[deliverable/linux.git] / fs / f2fs / file.c

/*
 * fs/f2fs/file.c
 *
 * Copyright (c) 2012 Samsung Electronics Co., Ltd.
 *             http://www.samsung.com/
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 */
#include <linux/fs.h>
#include <linux/f2fs_fs.h>
#include <linux/stat.h>
#include <linux/buffer_head.h>
#include <linux/writeback.h>
#include <linux/blkdev.h>
#include <linux/falloc.h>
#include <linux/types.h>
#include <linux/compat.h>
#include <linux/uaccess.h>
#include <linux/mount.h>

#include "f2fs.h"
#include "node.h"
#include "segment.h"
#include "xattr.h"
#include "acl.h"
#include <trace/events/f2fs.h>

static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma,
						struct vm_fault *vmf)
{
	struct page *page = vmf->page;
	struct inode *inode = file_inode(vma->vm_file);
	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
	block_t old_blk_addr;
	struct dnode_of_data dn;
	int err, ilock;

	f2fs_balance_fs(sbi);

	sb_start_pagefault(inode->i_sb);

	/* block allocation */
	ilock = mutex_lock_op(sbi);
	set_new_dnode(&dn, inode, NULL, NULL, 0);
	err = get_dnode_of_data(&dn, page->index, ALLOC_NODE);
	if (err) {
		mutex_unlock_op(sbi, ilock);
		goto out;
	}

	old_blk_addr = dn.data_blkaddr;

	if (old_blk_addr == NULL_ADDR) {
		err = reserve_new_block(&dn);
		if (err) {
			f2fs_put_dnode(&dn);
			mutex_unlock_op(sbi, ilock);
			goto out;
		}
	}
	f2fs_put_dnode(&dn);
	mutex_unlock_op(sbi, ilock);

	lock_page(page);
	if (page->mapping != inode->i_mapping ||
			page_offset(page) >= i_size_read(inode) ||
			!PageUptodate(page)) {
		unlock_page(page);
		err = -EFAULT;
		goto out;
	}

	/*
	 * check to see if the page is mapped already (no holes)
	 */
	if (PageMappedToDisk(page))
		goto out;

	/* fill the page */
	wait_on_page_writeback(page);

	/* page is wholly or partially inside EOF */
	if (((page->index + 1) << PAGE_CACHE_SHIFT) > i_size_read(inode)) {
		unsigned offset;
		offset = i_size_read(inode) & ~PAGE_CACHE_MASK;
		zero_user_segment(page, offset, PAGE_CACHE_SIZE);
	}
	set_page_dirty(page);
	SetPageUptodate(page);

	file_update_time(vma->vm_file);
out:
	sb_end_pagefault(inode->i_sb);
	return block_page_mkwrite_return(err);
}

static const struct vm_operations_struct f2fs_file_vm_ops = {
	.fault		= filemap_fault,
	.page_mkwrite	= f2fs_vm_page_mkwrite,
	.remap_pages	= generic_file_remap_pages,
};

int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
{
	struct inode *inode = file->f_mapping->host;
	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
	int ret = 0;
	bool need_cp = false;
	struct writeback_control wbc = {
		.sync_mode = WB_SYNC_ALL,
		.nr_to_write = LONG_MAX,
		.for_reclaim = 0,
	};

	if (inode->i_sb->s_flags & MS_RDONLY)
		return 0;

	trace_f2fs_sync_file_enter(inode);
	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
	if (ret) {
		trace_f2fs_sync_file_exit(inode, need_cp, datasync, ret);
		return ret;
	}

	/* guarantee free sections for fsync */
	f2fs_balance_fs(sbi);

	mutex_lock(&inode->i_mutex);

	if (datasync && !(inode->i_state & I_DIRTY_DATASYNC))
		goto out;

	if (!S_ISREG(inode->i_mode) || inode->i_nlink != 1)
		need_cp = true;
	else if (is_cp_file(inode))
		need_cp = true;
	else if (!space_for_roll_forward(sbi))
		need_cp = true;
	else if (!is_checkpointed_node(sbi, F2FS_I(inode)->i_pino))
		need_cp = true;

	if (need_cp) {
		/* all the dirty node pages should be flushed for POR */
		ret = f2fs_sync_fs(inode->i_sb, 1);
	} else {
		/* if there is no written node page, write its inode page */
		while (!sync_node_pages(sbi, inode->i_ino, &wbc)) {
			ret = f2fs_write_inode(inode, NULL);
			if (ret)
				goto out;
		}
		filemap_fdatawait_range(sbi->node_inode->i_mapping,
							0, LONG_MAX);
		ret = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
	}
out:
	mutex_unlock(&inode->i_mutex);
	trace_f2fs_sync_file_exit(inode, need_cp, datasync, ret);
	return ret;
}

static int f2fs_file_mmap(struct file *file, struct vm_area_struct *vma)
{
	file_accessed(file);
	vma->vm_ops = &f2fs_file_vm_ops;
	return 0;
}

static int truncate_data_blocks_range(struct dnode_of_data *dn, int count)
{
	int nr_free = 0, ofs = dn->ofs_in_node;
	struct f2fs_sb_info *sbi = F2FS_SB(dn->inode->i_sb);
	struct f2fs_node *raw_node;
	__le32 *addr;

	raw_node = page_address(dn->node_page);
	addr = blkaddr_in_node(raw_node) + ofs;

	for ( ; count > 0; count--, addr++, dn->ofs_in_node++) {
		block_t blkaddr = le32_to_cpu(*addr);
		if (blkaddr == NULL_ADDR)
			continue;

		update_extent_cache(NULL_ADDR, dn);
		invalidate_blocks(sbi, blkaddr);
		dec_valid_block_count(sbi, dn->inode, 1);
		nr_free++;
	}
	if (nr_free) {
		set_page_dirty(dn->node_page);
		sync_inode_page(dn);
	}
	dn->ofs_in_node = ofs;

