sure to have correct a_ops only after f_op->open */
if (_f->f_flags & O_DIRECT) {
if (!f->f_mapping->a_ops ||
((!f->f_mapping->a_ops->direct_IO) &&
(!f->f_mapping->a_ops->get_xip_page))) {
fput(f);
f = ERR_PTR(-EINVAL);
}
}
return f;
cleanup_all:
fops_put(f->f_op);
if (_f->f_mode & FMODE_WRITE)
put_write_access(parent(inode));
file_kill(f);
_f->f_path.dentry = NULL;
_f->f_path.mnt = NULL;
cleanup_file:
/* Avoid issues if we recycle this object */
if(live_transaction())
early_release(&f->xobj, 1);
put_filp(f);
dput(parent(dentry));
mntput(mnt);
return ERR_PTR(error);
}
/*
* add a cache
*/
static int cachefiles_daemon_add_cache(struct cachefiles_cache *cache)
{
struct cachefiles_object *fsdef;
struct path path;
struct kstatfs stats;
struct dentry *graveyard, *cachedir, *root;
const struct cred *saved_cred;
int ret;
_enter("");
/* we want to work under the module's security ID */
ret = cachefiles_get_security_ID(cache);
if (ret < 0)
return ret;
cachefiles_begin_secure(cache, &saved_cred);
/* allocate the root index object */
ret = -ENOMEM;
fsdef = kmem_cache_alloc(cachefiles_object_jar, GFP_KERNEL);
if (!fsdef)
goto error_root_object;
ASSERTCMP(fsdef->backer, ==, NULL);
atomic_set(&fsdef->usage, 1);
fsdef->type = FSCACHE_COOKIE_TYPE_INDEX;
_debug("- fsdef %p", fsdef);
/* look up the directory at the root of the cache */
ret = kern_path(cache->rootdirname, LOOKUP_DIRECTORY, &path);
if (ret < 0)
goto error_open_root;
cache->mnt = path.mnt;
root = path.dentry;
/* check parameters */
ret = -EOPNOTSUPP;
if (!root->d_inode ||
!root->d_inode->i_op->lookup ||
!root->d_inode->i_op->mkdir ||
!root->d_inode->i_op->setxattr ||
!root->d_inode->i_op->getxattr ||
!root->d_sb->s_op->statfs ||
!root->d_sb->s_op->sync_fs)
goto error_unsupported;
ret = -EROFS;
if (root->d_sb->s_flags & MS_RDONLY)
goto error_unsupported;
/* determine the security of the on-disk cache as this governs
* security ID of files we create */
ret = cachefiles_determine_cache_security(cache, root, &saved_cred);
if (ret < 0)
goto error_unsupported;
/* get the cache size and blocksize */
ret = vfs_statfs(&path, &stats);
if (ret < 0)
goto error_unsupported;
ret = -ERANGE;
if (stats.f_bsize <= 0)
goto error_unsupported;
ret = -EOPNOTSUPP;
if (stats.f_bsize > PAGE_SIZE)
goto error_unsupported;
cache->bsize = stats.f_bsize;
cache->bshift = 0;
if (stats.f_bsize < PAGE_SIZE)
cache->bshift = PAGE_SHIFT - ilog2(stats.f_bsize);
_debug("blksize %u (shift %u)",
cache->bsize, cache->bshift);
_debug("size %llu, avail %llu",
(unsigned long long) stats.f_blocks,
(unsigned long long) stats.f_bavail);
/* set up caching limits */
do_div(stats.f_files, 100);
cache->fstop = stats.f_files * cache->fstop_percent;
cache->fcull = stats.f_files * cache->fcull_percent;
cache->frun = stats.f_files * cache->frun_percent;
_debug("limits {%llu,%llu,%llu} files",
(unsigned long long) cache->frun,
(unsigned long long) cache->fcull,
(unsigned long long) cache->fstop);
stats.f_blocks >>= cache->bshift;
do_div(stats.f_blocks, 100);
cache->bstop = stats.f_blocks * cache->bstop_percent;
cache->bcull = stats.f_blocks * cache->bcull_percent;
cache->brun = stats.f_blocks * cache->brun_percent;
_debug("limits {%llu,%llu,%llu} blocks",
(unsigned long long) cache->brun,
(unsigned long long) cache->bcull,
(unsigned long long) cache->bstop);
/* get the cache directory and check its type */
cachedir = cachefiles_get_directory(cache, root, "cache");
if (IS_ERR(cachedir)) {
ret = PTR_ERR(cachedir);
goto error_unsupported;
}
fsdef->dentry = cachedir;
fsdef->fscache.cookie = NULL;
ret = cachefiles_check_object_type(fsdef);
if (ret < 0)
goto error_unsupported;
/* get the graveyard directory */
graveyard = cachefiles_get_directory(cache, root, "graveyard");
if (IS_ERR(graveyard)) {
ret = PTR_ERR(graveyard);
goto error_unsupported;
}
cache->graveyard = graveyard;
/* publish the cache */
fscache_init_cache(&cache->cache,
&cachefiles_cache_ops,
"%s",
fsdef->dentry->d_sb->s_id);
fscache_object_init(&fsdef->fscache, NULL, &cache->cache);
ret = fscache_add_cache(&cache->cache, &fsdef->fscache, cache->tag);
if (ret < 0)
goto error_add_cache;
/* done */
set_bit(CACHEFILES_READY, &cache->flags);
dput(root);
pr_info("File cache on %s registered\n", cache->cache.identifier);
/* check how much space the cache has */
cachefiles_has_space(cache, 0, 0);
cachefiles_end_secure(cache, saved_cred);
return 0;
error_add_cache:
dput(cache->graveyard);
cache->graveyard = NULL;
error_unsupported:
mntput(cache->mnt);
cache->mnt = NULL;
dput(fsdef->dentry);
fsdef->dentry = NULL;
dput(root);
error_open_root:
kmem_cache_free(cachefiles_object_jar, fsdef);
error_root_object:
cachefiles_end_secure(cache, saved_cred);
pr_err("Failed to register: %d\n", ret);
return ret;
}
/*
* nfs_follow_mountpoint - handle crossing a mountpoint on the server
* @dentry - dentry of mountpoint
* @nd - nameidata info
*
* When we encounter a mountpoint on the server, we want to set up
* a mountpoint on the client too, to prevent inode numbers from
* colliding, and to allow "df" to work properly.
* On NFSv4, we also want to allow for the fact that different
* filesystems may be migrated to different servers in a failover
* situation, and that different filesystems may want to use
* different security flavours.
