#include <linux/fs_struct.h>
#include <asm/uaccess.h>
-#define ACC_MODE(x) ("\000\004\002\006"[(x)&O_ACCMODE])
+#include "internal.h"
/* [Feb-1997 T. Schoebel-Theuer]
* Fundamental changes in the pathname lookup mechanisms (namei)
* any extra contention...
*/
-static int __link_path_walk(const char *name, struct nameidata *nd);
-
/* In order to reduce some races, while at the same time doing additional
* checking and hopefully speeding things up, we copy filenames to the
* kernel data space before using them..
EXPORT_SYMBOL(putname);
#endif
-
-/**
- * generic_permission - check for access rights on a Posix-like filesystem
- * @inode: inode to check access rights for
- * @mask: right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC)
- * @check_acl: optional callback to check for Posix ACLs
- *
- * Used to check for read/write/execute permissions on a file.
- * We use "fsuid" for this, letting us set arbitrary permissions
- * for filesystem access without changing the "normal" uids which
- * are used for other things..
+/*
+ * This does basic POSIX ACL permission checking
*/
-int generic_permission(struct inode *inode, int mask,
+static int acl_permission_check(struct inode *inode, int mask,
int (*check_acl)(struct inode *inode, int mask))
{
umode_t mode = inode->i_mode;
else {
if (IS_POSIXACL(inode) && (mode & S_IRWXG) && check_acl) {
int error = check_acl(inode, mask);
- if (error == -EACCES)
- goto check_capabilities;
- else if (error != -EAGAIN)
+ if (error != -EAGAIN)
return error;
}
*/
if ((mask & ~mode) == 0)
return 0;
+ return -EACCES;
+}
+
+/**
+ * generic_permission - check for access rights on a Posix-like filesystem
+ * @inode: inode to check access rights for
+ * @mask: right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC)
+ * @check_acl: optional callback to check for Posix ACLs
+ *
+ * Used to check for read/write/execute permissions on a file.
+ * We use "fsuid" for this, letting us set arbitrary permissions
+ * for filesystem access without changing the "normal" uids which
+ * are used for other things..
+ */
+int generic_permission(struct inode *inode, int mask,
+ int (*check_acl)(struct inode *inode, int mask))
+{
+ int ret;
+
+ /*
+ * Do the basic POSIX ACL permission checks.
+ */
+ ret = acl_permission_check(inode, mask, check_acl);
+ if (ret != -EACCES)
+ return ret;
- check_capabilities:
/*
* Read/write DACs are always overridable.
* Executable DACs are overridable if at least one exec bit is set.
/*
* Searching includes executable on directories, else just read.
*/
+ mask &= MAY_READ | MAY_WRITE | MAY_EXEC;
if (mask == MAY_READ || (S_ISDIR(inode->i_mode) && !(mask & MAY_WRITE)))
if (capable(CAP_DAC_READ_SEARCH))
return 0;
if (inode->i_op->permission)
retval = inode->i_op->permission(inode, mask);
else
- retval = generic_permission(inode, mask, NULL);
+ retval = generic_permission(inode, mask, inode->i_op->check_acl);
if (retval)
return retval;
}
/*
- * Internal lookup() using the new generic dcache.
- * SMP-safe
+ * force_reval_path - force revalidation of a dentry
+ *
+ * In some situations the path walking code will trust dentries without
+ * revalidating them. This causes problems for filesystems that depend on
+ * d_revalidate to handle file opens (e.g. NFSv4). When FS_REVAL_DOT is set
+ * (which indicates that it's possible for the dentry to go stale), force
+ * a d_revalidate call before proceeding.
+ *
+ * Returns 0 if the revalidation was successful. If the revalidation fails,
+ * either return the error returned by d_revalidate or -ESTALE if the
+ * revalidation it just returned 0. If d_revalidate returns 0, we attempt to
+ * invalidate the dentry. It's up to the caller to handle putting references
+ * to the path if necessary.
*/
-static struct dentry * cached_lookup(struct dentry * parent, struct qstr * name, struct nameidata *nd)
+static int
+force_reval_path(struct path *path, struct nameidata *nd)
{
- struct dentry * dentry = __d_lookup(parent, name);
+ int status;
+ struct dentry *dentry = path->dentry;
- /* lockess __d_lookup may fail due to concurrent d_move()
- * in some unrelated directory, so try with d_lookup
+ /*
+ * only check on filesystems where it's possible for the dentry to
+ * become stale. It's assumed that if this flag is set then the
+ * d_revalidate op will also be defined.
*/
- if (!dentry)
- dentry = d_lookup(parent, name);
+ if (!(dentry->d_sb->s_type->fs_flags & FS_REVAL_DOT))
+ return 0;
- if (dentry && dentry->d_op && dentry->d_op->d_revalidate)
- dentry = do_revalidate(dentry, nd);
+ status = dentry->d_op->d_revalidate(dentry, nd);
+ if (status > 0)
+ return 0;
- return dentry;
+ if (!status) {
+ d_invalidate(dentry);
+ status = -ESTALE;
+ }
+ return status;
}
/*
- * Short-cut version of permission(), for calling by
- * path_walk(), when dcache lock is held. Combines parts
- * of permission() and generic_permission(), and tests ONLY for
- * MAY_EXEC permission.
