[deliverable/linux.git] / fs / proc / inode.c

/*
 *  linux/fs/proc/inode.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 */

#include <linux/time.h>
#include <linux/proc_fs.h>
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/string.h>
#include <linux/stat.h>
#include <linux/completion.h>
#include <linux/poll.h>
#include <linux/file.h>
#include <linux/limits.h>
#include <linux/init.h>
#include <linux/module.h>
#include <linux/smp_lock.h>

#include <asm/system.h>
#include <asm/uaccess.h>

#include "internal.h"

struct proc_dir_entry *de_get(struct proc_dir_entry *de)
{
	atomic_inc(&de->count);
	return de;
}

/*
 * Decrements the use count and checks for deferred deletion.
 */
void de_put(struct proc_dir_entry *de)
{
	lock_kernel();
	if (!atomic_read(&de->count)) {
		printk("de_put: entry %s already free!\n", de->name);
		unlock_kernel();
		return;
	}

	if (atomic_dec_and_test(&de->count))
		free_proc_entry(de);
	unlock_kernel();
}

/*
 * Decrement the use count of the proc_dir_entry.
 */
static void proc_delete_inode(struct inode *inode)
{
	struct proc_dir_entry *de;

	truncate_inode_pages(&inode->i_data, 0);

	/* Stop tracking associated processes */
	put_pid(PROC_I(inode)->pid);

	/* Let go of any associated proc directory entry */
	de = PROC_I(inode)->pde;
	if (de) {
		if (de->owner)
			module_put(de->owner);
		de_put(de);
	}
	clear_inode(inode);
}

struct vfsmount *proc_mnt;

static struct kmem_cache * proc_inode_cachep;

static struct inode *proc_alloc_inode(struct super_block *sb)
{
	struct proc_inode *ei;
	struct inode *inode;

	ei = (struct proc_inode *)kmem_cache_alloc(proc_inode_cachep, GFP_KERNEL);
	if (!ei)
		return NULL;
	ei->pid = NULL;
	ei->fd = 0;
	ei->op.proc_get_link = NULL;
	ei->pde = NULL;
	inode = &ei->vfs_inode;
	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
	return inode;
}

static void proc_destroy_inode(struct inode *inode)
{
	kmem_cache_free(proc_inode_cachep, PROC_I(inode));
}

static void init_once(struct kmem_cache * cachep, void *foo)
{
	struct proc_inode *ei = (struct proc_inode *) foo;

	inode_init_once(&ei->vfs_inode);
}

int __init proc_init_inodecache(void)
{
	proc_inode_cachep = kmem_cache_create("proc_inode_cache",
					     sizeof(struct proc_inode),
					     0, (SLAB_RECLAIM_ACCOUNT|
						SLAB_MEM_SPREAD|SLAB_PANIC),
					     init_once);
	return 0;
}

static const struct super_operations proc_sops = {
	.alloc_inode	= proc_alloc_inode,
	.destroy_inode	= proc_destroy_inode,
	.drop_inode	= generic_delete_inode,
	.delete_inode	= proc_delete_inode,
	.statfs		= simple_statfs,
};

static void __pde_users_dec(struct proc_dir_entry *pde)
{
	pde->pde_users--;
	if (pde->pde_unload_completion && pde->pde_users == 0)
		complete(pde->pde_unload_completion);
}

static void pde_users_dec(struct proc_dir_entry *pde)
{
	spin_lock(&pde->pde_unload_lock);
	__pde_users_dec(pde);
	spin_unlock(&pde->pde_unload_lock);
}

static loff_t proc_reg_llseek(struct file *file, loff_t offset, int whence)
{
	struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
	loff_t rv = -EINVAL;
	loff_t (*llseek)(struct file *, loff_t, int);

	spin_lock(&pde->pde_unload_lock);
	/*
	 * remove_proc_entry() is going to delete PDE (as part of module
	 * cleanup sequence). No new callers into module allowed.
	 */
	if (!pde->proc_fops) {
		spin_unlock(&pde->pde_unload_lock);
		return rv;
	}
	/*
	 * Bump refcount so that remove_proc_entry will wail for ->llseek to
	 * complete.
	 */
	pde->pde_users++;
	/*
	 * Save function pointer under lock, to protect against ->proc_fops
	 * NULL'ifying right after ->pde_unload_lock is dropped.
	 */
	llseek = pde->proc_fops->llseek;
	spin_unlock(&pde->pde_unload_lock);

