[deliverable/linux.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)
{
	int len;
	for ( ; p->procname; p++) {

		if (!p->procname)
			continue;

		len = strlen(p->procname);
		if (len != name->len)
			continue;

		if (memcmp(p->procname, 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;

		/* Can't do anything without a proc name */
		if (!table->procname)
			continue;

		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 EB	61	{
77b14db5 EB	62	int len;
2315ffa0	63	for ( ; p->procname; p++) {
77b14db5	64
9043476f	65	if (!p->procname)
77b14db5 EB	66	continue;
77b14db5 EB	67
9043476f	68	len = strlen(p->procname);
77b14db5 EB	69	if (len != name->len)
	70	continue;
	71
9043476f	72	if (memcmp(p->procname, name->name, len) != 0)
77b14db5 EB	73	continue;
	74
	75	/* I have a match */
9043476f	76	return p;
77b14db5 EB	77	}
	78	return NULL;
	79	}
	80
81324364	81	static struct ctl_table_header grab_header(struct inode inode)
77b14db5	82	{
9043476f AV	83	if (PROC_I(inode)->sysctl)
	84	return sysctl_head_grab(PROC_I(inode)->sysctl);
	85	else
	86	return sysctl_head_next(NULL);
	87	}
77b14db5	88
9043476f AV	89	static struct dentry proc_sys_lookup(struct inode dir, struct dentry *dentry,
	90	struct nameidata *nd)
	91	{
	92	struct ctl_table_header *head = grab_header(dir);
	93	struct ctl_table *table = PROC_I(dir)->sysctl_entry;
	94	struct ctl_table_header *h = NULL;
	95	struct qstr *name = &dentry->d_name;
	96	struct ctl_table *p;
	97	struct inode *inode;
	98	struct dentry *err = ERR_PTR(-ENOENT);
77b14db5	99
9043476f AV	100	if (IS_ERR(head))
9043476f AV	101	return ERR_CAST(head);
77b14db5	102
9043476f AV	103	if (table && !table->child) {
	104	WARN_ON(1);
	105	goto out;
77b14db5	106	}
77b14db5	107
9043476f	108	table = table ? table->child : head->ctl_table;
77b14db5	109
9043476f AV	110	p = find_in_table(table, name);
	111	if (!p) {
	112	for (h = sysctl_head_next(NULL); h; h = sysctl_head_next(h)) {
	113	if (h->attached_to != table)
	114	continue;
	115	p = find_in_table(h->attached_by, name);
	116	if (p)
	117	break;
	118	}
77b14db5	119	}
77b14db5	120
9043476f	121	if (!p)
77b14db5 EB	122	goto out;
	123
	124	err = ERR_PTR(-ENOMEM);
9043476f AV	125	inode = proc_sys_make_inode(dir->i_sb, h ? h : head, p);
	126	if (h)
	127	sysctl_head_finish(h);
	128
77b14db5 EB	129	if (!inode)
	130	goto out;
	131
	132	err = NULL;
fb045adb	133	d_set_d_op(dentry, &proc_sys_dentry_operations);
77b14db5 EB	134	d_add(dentry, inode);
	135
	136	out:
	137	sysctl_head_finish(head);
	138	return err;
	139	}
	140
7708bfb1 PE	141	static ssize_t proc_sys_call_handler(struct file filp, void __user buf,
7708bfb1 PE	142	size_t count, loff_t *ppos, int write)
77b14db5	143	{
9043476f AV	144	struct inode *inode = filp->f_path.dentry->d_inode;
	145	struct ctl_table_header *head = grab_header(inode);
	146	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
2a2da53b DH	147	ssize_t error;
2a2da53b DH	148	size_t res;
77b14db5	149
9043476f AV	150	if (IS_ERR(head))
9043476f AV	151	return PTR_ERR(head);
77b14db5 EB	152
	153	/*
	154	* At this point we know that the sysctl was not unregistered
	155	* and won't be until we finish.
