file.c 20.1 KB
Newer Older
J
Jaegeuk Kim 已提交
1
/*
J
Jaegeuk Kim 已提交
2 3 4 5 6 7 8 9 10 11 12 13 14 15
 * 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>
16
#include <linux/blkdev.h>
J
Jaegeuk Kim 已提交
17 18
#include <linux/falloc.h>
#include <linux/types.h>
19
#include <linux/compat.h>
J
Jaegeuk Kim 已提交
20 21
#include <linux/uaccess.h>
#include <linux/mount.h>
22
#include <linux/pagevec.h>
J
Jaegeuk Kim 已提交
23 24 25 26 27 28

#include "f2fs.h"
#include "node.h"
#include "segment.h"
#include "xattr.h"
#include "acl.h"
29
#include <trace/events/f2fs.h>
J
Jaegeuk Kim 已提交
30 31 32 33 34

static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma,
						struct vm_fault *vmf)
{
	struct page *page = vmf->page;
A
Al Viro 已提交
35
	struct inode *inode = file_inode(vma->vm_file);
36
	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
J
Jaegeuk Kim 已提交
37
	struct dnode_of_data dn;
38
	int err;
J
Jaegeuk Kim 已提交
39 40 41 42 43

	f2fs_balance_fs(sbi);

	sb_start_pagefault(inode->i_sb);

44 45 46 47 48
	/* force to convert with normal data indices */
	err = f2fs_convert_inline_data(inode, MAX_INLINE_DATA + 1, page);
	if (err)
		goto out;

J
Jaegeuk Kim 已提交
49
	/* block allocation */
50
	f2fs_lock_op(sbi);
J
Jaegeuk Kim 已提交
51
	set_new_dnode(&dn, inode, NULL, NULL, 0);
52
	err = f2fs_reserve_block(&dn, page->index);
53
	f2fs_unlock_op(sbi);
54 55
	if (err)
		goto out;
J
Jaegeuk Kim 已提交
56

57
	file_update_time(vma->vm_file);
J
Jaegeuk Kim 已提交
58
	lock_page(page);
59
	if (unlikely(page->mapping != inode->i_mapping ||
60
			page_offset(page) > i_size_read(inode) ||
61
			!PageUptodate(page))) {
J
Jaegeuk Kim 已提交
62 63 64 65 66 67 68 69 70
		unlock_page(page);
		err = -EFAULT;
		goto out;
	}

	/*
	 * check to see if the page is mapped already (no holes)
	 */
	if (PageMappedToDisk(page))
71
		goto mapped;
J
Jaegeuk Kim 已提交
72 73 74 75 76 77 78 79 80 81

	/* 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);

82
	trace_f2fs_vm_page_mkwrite(page, DATA);
83 84
mapped:
	/* fill the page */
85
	f2fs_wait_on_page_writeback(page, DATA);
J
Jaegeuk Kim 已提交
86 87 88 89 90 91
out:
	sb_end_pagefault(inode->i_sb);
	return block_page_mkwrite_return(err);
}

static const struct vm_operations_struct f2fs_file_vm_ops = {
92
	.fault		= filemap_fault,
93
	.map_pages	= filemap_map_pages,
94 95
	.page_mkwrite	= f2fs_vm_page_mkwrite,
	.remap_pages	= generic_file_remap_pages,
J
Jaegeuk Kim 已提交
96 97
};

98 99 100 101 102 103 104 105 106 107
static int get_parent_ino(struct inode *inode, nid_t *pino)
{
	struct dentry *dentry;

	inode = igrab(inode);
	dentry = d_find_any_alias(inode);
	iput(inode);
	if (!dentry)
		return 0;

108 109 110 111
	if (update_dent_inode(inode, &dentry->d_name)) {
		dput(dentry);
		return 0;
	}
112

113 114
	*pino = parent_ino(dentry);
	dput(dentry);
115 116 117
	return 1;
}

118 119
static inline bool need_do_checkpoint(struct inode *inode)
{
120
	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136
	bool need_cp = false;

	if (!S_ISREG(inode->i_mode) || inode->i_nlink != 1)
		need_cp = true;
	else if (file_wrong_pino(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;
	else if (F2FS_I(inode)->xattr_ver == cur_cp_version(F2FS_CKPT(sbi)))
		need_cp = true;

	return need_cp;
}

J
Jaegeuk Kim 已提交
137 138 139
int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
{
	struct inode *inode = file->f_mapping->host;
140
	struct f2fs_inode_info *fi = F2FS_I(inode);
141
	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
J
Jaegeuk Kim 已提交
142 143 144
	int ret = 0;
	bool need_cp = false;
	struct writeback_control wbc = {
145
		.sync_mode = WB_SYNC_ALL,
J
Jaegeuk Kim 已提交
146 147 148 149
		.nr_to_write = LONG_MAX,
		.for_reclaim = 0,
	};