	if (!llseek)
		llseek = default_llseek;
	rv = llseek(file, offset, whence);

	pde_users_dec(pde);
	return rv;
}

static ssize_t proc_reg_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
{
	struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
	ssize_t rv = -EIO;
	ssize_t (*read)(struct file *, char __user *, size_t, loff_t *);

	spin_lock(&pde->pde_unload_lock);
	if (!pde->proc_fops) {
		spin_unlock(&pde->pde_unload_lock);
		return rv;
	}
	pde->pde_users++;
	read = pde->proc_fops->read;
	spin_unlock(&pde->pde_unload_lock);

	if (read)
		rv = read(file, buf, count, ppos);

	pde_users_dec(pde);
	return rv;
}

static ssize_t proc_reg_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
{
	struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
	ssize_t rv = -EIO;
	ssize_t (*write)(struct file *, const char __user *, size_t, loff_t *);

	spin_lock(&pde->pde_unload_lock);
	if (!pde->proc_fops) {
		spin_unlock(&pde->pde_unload_lock);
		return rv;
	}
	pde->pde_users++;
	write = pde->proc_fops->write;
	spin_unlock(&pde->pde_unload_lock);

	if (write)
		rv = write(file, buf, count, ppos);

	pde_users_dec(pde);
	return rv;
}

static unsigned int proc_reg_poll(struct file *file, struct poll_table_struct *pts)
{
	struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
	unsigned int rv = DEFAULT_POLLMASK;
	unsigned int (*poll)(struct file *, struct poll_table_struct *);

	spin_lock(&pde->pde_unload_lock);
	if (!pde->proc_fops) {
		spin_unlock(&pde->pde_unload_lock);
		return rv;
	}
	pde->pde_users++;
	poll = pde->proc_fops->poll;
	spin_unlock(&pde->pde_unload_lock);

	if (poll)
		rv = poll(file, pts);

	pde_users_dec(pde);
	return rv;
}

static long proc_reg_unlocked_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
	struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
	long rv = -ENOTTY;
	long (*unlocked_ioctl)(struct file *, unsigned int, unsigned long);
	int (*ioctl)(struct inode *, struct file *, unsigned int, unsigned long);

	spin_lock(&pde->pde_unload_lock);
	if (!pde->proc_fops) {
		spin_unlock(&pde->pde_unload_lock);
		return rv;
	}
	pde->pde_users++;
	unlocked_ioctl = pde->proc_fops->unlocked_ioctl;
	ioctl = pde->proc_fops->ioctl;
	spin_unlock(&pde->pde_unload_lock);

	if (unlocked_ioctl) {
		rv = unlocked_ioctl(file, cmd, arg);
		if (rv == -ENOIOCTLCMD)
			rv = -EINVAL;
	} else if (ioctl) {
		lock_kernel();
		rv = ioctl(file->f_path.dentry->d_inode, file, cmd, arg);
		unlock_kernel();
	}

	pde_users_dec(pde);
	return rv;
}

#ifdef CONFIG_COMPAT
static long proc_reg_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
	struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
	long rv = -ENOTTY;
	long (*compat_ioctl)(struct file *, unsigned int, unsigned long);

	spin_lock(&pde->pde_unload_lock);
	if (!pde->proc_fops) {
		spin_unlock(&pde->pde_unload_lock);
		return rv;
	}
	pde->pde_users++;
	compat_ioctl = pde->proc_fops->compat_ioctl;
	spin_unlock(&pde->pde_unload_lock);

	if (compat_ioctl)
		rv = compat_ioctl(file, cmd, arg);

	pde_users_dec(pde);
	return rv;
}
#endif

static int proc_reg_mmap(struct file *file, struct vm_area_struct *vma)
{
	struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
	int rv = -EIO;
	int (*mmap)(struct file *, struct vm_area_struct *);

	spin_lock(&pde->pde_unload_lock);
	if (!pde->proc_fops) {
		spin_unlock(&pde->pde_unload_lock);
		return rv;
	}
	pde->pde_users++;
	mmap = pde->proc_fops->mmap;
	spin_unlock(&pde->pde_unload_lock);

	if (mmap)
		rv = mmap(file, vma);

	pde_users_dec(pde);
	return rv;
}

static int proc_reg_open(struct inode *inode, struct file *file)
{
	struct proc_dir_entry *pde = PDE(inode);
	int rv = 0;
	int (*open)(struct inode *, struct file *);
	int (*release)(struct inode *, struct file *);
	struct pde_opener *pdeo;

