*/
static int lookup_device(const char *path, dev_t *dev)
{
- int r;
- struct nameidata nd;
- struct inode *inode;
-
- if ((r = path_lookup(path, LOOKUP_FOLLOW, &nd)))
- return r;
-
- inode = nd.path.dentry->d_inode;
- if (!inode) {
- r = -ENOENT;
- goto out;
- }
-
- if (!S_ISBLK(inode->i_mode)) {
- r = -ENOTBLK;
- goto out;
- }
-
- *dev = inode->i_rdev;
-
- out:
- path_put(&nd.path);
- return r;
+ struct block_device *bdev = lookup_bdev(path);
+ if (IS_ERR(bdev))
+ return PTR_ERR(bdev);
+ *dev = bdev->bd_dev;
+ bdput(bdev);
+ return 0;
}
/*
int (*fill_super)(struct super_block *, void *, int),
struct vfsmount *mnt)
{
- struct nameidata nd;
+ struct block_device *bdev;
int mtdnr, ret;
if (!dev_name)
/* try the old way - the hack where we allowed users to mount
* /dev/mtdblock$(n) but didn't actually _use_ the blockdev
*/
- ret = path_lookup(dev_name, LOOKUP_FOLLOW, &nd);
-
- DEBUG(1, "MTDSB: path_lookup() returned %d, inode %p\n",
- ret, nd.path.dentry ? nd.path.dentry->d_inode : NULL);
-
- if (ret)
+ bdev = lookup_bdev(dev_name);
+ if (IS_ERR(bdev)) {
+ ret = PTR_ERR(bdev);
+ DEBUG(1, "MTDSB: lookup_bdev() returned %d\n", ret);
return ret;
-
- ret = -EINVAL;
-
- if (!S_ISBLK(nd.path.dentry->d_inode->i_mode))
- goto out;
-
- if (nd.path.mnt->mnt_flags & MNT_NODEV) {
- ret = -EACCES;
- goto out;
}
+ DEBUG(1, "MTDSB: lookup_bdev() returned 0\n");
- if (imajor(nd.path.dentry->d_inode) != MTD_BLOCK_MAJOR)
+ ret = -EINVAL;
+ if (MAJOR(bdev->bd_dev) != MTD_BLOCK_MAJOR)
goto not_an_MTD_device;
- mtdnr = iminor(nd.path.dentry->d_inode);
- path_put(&nd.path);
+ mtdnr = MINOR(bdev->bd_dev);
+ bdput(bdev);
return get_sb_mtd_nr(fs_type, flags, dev_name, data, mtdnr, fill_super,
mnt);
printk(KERN_NOTICE
"MTD: Attempt to mount non-MTD device \"%s\"\n",
dev_name);
-out:
- path_put(&nd.path);
+ bdput(bdev);
return ret;
-
}
EXPORT_SYMBOL_GPL(get_sb_mtd);
}
mntget(newmnt);
- err = do_add_mount(newmnt, nd, MNT_SHRINKABLE, &afs_vfsmounts);
+ err = do_add_mount(newmnt, &nd->path, MNT_SHRINKABLE, &afs_vfsmounts);
switch (err) {
case 0:
path_put(&nd->path);
* hooks: /n/, see "layering violations".