+ * Short-cut version of permission(), for calling on directories
+ * during pathname resolution. Combines parts of permission()
+ * and generic_permission(), and tests ONLY for MAY_EXEC permission.
*
* If appropriate, check DAC only. If not appropriate, or
- * short-cut DAC fails, then call permission() to do more
+ * short-cut DAC fails, then call ->permission() to do more
* complete permission check.
*/
-static int exec_permission_lite(struct inode *inode)
+static int exec_permission(struct inode *inode)
{
- umode_t mode = inode->i_mode;
-
- if (inode->i_op->permission)
- return -EAGAIN;
+ int ret;
- if (current_fsuid() == inode->i_uid)
- mode >>= 6;
- else if (in_group_p(inode->i_gid))
- mode >>= 3;
-
- if (mode & MAY_EXEC)
- goto ok;
-
- if ((inode->i_mode & S_IXUGO) && capable(CAP_DAC_OVERRIDE))
- goto ok;
-
- if (S_ISDIR(inode->i_mode) && capable(CAP_DAC_OVERRIDE))
+ if (inode->i_op->permission) {
+ ret = inode->i_op->permission(inode, MAY_EXEC);
+ if (!ret)
+ goto ok;
+ return ret;
+ }
+ ret = acl_permission_check(inode, MAY_EXEC, inode->i_op->check_acl);
+ if (!ret)
goto ok;
- if (S_ISDIR(inode->i_mode) && capable(CAP_DAC_READ_SEARCH))
+ if (capable(CAP_DAC_OVERRIDE) || capable(CAP_DAC_READ_SEARCH))
goto ok;
- return -EACCES;
+ return ret;
ok:
return security_inode_permission(inode, MAY_EXEC);
}
-/*
- * This is called when everything else fails, and we actually have
- * to go to the low-level filesystem to find out what we should do..
- *
- * We get the directory semaphore, and after getting that we also
- * make sure that nobody added the entry to the dcache in the meantime..
- * SMP-safe
- */
-static struct dentry * real_lookup(struct dentry * parent, struct qstr * name, struct nameidata *nd)
-{
- struct dentry * result;
- struct inode *dir = parent->d_inode;
-
- mutex_lock(&dir->i_mutex);
- /*
- * First re-do the cached lookup just in case it was created
- * while we waited for the directory semaphore..
- *
- * FIXME! This could use version numbering or similar to
- * avoid unnecessary cache lookups.
- *
- * The "dcache_lock" is purely to protect the RCU list walker
- * from concurrent renames at this point (we mustn't get false
- * negatives from the RCU list walk here, unlike the optimistic
- * fast walk).
- *
- * so doing d_lookup() (with seqlock), instead of lockfree __d_lookup
- */
- result = d_lookup(parent, name);
- if (!result) {
- struct dentry *dentry;
-
- /* Don't create child dentry for a dead directory. */
- result = ERR_PTR(-ENOENT);
- if (IS_DEADDIR(dir))
- goto out_unlock;
-
- dentry = d_alloc(parent, name);
- result = ERR_PTR(-ENOMEM);
- if (dentry) {
- result = dir->i_op->lookup(dir, dentry, nd);
- if (result)
- dput(dentry);
- else
- result = dentry;
- }
-out_unlock:
- mutex_unlock(&dir->i_mutex);
- return result;
- }
-
- /*
- * Uhhuh! Nasty case: the cache was re-populated while
- * we waited on the semaphore. Need to revalidate.
- */
- mutex_unlock(&dir->i_mutex);
- if (result->d_op && result->d_op->d_revalidate) {
- result = do_revalidate(result, nd);
- if (!result)
- result = ERR_PTR(-ENOENT);
- }
- return result;
-}
-
-/*
- * Wrapper to retry pathname resolution whenever the underlying
- * file system returns an ESTALE.
- *
- * Retry the whole path once, forcing real lookup requests
- * instead of relying on the dcache.
