[net-next-2.6.git] / fs / read_write.c

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

#include <linux/slab.h> 
#include <linux/stat.h>
#include <linux/fcntl.h>
#include <linux/file.h>
#include <linux/uio.h>
#include <linux/smp_lock.h>
#include <linux/fsnotify.h>
#include <linux/security.h>
#include <linux/module.h>
#include <linux/syscalls.h>
#include <linux/pagemap.h>

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

const struct file_operations generic_ro_fops = {
	.llseek		= generic_file_llseek,
	.read		= generic_file_read,
	.mmap		= generic_file_readonly_mmap,
	.sendfile	= generic_file_sendfile,
};

EXPORT_SYMBOL(generic_ro_fops);

loff_t generic_file_llseek(struct file *file, loff_t offset, int origin)
{
	long long retval;
	struct inode *inode = file->f_mapping->host;

	mutex_lock(&inode->i_mutex);
	switch (origin) {
		case 2:
			offset += inode->i_size;
			break;
		case 1:
			offset += file->f_pos;
	}
	retval = -EINVAL;
	if (offset>=0 && offset<=inode->i_sb->s_maxbytes) {
		if (offset != file->f_pos) {
			file->f_pos = offset;
			file->f_version = 0;
		}
		retval = offset;
	}
	mutex_unlock(&inode->i_mutex);
	return retval;
}

EXPORT_SYMBOL(generic_file_llseek);

loff_t remote_llseek(struct file *file, loff_t offset, int origin)
{
	long long retval;

	lock_kernel();
	switch (origin) {
		case 2:
			offset += i_size_read(file->f_dentry->d_inode);
			break;
		case 1:
			offset += file->f_pos;
	}
	retval = -EINVAL;
	if (offset>=0 && offset<=file->f_dentry->d_inode->i_sb->s_maxbytes) {
		if (offset != file->f_pos) {
			file->f_pos = offset;
			file->f_version = 0;
		}
		retval = offset;
	}
	unlock_kernel();
	return retval;
}
EXPORT_SYMBOL(remote_llseek);

loff_t no_llseek(struct file *file, loff_t offset, int origin)
{
	return -ESPIPE;
}
EXPORT_SYMBOL(no_llseek);

loff_t default_llseek(struct file *file, loff_t offset, int origin)
{
	long long retval;

	lock_kernel();
	switch (origin) {
		case 2:
			offset += i_size_read(file->f_dentry->d_inode);
			break;
		case 1:
			offset += file->f_pos;
	}
	retval = -EINVAL;
	if (offset >= 0) {
		if (offset != file->f_pos) {
			file->f_pos = offset;
			file->f_version = 0;
		}
		retval = offset;
	}
	unlock_kernel();
	return retval;
}
EXPORT_SYMBOL(default_llseek);

loff_t vfs_llseek(struct file *file, loff_t offset, int origin)
{
	loff_t (*fn)(struct file *, loff_t, int);

	fn = no_llseek;
	if (file->f_mode & FMODE_LSEEK) {
		fn = default_llseek;
		if (file->f_op && file->f_op->llseek)
			fn = file->f_op->llseek;
	}
	return fn(file, offset, origin);
}
EXPORT_SYMBOL(vfs_llseek);

asmlinkage off_t sys_lseek(unsigned int fd, off_t offset, unsigned int origin)
{
	off_t retval;
	struct file * file;
	int fput_needed;

	retval = -EBADF;
	file = fget_light(fd, &fput_needed);
	if (!file)
		goto bad;

	retval = -EINVAL;
	if (origin <= 2) {
		loff_t res = vfs_llseek(file, offset, origin);
		retval = res;
		if (res != (loff_t)retval)
			retval = -EOVERFLOW;	/* LFS: should only happen on 32 bit platforms */
	}
	fput_light(file, fput_needed);
bad:
	return retval;
}

#ifdef __ARCH_WANT_SYS_LLSEEK
asmlinkage long sys_llseek(unsigned int fd, unsigned long offset_high,
			   unsigned long offset_low, loff_t __user * result,
			   unsigned int origin)
{
	int retval;
	struct file * file;
	loff_t offset;
	int fput_needed;

	retval = -EBADF;
	file = fget_light(fd, &fput_needed);
	if (!file)
		goto bad;

	retval = -EINVAL;
	if (origin > 2)
		goto out_putf;

	offset = vfs_llseek(file, ((loff_t) offset_high << 32) | offset_low,
			origin);

	retval = (int)offset;
	if (offset >= 0) {
		retval = -EFAULT;
		if (!copy_to_user(result, &offset, sizeof(offset)))
			retval = 0;
	}
out_putf:
	fput_light(file, fput_needed);
bad:
	return retval;
}
#endif

/*
 * rw_verify_area doesn't like huge counts. We limit
 * them to something that fits in "int" so that others
 * won't have to do range checks all the time.
 */
#define MAX_RW_COUNT (INT_MAX & PAGE_CACHE_MASK)

int rw_verify_area(int read_write, struct file *file, loff_t *ppos, size_t count)
{
	struct inode *inode;
	loff_t pos;

	if (unlikely((ssize_t) count < 0))
		goto Einval;
	pos = *ppos;
	if (unlikely((pos < 0) || (loff_t) (pos + count) < 0))
		goto Einval;

	inode = file->f_dentry->d_inode;
	if (unlikely(inode->i_flock && MANDATORY_LOCK(inode))) {
		int retval = locks_mandatory_area(
			read_write == READ ? FLOCK_VERIFY_READ : FLOCK_VERIFY_WRITE,
			inode, file, pos, count);
		if (retval < 0)
			return retval;
	}
	return count > MAX_RW_COUNT ? MAX_RW_COUNT : count;

