/* SPDX-License-Identifier: GPL-2.0 */
#include <linux/syscalls.h>
#include <linux/export.h>
#include <linux/uaccess.h>
#include <linux/fs_struct.h>
#include <linux/fs.h>
#include <linux/slab.h>
#include <linux/prefetch.h>
#include "mount.h"
static void prepend(char **buffer, int *buflen, const char *str, int namelen)
{
*buflen -= namelen;
if (likely(*buflen >= 0)) {
*buffer -= namelen;
memcpy(*buffer, str, namelen);
}
}
/**
* prepend_name - prepend a pathname in front of current buffer pointer
* @buffer: buffer pointer
* @buflen: allocated length of the buffer
* @name: name string and length qstr structure
*
* With RCU path tracing, it may race with d_move(). Use READ_ONCE() to
* make sure that either the old or the new name pointer and length are
* fetched. However, there may be mismatch between length and pointer.
* The length cannot be trusted, we need to copy it byte-by-byte until
* the length is reached or a null byte is found. It also prepends "/" at
* the beginning of the name. The sequence number check at the caller will
* retry it again when a d_move() does happen. So any garbage in the buffer
* due to mismatched pointer and length will be discarded.
*
* Load acquire is needed to make sure that we see that terminating NUL.
*/
static bool prepend_name(char **buffer, int *buflen, const struct qstr *name)
{
const char *dname = smp_load_acquire(&name->name); /* ^^^ */
u32 dlen = READ_ONCE(name->len);
char *p;
*buflen -= dlen + 1;
if (unlikely(*buflen < 0))
return false;
p = *buffer -= dlen + 1;
*p++ = '/';
while (dlen--) {
char c = *dname++;
if (!c)
break;
*p++ = c;
}
return true;
}
/**
* prepend_path - Prepend path string to a buffer
* @path: the dentry/vfsmount to report
* @root: root vfsmnt/dentry
* @buffer: pointer to the end of the buffer
* @buflen: pointer to buffer length
*
* The function will first try to write out the pathname without taking any
* lock other than the RCU read lock to make sure that dentries won't go away.
* It only checks the sequence number of the global rename_lock as any change
* in the dentry's d_seq will be preceded by changes in the rename_lock
* sequence number. If the sequence number had been changed, it will restart
* the whole pathname back-tracing sequence again by taking the rename_lock.
* In this case, there is no need to take the RCU read lock as the recursive
* parent pointer references will keep the dentry chain alive as long as no
* rename operation is performed.
*/
static int prepend_path(const struct path *path,
const struct path *root,
char **buffer, int *buflen)
{
struct dentry *dentry;
struct vfsmount *vfsmnt;
struct mount *mnt;
int error = 0;
unsigned seq, m_seq = 0;
char *bptr;
int blen;
rcu_read_lock();
restart_mnt:
read_seqbegin_or_lock(&mount_lock, &m_seq);
seq = 0;
rcu_read_lock();
restart:
bptr = *buffer;
blen = *buflen;
error = 0;
dentry = path->dentry;
vfsmnt = path->mnt;
mnt = real_mount(vfsmnt);
read_seqbegin_or_lock(&rename_lock, &seq);
while (dentry != root->dentry || vfsmnt != root->mnt) {
struct dentry * parent;
if (dentry == vfsmnt->mnt_root || IS_ROOT(dentry)) {
struct mount *parent = READ_ONCE(mnt->mnt_parent);
struct mnt_namespace *mnt_ns;
/* Escaped? */
if (dentry != vfsmnt->mnt_root) {
bptr = *buffer;
blen = *buflen;
error = 3;
break;
}
/* Global root? */
if (mnt != parent) {
dentry = READ_ONCE(mnt->mnt_mountpoint);
mnt = parent;
vfsmnt = &mnt->mnt;
continue;
}
mnt_ns = READ_ONCE(mnt->mnt_ns);
/* open-coded is_mounted() to use local mnt_ns */
if (!IS_ERR_OR_NULL(mnt_ns) && !is_anon_ns(mnt_ns))
error = 1; // absolute root
else
error = 2; // detached or not attached yet
break;
}
parent = dentry->d_parent;
prefetch(parent);
if (!prepend_name(&bptr, &blen, &dentry->d_name))
break;
dentry = parent;
}
if (!(seq & 1))
rcu_read_unlock();
if (need_seqretry(&rename_lock, seq)) {
seq = 1;
goto restart;
}
done_seqretry(&rename_lock, seq);
if (!(m_seq & 1))
rcu_read_unlock();
if (need_seqretry(&mount_lock, m_seq)) {
m_seq = 1;
goto restart_mnt;
}
done_seqretry(&mount_lock, m_seq);
if (blen == *buflen)
prepend(&bptr, &blen, "/", 1);
*buffer = bptr;
*buflen = blen;
return error;
}
/**
* __d_path - return the path of a dentry
* @path: the dentry/vfsmount to report
* @root: root vfsmnt/dentry
* @buf: buffer to return value in
* @buflen: buffer length
*
* Convert a dentry into an ASCII path name.