	trace_f2fs_truncate_data_blocks_range(dn->inode, dn->nid,
					 dn->ofs_in_node, nr_free);
	return nr_free;
}

void truncate_data_blocks(struct dnode_of_data *dn)
{
	truncate_data_blocks_range(dn, ADDRS_PER_BLOCK);
}

static void truncate_partial_data_page(struct inode *inode, u64 from)
{
	unsigned offset = from & (PAGE_CACHE_SIZE - 1);
	struct page *page;

	if (!offset)
		return;

	page = find_data_page(inode, from >> PAGE_CACHE_SHIFT, false);
	if (IS_ERR(page))
		return;

	lock_page(page);
	wait_on_page_writeback(page);
	zero_user(page, offset, PAGE_CACHE_SIZE - offset);
	set_page_dirty(page);
	f2fs_put_page(page, 1);
}

static int truncate_blocks(struct inode *inode, u64 from)
{
	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
	unsigned int blocksize = inode->i_sb->s_blocksize;
	struct dnode_of_data dn;
	pgoff_t free_from;
	int count = 0, ilock = -1;
	int err;

	trace_f2fs_truncate_blocks_enter(inode, from);

	free_from = (pgoff_t)
			((from + blocksize - 1) >> (sbi->log_blocksize));

	ilock = mutex_lock_op(sbi);
	set_new_dnode(&dn, inode, NULL, NULL, 0);
	err = get_dnode_of_data(&dn, free_from, LOOKUP_NODE);
	if (err) {
		if (err == -ENOENT)
			goto free_next;
		mutex_unlock_op(sbi, ilock);
		trace_f2fs_truncate_blocks_exit(inode, err);
		return err;
	}

	if (IS_INODE(dn.node_page))
		count = ADDRS_PER_INODE;
	else
		count = ADDRS_PER_BLOCK;

	count -= dn.ofs_in_node;
	BUG_ON(count < 0);

	if (dn.ofs_in_node || IS_INODE(dn.node_page)) {
		truncate_data_blocks_range(&dn, count);
		free_from += count;
	}

	f2fs_put_dnode(&dn);
free_next:
	err = truncate_inode_blocks(inode, free_from);
	mutex_unlock_op(sbi, ilock);

	/* lastly zero out the first data page */
	truncate_partial_data_page(inode, from);

	trace_f2fs_truncate_blocks_exit(inode, err);
	return err;
}

void f2fs_truncate(struct inode *inode)
{
	if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) ||
				S_ISLNK(inode->i_mode)))
		return;

	trace_f2fs_truncate(inode);

	if (!truncate_blocks(inode, i_size_read(inode))) {
		inode->i_mtime = inode->i_ctime = CURRENT_TIME;
		mark_inode_dirty(inode);
	}
}

static int f2fs_getattr(struct vfsmount *mnt,
			 struct dentry *dentry, struct kstat *stat)
{
	struct inode *inode = dentry->d_inode;
	generic_fillattr(inode, stat);
	stat->blocks <<= 3;
	return 0;
}

#ifdef CONFIG_F2FS_FS_POSIX_ACL
static void __setattr_copy(struct inode *inode, const struct iattr *attr)
{
	struct f2fs_inode_info *fi = F2FS_I(inode);
	unsigned int ia_valid = attr->ia_valid;

	if (ia_valid & ATTR_UID)
		inode->i_uid = attr->ia_uid;
	if (ia_valid & ATTR_GID)
		inode->i_gid = attr->ia_gid;
	if (ia_valid & ATTR_ATIME)
		inode->i_atime = timespec_trunc(attr->ia_atime,
						inode->i_sb->s_time_gran);
	if (ia_valid & ATTR_MTIME)
		inode->i_mtime = timespec_trunc(attr->ia_mtime,
						inode->i_sb->s_time_gran);
	if (ia_valid & ATTR_CTIME)
		inode->i_ctime = timespec_trunc(attr->ia_ctime,
						inode->i_sb->s_time_gran);
	if (ia_valid & ATTR_MODE) {
		umode_t mode = attr->ia_mode;

		if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID))
			mode &= ~S_ISGID;
		set_acl_inode(fi, mode);
	}
}
#else
#define __setattr_copy setattr_copy
#endif

int f2fs_setattr(struct dentry *dentry, struct iattr *attr)
{
	struct inode *inode = dentry->d_inode;
	struct f2fs_inode_info *fi = F2FS_I(inode);
	int err;

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

	if ((attr->ia_valid & ATTR_SIZE) &&
			attr->ia_size != i_size_read(inode)) {
		truncate_setsize(inode, attr->ia_size);
		f2fs_truncate(inode);
		f2fs_balance_fs(F2FS_SB(inode->i_sb));
	}

	__setattr_copy(inode, attr);

	if (attr->ia_valid & ATTR_MODE) {
		err = f2fs_acl_chmod(inode);
		if (err || is_inode_flag_set(fi, FI_ACL_MODE)) {
			inode->i_mode = fi->i_acl_mode;
			clear_inode_flag(fi, FI_ACL_MODE);
		}
	}

	mark_inode_dirty(inode);
	return err;
}

const struct inode_operations f2fs_file_inode_operations = {
	.getattr	= f2fs_getattr,
	.setattr	= f2fs_setattr,
	.get_acl	= f2fs_get_acl,
#ifdef CONFIG_F2FS_FS_XATTR
	.setxattr	= generic_setxattr,
	.getxattr	= generic_getxattr,
	.listxattr	= f2fs_listxattr,
	.removexattr	= generic_removexattr,
#endif
};

static void fill_zero(struct inode *inode, pgoff_t index,
					loff_t start, loff_t len)
{
	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
	struct page *page;
	int ilock;

	if (!len)
		return;

	f2fs_balance_fs(sbi);

	ilock = mutex_lock_op(sbi);
	page = get_new_data_page(inode, index, false);
	mutex_unlock_op(sbi, ilock);

