[linux-2.6-block.git] / fs / proc_namespace.c

/*
 * fs/proc_namespace.c - handling of /proc/<pid>/{mounts,mountinfo,mountstats}
 *
 * In fact, that's a piece of procfs; it's *almost* isolated from
 * the rest of fs/proc, but has rather close relationships with
 * fs/namespace.c, thus here instead of fs/proc
 *
 */
#include <linux/mnt_namespace.h>
#include <linux/nsproxy.h>
#include <linux/security.h>
#include <linux/fs_struct.h>
#include "proc/internal.h" /* only for get_proc_task() in ->open() */

#include "pnode.h"
#include "internal.h"

static unsigned mounts_poll(struct file *file, poll_table *wait)
{
	struct proc_mounts *p = proc_mounts(file->private_data);
	struct mnt_namespace *ns = p->ns;
	unsigned res = POLLIN | POLLRDNORM;
	int event;

	poll_wait(file, &p->ns->poll, wait);

	event = ACCESS_ONCE(ns->event);
	if (p->m.poll_event != event) {
		p->m.poll_event = event;
		res |= POLLERR | POLLPRI;
	}

	return res;
}

struct proc_fs_info {
	int flag;
	const char *str;
};

static int show_sb_opts(struct seq_file *m, struct super_block *sb)
{
	static const struct proc_fs_info fs_info[] = {
		{ MS_SYNCHRONOUS, ",sync" },
		{ MS_DIRSYNC, ",dirsync" },
		{ MS_MANDLOCK, ",mand" },
		{ MS_LAZYTIME, ",lazytime" },
		{ 0, NULL }
	};
	const struct proc_fs_info *fs_infop;

	for (fs_infop = fs_info; fs_infop->flag; fs_infop++) {
		if (sb->s_flags & fs_infop->flag)
			seq_puts(m, fs_infop->str);
	}

	return security_sb_show_options(m, sb);
}

static void show_mnt_opts(struct seq_file *m, struct vfsmount *mnt)
{
	static const struct proc_fs_info mnt_info[] = {
		{ MNT_NOSUID, ",nosuid" },
		{ MNT_NODEV, ",nodev" },
		{ MNT_NOEXEC, ",noexec" },
		{ MNT_NOATIME, ",noatime" },
		{ MNT_NODIRATIME, ",nodiratime" },
		{ MNT_RELATIME, ",relatime" },
		{ 0, NULL }
	};
	const struct proc_fs_info *fs_infop;

	for (fs_infop = mnt_info; fs_infop->flag; fs_infop++) {
		if (mnt->mnt_flags & fs_infop->flag)
			seq_puts(m, fs_infop->str);
	}
}

static inline void mangle(struct seq_file *m, const char *s)
{
	seq_escape(m, s, " \t\n\\");
}

static void show_type(struct seq_file *m, struct super_block *sb)
{
	mangle(m, sb->s_type->name);
	if (sb->s_subtype && sb->s_subtype[0]) {
		seq_putc(m, '.');
		mangle(m, sb->s_subtype);
	}
}

static int show_vfsmnt(struct seq_file *m, struct vfsmount *mnt)
{
	struct proc_mounts *p = proc_mounts(m);
	struct mount *r = real_mount(mnt);
	int err = 0;
	struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt };
	struct super_block *sb = mnt_path.dentry->d_sb;

	if (sb->s_op->show_devname) {
		err = sb->s_op->show_devname(m, mnt_path.dentry);
		if (err)
			goto out;
	} else {
		mangle(m, r->mnt_devname ? r->mnt_devname : "none");
	}
	seq_putc(m, ' ');
	/* mountpoints outside of chroot jail will give SEQ_SKIP on this */
	err = seq_path_root(m, &mnt_path, &p->root, " \t\n\\");
	if (err)
		goto out;
	seq_putc(m, ' ');
	show_type(m, sb);
	seq_puts(m, __mnt_is_readonly(mnt) ? " ro" : " rw");
	err = show_sb_opts(m, sb);
	if (err)
		goto out;
	show_mnt_opts(m, mnt);
	if (sb->s_op->show_options)
		err = sb->s_op->show_options(m, mnt_path.dentry);
	seq_puts(m, " 0 0\n");
out:
	return err;
}