*/
static void * nfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd)
{
struct vfsmount *mnt;
struct nfs_server *server = NFS_SERVER(dentry->d_inode);
struct dentry *parent;
struct nfs_fh fh;
struct nfs_fattr fattr;
int err;
dprintk("--> nfs_follow_mountpoint()\n");
err = -ESTALE;
if (IS_ROOT(dentry))
goto out_err;
dprintk("%s: enter\n", __func__);
dput(nd->path.dentry);
nd->path.dentry = dget(dentry);
/* Look it up again */
parent = dget_parent(nd->path.dentry);
err = server->nfs_client->rpc_ops->lookup(parent->d_inode,
&nd->path.dentry->d_name,
&fh, &fattr);
dput(parent);
if (err != 0)
goto out_err;
if (fattr.valid & NFS_ATTR_FATTR_V4_REFERRAL)
mnt = nfs_do_refmount(nd->path.mnt, nd->path.dentry);
else
mnt = nfs_do_submount(nd->path.mnt, nd->path.dentry, &fh,
&fattr);
err = PTR_ERR(mnt);
if (IS_ERR(mnt))
goto out_err;
mntget(mnt);
err = do_add_mount(mnt, &nd->path, nd->path.mnt->mnt_flags|MNT_SHRINKABLE,
&nfs_automount_list);
if (err < 0) {
mntput(mnt);
if (err == -EBUSY)
goto out_follow;
goto out_err;
}
path_put(&nd->path);
nd->path.mnt = mnt;
nd->path.dentry = dget(mnt->mnt_root);
schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout);
out:
dprintk("%s: done, returned %d\n", __func__, err);
dprintk("<-- nfs_follow_mountpoint() = %d\n", err);
return ERR_PTR(err);
out_err:
path_put(&nd->path);
goto out;
out_follow:
while (d_mountpoint(nd->path.dentry) &&
follow_down(&nd->path.mnt, &nd->path.dentry))
;
err = 0;
goto out;
}
/**
* ecryptfs_open
* @inode: inode speciying file to open
* @file: Structure to return filled in
*
* Opens the file specified by inode.
*
* Returns zero on success; non-zero otherwise
*/
static int ecryptfs_open(struct inode *inode, struct file *file)
{
int rc = 0;
struct ecryptfs_crypt_stat *crypt_stat = NULL;
struct ecryptfs_mount_crypt_stat *mount_crypt_stat;
struct dentry *ecryptfs_dentry = file->f_path.dentry;
/* Private value of ecryptfs_dentry allocated in
* ecryptfs_lookup() */
struct dentry *lower_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry);
struct inode *lower_inode = NULL;
struct file *lower_file = NULL;
struct vfsmount *lower_mnt;
struct ecryptfs_file_info *file_info;
int lower_flags;
/* Released in ecryptfs_release or end of function if failure */
file_info = kmem_cache_alloc(ecryptfs_file_info_cache, GFP_KERNEL);
ecryptfs_set_file_private(file, file_info);
if (!file_info) {
ecryptfs_printk(KERN_ERR,
"Error attempting to allocate memory\n");
rc = -ENOMEM;
goto out;
}
memset(file_info, 0, sizeof(*file_info));
lower_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry);
crypt_stat = &ecryptfs_inode_to_private(inode)->crypt_stat;
mount_crypt_stat = &ecryptfs_superblock_to_private(
ecryptfs_dentry->d_sb)->mount_crypt_stat;
mutex_lock(&crypt_stat->cs_mutex);
if (!ECRYPTFS_CHECK_FLAG(crypt_stat->flags, ECRYPTFS_POLICY_APPLIED)) {
ecryptfs_printk(KERN_DEBUG, "Setting flags for stat...\n");
/* Policy code enabled in future release */
ECRYPTFS_SET_FLAG(crypt_stat->flags, ECRYPTFS_POLICY_APPLIED);
ECRYPTFS_SET_FLAG(crypt_stat->flags, ECRYPTFS_ENCRYPTED);
}
mutex_unlock(&crypt_stat->cs_mutex);
lower_flags = file->f_flags;
if ((lower_flags & O_ACCMODE) == O_WRONLY)
lower_flags = (lower_flags & O_ACCMODE) | O_RDWR;
if (file->f_flags & O_APPEND)
lower_flags &= ~O_APPEND;
lower_mnt = ecryptfs_dentry_to_lower_mnt(ecryptfs_dentry);
/* Corresponding fput() in ecryptfs_release() */
if ((rc = ecryptfs_open_lower_file(&lower_file, lower_dentry, lower_mnt,
lower_flags))) {
ecryptfs_printk(KERN_ERR, "Error opening lower file\n");
goto out_puts;
}
ecryptfs_set_file_lower(file, lower_file);
/* Isn't this check the same as the one in lookup? */
lower_inode = lower_dentry->d_inode;
if (S_ISDIR(ecryptfs_dentry->d_inode->i_mode)) {
ecryptfs_printk(KERN_DEBUG, "This is a directory\n");
ECRYPTFS_CLEAR_FLAG(crypt_stat->flags, ECRYPTFS_ENCRYPTED);
rc = 0;
goto out;
}
mutex_lock(&crypt_stat->cs_mutex);
if (i_size_read(lower_inode) < ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE) {
if (!(mount_crypt_stat->flags
& ECRYPTFS_PLAINTEXT_PASSTHROUGH_ENABLED)) {
rc = -EIO;
printk(KERN_WARNING "Attempt to read file that is "
"not in a valid eCryptfs format, and plaintext "
"passthrough mode is not enabled; returning "
"-EIO\n");
mutex_unlock(&crypt_stat->cs_mutex);
goto out_puts;
}
crypt_stat->flags &= ~(ECRYPTFS_ENCRYPTED);
rc = 0;
mutex_unlock(&crypt_stat->cs_mutex);
goto out;
} else if (!ECRYPTFS_CHECK_FLAG(crypt_stat->flags,
ECRYPTFS_POLICY_APPLIED)
|| !ECRYPTFS_CHECK_FLAG(crypt_stat->flags,
ECRYPTFS_KEY_VALID)) {
rc = ecryptfs_read_headers(ecryptfs_dentry, lower_file);
if (rc) {
ecryptfs_printk(KERN_DEBUG,
"Valid headers not found\n");
if (!(mount_crypt_stat->flags
& ECRYPTFS_PLAINTEXT_PASSTHROUGH_ENABLED)) {
rc = -EIO;
printk(KERN_WARNING "Attempt to read file that "
"is not in a valid eCryptfs format, "
"and plaintext passthrough mode is not "
"enabled; returning -EIO\n");
mutex_unlock(&crypt_stat->cs_mutex);
goto out_puts;
}
ECRYPTFS_CLEAR_FLAG(crypt_stat->flags,
ECRYPTFS_ENCRYPTED);
rc = 0;
mutex_unlock(&crypt_stat->cs_mutex);
goto out;
}
}
mutex_unlock(&crypt_stat->cs_mutex);
ecryptfs_printk(KERN_DEBUG, "inode w/ addr = [0x%p], i_ino = [0x%.16x] "
"size: [0x%.16x]\n", inode, inode->i_ino,
i_size_read(inode));
ecryptfs_set_file_lower(file, lower_file);
goto out;
out_puts:
mntput(lower_mnt);
dput(lower_dentry);
kmem_cache_free(ecryptfs_file_info_cache,
ecryptfs_file_to_private(file));
out:
return rc;
}
static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt, int flags, struct file *f)
{
struct inode *inode;
static LIST_HEAD(kill_list);
int error;
f->f_flags = flags;
f->f_mode = (flags+1) & O_ACCMODE;
inode = dentry->d_inode;
if (f->f_mode & FMODE_WRITE) {
error = get_write_access(inode);
if (error)
goto cleanup_file;
}
f->f_dentry = dentry;
f->f_vfsmnt = mnt;
f->f_pos = 0;
f->f_reada = 0;
f->f_op = fops_get(inode->i_fop);
file_move(f, &inode->i_sb->s_files);
/* preallocate kiobuf for O_DIRECT */
f->f_iobuf = NULL;
f->f_iobuf_lock = 0;
if (f->f_flags & O_DIRECT) {
error = alloc_kiovec(1, &f->f_iobuf);
if (error)
goto cleanup_all;
}
if (f->f_op && f->f_op->open) {
error = f->f_op->open(inode,f);
if (error)
goto cleanup_all;
}
f->f_flags &= ~(O_CREAT | O_EXCL | O_NOCTTY | O_TRUNC);
/* NB: we're sure to have correct a_ops only after f_op->open */
if (f->f_flags & O_DIRECT) {
if (!inode->i_mapping || !inode->i_mapping->a_ops ||
!(inode->i_mapping->a_ops->direct_IO ||
inode->i_mapping->a_ops->direct_sector_IO)) {
fput(f);
f = ERR_PTR(-EINVAL);
}
}
return f;
cleanup_all:
if (f->f_iobuf)
free_kiovec(1, &f->f_iobuf);
fops_put(f->f_op);
if (f->f_mode & FMODE_WRITE)
put_write_access(inode);
file_move(f, &kill_list); /* out of the way.. */
f->f_dentry = NULL;
f->f_vfsmnt = NULL;
cleanup_file:
put_filp(f);
dput(dentry);
mntput(mnt);
return ERR_PTR(error);
}
/*
* Object creation/destruction.
