[linux-2.6-block.git] / fs / proc / proc_sysctl.c

/*
 * /proc/sys support
 */
#include <linux/init.h>
#include <linux/sysctl.h>
#include <linux/poll.h>
#include <linux/proc_fs.h>
#include <linux/security.h>
#include <linux/namei.h>
#include "internal.h"

static const struct dentry_operations proc_sys_dentry_operations;
static const struct file_operations proc_sys_file_operations;
static const struct inode_operations proc_sys_inode_operations;
static const struct file_operations proc_sys_dir_file_operations;
static const struct inode_operations proc_sys_dir_operations;

void proc_sys_poll_notify(struct ctl_table_poll *poll)
{
	if (!poll)
		return;

	atomic_inc(&poll->event);
	wake_up_interruptible(&poll->wait);
}

static struct inode *proc_sys_make_inode(struct super_block *sb,
		struct ctl_table_header *head, struct ctl_table *table)
{
	struct inode *inode;
	struct proc_inode *ei;

	inode = new_inode(sb);
	if (!inode)
		goto out;

	inode->i_ino = get_next_ino();

	sysctl_head_get(head);
	ei = PROC_I(inode);
	ei->sysctl = head;
	ei->sysctl_entry = table;

	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
	inode->i_mode = table->mode;
	if (!table->child) {
		inode->i_mode |= S_IFREG;
		inode->i_op = &proc_sys_inode_operations;
		inode->i_fop = &proc_sys_file_operations;
	} else {
		inode->i_mode |= S_IFDIR;
		clear_nlink(inode);
		inode->i_op = &proc_sys_dir_operations;
		inode->i_fop = &proc_sys_dir_file_operations;
	}
out:
	return inode;
}

static struct ctl_table *find_in_table(struct ctl_table *p, struct qstr *name)
{
	for ( ; p->procname; p++) {
		if (strlen(p->procname) != name->len)
			continue;

		if (memcmp(p->procname, name->name, name->len) != 0)
			continue;

		/* I have a match */
		return p;
	}
	return NULL;
}

static struct ctl_table_header *grab_header(struct inode *inode)
{
	if (PROC_I(inode)->sysctl)
		return sysctl_head_grab(PROC_I(inode)->sysctl);
	else
		return sysctl_head_next(NULL);
}

static struct dentry *proc_sys_lookup(struct inode *dir, struct dentry *dentry,
					struct nameidata *nd)
{
	struct ctl_table_header *head = grab_header(dir);
	struct ctl_table *table = PROC_I(dir)->sysctl_entry;
	struct ctl_table_header *h = NULL;
	struct qstr *name = &dentry->d_name;
	struct ctl_table *p;
	struct inode *inode;
	struct dentry *err = ERR_PTR(-ENOENT);

	if (IS_ERR(head))
		return ERR_CAST(head);

	if (table && !table->child) {
		WARN_ON(1);
		goto out;
	}

	table = table ? table->child : head->ctl_table;

	p = find_in_table(table, name);
	if (!p) {
		for (h = sysctl_head_next(NULL); h; h = sysctl_head_next(h)) {
			if (h->attached_to != table)
				continue;
			p = find_in_table(h->attached_by, name);
			if (p)
				break;
		}
	}

	if (!p)
		goto out;

	err = ERR_PTR(-ENOMEM);
	inode = proc_sys_make_inode(dir->i_sb, h ? h : head, p);
	if (h)
		sysctl_head_finish(h);

	if (!inode)
		goto out;

	err = NULL;
	d_set_d_op(dentry, &proc_sys_dentry_operations);
	d_add(dentry, inode);

out:
	sysctl_head_finish(head);
	return err;
}

static ssize_t proc_sys_call_handler(struct file *filp, void __user *buf,
		size_t count, loff_t *ppos, int write)
{
	struct inode *inode = filp->f_path.dentry->d_inode;
	struct ctl_table_header *head = grab_header(inode);
	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
	ssize_t error;
	size_t res;

	if (IS_ERR(head))
		return PTR_ERR(head);