static int show_mountinfo(struct seq_file *m, struct vfsmount *mnt)
{
	struct proc_mounts *p = proc_mounts(m);
	struct mount *r = real_mount(mnt);
	struct super_block *sb = mnt->mnt_sb;
	struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt };
	int err = 0;

	seq_printf(m, "%i %i %u:%u ", r->mnt_id, r->mnt_parent->mnt_id,
		   MAJOR(sb->s_dev), MINOR(sb->s_dev));
	if (sb->s_op->show_path)
		err = sb->s_op->show_path(m, mnt->mnt_root);
	else
		seq_dentry(m, mnt->mnt_root, " \t\n\\");
	if (err)
		goto out;
	seq_putc(m, ' ');

	/* mountpoints outside of chroot jail will give SEQ_SKIP on this */
	err = seq_path_root(m, &mnt_path, &p->root, " \t\n\\");
	if (err)
		goto out;

	seq_puts(m, mnt->mnt_flags & MNT_READONLY ? " ro" : " rw");
	show_mnt_opts(m, mnt);

	/* Tagged fields ("foo:X" or "bar") */
	if (IS_MNT_SHARED(r))
		seq_printf(m, " shared:%i", r->mnt_group_id);
	if (IS_MNT_SLAVE(r)) {
		int master = r->mnt_master->mnt_group_id;
		int dom = get_dominating_id(r, &p->root);
		seq_printf(m, " master:%i", master);
		if (dom && dom != master)
			seq_printf(m, " propagate_from:%i", dom);
	}
	if (IS_MNT_UNBINDABLE(r))
		seq_puts(m, " unbindable");

	/* Filesystem specific data */
	seq_puts(m, " - ");
	show_type(m, sb);
	seq_putc(m, ' ');
	if (sb->s_op->show_devname)
		err = sb->s_op->show_devname(m, mnt->mnt_root);
	else
		mangle(m, r->mnt_devname ? r->mnt_devname : "none");
	if (err)
		goto out;
	seq_puts(m, sb->s_flags & MS_RDONLY ? " ro" : " rw");
	err = show_sb_opts(m, sb);
	if (err)
		goto out;
	if (sb->s_op->show_options)
		err = sb->s_op->show_options(m, mnt->mnt_root);
	seq_putc(m, '\n');
out:
	return err;
}

static int show_vfsstat(struct seq_file *m, struct vfsmount *mnt)
{
	struct proc_mounts *p = proc_mounts(m);
	struct mount *r = real_mount(mnt);
	struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt };
	struct super_block *sb = mnt_path.dentry->d_sb;
	int err = 0;

	/* device */
	if (sb->s_op->show_devname) {
		seq_puts(m, "device ");
		err = sb->s_op->show_devname(m, mnt_path.dentry);
	} else {
		if (r->mnt_devname) {
			seq_puts(m, "device ");
			mangle(m, r->mnt_devname);
		} else
			seq_puts(m, "no device");
	}

	/* mount point */
	seq_puts(m, " mounted on ");
	/* mountpoints outside of chroot jail will give SEQ_SKIP on this */
	err = seq_path_root(m, &mnt_path, &p->root, " \t\n\\");
	if (err)
		goto out;
	seq_putc(m, ' ');

	/* file system type */
	seq_puts(m, "with fstype ");
	show_type(m, sb);

