aboutsummaryrefslogtreecommitdiff
path: root/ipc/mqueue.c
diff options
context:
space:
mode:
Diffstat (limited to 'ipc/mqueue.c')
-rw-r--r--ipc/mqueue.c298
1 files changed, 140 insertions, 158 deletions
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 2dee38d53c7..4fcf39af177 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -6,7 +6,7 @@
*
* Spinlocks: Mohamed Abbas (abbas.mohamed@intel.com)
* Lockless receive & send, fd based notify:
- * Manfred Spraul (manfred@colorfullife.com)
+ * Manfred Spraul (manfred@colorfullife.com)
*
* Audit: George Wilson (ltcgcw@us.ibm.com)
*
@@ -73,7 +73,7 @@ struct mqueue_inode_info {
struct mq_attr attr;
struct sigevent notify;
- struct pid* notify_owner;
+ struct pid *notify_owner;
struct user_namespace *notify_user_ns;
struct user_struct *user; /* user who created, for accounting */
struct sock *notify_sock;
@@ -92,7 +92,7 @@ static void remove_notification(struct mqueue_inode_info *info);
static struct kmem_cache *mqueue_inode_cachep;
-static struct ctl_table_header * mq_sysctl_table;
+static struct ctl_table_header *mq_sysctl_table;
static inline struct mqueue_inode_info *MQUEUE_I(struct inode *inode)
{
@@ -142,7 +142,6 @@ static int msg_insert(struct msg_msg *msg, struct mqueue_inode_info *info)
leaf = kmalloc(sizeof(*leaf), GFP_ATOMIC);
if (!leaf)
return -ENOMEM;
- rb_init_node(&leaf->rb_node);
INIT_LIST_HEAD(&leaf->msg_list);
info->qsize += sizeof(*leaf);
}
@@ -331,8 +330,16 @@ static struct dentry *mqueue_mount(struct file_system_type *fs_type,
int flags, const char *dev_name,
void *data)
{
- if (!(flags & MS_KERNMOUNT))
- data = current->nsproxy->ipc_ns;
+ if (!(flags & MS_KERNMOUNT)) {
+ struct ipc_namespace *ns = current->nsproxy->ipc_ns;
+ /* Don't allow mounting unless the caller has CAP_SYS_ADMIN
+ * over the ipc namespace.
+ */
+ if (!ns_capable(ns->user_ns, CAP_SYS_ADMIN))
+ return ERR_PTR(-EPERM);
+
+ data = ns;
+ }
return mount_ns(fs_type, flags, data, mqueue_fill_super);
}
@@ -426,9 +433,9 @@ static int mqueue_create(struct inode *dir, struct dentry *dentry,
error = -EACCES;
goto out_unlock;
}
- if (ipc_ns->mq_queues_count >= HARD_QUEUESMAX ||
- (ipc_ns->mq_queues_count >= ipc_ns->mq_queues_max &&
- !capable(CAP_SYS_RESOURCE))) {
+
+ if (ipc_ns->mq_queues_count >= ipc_ns->mq_queues_max &&
+ !capable(CAP_SYS_RESOURCE)) {
error = -ENOSPC;
goto out_unlock;
}
@@ -459,13 +466,13 @@ out_unlock:
static int mqueue_unlink(struct inode *dir, struct dentry *dentry)
{
- struct inode *inode = dentry->d_inode;
+ struct inode *inode = dentry->d_inode;
dir->i_ctime = dir->i_mtime = dir->i_atime = CURRENT_TIME;
dir->i_size -= DIRENT_SIZE;
- drop_nlink(inode);
- dput(dentry);
- return 0;
+ drop_nlink(inode);
+ dput(dentry);
+ return 0;
}
/*
@@ -478,7 +485,7 @@ static int mqueue_unlink(struct inode *dir, struct dentry *dentry)