	/*
	 * What for, you ask? Well, we can have open, rmmod, remove_proc_entry
	 * sequence. ->release won't be called because ->proc_fops will be
	 * cleared. Depending on complexity of ->release, consequences vary.
	 *
	 * We can't wait for mercy when close will be done for real, it's
	 * deadlockable: rmmod foo </proc/foo . So, we're going to do ->release
	 * by hand in remove_proc_entry(). For this, save opener's credentials
	 * for later.
	 */
	pdeo = kmalloc(sizeof(struct pde_opener), GFP_KERNEL);
	if (!pdeo)
		return -ENOMEM;

	spin_lock(&pde->pde_unload_lock);
	if (!pde->proc_fops) {
		spin_unlock(&pde->pde_unload_lock);
		kfree(pdeo);
		return rv;
	}
	pde->pde_users++;
	open = pde->proc_fops->open;
	release = pde->proc_fops->release;
	spin_unlock(&pde->pde_unload_lock);

	if (open)
		rv = open(inode, file);

	spin_lock(&pde->pde_unload_lock);
	if (rv == 0 && release) {
		/* To know what to release. */
		pdeo->inode = inode;
		pdeo->file = file;
		/* Strictly for "too late" ->release in proc_reg_release(). */
		pdeo->release = release;
		list_add(&pdeo->lh, &pde->pde_openers);
	} else
		kfree(pdeo);
	__pde_users_dec(pde);
	spin_unlock(&pde->pde_unload_lock);
	return rv;
}

static struct pde_opener *find_pde_opener(struct proc_dir_entry *pde,
					struct inode *inode, struct file *file)
{
	struct pde_opener *pdeo;

	list_for_each_entry(pdeo, &pde->pde_openers, lh) {
		if (pdeo->inode == inode && pdeo->file == file)
			return pdeo;
	}
	return NULL;
}

static int proc_reg_release(struct inode *inode, struct file *file)
{
	struct proc_dir_entry *pde = PDE(inode);
	int rv = 0;
	int (*release)(struct inode *, struct file *);
	struct pde_opener *pdeo;

	spin_lock(&pde->pde_unload_lock);
	pdeo = find_pde_opener(pde, inode, file);
	if (!pde->proc_fops) {
		/*
		 * Can't simply exit, __fput() will think that everything is OK,
		 * and move on to freeing struct file. remove_proc_entry() will
		 * find slacker in opener's list and will try to do non-trivial
		 * things with struct file. Therefore, remove opener from list.
		 *
		 * But if opener is removed from list, who will ->release it?
		 */
		if (pdeo) {
			list_del(&pdeo->lh);
			spin_unlock(&pde->pde_unload_lock);
			rv = pdeo->release(inode, file);
			kfree(pdeo);
		} else
			spin_unlock(&pde->pde_unload_lock);
		return rv;
	}
	pde->pde_users++;
	release = pde->proc_fops->release;
	if (pdeo) {
		list_del(&pdeo->lh);
		kfree(pdeo);
	}
	spin_unlock(&pde->pde_unload_lock);

	if (release)
		rv = release(inode, file);

	pde_users_dec(pde);
	return rv;
}

static const struct file_operations proc_reg_file_ops = {
	.llseek		= proc_reg_llseek,
	.read		= proc_reg_read,
	.write		= proc_reg_write,
	.poll		= proc_reg_poll,
	.unlocked_ioctl	= proc_reg_unlocked_ioctl,
#ifdef CONFIG_COMPAT
	.compat_ioctl	= proc_reg_compat_ioctl,
#endif
	.mmap		= proc_reg_mmap,
	.open		= proc_reg_open,
	.release	= proc_reg_release,
};

#ifdef CONFIG_COMPAT
static const struct file_operations proc_reg_file_ops_no_compat = {
	.llseek		= proc_reg_llseek,
	.read		= proc_reg_read,
	.write		= proc_reg_write,
	.poll		= proc_reg_poll,
	.unlocked_ioctl	= proc_reg_unlocked_ioctl,
	.mmap		= proc_reg_mmap,
	.open		= proc_reg_open,
	.release	= proc_reg_release,
};
#endif

struct inode *proc_get_inode(struct super_block *sb, unsigned int ino,
				struct proc_dir_entry *de)
{
	struct inode * inode;

	if (!try_module_get(de->owner))
		goto out_mod;