	156	*/
	157	error = -EPERM;
d7321cd6	158	if (sysctl_perm(head->root, table, write ? MAY_WRITE : MAY_READ))
77b14db5 EB	159	goto out;
77b14db5 EB	160
9043476f AV	161	/* if that can happen at all, it should be -EINVAL, not -EISDIR */
	162	error = -EINVAL;
	163	if (!table->proc_handler)
	164	goto out;
	165
77b14db5 EB	166	/* careful: calling conventions are nasty here */
77b14db5 EB	167	res = count;
8d65af78	168	error = table->proc_handler(table, write, buf, &res, ppos);
77b14db5 EB	169	if (!error)
	170	error = res;
	171	out:
	172	sysctl_head_finish(head);
	173
	174	return error;
	175	}
	176
7708bfb1	177	static ssize_t proc_sys_read(struct file filp, char __user buf,
77b14db5 EB	178	size_t count, loff_t *ppos)
77b14db5 EB	179	{
7708bfb1 PE	180	return proc_sys_call_handler(filp, (void __user *)buf, count, ppos, 0);
7708bfb1 PE	181	}
77b14db5	182
7708bfb1 PE	183	static ssize_t proc_sys_write(struct file filp, const char __user buf,
	184	size_t count, loff_t *ppos)
	185	{
	186	return proc_sys_call_handler(filp, (void __user *)buf, count, ppos, 1);
77b14db5 EB	187	}
77b14db5 EB	188
f1ecf068 LDM	189	static int proc_sys_open(struct inode inode, struct file filp)
	190	{
	191	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
	192
	193	if (table->poll)
	194	filp->private_data = proc_sys_poll_event(table->poll);
	195
	196	return 0;
	197	}
	198
	199	static unsigned int proc_sys_poll(struct file filp, poll_table wait)
	200	{
	201	struct inode *inode = filp->f_path.dentry->d_inode;
	202	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
	203	unsigned long event = (unsigned long)filp->private_data;
	204	unsigned int ret = DEFAULT_POLLMASK;
	205
	206	if (!table->proc_handler)
	207	goto out;
	208
	209	if (!table->poll)
	210	goto out;
	211
	212	poll_wait(filp, &table->poll->wait, wait);
	213
	214	if (event != atomic_read(&table->poll->event)) {
	215	filp->private_data = proc_sys_poll_event(table->poll);
	216	ret = POLLIN \| POLLRDNORM \| POLLERR \| POLLPRI;
	217	}
	218
	219	out:
	220	return ret;
	221	}
77b14db5 EB	222
77b14db5 EB	223	static int proc_sys_fill_cache(struct file filp, void dirent,
9043476f AV	224	filldir_t filldir,
	225	struct ctl_table_header *head,
	226	struct ctl_table *table)
77b14db5	227	{
77b14db5 EB	228	struct dentry child, dir = filp->f_path.dentry;
	229	struct inode *inode;
	230	struct qstr qname;
	231	ino_t ino = 0;
	232	unsigned type = DT_UNKNOWN;
77b14db5 EB	233
	234	qname.name = table->procname;
	235	qname.len = strlen(table->procname);
	236	qname.hash = full_name_hash(qname.name, qname.len);
	237
77b14db5 EB	238	child = d_lookup(dir, &qname);
77b14db5 EB	239	if (!child) {
9043476f AV	240	child = d_alloc(dir, &qname);
	241	if (child) {
	242	inode = proc_sys_make_inode(dir->d_sb, head, table);
	243	if (!inode) {
	244	dput(child);
	245	return -ENOMEM;
	246	} else {
fb045adb	247	d_set_d_op(child, &proc_sys_dentry_operations);
9043476f	248	d_add(child, inode);
77b14db5	249	}
9043476f AV	250	} else {
9043476f AV	251	return -ENOMEM;
77b14db5 EB	252	}
77b14db5 EB	253	}
77b14db5	254	inode = child->d_inode;
9043476f AV	255	ino = inode->i_ino;
9043476f AV	256	type = inode->i_mode >> 12;
77b14db5	257	dput(child);
9043476f AV	258	return !!filldir(dirent, qname.name, qname.len, filp->f_pos, ino, type);
	259	}
	260
	261	static int scan(struct ctl_table_header head, ctl_table table,
	262	unsigned long pos, struct file file,
	263	void *dirent, filldir_t filldir)
	264	{
	265
2315ffa0	266	for (; table->procname; table++, (*pos)++) {