Einval:
	return -EINVAL;
}

static void wait_on_retry_sync_kiocb(struct kiocb *iocb)
{
	set_current_state(TASK_UNINTERRUPTIBLE);
	if (!kiocbIsKicked(iocb))
		schedule();
	else
		kiocbClearKicked(iocb);
	__set_current_state(TASK_RUNNING);
}

ssize_t do_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos)
{
	struct iovec iov = { .iov_base = buf, .iov_len = len };
	struct kiocb kiocb;
	ssize_t ret;

	init_sync_kiocb(&kiocb, filp);
	kiocb.ki_pos = *ppos;
	kiocb.ki_left = len;

	for (;;) {
		ret = filp->f_op->aio_read(&kiocb, &iov, 1, kiocb.ki_pos);
		if (ret != -EIOCBRETRY)
			break;
		wait_on_retry_sync_kiocb(&kiocb);
	}

	if (-EIOCBQUEUED == ret)
		ret = wait_on_sync_kiocb(&kiocb);
	*ppos = kiocb.ki_pos;
	return ret;
}

EXPORT_SYMBOL(do_sync_read);

ssize_t vfs_read(struct file *file, char __user *buf, size_t count, loff_t *pos)
{
	ssize_t ret;

	if (!(file->f_mode & FMODE_READ))
		return -EBADF;
	if (!file->f_op || (!file->f_op->read && !file->f_op->aio_read))
		return -EINVAL;
	if (unlikely(!access_ok(VERIFY_WRITE, buf, count)))
		return -EFAULT;

	ret = rw_verify_area(READ, file, pos, count);
	if (ret >= 0) {
		count = ret;
		ret = security_file_permission (file, MAY_READ);
		if (!ret) {
			if (file->f_op->read)
				ret = file->f_op->read(file, buf, count, pos);
			else
				ret = do_sync_read(file, buf, count, pos);
			if (ret > 0) {
				fsnotify_access(file->f_dentry);
				current->rchar += ret;
			}
			current->syscr++;
		}
	}

	return ret;
}

EXPORT_SYMBOL(vfs_read);

ssize_t do_sync_write(struct file *filp, const char __user *buf, size_t len, loff_t *ppos)
{
	struct iovec iov = { .iov_base = (void __user *)buf, .iov_len = len };
	struct kiocb kiocb;
	ssize_t ret;

	init_sync_kiocb(&kiocb, filp);
	kiocb.ki_pos = *ppos;
	kiocb.ki_left = len;

	for (;;) {
		ret = filp->f_op->aio_write(&kiocb, &iov, 1, kiocb.ki_pos);
		if (ret != -EIOCBRETRY)
			break;
		wait_on_retry_sync_kiocb(&kiocb);
	}

	if (-EIOCBQUEUED == ret)
		ret = wait_on_sync_kiocb(&kiocb);
	*ppos = kiocb.ki_pos;
	return ret;
}

EXPORT_SYMBOL(do_sync_write);

ssize_t vfs_write(struct file *file, const char __user *buf, size_t count, loff_t *pos)
{
	ssize_t ret;

	if (!(file->f_mode & FMODE_WRITE))
		return -EBADF;
	if (!file->f_op || (!file->f_op->write && !file->f_op->aio_write))
		return -EINVAL;
	if (unlikely(!access_ok(VERIFY_READ, buf, count)))
		return -EFAULT;

	ret = rw_verify_area(WRITE, file, pos, count);
	if (ret >= 0) {
		count = ret;
		ret = security_file_permission (file, MAY_WRITE);
		if (!ret) {
			if (file->f_op->write)
				ret = file->f_op->write(file, buf, count, pos);
			else
				ret = do_sync_write(file, buf, count, pos);
			if (ret > 0) {
				fsnotify_modify(file->f_dentry);
				current->wchar += ret;
			}
			current->syscw++;
		}
	}

	return ret;
}

EXPORT_SYMBOL(vfs_write);

static inline loff_t file_pos_read(struct file *file)
{
	return file->f_pos;
}

static inline void file_pos_write(struct file *file, loff_t pos)
{
	file->f_pos = pos;
}

asmlinkage ssize_t sys_read(unsigned int fd, char __user * buf, size_t count)
{
	struct file *file;
	ssize_t ret = -EBADF;
	int fput_needed;

	file = fget_light(fd, &fput_needed);
	if (file) {
		loff_t pos = file_pos_read(file);
		ret = vfs_read(file, buf, count, &pos);
		file_pos_write(file, pos);
		fput_light(file, fput_needed);
	}

	return ret;
}
EXPORT_SYMBOL_GPL(sys_read);

asmlinkage ssize_t sys_write(unsigned int fd, const char __user * buf, size_t count)
{
	struct file *file;
	ssize_t ret = -EBADF;
	int fput_needed;

	file = fget_light(fd, &fput_needed);
	if (file) {
		loff_t pos = file_pos_read(file);
		ret = vfs_write(file, buf, count, &pos);
		file_pos_write(file, pos);
		fput_light(file, fput_needed);
	}

	return ret;
}

asmlinkage ssize_t sys_pread64(unsigned int fd, char __user *buf,
			     size_t count, loff_t pos)
{
	struct file *file;
	ssize_t ret = -EBADF;
	int fput_needed;

	if (pos < 0)
		return -EINVAL;

	file = fget_light(fd, &fput_needed);
	if (file) {
		ret = -ESPIPE;
		if (file->f_mode & FMODE_PREAD)
			ret = vfs_read(file, buf, count, &pos);
		fput_light(file, fput_needed);
	}

	return ret;
}

asmlinkage ssize_t sys_pwrite64(unsigned int fd, const char __user *buf,
			      size_t count, loff_t pos)
{
	struct file *file;
	ssize_t ret = -EBADF;
	int fput_needed;

	if (pos < 0)
		return -EINVAL;

	file = fget_light(fd, &fput_needed);
	if (file) {
		ret = -ESPIPE;
		if (file->f_mode & FMODE_PWRITE)  
			ret = vfs_write(file, buf, count, &pos);
		fput_light(file, fput_needed);
	}

	return ret;
}

/*
 * Reduce an iovec's length in-place.  Return the resulting number of segments
 */
unsigned long iov_shorten(struct iovec *iov, unsigned long nr_segs, size_t to)
{
	unsigned long seg = 0;
	size_t len = 0;

	while (seg < nr_segs) {
		seg++;
		if (len + iov->iov_len >= to) {
			iov->iov_len = to - len;
			break;
		}
		len += iov->iov_len;
		iov++;
	}
	return seg;
}