	/*
	 * At this point we know that the sysctl was not unregistered
	 * and won't be until we finish.
	 */
	error = -EPERM;
	if (sysctl_perm(head->root, table, write ? MAY_WRITE : MAY_READ))
		goto out;

	/* if that can happen at all, it should be -EINVAL, not -EISDIR */
	error = -EINVAL;
	if (!table->proc_handler)
		goto out;

	/* careful: calling conventions are nasty here */
	res = count;
	error = table->proc_handler(table, write, buf, &res, ppos);
	if (!error)
		error = res;
out:
	sysctl_head_finish(head);

	return error;
}

static ssize_t proc_sys_read(struct file *filp, char __user *buf,
				size_t count, loff_t *ppos)
{
	return proc_sys_call_handler(filp, (void __user *)buf, count, ppos, 0);
}

static ssize_t proc_sys_write(struct file *filp, const char __user *buf,
				size_t count, loff_t *ppos)
{
	return proc_sys_call_handler(filp, (void __user *)buf, count, ppos, 1);
}

static int proc_sys_open(struct inode *inode, struct file *filp)
{
	struct ctl_table *table = PROC_I(inode)->sysctl_entry;

	if (table->poll)
		filp->private_data = proc_sys_poll_event(table->poll);

	return 0;
}

static unsigned int proc_sys_poll(struct file *filp, poll_table *wait)
{
	struct inode *inode = filp->f_path.dentry->d_inode;
	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
	unsigned long event = (unsigned long)filp->private_data;
	unsigned int ret = DEFAULT_POLLMASK;

	if (!table->proc_handler)
		goto out;

	if (!table->poll)
		goto out;

	poll_wait(filp, &table->poll->wait, wait);

	if (event != atomic_read(&table->poll->event)) {
		filp->private_data = proc_sys_poll_event(table->poll);
		ret = POLLIN | POLLRDNORM | POLLERR | POLLPRI;
	}

out:
	return ret;
}

static int proc_sys_fill_cache(struct file *filp, void *dirent,
				filldir_t filldir,
				struct ctl_table_header *head,
				struct ctl_table *table)
{
	struct dentry *child, *dir = filp->f_path.dentry;
	struct inode *inode;
	struct qstr qname;
	ino_t ino = 0;
	unsigned type = DT_UNKNOWN;

	qname.name = table->procname;
	qname.len  = strlen(table->procname);
	qname.hash = full_name_hash(qname.name, qname.len);

	child = d_lookup(dir, &qname);
	if (!child) {
		child = d_alloc(dir, &qname);
		if (child) {
			inode = proc_sys_make_inode(dir->d_sb, head, table);
			if (!inode) {
				dput(child);
				return -ENOMEM;
			} else {
				d_set_d_op(child, &proc_sys_dentry_operations);
				d_add(child, inode);
			}
		} else {
			return -ENOMEM;
		}
	}
	inode = child->d_inode;
	ino  = inode->i_ino;
	type = inode->i_mode >> 12;
	dput(child);
	return !!filldir(dirent, qname.name, qname.len, filp->f_pos, ino, type);
}

static int scan(struct ctl_table_header *head, ctl_table *table,
		unsigned long *pos, struct file *file,
		void *dirent, filldir_t filldir)
{

	for (; table->procname; table++, (*pos)++) {
		int res;

		if (*pos < file->f_pos)
			continue;

		res = proc_sys_fill_cache(file, dirent, filldir, head, table);
		if (res)
			return res;

		file->f_pos = *pos + 1;
	}
	return 0;
}

static int proc_sys_readdir(struct file *filp, void *dirent, filldir_t filldir)
{
	struct dentry *dentry = filp->f_path.dentry;
	struct inode *inode = dentry->d_inode;
	struct ctl_table_header *head = grab_header(inode);
	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
	struct ctl_table_header *h = NULL;
	unsigned long pos;
	int ret = -EINVAL;

	if (IS_ERR(head))
		return PTR_ERR(head);

	if (table && !table->child) {
		WARN_ON(1);
		goto out;
	}

	table = table ? table->child : head->ctl_table;