	inode = iget_locked(sb, ino);
	if (!inode)
		goto out_ino;
	if (inode->i_state & I_NEW) {
		inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
		PROC_I(inode)->fd = 0;
		PROC_I(inode)->pde = de;

		if (de->mode) {
			inode->i_mode = de->mode;
			inode->i_uid = de->uid;
			inode->i_gid = de->gid;
		}
		if (de->size)
			inode->i_size = de->size;
		if (de->nlink)
			inode->i_nlink = de->nlink;
		if (de->proc_iops)
			inode->i_op = de->proc_iops;
		if (de->proc_fops) {
			if (S_ISREG(inode->i_mode)) {
#ifdef CONFIG_COMPAT
				if (!de->proc_fops->compat_ioctl)
					inode->i_fop =
						&proc_reg_file_ops_no_compat;
				else
#endif
					inode->i_fop = &proc_reg_file_ops;
			} else {
				inode->i_fop = de->proc_fops;
			}
		}
		unlock_new_inode(inode);
	} else
	       module_put(de->owner);
	return inode;

out_ino:
	module_put(de->owner);
out_mod:
	return NULL;
}			

int proc_fill_super(struct super_block *s)
{
	struct inode * root_inode;

	s->s_flags |= MS_NODIRATIME | MS_NOSUID | MS_NOEXEC;
	s->s_blocksize = 1024;
	s->s_blocksize_bits = 10;
	s->s_magic = PROC_SUPER_MAGIC;
	s->s_op = &proc_sops;
	s->s_time_gran = 1;
	
	de_get(&proc_root);
	root_inode = proc_get_inode(s, PROC_ROOT_INO, &proc_root);
	if (!root_inode)
		goto out_no_root;
	root_inode->i_uid = 0;
	root_inode->i_gid = 0;
	s->s_root = d_alloc_root(root_inode);
	if (!s->s_root)
		goto out_no_root;
	return 0;