- */
-static __always_inline int link_path_walk(const char *name, struct nameidata *nd)
-{
- struct path save = nd->path;
- int result;
-
- /* make sure the stuff we saved doesn't go away */
- path_get(&save);
-
- result = __link_path_walk(name, nd);
- if (result == -ESTALE) {
- /* nd->path had been dropped */
- nd->path = save;
- path_get(&nd->path);
- nd->flags |= LOOKUP_REVAL;
- result = __link_path_walk(name, nd);
- }
-
- path_put(&save);
-
- return result;
-}
-
static __always_inline void set_root(struct nameidata *nd)
{
if (!nd->root.mnt) {
}
}
+static int link_path_walk(const char *, struct nameidata *);
+
static __always_inline int __vfs_follow_link(struct nameidata *nd, const char *link)
{
int res = 0;
dget(dentry);
}
mntget(path->mnt);
+ nd->last_type = LAST_BIND;
cookie = dentry->d_inode->i_op->follow_link(dentry, nd);
error = PTR_ERR(cookie);
if (!IS_ERR(cookie)) {
error = 0;
if (s)
error = __vfs_follow_link(nd, s);
+ else if (nd->last_type == LAST_BIND) {
+ error = force_reval_path(&nd->path, nd);
+ if (error)
+ path_put(&nd->path);
+ }
if (dentry->d_inode->i_op->put_link)
dentry->d_inode->i_op->put_link(dentry, nd, cookie);
}
- path_put(path);
-
return error;
}
current->total_link_count++;
nd->depth++;
err = __do_follow_link(path, nd);
+ path_put(path);
current->link_count--;
nd->depth--;
return err;
return err;
}
-int follow_up(struct vfsmount **mnt, struct dentry **dentry)
+int follow_up(struct path *path)
{
struct vfsmount *parent;
struct dentry *mountpoint;
spin_lock(&vfsmount_lock);
- parent=(*mnt)->mnt_parent;
- if (parent == *mnt) {
+ parent = path->mnt->mnt_parent;
+ if (parent == path->mnt) {
spin_unlock(&vfsmount_lock);
return 0;
}
mntget(parent);
- mountpoint=dget((*mnt)->mnt_mountpoint);
+ mountpoint = dget(path->mnt->mnt_mountpoint);
spin_unlock(&vfsmount_lock);
- dput(*dentry);
- *dentry = mountpoint;
- mntput(*mnt);
- *mnt = parent;
+ dput(path->dentry);
+ path->dentry = mountpoint;
+ mntput(path->mnt);
+ path->mnt = parent;
return 1;
}
{
int res = 0;
while (d_mountpoint(path->dentry)) {
- struct vfsmount *mounted = lookup_mnt(path->mnt, path->dentry);
+ struct vfsmount *mounted = lookup_mnt(path);
if (!mounted)
break;
dput(path->dentry);
return res;
}
-static void follow_mount(struct vfsmount **mnt, struct dentry **dentry)
+static void follow_mount(struct path *path)
{
- while (d_mountpoint(*dentry)) {
- struct vfsmount *mounted = lookup_mnt(*mnt, *dentry);
+ while (d_mountpoint(path->dentry)) {
+ struct vfsmount *mounted = lookup_mnt(path);
if (!mounted)
break;
- dput(*dentry);
- mntput(*mnt);
- *mnt = mounted;
- *dentry = dget(mounted->mnt_root);
+ dput(path->dentry);
+ mntput(path->mnt);
+ path->mnt = mounted;
+ path->dentry = dget(mounted->mnt_root);
}
}
/* no need for dcache_lock, as serialization is taken care in
* namespace.c
*/
-int follow_down(struct vfsmount **mnt, struct dentry **dentry)
+int follow_down(struct path *path)
{
struct vfsmount *mounted;
- mounted = lookup_mnt(*mnt, *dentry);
+ mounted = lookup_mnt(path);
if (mounted) {
- dput(*dentry);
- mntput(*mnt);
- *mnt = mounted;
- *dentry = dget(mounted->mnt_root);
+ dput(path->dentry);
+ mntput(path->mnt);
+ path->mnt = mounted;
+ path->dentry = dget(mounted->mnt_root);
return 1;
}
return 0;
set_root(nd);
while(1) {
- struct vfsmount *parent;
struct dentry *old = nd->path.dentry;
if (nd->path.dentry == nd->root.dentry &&
nd->path.mnt == nd->root.mnt) {
break;
}
- spin_lock(&dcache_lock);
if (nd->path.dentry != nd->path.mnt->mnt_root) {
- nd->path.dentry = dget(nd->path.dentry->d_parent);
- spin_unlock(&dcache_lock);
+ /* rare case of legitimate dget_parent()... */
+ nd->path.dentry = dget_parent(nd->path.dentry);
dput(old);
break;
}
- spin_unlock(&dcache_lock);
- spin_lock(&vfsmount_lock);
- parent = nd->path.mnt->mnt_parent;
- if (parent == nd->path.mnt) {
- spin_unlock(&vfsmount_lock);
+ if (!follow_up(&nd->path))
break;
- }
- mntget(parent);
- nd->path.dentry = dget(nd->path.mnt->mnt_mountpoint);
- spin_unlock(&vfsmount_lock);
- dput(old);
- mntput(nd->path.mnt);
- nd->path.mnt = parent;
}
- follow_mount(&nd->path.mnt, &nd->path.dentry);
+ follow_mount(&nd->path);
}
/*
struct path *path)
{
struct vfsmount *mnt = nd->path.mnt;
- struct dentry *dentry = __d_lookup(nd->path.dentry, name);
+ struct dentry *dentry, *parent;
+ struct inode *dir;
+ /*
+ * See if the low-level filesystem might want
+ * to use its own hash..
+ */
+ if (nd->path.dentry->d_op && nd->path.dentry->d_op->d_hash) {
+ int err = nd->path.dentry->d_op->d_hash(nd->path.dentry, name);
+ if (err < 0)
+ return err;
+ }
+ dentry = __d_lookup(nd->path.dentry, name);
if (!dentry)
goto need_lookup;
if (dentry->d_op && dentry->d_op->d_revalidate)
return 0;
need_lookup:
- dentry = real_lookup(nd->path.dentry, name, nd);
+ parent = nd->path.dentry;
+ dir = parent->d_inode;
+
+ mutex_lock(&dir->i_mutex);
+ /*
+ * First re-do the cached lookup just in case it was created
+ * while we waited for the directory semaphore..
+ *
+ * FIXME! This could use version numbering or similar to
+ * avoid unnecessary cache lookups.
+ *
+ * The "dcache_lock" is purely to protect the RCU list walker
+ * from concurrent renames at this point (we mustn't get false
+ * negatives from the RCU list walk here, unlike the optimistic
+ * fast walk).