	ret = 0;
	/* Avoid a switch here: arm builds fail with missing __cmpdi2 */
	if (filp->f_pos == 0) {
		if (filldir(dirent, ".", 1, filp->f_pos,
				inode->i_ino, DT_DIR) < 0)
			goto out;
		filp->f_pos++;
	}
	if (filp->f_pos == 1) {
		if (filldir(dirent, "..", 2, filp->f_pos,
				parent_ino(dentry), DT_DIR) < 0)
			goto out;
		filp->f_pos++;
	}
	pos = 2;

	ret = scan(head, table, &pos, filp, dirent, filldir);
	if (ret)
		goto out;

	for (h = sysctl_head_next(NULL); h; h = sysctl_head_next(h)) {
		if (h->attached_to != table)
			continue;
		ret = scan(h, h->attached_by, &pos, filp, dirent, filldir);
		if (ret) {
			sysctl_head_finish(h);
			break;
		}
	}
	ret = 1;
out:
	sysctl_head_finish(head);
	return ret;
}

static int proc_sys_permission(struct inode *inode, int mask)
{
	/*
	 * sysctl entries that are not writeable,
	 * are _NOT_ writeable, capabilities or not.
	 */
	struct ctl_table_header *head;
	struct ctl_table *table;
	int error;

	/* Executable files are not allowed under /proc/sys/ */
	if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode))
		return -EACCES;

	head = grab_header(inode);
	if (IS_ERR(head))
		return PTR_ERR(head);

	table = PROC_I(inode)->sysctl_entry;
	if (!table) /* global root - r-xr-xr-x */
		error = mask & MAY_WRITE ? -EACCES : 0;
	else /* Use the permissions on the sysctl table entry */
		error = sysctl_perm(head->root, table, mask & ~MAY_NOT_BLOCK);

	sysctl_head_finish(head);
	return error;
}

static int proc_sys_setattr(struct dentry *dentry, struct iattr *attr)
{
	struct inode *inode = dentry->d_inode;
	int error;

	if (attr->ia_valid & (ATTR_MODE | ATTR_UID | ATTR_GID))
		return -EPERM;

	error = inode_change_ok(inode, attr);
	if (error)
		return error;

	if ((attr->ia_valid & ATTR_SIZE) &&
	    attr->ia_size != i_size_read(inode)) {
		error = vmtruncate(inode, attr->ia_size);
		if (error)
			return error;
	}

	setattr_copy(inode, attr);
	mark_inode_dirty(inode);
	return 0;
}

static int proc_sys_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
{
	struct inode *inode = dentry->d_inode;
	struct ctl_table_header *head = grab_header(inode);
	struct ctl_table *table = PROC_I(inode)->sysctl_entry;

	if (IS_ERR(head))
		return PTR_ERR(head);

	generic_fillattr(inode, stat);
	if (table)
		stat->mode = (stat->mode & S_IFMT) | table->mode;

	sysctl_head_finish(head);
	return 0;
}

static const struct file_operations proc_sys_file_operations = {
	.open		= proc_sys_open,
	.poll		= proc_sys_poll,
	.read		= proc_sys_read,
	.write		= proc_sys_write,
	.llseek		= default_llseek,
};

static const struct file_operations proc_sys_dir_file_operations = {
	.read		= generic_read_dir,
	.readdir	= proc_sys_readdir,
	.llseek		= generic_file_llseek,
};

static const struct inode_operations proc_sys_inode_operations = {
	.permission	= proc_sys_permission,
	.setattr	= proc_sys_setattr,
	.getattr	= proc_sys_getattr,
};

static const struct inode_operations proc_sys_dir_operations = {
	.lookup		= proc_sys_lookup,
	.permission	= proc_sys_permission,
	.setattr	= proc_sys_setattr,
	.getattr	= proc_sys_getattr,
};

static int proc_sys_revalidate(struct dentry *dentry, struct nameidata *nd)
{
	if (nd->flags & LOOKUP_RCU)
		return -ECHILD;
	return !PROC_I(dentry->d_inode)->sysctl->unregistering;
}

static int proc_sys_delete(const struct dentry *dentry)
{
	return !!PROC_I(dentry->d_inode)->sysctl->unregistering;
}