150
	if (unlikely(f2fs_readonly(inode->i_sb)))
151 152
		return 0;

153
	trace_f2fs_sync_file_enter(inode);
154 155 156 157 158

	/* if fdatasync is triggered, let's do in-place-update */
	if (datasync)
		set_inode_flag(fi, FI_NEED_IPU);

J
Jaegeuk Kim 已提交
159
	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
160 161
	if (datasync)
		clear_inode_flag(fi, FI_NEED_IPU);
162 163
	if (ret) {
		trace_f2fs_sync_file_exit(inode, need_cp, datasync, ret);
J
Jaegeuk Kim 已提交
164
		return ret;
165
	}
J
Jaegeuk Kim 已提交
166

167 168 169 170 171 172 173 174 175 176 177
	/*
	 * if there is no written data, don't waste time to write recovery info.
	 */
	if (!is_inode_flag_set(fi, FI_APPEND_WRITE) &&
		!exist_written_data(sbi, inode->i_ino, APPEND_INO)) {
		if (is_inode_flag_set(fi, FI_UPDATE_WRITE) ||
			exist_written_data(sbi, inode->i_ino, UPDATE_INO))
			goto flush_out;
		goto out;
	}

178 179 180
	/* guarantee free sections for fsync */
	f2fs_balance_fs(sbi);

181 182 183 184
	/*
	 * Both of fdatasync() and fsync() are able to be recovered from
	 * sudden-power-off.
	 */
185 186
	down_read(&fi->i_sem);
	need_cp = need_do_checkpoint(inode);
187 188
	up_read(&fi->i_sem);

J
Jaegeuk Kim 已提交
189
	if (need_cp) {
190 191
		nid_t pino;

J
Jaegeuk Kim 已提交
192 193
		/* all the dirty node pages should be flushed for POR */
		ret = f2fs_sync_fs(inode->i_sb, 1);
194 195 196

		down_write(&fi->i_sem);
		F2FS_I(inode)->xattr_ver = 0;
197 198 199 200
		if (file_wrong_pino(inode) && inode->i_nlink == 1 &&
					get_parent_ino(inode, &pino)) {
			F2FS_I(inode)->i_pino = pino;
			file_got_pino(inode);
201
			up_write(&fi->i_sem);
202 203 204 205
			mark_inode_dirty_sync(inode);
			ret = f2fs_write_inode(inode, NULL);
			if (ret)
				goto out;
206 207
		} else {
			up_write(&fi->i_sem);
208
		}
J
Jaegeuk Kim 已提交
209
	} else {
210 211
		/* if there is no written node page, write its inode page */
		while (!sync_node_pages(sbi, inode->i_ino, &wbc)) {
212 213
			if (fsync_mark_done(sbi, inode->i_ino))
				goto out;
214
			mark_inode_dirty_sync(inode);
215 216 217 218
			ret = f2fs_write_inode(inode, NULL);
			if (ret)
				goto out;
		}
219 220 221
		ret = wait_on_node_pages_writeback(sbi, inode->i_ino);
		if (ret)
			goto out;
222 223 224 225 226 227 228

		/* once recovery info is written, don't need to tack this */
		remove_dirty_inode(sbi, inode->i_ino, APPEND_INO);
		clear_inode_flag(fi, FI_APPEND_WRITE);
flush_out:
		remove_dirty_inode(sbi, inode->i_ino, UPDATE_INO);
		clear_inode_flag(fi, FI_UPDATE_WRITE);
229
		ret = f2fs_issue_flush(F2FS_I_SB(inode));
J
Jaegeuk Kim 已提交
230 231
	}
out:
232
	trace_f2fs_sync_file_exit(inode, need_cp, datasync, ret);
J
Jaegeuk Kim 已提交
233 234 235
	return ret;
}

236 237 238 239 240 241 242 243 244 245 246
static pgoff_t __get_first_dirty_index(struct address_space *mapping,
						pgoff_t pgofs, int whence)
{
	struct pagevec pvec;
	int nr_pages;

	if (whence != SEEK_DATA)
		return 0;