out_no_root:
	printk("proc_read_super: get root inode failed\n");
	iput(root_inode);
	de_put(&proc_root);
	return -ENOMEM;
}
Commit	Line	Data
1da177e4 LT	1	/*
	2	* linux/fs/proc/inode.c
	3	*
	4	* Copyright (C) 1991, 1992 Linus Torvalds
	5	*/
	6
	7	#include <linux/time.h>
	8	#include <linux/proc_fs.h>
	9	#include <linux/kernel.h>
	10	#include <linux/mm.h>
	11	#include <linux/string.h>
	12	#include <linux/stat.h>
786d7e16	13	#include <linux/completion.h>
dd23aae4	14	#include <linux/poll.h>
1da177e4 LT	15	#include <linux/file.h>
	16	#include <linux/limits.h>
	17	#include <linux/init.h>
	18	#include <linux/module.h>
	19	#include <linux/smp_lock.h>
	20
	21	#include <asm/system.h>
	22	#include <asm/uaccess.h>
	23
fee781e6	24	#include "internal.h"
1da177e4	25
7695650a	26	struct proc_dir_entry de_get(struct proc_dir_entry de)
1da177e4	27	{
5e971dce	28	atomic_inc(&de->count);
1da177e4 LT	29	return de;
	30	}
	31
	32	/*
	33	* Decrements the use count and checks for deferred deletion.
	34	*/
7695650a	35	void de_put(struct proc_dir_entry *de)
1da177e4	36	{
5e971dce AD	37	lock_kernel();
	38	if (!atomic_read(&de->count)) {
	39	printk("de_put: entry %s already free!\n", de->name);
1da177e4	40	unlock_kernel();
5e971dce	41	return;
1da177e4	42	}
5e971dce AD	43
	44	if (atomic_dec_and_test(&de->count))
	45	free_proc_entry(de);
	46	unlock_kernel();
1da177e4 LT	47	}
	48
	49	/*
	50	* Decrement the use count of the proc_dir_entry.
	51	*/
	52	static void proc_delete_inode(struct inode *inode)
	53	{
	54	struct proc_dir_entry *de;
1da177e4	55
fef26658 MF	56	truncate_inode_pages(&inode->i_data, 0);
fef26658 MF	57
99f89551	58	/* Stop tracking associated processes */
13b41b09	59	put_pid(PROC_I(inode)->pid);
1da177e4 LT	60
	61	/* Let go of any associated proc directory entry */
	62	de = PROC_I(inode)->pde;
	63	if (de) {
	64	if (de->owner)
	65	module_put(de->owner);
	66	de_put(de);
	67	}
	68	clear_inode(inode);
	69	}
	70
	71	struct vfsmount *proc_mnt;
	72
e18b890b	73	static struct kmem_cache * proc_inode_cachep;
1da177e4 LT	74
	75	static struct inode proc_alloc_inode(struct super_block sb)
	76	{
	77	struct proc_inode *ei;
	78	struct inode *inode;
	79
e94b1766	80	ei = (struct proc_inode *)kmem_cache_alloc(proc_inode_cachep, GFP_KERNEL);
1da177e4 LT	81	if (!ei)
1da177e4 LT	82	return NULL;
13b41b09	83	ei->pid = NULL;
aed7a6c4	84	ei->fd = 0;
1da177e4 LT	85	ei->op.proc_get_link = NULL;
	86	ei->pde = NULL;
	87	inode = &ei->vfs_inode;
	88	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
	89	return inode;
	90	}
	91
	92	static void proc_destroy_inode(struct inode *inode)
	93	{
	94	kmem_cache_free(proc_inode_cachep, PROC_I(inode));
	95	}
	96
4ba9b9d0	97	static void init_once(struct kmem_cache * cachep, void *foo)
1da177e4 LT	98	{
	99	struct proc_inode ei = (struct proc_inode ) foo;
	100
a35afb83	101	inode_init_once(&ei->vfs_inode);
1da177e4	102	}
20c2df83	103
1da177e4 LT	104	int __init proc_init_inodecache(void)
	105	{
	106	proc_inode_cachep = kmem_cache_create("proc_inode_cache",
	107	sizeof(struct proc_inode),
fffb60f9	108	0, (SLAB_RECLAIM_ACCOUNT\|
040b5c6f	109	SLAB_MEM_SPREAD\|SLAB_PANIC),
20c2df83	110	init_once);
1da177e4 LT	111	return 0;
	112	}
	113
ee9b6d61	114	static const struct super_operations proc_sops = {
1da177e4 LT	115	.alloc_inode = proc_alloc_inode,
1da177e4 LT	116	.destroy_inode = proc_destroy_inode,
1da177e4 LT	117	.drop_inode = generic_delete_inode,
	118	.delete_inode = proc_delete_inode,
	119	.statfs = simple_statfs,
1da177e4 LT	120	};
1da177e4 LT	121
881adb85	122	static void __pde_users_dec(struct proc_dir_entry *pde)
786d7e16	123	{
786d7e16 AD	124	pde->pde_users--;
	125	if (pde->pde_unload_completion && pde->pde_users == 0)
	126	complete(pde->pde_unload_completion);
881adb85 AD	127	}
	128
	129	static void pde_users_dec(struct proc_dir_entry *pde)
	130	{
	131	spin_lock(&pde->pde_unload_lock);