static int proc_sys_compare(const struct dentry *parent,
		const struct inode *pinode,
		const struct dentry *dentry, const struct inode *inode,
		unsigned int len, const char *str, const struct qstr *name)
{
	struct ctl_table_header *head;
	/* Although proc doesn't have negative dentries, rcu-walk means
	 * that inode here can be NULL */
	/* AV: can it, indeed? */
	if (!inode)
		return 1;
	if (name->len != len)
		return 1;
	if (memcmp(name->name, str, len))
		return 1;
	head = rcu_dereference(PROC_I(inode)->sysctl);
	return !head || !sysctl_is_seen(head);
}

static const struct dentry_operations proc_sys_dentry_operations = {
	.d_revalidate	= proc_sys_revalidate,
	.d_delete	= proc_sys_delete,
	.d_compare	= proc_sys_compare,
};

int __init proc_sys_init(void)
{
	struct proc_dir_entry *proc_sys_root;

	proc_sys_root = proc_mkdir("sys", NULL);
	proc_sys_root->proc_iops = &proc_sys_dir_operations;
	proc_sys_root->proc_fops = &proc_sys_dir_file_operations;
	proc_sys_root->nlink = 0;
	return 0;
}
Commit	Line	Data
77b14db5 EB	1	/*
	2	* /proc/sys support
	3	*/
1e0edd3f	4	#include <linux/init.h>
77b14db5	5	#include <linux/sysctl.h>
f1ecf068	6	#include <linux/poll.h>
77b14db5 EB	7	#include <linux/proc_fs.h>
77b14db5 EB	8	#include <linux/security.h>
34286d66	9	#include <linux/namei.h>
77b14db5 EB	10	#include "internal.h"
77b14db5 EB	11
d72f71eb	12	static const struct dentry_operations proc_sys_dentry_operations;
77b14db5	13	static const struct file_operations proc_sys_file_operations;
03a44825	14	static const struct inode_operations proc_sys_inode_operations;
9043476f AV	15	static const struct file_operations proc_sys_dir_file_operations;
9043476f AV	16	static const struct inode_operations proc_sys_dir_operations;
77b14db5	17
f1ecf068 LDM	18	void proc_sys_poll_notify(struct ctl_table_poll *poll)
	19	{
	20	if (!poll)
	21	return;
	22
	23	atomic_inc(&poll->event);
	24	wake_up_interruptible(&poll->wait);
	25	}
	26
9043476f AV	27	static struct inode proc_sys_make_inode(struct super_block sb,
9043476f AV	28	struct ctl_table_header head, struct ctl_table table)
77b14db5 EB	29	{
77b14db5 EB	30	struct inode *inode;
9043476f	31	struct proc_inode *ei;
77b14db5	32
9043476f	33	inode = new_inode(sb);
77b14db5 EB	34	if (!inode)
	35	goto out;
	36
85fe4025 CH	37	inode->i_ino = get_next_ino();
85fe4025 CH	38
9043476f	39	sysctl_head_get(head);
77b14db5	40	ei = PROC_I(inode);
9043476f AV	41	ei->sysctl = head;
	42	ei->sysctl_entry = table;
	43
77b14db5	44	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
9043476f AV	45	inode->i_mode = table->mode;
	46	if (!table->child) {
	47	inode->i_mode \|= S_IFREG;
	48	inode->i_op = &proc_sys_inode_operations;
	49	inode->i_fop = &proc_sys_file_operations;
	50	} else {
	51	inode->i_mode \|= S_IFDIR;
6d6b77f1	52	clear_nlink(inode);
9043476f AV	53	inode->i_op = &proc_sys_dir_operations;
	54	inode->i_fop = &proc_sys_dir_file_operations;
	55	}
77b14db5 EB	56	out:
	57	return inode;
	58	}
	59
9043476f	60	static struct ctl_table find_in_table(struct ctl_table p, struct qstr *name)
77b14db5	61	{
2315ffa0	62	for ( ; p->procname; p++) {
36885d7b	63	if (strlen(p->procname) != name->len)
77b14db5 EB	64	continue;
77b14db5 EB	65