	if (!IS_ERR(page)) {
		wait_on_page_writeback(page);
		zero_user(page, start, len);
		set_page_dirty(page);
		f2fs_put_page(page, 1);
	}
}

int truncate_hole(struct inode *inode, pgoff_t pg_start, pgoff_t pg_end)
{
	pgoff_t index;
	int err;

	for (index = pg_start; index < pg_end; index++) {
		struct dnode_of_data dn;

		set_new_dnode(&dn, inode, NULL, NULL, 0);
		err = get_dnode_of_data(&dn, index, LOOKUP_NODE);
		if (err) {
			if (err == -ENOENT)
				continue;
			return err;
		}

		if (dn.data_blkaddr != NULL_ADDR)
			truncate_data_blocks_range(&dn, 1);
		f2fs_put_dnode(&dn);
	}
	return 0;
}

static int punch_hole(struct inode *inode, loff_t offset, loff_t len, int mode)
{
	pgoff_t pg_start, pg_end;
	loff_t off_start, off_end;
	int ret = 0;

	pg_start = ((unsigned long long) offset) >> PAGE_CACHE_SHIFT;
	pg_end = ((unsigned long long) offset + len) >> PAGE_CACHE_SHIFT;

	off_start = offset & (PAGE_CACHE_SIZE - 1);
	off_end = (offset + len) & (PAGE_CACHE_SIZE - 1);

	if (pg_start == pg_end) {
		fill_zero(inode, pg_start, off_start,
						off_end - off_start);
	} else {
		if (off_start)
			fill_zero(inode, pg_start++, off_start,
					PAGE_CACHE_SIZE - off_start);
		if (off_end)
			fill_zero(inode, pg_end, 0, off_end);

		if (pg_start < pg_end) {
			struct address_space *mapping = inode->i_mapping;
			loff_t blk_start, blk_end;
			struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
			int ilock;

			f2fs_balance_fs(sbi);

			blk_start = pg_start << PAGE_CACHE_SHIFT;
			blk_end = pg_end << PAGE_CACHE_SHIFT;
			truncate_inode_pages_range(mapping, blk_start,
					blk_end - 1);

			ilock = mutex_lock_op(sbi);
			ret = truncate_hole(inode, pg_start, pg_end);
			mutex_unlock_op(sbi, ilock);
		}
	}

	if (!(mode & FALLOC_FL_KEEP_SIZE) &&
		i_size_read(inode) <= (offset + len)) {
		i_size_write(inode, offset);
		mark_inode_dirty(inode);
	}

	return ret;
}

static int expand_inode_data(struct inode *inode, loff_t offset,
					loff_t len, int mode)
{
	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
	pgoff_t index, pg_start, pg_end;
	loff_t new_size = i_size_read(inode);
	loff_t off_start, off_end;
	int ret = 0;

	ret = inode_newsize_ok(inode, (len + offset));
	if (ret)
		return ret;

	pg_start = ((unsigned long long) offset) >> PAGE_CACHE_SHIFT;
	pg_end = ((unsigned long long) offset + len) >> PAGE_CACHE_SHIFT;

	off_start = offset & (PAGE_CACHE_SIZE - 1);
	off_end = (offset + len) & (PAGE_CACHE_SIZE - 1);

	for (index = pg_start; index <= pg_end; index++) {
		struct dnode_of_data dn;
		int ilock;

		ilock = mutex_lock_op(sbi);
		set_new_dnode(&dn, inode, NULL, NULL, 0);
		ret = get_dnode_of_data(&dn, index, ALLOC_NODE);
		if (ret) {
			mutex_unlock_op(sbi, ilock);
			break;
		}

		if (dn.data_blkaddr == NULL_ADDR) {
			ret = reserve_new_block(&dn);
			if (ret) {
				f2fs_put_dnode(&dn);
				mutex_unlock_op(sbi, ilock);
				break;
			}
		}
		f2fs_put_dnode(&dn);
		mutex_unlock_op(sbi, ilock);

		if (pg_start == pg_end)
			new_size = offset + len;
		else if (index == pg_start && off_start)
			new_size = (index + 1) << PAGE_CACHE_SHIFT;
		else if (index == pg_end)
			new_size = (index << PAGE_CACHE_SHIFT) + off_end;
		else
			new_size += PAGE_CACHE_SIZE;
	}

	if (!(mode & FALLOC_FL_KEEP_SIZE) &&
		i_size_read(inode) < new_size) {
		i_size_write(inode, new_size);
		mark_inode_dirty(inode);
	}

	return ret;
}

static long f2fs_fallocate(struct file *file, int mode,
				loff_t offset, loff_t len)
{
	struct inode *inode = file_inode(file);
	long ret;

	if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE))
		return -EOPNOTSUPP;

	if (mode & FALLOC_FL_PUNCH_HOLE)
		ret = punch_hole(inode, offset, len, mode);
	else
		ret = expand_inode_data(inode, offset, len, mode);

	if (!ret) {
		inode->i_mtime = inode->i_ctime = CURRENT_TIME;
		mark_inode_dirty(inode);
	}
	trace_f2fs_fallocate(inode, mode, offset, len, ret);
	return ret;
}

#define F2FS_REG_FLMASK		(~(FS_DIRSYNC_FL | FS_TOPDIR_FL))
#define F2FS_OTHER_FLMASK	(FS_NODUMP_FL | FS_NOATIME_FL)

static inline __u32 f2fs_mask_flags(umode_t mode, __u32 flags)
{
	if (S_ISDIR(mode))
		return flags;
	else if (S_ISREG(mode))
		return flags & F2FS_REG_FLMASK;
	else
		return flags & F2FS_OTHER_FLMASK;
}

long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
{
	struct inode *inode = file_inode(filp);
	struct f2fs_inode_info *fi = F2FS_I(inode);
	unsigned int flags;
	int ret;

	switch (cmd) {
	case FS_IOC_GETFLAGS:
		flags = fi->i_flags & FS_FL_USER_VISIBLE;
		return put_user(flags, (int __user *) arg);
	case FS_IOC_SETFLAGS:
	{
		unsigned int oldflags;