*/
ret = devcgroup_inode_permission(bdev->bd_inode, perm);
- if (ret != 0)
+ if (ret != 0) {
+ bdput(bdev);
return ret;
+ }
ret = -ENXIO;
file->f_mapping = bdev->bd_inode->i_mapping;
bdev = ERR_PTR(error);
goto out;
}
+EXPORT_SYMBOL(lookup_bdev);
/**
* open_bdev_excl - open a block device by name and set it up for use
int err;
mntget(newmnt);
- err = do_add_mount(newmnt, nd, nd->path.mnt->mnt_flags, mntlist);
+ err = do_add_mount(newmnt, &nd->path, nd->path.mnt->mnt_flags, mntlist);
switch (err) {
case 0:
path_put(&nd->path);
#define DEVPTS_DEFAULT_MODE 0600
extern int pty_limit; /* Config limit on Unix98 ptys */
-static DEFINE_IDR(allocated_ptys);
+static DEFINE_IDA(allocated_ptys);
static DEFINE_MUTEX(allocated_ptys_lock);
static struct vfsmount *devpts_mnt;
int devpts_new_index(void)
{
int index;
- int idr_ret;
+ int ida_ret;
retry:
- if (!idr_pre_get(&allocated_ptys, GFP_KERNEL)) {
+ if (!ida_pre_get(&allocated_ptys, GFP_KERNEL)) {
return -ENOMEM;
}
mutex_lock(&allocated_ptys_lock);
- idr_ret = idr_get_new(&allocated_ptys, NULL, &index);
- if (idr_ret < 0) {
+ ida_ret = ida_get_new(&allocated_ptys, &index);
+ if (ida_ret < 0) {
mutex_unlock(&allocated_ptys_lock);
- if (idr_ret == -EAGAIN)
+ if (ida_ret == -EAGAIN)
goto retry;
return -EIO;
}
if (index >= pty_limit) {
- idr_remove(&allocated_ptys, index);
+ ida_remove(&allocated_ptys, index);
mutex_unlock(&allocated_ptys_lock);
return -EIO;
}
void devpts_kill_index(int idx)
{
mutex_lock(&allocated_ptys_lock);
- idr_remove(&allocated_ptys, idx);
+ ida_remove(&allocated_ptys, idx);
mutex_unlock(&allocated_ptys_lock);
}
return ret;
}
+int vfs_quota_on_path(struct super_block *sb, int type, int format_id,
+ struct path *path)
+{
+ int error = security_quota_on(path->dentry);
+ if (error)
+ return error;
+ /* Quota file not on the same filesystem? */
+ if (path->mnt->mnt_sb != sb)
+ error = -EXDEV;
+ else
+ error = vfs_quota_on_inode(path->dentry->d_inode, type,
+ format_id);
+ return error;
+}
+
/* Actual function called from quotactl() */
int vfs_quota_on(struct super_block *sb, int type, int format_id, char *path,
int remount)
return vfs_quota_on_remount(sb, type);
error = path_lookup(path, LOOKUP_FOLLOW, &nd);
- if (error < 0)
- return error;
- error = security_quota_on(nd.path.dentry);
- if (error)
- goto out_path;
- /* Quota file not on the same filesystem? */
- if (nd.path.mnt->mnt_sb != sb)
- error = -EXDEV;
- else
- error = vfs_quota_on_inode(nd.path.dentry->d_inode, type,
- format_id);
-out_path:
- path_put(&nd.path);
+ if (!error) {
+ error = vfs_quota_on_path(sb, type, format_id, &nd.path);
+ path_put(&nd.path);
+ }
return error;
}
EXPORT_SYMBOL(dqstats);
EXPORT_SYMBOL(dq_data_lock);
EXPORT_SYMBOL(vfs_quota_on);
+EXPORT_SYMBOL(vfs_quota_on_path);
EXPORT_SYMBOL(vfs_quota_on_mount);
EXPORT_SYMBOL(vfs_quota_off);
EXPORT_SYMBOL(vfs_quota_sync);
journal_unlock_updates(EXT3_SB(sb)->s_journal);
}
+ err = vfs_quota_on_path(sb, type, format_id, &nd.path);
path_put(&nd.path);
- return vfs_quota_on(sb, type, format_id, path, remount);
+ return err;
}
/* Read data from quotafile - avoid pagecache and such because we cannot afford
jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal);
}
+ err = vfs_quota_on_path(sb, type, format_id, &nd.path);
path_put(&nd.path);
- return vfs_quota_on(sb, type, format_id, path, remount);
+ return err;
}
/* Read data from quotafile - avoid pagecache and such because we cannot afford
return res;
}
-/*
- * locate_fd finds a free file descriptor in the open_fds fdset,
- * expanding the fd arrays if necessary. Must be called with the
- * file_lock held for write.