	/* optional statistics */
	if (sb->s_op->show_stats) {
		seq_putc(m, ' ');
		if (!err)
			err = sb->s_op->show_stats(m, mnt_path.dentry);
	}

	seq_putc(m, '\n');
out:
	return err;
}

static int mounts_open_common(struct inode *inode, struct file *file,
			      int (*show)(struct seq_file *, struct vfsmount *))
{
	struct task_struct *task = get_proc_task(inode);
	struct nsproxy *nsp;
	struct mnt_namespace *ns = NULL;
	struct path root;
	struct proc_mounts *p;
	int ret = -EINVAL;

	if (!task)
		goto err;

	task_lock(task);
	nsp = task->nsproxy;
	if (!nsp || !nsp->mnt_ns) {
		task_unlock(task);
		put_task_struct(task);
		goto err;
	}
	ns = nsp->mnt_ns;
	get_mnt_ns(ns);
	if (!task->fs) {
		task_unlock(task);
		put_task_struct(task);
		ret = -ENOENT;
		goto err_put_ns;
	}
	get_fs_root(task->fs, &root);
	task_unlock(task);
	put_task_struct(task);

	ret = -ENOMEM;
	p = kmalloc(sizeof(struct proc_mounts), GFP_KERNEL);
	if (!p)
		goto err_put_path;

	file->private_data = &p->m;
	ret = seq_open(file, &mounts_op);
	if (ret)
		goto err_free;

	p->ns = ns;
	p->root = root;
	p->m.poll_event = ns->event;
	p->show = show;
	p->cached_event = ~0ULL;

	return 0;

 err_free:
	kfree(p);
 err_put_path:
	path_put(&root);
 err_put_ns:
	put_mnt_ns(ns);
 err:
	return ret;
}

static int mounts_release(struct inode *inode, struct file *file)
{
	struct proc_mounts *p = proc_mounts(file->private_data);
	path_put(&p->root);
	put_mnt_ns(p->ns);
	return seq_release(inode, file);
}

static int mounts_open(struct inode *inode, struct file *file)
{
	return mounts_open_common(inode, file, show_vfsmnt);
}

static int mountinfo_open(struct inode *inode, struct file *file)
{
	return mounts_open_common(inode, file, show_mountinfo);
}

static int mountstats_open(struct inode *inode, struct file *file)
{
	return mounts_open_common(inode, file, show_vfsstat);
}

const struct file_operations proc_mounts_operations = {
	.open		= mounts_open,
	.read		= seq_read,
	.llseek		= seq_lseek,
	.release	= mounts_release,
	.poll		= mounts_poll,
};

const struct file_operations proc_mountinfo_operations = {
	.open		= mountinfo_open,
	.read		= seq_read,
	.llseek		= seq_lseek,
	.release	= mounts_release,
	.poll		= mounts_poll,
};