	/* find first dirty page index */
	pagevec_init(&pvec, 0);
J
Jaegeuk Kim 已提交
247 248 249
	nr_pages = pagevec_lookup_tag(&pvec, mapping, &pgofs,
					PAGECACHE_TAG_DIRTY, 1);
	pgofs = nr_pages ? pvec.pages[0]->index : LONG_MAX;
250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270
	pagevec_release(&pvec);
	return pgofs;
}

static bool __found_offset(block_t blkaddr, pgoff_t dirty, pgoff_t pgofs,
							int whence)
{
	switch (whence) {
	case SEEK_DATA:
		if ((blkaddr == NEW_ADDR && dirty == pgofs) ||
			(blkaddr != NEW_ADDR && blkaddr != NULL_ADDR))
			return true;
		break;
	case SEEK_HOLE:
		if (blkaddr == NULL_ADDR)
			return true;
		break;
	}
	return false;
}

271 272 273 274 275
static loff_t f2fs_seek_block(struct file *file, loff_t offset, int whence)
{
	struct inode *inode = file->f_mapping->host;
	loff_t maxbytes = inode->i_sb->s_maxbytes;
	struct dnode_of_data dn;
276 277 278
	pgoff_t pgofs, end_offset, dirty;
	loff_t data_ofs = offset;
	loff_t isize;
279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295
	int err = 0;

	mutex_lock(&inode->i_mutex);

	isize = i_size_read(inode);
	if (offset >= isize)
		goto fail;

	/* handle inline data case */
	if (f2fs_has_inline_data(inode)) {
		if (whence == SEEK_HOLE)
			data_ofs = isize;
		goto found;
	}

	pgofs = (pgoff_t)(offset >> PAGE_CACHE_SHIFT);

296 297
	dirty = __get_first_dirty_index(inode->i_mapping, pgofs, whence);

298 299 300 301 302 303
	for (; data_ofs < isize; data_ofs = pgofs << PAGE_CACHE_SHIFT) {
		set_new_dnode(&dn, inode, NULL, NULL, 0);
		err = get_dnode_of_data(&dn, pgofs, LOOKUP_NODE_RA);
		if (err && err != -ENOENT) {
			goto fail;
		} else if (err == -ENOENT) {
A
arter97 已提交
304
			/* direct node does not exists */
305 306 307 308 309 310 311 312 313
			if (whence == SEEK_DATA) {
				pgofs = PGOFS_OF_NEXT_DNODE(pgofs,
							F2FS_I(inode));
				continue;
			} else {
				goto found;
			}
		}

314
		end_offset = ADDRS_PER_PAGE(dn.node_page, F2FS_I(inode));
315 316 317 318 319 320 321 322

		/* find data/hole in dnode block */
		for (; dn.ofs_in_node < end_offset;
				dn.ofs_in_node++, pgofs++,
				data_ofs = pgofs << PAGE_CACHE_SHIFT) {
			block_t blkaddr;
			blkaddr = datablock_addr(dn.node_page, dn.ofs_in_node);

323
			if (__found_offset(blkaddr, dirty, pgofs, whence)) {
324 325 326 327 328 329 330 331 332 333
				f2fs_put_dnode(&dn);
				goto found;
			}
		}
		f2fs_put_dnode(&dn);
	}

	if (whence == SEEK_DATA)
		goto fail;
found:
334 335
	if (whence == SEEK_HOLE && data_ofs > isize)
		data_ofs = isize;
336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355
	mutex_unlock(&inode->i_mutex);
	return vfs_setpos(file, data_ofs, maxbytes);
fail:
	mutex_unlock(&inode->i_mutex);
	return -ENXIO;
}

static loff_t f2fs_llseek(struct file *file, loff_t offset, int whence)
{
	struct inode *inode = file->f_mapping->host;
	loff_t maxbytes = inode->i_sb->s_maxbytes;

	switch (whence) {
	case SEEK_SET:
	case SEEK_CUR:
	case SEEK_END:
		return generic_file_llseek_size(file, offset, whence,
						maxbytes, i_size_read(inode));
	case SEEK_DATA:
	case SEEK_HOLE:
356 357
		if (offset < 0)
			return -ENXIO;
358 359 360 361 362 363
		return f2fs_seek_block(file, offset, whence);
	}

	return -EINVAL;
}

J
Jaegeuk Kim 已提交
364 365 366 367 368 369 370
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;
}