9043476f AV	267	int res;
	268
	269	/* Can't do anything without a proc name */
	270	if (!table->procname)
	271	continue;
	272
	273	if (*pos < file->f_pos)
	274	continue;
	275
	276	res = proc_sys_fill_cache(file, dirent, filldir, head, table);
	277	if (res)
	278	return res;
	279
	280	file->f_pos = *pos + 1;
	281	}
	282	return 0;
77b14db5 EB	283	}
	284
	285	static int proc_sys_readdir(struct file filp, void dirent, filldir_t filldir)
	286	{
9043476f	287	struct dentry *dentry = filp->f_path.dentry;
77b14db5	288	struct inode *inode = dentry->d_inode;
9043476f AV	289	struct ctl_table_header *head = grab_header(inode);
	290	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
	291	struct ctl_table_header *h = NULL;
77b14db5	292	unsigned long pos;
9043476f AV	293	int ret = -EINVAL;
	294
	295	if (IS_ERR(head))
	296	return PTR_ERR(head);
77b14db5	297
9043476f AV	298	if (table && !table->child) {
9043476f AV	299	WARN_ON(1);
77b14db5	300	goto out;
9043476f AV	301	}
	302
	303	table = table ? table->child : head->ctl_table;
77b14db5 EB	304
	305	ret = 0;
	306	/* Avoid a switch here: arm builds fail with missing __cmpdi2 */
	307	if (filp->f_pos == 0) {
	308	if (filldir(dirent, ".", 1, filp->f_pos,
	309	inode->i_ino, DT_DIR) < 0)
	310	goto out;
	311	filp->f_pos++;
	312	}
	313	if (filp->f_pos == 1) {
	314	if (filldir(dirent, "..", 2, filp->f_pos,
	315	parent_ino(dentry), DT_DIR) < 0)
	316	goto out;
	317	filp->f_pos++;
	318	}
	319	pos = 2;
	320
9043476f AV	321	ret = scan(head, table, &pos, filp, dirent, filldir);
	322	if (ret)
	323	goto out;
77b14db5	324
9043476f AV	325	for (h = sysctl_head_next(NULL); h; h = sysctl_head_next(h)) {
9043476f AV	326	if (h->attached_to != table)
77b14db5	327	continue;
9043476f AV	328	ret = scan(h, h->attached_by, &pos, filp, dirent, filldir);
	329	if (ret) {
	330	sysctl_head_finish(h);
	331	break;
77b14db5 EB	332	}
	333	}
	334	ret = 1;
	335	out:
	336	sysctl_head_finish(head);
	337	return ret;
	338	}
	339
10556cb2	340	static int proc_sys_permission(struct inode *inode, int mask)
77b14db5 EB	341	{
	342	/*
	343	* sysctl entries that are not writeable,
	344	* are _NOT_ writeable, capabilities or not.
	345	*/
f696a365 MS	346	struct ctl_table_header *head;
f696a365 MS	347	struct ctl_table *table;
77b14db5 EB	348	int error;
77b14db5 EB	349
f696a365 MS	350	/* Executable files are not allowed under /proc/sys/ */
	351	if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode))
	352	return -EACCES;
	353
	354	head = grab_header(inode);
9043476f AV	355	if (IS_ERR(head))
9043476f AV	356	return PTR_ERR(head);
77b14db5	357
f696a365	358	table = PROC_I(inode)->sysctl_entry;
9043476f AV	359	if (!table) /* global root - r-xr-xr-x */
	360	error = mask & MAY_WRITE ? -EACCES : 0;
	361	else /* Use the permissions on the sysctl table entry */
1fc0f78c	362	error = sysctl_perm(head->root, table, mask & ~MAY_NOT_BLOCK);
77b14db5	363
77b14db5 EB	364	sysctl_head_finish(head);
	365	return error;
	366	}
	367
	368	static int proc_sys_setattr(struct dentry dentry, struct iattr attr)
	369	{
	370	struct inode *inode = dentry->d_inode;
	371	int error;
	372
	373	if (attr->ia_valid & (ATTR_MODE \| ATTR_UID \| ATTR_GID))
	374	return -EPERM;
	375
	376	error = inode_change_ok(inode, attr);
1025774c CH	377	if (error)
	378	return error;
	379
	380	if ((attr->ia_valid & ATTR_SIZE) &&
	381	attr->ia_size != i_size_read(inode)) {
	382	error = vmtruncate(inode, attr->ia_size);
	383	if (error)
	384	return error;
	385	}