+ *
+ * so doing d_lookup() (with seqlock), instead of lockfree __d_lookup
+ */
+ dentry = d_lookup(parent, name);
+ if (!dentry) {
+ struct dentry *new;
+
+ /* Don't create child dentry for a dead directory. */
+ dentry = ERR_PTR(-ENOENT);
+ if (IS_DEADDIR(dir))
+ goto out_unlock;
+
+ new = d_alloc(parent, name);
+ dentry = ERR_PTR(-ENOMEM);
+ if (new) {
+ dentry = dir->i_op->lookup(dir, new, nd);
+ if (dentry)
+ dput(new);
+ else
+ dentry = new;
+ }
+out_unlock:
+ mutex_unlock(&dir->i_mutex);
+ if (IS_ERR(dentry))
+ goto fail;
+ goto done;
+ }
+
+ /*
+ * Uhhuh! Nasty case: the cache was re-populated while
+ * we waited on the semaphore. Need to revalidate.
+ */
+ mutex_unlock(&dir->i_mutex);
+ if (dentry->d_op && dentry->d_op->d_revalidate) {
+ dentry = do_revalidate(dentry, nd);
+ if (!dentry)
+ dentry = ERR_PTR(-ENOENT);
+ }
if (IS_ERR(dentry))
goto fail;
goto done;
}
/*
+ * This is a temporary kludge to deal with "automount" symlinks; proper
+ * solution is to trigger them on follow_mount(), so that do_lookup()
+ * would DTRT. To be killed before 2.6.34-final.
+ */
+static inline int follow_on_final(struct inode *inode, unsigned lookup_flags)
+{
+ return inode && unlikely(inode->i_op->follow_link) &&
+ ((lookup_flags & LOOKUP_FOLLOW) || S_ISDIR(inode->i_mode));
+}
+
+/*
* Name resolution.
* This is the basic name resolution function, turning a pathname into
* the final dentry. We expect 'base' to be positive and a directory.
* Returns 0 and nd will have valid dentry and mnt on success.
* Returns error and drops reference to input namei data on failure.
*/
-static int __link_path_walk(const char *name, struct nameidata *nd)
+static int link_path_walk(const char *name, struct nameidata *nd)
{
struct path next;
struct inode *inode;
unsigned int c;
nd->flags |= LOOKUP_CONTINUE;
- err = exec_permission_lite(inode);
- if (err == -EAGAIN)
- err = inode_permission(nd->path.dentry->d_inode,
- MAY_EXEC);
- if (!err)
- err = ima_path_check(&nd->path, MAY_EXEC,
- IMA_COUNT_UPDATE);
+ err = exec_permission(inode);
if (err)
break;
case 1:
continue;
}
- /*
- * See if the low-level filesystem might want
- * to use its own hash..
- */
- if (nd->path.dentry->d_op && nd->path.dentry->d_op->d_hash) {
- err = nd->path.dentry->d_op->d_hash(nd->path.dentry,
- &this);
- if (err < 0)
- break;
- }
/* This does the actual lookups.. */
err = do_lookup(nd, &this, &next);
if (err)
case 1:
goto return_reval;
}
- if (nd->path.dentry->d_op && nd->path.dentry->d_op->d_hash) {
- err = nd->path.dentry->d_op->d_hash(nd->path.dentry,
- &this);
- if (err < 0)
- break;
- }
err = do_lookup(nd, &this, &next);
if (err)
break;
inode = next.dentry->d_inode;
- if ((lookup_flags & LOOKUP_FOLLOW)
- && inode && inode->i_op->follow_link) {
+ if (follow_on_final(inode, lookup_flags)) {
err = do_follow_link(&next, nd);
if (err)
goto return_err;
static int path_walk(const char *name, struct nameidata *nd)
{
+ struct path save = nd->path;
+ int result;
+
current->total_link_count = 0;
- return link_path_walk(name, nd);
+
+ /* make sure the stuff we saved doesn't go away */
+ path_get(&save);
+
+ result = link_path_walk(name, nd);
+ if (result == -ESTALE) {
+ /* nd->path had been dropped */
+ current->total_link_count = 0;
+ nd->path = save;
+ path_get(&nd->path);
+ nd->flags |= LOOKUP_REVAL;
+ result = link_path_walk(name, nd);
+ }
+
+ path_put(&save);
+
+ return result;
}
static int path_init(int dfd, const char *name, unsigned int flags, struct nameidata *nd)
nd->last_type = LAST_ROOT;
nd->flags = flags;
nd->depth = 0;
- nd->root.mnt = NULL;
nd->path.dentry = dentry;
nd->path.mnt = mnt;
path_get(&nd->path);
+ nd->root = nd->path;
+ path_get(&nd->root);
retval = path_walk(name, nd);
if (unlikely(!retval && !audit_dummy_context() && nd->path.dentry &&
nd->path.dentry->d_inode))
audit_inode(name, nd->path.dentry);
- if (nd->root.mnt) {
- path_put(&nd->root);
- nd->root.mnt = NULL;
- }
+ path_put(&nd->root);
+ nd->root.mnt = NULL;
return retval;
}
-/**
- * path_lookup_open - lookup a file path with open intent
- * @dfd: the directory to use as base, or AT_FDCWD
- * @name: pointer to file name
- * @lookup_flags: lookup intent flags
- * @nd: pointer to nameidata
