/**
* anon_inode_getfile - creates a new file instance by hooking it up to an
* anonymous inode, and a dentry that describe the "class"
* of the file
*
* @name: [in] name of the "class" of the new file
* @fops: [in] file operations for the new file
* @priv: [in] private data for the new file (will be file's private_data)
* @flags: [in] flags
*
* Creates a new file by hooking it on a single inode. This is useful for files
* that do not need to have a full-fledged inode in order to operate correctly.
* All the files created with anon_inode_getfile() will share a single inode,
* hence saving memory and avoiding code duplication for the file/inode/dentry
* setup. Returns the newly created file* or an error pointer.
*/
struct file *anon_inode_getfile(const char *name,
const struct file_operations *fops,
void *priv, int flags)
{
struct qstr this;
struct path path;
struct file *file;
int error;
if (IS_ERR(anon_inode_inode))
return ERR_PTR(-ENODEV);
if (fops->owner && !try_module_get(fops->owner))
return ERR_PTR(-ENOENT);
/*
* Link the inode to a directory entry by creating a unique name
* using the inode sequence number.
*/
error = -ENOMEM;
this.name = name;
this.len = strlen(name);
this.hash = 0;
path.dentry = d_alloc_pseudo(anon_inode_mnt->mnt_sb, &this);
if (!path.dentry)
goto err_module;
path.mnt = mntget(anon_inode_mnt);
/*
* We know the anon_inode inode count is always greater than zero,
* so ihold() is safe.
*/
ihold(anon_inode_inode);
d_instantiate(path.dentry, anon_inode_inode);
error = -ENFILE;
file = alloc_file(&path, OPEN_FMODE(flags), fops);
if (!file)
goto err_dput;
file->f_mapping = anon_inode_inode->i_mapping;
file->f_pos = 0;
file->f_flags = flags & (O_ACCMODE | O_NONBLOCK);
file->f_version = 0;
file->private_data = priv;
return file;
err_dput:
path_put(&path);
err_module:
module_put(fops->owner);
return ERR_PTR(error);
}
struct file *anon_inode_getfile(const char *name,
const struct file_operations *fops,
void *priv, int flags)
{
struct qstr this;
struct path path;
struct file *file;
int error;
if (IS_ERR(anon_inode_inode))
return ERR_PTR(-ENODEV);
if (fops->owner && !try_module_get(fops->owner))
return ERR_PTR(-ENOENT);
error = -ENOMEM;
this.name = name;
this.len = strlen(name);
this.hash = 0;
path.dentry = d_alloc_pseudo(anon_inode_mnt->mnt_sb, &this);
if (!path.dentry)
goto err_module;
path.mnt = mntget(anon_inode_mnt);
ihold(anon_inode_inode);
d_instantiate(path.dentry, anon_inode_inode);
error = -ENFILE;
file = alloc_file(&path, OPEN_FMODE(flags), fops);
if (!file)
goto err_dput;
file->f_mapping = anon_inode_inode->i_mapping;
file->f_pos = 0;
file->f_flags = flags & (O_ACCMODE | O_NONBLOCK);
file->f_version = 0;
file->private_data = priv;
return file;
err_dput:
path_put(&path);
err_module:
module_put(fops->owner);
return ERR_PTR(error);
}
static void *__ns_get_path(struct path *path, struct ns_common *ns)