36885d7b	66	if (memcmp(p->procname, name->name, name->len) != 0)
77b14db5 EB	67	continue;
	68
	69	/* I have a match */
9043476f	70	return p;
77b14db5 EB	71	}
	72	return NULL;
	73	}
	74
81324364	75	static struct ctl_table_header grab_header(struct inode inode)
77b14db5	76	{
9043476f AV	77	if (PROC_I(inode)->sysctl)
	78	return sysctl_head_grab(PROC_I(inode)->sysctl);
	79	else
	80	return sysctl_head_next(NULL);
	81	}
77b14db5	82
9043476f AV	83	static struct dentry proc_sys_lookup(struct inode dir, struct dentry *dentry,
	84	struct nameidata *nd)
	85	{
	86	struct ctl_table_header *head = grab_header(dir);
	87	struct ctl_table *table = PROC_I(dir)->sysctl_entry;
	88	struct ctl_table_header *h = NULL;
	89	struct qstr *name = &dentry->d_name;
	90	struct ctl_table *p;
	91	struct inode *inode;
	92	struct dentry *err = ERR_PTR(-ENOENT);
77b14db5	93
9043476f AV	94	if (IS_ERR(head))
9043476f AV	95	return ERR_CAST(head);
77b14db5	96
9043476f AV	97	if (table && !table->child) {
	98	WARN_ON(1);
	99	goto out;
77b14db5	100	}
77b14db5	101
9043476f	102	table = table ? table->child : head->ctl_table;
77b14db5	103
9043476f AV	104	p = find_in_table(table, name);
	105	if (!p) {
	106	for (h = sysctl_head_next(NULL); h; h = sysctl_head_next(h)) {
	107	if (h->attached_to != table)
	108	continue;
	109	p = find_in_table(h->attached_by, name);
	110	if (p)
	111	break;
	112	}
77b14db5	113	}
77b14db5	114
9043476f	115	if (!p)
77b14db5 EB	116	goto out;
	117
	118	err = ERR_PTR(-ENOMEM);
9043476f AV	119	inode = proc_sys_make_inode(dir->i_sb, h ? h : head, p);
	120	if (h)
	121	sysctl_head_finish(h);
	122
77b14db5 EB	123	if (!inode)
	124	goto out;
	125
	126	err = NULL;
fb045adb	127	d_set_d_op(dentry, &proc_sys_dentry_operations);
77b14db5 EB	128	d_add(dentry, inode);
	129
	130	out:
	131	sysctl_head_finish(head);
	132	return err;
	133	}
	134
7708bfb1 PE	135	static ssize_t proc_sys_call_handler(struct file filp, void __user buf,
7708bfb1 PE	136	size_t count, loff_t *ppos, int write)
77b14db5	137	{
9043476f AV	138	struct inode *inode = filp->f_path.dentry->d_inode;
	139	struct ctl_table_header *head = grab_header(inode);
	140	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
2a2da53b DH	141	ssize_t error;
2a2da53b DH	142	size_t res;
77b14db5	143
9043476f AV	144	if (IS_ERR(head))
9043476f AV	145	return PTR_ERR(head);
77b14db5 EB	146
	147	/*
	148	* At this point we know that the sysctl was not unregistered
	149	* and won't be until we finish.
	150	*/
	151	error = -EPERM;
d7321cd6	152	if (sysctl_perm(head->root, table, write ? MAY_WRITE : MAY_READ))
77b14db5 EB	153	goto out;
77b14db5 EB	154
9043476f AV	155	/* if that can happen at all, it should be -EINVAL, not -EISDIR */
	156	error = -EINVAL;
	157	if (!table->proc_handler)
	158	goto out;
	159
77b14db5 EB	160	/* careful: calling conventions are nasty here */
77b14db5 EB	161	res = count;
8d65af78	162	error = table->proc_handler(table, write, buf, &res, ppos);
77b14db5 EB	163	if (!error)
	164	error = res;
	165	out:
	166	sysctl_head_finish(head);
	167
	168	return error;
	169	}
	170
7708bfb1	171	static ssize_t proc_sys_read(struct file filp, char __user buf,
77b14db5 EB	172	size_t count, loff_t *ppos)
77b14db5 EB	173	{