*
* Returns a pointer into the buffer or an error code if the
* path was too long.
*
* "buflen" should be positive.
*
* If the path is not reachable from the supplied root, return %NULL.
*/
char *__d_path(const struct path *path,
const struct path *root,
char *buf, int buflen)
{
char *res = buf + buflen;
prepend(&res, &buflen, "", 1);
if (prepend_path(path, root, &res, &buflen) > 0)
return NULL;
return buflen >= 0 ? res : ERR_PTR(-ENAMETOOLONG);
}
char *d_absolute_path(const struct path *path,
char *buf, int buflen)
{
struct path root = {};
char *res = buf + buflen;
prepend(&res, &buflen, "", 1);
if (prepend_path(path, &root, &res, &buflen) > 1)
return ERR_PTR(-EINVAL);
return buflen >= 0 ? res : ERR_PTR(-ENAMETOOLONG);
}
static void get_fs_root_rcu(struct fs_struct *fs, struct path *root)
{
unsigned seq;
do {
seq = read_seqcount_begin(&fs->seq);
*root = fs->root;
} while (read_seqcount_retry(&fs->seq, seq));
}
/**
* d_path - return the path of a dentry
* @path: path to report
* @buf: buffer to return value in
* @buflen: buffer length
*
* Convert a dentry into an ASCII path name. If the entry has been deleted
* the string " (deleted)" is appended. Note that this is ambiguous.
*
* Returns a pointer into the buffer or an error code if the path was
* too long. Note: Callers should use the returned pointer, not the passed
* in buffer, to use the name! The implementation often starts at an offset
* into the buffer, and may leave 0 bytes at the start.
*
* "buflen" should be positive.
*/
char *d_path(const struct path *path, char *buf, int buflen)
{
char *res = buf + buflen;
struct path root;
/*
* We have various synthetic filesystems that never get mounted. On
* these filesystems dentries are never used for lookup purposes, and
* thus don't need to be hashed. They also don't need a name until a
* user wants to identify the object in /proc/pid/fd/. The little hack
* below allows us to generate a name for these objects on demand:
*
* Some pseudo inodes are mountable. When they are mounted
* path->dentry == path->mnt->mnt_root. In that case don't call d_dname
* and instead have d_path return the mounted path.
*/
if (path->dentry->d_op && path->dentry->d_op->d_dname &&
(!IS_ROOT(path->dentry) || path->dentry != path->mnt->mnt_root))
return path->dentry->d_op->d_dname(path->dentry, buf, buflen);
rcu_read_lock();
get_fs_root_rcu(current->fs, &root);
if (unlikely(d_unlinked(path->dentry)))
prepend(&res, &buflen, " (deleted)", 11);
else
prepend(&res, &buflen, "", 1);
prepend_path(path, &root, &res, &buflen);
rcu_read_unlock();
return buflen >= 0 ? res : ERR_PTR(-ENAMETOOLONG);
}
EXPORT_SYMBOL(d_path);
/*
* Helper function for dentry_operations.d_dname() members
*/
char *dynamic_dname(struct dentry *dentry, char *buffer, int buflen,
const char *fmt, ...)
{
va_list args;
char temp[64];
int sz;
va_start(args, fmt);
sz = vsnprintf(temp, sizeof(temp), fmt, args) + 1;
va_end(args);
if (sz > sizeof(temp) || sz > buflen)
return ERR_PTR(-ENAMETOOLONG);
buffer += buflen - sz;
return memcpy(buffer, temp, sz);
}
char *simple_dname(struct dentry *dentry, char *buffer, int buflen)
{
char *end = buffer + buflen;
/* these dentries are never renamed, so d_lock is not needed */
prepend(&end, &buflen, " (deleted)", 11);
prepend(&end, &buflen, dentry->d_name.name, dentry->d_name.len);
prepend(&end, &buflen, "/", 1);
return buflen >= 0 ? end : ERR_PTR(-ENAMETOOLONG);
}
/*
* Write full pathname from the root of the filesystem into the buffer.
*/
static char *__dentry_path(const struct dentry *d, char *p, int buflen)
{
const struct dentry *dentry;
char *end;
int len, seq = 0;
rcu_read_lock();
restart:
dentry = d;
end = p;
len = buflen;
read_seqbegin_or_lock(&rename_lock, &seq);
while (!IS_ROOT(dentry)) {
const struct dentry *parent = dentry->d_parent;
prefetch(parent);
if (!prepend_name(&end, &len, &dentry->d_name))
break;
dentry = parent;
}
if (!(seq & 1))
rcu_read_unlock();
if (need_seqretry(&rename_lock, seq)) {
seq = 1;
goto restart;
}
done_seqretry(&rename_lock, seq);
if (len == buflen)
prepend(&end, &len, "/", 1);
return len >= 0 ? end : ERR_PTR(-ENAMETOOLONG);
}
char *dentry_path_raw(const struct dentry *dentry, char *buf, int buflen)
{
char *p = buf + buflen;
prepend(&p, &buflen, "", 1);
return __dentry_path(dentry, p, buflen);
}
EXPORT_SYMBOL(dentry_path_raw);
char *dentry_path(const struct dentry *dentry, char *buf, int buflen)
{
char *p = buf + buflen;
if (unlikely(d_unlinked(dentry)))
prepend(&p, &buflen, "//deleted", 10);
else
prepend(&p, &buflen, "", 1);
return __dentry_path(dentry, p, buflen);
}
static void get_fs_root_and_pwd_rcu(struct fs_struct *fs, struct path *root,
struct path *pwd)
{
unsigned seq;
do {
seq = read_seqcount_begin(&fs->seq);
*root = fs->root;
*pwd = fs->pwd;
} while (read_seqcount_retry(&fs->seq, seq));
}
/*
* NOTE! The user-level library version returns a
* character pointer. The kernel system call just
* returns the length of the buffer filled (which
* includes the ending '\0' character), or a negative
* error value. So libc would do something like
*
* char *getcwd(char * buf, size_t size)
* {
* int retval;
*
* retval = sys_getcwd(buf, size);
* if (retval >= 0)
* return buf;
* errno = -retval;
* return NULL;
* }
*/
SYSCALL_DEFINE2(getcwd, char __user *, buf, unsigned long, size)
{
int error;
struct path pwd, root;
char *page = __getname();
if (!page)
return -ENOMEM;
rcu_read_lock();
get_fs_root_and_pwd_rcu(current->fs, &root, &pwd);
error = -ENOENT;
if (!d_unlinked(pwd.dentry)) {
unsigned long len;
char *cwd = page + PATH_MAX;
int buflen = PATH_MAX;
prepend(&cwd, &buflen, "", 1);
if (prepend_path(&pwd, &root, &cwd, &buflen) > 0)
prepend(&cwd, &buflen, "(unreachable)", 13);
rcu_read_unlock();
if (buflen < 0) {
error = -ENAMETOOLONG;
goto out;
}
error = -ERANGE;
len = PATH_MAX + page - cwd;
if (len <= size) {
error = len;
if (copy_to_user(buf, cwd, len))
error = -EFAULT;
}
} else {
rcu_read_unlock();
}
out:
__putname(page);
return error;
}