371
int truncate_data_blocks_range(struct dnode_of_data *dn, int count)
J
Jaegeuk Kim 已提交
372 373
{
	int nr_free = 0, ofs = dn->ofs_in_node;
374
	struct f2fs_sb_info *sbi = F2FS_I_SB(dn->inode);
J
Jaegeuk Kim 已提交
375 376 377
	struct f2fs_node *raw_node;
	__le32 *addr;

378
	raw_node = F2FS_NODE(dn->node_page);
J
Jaegeuk Kim 已提交
379 380
	addr = blkaddr_in_node(raw_node) + ofs;

C
Chris Fries 已提交
381
	for (; count > 0; count--, addr++, dn->ofs_in_node++) {
J
Jaegeuk Kim 已提交
382 383 384 385 386 387 388 389 390
		block_t blkaddr = le32_to_cpu(*addr);
		if (blkaddr == NULL_ADDR)
			continue;

		update_extent_cache(NULL_ADDR, dn);
		invalidate_blocks(sbi, blkaddr);
		nr_free++;
	}
	if (nr_free) {
391
		dec_valid_block_count(sbi, dn->inode, nr_free);
J
Jaegeuk Kim 已提交
392 393 394 395
		set_page_dirty(dn->node_page);
		sync_inode_page(dn);
	}
	dn->ofs_in_node = ofs;
396 397 398

	trace_f2fs_truncate_data_blocks_range(dn->inode, dn->nid,
					 dn->ofs_in_node, nr_free);
J
Jaegeuk Kim 已提交
399 400 401 402 403 404 405 406 407 408 409 410 411
	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;

412 413 414
	if (f2fs_has_inline_data(inode))
		return truncate_inline_data(inode, from);

J
Jaegeuk Kim 已提交
415 416 417
	if (!offset)
		return;

418
	page = find_data_page(inode, from >> PAGE_CACHE_SHIFT, false);
J
Jaegeuk Kim 已提交
419 420 421 422
	if (IS_ERR(page))
		return;

	lock_page(page);
423 424 425 426
	if (unlikely(!PageUptodate(page) ||
			page->mapping != inode->i_mapping))
		goto out;

427
	f2fs_wait_on_page_writeback(page, DATA);
J
Jaegeuk Kim 已提交
428 429
	zero_user(page, offset, PAGE_CACHE_SIZE - offset);
	set_page_dirty(page);
430 431

out:
J
Jaegeuk Kim 已提交
432 433 434
	f2fs_put_page(page, 1);
}

435
int truncate_blocks(struct inode *inode, u64 from, bool lock)
J
Jaegeuk Kim 已提交
436
{
437
	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
J
Jaegeuk Kim 已提交
438 439 440
	unsigned int blocksize = inode->i_sb->s_blocksize;
	struct dnode_of_data dn;
	pgoff_t free_from;
H
Huajun Li 已提交
441
	int count = 0, err = 0;
J
Jaegeuk Kim 已提交
442

443 444
	trace_f2fs_truncate_blocks_enter(inode, from);

445 446 447
	if (f2fs_has_inline_data(inode))
		goto done;

J
Jaegeuk Kim 已提交
448 449 450
	free_from = (pgoff_t)
			((from + blocksize - 1) >> (sbi->log_blocksize));

451 452
	if (lock)
		f2fs_lock_op(sbi);
H
Huajun Li 已提交
453

J
Jaegeuk Kim 已提交
454
	set_new_dnode(&dn, inode, NULL, NULL, 0);
455
	err = get_dnode_of_data(&dn, free_from, LOOKUP_NODE);
J
Jaegeuk Kim 已提交
456 457 458
	if (err) {
		if (err == -ENOENT)
			goto free_next;
459 460
		if (lock)
			f2fs_unlock_op(sbi);
461
		trace_f2fs_truncate_blocks_exit(inode, err);
J
Jaegeuk Kim 已提交
462 463 464
		return err;
	}

465
	count = ADDRS_PER_PAGE(dn.node_page, F2FS_I(inode));
J
Jaegeuk Kim 已提交
466 467

	count -= dn.ofs_in_node;
468
	f2fs_bug_on(sbi, count < 0);
469

J
Jaegeuk Kim 已提交
470 471 472 473 474 475 476 477
	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);
478 479
	if (lock)
		f2fs_unlock_op(sbi);
480
done:
J
Jaegeuk Kim 已提交
481 482 483
	/* lastly zero out the first data page */
	truncate_partial_data_page(inode, from);