static ssize_t mqueue_read_file(struct file *filp, char __user *u_data,
size_t count, loff_t *off)
{
- struct mqueue_inode_info *info = MQUEUE_I(filp->f_path.dentry->d_inode);
+ struct mqueue_inode_info *info = MQUEUE_I(file_inode(filp));
char buffer[FILENT_SIZE];
ssize_t ret;
@@ -499,13 +506,13 @@ static ssize_t mqueue_read_file(struct file *filp, char __user *u_data,
if (ret <= 0)
return ret;
- filp->f_path.dentry->d_inode->i_atime = filp->f_path.dentry->d_inode->i_ctime = CURRENT_TIME;
+ file_inode(filp)->i_atime = file_inode(filp)->i_ctime = CURRENT_TIME;
return ret;
}
static int mqueue_flush_file(struct file *filp, fl_owner_t id)
{
- struct mqueue_inode_info *info = MQUEUE_I(filp->f_path.dentry->d_inode);
+ struct mqueue_inode_info *info = MQUEUE_I(file_inode(filp));
spin_lock(&info->lock);
if (task_tgid(current) == info->notify_owner)
@@ -517,7 +524,7 @@ static int mqueue_flush_file(struct file *filp, fl_owner_t id)
static unsigned int mqueue_poll_file(struct file *filp, struct poll_table_struct *poll_tab)
{
- struct mqueue_inode_info *info = MQUEUE_I(filp->f_path.dentry->d_inode);
+ struct mqueue_inode_info *info = MQUEUE_I(file_inode(filp));
int retval = 0;
poll_wait(filp, &info->wait_q, poll_tab);
@@ -615,7 +622,7 @@ static struct ext_wait_queue *wq_get_first_waiter(
static inline void set_cookie(struct sk_buff *skb, char code)
{
- ((char*)skb->data)[NOTIFY_COOKIE_LEN-1] = code;
+ ((char *)skb->data)[NOTIFY_COOKIE_LEN-1] = code;
}
/*
@@ -721,20 +728,19 @@ static int mq_attr_ok(struct ipc_namespace *ipc_ns, struct mq_attr *attr)
/*
* Invoked when creating a new queue via sys_mq_open
*/
-static struct file *do_create(struct ipc_namespace *ipc_ns, struct dentry *dir,
- struct dentry *dentry, int oflag, umode_t mode,
+static struct file *do_create(struct ipc_namespace *ipc_ns, struct inode *dir,
+ struct path *path, int oflag, umode_t mode,
struct mq_attr *attr)
{
const struct cred *cred = current_cred();
- struct file *result;
int ret;
if (attr) {
ret = mq_attr_ok(ipc_ns, attr);
if (ret)
- goto out;
+ return ERR_PTR(ret);
/* store for use during create */
- dentry->d_fsdata = attr;
+ path->dentry->d_fsdata = attr;
} else {
struct mq_attr def_attr;
@@ -744,71 +750,43 @@ static struct file *do_create(struct ipc_namespace *ipc_ns, struct dentry *dir,
ipc_ns->mq_msgsize_default);
ret = mq_attr_ok(ipc_ns, &def_attr);
if (ret)
- goto out;
+ return ERR_PTR(ret);
}
mode &= ~current_umask();
- ret = mnt_want_write(ipc_ns->mq_mnt);
+ ret = vfs_create(dir, path->dentry, mode, true);
+ path->dentry->d_fsdata = NULL;
if (ret)
- goto out;
- ret = vfs_create(dir->d_inode, dentry, mode, true);
- dentry->d_fsdata = NULL;
- if (ret)
- goto out_drop_write;
-
- result = dentry_open(dentry, ipc_ns->mq_mnt, oflag, cred);
- /*
- * dentry_open() took a persistent mnt_want_write(),
- * so we can now drop this one.