- * @open_flags: open intent flags
- */
-static int path_lookup_open(int dfd, const char *name,
- unsigned int lookup_flags, struct nameidata *nd, int open_flags)
-{
- struct file *filp = get_empty_filp();
- int err;
-
- if (filp == NULL)
- return -ENFILE;
- nd->intent.open.file = filp;
- nd->intent.open.flags = open_flags;
- nd->intent.open.create_mode = 0;
- err = do_path_lookup(dfd, name, lookup_flags|LOOKUP_OPEN, nd);
- if (IS_ERR(nd->intent.open.file)) {
- if (err == 0) {
- err = PTR_ERR(nd->intent.open.file);
- path_put(&nd->path);
- }
- } else if (err != 0)
- release_open_intent(nd);
- return err;
-}
-
static struct dentry *__lookup_hash(struct qstr *name,
struct dentry *base, struct nameidata *nd)
{
goto out;
}
- dentry = cached_lookup(base, name, nd);
+ dentry = __d_lookup(base, name);
+
+ /* lockess __d_lookup may fail due to concurrent d_move()
+ * in some unrelated directory, so try with d_lookup
+ */
+ if (!dentry)
+ dentry = d_lookup(base, name);
+
+ if (dentry && dentry->d_op && dentry->d_op->d_revalidate)
+ dentry = do_revalidate(dentry, nd);
+
if (!dentry) {
struct dentry *new;
{
int err;
- err = inode_permission(nd->path.dentry->d_inode, MAY_EXEC);
+ err = exec_permission(nd->path.dentry->d_inode);
if (err)
return ERR_PTR(err);
return __lookup_hash(&nd->last, nd->path.dentry, nd);
if (err)
return ERR_PTR(err);
- err = inode_permission(base->d_inode, MAY_EXEC);
- if (err)
- return ERR_PTR(err);
- return __lookup_hash(&this, base, NULL);
-}
-
-/**
- * lookup_one_noperm - bad hack for sysfs
- * @name: pathname component to lookup
- * @base: base directory to lookup from
- *
- * This is a variant of lookup_one_len that doesn't perform any permission
- * checks. It's a horrible hack to work around the braindead sysfs
- * architecture and should not be used anywhere else.
- *
- * DON'T USE THIS FUNCTION EVER, thanks.
- */
-struct dentry *lookup_one_noperm(const char *name, struct dentry *base)
-{
- int err;
- struct qstr this;
-
- err = __lookup_one_len(name, &this, base, strlen(name));
+ err = exec_permission(base->d_inode);
if (err)
return ERR_PTR(err);
return __lookup_hash(&this, base, NULL);
return -ENOENT;
BUG_ON(victim->d_parent->d_inode != dir);
- audit_inode_child(victim->d_name.name, victim, dir);
+ audit_inode_child(victim, dir);
error = inode_permission(dir, MAY_WRITE | MAY_EXEC);
if (error)
if (error)
return error;
- error = ima_path_check(path,
- acc_mode & (MAY_READ | MAY_WRITE | MAY_EXEC),
- IMA_COUNT_UPDATE);
- if (error)
- return error;
/*
* An append-only file must be opened in append mode for writing.
*/
if (IS_APPEND(inode)) {
- if ((flag & FMODE_WRITE) && !(flag & O_APPEND))
+ if ((flag & O_ACCMODE) != O_RDONLY && !(flag & O_APPEND))
return -EPERM;
if (flag & O_TRUNC)
return -EPERM;
}
/* O_NOATIME can only be set by the owner or superuser */
- if (flag & O_NOATIME)
- if (!is_owner_or_cap(inode))
- return -EPERM;
+ if (flag & O_NOATIME && !is_owner_or_cap(inode))
+ return -EPERM;
/*
* Ensure there are no outstanding leases on the file.
*/
- error = break_lease(inode, flag);
+ return break_lease(inode, flag);
+}
+
+static int handle_truncate(struct path *path)
+{
+ struct inode *inode = path->dentry->d_inode;
+ int error = get_write_access(inode);
if (error)
return error;
-
- if (flag & O_TRUNC) {
- error = get_write_access(inode);
- if (error)
- return error;
-
- /*
- * Refuse to truncate files with mandatory locks held on them.
- */
- error = locks_verify_locked(inode);
- if (!error)
- error = security_path_truncate(path, 0,
- ATTR_MTIME|ATTR_CTIME|ATTR_OPEN);
- if (!error) {
- vfs_dq_init(inode);
-
- error = do_truncate(dentry, 0,
- ATTR_MTIME|ATTR_CTIME|ATTR_OPEN,
- NULL);
- }
- put_write_access(inode);
- if (error)
- return error;
- } else
- if (flag & FMODE_WRITE)
- vfs_dq_init(inode);
-
- return 0;
+ /*
+ * Refuse to truncate files with mandatory locks held on them.
+ */
+ error = locks_verify_locked(inode);
+ if (!error)
+ error = security_path_truncate(path, 0,
+ ATTR_MTIME|ATTR_CTIME|ATTR_OPEN);
+ if (!error) {
+ error = do_truncate(path->dentry, 0,
+ ATTR_MTIME|ATTR_CTIME|ATTR_OPEN,
+ NULL);
+ }
+ put_write_access(inode);
+ return error;
}
/*
* what get passed to sys_open().