EXPORT_UNUSED_SYMBOL(iov_shorten);  /*  June 2006  */

/* A write operation does a read from user space and vice versa */
#define vrfy_dir(type) ((type) == READ ? VERIFY_WRITE : VERIFY_READ)

static ssize_t do_readv_writev(int type, struct file *file,
			       const struct iovec __user * uvector,
			       unsigned long nr_segs, loff_t *pos)
{
	typedef ssize_t (*io_fn_t)(struct file *, char __user *, size_t, loff_t *);
	typedef ssize_t (*iov_fn_t)(struct file *, const struct iovec *, unsigned long, loff_t *);

	size_t tot_len;
	struct iovec iovstack[UIO_FASTIOV];
	struct iovec *iov=iovstack, *vector;
	ssize_t ret;
	int seg;
	io_fn_t fn;
	iov_fn_t fnv;

	/*
	 * SuS says "The readv() function *may* fail if the iovcnt argument
	 * was less than or equal to 0, or greater than {IOV_MAX}.  Linux has
	 * traditionally returned zero for zero segments, so...
	 */
	ret = 0;
	if (nr_segs == 0)
		goto out;

	/*
	 * First get the "struct iovec" from user memory and
	 * verify all the pointers
	 */
	ret = -EINVAL;
	if (nr_segs > UIO_MAXIOV)
		goto out;
	if (!file->f_op)
		goto out;
	if (nr_segs > UIO_FASTIOV) {
		ret = -ENOMEM;
		iov = kmalloc(nr_segs*sizeof(struct iovec), GFP_KERNEL);
		if (!iov)
			goto out;
	}
	ret = -EFAULT;
	if (copy_from_user(iov, uvector, nr_segs*sizeof(*uvector)))
		goto out;

	/*
	 * Single unix specification:
	 * We should -EINVAL if an element length is not >= 0 and fitting an
	 * ssize_t.  The total length is fitting an ssize_t
	 *
	 * Be careful here because iov_len is a size_t not an ssize_t
	 */
	tot_len = 0;
	ret = -EINVAL;
	for (seg = 0; seg < nr_segs; seg++) {
		void __user *buf = iov[seg].iov_base;
		ssize_t len = (ssize_t)iov[seg].iov_len;

		if (len < 0)	/* size_t not fitting an ssize_t .. */
			goto out;
		if (unlikely(!access_ok(vrfy_dir(type), buf, len)))
			goto Efault;
		tot_len += len;
		if ((ssize_t)tot_len < 0) /* maths overflow on the ssize_t */
			goto out;
	}
	if (tot_len == 0) {
		ret = 0;
		goto out;
	}

	ret = rw_verify_area(type, file, pos, tot_len);
	if (ret < 0)
		goto out;
	ret = security_file_permission(file, type == READ ? MAY_READ : MAY_WRITE);
	if (ret)
		goto out;

	fnv = NULL;
	if (type == READ) {
		fn = file->f_op->read;
		fnv = file->f_op->readv;
	} else {
		fn = (io_fn_t)file->f_op->write;
		fnv = file->f_op->writev;
	}
	if (fnv) {
		ret = fnv(file, iov, nr_segs, pos);
		goto out;
	}

	/* Do it by hand, with file-ops */
	ret = 0;
	vector = iov;
	while (nr_segs > 0) {
		void __user * base;
		size_t len;
		ssize_t nr;

		base = vector->iov_base;
		len = vector->iov_len;
		vector++;
		nr_segs--;

		nr = fn(file, base, len, pos);

		if (nr < 0) {
			if (!ret) ret = nr;
			break;
		}
		ret += nr;
		if (nr != len)
			break;
	}
out:
	if (iov != iovstack)
		kfree(iov);
	if ((ret + (type == READ)) > 0) {
		if (type == READ)
			fsnotify_access(file->f_dentry);
		else
			fsnotify_modify(file->f_dentry);
	}
	return ret;
Efault:
	ret = -EFAULT;
	goto out;
}

ssize_t vfs_readv(struct file *file, const struct iovec __user *vec,
		  unsigned long vlen, loff_t *pos)
{
	if (!(file->f_mode & FMODE_READ))
		return -EBADF;
	if (!file->f_op || (!file->f_op->readv && !file->f_op->read))
		return -EINVAL;

	return do_readv_writev(READ, file, vec, vlen, pos);
}

EXPORT_SYMBOL(vfs_readv);

ssize_t vfs_writev(struct file *file, const struct iovec __user *vec,
		   unsigned long vlen, loff_t *pos)
{
	if (!(file->f_mode & FMODE_WRITE))
		return -EBADF;
	if (!file->f_op || (!file->f_op->writev && !file->f_op->write))
		return -EINVAL;

	return do_readv_writev(WRITE, file, vec, vlen, pos);
}

EXPORT_SYMBOL(vfs_writev);

asmlinkage ssize_t
sys_readv(unsigned long fd, const struct iovec __user *vec, unsigned long vlen)
{
	struct file *file;
	ssize_t ret = -EBADF;
	int fput_needed;

	file = fget_light(fd, &fput_needed);
	if (file) {
		loff_t pos = file_pos_read(file);
		ret = vfs_readv(file, vec, vlen, &pos);
		file_pos_write(file, pos);
		fput_light(file, fput_needed);
	}

	if (ret > 0)
		current->rchar += ret;
	current->syscr++;
	return ret;
}

asmlinkage ssize_t
sys_writev(unsigned long fd, const struct iovec __user *vec, unsigned long vlen)
{
	struct file *file;
	ssize_t ret = -EBADF;
	int fput_needed;

	file = fget_light(fd, &fput_needed);
	if (file) {
		loff_t pos = file_pos_read(file);
		ret = vfs_writev(file, vec, vlen, &pos);
		file_pos_write(file, pos);
		fput_light(file, fput_needed);
	}

	if (ret > 0)
		current->wchar += ret;
	current->syscw++;
	return ret;
}

static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos,
			   size_t count, loff_t max)
{
	struct file * in_file, * out_file;
	struct inode * in_inode, * out_inode;
	loff_t pos;
	ssize_t retval;
	int fput_needed_in, fput_needed_out;