- */
- mnt_drop_write(ipc_ns->mq_mnt);
- return result;
-
-out_drop_write:
- mnt_drop_write(ipc_ns->mq_mnt);
-out:
- dput(dentry);
- mntput(ipc_ns->mq_mnt);
- return ERR_PTR(ret);
+ return ERR_PTR(ret);
+ return dentry_open(path, oflag, cred);
}
/* Opens existing queue */
-static struct file *do_open(struct ipc_namespace *ipc_ns,
- struct dentry *dentry, int oflag)
+static struct file *do_open(struct path *path, int oflag)
{
- int ret;
- const struct cred *cred = current_cred();
-
static const int oflag2acc[O_ACCMODE] = { MAY_READ, MAY_WRITE,
MAY_READ | MAY_WRITE };
-
- if ((oflag & O_ACCMODE) == (O_RDWR | O_WRONLY)) {
- ret = -EINVAL;
- goto err;
- }
-
- if (inode_permission(dentry->d_inode, oflag2acc[oflag & O_ACCMODE])) {
- ret = -EACCES;
- goto err;
- }
-
- return dentry_open(dentry, ipc_ns->mq_mnt, oflag, cred);
-
-err:
- dput(dentry);
- mntput(ipc_ns->mq_mnt);
- return ERR_PTR(ret);
+ int acc;
+ if ((oflag & O_ACCMODE) == (O_RDWR | O_WRONLY))
+ return ERR_PTR(-EINVAL);
+ acc = oflag2acc[oflag & O_ACCMODE];
+ if (inode_permission(path->dentry->d_inode, acc))
+ return ERR_PTR(-EACCES);
+ return dentry_open(path, oflag, current_cred());
}
SYSCALL_DEFINE4(mq_open, const char __user *, u_name, int, oflag, umode_t, mode,
struct mq_attr __user *, u_attr)
{
- struct dentry *dentry;
+ struct path path;
struct file *filp;
- char *name;
+ struct filename *name;
struct mq_attr attr;
int fd, error;
struct ipc_namespace *ipc_ns = current->nsproxy->ipc_ns;
+ struct vfsmount *mnt = ipc_ns->mq_mnt;
+ struct dentry *root = mnt->mnt_root;
+ int ro;
if (u_attr && copy_from_user(&attr, u_attr, sizeof(struct mq_attr)))
return -EFAULT;
@@ -822,52 +800,57 @@ SYSCALL_DEFINE4(mq_open, const char __user *, u_name, int, oflag, umode_t, mode,
if (fd < 0)
goto out_putname;
- mutex_lock(&ipc_ns->mq_mnt->mnt_root->d_inode->i_mutex);
- dentry = lookup_one_len(name, ipc_ns->mq_mnt->mnt_root, strlen(name));
- if (IS_ERR(dentry)) {
- error = PTR_ERR(dentry);
+ ro = mnt_want_write(mnt); /* we'll drop it in any case */
+ error = 0;
+ mutex_lock(&root->d_inode->i_mutex);
+ path.dentry = lookup_one_len(name->name, root, strlen(name->name));
+ if (IS_ERR(path.dentry)) {
+ error = PTR_ERR(path.dentry);
goto out_putfd;
}
- mntget(ipc_ns->mq_mnt);
+ path.mnt = mntget(mnt);
if (oflag & O_CREAT) {
- if (dentry->d_inode) { /* entry already exists */
- audit_inode(name, dentry);
+ if (path.dentry->d_inode) { /* entry already exists */
+ audit_inode(name, path.dentry, 0);
if (oflag & O_EXCL) {
error = -EEXIST;
goto out;
}
- filp = do_open(ipc_ns, dentry, oflag);
+ filp = do_open(&path, oflag);
} else {
- filp = do_create(ipc_ns, ipc_ns->mq_mnt->mnt_root,
- dentry, oflag, mode,
+ if (ro) {
+ error = ro;
+ goto out;
+ }
+ audit_inode_parent_hidden(name, root);
+ filp = do_create(ipc_ns, root->d_inode,
+ &path, oflag, mode,
u_attr ? &attr : NULL);
}
} else {
- if (!dentry->d_inode) {