77b14db5	386
1025774c CH	387	setattr_copy(inode, attr);
	388	mark_inode_dirty(inode);
	389	return 0;
77b14db5 EB	390	}
77b14db5 EB	391
9043476f AV	392	static int proc_sys_getattr(struct vfsmount mnt, struct dentry dentry, struct kstat *stat)
	393	{
	394	struct inode *inode = dentry->d_inode;
	395	struct ctl_table_header *head = grab_header(inode);
	396	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
	397
	398	if (IS_ERR(head))
	399	return PTR_ERR(head);
	400
	401	generic_fillattr(inode, stat);
	402	if (table)
	403	stat->mode = (stat->mode & S_IFMT) \| table->mode;
	404
	405	sysctl_head_finish(head);
	406	return 0;
	407	}
	408
77b14db5	409	static const struct file_operations proc_sys_file_operations = {
f1ecf068 LDM	410	.open = proc_sys_open,
f1ecf068 LDM	411	.poll = proc_sys_poll,
77b14db5 EB	412	.read = proc_sys_read,
77b14db5 EB	413	.write = proc_sys_write,
6038f373	414	.llseek = default_llseek,
9043476f AV	415	};
	416
	417	static const struct file_operations proc_sys_dir_file_operations = {
887df078	418	.read = generic_read_dir,
77b14db5	419	.readdir = proc_sys_readdir,
3222a3e5	420	.llseek = generic_file_llseek,
77b14db5 EB	421	};
77b14db5 EB	422
03a44825	423	static const struct inode_operations proc_sys_inode_operations = {
9043476f AV	424	.permission = proc_sys_permission,
	425	.setattr = proc_sys_setattr,
	426	.getattr = proc_sys_getattr,
	427	};
	428
	429	static const struct inode_operations proc_sys_dir_operations = {
77b14db5 EB	430	.lookup = proc_sys_lookup,
	431	.permission = proc_sys_permission,
	432	.setattr = proc_sys_setattr,
9043476f	433	.getattr = proc_sys_getattr,
77b14db5 EB	434	};
	435
	436	static int proc_sys_revalidate(struct dentry dentry, struct nameidata nd)
	437	{
34286d66 NP	438	if (nd->flags & LOOKUP_RCU)
34286d66 NP	439	return -ECHILD;
9043476f AV	440	return !PROC_I(dentry->d_inode)->sysctl->unregistering;
	441	}
	442
fe15ce44	443	static int proc_sys_delete(const struct dentry *dentry)
9043476f AV	444	{
	445	return !!PROC_I(dentry->d_inode)->sysctl->unregistering;
	446	}
	447
621e155a NP	448	static int proc_sys_compare(const struct dentry *parent,
	449	const struct inode *pinode,
	450	const struct dentry dentry, const struct inode inode,
	451	unsigned int len, const char str, const struct qstr name)
9043476f	452	{
dfef6dcd	453	struct ctl_table_header *head;
31e6b01f NP	454	/* Although proc doesn't have negative dentries, rcu-walk means
31e6b01f NP	455	* that inode here can be NULL */
dfef6dcd	456	/* AV: can it, indeed? */
31e6b01f	457	if (!inode)
dfef6dcd	458	return 1;
621e155a	459	if (name->len != len)
9043476f	460	return 1;
621e155a	461	if (memcmp(name->name, str, len))
9043476f	462	return 1;
dfef6dcd AV	463	head = rcu_dereference(PROC_I(inode)->sysctl);
dfef6dcd AV	464	return !head \|\| !sysctl_is_seen(head);
77b14db5 EB	465	}
77b14db5 EB	466
d72f71eb	467	static const struct dentry_operations proc_sys_dentry_operations = {
77b14db5	468	.d_revalidate = proc_sys_revalidate,
9043476f AV	469	.d_delete = proc_sys_delete,
9043476f AV	470	.d_compare = proc_sys_compare,
77b14db5 EB	471	};
77b14db5 EB	472
1e0edd3f	473	int __init proc_sys_init(void)
77b14db5	474	{
e1675231 AD	475	struct proc_dir_entry *proc_sys_root;
e1675231 AD	476
77b14db5	477	proc_sys_root = proc_mkdir("sys", NULL);
9043476f AV	478	proc_sys_root->proc_iops = &proc_sys_dir_operations;
9043476f AV	479	proc_sys_root->proc_fops = &proc_sys_dir_file_operations;
77b14db5 EB	480	proc_sys_root->nlink = 0;
	481	return 0;
	482	}