7708bfb1 PE	174	return proc_sys_call_handler(filp, (void __user *)buf, count, ppos, 0);
7708bfb1 PE	175	}
77b14db5	176
7708bfb1 PE	177	static ssize_t proc_sys_write(struct file filp, const char __user buf,
	178	size_t count, loff_t *ppos)
	179	{
	180	return proc_sys_call_handler(filp, (void __user *)buf, count, ppos, 1);
77b14db5 EB	181	}
77b14db5 EB	182
f1ecf068 LDM	183	static int proc_sys_open(struct inode inode, struct file filp)
	184	{
	185	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
	186
	187	if (table->poll)
	188	filp->private_data = proc_sys_poll_event(table->poll);
	189
	190	return 0;
	191	}
	192
	193	static unsigned int proc_sys_poll(struct file filp, poll_table wait)
	194	{
	195	struct inode *inode = filp->f_path.dentry->d_inode;
	196	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
	197	unsigned long event = (unsigned long)filp->private_data;
	198	unsigned int ret = DEFAULT_POLLMASK;
	199
	200	if (!table->proc_handler)
	201	goto out;
	202
	203	if (!table->poll)
	204	goto out;
	205
	206	poll_wait(filp, &table->poll->wait, wait);
	207
	208	if (event != atomic_read(&table->poll->event)) {
	209	filp->private_data = proc_sys_poll_event(table->poll);
	210	ret = POLLIN \| POLLRDNORM \| POLLERR \| POLLPRI;
	211	}
	212
	213	out:
	214	return ret;
	215	}
77b14db5 EB	216
77b14db5 EB	217	static int proc_sys_fill_cache(struct file filp, void dirent,
9043476f AV	218	filldir_t filldir,
	219	struct ctl_table_header *head,
	220	struct ctl_table *table)
77b14db5	221	{
77b14db5 EB	222	struct dentry child, dir = filp->f_path.dentry;
	223	struct inode *inode;
	224	struct qstr qname;
	225	ino_t ino = 0;
	226	unsigned type = DT_UNKNOWN;
77b14db5 EB	227
	228	qname.name = table->procname;
	229	qname.len = strlen(table->procname);
	230	qname.hash = full_name_hash(qname.name, qname.len);
	231
77b14db5 EB	232	child = d_lookup(dir, &qname);
77b14db5 EB	233	if (!child) {
9043476f AV	234	child = d_alloc(dir, &qname);
	235	if (child) {
	236	inode = proc_sys_make_inode(dir->d_sb, head, table);
	237	if (!inode) {
	238	dput(child);
	239	return -ENOMEM;
	240	} else {
fb045adb	241	d_set_d_op(child, &proc_sys_dentry_operations);
9043476f	242	d_add(child, inode);
77b14db5	243	}
9043476f AV	244	} else {
9043476f AV	245	return -ENOMEM;
77b14db5 EB	246	}
77b14db5 EB	247	}
77b14db5	248	inode = child->d_inode;
9043476f AV	249	ino = inode->i_ino;
9043476f AV	250	type = inode->i_mode >> 12;
77b14db5	251	dput(child);
9043476f AV	252	return !!filldir(dirent, qname.name, qname.len, filp->f_pos, ino, type);
	253	}
	254
	255	static int scan(struct ctl_table_header head, ctl_table table,
	256	unsigned long pos, struct file file,
	257	void *dirent, filldir_t filldir)
	258	{
	259
2315ffa0	260	for (; table->procname; table++, (*pos)++) {
9043476f AV	261	int res;
9043476f AV	262
9043476f AV	263	if (*pos < file->f_pos)
	264	continue;
	265
	266	res = proc_sys_fill_cache(file, dirent, filldir, head, table);
	267	if (res)
	268	return res;
	269
	270	file->f_pos = *pos + 1;
	271	}
	272	return 0;
77b14db5 EB	273	}
	274
	275	static int proc_sys_readdir(struct file filp, void dirent, filldir_t filldir)
	276	{
9043476f	277	struct dentry *dentry = filp->f_path.dentry;
77b14db5	278	struct inode *inode = dentry->d_inode;