*/
static int __open_namei_create(struct nameidata *nd, struct path *path,
- int flag, int mode)
+ int open_flag, int mode)
{
int error;
struct dentry *dir = nd->path.dentry;
if (error)
return error;
/* Don't check for write permission, don't truncate */
- return may_open(&nd->path, 0, flag & ~O_TRUNC);
+ return may_open(&nd->path, 0, open_flag & ~O_TRUNC);
}
/*
return flag;
}
-static int open_will_write_to_fs(int flag, struct inode *inode)
+static int open_will_truncate(int flag, struct inode *inode)
{
/*
* We'll never write to the fs underlying
return (flag & O_TRUNC);
}
-/*
- * Note that the low bits of the passed in "open_flag"
- * are not the same as in the local variable "flag". See
- * open_to_namei_flags() for more details.
- */
-struct file *do_filp_open(int dfd, const char *pathname,
- int open_flag, int mode, int acc_mode)
+static struct file *finish_open(struct nameidata *nd,
+ int open_flag, int acc_mode)
{
struct file *filp;
- struct nameidata nd;
+ int will_truncate;
int error;
- struct path path;
- struct dentry *dir;
- int count = 0;
- int will_write;
- int flag = open_to_namei_flags(open_flag);
-
- if (!acc_mode)
- acc_mode = MAY_OPEN | ACC_MODE(flag);
- /* O_TRUNC implies we need access checks for write permissions */
- if (flag & O_TRUNC)
- acc_mode |= MAY_WRITE;
-
- /* Allow the LSM permission hook to distinguish append
- access from general write access. */
- if (flag & O_APPEND)
- acc_mode |= MAY_APPEND;
-
- /*
- * The simplest case - just a plain lookup.
- */
- if (!(flag & O_CREAT)) {
- error = path_lookup_open(dfd, pathname, lookup_flags(flag),
- &nd, flag);
+ will_truncate = open_will_truncate(open_flag, nd->path.dentry->d_inode);
+ if (will_truncate) {
+ error = mnt_want_write(nd->path.mnt);
if (error)
- return ERR_PTR(error);
- goto ok;
+ goto exit;
+ }
+ error = may_open(&nd->path, acc_mode, open_flag);
+ if (error) {
+ if (will_truncate)
+ mnt_drop_write(nd->path.mnt);
+ goto exit;
+ }
+ filp = nameidata_to_filp(nd);
+ if (!IS_ERR(filp)) {
+ error = ima_file_check(filp, acc_mode);
+ if (error) {
+ fput(filp);
+ filp = ERR_PTR(error);
+ }
}
+ if (!IS_ERR(filp)) {
+ if (acc_mode & MAY_WRITE)
+ vfs_dq_init(nd->path.dentry->d_inode);
+ if (will_truncate) {
+ error = handle_truncate(&nd->path);
+ if (error) {
+ fput(filp);
+ filp = ERR_PTR(error);
+ }
+ }
+ }
/*
- * Create - we need to know the parent.
+ * It is now safe to drop the mnt write
+ * because the filp has had a write taken
+ * on its behalf.
*/
- error = path_init(dfd, pathname, LOOKUP_PARENT, &nd);
- if (error)
- return ERR_PTR(error);
- error = path_walk(pathname, &nd);
- if (error)
- return ERR_PTR(error);
- if (unlikely(!audit_dummy_context()))
- audit_inode(pathname, nd.path.dentry);
+ if (will_truncate)
+ mnt_drop_write(nd->path.mnt);
+ return filp;
+
+exit:
+ if (!IS_ERR(nd->intent.open.file))
+ release_open_intent(nd);
+ path_put(&nd->path);
+ return ERR_PTR(error);
+}
+
+static struct file *do_last(struct nameidata *nd, struct path *path,
+ int open_flag, int acc_mode,
+ int mode, const char *pathname)
+{
+ struct dentry *dir = nd->path.dentry;
+ struct file *filp;
+ int error;
+
+ if (nd->last_type == LAST_BIND)
+ goto ok;
- /*
- * We have the parent and last component. First of all, check
- * that we are not asked to creat(2) an obvious directory - that
- * will not do.
- */
error = -EISDIR;
- if (nd.last_type != LAST_NORM || nd.last.name[nd.last.len])
- goto exit_parent;
+ if (nd->last_type != LAST_NORM || nd->last.name[nd->last.len])
+ goto exit;
- error = -ENFILE;
- filp = get_empty_filp();
- if (filp == NULL)
- goto exit_parent;
- nd.intent.open.file = filp;
- nd.intent.open.flags = flag;
- nd.intent.open.create_mode = mode;
- dir = nd.path.dentry;
- nd.flags &= ~LOOKUP_PARENT;
- nd.flags |= LOOKUP_CREATE | LOOKUP_OPEN;
- if (flag & O_EXCL)
- nd.flags |= LOOKUP_EXCL;
mutex_lock(&dir->d_inode->i_mutex);
- path.dentry = lookup_hash(&nd);
- path.mnt = nd.path.mnt;
-do_last:
- error = PTR_ERR(path.dentry);
- if (IS_ERR(path.dentry)) {
+ path->dentry = lookup_hash(nd);
+ path->mnt = nd->path.mnt;
+
+ error = PTR_ERR(path->dentry);
+ if (IS_ERR(path->dentry)) {
mutex_unlock(&dir->d_inode->i_mutex);
goto exit;
}
- if (IS_ERR(nd.intent.open.file)) {
- error = PTR_ERR(nd.intent.open.file);
+ if (IS_ERR(nd->intent.open.file)) {
+ error = PTR_ERR(nd->intent.open.file);
goto exit_mutex_unlock;
}
/* Negative dentry, just create the file */
- if (!path.dentry->d_inode) {
+ if (!path->dentry->d_inode) {
/*
* This write is needed to ensure that a
* ro->rw transition does not occur between
* a permanent write count is taken through
* the 'struct file' in nameidata_to_filp().