		ret = mnt_want_write(filp->f_path.mnt);
		if (ret)
			return ret;

		if (!inode_owner_or_capable(inode)) {
			ret = -EACCES;
			goto out;
		}

		if (get_user(flags, (int __user *) arg)) {
			ret = -EFAULT;
			goto out;
		}

		flags = f2fs_mask_flags(inode->i_mode, flags);

		mutex_lock(&inode->i_mutex);

		oldflags = fi->i_flags;

		if ((flags ^ oldflags) & (FS_APPEND_FL | FS_IMMUTABLE_FL)) {
			if (!capable(CAP_LINUX_IMMUTABLE)) {
				mutex_unlock(&inode->i_mutex);
				ret = -EPERM;
				goto out;
			}
		}

		flags = flags & FS_FL_USER_MODIFIABLE;
		flags |= oldflags & ~FS_FL_USER_MODIFIABLE;
		fi->i_flags = flags;
		mutex_unlock(&inode->i_mutex);

		f2fs_set_inode_flags(inode);
		inode->i_ctime = CURRENT_TIME;
		mark_inode_dirty(inode);
out:
		mnt_drop_write(filp->f_path.mnt);
		return ret;
	}
	default:
		return -ENOTTY;
	}
}

#ifdef CONFIG_COMPAT
long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
	switch (cmd) {
	case F2FS_IOC32_GETFLAGS:
		cmd = F2FS_IOC_GETFLAGS;
		break;
	case F2FS_IOC32_SETFLAGS:
		cmd = F2FS_IOC_SETFLAGS;
		break;
	default:
		return -ENOIOCTLCMD;
	}
	return f2fs_ioctl(file, cmd, (unsigned long) compat_ptr(arg));
}
#endif