9043476f AV	279	struct ctl_table_header *head = grab_header(inode);
	280	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
	281	struct ctl_table_header *h = NULL;
77b14db5	282	unsigned long pos;
9043476f AV	283	int ret = -EINVAL;
	284
	285	if (IS_ERR(head))
	286	return PTR_ERR(head);
77b14db5	287
9043476f AV	288	if (table && !table->child) {
9043476f AV	289	WARN_ON(1);
77b14db5	290	goto out;
9043476f AV	291	}
	292
	293	table = table ? table->child : head->ctl_table;
77b14db5 EB	294
	295	ret = 0;
	296	/* Avoid a switch here: arm builds fail with missing __cmpdi2 */
	297	if (filp->f_pos == 0) {
	298	if (filldir(dirent, ".", 1, filp->f_pos,
	299	inode->i_ino, DT_DIR) < 0)
	300	goto out;
	301	filp->f_pos++;
	302	}
	303	if (filp->f_pos == 1) {
	304	if (filldir(dirent, "..", 2, filp->f_pos,
	305	parent_ino(dentry), DT_DIR) < 0)
	306	goto out;
	307	filp->f_pos++;
	308	}
	309	pos = 2;
	310
9043476f AV	311	ret = scan(head, table, &pos, filp, dirent, filldir);
	312	if (ret)
	313	goto out;
77b14db5	314
9043476f AV	315	for (h = sysctl_head_next(NULL); h; h = sysctl_head_next(h)) {
9043476f AV	316	if (h->attached_to != table)
77b14db5	317	continue;
9043476f AV	318	ret = scan(h, h->attached_by, &pos, filp, dirent, filldir);
	319	if (ret) {
	320	sysctl_head_finish(h);
	321	break;
77b14db5 EB	322	}
	323	}
	324	ret = 1;
	325	out:
	326	sysctl_head_finish(head);
	327	return ret;
	328	}
	329
10556cb2	330	static int proc_sys_permission(struct inode *inode, int mask)
77b14db5 EB	331	{
	332	/*
	333	* sysctl entries that are not writeable,
	334	* are _NOT_ writeable, capabilities or not.
	335	*/
f696a365 MS	336	struct ctl_table_header *head;
f696a365 MS	337	struct ctl_table *table;
77b14db5 EB	338	int error;
77b14db5 EB	339
f696a365 MS	340	/* Executable files are not allowed under /proc/sys/ */
	341	if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode))
	342	return -EACCES;
	343
	344	head = grab_header(inode);
9043476f AV	345	if (IS_ERR(head))
9043476f AV	346	return PTR_ERR(head);
77b14db5	347
f696a365	348	table = PROC_I(inode)->sysctl_entry;
9043476f AV	349	if (!table) /* global root - r-xr-xr-x */
	350	error = mask & MAY_WRITE ? -EACCES : 0;
	351	else /* Use the permissions on the sysctl table entry */
1fc0f78c	352	error = sysctl_perm(head->root, table, mask & ~MAY_NOT_BLOCK);
77b14db5	353
77b14db5 EB	354	sysctl_head_finish(head);
	355	return error;
	356	}
	357
	358	static int proc_sys_setattr(struct dentry dentry, struct iattr attr)
	359	{
	360	struct inode *inode = dentry->d_inode;
	361	int error;
	362
	363	if (attr->ia_valid & (ATTR_MODE \| ATTR_UID \| ATTR_GID))
	364	return -EPERM;
	365
	366	error = inode_change_ok(inode, attr);
1025774c CH	367	if (error)
	368	return error;
	369
	370	if ((attr->ia_valid & ATTR_SIZE) &&
	371	attr->ia_size != i_size_read(inode)) {
	372	error = vmtruncate(inode, attr->ia_size);
	373	if (error)
	374	return error;
	375	}
77b14db5	376
1025774c CH	377	setattr_copy(inode, attr);
	378	mark_inode_dirty(inode);
	379	return 0;
77b14db5 EB	380	}
77b14db5 EB	381
9043476f AV	382	static int proc_sys_getattr(struct vfsmount mnt, struct dentry dentry, struct kstat *stat)
	383	{
	384	struct inode *inode = dentry->d_inode;