	/*
	 * Get input file, and verify that it is ok..
	 */
	retval = -EBADF;
	in_file = fget_light(in_fd, &fput_needed_in);
	if (!in_file)
		goto out;
	if (!(in_file->f_mode & FMODE_READ))
		goto fput_in;
	retval = -EINVAL;
	in_inode = in_file->f_dentry->d_inode;
	if (!in_inode)
		goto fput_in;
	if (!in_file->f_op || !in_file->f_op->sendfile)
		goto fput_in;
	retval = -ESPIPE;
	if (!ppos)
		ppos = &in_file->f_pos;
	else
		if (!(in_file->f_mode & FMODE_PREAD))
			goto fput_in;
	retval = rw_verify_area(READ, in_file, ppos, count);
	if (retval < 0)
		goto fput_in;
	count = retval;

	retval = security_file_permission (in_file, MAY_READ);
	if (retval)
		goto fput_in;

	/*
	 * Get output file, and verify that it is ok..
	 */
	retval = -EBADF;
	out_file = fget_light(out_fd, &fput_needed_out);
	if (!out_file)
		goto fput_in;
	if (!(out_file->f_mode & FMODE_WRITE))
		goto fput_out;
	retval = -EINVAL;
	if (!out_file->f_op || !out_file->f_op->sendpage)
		goto fput_out;
	out_inode = out_file->f_dentry->d_inode;
	retval = rw_verify_area(WRITE, out_file, &out_file->f_pos, count);
	if (retval < 0)
		goto fput_out;
	count = retval;

	retval = security_file_permission (out_file, MAY_WRITE);
	if (retval)
		goto fput_out;

	if (!max)
		max = min(in_inode->i_sb->s_maxbytes, out_inode->i_sb->s_maxbytes);

	pos = *ppos;
	retval = -EINVAL;
	if (unlikely(pos < 0))
		goto fput_out;
	if (unlikely(pos + count > max)) {
		retval = -EOVERFLOW;
		if (pos >= max)
			goto fput_out;
		count = max - pos;
	}

	retval = in_file->f_op->sendfile(in_file, ppos, count, file_send_actor, out_file);

	if (retval > 0) {
		current->rchar += retval;
		current->wchar += retval;
	}
	current->syscr++;
	current->syscw++;

	if (*ppos > max)
		retval = -EOVERFLOW;

fput_out:
	fput_light(out_file, fput_needed_out);
fput_in:
	fput_light(in_file, fput_needed_in);
out:
	return retval;
}

asmlinkage ssize_t sys_sendfile(int out_fd, int in_fd, off_t __user *offset, size_t count)
{
	loff_t pos;
	off_t off;
	ssize_t ret;

	if (offset) {
		if (unlikely(get_user(off, offset)))
			return -EFAULT;
		pos = off;
		ret = do_sendfile(out_fd, in_fd, &pos, count, MAX_NON_LFS);
		if (unlikely(put_user(pos, offset)))
			return -EFAULT;
		return ret;
	}

	return do_sendfile(out_fd, in_fd, NULL, count, 0);
}

asmlinkage ssize_t sys_sendfile64(int out_fd, int in_fd, loff_t __user *offset, size_t count)
{
	loff_t pos;
	ssize_t ret;

	if (offset) {
		if (unlikely(copy_from_user(&pos, offset, sizeof(loff_t))))
			return -EFAULT;
		ret = do_sendfile(out_fd, in_fd, &pos, count, 0);
		if (unlikely(put_user(pos, offset)))
			return -EFAULT;
		return ret;
	}