	132	__pde_users_dec(pde);
786d7e16 AD	133	spin_unlock(&pde->pde_unload_lock);
	134	}
	135
	136	static loff_t proc_reg_llseek(struct file *file, loff_t offset, int whence)
	137	{
	138	struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
	139	loff_t rv = -EINVAL;
	140	loff_t (llseek)(struct file , loff_t, int);
	141
	142	spin_lock(&pde->pde_unload_lock);
	143	/*
	144	* remove_proc_entry() is going to delete PDE (as part of module
	145	* cleanup sequence). No new callers into module allowed.
	146	*/
	147	if (!pde->proc_fops) {
	148	spin_unlock(&pde->pde_unload_lock);
	149	return rv;
	150	}
	151	/*
	152	* Bump refcount so that remove_proc_entry will wail for ->llseek to
	153	* complete.
	154	*/
	155	pde->pde_users++;
	156	/*
	157	* Save function pointer under lock, to protect against ->proc_fops
	158	* NULL'ifying right after ->pde_unload_lock is dropped.
	159	*/
	160	llseek = pde->proc_fops->llseek;
	161	spin_unlock(&pde->pde_unload_lock);
	162
	163	if (!llseek)
	164	llseek = default_llseek;
	165	rv = llseek(file, offset, whence);
	166
	167	pde_users_dec(pde);
	168	return rv;
	169	}
	170
	171	static ssize_t proc_reg_read(struct file file, char __user buf, size_t count, loff_t *ppos)
	172	{
	173	struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
	174	ssize_t rv = -EIO;
	175	ssize_t (read)(struct file , char __user , size_t, loff_t );
	176
	177	spin_lock(&pde->pde_unload_lock);
	178	if (!pde->proc_fops) {
	179	spin_unlock(&pde->pde_unload_lock);
	180	return rv;
	181	}
	182	pde->pde_users++;
	183	read = pde->proc_fops->read;
	184	spin_unlock(&pde->pde_unload_lock);
	185
	186	if (read)
	187	rv = read(file, buf, count, ppos);
	188
	189	pde_users_dec(pde);
	190	return rv;
	191	}
	192
	193	static ssize_t proc_reg_write(struct file file, const char __user buf, size_t count, loff_t *ppos)
	194	{
	195	struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
	196	ssize_t rv = -EIO;
197	ssize_t (write)(struct file , const char __user , size_t, loff_t );
198
199	spin_lock(&pde->pde_unload_lock);
200	if (!pde->proc_fops) {
201	spin_unlock(&pde->pde_unload_lock);
202	return rv;
203	}
204	pde->pde_users++;
205	write = pde->proc_fops->write;
206	spin_unlock(&pde->pde_unload_lock);
207
208	if (write)
209	rv = write(file, buf, count, ppos);
210
211	pde_users_dec(pde);
212	return rv;
213	}
214
215	static unsigned int proc_reg_poll(struct file file, struct poll_table_struct pts)
216	{
217	struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
dd23aae4	218	unsigned int rv = DEFAULT_POLLMASK;
786d7e16 AD	219	unsigned int (poll)(struct file , struct poll_table_struct *);
	220
	221	spin_lock(&pde->pde_unload_lock);
	222	if (!pde->proc_fops) {
	223	spin_unlock(&pde->pde_unload_lock);
	224	return rv;
	225	}
	226	pde->pde_users++;
	227	poll = pde->proc_fops->poll;
	228	spin_unlock(&pde->pde_unload_lock);
	229
	230	if (poll)
	231	rv = poll(file, pts);
	232
	233	pde_users_dec(pde);
	234	return rv;
	235	}
	236
	237	static long proc_reg_unlocked_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
	238	{
	239	struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
	240	long rv = -ENOTTY;
	241	long (unlocked_ioctl)(struct file , unsigned int, unsigned long);
	242	int (ioctl)(struct inode , struct file *, unsigned int, unsigned long);
	243
	244	spin_lock(&pde->pde_unload_lock);
	245	if (!pde->proc_fops) {
	246	spin_unlock(&pde->pde_unload_lock);
	247	return rv;
	248	}
	249	pde->pde_users++;
	250	unlocked_ioctl = pde->proc_fops->unlocked_ioctl;
	251	ioctl = pde->proc_fops->ioctl;
	252	spin_unlock(&pde->pde_unload_lock);
	253
	254	if (unlocked_ioctl) {
	255	rv = unlocked_ioctl(file, cmd, arg);
	256	if (rv == -ENOIOCTLCMD)
	257	rv = -EINVAL;
	258	} else if (ioctl) {
	259	lock_kernel();
	260	rv = ioctl(file->f_path.dentry->d_inode, file, cmd, arg);
	261	unlock_kernel();
	262	}