+ if (!path.dentry->d_inode) {
error = -ENOENT;
goto out;
}
- audit_inode(name, dentry);
- filp = do_open(ipc_ns, dentry, oflag);
+ audit_inode(name, path.dentry, 0);
+ filp = do_open(&path, oflag);
}
- if (IS_ERR(filp)) {
+ if (!IS_ERR(filp))
+ fd_install(fd, filp);
+ else
error = PTR_ERR(filp);
- goto out_putfd;
- }
-
- fd_install(fd, filp);
- goto out_upsem;
-
out:
- dput(dentry);
- mntput(ipc_ns->mq_mnt);
+ path_put(&path);
out_putfd:
- put_unused_fd(fd);
- fd = error;
-out_upsem:
- mutex_unlock(&ipc_ns->mq_mnt->mnt_root->d_inode->i_mutex);
+ if (error) {
+ put_unused_fd(fd);
+ fd = error;
+ }
+ mutex_unlock(&root->d_inode->i_mutex);
+ if (!ro)
+ mnt_drop_write(mnt);
out_putname:
putname(name);
return fd;
@@ -876,44 +859,44 @@ out_putname:
SYSCALL_DEFINE1(mq_unlink, const char __user *, u_name)
{
int err;
- char *name;
+ struct filename *name;
struct dentry *dentry;
struct inode *inode = NULL;
struct ipc_namespace *ipc_ns = current->nsproxy->ipc_ns;
+ struct vfsmount *mnt = ipc_ns->mq_mnt;
name = getname(u_name);
if (IS_ERR(name))
return PTR_ERR(name);
- mutex_lock_nested(&ipc_ns->mq_mnt->mnt_root->d_inode->i_mutex,
- I_MUTEX_PARENT);
- dentry = lookup_one_len(name, ipc_ns->mq_mnt->mnt_root, strlen(name));
+ audit_inode_parent_hidden(name, mnt->mnt_root);
+ err = mnt_want_write(mnt);
+ if (err)
+ goto out_name;
+ mutex_lock_nested(&mnt->mnt_root->d_inode->i_mutex, I_MUTEX_PARENT);
+ dentry = lookup_one_len(name->name, mnt->mnt_root,
+ strlen(name->name));
if (IS_ERR(dentry)) {
err = PTR_ERR(dentry);
goto out_unlock;
}
- if (!dentry->d_inode) {
- err = -ENOENT;
- goto out_err;
- }
-
inode = dentry->d_inode;
- if (inode)
+ if (!inode) {
+ err = -ENOENT;
+ } else {
ihold(inode);
- err = mnt_want_write(ipc_ns->mq_mnt);
- if (err)
- goto out_err;
- err = vfs_unlink(dentry->d_parent->d_inode, dentry);
- mnt_drop_write(ipc_ns->mq_mnt);
-out_err:
+ err = vfs_unlink(dentry->d_parent->d_inode, dentry, NULL);
+ }
dput(dentry);
out_unlock:
- mutex_unlock(&ipc_ns->mq_mnt->mnt_root->d_inode->i_mutex);
- putname(name);
+ mutex_unlock(&mnt->mnt_root->d_inode->i_mutex);
if (inode)
iput(inode);
+ mnt_drop_write(mnt);
+out_name:
+ putname(name);
return err;
}
@@ -972,7 +955,7 @@ SYSCALL_DEFINE5(mq_timedsend, mqd_t, mqdes, const char __user *, u_msg_ptr,
size_t, msg_len, unsigned int, msg_prio,
const struct timespec __user *, u_abs_timeout)
{
- struct file *filp;
+ struct fd f;
struct inode *inode;
struct ext_wait_queue wait;
struct ext_wait_queue *receiver;
@@ -995,21 +978,21 @@ SYSCALL_DEFINE5(mq_timedsend, mqd_t, mqdes, const char __user *, u_msg_ptr,
audit_mq_sendrecv(mqdes, msg_len, msg_prio, timeout ? &ts : NULL);
- filp = fget(mqdes);
- if (unlikely(!filp)) {
+ f = fdget(mqdes);
+ if (unlikely(!f.file)) {
ret = -EBADF;
goto out;
}
- inode = filp->f_path.dentry->d_inode;
- if (unlikely(filp->f_op != &mqueue_file_operations)) {
+ inode = file_inode(f.file);