	return do_sendfile(out_fd, in_fd, NULL, count, 0);
}
Commit	Line	Data
1da177e4 LT	1	/*
	2	* linux/fs/read_write.c
	3	*
	4	* Copyright (C) 1991, 1992 Linus Torvalds
	5	*/
	6
	7	#include <linux/slab.h>
	8	#include <linux/stat.h>
	9	#include <linux/fcntl.h>
	10	#include <linux/file.h>
	11	#include <linux/uio.h>
	12	#include <linux/smp_lock.h>
0eeca283	13	#include <linux/fsnotify.h>
1da177e4 LT	14	#include <linux/security.h>
	15	#include <linux/module.h>
	16	#include <linux/syscalls.h>
e28cc715	17	#include <linux/pagemap.h>
1da177e4 LT	18
	19	#include <asm/uaccess.h>
	20	#include <asm/unistd.h>
	21
4b6f5d20	22	const struct file_operations generic_ro_fops = {
1da177e4 LT	23	.llseek = generic_file_llseek,
	24	.read = generic_file_read,
	25	.mmap = generic_file_readonly_mmap,
	26	.sendfile = generic_file_sendfile,
	27	};
	28
	29	EXPORT_SYMBOL(generic_ro_fops);
	30
	31	loff_t generic_file_llseek(struct file *file, loff_t offset, int origin)
	32	{
	33	long long retval;
	34	struct inode *inode = file->f_mapping->host;
	35
1b1dcc1b	36	mutex_lock(&inode->i_mutex);
1da177e4 LT	37	switch (origin) {
	38	case 2:
	39	offset += inode->i_size;
	40	break;
	41	case 1:
	42	offset += file->f_pos;
	43	}
	44	retval = -EINVAL;
	45	if (offset>=0 && offset<=inode->i_sb->s_maxbytes) {
	46	if (offset != file->f_pos) {
	47	file->f_pos = offset;
	48	file->f_version = 0;
	49	}
	50	retval = offset;
	51	}
1b1dcc1b	52	mutex_unlock(&inode->i_mutex);
1da177e4 LT	53	return retval;
	54	}
	55
	56	EXPORT_SYMBOL(generic_file_llseek);
	57
	58	loff_t remote_llseek(struct file *file, loff_t offset, int origin)
	59	{
	60	long long retval;
	61
	62	lock_kernel();
	63	switch (origin) {
	64	case 2:
	65	offset += i_size_read(file->f_dentry->d_inode);
	66	break;
	67	case 1:
	68	offset += file->f_pos;
	69	}
	70	retval = -EINVAL;
	71	if (offset>=0 && offset<=file->f_dentry->d_inode->i_sb->s_maxbytes) {
	72	if (offset != file->f_pos) {
	73	file->f_pos = offset;
	74	file->f_version = 0;
	75	}
	76	retval = offset;
	77	}
	78	unlock_kernel();
	79	return retval;
	80	}
	81	EXPORT_SYMBOL(remote_llseek);
	82
	83	loff_t no_llseek(struct file *file, loff_t offset, int origin)
	84	{
	85	return -ESPIPE;
	86	}
	87	EXPORT_SYMBOL(no_llseek);
	88
	89	loff_t default_llseek(struct file *file, loff_t offset, int origin)
	90	{
	91	long long retval;
	92
	93	lock_kernel();
	94	switch (origin) {
	95	case 2:
	96	offset += i_size_read(file->f_dentry->d_inode);
	97	break;
	98	case 1:
	99	offset += file->f_pos;
	100	}
	101	retval = -EINVAL;
	102	if (offset >= 0) {
	103	if (offset != file->f_pos) {
	104	file->f_pos = offset;
	105	file->f_version = 0;
	106	}
	107	retval = offset;
	108	}
	109	unlock_kernel();
	110	return retval;
	111	}
	112	EXPORT_SYMBOL(default_llseek);
	113
	114	loff_t vfs_llseek(struct file *file, loff_t offset, int origin)
	115	{
	116	loff_t (fn)(struct file , loff_t, int);
117
118	fn = no_llseek;
119	if (file->f_mode & FMODE_LSEEK) {
120	fn = default_llseek;
121	if (file->f_op && file->f_op->llseek)
122	fn = file->f_op->llseek;
123	}
124	return fn(file, offset, origin);
125	}
126	EXPORT_SYMBOL(vfs_llseek);
127
128	asmlinkage off_t sys_lseek(unsigned int fd, off_t offset, unsigned int origin)
129	{
130	off_t retval;
131	struct file * file;
132	int fput_needed;
133
134	retval = -EBADF;
135	file = fget_light(fd, &fput_needed);
136	if (!file)
137	goto bad;
138
139	retval = -EINVAL;
140	if (origin <= 2) {
141	loff_t res = vfs_llseek(file, offset, origin);
142	retval = res;
143	if (res != (loff_t)retval)
144	retval = -EOVERFLOW; /* LFS: should only happen on 32 bit platforms */
145	}
146	fput_light(file, fput_needed);
147	bad:
148	return retval;
149	}
150
151	#ifdef __ARCH_WANT_SYS_LLSEEK
152	asmlinkage long sys_llseek(unsigned int fd, unsigned long offset_high,
153	unsigned long offset_low, loff_t __user * result,
154	unsigned int origin)
155	{
156	int retval;
157	struct file * file;
158	loff_t offset;
159	int fput_needed;
160
161	retval = -EBADF;
162	file = fget_light(fd, &fput_needed);
163	if (!file)
164	goto bad;
165
166	retval = -EINVAL;
167	if (origin > 2)
168	goto out_putf;
169
170	offset = vfs_llseek(file, ((loff_t) offset_high << 32) \| offset_low,
171	origin);
172
173	retval = (int)offset;
174	if (offset >= 0) {
175	retval = -EFAULT;
176	if (!copy_to_user(result, &offset, sizeof(offset)))
177	retval = 0;
178	}
179	out_putf:
180	fput_light(file, fput_needed);
181	bad:
182	return retval;
183	}
184	#endif
185
e28cc715 LT	186	/*
	187	* rw_verify_area doesn't like huge counts. We limit
	188	* them to something that fits in "int" so that others
	189	* won't have to do range checks all the time.
	190	*/
	191	#define MAX_RW_COUNT (INT_MAX & PAGE_CACHE_MASK)
1da177e4 LT	192
	193	int rw_verify_area(int read_write, struct file file, loff_t ppos, size_t count)
	194	{
	195	struct inode *inode;
	196	loff_t pos;
	197
e28cc715	198	if (unlikely((ssize_t) count < 0))
1da177e4 LT	199	goto Einval;
	200	pos = *ppos;
	201	if (unlikely((pos < 0) \|\| (loff_t) (pos + count) < 0))
	202	goto Einval;
	203
	204	inode = file->f_dentry->d_inode;