	263
	264	pde_users_dec(pde);
	265	return rv;
	266	}
	267
	268	#ifdef CONFIG_COMPAT
	269	static long proc_reg_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
	270	{
	271	struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
	272	long rv = -ENOTTY;
	273	long (compat_ioctl)(struct file , unsigned int, unsigned long);
	274
	275	spin_lock(&pde->pde_unload_lock);
	276	if (!pde->proc_fops) {
	277	spin_unlock(&pde->pde_unload_lock);
	278	return rv;
	279	}
	280	pde->pde_users++;
	281	compat_ioctl = pde->proc_fops->compat_ioctl;
	282	spin_unlock(&pde->pde_unload_lock);
283
284	if (compat_ioctl)
285	rv = compat_ioctl(file, cmd, arg);
286
287	pde_users_dec(pde);
288	return rv;
289	}
290	#endif
291
292	static int proc_reg_mmap(struct file file, struct vm_area_struct vma)
293	{
294	struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
295	int rv = -EIO;
296	int (mmap)(struct file , struct vm_area_struct *);
297
298	spin_lock(&pde->pde_unload_lock);
299	if (!pde->proc_fops) {
300	spin_unlock(&pde->pde_unload_lock);
301	return rv;
302	}
303	pde->pde_users++;
304	mmap = pde->proc_fops->mmap;
305	spin_unlock(&pde->pde_unload_lock);
306
307	if (mmap)
308	rv = mmap(file, vma);
309
310	pde_users_dec(pde);
311	return rv;
312	}
313
314	static int proc_reg_open(struct inode inode, struct file file)
315	{
316	struct proc_dir_entry *pde = PDE(inode);
317	int rv = 0;
318	int (open)(struct inode , struct file *);
881adb85 AD	319	int (release)(struct inode , struct file *);
	320	struct pde_opener *pdeo;
	321
	322	/*
	323	* What for, you ask? Well, we can have open, rmmod, remove_proc_entry
	324	* sequence. ->release won't be called because ->proc_fops will be
	325	* cleared. Depending on complexity of ->release, consequences vary.
	326	*
	327	* We can't wait for mercy when close will be done for real, it's
	328	* deadlockable: rmmod foo </proc/foo . So, we're going to do ->release
	329	* by hand in remove_proc_entry(). For this, save opener's credentials
	330	* for later.
	331	*/
	332	pdeo = kmalloc(sizeof(struct pde_opener), GFP_KERNEL);
	333	if (!pdeo)
	334	return -ENOMEM;
786d7e16 AD	335
	336	spin_lock(&pde->pde_unload_lock);
	337	if (!pde->proc_fops) {
	338	spin_unlock(&pde->pde_unload_lock);
881adb85	339	kfree(pdeo);
786d7e16 AD	340	return rv;
	341	}
	342	pde->pde_users++;
	343	open = pde->proc_fops->open;
881adb85	344	release = pde->proc_fops->release;
786d7e16 AD	345	spin_unlock(&pde->pde_unload_lock);
	346
	347	if (open)
	348	rv = open(inode, file);
	349
881adb85 AD	350	spin_lock(&pde->pde_unload_lock);
	351	if (rv == 0 && release) {
	352	/* To know what to release. */
	353	pdeo->inode = inode;
	354	pdeo->file = file;
	355	/* Strictly for "too late" ->release in proc_reg_release(). */
	356	pdeo->release = release;
	357	list_add(&pdeo->lh, &pde->pde_openers);
	358	} else
	359	kfree(pdeo);
	360	__pde_users_dec(pde);
	361	spin_unlock(&pde->pde_unload_lock);
786d7e16 AD	362	return rv;
	363	}
	364
881adb85 AD	365	static struct pde_opener find_pde_opener(struct proc_dir_entry pde,
	366	struct inode inode, struct file file)
	367	{
	368	struct pde_opener *pdeo;
	369
	370	list_for_each_entry(pdeo, &pde->pde_openers, lh) {
	371	if (pdeo->inode == inode && pdeo->file == file)
	372	return pdeo;
	373	}
	374	return NULL;
	375	}
	376
786d7e16 AD	377	static int proc_reg_release(struct inode inode, struct file file)
	378	{
	379	struct proc_dir_entry *pde = PDE(inode);
	380	int rv = 0;
	381	int (release)(struct inode , struct file *);
881adb85	382	struct pde_opener *pdeo;
786d7e16 AD	383
786d7e16 AD	384	spin_lock(&pde->pde_unload_lock);
881adb85	385	pdeo = find_pde_opener(pde, inode, file);
786d7e16	386	if (!pde->proc_fops) {
881adb85 AD	387	/*
	388	* Can't simply exit, __fput() will think that everything is OK,
	389	* and move on to freeing struct file. remove_proc_entry() will