	385	struct ctl_table_header *head = grab_header(inode);
	386	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
	387
	388	if (IS_ERR(head))
	389	return PTR_ERR(head);
	390
	391	generic_fillattr(inode, stat);
	392	if (table)
	393	stat->mode = (stat->mode & S_IFMT) \| table->mode;
	394
	395	sysctl_head_finish(head);
	396	return 0;
	397	}
	398
77b14db5	399	static const struct file_operations proc_sys_file_operations = {
f1ecf068 LDM	400	.open = proc_sys_open,
f1ecf068 LDM	401	.poll = proc_sys_poll,
77b14db5 EB	402	.read = proc_sys_read,
77b14db5 EB	403	.write = proc_sys_write,
6038f373	404	.llseek = default_llseek,
9043476f AV	405	};
	406
	407	static const struct file_operations proc_sys_dir_file_operations = {
887df078	408	.read = generic_read_dir,
77b14db5	409	.readdir = proc_sys_readdir,
3222a3e5	410	.llseek = generic_file_llseek,
77b14db5 EB	411	};
77b14db5 EB	412
03a44825	413	static const struct inode_operations proc_sys_inode_operations = {
9043476f AV	414	.permission = proc_sys_permission,
	415	.setattr = proc_sys_setattr,
	416	.getattr = proc_sys_getattr,
	417	};
	418
	419	static const struct inode_operations proc_sys_dir_operations = {
77b14db5 EB	420	.lookup = proc_sys_lookup,
	421	.permission = proc_sys_permission,
	422	.setattr = proc_sys_setattr,
9043476f	423	.getattr = proc_sys_getattr,
77b14db5 EB	424	};
	425
	426	static int proc_sys_revalidate(struct dentry dentry, struct nameidata nd)
	427	{
34286d66 NP	428	if (nd->flags & LOOKUP_RCU)
34286d66 NP	429	return -ECHILD;
9043476f AV	430	return !PROC_I(dentry->d_inode)->sysctl->unregistering;
	431	}
	432
fe15ce44	433	static int proc_sys_delete(const struct dentry *dentry)
9043476f AV	434	{
	435	return !!PROC_I(dentry->d_inode)->sysctl->unregistering;
	436	}
	437
621e155a NP	438	static int proc_sys_compare(const struct dentry *parent,
	439	const struct inode *pinode,
	440	const struct dentry dentry, const struct inode inode,
	441	unsigned int len, const char str, const struct qstr name)
9043476f	442	{
dfef6dcd	443	struct ctl_table_header *head;
31e6b01f NP	444	/* Although proc doesn't have negative dentries, rcu-walk means
31e6b01f NP	445	* that inode here can be NULL */
dfef6dcd	446	/* AV: can it, indeed? */
31e6b01f	447	if (!inode)
dfef6dcd	448	return 1;
621e155a	449	if (name->len != len)
9043476f	450	return 1;
621e155a	451	if (memcmp(name->name, str, len))
9043476f	452	return 1;
dfef6dcd AV	453	head = rcu_dereference(PROC_I(inode)->sysctl);
dfef6dcd AV	454	return !head \|\| !sysctl_is_seen(head);
77b14db5 EB	455	}
77b14db5 EB	456
d72f71eb	457	static const struct dentry_operations proc_sys_dentry_operations = {
77b14db5	458	.d_revalidate = proc_sys_revalidate,
9043476f AV	459	.d_delete = proc_sys_delete,
9043476f AV	460	.d_compare = proc_sys_compare,
77b14db5 EB	461	};
77b14db5 EB	462
1e0edd3f	463	int __init proc_sys_init(void)
77b14db5	464	{
e1675231 AD	465	struct proc_dir_entry *proc_sys_root;
e1675231 AD	466
77b14db5	467	proc_sys_root = proc_mkdir("sys", NULL);
9043476f AV	468	proc_sys_root->proc_iops = &proc_sys_dir_operations;
9043476f AV	469	proc_sys_root->proc_fops = &proc_sys_dir_file_operations;
77b14db5 EB	470	proc_sys_root->nlink = 0;
	471	return 0;
	472	}