{
struct vfsmount *mnt = nsfs_mnt;
struct qstr qname = { .name = "", };
struct dentry *dentry;
struct inode *inode;
unsigned long d;
rcu_read_lock();
d = atomic_long_read(&ns->stashed);
if (!d)
goto slow;
dentry = (struct dentry *)d;
if (!lockref_get_not_dead(&dentry->d_lockref))
goto slow;
rcu_read_unlock();
ns->ops->put(ns);
got_it:
path->mnt = mntget(mnt);
path->dentry = dentry;
return NULL;
slow:
rcu_read_unlock();
inode = new_inode_pseudo(mnt->mnt_sb);
if (!inode) {
ns->ops->put(ns);
return ERR_PTR(-ENOMEM);
}
inode->i_ino = ns->inum;
inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
inode->i_flags |= S_IMMUTABLE;
inode->i_mode = S_IFREG | S_IRUGO;
inode->i_fop = &ns_file_operations;
inode->i_private = ns;
dentry = d_alloc_pseudo(mnt->mnt_sb, &qname);
if (!dentry) {
iput(inode);
return ERR_PTR(-ENOMEM);
}
d_instantiate(dentry, inode);
dentry->d_flags |= DCACHE_RCUACCESS;
dentry->d_fsdata = (void *)ns->ops;
d = atomic_long_cmpxchg(&ns->stashed, 0, (unsigned long)dentry);
if (d) {
d_delete(dentry); /* make sure ->d_prune() does nothing */
dput(dentry);
cpu_relax();
return ERR_PTR(-EAGAIN);
}
goto got_it;
}
void *ns_get_path(struct path *path, struct task_struct *task,
const struct proc_ns_operations *ns_ops)
{
struct ns_common *ns;
void *ret;
again:
ns = ns_ops->get(task);
if (!ns)
return ERR_PTR(-ENOENT);
ret = __ns_get_path(path, ns);
if (IS_ERR(ret) && PTR_ERR(ret) == -EAGAIN)
goto again;
return ret;
}
int open_related_ns(struct ns_common *ns,
struct ns_common *(*get_ns)(struct ns_common *ns))
{
struct path path = {};
struct file *f;
void *err;
int fd;
fd = get_unused_fd_flags(O_CLOEXEC);
if (fd < 0)
return fd;
while (1) {
struct ns_common *relative;
relative = get_ns(ns);
if (IS_ERR(relative)) {
put_unused_fd(fd);
return PTR_ERR(relative);
}
err = __ns_get_path(&path, relative);
if (IS_ERR(err) && PTR_ERR(err) == -EAGAIN)
continue;
break;
}
if (IS_ERR(err)) {
put_unused_fd(fd);
return PTR_ERR(err);
}
f = dentry_open(&path, O_RDONLY, current_cred());
path_put(&path);
if (IS_ERR(f)) {
put_unused_fd(fd);
fd = PTR_ERR(f);
} else
fd_install(fd, f);
return fd;
}
static long ns_ioctl(struct file *filp, unsigned int ioctl,
unsigned long arg)
{
struct user_namespace *user_ns;
struct ns_common *ns = get_proc_ns(file_inode(filp));
uid_t __user *argp;
uid_t uid;
switch (ioctl) {
case NS_GET_USERNS:
return open_related_ns(ns, ns_get_owner);
case NS_GET_PARENT:
if (!ns->ops->get_parent)
return -EINVAL;
return open_related_ns(ns, ns->ops->get_parent);
case NS_GET_NSTYPE:
return ns->ops->type;
case NS_GET_OWNER_UID:
if (ns->ops->type != CLONE_NEWUSER)
return -EINVAL;
user_ns = container_of(ns, struct user_namespace, ns);
argp = (uid_t __user *) arg;
uid = from_kuid_munged(current_user_ns(), user_ns->owner);
return put_user(uid, argp);
default:
return -ENOTTY;
}
}
int ns_get_name(char *buf, size_t size, struct task_struct *task,
const struct proc_ns_operations *ns_ops)
{
struct ns_common *ns;
int res = -ENOENT;
ns = ns_ops->get(task);
if (ns) {
res = snprintf(buf, size, "%s:[%u]", ns_ops->name, ns->inum);
ns_ops->put(ns);
}
return res;
}
struct file *proc_ns_fget(int fd)
{
struct file *file;
file = fget(fd);
if (!file)
return ERR_PTR(-EBADF);
if (file->f_op != &ns_file_operations)
goto out_invalid;
return file;
out_invalid:
fput(file);
return ERR_PTR(-EINVAL);
}
static int nsfs_show_path(struct seq_file *seq, struct dentry *dentry)
{
struct inode *inode = d_inode(dentry);
const struct proc_ns_operations *ns_ops = dentry->d_fsdata;
seq_printf(seq, "%s:[%lu]", ns_ops->name, inode->i_ino);
return 0;
}
static const struct super_operations nsfs_ops = {
.statfs = simple_statfs,
.evict_inode = nsfs_evict,
.show_path = nsfs_show_path,
};
static struct dentry *nsfs_mount(struct file_system_type *fs_type,
int flags, const char *dev_name, void *data)
{
return mount_pseudo(fs_type, "nsfs:", &nsfs_ops,
&ns_dentry_operations, NSFS_MAGIC);
}
static struct file_system_type nsfs = {
.name = "nsfs",
.mount = nsfs_mount,
.kill_sb = kill_anon_super,
};
void __init nsfs_init(void)
{
nsfs_mnt = kern_mount(&nsfs);
if (IS_ERR(nsfs_mnt))
panic("can't set nsfs up\n");
nsfs_mnt->mnt_sb->s_flags &= ~MS_NOUSER;
}
void *ns_get_path(struct path *path, struct task_struct *task,
const struct proc_ns_operations *ns_ops)
{
struct vfsmount *mnt = mntget(nsfs_mnt);
struct qstr qname = { .name = "", };
struct dentry *dentry;
struct inode *inode;
struct ns_common *ns;
unsigned long d;
again:
ns = ns_ops->get(task);
if (!ns) {
mntput(mnt);
return ERR_PTR(-ENOENT);
}
rcu_read_lock();
d = atomic_long_read(&ns->stashed);
if (!d)
goto slow;
dentry = (struct dentry *)d;
if (!lockref_get_not_dead(&dentry->d_lockref))
goto slow;
rcu_read_unlock();
ns_ops->put(ns);
got_it:
path->mnt = mnt;
path->dentry = dentry;
return NULL;
slow:
rcu_read_unlock();
inode = new_inode_pseudo(mnt->mnt_sb);
if (!inode) {
ns_ops->put(ns);
mntput(mnt);
return ERR_PTR(-ENOMEM);
}
inode->i_ino = ns->inum;
inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
inode->i_flags |= S_IMMUTABLE;
inode->i_mode = S_IFREG | S_IRUGO;
inode->i_fop = &ns_file_operations;
inode->i_private = ns;
dentry = d_alloc_pseudo(mnt->mnt_sb, &qname);
if (!dentry) {
iput(inode);
mntput(mnt);
return ERR_PTR(-ENOMEM);
}
d_instantiate(dentry, inode);
dentry->d_fsdata = (void *)ns_ops;
d = atomic_long_cmpxchg(&ns->stashed, 0, (unsigned long)dentry);
if (d) {
d_delete(dentry); /* make sure ->d_prune() does nothing */
dput(dentry);
cpu_relax();
goto again;
}
goto got_it;
}
int ns_get_name(char *buf, size_t size, struct task_struct *task,
const struct proc_ns_operations *ns_ops)
{
struct ns_common *ns;
int res = -ENOENT;
ns = ns_ops->get(task);
if (ns) {
res = snprintf(buf, size, "%s:[%u]", ns_ops->name, ns->inum);
ns_ops->put(ns);
}
return res;
}
struct file *proc_ns_fget(int fd)
{
struct file *file;
file = fget(fd);
if (!file)
return ERR_PTR(-EBADF);
if (file->f_op != &ns_file_operations)
goto out_invalid;
return file;
out_invalid:
fput(file);
return ERR_PTR(-EINVAL);
}
static const struct super_operations nsfs_ops = {
.statfs = simple_statfs,
.evict_inode = nsfs_evict,
};
static struct dentry *nsfs_mount(struct file_system_type *fs_type,
int flags, const char *dev_name, void *data)
{
return mount_pseudo(fs_type, "nsfs:", &nsfs_ops,
&ns_dentry_operations, NSFS_MAGIC);
}
static struct file_system_type nsfs = {
.name = "nsfs",
.mount = nsfs_mount,
.kill_sb = kill_anon_super,
};
void __init nsfs_init(void)
{
nsfs_mnt = kern_mount(&nsfs);
if (IS_ERR(nsfs_mnt))
panic("can't set nsfs up\n");
nsfs_mnt->mnt_sb->s_flags &= ~MS_NOUSER;
}
static int
onload_alloc_file(tcp_helper_resource_t *thr, oo_sp ep_id, int flags,
int fd_type)
{
struct qstr name = { .name = "" };
#ifdef EFX_HAVE_STRUCT_PATH
struct path path;
#define my_dentry path.dentry
#else
struct dentry *dentry;
#define my_dentry dentry
#endif
struct file *file;
int fd;
struct inode *inode;
ci_private_t *priv;
struct file_operations *fops;
fops = oo_fops_by_type(fd_type);
if( fops == NULL )
return -EINVAL;
ci_assert_equal(fops->owner, THIS_MODULE);
inode = new_inode(onload_mnt->mnt_sb);
if( inode == NULL )
return -ENOMEM;
#ifdef EFX_FSTYPE_HAS_MOUNT
inode->i_ino = get_next_ino();
#endif
if( fd_type == CI_PRIV_TYPE_NETIF )
inode->i_mode = S_IRWXUGO;
if( fd_type == CI_PRIV_TYPE_TCP_EP || fd_type == CI_PRIV_TYPE_UDP_EP )
inode->i_mode =
#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,21)
/* in 2.6.18 this flag makes us "socket" and sendmsg crashes;
* see sock_from_file() */
S_IFSOCK |
#endif
S_IRWXUGO;
else
inode->i_mode = S_IFIFO | S_IRUSR | S_IWUSR;
inode->i_uid = current_fsuid();
inode->i_gid = current_fsgid();
priv = &container_of(inode, struct onload_inode, vfs_inode)->priv;
priv->thr = thr;
priv->sock_id = ep_id;
priv->fd_type = fd_type;
fd = get_unused_fd();
if( fd < 0 ) {
iput(inode);
return fd;
}
/*ci_log("[%d]%s(%d:%d) return %d priv=%p", current->pid, __func__,
thr->id, ep_id, fd, priv);*/
#ifdef EFX_FSTYPE_HAS_MOUNT
#if LINUX_VERSION_CODE <= KERNEL_VERSION(2,6,37)
path.dentry = d_alloc(onload_mnt->mnt_sb->s_root, &name);
if( path.dentry != NULL )
path.dentry->d_op = &onloadfs_dentry_operations;
#else
path.dentry = d_alloc_pseudo(onload_mnt->mnt_sb, &name);
#endif
#else /* EFX_FSTYPE_HAS_MOUNT */
#ifdef EFX_HAVE_D_DNAME
my_dentry = d_alloc(onload_mnt->mnt_sb->s_root, &name);
#else
{
char str[32];
name.len = onloadfs_name(&container_of(inode, struct onload_inode,
vfs_inode)->priv,
str, sizeof(str));
name.name = str;
name.hash = inode->i_ino;
my_dentry = d_alloc(onload_mnt->mnt_sb->s_root, &name);
}
#endif
#endif /* EFX_FSTYPE_HAS_MOUNT */
if( my_dentry == NULL ) {
put_unused_fd(fd);
iput(inode);
return -ENOMEM;
}
#if !defined(EFX_FSTYPE_HAS_MOUNT) || defined(EFX_OLD_MOUNT_PSEUDO)
my_dentry->d_op = &onloadfs_dentry_operations;
#if !defined(EFX_HAVE_STRUCT_PATH) && defined(EFX_HAVE_D_DNAME)
my_dentry->d_flags &= ~DCACHE_UNHASHED;
#endif
#endif
d_instantiate(my_dentry, inode);
#ifndef EFX_HAVE_D_DNAME
d_rehash(my_dentry);
#endif
inode->i_fop = fops;
#ifdef EFX_HAVE_STRUCT_PATH
path.mnt = mntget(onload_mnt);
file = alloc_file(&path, FMODE_READ | FMODE_WRITE, fops);
#else
file = alloc_file(onload_mnt, dentry, FMODE_READ | FMODE_WRITE, fops);
#endif
if( file == NULL) {
#ifdef EFX_HAVE_STRUCT_PATH
path_put(&path);
#else
dput(dentry);
iput(inode);
#endif
put_unused_fd(fd);
return -ENFILE;
}
priv->_filp = file;
file->f_flags = O_RDWR | (flags & O_NONBLOCK);
file->f_pos = 0;
file->private_data = priv;
if( flags & O_CLOEXEC ) {
struct files_struct *files = current->files;
struct fdtable *fdt;
spin_lock(&files->file_lock);
fdt = files_fdtable(files);
rcu_assign_pointer(fdt->fd[fd], file);
efx_set_close_on_exec(fd, fdt);
spin_unlock(&files->file_lock);
} else
fd_install(fd, file);
try_module_get(THIS_MODULE);
ci_assert_equal(file->f_op, fops);
return fd;
}
void onload_priv_free(ci_private_t *priv)
{
if( priv->_filp->f_vfsmnt != onload_mnt)
ci_free(priv);
/* inode will free the priv automatically */
}
int
oo_create_fd(tcp_helper_endpoint_t* ep, int flags, int fd_type)
{
int fd;
tcp_helper_resource_t *trs = ep->thr;
citp_waitable_obj *wo = SP_TO_WAITABLE_OBJ(&trs->netif, ep->id);
efab_thr_ref(trs);
fd = onload_alloc_file(trs, ep->id, flags, fd_type);
if( fd < 0 ) {
efab_thr_release(trs);
OO_DEBUG_ERR(ci_log("%s: onload_alloc_file failed (%d)", __FUNCTION__, fd));
return fd;
}
ci_atomic32_and(&wo-> waitable.sb_aflags,
~(CI_SB_AFLAG_ORPHAN | CI_SB_AFLAG_TCP_IN_ACCEPTQ));
return fd;
}
/*
* We have a vnode/inode, now get the dentry. If there isn't
* one already pointing to the inode, we must fabricate a
* disconnected entry for NFS to connect up to the known
* dcache tree (in the paranoia/export directory checking
* cases--but performance will be better if
* NFSEXP_NOSUBTREECHECK is present in the export options--but
* will this break our use of dentry->d_parent->d_inode in
* dentry_to_fh()? XXX).
* Returns a valid dentry pointer or a negative error code.
* If the returned dentry is a new one, the inode's refcount is
* incremented.
*/
STATIC struct dentry *
vnlayer_find_dentry(VNODE_T *vp)
{
static const struct qstr this_is_anon = { .name = ""};
struct dentry *dp;
struct dentry *dnew;
INODE_T *ip;
ip = VTOI(vp);
/*
* We create an anonymous dentry for NFS, but it will be used
* only if we don't find a suitable one for this inode.
* d_alloc is here because before 2.6.38 it acquires dcache_lock.
*/
#if LINUX_VERSION_CODE < KERNEL_VERSION(3,0,0)
dnew = d_alloc(NULL, &this_is_anon);
#else
dnew = d_alloc_pseudo(ip->i_sb, &this_is_anon);
#endif
if (dnew == NULL) {
dp = ERR_PTR(-ENOMEM);
} else {
dnew->d_parent = dnew;
#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,38)
spin_lock(&dcache_lock);
#else
spin_lock(&ip->i_lock);
#endif
/*
* equivalent of d_splice_alias,
* we only want view-extended dentries
*/
dp = vnlayer_inode2dentry_internal_no_lock(ip,
NULL,
NULL,
&vnode_dentry_ops);
if (dp == NULL) {
/* new dentry, increase the refcount for this v/inode */
#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,38)
VN_HOLD(vp);
#else
/* can't igrab because the spinlock is acquired */
ihold(ip);
#endif
/* found no suitable dentry, add a new one */
MDKI_SET_DOPS(dnew, &vnode_dentry_ops);
dnew->d_sb = ip->i_sb;
dnew->d_inode = ip;
#if LINUX_VERSION_CODE < KERNEL_VERSION(3,0,0)
dnew->d_flags |= NFSD_DCACHE_DISCON;
dnew->d_flags &= ~DCACHE_UNHASHED;
#else /* LINUX_VERSION_CODE < KERNEL_VERSION(3,0,0) */
/* it is not clear if it needs DCACHE_RCUACCESS */
dnew->d_flags |= NFSD_DCACHE_DISCON;
#endif /* else LINUX_VERSION_CODE < KERNEL_VERSION(3,0,0) */
#if LINUX_VERSION_CODE >= KERNEL_VERSION(3,6,0)
hlist_add_head(&dnew->d_alias, &ip->i_dentry);
#else
list_add(&dnew->d_alias, &ip->i_dentry);
#endif /* LINUX_VERSION_CODE >= KERNEL_VERSION(3,6,0) */
#if LINUX_VERSION_CODE < KERNEL_VERSION(3,0,0)
hlist_add_head(&dnew->d_hash, &ip->i_sb->s_anon);
#else /* LINUX_VERSION_CODE < KERNEL_VERSION(3,0,0) */
hlist_bl_lock(&ip->i_sb->s_anon);
hlist_bl_add_head_rcu(&dnew->d_hash, &ip->i_sb->s_anon);
hlist_bl_unlock(&ip->i_sb->s_anon);
#endif /* else LINUX_VERSION_CODE < KERNEL_VERSION(3,0,0) */
dp = dnew;
/* skip the dput call for dnew, we will need it */
dnew = NULL;
}
#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,38)
spin_unlock(&dcache_lock);
#else
spin_unlock(&ip->i_lock);
#endif
}
/* dput dnew if we don't use it */
if (dnew != NULL)
dput(dnew);
return dp;
}
static const char vnode_verid_mvfs_linux_sops_c[] = "$Id: c9a2d7b9.e2bd11e3.8cd7.00:11:25:27:c4:b4 $";
static struct dentry *proc_ns_get_dentry(struct super_block *sb,
struct task_struct *task, const struct proc_ns_operations *ns_ops)
{
struct dentry *dentry, *result;
struct inode *inode;
struct proc_inode *ei;
struct qstr qname = { .name = "", };
void *ns;
ns = ns_ops->get(task);
if (!ns)
return ERR_PTR(-ENOENT);
dentry = d_alloc_pseudo(sb, &qname);
if (!dentry) {
ns_ops->put(ns);
return ERR_PTR(-ENOMEM);
}
inode = iget_locked(sb, ns_ops->inum(ns));
if (!inode) {
dput(dentry);
ns_ops->put(ns);
return ERR_PTR(-ENOMEM);
}
ei = PROC_I(inode);
if (inode->i_state & I_NEW) {
inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
inode->i_op = &ns_inode_operations;
inode->i_mode = S_IFREG | S_IRUGO;
inode->i_fop = &ns_file_operations;
ei->ns.ns_ops = ns_ops;
ei->ns.ns = ns;
unlock_new_inode(inode);
} else {
ns_ops->put(ns);
}
d_set_d_op(dentry, &ns_dentry_operations);
result = d_instantiate_unique(dentry, inode);
if (result) {
dput(dentry);
dentry = result;
}
return dentry;
}
static void *proc_ns_follow_link(struct dentry *dentry, struct nameidata *nd)
{
struct inode *inode = dentry->d_inode;
struct super_block *sb = inode->i_sb;
struct proc_inode *ei = PROC_I(inode);
struct task_struct *task;
struct path ns_path;
void *error = ERR_PTR(-EACCES);
task = get_proc_task(inode);
if (!task)
goto out;
if (!ptrace_may_access(task, PTRACE_MODE_READ))
goto out_put_task;
ns_path.dentry = proc_ns_get_dentry(sb, task, ei->ns.ns_ops);
if (IS_ERR(ns_path.dentry)) {
error = ERR_CAST(ns_path.dentry);
goto out_put_task;
}
ns_path.mnt = mntget(nd->path.mnt);
nd_jump_link(nd, &ns_path);
error = NULL;
out_put_task:
put_task_struct(task);
out:
return error;
}
static int proc_ns_readlink(struct dentry *dentry, char __user *buffer, int buflen)
{
struct inode *inode = dentry->d_inode;
struct proc_inode *ei = PROC_I(inode);
const struct proc_ns_operations *ns_ops = ei->ns.ns_ops;
struct task_struct *task;
void *ns;
char name[50];
int len = -EACCES;
task = get_proc_task(inode);
if (!task)
goto out;
if (!ptrace_may_access(task, PTRACE_MODE_READ))
goto out_put_task;
len = -ENOENT;
ns = ns_ops->get(task);
if (!ns)
goto out_put_task;
snprintf(name, sizeof(name), "%s:[%u]", ns_ops->name, ns_ops->inum(ns));
len = strlen(name);
if (len > buflen)
len = buflen;
if (copy_to_user(buffer, name, len))
len = -EFAULT;
ns_ops->put(ns);
out_put_task:
put_task_struct(task);
out:
return len;
}
static const struct inode_operations proc_ns_link_inode_operations = {
.readlink = proc_ns_readlink,
.follow_link = proc_ns_follow_link,
.setattr = proc_setattr,
};
static int proc_ns_instantiate(struct inode *dir,
struct dentry *dentry, struct task_struct *task, const void *ptr)
{
const struct proc_ns_operations *ns_ops = ptr;
struct inode *inode;
struct proc_inode *ei;
inode = proc_pid_make_inode(dir->i_sb, task);
if (!inode)
goto out;
ei = PROC_I(inode);
inode->i_mode = S_IFLNK|S_IRWXUGO;
inode->i_op = &proc_ns_link_inode_operations;
ei->ns.ns_ops = ns_ops;
d_set_d_op(dentry, &pid_dentry_operations);
d_add(dentry, inode);
/* Close the race of the process dying before we return the dentry */
if (pid_revalidate(dentry, 0))
return 0;
out:
return -ENOENT;
}
static int proc_ns_dir_readdir(struct file *file, struct dir_context *ctx)
{
struct task_struct *task = get_proc_task(file_inode(file));
const struct proc_ns_operations **entry, **last;
if (!task)
return -ENOENT;
if (!dir_emit_dots(file, ctx))
goto out;
if (ctx->pos >= 2 + ARRAY_SIZE(ns_entries))
goto out;
entry = ns_entries + (ctx->pos - 2);
last = &ns_entries[ARRAY_SIZE(ns_entries) - 1];
while (entry <= last) {
const struct proc_ns_operations *ops = *entry;
if (!proc_fill_cache(file, ctx, ops->name, strlen(ops->name),
proc_ns_instantiate, task, ops))
break;
ctx->pos++;
entry++;
}
out:
put_task_struct(task);
return 0;
}
const struct file_operations proc_ns_dir_operations = {
.read = generic_read_dir,
.iterate = proc_ns_dir_readdir,
};
static struct dentry *proc_ns_dir_lookup(struct inode *dir,
struct dentry *dentry, unsigned int flags)
{
int error;
struct task_struct *task = get_proc_task(dir);
const struct proc_ns_operations **entry, **last;
unsigned int len = dentry->d_name.len;
error = -ENOENT;
if (!task)
goto out_no_task;
last = &ns_entries[ARRAY_SIZE(ns_entries)];
for (entry = ns_entries; entry < last; entry++) {
if (strlen((*entry)->name) != len)
continue;
if (!memcmp(dentry->d_name.name, (*entry)->name, len))
break;
}
if (entry == last)
goto out;
error = proc_ns_instantiate(dir, dentry, task, *entry);
out:
put_task_struct(task);
out_no_task:
return ERR_PTR(error);
}