+ if (unlikely(f.file->f_op != &mqueue_file_operations)) {
ret = -EBADF;
goto out_fput;
}
info = MQUEUE_I(inode);
- audit_inode(NULL, filp->f_path.dentry);
+ audit_inode(NULL, f.file->f_path.dentry, 0);
- if (unlikely(!(filp->f_mode & FMODE_WRITE))) {
+ if (unlikely(!(f.file->f_mode & FMODE_WRITE))) {
ret = -EBADF;
goto out_fput;
}
@@ -1041,7 +1024,6 @@ SYSCALL_DEFINE5(mq_timedsend, mqd_t, mqdes, const char __user *, u_msg_ptr,
if (!info->node_cache && new_leaf) {
/* Save our speculative allocation into the cache */
- rb_init_node(&new_leaf->rb_node);
INIT_LIST_HEAD(&new_leaf->msg_list);
info->node_cache = new_leaf;
info->qsize += sizeof(*new_leaf);
@@ -1051,7 +1033,7 @@ SYSCALL_DEFINE5(mq_timedsend, mqd_t, mqdes, const char __user *, u_msg_ptr,
}
if (info->attr.mq_curmsgs == info->attr.mq_maxmsg) {
- if (filp->f_flags & O_NONBLOCK) {
+ if (f.file->f_flags & O_NONBLOCK) {
ret = -EAGAIN;
} else {
wait.task = current;
@@ -1084,7 +1066,7 @@ out_free:
if (ret)
free_msg(msg_ptr);
out_fput:
- fput(filp);
+ fdput(f);
out:
return ret;
}
@@ -1095,7 +1077,7 @@ SYSCALL_DEFINE5(mq_timedreceive, mqd_t, mqdes, char __user *, u_msg_ptr,
{
ssize_t ret;
struct msg_msg *msg_ptr;
- struct file *filp;
+ struct fd f;
struct inode *inode;
struct mqueue_inode_info *info;
struct ext_wait_queue wait;
@@ -1112,21 +1094,21 @@ SYSCALL_DEFINE5(mq_timedreceive, mqd_t, mqdes, char __user *, u_msg_ptr,
audit_mq_sendrecv(mqdes, msg_len, 0, timeout ? &ts : NULL);
- filp = fget(mqdes);
- if (unlikely(!filp)) {
+ f = fdget(mqdes);
+ if (unlikely(!f.file)) {
ret = -EBADF;
goto out;
}
- inode = filp->f_path.dentry->d_inode;
- if (unlikely(filp->f_op != &mqueue_file_operations)) {
+ inode = file_inode(f.file);
+ if (unlikely(f.file->f_op != &mqueue_file_operations)) {
ret = -EBADF;
goto out_fput;
}
info = MQUEUE_I(inode);
- audit_inode(NULL, filp->f_path.dentry);
+ audit_inode(NULL, f.file->f_path.dentry, 0);
- if (unlikely(!(filp->f_mode & FMODE_READ))) {
+ if (unlikely(!(f.file->f_mode & FMODE_READ))) {
ret = -EBADF;
goto out_fput;
}
@@ -1149,7 +1131,6 @@ SYSCALL_DEFINE5(mq_timedreceive, mqd_t, mqdes, char __user *, u_msg_ptr,
if (!info->node_cache && new_leaf) {
/* Save our speculative allocation into the cache */
- rb_init_node(&new_leaf->rb_node);
INIT_LIST_HEAD(&new_leaf->msg_list);
info->node_cache = new_leaf;
info->qsize += sizeof(*new_leaf);
@@ -1158,7 +1139,7 @@ SYSCALL_DEFINE5(mq_timedreceive, mqd_t, mqdes, char __user *, u_msg_ptr,
}
if (info->attr.mq_curmsgs == 0) {
- if (filp->f_flags & O_NONBLOCK) {
+ if (f.file->f_flags & O_NONBLOCK) {
spin_unlock(&info->lock);
ret = -EAGAIN;
} else {
@@ -1188,7 +1169,7 @@ SYSCALL_DEFINE5(mq_timedreceive, mqd_t, mqdes, char __user *, u_msg_ptr,
free_msg(msg_ptr);
}
out_fput:
- fput(filp);
+ fdput(f);
out:
return ret;
}
@@ -1202,7 +1183,7 @@ SYSCALL_DEFINE2(mq_notify, mqd_t, mqdes,
const struct sigevent __user *, u_notification)