49570e9b	205	if (unlikely(inode->i_flock && MANDATORY_LOCK(inode))) {
e28cc715 LT	206	int retval = locks_mandatory_area(
	207	read_write == READ ? FLOCK_VERIFY_READ : FLOCK_VERIFY_WRITE,
	208	inode, file, pos, count);
	209	if (retval < 0)
	210	return retval;
	211	}
	212	return count > MAX_RW_COUNT ? MAX_RW_COUNT : count;
1da177e4 LT	213
	214	Einval:
	215	return -EINVAL;
	216	}
	217
63e68809 BL	218	static void wait_on_retry_sync_kiocb(struct kiocb *iocb)
	219	{
	220	set_current_state(TASK_UNINTERRUPTIBLE);
	221	if (!kiocbIsKicked(iocb))
	222	schedule();
	223	else
	224	kiocbClearKicked(iocb);
	225	__set_current_state(TASK_RUNNING);
	226	}
	227
1da177e4 LT	228	ssize_t do_sync_read(struct file filp, char __user buf, size_t len, loff_t *ppos)
1da177e4 LT	229	{
027445c3	230	struct iovec iov = { .iov_base = buf, .iov_len = len };
1da177e4 LT	231	struct kiocb kiocb;
	232	ssize_t ret;
	233
	234	init_sync_kiocb(&kiocb, filp);
	235	kiocb.ki_pos = *ppos;
027445c3 BP	236	kiocb.ki_left = len;
	237
	238	for (;;) {
	239	ret = filp->f_op->aio_read(&kiocb, &iov, 1, kiocb.ki_pos);
	240	if (ret != -EIOCBRETRY)
	241	break;
63e68809	242	wait_on_retry_sync_kiocb(&kiocb);
027445c3	243	}
63e68809	244
1da177e4 LT	245	if (-EIOCBQUEUED == ret)
	246	ret = wait_on_sync_kiocb(&kiocb);
	247	*ppos = kiocb.ki_pos;
	248	return ret;
	249	}
	250
	251	EXPORT_SYMBOL(do_sync_read);
	252
	253	ssize_t vfs_read(struct file file, char __user buf, size_t count, loff_t *pos)
	254	{
	255	ssize_t ret;
	256
	257	if (!(file->f_mode & FMODE_READ))
	258	return -EBADF;
	259	if (!file->f_op \|\| (!file->f_op->read && !file->f_op->aio_read))
	260	return -EINVAL;
	261	if (unlikely(!access_ok(VERIFY_WRITE, buf, count)))
	262	return -EFAULT;
	263
	264	ret = rw_verify_area(READ, file, pos, count);
e28cc715 LT	265	if (ret >= 0) {
e28cc715 LT	266	count = ret;
1da177e4 LT	267	ret = security_file_permission (file, MAY_READ);
	268	if (!ret) {
	269	if (file->f_op->read)
	270	ret = file->f_op->read(file, buf, count, pos);
	271	else
	272	ret = do_sync_read(file, buf, count, pos);
	273	if (ret > 0) {
0eeca283	274	fsnotify_access(file->f_dentry);
1da177e4 LT	275	current->rchar += ret;
	276	}
	277	current->syscr++;
	278	}
	279	}
	280
	281	return ret;
	282	}
	283
	284	EXPORT_SYMBOL(vfs_read);
	285
	286	ssize_t do_sync_write(struct file filp, const char __user buf, size_t len, loff_t *ppos)
	287	{
027445c3	288	struct iovec iov = { .iov_base = (void __user *)buf, .iov_len = len };
1da177e4 LT	289	struct kiocb kiocb;
	290	ssize_t ret;
	291
	292	init_sync_kiocb(&kiocb, filp);
	293	kiocb.ki_pos = *ppos;
027445c3 BP	294	kiocb.ki_left = len;
	295
	296	for (;;) {
	297	ret = filp->f_op->aio_write(&kiocb, &iov, 1, kiocb.ki_pos);
	298	if (ret != -EIOCBRETRY)
	299	break;
63e68809	300	wait_on_retry_sync_kiocb(&kiocb);
027445c3	301	}
63e68809	302
1da177e4 LT	303	if (-EIOCBQUEUED == ret)
	304	ret = wait_on_sync_kiocb(&kiocb);
	305	*ppos = kiocb.ki_pos;
	306	return ret;
	307	}
	308
	309	EXPORT_SYMBOL(do_sync_write);
	310
	311	ssize_t vfs_write(struct file file, const char __user buf, size_t count, loff_t *pos)
	312	{
	313	ssize_t ret;
	314
	315	if (!(file->f_mode & FMODE_WRITE))
	316	return -EBADF;
	317	if (!file->f_op \|\| (!file->f_op->write && !file->f_op->aio_write))
	318	return -EINVAL;
	319	if (unlikely(!access_ok(VERIFY_READ, buf, count)))
	320	return -EFAULT;
	321
	322	ret = rw_verify_area(WRITE, file, pos, count);
e28cc715 LT	323	if (ret >= 0) {
e28cc715 LT	324	count = ret;
1da177e4 LT	325	ret = security_file_permission (file, MAY_WRITE);
	326	if (!ret) {
	327	if (file->f_op->write)
	328	ret = file->f_op->write(file, buf, count, pos);
	329	else
	330	ret = do_sync_write(file, buf, count, pos);
	331	if (ret > 0) {
0eeca283	332	fsnotify_modify(file->f_dentry);
1da177e4 LT	333	current->wchar += ret;
	334	}
	335	current->syscw++;
	336	}
	337	}
	338
	339	return ret;
	340	}
	341
	342	EXPORT_SYMBOL(vfs_write);
	343
	344	static inline loff_t file_pos_read(struct file *file)
	345	{
	346	return file->f_pos;
	347	}
	348
	349	static inline void file_pos_write(struct file *file, loff_t pos)
	350	{
	351	file->f_pos = pos;
	352	}
	353
	354	asmlinkage ssize_t sys_read(unsigned int fd, char __user * buf, size_t count)
	355	{
	356	struct file *file;
	357	ssize_t ret = -EBADF;
	358	int fput_needed;
	359
	360	file = fget_light(fd, &fput_needed);
	361	if (file) {
	362	loff_t pos = file_pos_read(file);
	363	ret = vfs_read(file, buf, count, &pos);
	364	file_pos_write(file, pos);
	365	fput_light(file, fput_needed);
	366	}
	367
	368	return ret;
	369	}
	370	EXPORT_SYMBOL_GPL(sys_read);
	371
	372	asmlinkage ssize_t sys_write(unsigned int fd, const char __user * buf, size_t count)
	373	{
	374	struct file *file;
	375	ssize_t ret = -EBADF;
	376	int fput_needed;
	377
	378	file = fget_light(fd, &fput_needed);
	379	if (file) {
	380	loff_t pos = file_pos_read(file);
	381	ret = vfs_write(file, buf, count, &pos);
	382	file_pos_write(file, pos);
	383	fput_light(file, fput_needed);
	384	}
	385
	386	return ret;
	387	}
	388