const struct file_operations proc_mountstats_operations = {
	.open		= mountstats_open,
	.read		= seq_read,
	.llseek		= seq_lseek,
	.release	= mounts_release,
};
Commit	Line	Data
0226f492 AV	1	/*
	2	* fs/proc_namespace.c - handling of /proc/<pid>/{mounts,mountinfo,mountstats}
	3	*
	4	* In fact, that's a piece of procfs; it's almost isolated from
	5	* the rest of fs/proc, but has rather close relationships with
	6	* fs/namespace.c, thus here instead of fs/proc
	7	*
	8	*/
	9	#include <linux/mnt_namespace.h>
	10	#include <linux/nsproxy.h>
	11	#include <linux/security.h>
	12	#include <linux/fs_struct.h>
	13	#include "proc/internal.h" /* only for get_proc_task() in ->open() */
	14
	15	#include "pnode.h"
	16	#include "internal.h"
	17
	18	static unsigned mounts_poll(struct file file, poll_table wait)
	19	{
6ce6e24e	20	struct proc_mounts *p = proc_mounts(file->private_data);
0226f492 AV	21	struct mnt_namespace *ns = p->ns;
0226f492 AV	22	unsigned res = POLLIN \| POLLRDNORM;
aab407fc	23	int event;
0226f492 AV	24
	25	poll_wait(file, &p->ns->poll, wait);
	26
aab407fc AV	27	event = ACCESS_ONCE(ns->event);
	28	if (p->m.poll_event != event) {
	29	p->m.poll_event = event;
0226f492 AV	30	res \|= POLLERR \| POLLPRI;
0226f492 AV	31	}
0226f492 AV	32
	33	return res;
	34	}
	35
	36	struct proc_fs_info {
	37	int flag;
	38	const char *str;
	39	};
	40
	41	static int show_sb_opts(struct seq_file m, struct super_block sb)
	42	{
	43	static const struct proc_fs_info fs_info[] = {
	44	{ MS_SYNCHRONOUS, ",sync" },
	45	{ MS_DIRSYNC, ",dirsync" },
	46	{ MS_MANDLOCK, ",mand" },
0ae45f63	47	{ MS_LAZYTIME, ",lazytime" },
0226f492 AV	48	{ 0, NULL }
	49	};
	50	const struct proc_fs_info *fs_infop;
	51
	52	for (fs_infop = fs_info; fs_infop->flag; fs_infop++) {
	53	if (sb->s_flags & fs_infop->flag)
	54	seq_puts(m, fs_infop->str);
	55	}
	56
	57	return security_sb_show_options(m, sb);
	58	}
	59
	60	static void show_mnt_opts(struct seq_file m, struct vfsmount mnt)
	61	{
	62	static const struct proc_fs_info mnt_info[] = {
	63	{ MNT_NOSUID, ",nosuid" },
	64	{ MNT_NODEV, ",nodev" },
	65	{ MNT_NOEXEC, ",noexec" },
	66	{ MNT_NOATIME, ",noatime" },
	67	{ MNT_NODIRATIME, ",nodiratime" },
	68	{ MNT_RELATIME, ",relatime" },
	69	{ 0, NULL }
	70	};
	71	const struct proc_fs_info *fs_infop;
	72
	73	for (fs_infop = mnt_info; fs_infop->flag; fs_infop++) {
	74	if (mnt->mnt_flags & fs_infop->flag)
	75	seq_puts(m, fs_infop->str);
	76	}
	77	}
	78
	79	static inline void mangle(struct seq_file m, const char s)
	80	{
	81	seq_escape(m, s, " \t\n\\");
	82	}
	83
	84	static void show_type(struct seq_file m, struct super_block sb)
	85	{
	86	mangle(m, sb->s_type->name);
	87	if (sb->s_subtype && sb->s_subtype[0]) {
	88	seq_putc(m, '.');
	89	mangle(m, sb->s_subtype);
	90	}
	91	}
	92
	93	static int show_vfsmnt(struct seq_file m, struct vfsmount mnt)
	94	{
9d4d6574	95	struct proc_mounts *p = proc_mounts(m);
0226f492 AV	96	struct mount *r = real_mount(mnt);
	97	int err = 0;
	98	struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt };
d861c630	99	struct super_block *sb = mnt_path.dentry->d_sb;
0226f492	100
d861c630 AV	101	if (sb->s_op->show_devname) {
d861c630 AV	102	err = sb->s_op->show_devname(m, mnt_path.dentry);
0226f492 AV	103	if (err)
	104	goto out;
	105	} else {
	106	mangle(m, r->mnt_devname ? r->mnt_devname : "none");
	107	}