484
	trace_f2fs_truncate_blocks_exit(inode, err);
J
Jaegeuk Kim 已提交
485 486 487 488 489 490 491 492 493
	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;

494 495
	trace_f2fs_truncate(inode);

496
	if (!truncate_blocks(inode, i_size_read(inode), true)) {
J
Jaegeuk Kim 已提交
497 498 499 500 501
		inode->i_mtime = inode->i_ctime = CURRENT_TIME;
		mark_inode_dirty(inode);
	}
}

502
int f2fs_getattr(struct vfsmount *mnt,
J
Jaegeuk Kim 已提交
503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553
			 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)) {
554
		err = f2fs_convert_inline_data(inode, attr->ia_size, NULL);
555 556
		if (err)
			return err;
H
Huajun Li 已提交
557

J
Jaegeuk Kim 已提交
558
		truncate_setsize(inode, attr->ia_size);
559
		f2fs_truncate(inode);
560
		f2fs_balance_fs(F2FS_I_SB(inode));
J
Jaegeuk Kim 已提交
561 562 563 564 565
	}

	__setattr_copy(inode, attr);

	if (attr->ia_valid & ATTR_MODE) {
566
		err = posix_acl_chmod(inode, get_inode_mode(inode));
J
Jaegeuk Kim 已提交
567 568 569 570 571 572 573 574 575 576 577 578 579 580
		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,
581
	.set_acl	= f2fs_set_acl,
J
Jaegeuk Kim 已提交
582 583 584 585 586 587
#ifdef CONFIG_F2FS_FS_XATTR
	.setxattr	= generic_setxattr,
	.getxattr	= generic_getxattr,
	.listxattr	= f2fs_listxattr,
	.removexattr	= generic_removexattr,
#endif
J
Jaegeuk Kim 已提交
588
	.fiemap		= f2fs_fiemap,
J
Jaegeuk Kim 已提交
589 590 591 592 593
};

static void fill_zero(struct inode *inode, pgoff_t index,
					loff_t start, loff_t len)
{
594
	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
J
Jaegeuk Kim 已提交
595 596 597 598 599
	struct page *page;

	if (!len)
		return;

600 601
	f2fs_balance_fs(sbi);

602
	f2fs_lock_op(sbi);
603
	page = get_new_data_page(inode, NULL, index, false);
604
	f2fs_unlock_op(sbi);
J
Jaegeuk Kim 已提交
605 606

	if (!IS_ERR(page)) {
607
		f2fs_wait_on_page_writeback(page, DATA);
J
Jaegeuk Kim 已提交
608 609 610 611 612 613 614 615 616 617 618 619 620
		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;
621

J
Jaegeuk Kim 已提交
622
		set_new_dnode(&dn, inode, NULL, NULL, 0);
623
		err = get_dnode_of_data(&dn, index, LOOKUP_NODE);
J
Jaegeuk Kim 已提交
624 625 626 627 628 629 630 631 632 633 634 635 636
		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;
}

C
Chao Yu 已提交
637
static int punch_hole(struct inode *inode, loff_t offset, loff_t len)
J
Jaegeuk Kim 已提交
638 639 640 641 642
{
	pgoff_t pg_start, pg_end;
	loff_t off_start, off_end;
	int ret = 0;

643
	ret = f2fs_convert_inline_data(inode, MAX_INLINE_DATA + 1, NULL);
644 645
	if (ret)
		return ret;
H
Huajun Li 已提交
646

J
Jaegeuk Kim 已提交
647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665
	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;
666
			struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
667 668

			f2fs_balance_fs(sbi);
J
Jaegeuk Kim 已提交
669 670 671 672 673

			blk_start = pg_start << PAGE_CACHE_SHIFT;
			blk_end = pg_end << PAGE_CACHE_SHIFT;
			truncate_inode_pages_range(mapping, blk_start,
					blk_end - 1);
674

675
			f2fs_lock_op(sbi);
J
Jaegeuk Kim 已提交
676
			ret = truncate_hole(inode, pg_start, pg_end);
677
			f2fs_unlock_op(sbi);
J
Jaegeuk Kim 已提交
678 679 680 681 682 683 684 685 686
		}
	}

	return ret;
}

static int expand_inode_data(struct inode *inode, loff_t offset,
					loff_t len, int mode)
{
687
	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
J
Jaegeuk Kim 已提交
688 689 690 691 692
	pgoff_t index, pg_start, pg_end;
	loff_t new_size = i_size_read(inode);
	loff_t off_start, off_end;
	int ret = 0;