- */
-
-static int locate_fd(unsigned int orig_start, int cloexec)
-{
- struct files_struct *files = current->files;
- unsigned int newfd;
- unsigned int start;
- int error;
- struct fdtable *fdt;
-
- spin_lock(&files->file_lock);
-repeat:
- fdt = files_fdtable(files);
- /*
- * Someone might have closed fd's in the range
- * orig_start..fdt->next_fd
- */
- start = orig_start;
- if (start < files->next_fd)
- start = files->next_fd;
-
- newfd = start;
- if (start < fdt->max_fds)
- newfd = find_next_zero_bit(fdt->open_fds->fds_bits,
- fdt->max_fds, start);
-
- error = expand_files(files, newfd);
- if (error < 0)
- goto out;
-
- /*
- * If we needed to expand the fs array we
- * might have blocked - try again.
- */
- if (error)
- goto repeat;
-
- if (start <= files->next_fd)
- files->next_fd = newfd + 1;
-
- FD_SET(newfd, fdt->open_fds);
- if (cloexec)
- FD_SET(newfd, fdt->close_on_exec);
- else
- FD_CLR(newfd, fdt->close_on_exec);
- error = newfd;
-
-out:
- spin_unlock(&files->file_lock);
- return error;
-}
-
-static int dupfd(struct file *file, unsigned int start, int cloexec)
-{
- int fd = locate_fd(start, cloexec);
- if (fd >= 0)
- fd_install(fd, file);
- else
- fput(file);
-
- return fd;
-}
-
asmlinkage long sys_dup3(unsigned int oldfd, unsigned int newfd, int flags)
{
int err = -EBADF;
return -EINVAL;
spin_lock(&files->file_lock);
- if (!(file = fcheck(oldfd)))
- goto out_unlock;
- get_file(file); /* We are now finished with oldfd */
-
err = expand_files(files, newfd);
+ file = fcheck(oldfd);
+ if (unlikely(!file))
+ goto Ebadf;
if (unlikely(err < 0)) {
if (err == -EMFILE)
- err = -EBADF;
- goto out_fput;
+ goto Ebadf;
+ goto out_unlock;
}
-
- /* To avoid races with open() and dup(), we will mark the fd as
- * in-use in the open-file bitmap throughout the entire dup2()
- * process. This is quite safe: do_close() uses the fd array
- * entry, not the bitmap, to decide what work needs to be
- * done. --sct */
- /* Doesn't work. open() might be there first. --AV */
-
- /* Yes. It's a race. In user space. Nothing sane to do */
+ /*
+ * We need to detect attempts to do dup2() over allocated but still
+ * not finished descriptor. NB: OpenBSD avoids that at the price of
+ * extra work in their equivalent of fget() - they insert struct
+ * file immediately after grabbing descriptor, mark it larval if
+ * more work (e.g. actual opening) is needed and make sure that
+ * fget() treats larval files as absent. Potentially interesting,
+ * but while extra work in fget() is trivial, locking implications
+ * and amount of surgery on open()-related paths in VFS are not.
+ * FreeBSD fails with -EBADF in the same situation, NetBSD "solution"
+ * deadlocks in rather amusing ways, AFAICS. All of that is out of
+ * scope of POSIX or SUS, since neither considers shared descriptor
+ * tables and this condition does not arise without those.
+ */
err = -EBUSY;
fdt = files_fdtable(files);
tofree = fdt->fd[newfd];
if (!tofree && FD_ISSET(newfd, fdt->open_fds))
- goto out_fput;
-
+ goto out_unlock;
+ get_file(file);
rcu_assign_pointer(fdt->fd[newfd], file);
FD_SET(newfd, fdt->open_fds);
if (flags & O_CLOEXEC)
if (tofree)
filp_close(tofree, files);
- err = newfd;
-out:
- return err;
-out_unlock:
- spin_unlock(&files->file_lock);
- goto out;
-out_fput:
+ return newfd;
+
+Ebadf:
+ err = -EBADF;
+out_unlock:
spin_unlock(&files->file_lock);
- fput(file);
- goto out;
+ return err;
}
asmlinkage long sys_dup2(unsigned int oldfd, unsigned int newfd)
asmlinkage long sys_dup(unsigned int fildes)
{
int ret = -EBADF;
- struct file * file = fget(fildes);
-
- if (file)
- ret = dupfd(file, 0, 0);
+ struct file *file = fget(fildes);
+
+ if (file) {
+ ret = get_unused_fd();
+ if (ret >= 0)
+ fd_install(ret, file);
+ else
+ fput(file);
+ }
return ret;
}
case F_DUPFD_CLOEXEC:
if (arg >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
break;
- get_file(filp);
- err = dupfd(filp, arg, cmd == F_DUPFD_CLOEXEC);
+ err = alloc_fd(arg, cmd == F_DUPFD_CLOEXEC ? O_CLOEXEC : 0);
+ if (err >= 0) {
+ get_file(filp);
+ fd_install(err, filp);
+ }
break;
case F_GETFD:
err = get_close_on_exec(fd) ? FD_CLOEXEC : 0;
* Manage the dynamic fd arrays in the process files_struct.