	390	* find slacker in opener's list and will try to do non-trivial
	391	* things with struct file. Therefore, remove opener from list.
	392	*
	393	* But if opener is removed from list, who will ->release it?
	394	*/
	395	if (pdeo) {
	396	list_del(&pdeo->lh);
	397	spin_unlock(&pde->pde_unload_lock);
	398	rv = pdeo->release(inode, file);
	399	kfree(pdeo);
	400	} else
	401	spin_unlock(&pde->pde_unload_lock);
786d7e16 AD	402	return rv;
	403	}
	404	pde->pde_users++;
	405	release = pde->proc_fops->release;
881adb85 AD	406	if (pdeo) {
	407	list_del(&pdeo->lh);
	408	kfree(pdeo);
	409	}
786d7e16 AD	410	spin_unlock(&pde->pde_unload_lock);
	411
	412	if (release)
	413	rv = release(inode, file);
	414
	415	pde_users_dec(pde);
	416	return rv;
	417	}
	418
	419	static const struct file_operations proc_reg_file_ops = {
	420	.llseek = proc_reg_llseek,
	421	.read = proc_reg_read,
	422	.write = proc_reg_write,
	423	.poll = proc_reg_poll,
	424	.unlocked_ioctl = proc_reg_unlocked_ioctl,
	425	#ifdef CONFIG_COMPAT
	426	.compat_ioctl = proc_reg_compat_ioctl,
	427	#endif
	428	.mmap = proc_reg_mmap,
	429	.open = proc_reg_open,
	430	.release = proc_reg_release,
	431	};
	432
778f3dd5 DM	433	#ifdef CONFIG_COMPAT
	434	static const struct file_operations proc_reg_file_ops_no_compat = {
	435	.llseek = proc_reg_llseek,
	436	.read = proc_reg_read,
	437	.write = proc_reg_write,
	438	.poll = proc_reg_poll,
	439	.unlocked_ioctl = proc_reg_unlocked_ioctl,
	440	.mmap = proc_reg_mmap,
	441	.open = proc_reg_open,
	442	.release = proc_reg_release,
	443	};
	444	#endif
	445
1da177e4 LT	446	struct inode proc_get_inode(struct super_block sb, unsigned int ino,
	447	struct proc_dir_entry *de)
	448	{
	449	struct inode * inode;
	450
5e971dce	451	if (!try_module_get(de->owner))
e9543659 KK	452	goto out_mod;
e9543659 KK	453
a1d4aebb	454	inode = iget_locked(sb, ino);
1da177e4	455	if (!inode)
e9543659	456	goto out_ino;
a1d4aebb DH	457	if (inode->i_state & I_NEW) {
	458	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
	459	PROC_I(inode)->fd = 0;
	460	PROC_I(inode)->pde = de;
5e971dce AD	461
	462	if (de->mode) {
	463	inode->i_mode = de->mode;
	464	inode->i_uid = de->uid;
	465	inode->i_gid = de->gid;
	466	}
	467	if (de->size)
	468	inode->i_size = de->size;
	469	if (de->nlink)
	470	inode->i_nlink = de->nlink;
	471	if (de->proc_iops)
	472	inode->i_op = de->proc_iops;
	473	if (de->proc_fops) {
	474	if (S_ISREG(inode->i_mode)) {
778f3dd5	475	#ifdef CONFIG_COMPAT
5e971dce AD	476	if (!de->proc_fops->compat_ioctl)
	477	inode->i_fop =
	478	&proc_reg_file_ops_no_compat;
	479	else
778f3dd5	480	#endif
5e971dce AD	481	inode->i_fop = &proc_reg_file_ops;
	482	} else {
	483	inode->i_fop = de->proc_fops;
778f3dd5	484	}
786d7e16	485	}
a1d4aebb	486	unlock_new_inode(inode);
c4185a0e DL	487	} else
c4185a0e DL	488	module_put(de->owner);
1da177e4 LT	489	return inode;
1da177e4 LT	490
e9543659	491	out_ino:
5e971dce	492	module_put(de->owner);
e9543659	493	out_mod:
e9543659	494	return NULL;
1da177e4 LT	495	}
1da177e4 LT	496
07543f5c	497	int proc_fill_super(struct super_block *s)
1da177e4 LT	498	{
	499	struct inode * root_inode;
	500
92d03285	501	s->s_flags \|= MS_NODIRATIME \| MS_NOSUID \| MS_NOEXEC;
1da177e4 LT	502	s->s_blocksize = 1024;
	503	s->s_blocksize_bits = 10;
	504	s->s_magic = PROC_SUPER_MAGIC;
	505	s->s_op = &proc_sops;
	506	s->s_time_gran = 1;
	507
7695650a	508	de_get(&proc_root);
1da177e4 LT	509	root_inode = proc_get_inode(s, PROC_ROOT_INO, &proc_root);
	510	if (!root_inode)
	511	goto out_no_root;
1da177e4 LT	512	root_inode->i_uid = 0;
	513	root_inode->i_gid = 0;
	514	s->s_root = d_alloc_root(root_inode);
	515	if (!s->s_root)
	516	goto out_no_root;
	517	return 0;
	518
	519	out_no_root:
	520	printk("proc_read_super: get root inode failed\n");
	521	iput(root_inode);
7695650a	522	de_put(&proc_root);
1da177e4 LT	523	return -ENOMEM;
1da177e4 LT	524	}