*/
- error = mnt_want_write(nd.path.mnt);
+ error = mnt_want_write(nd->path.mnt);
if (error)
goto exit_mutex_unlock;
- error = __open_namei_create(&nd, &path, flag, mode);
+ error = __open_namei_create(nd, path, open_flag, mode);
if (error) {
- mnt_drop_write(nd.path.mnt);
+ mnt_drop_write(nd->path.mnt);
goto exit;
}
- filp = nameidata_to_filp(&nd, open_flag);
- mnt_drop_write(nd.path.mnt);
+ filp = nameidata_to_filp(nd);
+ mnt_drop_write(nd->path.mnt);
+ if (!IS_ERR(filp)) {
+ error = ima_file_check(filp, acc_mode);
+ if (error) {
+ fput(filp);
+ filp = ERR_PTR(error);
+ }
+ }
return filp;
}
* It already exists.
*/
mutex_unlock(&dir->d_inode->i_mutex);
- audit_inode(pathname, path.dentry);
+ audit_inode(pathname, path->dentry);
error = -EEXIST;
- if (flag & O_EXCL)
+ if (open_flag & O_EXCL)
goto exit_dput;
- if (__follow_mount(&path)) {
+ if (__follow_mount(path)) {
error = -ELOOP;
- if (flag & O_NOFOLLOW)
+ if (open_flag & O_NOFOLLOW)
goto exit_dput;
}
error = -ENOENT;
- if (!path.dentry->d_inode)
+ if (!path->dentry->d_inode)
goto exit_dput;
- if (path.dentry->d_inode->i_op->follow_link)
- goto do_link;
- path_to_nameidata(&path, &nd);
+ if (path->dentry->d_inode->i_op->follow_link)
+ return NULL;
+
+ path_to_nameidata(path, nd);
error = -EISDIR;
- if (path.dentry->d_inode && S_ISDIR(path.dentry->d_inode->i_mode))
+ if (S_ISDIR(path->dentry->d_inode->i_mode))
goto exit;
ok:
+ filp = finish_open(nd, open_flag, acc_mode);
+ return filp;
+
+exit_mutex_unlock:
+ mutex_unlock(&dir->d_inode->i_mutex);
+exit_dput:
+ path_put_conditional(path, nd);
+exit:
+ if (!IS_ERR(nd->intent.open.file))
+ release_open_intent(nd);
+ path_put(&nd->path);
+ return ERR_PTR(error);
+}
+
+/*
+ * Note that the low bits of the passed in "open_flag"
+ * are not the same as in the local variable "flag". See
+ * open_to_namei_flags() for more details.
+ */
+struct file *do_filp_open(int dfd, const char *pathname,
+ int open_flag, int mode, int acc_mode)
+{
+ struct file *filp;
+ struct nameidata nd;
+ int error;
+ struct path path;
+ int count = 0;
+ int flag = open_to_namei_flags(open_flag);
+ int force_reval = 0;
+
/*
- * Consider:
- * 1. may_open() truncates a file
- * 2. a rw->ro mount transition occurs
- * 3. nameidata_to_filp() fails due to
- * the ro mount.
- * That would be inconsistent, and should
- * be avoided. Taking this mnt write here
- * ensures that (2) can not occur.
+ * O_SYNC is implemented as __O_SYNC|O_DSYNC. As many places only
+ * check for O_DSYNC if the need any syncing at all we enforce it's
+ * always set instead of having to deal with possibly weird behaviour
+ * for malicious applications setting only __O_SYNC.
*/
- will_write = open_will_write_to_fs(flag, nd.path.dentry->d_inode);
- if (will_write) {
- error = mnt_want_write(nd.path.mnt);
+ if (open_flag & __O_SYNC)
+ open_flag |= O_DSYNC;
+
+ if (!acc_mode)
+ acc_mode = MAY_OPEN | ACC_MODE(open_flag);
+
+ /* O_TRUNC implies we need access checks for write permissions */
+ if (open_flag & O_TRUNC)
+ acc_mode |= MAY_WRITE;
+
+ /* Allow the LSM permission hook to distinguish append
+ access from general write access. */
+ if (open_flag & O_APPEND)
+ acc_mode |= MAY_APPEND;
+
+ /*
+ * The simplest case - just a plain lookup.
+ */
+ if (!(open_flag & O_CREAT)) {
+ filp = get_empty_filp();
+
+ if (filp == NULL)
+ return ERR_PTR(-ENFILE);
+ nd.intent.open.file = filp;
+ filp->f_flags = open_flag;
+ nd.intent.open.flags = flag;
+ nd.intent.open.create_mode = 0;
+ error = do_path_lookup(dfd, pathname,
+ lookup_flags(open_flag)|LOOKUP_OPEN, &nd);
+ if (IS_ERR(nd.intent.open.file)) {
+ if (error == 0) {
+ error = PTR_ERR(nd.intent.open.file);
+ path_put(&nd.path);
+ }
+ } else if (error)
+ release_open_intent(&nd);
if (error)
- goto exit;
+ return ERR_PTR(error);
+ return finish_open(&nd, open_flag, acc_mode);
}
- error = may_open(&nd.path, acc_mode, flag);
+
+ /*
+ * Create - we need to know the parent.