{
int ret;
- struct file *filp;
+ struct fd f;
struct sock *sock;
struct inode *inode;
struct sigevent notification;
@@ -1248,13 +1229,13 @@ SYSCALL_DEFINE2(mq_notify, mqd_t, mqdes,
skb_put(nc, NOTIFY_COOKIE_LEN);
/* and attach it to the socket */
retry:
- filp = fget(notification.sigev_signo);
- if (!filp) {
+ f = fdget(notification.sigev_signo);
+ if (!f.file) {
ret = -EBADF;
goto out;
}
- sock = netlink_getsockbyfilp(filp);
- fput(filp);
+ sock = netlink_getsockbyfilp(f.file);
+ fdput(f);
if (IS_ERR(sock)) {
ret = PTR_ERR(sock);
sock = NULL;
@@ -1273,14 +1254,14 @@ retry:
}
}
- filp = fget(mqdes);
- if (!filp) {
+ f = fdget(mqdes);
+ if (!f.file) {
ret = -EBADF;
goto out;
}
- inode = filp->f_path.dentry->d_inode;
- if (unlikely(filp->f_op != &mqueue_file_operations)) {
+ inode = file_inode(f.file);
+ if (unlikely(f.file->f_op != &mqueue_file_operations)) {
ret = -EBADF;
goto out_fput;
}
@@ -1320,13 +1301,13 @@ retry:
}
spin_unlock(&info->lock);
out_fput:
- fput(filp);
+ fdput(f);
out:
- if (sock) {
+ if (sock)
netlink_detachskb(sock, nc);
- } else if (nc) {
+ else if (nc)
dev_kfree_skb(nc);
- }
+
return ret;
}
@@ -1336,7 +1317,7 @@ SYSCALL_DEFINE3(mq_getsetattr, mqd_t, mqdes,
{
int ret;
struct mq_attr mqstat, omqstat;
- struct file *filp;
+ struct fd f;
struct inode *inode;
struct mqueue_inode_info *info;
@@ -1347,14 +1328,14 @@ SYSCALL_DEFINE3(mq_getsetattr, mqd_t, mqdes,
return -EINVAL;
}
- filp = fget(mqdes);
- if (!filp) {
+ f = fdget(mqdes);
+ if (!f.file) {
ret = -EBADF;
goto out;
}
- inode = filp->f_path.dentry->d_inode;
- if (unlikely(filp->f_op != &mqueue_file_operations)) {
+ inode = file_inode(f.file);
+ if (unlikely(f.file->f_op != &mqueue_file_operations)) {
ret = -EBADF;
goto out_fput;
}
@@ -1363,15 +1344,15 @@ SYSCALL_DEFINE3(mq_getsetattr, mqd_t, mqdes,
spin_lock(&info->lock);
omqstat = info->attr;
- omqstat.mq_flags = filp->f_flags & O_NONBLOCK;
+ omqstat.mq_flags = f.file->f_flags & O_NONBLOCK;
if (u_mqstat) {
audit_mq_getsetattr(mqdes, &mqstat);
- spin_lock(&filp->f_lock);
+ spin_lock(&f.file->f_lock);
if (mqstat.mq_flags & O_NONBLOCK)
- filp->f_flags |= O_NONBLOCK;
+ f.file->f_flags |= O_NONBLOCK;
else
- filp->f_flags &= ~O_NONBLOCK;
- spin_unlock(&filp->f_lock);
+ f.file->f_flags &= ~O_NONBLOCK;
+ spin_unlock(&f.file->f_lock);
inode->i_atime = inode->i_ctime = CURRENT_TIME;
}
@@ -1384,7 +1365,7 @@ SYSCALL_DEFINE3(mq_getsetattr, mqd_t, mqdes,
ret = -EFAULT;
out_fput:
- fput(filp);
+ fdput(f);
out:
return ret;
}
@@ -1413,6 +1394,7 @@ static struct file_system_type mqueue_fs_type = {
.name = "mqueue",
.mount = mqueue_mount,
.kill_sb = kill_litter_super,
+ .fs_flags = FS_USERNS_MOUNT,
};
int mq_init_ns(struct ipc_namespace *ns)
@@ -1477,4 +1459,4 @@ out_sysctl:
return error;
}
-__initcall(init_mqueue_fs);
+device_initcall(init_mqueue_fs);
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-11-10 15:14:28 +0000