	389	asmlinkage ssize_t sys_pread64(unsigned int fd, char __user *buf,
	390	size_t count, loff_t pos)
	391	{
	392	struct file *file;
	393	ssize_t ret = -EBADF;
	394	int fput_needed;
	395
	396	if (pos < 0)
397	return -EINVAL;
398
399	file = fget_light(fd, &fput_needed);
400	if (file) {
401	ret = -ESPIPE;
402	if (file->f_mode & FMODE_PREAD)
403	ret = vfs_read(file, buf, count, &pos);
404	fput_light(file, fput_needed);
405	}
406
407	return ret;
408	}
409
410	asmlinkage ssize_t sys_pwrite64(unsigned int fd, const char __user *buf,
411	size_t count, loff_t pos)
412	{
413	struct file *file;
414	ssize_t ret = -EBADF;
415	int fput_needed;
416
417	if (pos < 0)
418	return -EINVAL;
419
420	file = fget_light(fd, &fput_needed);
421	if (file) {
422	ret = -ESPIPE;
423	if (file->f_mode & FMODE_PWRITE)
424	ret = vfs_write(file, buf, count, &pos);
425	fput_light(file, fput_needed);
426	}
427
428	return ret;
429	}
430
431	/*
432	* Reduce an iovec's length in-place. Return the resulting number of segments
433	*/
434	unsigned long iov_shorten(struct iovec *iov, unsigned long nr_segs, size_t to)
435	{
436	unsigned long seg = 0;
437	size_t len = 0;
438
439	while (seg < nr_segs) {
440	seg++;
441	if (len + iov->iov_len >= to) {
442	iov->iov_len = to - len;
443	break;
444	}
445	len += iov->iov_len;
446	iov++;
447	}
448	return seg;
449	}
450
69c3a5b8	451	EXPORT_UNUSED_SYMBOL(iov_shorten); /* June 2006 */
1da177e4 LT	452
	453	/* A write operation does a read from user space and vice versa */
	454	#define vrfy_dir(type) ((type) == READ ? VERIFY_WRITE : VERIFY_READ)
	455
	456	static ssize_t do_readv_writev(int type, struct file *file,
	457	const struct iovec __user * uvector,
	458	unsigned long nr_segs, loff_t *pos)
	459	{
	460	typedef ssize_t (io_fn_t)(struct file , char __user , size_t, loff_t );
	461	typedef ssize_t (iov_fn_t)(struct file , const struct iovec , unsigned long, loff_t );
	462
	463	size_t tot_len;
	464	struct iovec iovstack[UIO_FASTIOV];
	465	struct iovec iov=iovstack, vector;
	466	ssize_t ret;
	467	int seg;
	468	io_fn_t fn;
	469	iov_fn_t fnv;
	470
	471	/*
	472	* SuS says "The readv() function may fail if the iovcnt argument
	473	* was less than or equal to 0, or greater than {IOV_MAX}. Linux has
	474	* traditionally returned zero for zero segments, so...
	475	*/
	476	ret = 0;
	477	if (nr_segs == 0)
	478	goto out;
	479
	480	/*
	481	* First get the "struct iovec" from user memory and
	482	* verify all the pointers
	483	*/
	484	ret = -EINVAL;
6cc6b122	485	if (nr_segs > UIO_MAXIOV)
1da177e4 LT	486	goto out;
	487	if (!file->f_op)
	488	goto out;
	489	if (nr_segs > UIO_FASTIOV) {
	490	ret = -ENOMEM;
	491	iov = kmalloc(nr_segs*sizeof(struct iovec), GFP_KERNEL);
	492	if (!iov)
	493	goto out;
	494	}
	495	ret = -EFAULT;
	496	if (copy_from_user(iov, uvector, nr_segssizeof(uvector)))
	497	goto out;
	498
	499	/*
	500	* Single unix specification:
	501	* We should -EINVAL if an element length is not >= 0 and fitting an
	502	* ssize_t. The total length is fitting an ssize_t
	503	*
	504	* Be careful here because iov_len is a size_t not an ssize_t
	505	*/
	506	tot_len = 0;
	507	ret = -EINVAL;
	508	for (seg = 0; seg < nr_segs; seg++) {
	509	void __user *buf = iov[seg].iov_base;
	510	ssize_t len = (ssize_t)iov[seg].iov_len;
	511
1da177e4 LT	512	if (len < 0) /* size_t not fitting an ssize_t .. */
1da177e4 LT	513	goto out;
1f08ad02 DH	514	if (unlikely(!access_ok(vrfy_dir(type), buf, len)))
1f08ad02 DH	515	goto Efault;
1da177e4 LT	516	tot_len += len;
	517	if ((ssize_t)tot_len < 0) /* maths overflow on the ssize_t */
	518	goto out;
	519	}
	520	if (tot_len == 0) {
	521	ret = 0;
	522	goto out;
	523	}
	524
	525	ret = rw_verify_area(type, file, pos, tot_len);
e28cc715	526	if (ret < 0)
411b67b4 KB	527	goto out;
411b67b4 KB	528	ret = security_file_permission(file, type == READ ? MAY_READ : MAY_WRITE);
1da177e4 LT	529	if (ret)
	530	goto out;
	531
	532	fnv = NULL;
	533	if (type == READ) {
	534	fn = file->f_op->read;
	535	fnv = file->f_op->readv;
	536	} else {
	537	fn = (io_fn_t)file->f_op->write;
	538	fnv = file->f_op->writev;
	539	}
	540	if (fnv) {
	541	ret = fnv(file, iov, nr_segs, pos);
	542	goto out;
	543	}
	544
	545	/* Do it by hand, with file-ops */
	546	ret = 0;
	547	vector = iov;
	548	while (nr_segs > 0) {
	549	void __user * base;
	550	size_t len;
	551	ssize_t nr;
	552
	553	base = vector->iov_base;
	554	len = vector->iov_len;
	555	vector++;
	556	nr_segs--;
	557
	558	nr = fn(file, base, len, pos);
	559
	560	if (nr < 0) {
	561	if (!ret) ret = nr;
	562	break;
	563	}
	564	ret += nr;
	565	if (nr != len)
	566	break;
	567	}
	568	out:
	569	if (iov != iovstack)
	570	kfree(iov);
0eeca283 RL	571	if ((ret + (type == READ)) > 0) {
	572	if (type == READ)
	573	fsnotify_access(file->f_dentry);
	574	else
	575	fsnotify_modify(file->f_dentry);
	576	}
1da177e4 LT	577	return ret;
	578	Efault:
	579	ret = -EFAULT;
	580	goto out;
	581	}
	582
	583	ssize_t vfs_readv(struct file file, const struct iovec __user vec,
	584	unsigned long vlen, loff_t *pos)
	585	{
	586	if (!(file->f_mode & FMODE_READ))
	587	return -EBADF;