*/
LinuxSystemRoot* newLinuxSystemRoot(void)
{
LinuxSystemRoot* object;
unsigned m_seq = 0;
object = talpa_alloc(sizeof(template_LinuxSystemRoot));
if ( object )
{
struct task_struct* inittask;
memcpy(object, &template_LinuxSystemRoot, sizeof(template_LinuxSystemRoot));
object->i_ISystemRoot.object = object;
talpa_tasklist_lock();
inittask = talpa_find_task_by_pid(1);
talpa_tasklist_unlock();
if ( inittask )
{
struct fs_struct *init_fs;
struct vfsmount *rootmnt;
task_lock(inittask);
init_fs = inittask->fs;
if ( init_fs )
{
#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,30)
#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,36)
spin_lock(&init_fs->lock);
#else
write_lock(&init_fs->lock);
#endif
init_fs->users++;
#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,36)
spin_unlock(&init_fs->lock);
#else
write_unlock(&init_fs->lock);
#endif
#else
atomic_inc(&init_fs->count);
#endif
}
task_unlock(inittask);
if ( init_fs )
{
#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,36)
spin_lock(&init_fs->lock);
#else
write_lock(&init_fs->lock);
#endif
restart_mnt:
talpa_vfsmount_lock(&m_seq); // locks dcache_lock on 2.4
for (rootmnt = talpa_fs_mnt(init_fs); rootmnt != getParent(rootmnt); rootmnt = getParent(rootmnt));
object->mMnt = mntget(rootmnt);
object->mDentry = dget(rootmnt->mnt_root);
if (talpa_vfsmount_unlock(&m_seq)) // unlocks dcache_lock on 2.4
{
goto restart_mnt;
}
#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,30)
init_fs->users--;
#else
atomic_dec(&init_fs->count);
#endif
#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,36)
spin_unlock(&init_fs->lock);
#else
write_unlock(&init_fs->lock);
#endif
}
}
if ( !object->mMnt || !object->mDentry )
{
if ( object->mMnt )
{
mntput(object->mMnt);
}
if ( object->mDentry )
{
dput(object->mDentry);
}
talpa_free(object);
return NULL;
}
}
return object;
}
/* This must be called with the super block already locked. */
int unionfs_ioctl_delbranch(struct super_block *sb, unsigned long arg)
{
struct dentry *hidden_dentry;
struct inode *hidden_inode;
struct super_block *hidden_sb;
struct vfsmount *hidden_mnt;
struct dentry *root_dentry;
struct inode *root_inode;
int err = 0;
int pmindex, i, gen;
print_entry("branch = %lu ", arg);
lock_dentry(sb->s_root);
err = -EBUSY;
if (sbmax(sb) == 1)
goto out;
err = -EINVAL;
if (arg < 0 || arg > stopd(sb)->b_end)
goto out;
err = -EBUSY;
if (branch_count(sb, arg))
goto out;
if ((err = newputmap(sb)))
goto out;
pmindex = stopd(sb)->usi_lastputmap;
pmindex -= stopd(sb)->usi_firstputmap;
atomic_inc(&stopd(sb)->usi_generation);
gen = atomic_read(&stopd(sb)->usi_generation);
root_dentry = sb->s_root;
root_inode = sb->s_root->d_inode;
hidden_dentry = dtohd_index(root_dentry, arg);
hidden_mnt = stohiddenmnt_index(sb, arg);
hidden_inode = itohi_index(root_inode, arg);
hidden_sb = stohs_index(sb, arg);
DPUT(hidden_dentry);
iput(hidden_inode);
mntput(hidden_mnt);
for (i = arg; i <= (sbend(sb) - 1); i++) {
set_branch_count(sb, i, branch_count(sb, i + 1));
set_stohiddenmnt_index(sb, i, stohiddenmnt_index(sb, i + 1));
set_stohs_index(sb, i, stohs_index(sb, i + 1));
set_branchperms(sb, i, branchperms(sb, i + 1));
set_dtohd_index(root_dentry, i,
dtohd_index(root_dentry, i + 1));
set_itohi_index(root_inode, i, itohi_index(root_inode, i + 1));
stopd(sb)->usi_putmaps[pmindex]->map[i + 1] = i;
}
set_dtohd_index(root_dentry, sbend(sb), NULL);
set_itohi_index(root_inode, sbend(sb), NULL);
set_stohiddenmnt_index(sb, sbend(sb), NULL);
set_stohs_index(sb, sbend(sb), NULL);
stopd(sb)->b_end--;
set_dbend(root_dentry, dbend(root_dentry) - 1);
dtopd(root_dentry)->udi_bcount--;
itopd(root_inode)->b_end--;
atomic_set(&dtopd(root_dentry)->udi_generation, gen);
atomic_set(&itopd(root_inode)->uii_generation, gen);
fixputmaps(sb);
/* This doesn't open a file, so we might have to free the map here. */
if (atomic_read(&stopd(sb)->usi_putmaps[pmindex]->count) == 0) {
KFREE(stopd(sb)->usi_putmaps[pmindex]);
stopd(sb)->usi_putmaps[pmindex] = NULL;
}
out:
unlock_dentry(sb->s_root);
print_exit_status(err);
return err;
}
static int ovl_fill_super(struct super_block *sb, void *data, int silent)
{
struct path lowerpath;
struct path upperpath;
struct path workpath;
struct inode *root_inode;
struct dentry *root_dentry;
struct ovl_entry *oe;
struct ovl_fs *ufs;
struct kstatfs statfs;
int err;
err = -ENOMEM;
ufs = kmalloc(sizeof(struct ovl_fs), GFP_KERNEL);
if (!ufs)
goto out;
err = ovl_parse_opt((char *) data, &ufs->config);
if (err)
goto out_free_ufs;
/* FIXME: workdir is not needed for a R/O mount */
err = -EINVAL;
if (!ufs->config.upperdir || !ufs->config.lowerdir ||
!ufs->config.workdir) {
pr_err("overlayfs: missing upperdir or lowerdir or workdir\n");
goto out_free_config;
}
oe = ovl_alloc_entry();
if (oe == NULL)
goto out_free_config;
err = ovl_mount_dir(ufs->config.upperdir, &upperpath);
if (err)
goto out_free_oe;
err = ovl_mount_dir(ufs->config.lowerdir, &lowerpath);
if (err)
goto out_put_upperpath;
err = ovl_mount_dir(ufs->config.workdir, &workpath);
if (err)
goto out_put_lowerpath;
err = -EINVAL;
if (!S_ISDIR(upperpath.dentry->d_inode->i_mode) ||
!S_ISDIR(lowerpath.dentry->d_inode->i_mode) ||
!S_ISDIR(workpath.dentry->d_inode->i_mode)) {
pr_err("overlayfs: upperdir or lowerdir or workdir not a directory\n");
goto out_put_workpath;
}
if (upperpath.mnt != workpath.mnt) {
pr_err("overlayfs: workdir and upperdir must reside under the same mount\n");
goto out_put_workpath;
}
if (upperpath.dentry == workpath.dentry ||
d_ancestor(upperpath.dentry, workpath.dentry) ||
d_ancestor(workpath.dentry, upperpath.dentry)) {
pr_err("overlayfs: workdir and upperdir must be separate subtrees\n");
goto out_put_workpath;
}
err = vfs_statfs(&lowerpath, &statfs);
if (err) {
pr_err("overlayfs: statfs failed on lowerpath\n");
goto out_put_workpath;
}
ufs->lower_namelen = statfs.f_namelen;
sb->s_stack_depth = max(upperpath.mnt->mnt_sb->s_stack_depth,
lowerpath.mnt->mnt_sb->s_stack_depth) + 1;
err = -EINVAL;
if (sb->s_stack_depth > FILESYSTEM_MAX_STACK_DEPTH) {
pr_err("overlayfs: maximum fs stacking depth exceeded\n");
goto out_put_lowerpath;
}
ufs->upper_mnt = clone_private_mount(&upperpath);
err = PTR_ERR(ufs->upper_mnt);
if (IS_ERR(ufs->upper_mnt)) {
pr_err("overlayfs: failed to clone upperpath\n");
goto out_put_workpath;
}
ufs->lower_mnt = clone_private_mount(&lowerpath);
err = PTR_ERR(ufs->lower_mnt);
if (IS_ERR(ufs->lower_mnt)) {
pr_err("overlayfs: failed to clone lowerpath\n");
goto out_put_upper_mnt;
}
ufs->workdir = ovl_workdir_create(ufs->upper_mnt, workpath.dentry);
err = PTR_ERR(ufs->workdir);
if (IS_ERR(ufs->workdir)) {
pr_err("overlayfs: failed to create directory %s/%s\n",
ufs->config.workdir, OVL_WORKDIR_NAME);
goto out_put_lower_mnt;
}
/*
* Make lower_mnt R/O. That way fchmod/fchown on lower file
* will fail instead of modifying lower fs.
*/
ufs->lower_mnt->mnt_flags |= MNT_READONLY;
/* If the upper fs is r/o, we mark overlayfs r/o too */
if (ufs->upper_mnt->mnt_sb->s_flags & MS_RDONLY)
sb->s_flags |= MS_RDONLY;
err = -ENOMEM;
root_inode = ovl_new_inode(sb, S_IFDIR, oe);
if (!root_inode)
goto out_put_workdir;
root_dentry = d_make_root(root_inode);
if (!root_dentry)
goto out_put_workdir;
mntput(upperpath.mnt);
mntput(lowerpath.mnt);
path_put(&workpath);
oe->__upperdentry = dget(upperpath.dentry);
oe->lowerdentry = lowerpath.dentry;
root_dentry->d_fsdata = oe;
root_dentry->d_op = &ovl_dentry_operations;
sb->s_magic = OVERLAYFS_SUPER_MAGIC;
sb->s_op = &ovl_super_operations;
sb->s_root = root_dentry;
sb->s_fs_info = ufs;
return 0;
out_put_workdir:
dput(ufs->workdir);
out_put_lower_mnt:
mntput(ufs->lower_mnt);
out_put_upper_mnt:
mntput(ufs->upper_mnt);
out_put_workpath:
path_put(&workpath);
out_put_lowerpath:
path_put(&lowerpath);
out_put_upperpath:
path_put(&upperpath);
out_free_oe:
kfree(oe);
out_free_config:
kfree(ufs->config.lowerdir);
kfree(ufs->config.upperdir);
kfree(ufs->config.workdir);
out_free_ufs:
kfree(ufs);
out:
return err;
}
/**
* ecryptfs_lookup_and_interpose_lower - Perform a lookup
*/
int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry,
struct dentry *lower_dentry,
struct inode *ecryptfs_dir_inode,
struct nameidata *ecryptfs_nd)
{
struct dentry *lower_dir_dentry;
struct vfsmount *lower_mnt;
struct inode *lower_inode;
struct ecryptfs_mount_crypt_stat *mount_crypt_stat;
struct ecryptfs_crypt_stat *crypt_stat;
char *page_virt = NULL;
u64 file_size;
int rc = 0;
lower_dir_dentry = lower_dentry->d_parent;
lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt(
ecryptfs_dentry->d_parent));
lower_inode = lower_dentry->d_inode;
fsstack_copy_attr_atime(ecryptfs_dir_inode, lower_dir_dentry->d_inode);
BUG_ON(!atomic_read(&lower_dentry->d_count));
ecryptfs_set_dentry_private(ecryptfs_dentry,
kmem_cache_alloc(ecryptfs_dentry_info_cache,
GFP_KERNEL));
if (!ecryptfs_dentry_to_private(ecryptfs_dentry)) {
rc = -ENOMEM;
printk(KERN_ERR "%s: Out of memory whilst attempting "
"to allocate ecryptfs_dentry_info struct\n",
__func__);
goto out_put;
}
ecryptfs_set_dentry_lower(ecryptfs_dentry, lower_dentry);
ecryptfs_set_dentry_lower_mnt(ecryptfs_dentry, lower_mnt);
if (!lower_dentry->d_inode) {
/* We want to add because we couldn't find in lower */
d_add(ecryptfs_dentry, NULL);
goto out;
}
rc = ecryptfs_interpose(lower_dentry, ecryptfs_dentry,
ecryptfs_dir_inode->i_sb, 1);
if (rc) {
printk(KERN_ERR "%s: Error interposing; rc = [%d]\n",
__func__, rc);
goto out;
}
if (S_ISDIR(lower_inode->i_mode))
goto out;
if (S_ISLNK(lower_inode->i_mode))
goto out;
if (special_file(lower_inode->i_mode))
goto out;
if (!ecryptfs_nd)
goto out;
/* Released in this function */
page_virt = kmem_cache_zalloc(ecryptfs_header_cache_2, GFP_USER);
if (!page_virt) {
printk(KERN_ERR "%s: Cannot kmem_cache_zalloc() a page\n",
__func__);
rc = -ENOMEM;
goto out;
}
if (!ecryptfs_inode_to_private(ecryptfs_dentry->d_inode)->lower_file) {
rc = ecryptfs_init_persistent_file(ecryptfs_dentry);
if (rc) {
printk(KERN_ERR "%s: Error attempting to initialize "
"the persistent file for the dentry with name "
"[%s]; rc = [%d]\n", __func__,
ecryptfs_dentry->d_name.name, rc);
goto out_free_kmem;
}
}
crypt_stat = &ecryptfs_inode_to_private(
ecryptfs_dentry->d_inode)->crypt_stat;
/* TODO: lock for crypt_stat comparison */
if (!(crypt_stat->flags & ECRYPTFS_POLICY_APPLIED))
ecryptfs_set_default_sizes(crypt_stat);
rc = ecryptfs_read_and_validate_header_region(page_virt,
ecryptfs_dentry->d_inode);
if (rc) {
rc = ecryptfs_read_and_validate_xattr_region(page_virt,
ecryptfs_dentry);
if (rc) {
rc = 0;
goto out_free_kmem;
}
crypt_stat->flags |= ECRYPTFS_METADATA_IN_XATTR;
}
mount_crypt_stat = &ecryptfs_superblock_to_private(
ecryptfs_dentry->d_sb)->mount_crypt_stat;
if (mount_crypt_stat->flags & ECRYPTFS_ENCRYPTED_VIEW_ENABLED) {
if (crypt_stat->flags & ECRYPTFS_METADATA_IN_XATTR)
file_size = (crypt_stat->num_header_bytes_at_front
+ i_size_read(lower_dentry->d_inode));
else
file_size = i_size_read(lower_dentry->d_inode);
} else {
file_size = get_unaligned_be64(page_virt);
}
i_size_write(ecryptfs_dentry->d_inode, (loff_t)file_size);
out_free_kmem:
kmem_cache_free(ecryptfs_header_cache_2, page_virt);
goto out;
out_put:
dput(lower_dentry);
mntput(lower_mnt);
d_drop(ecryptfs_dentry);
out:
return rc;
}
void redirect_namei(struct nameidata *dest, const struct nameidata *src) {
mntget(src->mnt); dget(src->dentry);
mntput(dest->mnt); dput(dest->dentry);
*dest = *src;
}
static int dazukofs_fill_super(struct super_block *sb, void *data, int silent)
{
struct dazukofs_sb_info *sbi;
struct dentry *root;
static const struct qstr name = { .name = "/", .len = 1 };
struct dazukofs_dentry_info *di;
sbi = kmem_cache_zalloc(dazukofs_sb_info_cachep, GFP_KERNEL);
if (!sbi)
return -ENOMEM;
sb->s_op = &dazukofs_sops;
root = d_alloc(NULL, &name);
if (!root) {
kmem_cache_free(dazukofs_sb_info_cachep, sbi);
return -ENOMEM;
}
sb->s_root = root;
sb->s_root->d_op = &dazukofs_dops;
sb->s_root->d_sb = sb;
sb->s_root->d_parent = sb->s_root;
di = kmem_cache_zalloc(dazukofs_dentry_info_cachep, GFP_KERNEL);
if (!di) {
kmem_cache_free(dazukofs_sb_info_cachep, sbi);
dput(sb->s_root);
return -ENOMEM;
}
set_dentry_private(sb->s_root, di);
set_sb_private(sb, sbi);
return 0;
}
static int dazukofs_read_super(struct super_block *sb, const char *dev_name)
{
struct nameidata nd;
struct dentry *lower_root;
struct vfsmount *lower_mnt;
int err;
memset(&nd, 0, sizeof(struct nameidata));
#if LINUX_VERSION_CODE >= KERNEL_VERSION(2, 6, 39)
err = kern_path(dev_name, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &nd.path);
#else
err = path_lookup(dev_name, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &nd);
#endif
if (err)
return err;
lower_root = dget(nd.path.dentry);
lower_mnt = mntget(nd.path.mnt);
if (IS_ERR(lower_root)) {
err = PTR_ERR(lower_root);
goto out_put;
}
if (!lower_root->d_inode) {
err = -ENOENT;
goto out_put;
}
if (!S_ISDIR(lower_root->d_inode->i_mode)) {
err = -EINVAL;
goto out_put;
}
set_lower_sb(sb, lower_root->d_sb);
sb->s_maxbytes = lower_root->d_sb->s_maxbytes;
set_lower_dentry(sb->s_root, lower_root, lower_mnt);
err = dazukofs_interpose(lower_root, sb->s_root, sb, 0);
if (err)
goto out_put;
goto out;
out_put:
dput(lower_root);
mntput(lower_mnt);
out:
path_put(&nd.path);
return err;
}
// FIXME!
#if LINUX_VERSION_CODE >= KERNEL_VERSION(2, 6, 39)
static struct dentry* dazukofs_get_sb(struct file_system_type *fs_type, int flags,
const char *dev_name, void *data)
{
return mount_nodev(fs_type, flags, data, dazukofs_fill_super);
}
#else
static int dazukofs_get_sb(struct file_system_type *fs_type, int flags,
const char *dev_name, void *data,
struct vfsmount *mnt)
{
struct super_block *sb;
int err;
err = get_sb_nodev(fs_type, flags, data, dazukofs_fill_super, mnt);
if (err)
goto out;
sb = mnt->mnt_sb;
err = dazukofs_parse_mount_options(data, sb);
if (err)
goto out_abort;
err = dazukofs_read_super(sb, dev_name);
if (err)
goto out_abort;
goto out;
out_abort:
dput(sb->s_root);
up_write(&sb->s_umount);
deactivate_super(sb);
out:
return err;
}
#endif
static void init_once(void *data)
{
struct dazukofs_inode_info *inode_info =
(struct dazukofs_inode_info *)data;
memset(inode_info, 0, sizeof(struct dazukofs_inode_info));
inode_init_once(&(inode_info->vfs_inode));
}
static void destroy_caches(void)
{
if (dazukofs_inode_info_cachep) {
kmem_cache_destroy(dazukofs_inode_info_cachep);
dazukofs_inode_info_cachep = NULL;
}
if (dazukofs_sb_info_cachep) {
kmem_cache_destroy(dazukofs_sb_info_cachep);
dazukofs_sb_info_cachep = NULL;
}
if (dazukofs_dentry_info_cachep) {
kmem_cache_destroy(dazukofs_dentry_info_cachep);
dazukofs_dentry_info_cachep = NULL;
}
if (dazukofs_file_info_cachep) {
kmem_cache_destroy(dazukofs_file_info_cachep);
dazukofs_file_info_cachep = NULL;
}
}
static int init_caches(void)
{
dazukofs_inode_info_cachep =
kmem_cache_create("dazukofs_inode_info_cache",
sizeof(struct dazukofs_inode_info), 0,
SLAB_HWCACHE_ALIGN,
init_once);
if (!dazukofs_inode_info_cachep)
goto out_nomem;
dazukofs_sb_info_cachep =
kmem_cache_create("dazukofs_sb_info_cache",
sizeof(struct dazukofs_sb_info), 0,
SLAB_HWCACHE_ALIGN,
NULL);
if (!dazukofs_sb_info_cachep)
goto out_nomem;
dazukofs_dentry_info_cachep =
kmem_cache_create("dazukofs_dentry_info_cache",
sizeof(struct dazukofs_dentry_info), 0,
SLAB_HWCACHE_ALIGN,
NULL);
if (!dazukofs_dentry_info_cachep)
goto out_nomem;
dazukofs_file_info_cachep =
kmem_cache_create("dazukofs_file_info_cache",
sizeof(struct dazukofs_file_info), 0,
SLAB_HWCACHE_ALIGN,
NULL);
if (!dazukofs_file_info_cachep)
goto out_nomem;
return 0;
out_nomem:
destroy_caches();
return -ENOMEM;
}
/**
* ecryptfs_lookup_and_interpose_lower - Perform a lookup
*/
int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry,
struct dentry *lower_dentry,
struct inode *ecryptfs_dir_inode)
{
struct dentry *lower_dir_dentry;
struct vfsmount *lower_mnt;
struct inode *lower_inode;
struct ecryptfs_crypt_stat *crypt_stat;
char *page_virt = NULL;
int put_lower = 0, rc = 0;
lower_dir_dentry = lower_dentry->d_parent;
lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt(
ecryptfs_dentry->d_parent));
lower_inode = lower_dentry->d_inode;
fsstack_copy_attr_atime(ecryptfs_dir_inode, lower_dir_dentry->d_inode);
BUG_ON(!lower_dentry->d_count);
ecryptfs_set_dentry_private(ecryptfs_dentry,
kmem_cache_alloc(ecryptfs_dentry_info_cache,
GFP_KERNEL));
if (!ecryptfs_dentry_to_private(ecryptfs_dentry)) {
rc = -ENOMEM;
printk(KERN_ERR "%s: Out of memory whilst attempting "
"to allocate ecryptfs_dentry_info struct\n",
__func__);
goto out_put;
}
ecryptfs_set_dentry_lower(ecryptfs_dentry, lower_dentry);
ecryptfs_set_dentry_lower_mnt(ecryptfs_dentry, lower_mnt);
if (!lower_dentry->d_inode) {
/* We want to add because we couldn't find in lower */
d_add(ecryptfs_dentry, NULL);
goto out;
}
rc = ecryptfs_interpose(lower_dentry, ecryptfs_dentry,
ecryptfs_dir_inode->i_sb,
ECRYPTFS_INTERPOSE_FLAG_D_ADD);
if (rc) {
printk(KERN_ERR "%s: Error interposing; rc = [%d]\n",
__func__, rc);
goto out;
}
if (S_ISDIR(lower_inode->i_mode))
goto out;
if (S_ISLNK(lower_inode->i_mode))
goto out;
if (special_file(lower_inode->i_mode))
goto out;
/* Released in this function */
page_virt = kmem_cache_zalloc(ecryptfs_header_cache_2, GFP_USER);
if (!page_virt) {
printk(KERN_ERR "%s: Cannot kmem_cache_zalloc() a page\n",
__func__);
rc = -ENOMEM;
goto out;
}
rc = ecryptfs_get_lower_file(ecryptfs_dentry);
if (rc) {
printk(KERN_ERR "%s: Error attempting to initialize "
"the lower file for the dentry with name "
"[%s]; rc = [%d]\n", __func__,
ecryptfs_dentry->d_name.name, rc);
goto out_free_kmem;
}
put_lower = 1;
crypt_stat = &ecryptfs_inode_to_private(
ecryptfs_dentry->d_inode)->crypt_stat;
/* TODO: lock for crypt_stat comparison */
if (!(crypt_stat->flags & ECRYPTFS_POLICY_APPLIED))
ecryptfs_set_default_sizes(crypt_stat);
rc = ecryptfs_read_and_validate_header_region(page_virt,
ecryptfs_dentry->d_inode);
if (rc) {
memset(page_virt, 0, PAGE_CACHE_SIZE);
rc = ecryptfs_read_and_validate_xattr_region(page_virt,
ecryptfs_dentry);
if (rc) {
rc = 0;
goto out_free_kmem;
}
crypt_stat->flags |= ECRYPTFS_METADATA_IN_XATTR;
}
ecryptfs_i_size_init(page_virt, ecryptfs_dentry->d_inode);
out_free_kmem:
kmem_cache_free(ecryptfs_header_cache_2, page_virt);
goto out;
out_put:
dput(lower_dentry);
mntput(lower_mnt);
d_drop(ecryptfs_dentry);
out:
if (put_lower)
ecryptfs_put_lower_file(ecryptfs_dentry->d_inode);
return rc;
}
static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt,
int flags, struct file *f,
int (*open)(struct inode *, struct file *))
{
struct inode *inode;
int error;
f->f_flags = flags;
f->f_mode = ((flags+1) & O_ACCMODE) | FMODE_LSEEK |
FMODE_PREAD | FMODE_PWRITE;
inode = dentry->d_inode;
if (f->f_mode & FMODE_WRITE) {
error = get_write_access(inode);
if (error)
goto cleanup_file;
}
f->f_mapping = inode->i_mapping;
f->f_path.dentry = dentry;
f->f_path.mnt = mnt;
f->f_pos = 0;
f->f_op = fops_get(inode->i_fop);
file_move(f, &inode->i_sb->s_files);
if (!open && f->f_op)
open = f->f_op->open;
if (open) {
error = open(inode, f);
if (error)
goto cleanup_all;
}
f->f_flags &= ~(O_CREAT | O_EXCL | O_NOCTTY | O_TRUNC);
file_ra_state_init(&f->f_ra, f->f_mapping->host->i_mapping);
/* NB: we're sure to have correct a_ops only after f_op->open */
if (f->f_flags & O_DIRECT) {
if (!f->f_mapping->a_ops ||
((!f->f_mapping->a_ops->direct_IO) &&
(!f->f_mapping->a_ops->get_xip_page))) {
fput(f);
f = ERR_PTR(-EINVAL);
}
}
return f;
cleanup_all:
fops_put(f->f_op);
if (f->f_mode & FMODE_WRITE)
put_write_access(inode);
file_kill(f);
f->f_path.dentry = NULL;
f->f_path.mnt = NULL;
cleanup_file:
put_filp(f);
dput(dentry);
mntput(mnt);
return ERR_PTR(error);
}
/*mds still need lov setup here*/
static int mds_cmd_setup(struct obd_device *obd, struct lustre_cfg *lcfg)
{
struct mds_obd *mds = &obd->u.mds;
struct lvfs_run_ctxt saved;
const char *dev;
struct vfsmount *mnt;
struct lustre_sb_info *lsi;
struct lustre_mount_info *lmi;
struct dentry *dentry;
int rc = 0;
ENTRY;
CDEBUG(D_INFO, "obd %s setup \n", obd->obd_name);
if (strncmp(obd->obd_name, MDD_OBD_NAME, strlen(MDD_OBD_NAME)))
RETURN(0);
if (lcfg->lcfg_bufcount < 5) {
CERROR("invalid arg for setup %s\n", MDD_OBD_NAME);
RETURN(-EINVAL);
}
dev = lustre_cfg_string(lcfg, 4);
lmi = server_get_mount(dev);
LASSERT(lmi != NULL);
lsi = s2lsi(lmi->lmi_sb);
mnt = lmi->lmi_mnt;
/* FIXME: MDD LOV initialize objects.