693 694
	f2fs_balance_fs(sbi);

J
Jaegeuk Kim 已提交
695 696 697 698
	ret = inode_newsize_ok(inode, (len + offset));
	if (ret)
		return ret;

699
	ret = f2fs_convert_inline_data(inode, offset + len, NULL);
700 701 702
	if (ret)
		return ret;

J
Jaegeuk Kim 已提交
703 704 705 706 707 708
	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);

709 710
	f2fs_lock_op(sbi);

J
Jaegeuk Kim 已提交
711 712 713
	for (index = pg_start; index <= pg_end; index++) {
		struct dnode_of_data dn;

714 715 716
		if (index == pg_end && !off_end)
			goto noalloc;

J
Jaegeuk Kim 已提交
717
		set_new_dnode(&dn, inode, NULL, NULL, 0);
718 719
		ret = f2fs_reserve_block(&dn, index);
		if (ret)
J
Jaegeuk Kim 已提交
720
			break;
721
noalloc:
J
Jaegeuk Kim 已提交
722 723 724 725 726 727 728 729 730 731 732 733 734 735
		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);
736
		update_inode_page(inode);
J
Jaegeuk Kim 已提交
737
	}
738
	f2fs_unlock_op(sbi);
J
Jaegeuk Kim 已提交
739 740 741 742 743 744 745

	return ret;
}

static long f2fs_fallocate(struct file *file, int mode,
				loff_t offset, loff_t len)
{
A
Al Viro 已提交
746
	struct inode *inode = file_inode(file);
J
Jaegeuk Kim 已提交
747 748 749 750 751
	long ret;

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

752 753
	mutex_lock(&inode->i_mutex);

J
Jaegeuk Kim 已提交
754
	if (mode & FALLOC_FL_PUNCH_HOLE)
C
Chao Yu 已提交
755
		ret = punch_hole(inode, offset, len);
J
Jaegeuk Kim 已提交
756 757 758
	else
		ret = expand_inode_data(inode, offset, len, mode);

759 760 761 762
	if (!ret) {
		inode->i_mtime = inode->i_ctime = CURRENT_TIME;
		mark_inode_dirty(inode);
	}
763 764 765

	mutex_unlock(&inode->i_mutex);

766
	trace_f2fs_fallocate(inode, mode, offset, len, ret);
J
Jaegeuk Kim 已提交
767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784
	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)
{
A
Al Viro 已提交
785
	struct inode *inode = file_inode(filp);
J
Jaegeuk Kim 已提交
786 787 788 789 790
	struct f2fs_inode_info *fi = F2FS_I(inode);
	unsigned int flags;
	int ret;

	switch (cmd) {
791
	case F2FS_IOC_GETFLAGS:
J
Jaegeuk Kim 已提交
792 793
		flags = fi->i_flags & FS_FL_USER_VISIBLE;
		return put_user(flags, (int __user *) arg);
794
	case F2FS_IOC_SETFLAGS:
J
Jaegeuk Kim 已提交
795 796 797
	{
		unsigned int oldflags;

798
		ret = mnt_want_write_file(filp);
J
Jaegeuk Kim 已提交
799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834
		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:
835
		mnt_drop_write_file(filp);
J
Jaegeuk Kim 已提交
836 837 838 839 840 841 842
		return ret;
	}
	default:
		return -ENOTTY;
	}
}

843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859
#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

J
Jaegeuk Kim 已提交
860
const struct file_operations f2fs_file_operations = {
861
	.llseek		= f2fs_llseek,
862
	.read		= new_sync_read,
863
	.write		= new_sync_write,
864
	.read_iter	= generic_file_read_iter,
865
	.write_iter	= generic_file_write_iter,
J
Jaegeuk Kim 已提交
866 867 868 869 870
	.open		= generic_file_open,
	.mmap		= f2fs_file_mmap,
	.fsync		= f2fs_sync_file,
	.fallocate	= f2fs_fallocate,
	.unlocked_ioctl	= f2fs_ioctl,
871 872 873
#ifdef CONFIG_COMPAT
	.compat_ioctl	= f2fs_compat_ioctl,
#endif
J
Jaegeuk Kim 已提交
874
	.splice_read	= generic_file_splice_read,
A
Al Viro 已提交
875
	.splice_write	= iter_file_splice_write,
J
Jaegeuk Kim 已提交
876
};