*/
+#include <linux/module.h>
#include <linux/fs.h>
#include <linux/mm.h>
#include <linux/time.h>
},
.file_lock = __SPIN_LOCK_UNLOCKED(init_task.file_lock),
};
+
+/*
+ * allocate a file descriptor, mark it busy.
+ */
+int alloc_fd(unsigned start, unsigned flags)
+{
+ struct files_struct *files = current->files;
+ unsigned int fd;
+ int error;
+ struct fdtable *fdt;
+
+ spin_lock(&files->file_lock);
+repeat:
+ fdt = files_fdtable(files);
+ fd = start;
+ if (fd < files->next_fd)
+ fd = files->next_fd;
+
+ if (fd < fdt->max_fds)
+ fd = find_next_zero_bit(fdt->open_fds->fds_bits,
+ fdt->max_fds, fd);
+
+ error = expand_files(files, fd);
+ if (error < 0)
+ goto out;
+
+ /*
+ * If we needed to expand the fs array we
+ * might have blocked - try again.
+ */
+ if (error)
+ goto repeat;
+
+ if (start <= files->next_fd)
+ files->next_fd = fd + 1;
+
+ FD_SET(fd, fdt->open_fds);
+ if (flags & O_CLOEXEC)
+ FD_SET(fd, fdt->close_on_exec);
+ else
+ FD_CLR(fd, fdt->close_on_exec);
+ error = fd;
+#if 1
+ /* Sanity check */
+ if (rcu_dereference(fdt->fd[fd]) != NULL) {
+ printk(KERN_WARNING "alloc_fd: slot %d not NULL!\n", fd);
+ rcu_assign_pointer(fdt->fd[fd], NULL);
+ }
+#endif
+
+out:
+ spin_unlock(&files->file_lock);
+ return error;
+}
+
+int get_unused_fd(void)
+{
+ return alloc_fd(0, 0);
+}
+EXPORT_SYMBOL(get_unused_fd);
return retval;
return security_inode_permission(inode,
- mask & (MAY_READ|MAY_WRITE|MAY_EXEC));
+ mask & (MAY_READ|MAY_WRITE|MAY_EXEC|MAY_APPEND));
}
/**
* 3. We should have write and exec permissions on dir
* 4. We can't do it if dir is immutable (done in permission())
*/
-static inline int may_create(struct inode *dir, struct dentry *child,
- struct nameidata *nd)
+static inline int may_create(struct inode *dir, struct dentry *child)
{
if (child->d_inode)
return -EEXIST;
int vfs_create(struct inode *dir, struct dentry *dentry, int mode,
struct nameidata *nd)
{
- int error = may_create(dir, dentry, nd);
+ int error = may_create(dir, dentry);
if (error)
return error;
int vfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev)
{
- int error = may_create(dir, dentry, NULL);
+ int error = may_create(dir, dentry);
if (error)
return error;
int vfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
{
- int error = may_create(dir, dentry, NULL);
+ int error = may_create(dir, dentry);
if (error)
return error;
int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname)
{
- int error = may_create(dir, dentry, NULL);
+ int error = may_create(dir, dentry);
if (error)
return error;
if (!inode)
return -ENOENT;
- error = may_create(dir, new_dentry, NULL);
+ error = may_create(dir, new_dentry);
if (error)
return error;
return error;
if (!new_dentry->d_inode)
- error = may_create(new_dir, new_dentry, NULL);
+ error = may_create(new_dir, new_dentry);
else
error = may_delete(new_dir, new_dentry, is_dir);
if (error)
if (IS_ERR(mnt))
return PTR_ERR(mnt);
- return do_add_mount(mnt, nd, mnt_flags, NULL);
+ return do_add_mount(mnt, &nd->path, mnt_flags, NULL);
}
/*
* add a mount into a namespace's mount tree
* - provide the option of adding the new mount to an expiration list
*/
-int do_add_mount(struct vfsmount *newmnt, struct nameidata *nd,