+ */
+reval:
+ error = path_init(dfd, pathname, LOOKUP_PARENT, &nd);
+ if (error)
+ return ERR_PTR(error);
+ if (force_reval)
+ nd.flags |= LOOKUP_REVAL;
+ error = path_walk(pathname, &nd);
if (error) {
- if (will_write)
- mnt_drop_write(nd.path.mnt);
- goto exit;
+ if (nd.root.mnt)
+ path_put(&nd.root);
+ return ERR_PTR(error);
}
- filp = nameidata_to_filp(&nd, open_flag);
+ if (unlikely(!audit_dummy_context()))
+ audit_inode(pathname, nd.path.dentry);
+
/*
- * It is now safe to drop the mnt write
- * because the filp has had a write taken
- * on its behalf.
+ * We have the parent and last component.
*/
- if (will_write)
- mnt_drop_write(nd.path.mnt);
+
+ error = -ENFILE;
+ filp = get_empty_filp();
+ if (filp == NULL)
+ goto exit_parent;
+ nd.intent.open.file = filp;
+ filp->f_flags = open_flag;
+ nd.intent.open.flags = flag;
+ nd.intent.open.create_mode = mode;
+ nd.flags &= ~LOOKUP_PARENT;
+ nd.flags |= LOOKUP_CREATE | LOOKUP_OPEN;
+ if (open_flag & O_EXCL)
+ nd.flags |= LOOKUP_EXCL;
+ filp = do_last(&nd, &path, open_flag, acc_mode, mode, pathname);
+ if (!filp)
+ goto do_link;
+ if (nd.root.mnt)
+ path_put(&nd.root);
return filp;
-exit_mutex_unlock:
- mutex_unlock(&dir->d_inode->i_mutex);
exit_dput:
path_put_conditional(&path, &nd);
-exit:
if (!IS_ERR(nd.intent.open.file))
release_open_intent(&nd);
exit_parent:
+ path_put(&nd.path);
if (nd.root.mnt)
path_put(&nd.root);
- path_put(&nd.path);
return ERR_PTR(error);
do_link:
error = -ELOOP;
- if (flag & O_NOFOLLOW)
+ if ((open_flag & O_NOFOLLOW) || count++ == 32)
goto exit_dput;
/*
* This is subtle. Instead of calling do_follow_link() we do the
if (error)
goto exit_dput;
error = __do_follow_link(&path, &nd);
+ path_put(&path);
if (error) {
/* Does someone understand code flow here? Or it is only
* me so stupid? Anathema to whoever designed this non-sense
* with "intent.open".
*/
release_open_intent(&nd);
+ if (nd.root.mnt)
+ path_put(&nd.root);
+ if (error == -ESTALE && !force_reval) {
+ force_reval = 1;
+ goto reval;
+ }
return ERR_PTR(error);
}
nd.flags &= ~LOOKUP_PARENT;
- if (nd.last_type == LAST_BIND)
- goto ok;
- error = -EISDIR;
- if (nd.last_type != LAST_NORM)
- goto exit;
- if (nd.last.name[nd.last.len]) {
+ filp = do_last(&nd, &path, open_flag, acc_mode, mode, pathname);
+ if (nd.last_type == LAST_NORM)
__putname(nd.last.name);
- goto exit;
- }
- error = -ELOOP;
- if (count++==32) {
- __putname(nd.last.name);
- goto exit;
- }
- dir = nd.path.dentry;
- mutex_lock(&dir->d_inode->i_mutex);
- path.dentry = lookup_hash(&nd);
- path.mnt = nd.path.mnt;
- __putname(nd.last.name);
- goto do_last;
+ if (!filp)
+ goto do_link;
+ if (nd.root.mnt)
+ path_put(&nd.root);
+ return filp;
}
/**
error = -EBUSY;
else {
error = security_inode_unlink(dir, dentry);
- if (!error)
+ if (!error) {
error = dir->i_op->unlink(dir, dentry);
+ if (!error)
+ dentry->d_inode->i_flags |= S_DEAD;
+ }
}
mutex_unlock(&dentry->d_inode->i_mutex);
else
error = old_dir->i_op->rename(old_dir, old_dentry, new_dir, new_dentry);
if (!error) {
+ if (target)
+ target->i_flags |= S_DEAD;
if (!(old_dir->i_sb->s_type->fs_flags & FS_RENAME_DOES_D_MOVE))
d_move(old_dentry, new_dentry);
}
error = vfs_rename_dir(old_dir,old_dentry,new_dir,new_dentry);
else
error = vfs_rename_other(old_dir,old_dentry,new_dir,new_dentry);
- if (!error) {
- const char *new_name = old_dentry->d_name.name;
- fsnotify_move(old_dir, new_dir, old_name, new_name, is_dir,
+ if (!error)
+ fsnotify_move(old_dir, new_dir, old_name, is_dir,
new_dentry->d_inode, old_dentry);
- }
fsnotify_oldname_free(old_name);
return error;
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff