nsfs.c 6.11 KB
Newer Older
1
// SPDX-License-Identifier: GPL-2.0
2
#include <linux/mount.h>
3
#include <linux/pseudo_fs.h>
4 5 6 7 8 9
#include <linux/file.h>
#include <linux/fs.h>
#include <linux/proc_ns.h>
#include <linux/magic.h>
#include <linux/ktime.h>
#include <linux/seq_file.h>
10 11
#include <linux/user_namespace.h>
#include <linux/nsfs.h>
12
#include <linux/uaccess.h>
13 14 15

static struct vfsmount *nsfs_mnt;

16 17
static long ns_ioctl(struct file *filp, unsigned int ioctl,
			unsigned long arg);
18 19
static const struct file_operations ns_file_operations = {
	.llseek		= no_llseek,
20
	.unlocked_ioctl = ns_ioctl,
21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54
};

static char *ns_dname(struct dentry *dentry, char *buffer, int buflen)
{
	struct inode *inode = d_inode(dentry);
	const struct proc_ns_operations *ns_ops = dentry->d_fsdata;

	return dynamic_dname(dentry, buffer, buflen, "%s:[%lu]",
		ns_ops->name, inode->i_ino);
}

static void ns_prune_dentry(struct dentry *dentry)
{
	struct inode *inode = d_inode(dentry);
	if (inode) {
		struct ns_common *ns = inode->i_private;
		atomic_long_set(&ns->stashed, 0);
	}
}

const struct dentry_operations ns_dentry_operations =
{
	.d_prune	= ns_prune_dentry,
	.d_delete	= always_delete_dentry,
	.d_dname	= ns_dname,
};

static void nsfs_evict(struct inode *inode)
{
	struct ns_common *ns = inode->i_private;
	clear_inode(inode);
	ns->ops->put(ns);
}

55
static void *__ns_get_path(struct path *path, struct ns_common *ns)
56
{
57
	struct vfsmount *mnt = nsfs_mnt;
58 59 60 61 62 63 64 65 66 67 68 69
	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();
70
	ns->ops->put(ns);
71
got_it:
72
	path->mnt = mntget(mnt);
73 74 75 76 77 78
	path->dentry = dentry;
	return NULL;
slow:
	rcu_read_unlock();
	inode = new_inode_pseudo(mnt->mnt_sb);
	if (!inode) {
79
		ns->ops->put(ns);
80 81 82
		return ERR_PTR(-ENOMEM);
	}
	inode->i_ino = ns->inum;
83
	inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
84 85 86 87 88
	inode->i_flags |= S_IMMUTABLE;
	inode->i_mode = S_IFREG | S_IRUGO;
	inode->i_fop = &ns_file_operations;
	inode->i_private = ns;

89
	dentry = d_alloc_anon(mnt->mnt_sb);
90 91 92 93 94
	if (!dentry) {
		iput(inode);
		return ERR_PTR(-ENOMEM);
	}
	d_instantiate(dentry, inode);
95
	dentry->d_fsdata = (void *)ns->ops;
96 97 98 99 100
	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();
101
		return ERR_PTR(-EAGAIN);
102 103 104 105
	}
	goto got_it;
}

106 107
void *ns_get_path_cb(struct path *path, ns_get_path_helper_t *ns_get_cb,
		     void *private_data)
108 109 110
{
	void *ret;

111 112 113 114 115 116 117
	do {
		struct ns_common *ns = ns_get_cb(private_data);
		if (!ns)
			return ERR_PTR(-ENOENT);

		ret = __ns_get_path(path, ns);
	} while (ret == ERR_PTR(-EAGAIN));
118 119 120 121

	return ret;
}

122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144
struct ns_get_path_task_args {
	const struct proc_ns_operations *ns_ops;
	struct task_struct *task;
};

static struct ns_common *ns_get_path_task(void *private_data)
{
	struct ns_get_path_task_args *args = private_data;

	return args->ns_ops->get(args->task);
}

void *ns_get_path(struct path *path, struct task_struct *task,
		  const struct proc_ns_operations *ns_ops)
{
	struct ns_get_path_task_args args = {
		.ns_ops	= ns_ops,
		.task	= task,
	};

	return ns_get_path_cb(path, ns_get_path_task, &args);
}

145
int open_related_ns(struct ns_common *ns,
146 147 148 149 150 151 152 153 154 155 156
		   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;

157
	do {
158 159 160 161 162 163 164 165 166
		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);
167 168
	} while (err == ERR_PTR(-EAGAIN));

169 170 171 172 173 174 175 176 177 178 179 180 181 182 183
	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;
}
184
EXPORT_SYMBOL_GPL(open_related_ns);
185 186 187 188

static long ns_ioctl(struct file *filp, unsigned int ioctl,
			unsigned long arg)
{
189
	struct user_namespace *user_ns;
190
	struct ns_common *ns = get_proc_ns(file_inode(filp));
191 192
	uid_t __user *argp;
	uid_t uid;
193 194 195 196 197 198 199 200

	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);
201 202 203 204 205 206 207 208 209
	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);
210 211 212 213 214
	default:
		return -ENOTTY;
	}
}

215 216 217 218 219
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;
220
	const char *name;
221 222
	ns = ns_ops->get(task);
	if (ns) {
223 224
		name = ns_ops->real_ns_name ? : ns_ops->name;
		res = snprintf(buf, size, "%s:[%u]", name, ns->inum);
225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261
		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,
};
262 263

static int nsfs_init_fs_context(struct fs_context *fc)
264
{
265 266 267 268 269 270
	struct pseudo_fs_context *ctx = init_pseudo(fc, NSFS_MAGIC);
	if (!ctx)
		return -ENOMEM;
	ctx->ops = &nsfs_ops;
	ctx->dops = &ns_dentry_operations;
	return 0;
271
}
272

273 274
static struct file_system_type nsfs = {
	.name = "nsfs",
275
	.init_fs_context = nsfs_init_fs_context,
276 277 278 279 280 281 282 283
	.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");
284
	nsfs_mnt->mnt_sb->s_flags &= ~SB_NOUSER;
285
}