* we need only lmi here but not get mount
* OSD did mount already, so put mount back
*/
cfs_atomic_dec(&lsi->lsi_mounts);
mntput(mnt);
cfs_init_rwsem(&mds->mds_notify_lock);
obd->obd_fsops = fsfilt_get_ops(MT_STR(lsi->lsi_ldd));
mds_init_ctxt(obd, mnt);
push_ctxt(&saved, &obd->obd_lvfs_ctxt, NULL);
dentry = simple_mkdir(current->fs->pwd, mnt, "OBJECTS", 0777, 1);
if (IS_ERR(dentry)) {
rc = PTR_ERR(dentry);
CERROR("cannot create OBJECTS directory: rc = %d\n", rc);
GOTO(err_putfs, rc);
}
mds->mds_objects_dir = dentry;
dentry = ll_lookup_one_len("__iopen__", current->fs->pwd,
strlen("__iopen__"));
if (IS_ERR(dentry)) {
rc = PTR_ERR(dentry);
CERROR("cannot lookup __iopen__ directory: rc = %d\n", rc);
GOTO(err_objects, rc);
}
mds->mds_fid_de = dentry;
if (!dentry->d_inode || is_bad_inode(dentry->d_inode)) {
rc = -ENOENT;
CERROR("__iopen__ directory has no inode? rc = %d\n", rc);
GOTO(err_fid, rc);
}
rc = mds_lov_init_objids(obd);
if (rc != 0) {
CERROR("cannot init lov objid rc = %d\n", rc);
GOTO(err_fid, rc );
}
rc = mds_lov_presetup(mds, lcfg);
if (rc < 0)
GOTO(err_objects, rc);
/* Don't wait for mds_postrecov trying to clear orphans */
obd->obd_async_recov = 1;
rc = mds_postsetup(obd);
/* Bug 11557 - allow async abort_recov start
FIXME can remove most of this obd_async_recov plumbing
obd->obd_async_recov = 0;
*/
if (rc)
GOTO(err_objects, rc);
err_pop:
pop_ctxt(&saved, &obd->obd_lvfs_ctxt, NULL);
RETURN(rc);
err_fid:
dput(mds->mds_fid_de);
err_objects:
dput(mds->mds_objects_dir);
err_putfs:
fsfilt_put_ops(obd->obd_fsops);
goto err_pop;
}
__be32
nfsd_lookup_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp,
const char *name, unsigned int len,
struct svc_export **exp_ret, struct dentry **dentry_ret)
{
struct svc_export *exp;
struct dentry *dparent;
struct dentry *dentry;
__be32 err;
int host_err;
dprintk("nfsd: nfsd_lookup(fh %s, %.*s)\n", SVCFH_fmt(fhp), len,name);
/* Obtain dentry and export. */
err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_EXEC);
if (err)
return err;
dparent = fhp->fh_dentry;
exp = fhp->fh_export;
exp_get(exp);
/* Lookup the name, but don't follow links */
if (isdotent(name, len)) {
if (len==1)
dentry = dget(dparent);
else if (dparent != exp->ex_path.dentry)
dentry = dget_parent(dparent);
else if (!EX_NOHIDE(exp))
dentry = dget(dparent); /* .. == . just like at / */
else {
/* checking mountpoint crossing is very different when stepping up */
struct svc_export *exp2 = NULL;
struct dentry *dp;
struct vfsmount *mnt = mntget(exp->ex_path.mnt);
dentry = dget(dparent);
while(dentry == mnt->mnt_root && follow_up(&mnt, &dentry))
;
dp = dget_parent(dentry);
dput(dentry);
dentry = dp;
exp2 = rqst_exp_parent(rqstp, mnt, dentry);
if (PTR_ERR(exp2) == -ENOENT) {
dput(dentry);
dentry = dget(dparent);
} else if (IS_ERR(exp2)) {
host_err = PTR_ERR(exp2);
dput(dentry);
mntput(mnt);
goto out_nfserr;
} else {
exp_put(exp);
exp = exp2;
}
mntput(mnt);
}
} else {
fh_lock(fhp);
dentry = lookup_one_len(name, dparent, len);
host_err = PTR_ERR(dentry);
if (IS_ERR(dentry))
goto out_nfserr;
/*
* check if we have crossed a mount point ...
*/
if (d_mountpoint(dentry)) {
if ((host_err = nfsd_cross_mnt(rqstp, &dentry, &exp))) {
dput(dentry);
goto out_nfserr;
}
}
}
*dentry_ret = dentry;
*exp_ret = exp;
return 0;
out_nfserr:
exp_put(exp);
return nfserrno(host_err);
}
long do_sys_open_by_handle(int mountdirfd, struct file_handle __user * ufh, int open_flag)
{
int fd;
long retval = 0;
struct file *filp;
struct dentry *dentry;
struct file_handle f_handle;
struct vfsmount *mnt = NULL;
struct file_handle *handle = NULL;
/* can't use O_CREATE with open_by_handle */
if(open_flag & O_CREAT)
{
retval = -EINVAL;
goto out_err;
}
if(copy_from_user(&f_handle, ufh, sizeof(struct file_handle)))
{
retval = -EFAULT;
goto out_err;
}
if((f_handle.handle_size > MAX_HANDLE_SZ) || (f_handle.handle_size <= 0))
{
retval = -EINVAL;
goto out_err;
}
handle = kmalloc(sizeof(struct file_handle) + f_handle.handle_size, GFP_KERNEL);
if(!handle)
{
retval = -ENOMEM;
goto out_err;
}
/* copy the full handle */
if(copy_from_user(handle, ufh, sizeof(struct file_handle) + f_handle.handle_size))
{
retval = -EFAULT;
goto out_handle;
}
dentry = handle_to_dentry(mountdirfd, handle, &mnt);
if(IS_ERR(dentry))
{
retval = PTR_ERR(dentry);
goto out_handle;
}
retval = may_handle_open(dentry, open_flag);
if(retval)
goto out_handle;
fd = get_unused_fd();
if(fd < 0)
{
retval = fd;
goto out_dentry;
}
filp = dentry_open(dentry, mnt, open_flag);
if(IS_ERR(filp))
{
put_unused_fd(fd);
retval = PTR_ERR(filp);
}
else
{
retval = fd;
fsnotify_open(filp->f_dentry);
fd_install(fd, filp);
}
kfree(handle);
return retval;
out_dentry:
dput(dentry);
mntput(mnt);
out_handle:
kfree(handle);
out_err:
return retval;
}
static void __exit capifs_exit(void)
{
unregister_filesystem(&capifs_fs_type);
mntput(capifs_mnt);
}
static void __exit exit_devpts_fs(void)
{
unregister_filesystem(&devpts_fs_type);
mntput(devpts_mnt);
}
static int ovl_fill_super(struct super_block *sb, void *data, int silent)
{
struct path lowerpath;
struct path upperpath;
struct inode *root_inode;
struct dentry *root_dentry;
struct ovl_entry *oe;
struct ovl_fs *ufs;
struct kstatfs statfs;
int err;
err = -ENOMEM;
ufs = kmalloc(sizeof(struct ovl_fs), GFP_KERNEL);
if (!ufs)
goto out;
err = ovl_parse_opt((char *) data, &ufs->config);
if (err)
goto out_free_ufs;
err = -EINVAL;
if (!ufs->config.upperdir || !ufs->config.lowerdir) {
pr_err("overlayfs: missing upperdir or lowerdir\n");
goto out_free_config;
}
oe = ovl_alloc_entry();
if (oe == NULL)
goto out_free_config;
err = kern_path(ufs->config.upperdir, LOOKUP_FOLLOW, &upperpath);
if (err)
goto out_free_oe;
err = kern_path(ufs->config.lowerdir, LOOKUP_FOLLOW, &lowerpath);
if (err)
goto out_put_upperpath;
err = -ENOTDIR;
if (!S_ISDIR(upperpath.dentry->d_inode->i_mode) ||
!S_ISDIR(lowerpath.dentry->d_inode->i_mode))
goto out_put_lowerpath;
err = vfs_statfs(&lowerpath, &statfs);
if (err) {
pr_err("overlayfs: statfs failed on lowerpath\n");
goto out_put_lowerpath;
}
ufs->lower_namelen = statfs.f_namelen;
sb->s_stack_depth = max(upperpath.mnt->mnt_sb->s_stack_depth,
lowerpath.mnt->mnt_sb->s_stack_depth) + 1;
err = -EINVAL;
if (sb->s_stack_depth > FILESYSTEM_MAX_STACK_DEPTH) {
pr_err("overlayfs: maximum fs stacking depth exceeded\n");
goto out_put_lowerpath;
}
ufs->upper_mnt = clone_private_mount(&upperpath);
err = PTR_ERR(ufs->upper_mnt);
if (IS_ERR(ufs->upper_mnt)) {
pr_err("overlayfs: failed to clone upperpath\n");
goto out_put_lowerpath;
}
ufs->lower_mnt = clone_private_mount(&lowerpath);
err = PTR_ERR(ufs->lower_mnt);
if (IS_ERR(ufs->lower_mnt)) {
pr_err("overlayfs: failed to clone lowerpath\n");
goto out_put_upper_mnt;
}
/*
* Make lower_mnt R/O. That way fchmod/fchown on lower file
* will fail instead of modifying lower fs.