const struct file_operations f2fs_file_operations = {
	.llseek		= generic_file_llseek,
	.read		= do_sync_read,
	.write		= do_sync_write,
	.aio_read	= generic_file_aio_read,
	.aio_write	= generic_file_aio_write,
	.open		= generic_file_open,
	.mmap		= f2fs_file_mmap,
	.fsync		= f2fs_sync_file,
	.fallocate	= f2fs_fallocate,
	.unlocked_ioctl	= f2fs_ioctl,
#ifdef CONFIG_COMPAT
	.compat_ioctl	= f2fs_compat_ioctl,
#endif
	.splice_read	= generic_file_splice_read,
	.splice_write	= generic_file_splice_write,
};
Commit	Line	Data
0a8165d7	1	/*
fbfa2cc5 JK	2	* fs/f2fs/file.c
	3	*
	4	* Copyright (c) 2012 Samsung Electronics Co., Ltd.
	5	* http://www.samsung.com/
	6	*
	7	* This program is free software; you can redistribute it and/or modify
	8	* it under the terms of the GNU General Public License version 2 as
	9	* published by the Free Software Foundation.
	10	*/
	11	#include <linux/fs.h>
	12	#include <linux/f2fs_fs.h>
	13	#include <linux/stat.h>
	14	#include <linux/buffer_head.h>
	15	#include <linux/writeback.h>
ae51fb31	16	#include <linux/blkdev.h>
fbfa2cc5 JK	17	#include <linux/falloc.h>
fbfa2cc5 JK	18	#include <linux/types.h>
e9750824	19	#include <linux/compat.h>
fbfa2cc5 JK	20	#include <linux/uaccess.h>
	21	#include <linux/mount.h>
	22
	23	#include "f2fs.h"
	24	#include "node.h"
	25	#include "segment.h"
	26	#include "xattr.h"
	27	#include "acl.h"
a2a4a7e4	28	#include <trace/events/f2fs.h>
fbfa2cc5 JK	29
	30	static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma,
	31	struct vm_fault *vmf)
	32	{
	33	struct page *page = vmf->page;
6131ffaa	34	struct inode *inode = file_inode(vma->vm_file);
fbfa2cc5	35	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
fbfa2cc5 JK	36	block_t old_blk_addr;
fbfa2cc5 JK	37	struct dnode_of_data dn;
39936837	38	int err, ilock;
fbfa2cc5 JK	39
	40	f2fs_balance_fs(sbi);
	41
	42	sb_start_pagefault(inode->i_sb);
	43
fbfa2cc5	44	/* block allocation */
39936837	45	ilock = mutex_lock_op(sbi);
fbfa2cc5	46	set_new_dnode(&dn, inode, NULL, NULL, 0);
266e97a8	47	err = get_dnode_of_data(&dn, page->index, ALLOC_NODE);
fbfa2cc5	48	if (err) {
39936837	49	mutex_unlock_op(sbi, ilock);
fbfa2cc5 JK	50	goto out;
	51	}
	52
	53	old_blk_addr = dn.data_blkaddr;
fbfa2cc5 JK	54
	55	if (old_blk_addr == NULL_ADDR) {
	56	err = reserve_new_block(&dn);
	57	if (err) {
	58	f2fs_put_dnode(&dn);
39936837	59	mutex_unlock_op(sbi, ilock);
fbfa2cc5 JK	60	goto out;
	61	}
	62	}
	63	f2fs_put_dnode(&dn);
39936837	64	mutex_unlock_op(sbi, ilock);
fbfa2cc5 JK	65
	66	lock_page(page);
	67	if (page->mapping != inode->i_mapping \|\|
	68	page_offset(page) >= i_size_read(inode) \|\|
	69	!PageUptodate(page)) {
	70	unlock_page(page);
	71	err = -EFAULT;
	72	goto out;
	73	}
	74
	75	/*
	76	* check to see if the page is mapped already (no holes)
	77	*/
	78	if (PageMappedToDisk(page))
	79	goto out;
	80
	81	/* fill the page */
	82	wait_on_page_writeback(page);
	83
	84	/* page is wholly or partially inside EOF */
	85	if (((page->index + 1) << PAGE_CACHE_SHIFT) > i_size_read(inode)) {
	86	unsigned offset;
	87	offset = i_size_read(inode) & ~PAGE_CACHE_MASK;
	88	zero_user_segment(page, offset, PAGE_CACHE_SIZE);
	89	}
	90	set_page_dirty(page);
	91	SetPageUptodate(page);
	92
	93	file_update_time(vma->vm_file);
	94	out:
	95	sb_end_pagefault(inode->i_sb);
	96	return block_page_mkwrite_return(err);
	97	}
	98
	99	static const struct vm_operations_struct f2fs_file_vm_ops = {
692bb55d JK	100	.fault = filemap_fault,
	101	.page_mkwrite = f2fs_vm_page_mkwrite,
	102	.remap_pages = generic_file_remap_pages,
fbfa2cc5 JK	103	};
fbfa2cc5 JK	104
fbfa2cc5 JK	105	int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
	106	{
	107	struct inode *inode = file->f_mapping->host;
	108	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
fbfa2cc5 JK	109	int ret = 0;
	110	bool need_cp = false;
	111	struct writeback_control wbc = {
	112	.sync_mode = WB_SYNC_ALL,
	113	.nr_to_write = LONG_MAX,
	114	.for_reclaim = 0,
	115	};
	116
1fa95b0b NJ	117	if (inode->i_sb->s_flags & MS_RDONLY)
	118	return 0;
	119
a2a4a7e4	120	trace_f2fs_sync_file_enter(inode);
fbfa2cc5	121	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
a2a4a7e4 NJ	122	if (ret) {
a2a4a7e4 NJ	123	trace_f2fs_sync_file_exit(inode, need_cp, datasync, ret);
fbfa2cc5	124	return ret;
a2a4a7e4	125	}
fbfa2cc5	126
7d82db83 JK	127	/* guarantee free sections for fsync */
	128	f2fs_balance_fs(sbi);
	129
fbfa2cc5 JK	130	mutex_lock(&inode->i_mutex);
fbfa2cc5 JK	131
fbfa2cc5 JK	132	if (datasync && !(inode->i_state & I_DIRTY_DATASYNC))
	133	goto out;
	134
fbfa2cc5 JK	135	if (!S_ISREG(inode->i_mode) \|\| inode->i_nlink != 1)
fbfa2cc5 JK	136	need_cp = true;
953a3e27	137	else if (is_cp_file(inode))
fbfa2cc5	138	need_cp = true;
facb0205	139	else if (!space_for_roll_forward(sbi))
fbfa2cc5	140	need_cp = true;
953a3e27	141	else if (!is_checkpointed_node(sbi, F2FS_I(inode)->i_pino))
fbfa2cc5 JK	142	need_cp = true;
fbfa2cc5 JK	143
fbfa2cc5 JK	144	if (need_cp) {