+int do_add_mount(struct vfsmount *newmnt, struct path *path,
int mnt_flags, struct list_head *fslist)
{
int err;
down_write(&namespace_sem);
/* Something was mounted here while we slept */
- while (d_mountpoint(nd->path.dentry) &&
- follow_down(&nd->path.mnt, &nd->path.dentry))
+ while (d_mountpoint(path->dentry) &&
+ follow_down(&path->mnt, &path->dentry))
;
err = -EINVAL;
- if (!check_mnt(nd->path.mnt))
+ if (!check_mnt(path->mnt))
goto unlock;
/* Refuse the same filesystem on the same mount point */
err = -EBUSY;
- if (nd->path.mnt->mnt_sb == newmnt->mnt_sb &&
- nd->path.mnt->mnt_root == nd->path.dentry)
+ if (path->mnt->mnt_sb == newmnt->mnt_sb &&
+ path->mnt->mnt_root == path->dentry)
goto unlock;
err = -EINVAL;
goto unlock;
newmnt->mnt_flags = mnt_flags;
- if ((err = graft_tree(newmnt, &nd->path)))
+ if ((err = graft_tree(newmnt, path)))
goto unlock;
if (fslist) /* add to the specified expiration list */
goto out_err;
mntget(mnt);
- err = do_add_mount(mnt, nd, nd->path.mnt->mnt_flags|MNT_SHRINKABLE,
+ err = do_add_mount(mnt, &nd->path, nd->path.mnt->mnt_flags|MNT_SHRINKABLE,
&nfs_automount_list);
if (err < 0) {
mntput(mnt);
}
EXPORT_SYMBOL(dentry_open);
-/*
- * Find an empty file descriptor entry, and mark it busy.
- */
-int get_unused_fd_flags(int flags)
-{
- struct files_struct * files = current->files;
- int fd, error;
- struct fdtable *fdt;
-
- spin_lock(&files->file_lock);
-
-repeat:
- fdt = files_fdtable(files);
- fd = find_next_zero_bit(fdt->open_fds->fds_bits, fdt->max_fds,
- files->next_fd);
-
- /* Do we need to expand the fd array or fd set? */
- error = expand_files(files, fd);
- if (error < 0)
- goto out;
-
- if (error) {
- /*
- * If we needed to expand the fs array we
- * might have blocked - try again.
- */
- goto repeat;
- }
-
- FD_SET(fd, fdt->open_fds);
- if (flags & O_CLOEXEC)
- FD_SET(fd, fdt->close_on_exec);
- else
- FD_CLR(fd, fdt->close_on_exec);
- files->next_fd = fd + 1;
-#if 1
- /* Sanity check */
- if (fdt->fd[fd] != NULL) {
- printk(KERN_WARNING "get_unused_fd: slot %d not NULL!\n", fd);
- fdt->fd[fd] = NULL;
- }
-#endif
- error = fd;
-
-out:
- spin_unlock(&files->file_lock);
- return error;
-}
-
-int get_unused_fd(void)
-{
- return get_unused_fd_flags(0);
-}
-
-EXPORT_SYMBOL(get_unused_fd);
-
static void __put_unused_fd(struct files_struct *files, unsigned int fd)
{
struct fdtable *fdt = files_fdtable(files);
return rtn;
}
-static DEFINE_IDR(proc_inum_idr);
+static DEFINE_IDA(proc_inum_ida);
static DEFINE_SPINLOCK(proc_inum_lock); /* protects the above */
-#define PROC_DYNAMIC_FIRST 0xF0000000UL
+#define PROC_DYNAMIC_FIRST 0xF0000000U
/*
* Return an inode number between PROC_DYNAMIC_FIRST and
*/
static unsigned int get_inode_number(void)
{
- int i, inum = 0;
+ unsigned int i;
int error;
retry:
- if (idr_pre_get(&proc_inum_idr, GFP_KERNEL) == 0)
+ if (ida_pre_get(&proc_inum_ida, GFP_KERNEL) == 0)
return 0;
spin_lock(&proc_inum_lock);
- error = idr_get_new(&proc_inum_idr, NULL, &i);
+ error = ida_get_new(&proc_inum_ida, &i);
spin_unlock(&proc_inum_lock);
if (error == -EAGAIN)
goto retry;
else if (error)
return 0;
- inum = (i & MAX_ID_MASK) + PROC_DYNAMIC_FIRST;
-
- /* inum will never be more than 0xf0ffffff, so no check
- * for overflow.