*/
ufs->lower_mnt->mnt_flags |= MNT_READONLY;
/* If the upper fs is r/o, we mark overlayfs r/o too */
if (ufs->upper_mnt->mnt_sb->s_flags & MS_RDONLY)
sb->s_flags |= MS_RDONLY;
if (!(sb->s_flags & MS_RDONLY)) {
err = mnt_want_write(ufs->upper_mnt);
if (err)
goto out_put_lower_mnt;
}
err = -ENOMEM;
root_inode = ovl_new_inode(sb, S_IFDIR, oe);
if (!root_inode)
goto out_drop_write;
root_dentry = d_make_root(root_inode);
if (!root_dentry)
goto out_drop_write;
mntput(upperpath.mnt);
mntput(lowerpath.mnt);
oe->__upperdentry = dget(upperpath.dentry);
oe->lowerdentry = lowerpath.dentry;
root_dentry->d_fsdata = oe;
root_dentry->d_op = &ovl_dentry_operations;
sb->s_magic = OVERLAYFS_SUPER_MAGIC;
sb->s_op = &ovl_super_operations;
sb->s_root = root_dentry;
sb->s_fs_info = ufs;
return 0;
out_drop_write:
if (!(sb->s_flags & MS_RDONLY))
mnt_drop_write(ufs->upper_mnt);
out_put_lower_mnt:
mntput(ufs->lower_mnt);
out_put_upper_mnt:
mntput(ufs->upper_mnt);
out_put_lowerpath:
path_put(&lowerpath);
out_put_upperpath:
path_put(&upperpath);
out_free_oe:
kfree(oe);
out_free_config:
kfree(ufs->config.lowerdir);
kfree(ufs->config.upperdir);
out_free_ufs:
kfree(ufs);
out:
return err;
}
/*
* Look up one component of a pathname.
* N.B. After this call _both_ fhp and resfh need an fh_put
*
* If the lookup would cross a mountpoint, and the mounted filesystem
* is exported to the client with NFSEXP_CROSSMNT, then the lookup is
* accepted as it stands and the mounted directory is
* returned. Otherwise the covered directory is returned.
* NOTE: this mountpoint crossing is not supported properly by all
* clients and is explicitly disallowed for NFSv3
* NeilBrown <*****@*****.**>
*/
int
nfsd_lookup(struct svc_rqst *rqstp, struct svc_fh *fhp, const char *name,
int len, struct svc_fh *resfh)
{
struct svc_export *exp;
struct dentry *dparent;
struct dentry *dentry;
int err;
dprintk("nfsd: nfsd_lookup(fh %s, %.*s)\n", SVCFH_fmt(fhp), len,name);
/* Obtain dentry and export. */
err = fh_verify(rqstp, fhp, S_IFDIR, MAY_EXEC);
if (err)
goto out;
dparent = fhp->fh_dentry;
exp = fhp->fh_export;
err = nfserr_acces;
/* Lookup the name, but don't follow links */
if (isdotent(name, len)) {
if (len==1)
dentry = dget(dparent);
else { /* must be ".." */
/* checking mountpoint crossing is very different when stepping up */
if (dparent == exp->ex_dentry) {
if (!EX_CROSSMNT(exp))
dentry = dget(dparent); /* .. == . just like at / */
else
{
struct svc_export *exp2 = NULL;
struct dentry *dp;
struct vfsmount *mnt = mntget(exp->ex_mnt);
dentry = dget(dparent);
while(follow_up(&mnt, &dentry))
;
dp = dget(dentry->d_parent);
dput(dentry);
dentry = dp;
for ( ; exp2 == NULL && dp->d_parent != dp;
dp=dp->d_parent)
exp2 = exp_get(exp->ex_client, dp->d_inode->i_dev, dp->d_inode->i_ino);
if (exp2==NULL) {
dput(dentry);
dentry = dget(dparent);
} else {
exp = exp2;
}
mntput(mnt);
}
} else
dentry = dget(dparent->d_parent);
}
} else {
fh_lock(fhp);
dentry = lookup_one_len(name, dparent, len);
err = PTR_ERR(dentry);
if (IS_ERR(dentry))
goto out_nfserr;
/*
* check if we have crossed a mount point ...
*/
if (d_mountpoint(dentry)) {
struct svc_export *exp2 = NULL;
struct vfsmount *mnt = mntget(exp->ex_mnt);
struct dentry *mounts = dget(dentry);
while (follow_down(&mnt,&mounts)&&d_mountpoint(mounts))
;
exp2 = exp_get(rqstp->rq_client,
mounts->d_inode->i_dev,
mounts->d_inode->i_ino);
if (exp2 && EX_CROSSMNT(exp2)) {
/* successfully crossed mount point */
exp = exp2;
dput(dentry);
dentry = mounts;
} else
dput(mounts);
mntput(mnt);
}
}
/*
* Note: we compose the file handle now, but as the
* dentry may be negative, it may need to be updated.
*/
err = fh_compose(resfh, exp, dentry, fhp);
if (!err && !dentry->d_inode)
err = nfserr_noent;
out:
return err;
out_nfserr:
err = nfserrno(err);
goto out;
}