	145	/* all the dirty node pages should be flushed for POR */
	146	ret = f2fs_sync_fs(inode->i_sb, 1);
fbfa2cc5	147	} else {
398b1ac5 JK	148	/* if there is no written node page, write its inode page */
	149	while (!sync_node_pages(sbi, inode->i_ino, &wbc)) {
	150	ret = f2fs_write_inode(inode, NULL);
	151	if (ret)
	152	goto out;
	153	}
fbfa2cc5 JK	154	filemap_fdatawait_range(sbi->node_inode->i_mapping,
fbfa2cc5 JK	155	0, LONG_MAX);
ae51fb31	156	ret = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
fbfa2cc5 JK	157	}
	158	out:
	159	mutex_unlock(&inode->i_mutex);
a2a4a7e4	160	trace_f2fs_sync_file_exit(inode, need_cp, datasync, ret);
fbfa2cc5 JK	161	return ret;
	162	}
	163
	164	static int f2fs_file_mmap(struct file file, struct vm_area_struct vma)
	165	{
	166	file_accessed(file);
	167	vma->vm_ops = &f2fs_file_vm_ops;
	168	return 0;
	169	}
	170
	171	static int truncate_data_blocks_range(struct dnode_of_data *dn, int count)
	172	{
	173	int nr_free = 0, ofs = dn->ofs_in_node;
	174	struct f2fs_sb_info *sbi = F2FS_SB(dn->inode->i_sb);
	175	struct f2fs_node *raw_node;
	176	__le32 *addr;
	177
	178	raw_node = page_address(dn->node_page);
	179	addr = blkaddr_in_node(raw_node) + ofs;
	180
	181	for ( ; count > 0; count--, addr++, dn->ofs_in_node++) {
	182	block_t blkaddr = le32_to_cpu(*addr);
	183	if (blkaddr == NULL_ADDR)
	184	continue;
	185
	186	update_extent_cache(NULL_ADDR, dn);
	187	invalidate_blocks(sbi, blkaddr);
	188	dec_valid_block_count(sbi, dn->inode, 1);
	189	nr_free++;
	190	}
	191	if (nr_free) {
	192	set_page_dirty(dn->node_page);
	193	sync_inode_page(dn);
	194	}
	195	dn->ofs_in_node = ofs;
51dd6249 NJ	196
	197	trace_f2fs_truncate_data_blocks_range(dn->inode, dn->nid,
	198	dn->ofs_in_node, nr_free);
fbfa2cc5 JK	199	return nr_free;
	200	}
	201
	202	void truncate_data_blocks(struct dnode_of_data *dn)
	203	{
	204	truncate_data_blocks_range(dn, ADDRS_PER_BLOCK);
	205	}
	206
	207	static void truncate_partial_data_page(struct inode *inode, u64 from)
	208	{
	209	unsigned offset = from & (PAGE_CACHE_SIZE - 1);
	210	struct page *page;
	211
	212	if (!offset)
	213	return;
	214
c718379b	215	page = find_data_page(inode, from >> PAGE_CACHE_SHIFT, false);
fbfa2cc5 JK	216	if (IS_ERR(page))
	217	return;
	218
	219	lock_page(page);
	220	wait_on_page_writeback(page);
	221	zero_user(page, offset, PAGE_CACHE_SIZE - offset);
	222	set_page_dirty(page);
	223	f2fs_put_page(page, 1);
	224	}
	225
	226	static int truncate_blocks(struct inode *inode, u64 from)
	227	{
	228	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
	229	unsigned int blocksize = inode->i_sb->s_blocksize;
	230	struct dnode_of_data dn;
	231	pgoff_t free_from;
39936837	232	int count = 0, ilock = -1;
fbfa2cc5 JK	233	int err;
fbfa2cc5 JK	234
51dd6249 NJ	235	trace_f2fs_truncate_blocks_enter(inode, from);
51dd6249 NJ	236
fbfa2cc5 JK	237	free_from = (pgoff_t)
	238	((from + blocksize - 1) >> (sbi->log_blocksize));
	239
39936837	240	ilock = mutex_lock_op(sbi);
fbfa2cc5	241	set_new_dnode(&dn, inode, NULL, NULL, 0);
266e97a8	242	err = get_dnode_of_data(&dn, free_from, LOOKUP_NODE);
fbfa2cc5 JK	243	if (err) {
	244	if (err == -ENOENT)
	245	goto free_next;
39936837	246	mutex_unlock_op(sbi, ilock);
51dd6249	247	trace_f2fs_truncate_blocks_exit(inode, err);
fbfa2cc5 JK	248	return err;
	249	}
	250
	251	if (IS_INODE(dn.node_page))
	252	count = ADDRS_PER_INODE;
	253	else
	254	count = ADDRS_PER_BLOCK;
	255
	256	count -= dn.ofs_in_node;
	257	BUG_ON(count < 0);
39936837	258
fbfa2cc5 JK	259	if (dn.ofs_in_node \|\| IS_INODE(dn.node_page)) {
	260	truncate_data_blocks_range(&dn, count);
	261	free_from += count;
	262	}
	263
	264	f2fs_put_dnode(&dn);
	265	free_next:
	266	err = truncate_inode_blocks(inode, free_from);
39936837	267	mutex_unlock_op(sbi, ilock);
fbfa2cc5 JK	268
	269	/* lastly zero out the first data page */
	270	truncate_partial_data_page(inode, from);
	271
51dd6249	272	trace_f2fs_truncate_blocks_exit(inode, err);
fbfa2cc5 JK	273	return err;
	274	}
	275
	276	void f2fs_truncate(struct inode *inode)
	277	{
	278	if (!(S_ISREG(inode->i_mode) \|\| S_ISDIR(inode->i_mode) \|\|
	279	S_ISLNK(inode->i_mode)))
	280	return;
	281
51dd6249 NJ	282	trace_f2fs_truncate(inode);
51dd6249 NJ	283
fbfa2cc5 JK	284	if (!truncate_blocks(inode, i_size_read(inode))) {
	285	inode->i_mtime = inode->i_ctime = CURRENT_TIME;
	286	mark_inode_dirty(inode);
	287	}
fbfa2cc5 JK	288	}
	289
	290	static int f2fs_getattr(struct vfsmount *mnt,
	291	struct dentry dentry, struct kstat stat)
	292	{
	293	struct inode *inode = dentry->d_inode;
	294	generic_fillattr(inode, stat);
	295	stat->blocks <<= 3;
	296	return 0;
	297	}
	298
	299	#ifdef CONFIG_F2FS_FS_POSIX_ACL
	300	static void __setattr_copy(struct inode inode, const struct iattr attr)
	301	{
	302	struct f2fs_inode_info *fi = F2FS_I(inode);
	303	unsigned int ia_valid = attr->ia_valid;
	304
	305	if (ia_valid & ATTR_UID)
	306	inode->i_uid = attr->ia_uid;
	307	if (ia_valid & ATTR_GID)
	308	inode->i_gid = attr->ia_gid;