- */
-
- return inum;
+ if (i > UINT_MAX - PROC_DYNAMIC_FIRST) {
+ spin_lock(&proc_inum_lock);
+ ida_remove(&proc_inum_ida, i);
+ spin_unlock(&proc_inum_lock);
+ }
+ return PROC_DYNAMIC_FIRST + i;
}
static void release_inode_number(unsigned int inum)
{
- int id = (inum - PROC_DYNAMIC_FIRST) | ~MAX_ID_MASK;
-
spin_lock(&proc_inum_lock);
- idr_remove(&proc_inum_idr, id);
+ ida_remove(&proc_inum_ida, inum - PROC_DYNAMIC_FIRST);
spin_unlock(&proc_inum_lock);
}
return err;
/* Quotafile not on the same filesystem? */
if (nd.path.mnt->mnt_sb != sb) {
- path_put(&nd.path);
- return -EXDEV;
+ err = -EXDEV;
+ goto out;
}
inode = nd.path.dentry->d_inode;
/* We must not pack tails for quota files on reiserfs for quota IO to work */
reiserfs_warning(sb,
"reiserfs: Unpacking tail of quota file failed"
" (%d). Cannot turn on quotas.", err);
- path_put(&nd.path);
- return -EINVAL;
+ err = -EINVAL;
+ goto out;
}
mark_inode_dirty(inode);
}
/* Just start temporary transaction and finish it */
err = journal_begin(&th, sb, 1);
if (err)
- return err;
+ goto out;
err = journal_end_sync(&th, sb, 1);
if (err)
- return err;
+ goto out;
}
+ err = vfs_quota_on_path(sb, type, format_id, &nd.path);
+out:
path_put(&nd.path);
- return vfs_quota_on(sb, type, format_id, path, 0);
+ return err;
}
/* Read data from quotafile - avoid pagecache and such because we cannot afford
extern struct file *fget_light(unsigned int fd, int *fput_needed);
extern void set_close_on_exec(unsigned int fd, int flag);
extern void put_filp(struct file *);
+extern int alloc_fd(unsigned start, unsigned flags);
extern int get_unused_fd(void);
-extern int get_unused_fd_flags(int flags);
+#define get_unused_fd_flags(flags) alloc_fd(0, (flags))
extern void put_unused_fd(unsigned int fd);
extern void fd_install(unsigned int fd, struct file *file);
struct nameidata;
-extern int do_add_mount(struct vfsmount *newmnt, struct nameidata *nd,
+struct path;
+extern int do_add_mount(struct vfsmount *newmnt, struct path *path,
int mnt_flags, struct list_head *fslist);
extern void mark_mounts_for_expiry(struct list_head *mounts);
int vfs_quota_on(struct super_block *sb, int type, int format_id,
char *path, int remount);
+int vfs_quota_on_path(struct super_block *sb, int type, int format_id,
+ struct path *path);
int vfs_quota_on_mount(struct super_block *sb, char *qf_name,
int format_id, int type);
int vfs_quota_off(struct super_block *sb, int type, int remount);
return rc;
}
+#ifdef CONFIG_SYSCTL
/*
* We really need to sanitize the damn ipv4 init order, then all
* this nonsense will go away.
{
register_sysctl_paths(ipv4_route_path, ipv4_route_table);
}
+#endif
EXPORT_SYMBOL(__ip_select_ident);
EXPORT_SYMBOL(ip_route_input);
git.openpandora.org / pandora-kernel.git / commitdiff