	588	if (!file->f_op \|\| (!file->f_op->readv && !file->f_op->read))
	589	return -EINVAL;
	590
	591	return do_readv_writev(READ, file, vec, vlen, pos);
	592	}
	593
	594	EXPORT_SYMBOL(vfs_readv);
	595
	596	ssize_t vfs_writev(struct file file, const struct iovec __user vec,
	597	unsigned long vlen, loff_t *pos)
	598	{
	599	if (!(file->f_mode & FMODE_WRITE))
	600	return -EBADF;
	601	if (!file->f_op \|\| (!file->f_op->writev && !file->f_op->write))
	602	return -EINVAL;
	603
	604	return do_readv_writev(WRITE, file, vec, vlen, pos);
	605	}
	606
	607	EXPORT_SYMBOL(vfs_writev);
	608
	609	asmlinkage ssize_t
	610	sys_readv(unsigned long fd, const struct iovec __user *vec, unsigned long vlen)
	611	{
	612	struct file *file;
	613	ssize_t ret = -EBADF;
	614	int fput_needed;
	615
	616	file = fget_light(fd, &fput_needed);
	617	if (file) {
	618	loff_t pos = file_pos_read(file);
	619	ret = vfs_readv(file, vec, vlen, &pos);
	620	file_pos_write(file, pos);
	621	fput_light(file, fput_needed);
	622	}
	623
	624	if (ret > 0)
	625	current->rchar += ret;
	626	current->syscr++;
	627	return ret;
	628	}
	629
	630	asmlinkage ssize_t
	631	sys_writev(unsigned long fd, const struct iovec __user *vec, unsigned long vlen)
	632	{
	633	struct file *file;
	634	ssize_t ret = -EBADF;
	635	int fput_needed;
	636
	637	file = fget_light(fd, &fput_needed);
	638	if (file) {
	639	loff_t pos = file_pos_read(file);
	640	ret = vfs_writev(file, vec, vlen, &pos);
641	file_pos_write(file, pos);
642	fput_light(file, fput_needed);
643	}
644
645	if (ret > 0)
646	current->wchar += ret;
647	current->syscw++;
648	return ret;
649	}
650
651	static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos,
652	size_t count, loff_t max)
653	{
654	struct file * in_file, * out_file;
655	struct inode * in_inode, * out_inode;
656	loff_t pos;
657	ssize_t retval;
658	int fput_needed_in, fput_needed_out;
659
660	/*
661	* Get input file, and verify that it is ok..
662	*/
663	retval = -EBADF;
664	in_file = fget_light(in_fd, &fput_needed_in);
665	if (!in_file)
666	goto out;
667	if (!(in_file->f_mode & FMODE_READ))
668	goto fput_in;
669	retval = -EINVAL;
670	in_inode = in_file->f_dentry->d_inode;
671	if (!in_inode)
672	goto fput_in;
673	if (!in_file->f_op \|\| !in_file->f_op->sendfile)
674	goto fput_in;
675	retval = -ESPIPE;
676	if (!ppos)
677	ppos = &in_file->f_pos;
678	else
679	if (!(in_file->f_mode & FMODE_PREAD))
680	goto fput_in;
681	retval = rw_verify_area(READ, in_file, ppos, count);
e28cc715	682	if (retval < 0)
1da177e4	683	goto fput_in;
e28cc715	684	count = retval;
1da177e4 LT	685
	686	retval = security_file_permission (in_file, MAY_READ);
	687	if (retval)
	688	goto fput_in;
	689
	690	/*
	691	* Get output file, and verify that it is ok..
	692	*/
	693	retval = -EBADF;
	694	out_file = fget_light(out_fd, &fput_needed_out);
	695	if (!out_file)
	696	goto fput_in;
	697	if (!(out_file->f_mode & FMODE_WRITE))
	698	goto fput_out;
	699	retval = -EINVAL;
	700	if (!out_file->f_op \|\| !out_file->f_op->sendpage)
	701	goto fput_out;
	702	out_inode = out_file->f_dentry->d_inode;
	703	retval = rw_verify_area(WRITE, out_file, &out_file->f_pos, count);
e28cc715	704	if (retval < 0)
1da177e4	705	goto fput_out;
e28cc715	706	count = retval;
1da177e4 LT	707
	708	retval = security_file_permission (out_file, MAY_WRITE);
	709	if (retval)
	710	goto fput_out;
	711
	712	if (!max)
	713	max = min(in_inode->i_sb->s_maxbytes, out_inode->i_sb->s_maxbytes);
	714
	715	pos = *ppos;
	716	retval = -EINVAL;
	717	if (unlikely(pos < 0))
	718	goto fput_out;
	719	if (unlikely(pos + count > max)) {
	720	retval = -EOVERFLOW;
	721	if (pos >= max)
	722	goto fput_out;
	723	count = max - pos;
	724	}
	725
	726	retval = in_file->f_op->sendfile(in_file, ppos, count, file_send_actor, out_file);
	727
	728	if (retval > 0) {
	729	current->rchar += retval;
	730	current->wchar += retval;
	731	}
	732	current->syscr++;
	733	current->syscw++;
	734
	735	if (*ppos > max)
	736	retval = -EOVERFLOW;
	737
	738	fput_out:
	739	fput_light(out_file, fput_needed_out);
	740	fput_in:
	741	fput_light(in_file, fput_needed_in);
	742	out:
	743	return retval;
	744	}
	745
	746	asmlinkage ssize_t sys_sendfile(int out_fd, int in_fd, off_t __user *offset, size_t count)
	747	{
	748	loff_t pos;
	749	off_t off;
	750	ssize_t ret;
	751
	752	if (offset) {
	753	if (unlikely(get_user(off, offset)))
	754	return -EFAULT;
	755	pos = off;
	756	ret = do_sendfile(out_fd, in_fd, &pos, count, MAX_NON_LFS);
	757	if (unlikely(put_user(pos, offset)))
	758	return -EFAULT;
	759	return ret;
	760	}
	761
	762	return do_sendfile(out_fd, in_fd, NULL, count, 0);
	763	}
	764
	765	asmlinkage ssize_t sys_sendfile64(int out_fd, int in_fd, loff_t __user *offset, size_t count)
	766	{
	767	loff_t pos;
	768	ssize_t ret;
	769
	770	if (offset) {
771	if (unlikely(copy_from_user(&pos, offset, sizeof(loff_t))))
772	return -EFAULT;
773	ret = do_sendfile(out_fd, in_fd, &pos, count, 0);
774	if (unlikely(put_user(pos, offset)))
775	return -EFAULT;
776	return ret;
777	}
778
779	return do_sendfile(out_fd, in_fd, NULL, count, 0);
780	}