	309	if (ia_valid & ATTR_ATIME)
	310	inode->i_atime = timespec_trunc(attr->ia_atime,
	311	inode->i_sb->s_time_gran);
	312	if (ia_valid & ATTR_MTIME)
	313	inode->i_mtime = timespec_trunc(attr->ia_mtime,
	314	inode->i_sb->s_time_gran);
	315	if (ia_valid & ATTR_CTIME)
	316	inode->i_ctime = timespec_trunc(attr->ia_ctime,
	317	inode->i_sb->s_time_gran);
	318	if (ia_valid & ATTR_MODE) {
	319	umode_t mode = attr->ia_mode;
	320
	321	if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID))
	322	mode &= ~S_ISGID;
	323	set_acl_inode(fi, mode);
	324	}
	325	}
	326	#else
	327	#define __setattr_copy setattr_copy
	328	#endif
	329
	330	int f2fs_setattr(struct dentry dentry, struct iattr attr)
	331	{
	332	struct inode *inode = dentry->d_inode;
	333	struct f2fs_inode_info *fi = F2FS_I(inode);
	334	int err;
	335
	336	err = inode_change_ok(inode, attr);
	337	if (err)
	338	return err;
	339
	340	if ((attr->ia_valid & ATTR_SIZE) &&
	341	attr->ia_size != i_size_read(inode)) {
	342	truncate_setsize(inode, attr->ia_size);
	343	f2fs_truncate(inode);
d4686d56	344	f2fs_balance_fs(F2FS_SB(inode->i_sb));
fbfa2cc5 JK	345	}
	346
	347	__setattr_copy(inode, attr);
	348
	349	if (attr->ia_valid & ATTR_MODE) {
	350	err = f2fs_acl_chmod(inode);
	351	if (err \|\| is_inode_flag_set(fi, FI_ACL_MODE)) {
	352	inode->i_mode = fi->i_acl_mode;
	353	clear_inode_flag(fi, FI_ACL_MODE);
	354	}
	355	}
	356
	357	mark_inode_dirty(inode);
	358	return err;
	359	}
	360
	361	const struct inode_operations f2fs_file_inode_operations = {
	362	.getattr = f2fs_getattr,
	363	.setattr = f2fs_setattr,
	364	.get_acl = f2fs_get_acl,
	365	#ifdef CONFIG_F2FS_FS_XATTR
	366	.setxattr = generic_setxattr,
	367	.getxattr = generic_getxattr,
	368	.listxattr = f2fs_listxattr,
	369	.removexattr = generic_removexattr,
	370	#endif
	371	};
	372
	373	static void fill_zero(struct inode *inode, pgoff_t index,
	374	loff_t start, loff_t len)
	375	{
bd43df02	376	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
fbfa2cc5	377	struct page *page;
39936837	378	int ilock;
fbfa2cc5 JK	379
	380	if (!len)
	381	return;
	382
bd43df02 JK	383	f2fs_balance_fs(sbi);
bd43df02 JK	384
39936837	385	ilock = mutex_lock_op(sbi);
fbfa2cc5	386	page = get_new_data_page(inode, index, false);
39936837	387	mutex_unlock_op(sbi, ilock);
fbfa2cc5 JK	388
	389	if (!IS_ERR(page)) {
	390	wait_on_page_writeback(page);
	391	zero_user(page, start, len);
	392	set_page_dirty(page);
	393	f2fs_put_page(page, 1);
	394	}
	395	}
	396
	397	int truncate_hole(struct inode *inode, pgoff_t pg_start, pgoff_t pg_end)
	398	{
	399	pgoff_t index;
	400	int err;
	401
	402	for (index = pg_start; index < pg_end; index++) {
	403	struct dnode_of_data dn;
fbfa2cc5	404
fbfa2cc5	405	set_new_dnode(&dn, inode, NULL, NULL, 0);
266e97a8	406	err = get_dnode_of_data(&dn, index, LOOKUP_NODE);
fbfa2cc5	407	if (err) {
fbfa2cc5 JK	408	if (err == -ENOENT)
	409	continue;
	410	return err;
	411	}
	412
	413	if (dn.data_blkaddr != NULL_ADDR)
	414	truncate_data_blocks_range(&dn, 1);
	415	f2fs_put_dnode(&dn);
fbfa2cc5 JK	416	}
	417	return 0;
	418	}
	419
	420	static int punch_hole(struct inode *inode, loff_t offset, loff_t len, int mode)
	421	{
	422	pgoff_t pg_start, pg_end;
	423	loff_t off_start, off_end;
	424	int ret = 0;
	425
	426	pg_start = ((unsigned long long) offset) >> PAGE_CACHE_SHIFT;
	427	pg_end = ((unsigned long long) offset + len) >> PAGE_CACHE_SHIFT;
	428
	429	off_start = offset & (PAGE_CACHE_SIZE - 1);
	430	off_end = (offset + len) & (PAGE_CACHE_SIZE - 1);
	431
	432	if (pg_start == pg_end) {
	433	fill_zero(inode, pg_start, off_start,
	434	off_end - off_start);
	435	} else {
	436	if (off_start)
	437	fill_zero(inode, pg_start++, off_start,
	438	PAGE_CACHE_SIZE - off_start);
	439	if (off_end)
	440	fill_zero(inode, pg_end, 0, off_end);
	441
	442	if (pg_start < pg_end) {
	443	struct address_space *mapping = inode->i_mapping;
	444	loff_t blk_start, blk_end;
1127a3d4	445	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
39936837	446	int ilock;
1127a3d4 JH	447
1127a3d4 JH	448	f2fs_balance_fs(sbi);
fbfa2cc5 JK	449
	450	blk_start = pg_start << PAGE_CACHE_SHIFT;
	451	blk_end = pg_end << PAGE_CACHE_SHIFT;
	452	truncate_inode_pages_range(mapping, blk_start,
	453	blk_end - 1);
39936837 JK	454
39936837 JK	455	ilock = mutex_lock_op(sbi);
fbfa2cc5	456	ret = truncate_hole(inode, pg_start, pg_end);
39936837	457	mutex_unlock_op(sbi, ilock);
fbfa2cc5 JK	458	}
	459	}
	460
	461	if (!(mode & FALLOC_FL_KEEP_SIZE) &&
	462	i_size_read(inode) <= (offset + len)) {
	463	i_size_write(inode, offset);
	464	mark_inode_dirty(inode);
	465	}
	466
	467	return ret;
	468	}
	469
	470	static int expand_inode_data(struct inode *inode, loff_t offset,
	471	loff_t len, int mode)
	472	{
	473	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
	474	pgoff_t index, pg_start, pg_end;
	475	loff_t new_size = i_size_read(inode);
	476	loff_t off_start, off_end;
	477	int ret = 0;
	478
	479	ret = inode_newsize_ok(inode, (len + offset));