	108	seq_putc(m, ' ');
9d4d6574 DL	109	/* mountpoints outside of chroot jail will give SEQ_SKIP on this */
	110	err = seq_path_root(m, &mnt_path, &p->root, " \t\n\\");
	111	if (err)
	112	goto out;
0226f492	113	seq_putc(m, ' ');
d861c630	114	show_type(m, sb);
0226f492	115	seq_puts(m, __mnt_is_readonly(mnt) ? " ro" : " rw");
d861c630	116	err = show_sb_opts(m, sb);
0226f492 AV	117	if (err)
	118	goto out;
	119	show_mnt_opts(m, mnt);
d861c630	120	if (sb->s_op->show_options)
34c80b1d	121	err = sb->s_op->show_options(m, mnt_path.dentry);
0226f492 AV	122	seq_puts(m, " 0 0\n");
	123	out:
	124	return err;
	125	}
	126
	127	static int show_mountinfo(struct seq_file m, struct vfsmount mnt)
	128	{
6ce6e24e	129	struct proc_mounts *p = proc_mounts(m);
0226f492 AV	130	struct mount *r = real_mount(mnt);
	131	struct super_block *sb = mnt->mnt_sb;
	132	struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt };
0226f492 AV	133	int err = 0;
	134
	135	seq_printf(m, "%i %i %u:%u ", r->mnt_id, r->mnt_parent->mnt_id,
	136	MAJOR(sb->s_dev), MINOR(sb->s_dev));
	137	if (sb->s_op->show_path)
a6322de6	138	err = sb->s_op->show_path(m, mnt->mnt_root);
0226f492 AV	139	else
	140	seq_dentry(m, mnt->mnt_root, " \t\n\\");
	141	if (err)
	142	goto out;
	143	seq_putc(m, ' ');
	144
	145	/* mountpoints outside of chroot jail will give SEQ_SKIP on this */
9ad4dc4f	146	err = seq_path_root(m, &mnt_path, &p->root, " \t\n\\");
0226f492 AV	147	if (err)
	148	goto out;
	149
	150	seq_puts(m, mnt->mnt_flags & MNT_READONLY ? " ro" : " rw");
	151	show_mnt_opts(m, mnt);
	152
	153	/* Tagged fields ("foo:X" or "bar") */
	154	if (IS_MNT_SHARED(r))
	155	seq_printf(m, " shared:%i", r->mnt_group_id);
	156	if (IS_MNT_SLAVE(r)) {
	157	int master = r->mnt_master->mnt_group_id;
	158	int dom = get_dominating_id(r, &p->root);
	159	seq_printf(m, " master:%i", master);
	160	if (dom && dom != master)
	161	seq_printf(m, " propagate_from:%i", dom);
	162	}
	163	if (IS_MNT_UNBINDABLE(r))
	164	seq_puts(m, " unbindable");
	165
	166	/* Filesystem specific data */
	167	seq_puts(m, " - ");
	168	show_type(m, sb);
	169	seq_putc(m, ' ');
	170	if (sb->s_op->show_devname)
d861c630	171	err = sb->s_op->show_devname(m, mnt->mnt_root);
0226f492 AV	172	else
	173	mangle(m, r->mnt_devname ? r->mnt_devname : "none");
	174	if (err)
	175	goto out;
	176	seq_puts(m, sb->s_flags & MS_RDONLY ? " ro" : " rw");
	177	err = show_sb_opts(m, sb);
	178	if (err)
	179	goto out;
	180	if (sb->s_op->show_options)
34c80b1d	181	err = sb->s_op->show_options(m, mnt->mnt_root);
0226f492 AV	182	seq_putc(m, '\n');
	183	out:
	184	return err;
	185	}
	186
	187	static int show_vfsstat(struct seq_file m, struct vfsmount mnt)
	188	{
9d4d6574	189	struct proc_mounts *p = proc_mounts(m);
0226f492 AV	190	struct mount *r = real_mount(mnt);
0226f492 AV	191	struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt };
64132379	192	struct super_block *sb = mnt_path.dentry->d_sb;
0226f492 AV	193	int err = 0;
	194
	195	/* device */
64132379	196	if (sb->s_op->show_devname) {
0226f492	197	seq_puts(m, "device ");
d861c630	198	err = sb->s_op->show_devname(m, mnt_path.dentry);
0226f492 AV	199	} else {
	200	if (r->mnt_devname) {
	201	seq_puts(m, "device ");
	202	mangle(m, r->mnt_devname);
	203	} else
	204	seq_puts(m, "no device");
	205	}
	206
	207	/* mount point */
	208	seq_puts(m, " mounted on ");
9d4d6574 DL	209	/* mountpoints outside of chroot jail will give SEQ_SKIP on this */
	210	err = seq_path_root(m, &mnt_path, &p->root, " \t\n\\");
	211	if (err)
	212	goto out;
0226f492 AV	213	seq_putc(m, ' ');
	214
	215	/* file system type */
	216	seq_puts(m, "with fstype ");
64132379	217	show_type(m, sb);
0226f492 AV	218
0226f492 AV	219	/* optional statistics */
64132379	220	if (sb->s_op->show_stats) {
0226f492 AV	221	seq_putc(m, ' ');
0226f492 AV	222	if (!err)
64132379	223	err = sb->s_op->show_stats(m, mnt_path.dentry);
0226f492 AV	224	}
	225
	226	seq_putc(m, '\n');
9d4d6574	227	out:
0226f492 AV	228	return err;
	229	}
	230
	231	static int mounts_open_common(struct inode inode, struct file file,
	232	int (show)(struct seq_file , struct vfsmount *))
	233	{
	234	struct task_struct *task = get_proc_task(inode);
	235	struct nsproxy *nsp;
	236	struct mnt_namespace *ns = NULL;
	237	struct path root;
	238	struct proc_mounts *p;
	239	int ret = -EINVAL;
	240
	241	if (!task)
	242	goto err;
	243
728dba3a EB	244	task_lock(task);
728dba3a EB	245	nsp = task->nsproxy;
3d93116c	246	if (!nsp \|\| !nsp->mnt_ns) {
728dba3a	247	task_unlock(task);
0226f492 AV	248	put_task_struct(task);
	249	goto err;
	250	}
	251	ns = nsp->mnt_ns;
0226f492	252	get_mnt_ns(ns);
0226f492 AV	253	if (!task->fs) {
	254	task_unlock(task);
	255	put_task_struct(task);
	256	ret = -ENOENT;
	257	goto err_put_ns;
	258	}
	259	get_fs_root(task->fs, &root);
	260	task_unlock(task);
	261	put_task_struct(task);
	262
	263	ret = -ENOMEM;
	264	p = kmalloc(sizeof(struct proc_mounts), GFP_KERNEL);
	265	if (!p)
	266	goto err_put_path;
	267
	268	file->private_data = &p->m;
	269	ret = seq_open(file, &mounts_op);
	270	if (ret)
	271	goto err_free;
	272
0226f492 AV	273	p->ns = ns;
	274	p->root = root;
	275	p->m.poll_event = ns->event;
	276	p->show = show;
c7999c36	277	p->cached_event = ~0ULL;
0226f492 AV	278
	279	return 0;
	280
	281	err_free:
	282	kfree(p);
	283	err_put_path:
	284	path_put(&root);
	285	err_put_ns:
	286	put_mnt_ns(ns);
	287	err:
	288	return ret;
	289	}
	290
	291	static int mounts_release(struct inode inode, struct file file)
	292	{
6ce6e24e	293	struct proc_mounts *p = proc_mounts(file->private_data);
0226f492 AV	294	path_put(&p->root);
	295	put_mnt_ns(p->ns);
	296	return seq_release(inode, file);
	297	}
	298
	299	static int mounts_open(struct inode inode, struct file file)
	300	{
	301	return mounts_open_common(inode, file, show_vfsmnt);
	302	}
	303
	304	static int mountinfo_open(struct inode inode, struct file file)
	305	{
	306	return mounts_open_common(inode, file, show_mountinfo);
	307	}
	308
	309	static int mountstats_open(struct inode inode, struct file file)
	310	{
	311	return mounts_open_common(inode, file, show_vfsstat);
	312	}
	313
	314	const struct file_operations proc_mounts_operations = {
	315	.open = mounts_open,
	316	.read = seq_read,
	317	.llseek = seq_lseek,
	318	.release = mounts_release,
	319	.poll = mounts_poll,
	320	};
	321
	322	const struct file_operations proc_mountinfo_operations = {
	323	.open = mountinfo_open,
	324	.read = seq_read,
	325	.llseek = seq_lseek,
	326	.release = mounts_release,
	327	.poll = mounts_poll,
	328	};
	329
	330	const struct file_operations proc_mountstats_operations = {
	331	.open = mountstats_open,
	332	.read = seq_read,
	333	.llseek = seq_lseek,
	334	.release = mounts_release,
	335	};