	480	if (ret)
	481	return ret;
	482
	483	pg_start = ((unsigned long long) offset) >> PAGE_CACHE_SHIFT;
	484	pg_end = ((unsigned long long) offset + len) >> PAGE_CACHE_SHIFT;
	485
	486	off_start = offset & (PAGE_CACHE_SIZE - 1);
	487	off_end = (offset + len) & (PAGE_CACHE_SIZE - 1);
	488
	489	for (index = pg_start; index <= pg_end; index++) {
	490	struct dnode_of_data dn;
39936837	491	int ilock;
fbfa2cc5	492
39936837	493	ilock = mutex_lock_op(sbi);
fbfa2cc5	494	set_new_dnode(&dn, inode, NULL, NULL, 0);
266e97a8	495	ret = get_dnode_of_data(&dn, index, ALLOC_NODE);
fbfa2cc5	496	if (ret) {
39936837	497	mutex_unlock_op(sbi, ilock);
fbfa2cc5 JK	498	break;
	499	}
	500
	501	if (dn.data_blkaddr == NULL_ADDR) {
	502	ret = reserve_new_block(&dn);
	503	if (ret) {
	504	f2fs_put_dnode(&dn);
39936837	505	mutex_unlock_op(sbi, ilock);
fbfa2cc5 JK	506	break;
	507	}
	508	}
	509	f2fs_put_dnode(&dn);
39936837	510	mutex_unlock_op(sbi, ilock);
fbfa2cc5 JK	511
	512	if (pg_start == pg_end)
	513	new_size = offset + len;
	514	else if (index == pg_start && off_start)
	515	new_size = (index + 1) << PAGE_CACHE_SHIFT;
	516	else if (index == pg_end)
	517	new_size = (index << PAGE_CACHE_SHIFT) + off_end;
	518	else
	519	new_size += PAGE_CACHE_SIZE;
	520	}
	521
	522	if (!(mode & FALLOC_FL_KEEP_SIZE) &&
	523	i_size_read(inode) < new_size) {
	524	i_size_write(inode, new_size);
	525	mark_inode_dirty(inode);
	526	}
	527
	528	return ret;
	529	}
	530
	531	static long f2fs_fallocate(struct file *file, int mode,
	532	loff_t offset, loff_t len)
	533	{
6131ffaa	534	struct inode *inode = file_inode(file);
fbfa2cc5 JK	535	long ret;
	536
	537	if (mode & ~(FALLOC_FL_KEEP_SIZE \| FALLOC_FL_PUNCH_HOLE))
	538	return -EOPNOTSUPP;
	539
	540	if (mode & FALLOC_FL_PUNCH_HOLE)
	541	ret = punch_hole(inode, offset, len, mode);
	542	else
	543	ret = expand_inode_data(inode, offset, len, mode);
	544
3af60a49 NJ	545	if (!ret) {
	546	inode->i_mtime = inode->i_ctime = CURRENT_TIME;
	547	mark_inode_dirty(inode);
	548	}
c01e2853	549	trace_f2fs_fallocate(inode, mode, offset, len, ret);
fbfa2cc5 JK	550	return ret;
	551	}
	552
	553	#define F2FS_REG_FLMASK (~(FS_DIRSYNC_FL \| FS_TOPDIR_FL))
	554	#define F2FS_OTHER_FLMASK (FS_NODUMP_FL \| FS_NOATIME_FL)
	555
	556	static inline __u32 f2fs_mask_flags(umode_t mode, __u32 flags)
	557	{
	558	if (S_ISDIR(mode))
	559	return flags;
	560	else if (S_ISREG(mode))
	561	return flags & F2FS_REG_FLMASK;
	562	else
	563	return flags & F2FS_OTHER_FLMASK;
	564	}
	565
	566	long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
	567	{
6131ffaa	568	struct inode *inode = file_inode(filp);
fbfa2cc5 JK	569	struct f2fs_inode_info *fi = F2FS_I(inode);
	570	unsigned int flags;
	571	int ret;
	572
	573	switch (cmd) {
	574	case FS_IOC_GETFLAGS:
	575	flags = fi->i_flags & FS_FL_USER_VISIBLE;
	576	return put_user(flags, (int __user *) arg);
	577	case FS_IOC_SETFLAGS:
	578	{
	579	unsigned int oldflags;
	580
	581	ret = mnt_want_write(filp->f_path.mnt);
	582	if (ret)
	583	return ret;
	584
	585	if (!inode_owner_or_capable(inode)) {
	586	ret = -EACCES;
	587	goto out;
	588	}
	589
	590	if (get_user(flags, (int __user *) arg)) {
	591	ret = -EFAULT;
	592	goto out;
	593	}
	594
	595	flags = f2fs_mask_flags(inode->i_mode, flags);
	596
	597	mutex_lock(&inode->i_mutex);
	598
	599	oldflags = fi->i_flags;
	600
	601	if ((flags ^ oldflags) & (FS_APPEND_FL \| FS_IMMUTABLE_FL)) {
	602	if (!capable(CAP_LINUX_IMMUTABLE)) {
	603	mutex_unlock(&inode->i_mutex);
	604	ret = -EPERM;
	605	goto out;
	606	}
	607	}
	608
	609	flags = flags & FS_FL_USER_MODIFIABLE;
	610	flags \|= oldflags & ~FS_FL_USER_MODIFIABLE;
	611	fi->i_flags = flags;
	612	mutex_unlock(&inode->i_mutex);
	613
	614	f2fs_set_inode_flags(inode);
	615	inode->i_ctime = CURRENT_TIME;
	616	mark_inode_dirty(inode);
	617	out:
	618	mnt_drop_write(filp->f_path.mnt);
	619	return ret;
	620	}
	621	default:
	622	return -ENOTTY;
	623	}
	624	}
	625
e9750824 NJ	626	#ifdef CONFIG_COMPAT
	627	long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
	628	{
	629	switch (cmd) {
	630	case F2FS_IOC32_GETFLAGS:
	631	cmd = F2FS_IOC_GETFLAGS;
	632	break;
	633	case F2FS_IOC32_SETFLAGS:
	634	cmd = F2FS_IOC_SETFLAGS;
	635	break;
	636	default:
	637	return -ENOIOCTLCMD;
	638	}
	639	return f2fs_ioctl(file, cmd, (unsigned long) compat_ptr(arg));
	640	}
	641	#endif
	642
fbfa2cc5 JK	643	const struct file_operations f2fs_file_operations = {
	644	.llseek = generic_file_llseek,
	645	.read = do_sync_read,
	646	.write = do_sync_write,
	647	.aio_read = generic_file_aio_read,
	648	.aio_write = generic_file_aio_write,
	649	.open = generic_file_open,
	650	.mmap = f2fs_file_mmap,
	651	.fsync = f2fs_sync_file,
	652	.fallocate = f2fs_fallocate,
	653	.unlocked_ioctl = f2fs_ioctl,
e9750824 NJ	654	#ifdef CONFIG_COMPAT
	655	.compat_ioctl = f2fs_compat_ioctl,
	656	#endif
fbfa2cc5 JK	657	.splice_read = generic_file_splice_read,
	658	.splice_write = generic_file_splice_write,
	659	};