bmap.c 49.2 KB
Newer Older
D
David Teigland 已提交
1 2
/*
 * Copyright (C) Sistina Software, Inc.  1997-2003 All rights reserved.
3
 * Copyright (C) 2004-2006 Red Hat, Inc.  All rights reserved.
D
David Teigland 已提交
4 5 6
 *
 * This copyrighted material is made available to anyone wishing to use,
 * modify, copy, or redistribute it subject to the terms and conditions
7
 * of the GNU General Public License version 2.
D
David Teigland 已提交
8 9 10 11 12
 */

#include <linux/spinlock.h>
#include <linux/completion.h>
#include <linux/buffer_head.h>
13
#include <linux/blkdev.h>
14
#include <linux/gfs2_ondisk.h>
15
#include <linux/crc32.h>
B
Bob Peterson 已提交
16
#include <linux/iomap.h>
D
David Teigland 已提交
17 18

#include "gfs2.h"
19
#include "incore.h"
D
David Teigland 已提交
20 21 22 23 24 25
#include "bmap.h"
#include "glock.h"
#include "inode.h"
#include "meta_io.h"
#include "quota.h"
#include "rgrp.h"
26
#include "log.h"
27
#include "super.h"
D
David Teigland 已提交
28
#include "trans.h"
29
#include "dir.h"
30
#include "util.h"
S
Steven Whitehouse 已提交
31
#include "trace_gfs2.h"
D
David Teigland 已提交
32 33 34 35 36 37

/* This doesn't need to be that large as max 64 bit pointers in a 4k
 * block is 512, so __u16 is fine for that. It saves stack space to
 * keep it small.
 */
struct metapath {
38
	struct buffer_head *mp_bh[GFS2_MAX_META_HEIGHT];
D
David Teigland 已提交
39
	__u16 mp_list[GFS2_MAX_META_HEIGHT];
40 41
	int mp_fheight; /* find_metapath height */
	int mp_aheight; /* actual height (lookup height) */
D
David Teigland 已提交
42 43
};

44 45 46 47 48
/**
 * gfs2_unstuffer_page - unstuff a stuffed inode into a block cached by a page
 * @ip: the inode
 * @dibh: the dinode buffer
 * @block: the block number that was allocated
S
Steven Whitehouse 已提交
49
 * @page: The (optional) page. This is looked up if @page is NULL
50 51 52 53 54
 *
 * Returns: errno
 */

static int gfs2_unstuffer_page(struct gfs2_inode *ip, struct buffer_head *dibh,
55
			       u64 block, struct page *page)
56 57 58 59 60 61
{
	struct inode *inode = &ip->i_inode;
	struct buffer_head *bh;
	int release = 0;

	if (!page || page->index) {
62
		page = find_or_create_page(inode->i_mapping, 0, GFP_NOFS);
63 64 65 66 67 68 69
		if (!page)
			return -ENOMEM;
		release = 1;
	}

	if (!PageUptodate(page)) {
		void *kaddr = kmap(page);
70 71 72 73
		u64 dsize = i_size_read(inode);
 
		if (dsize > (dibh->b_size - sizeof(struct gfs2_dinode)))
			dsize = dibh->b_size - sizeof(struct gfs2_dinode);
74

75
		memcpy(kaddr, dibh->b_data + sizeof(struct gfs2_dinode), dsize);
76
		memset(kaddr + dsize, 0, PAGE_SIZE - dsize);
77 78 79 80 81 82
		kunmap(page);

		SetPageUptodate(page);
	}

	if (!page_has_buffers(page))
F
Fabian Frederick 已提交
83 84
		create_empty_buffers(page, BIT(inode->i_blkbits),
				     BIT(BH_Uptodate));
85 86 87 88 89 90 91

	bh = page_buffers(page);

	if (!buffer_mapped(bh))
		map_bh(bh, inode->i_sb, block);

	set_buffer_uptodate(bh);
92 93
	if (!gfs2_is_jdata(ip))
		mark_buffer_dirty(bh);
94
	if (!gfs2_is_writeback(ip))
95
		gfs2_trans_add_data(ip->i_gl, bh);
96 97 98

	if (release) {
		unlock_page(page);
99
		put_page(page);
100 101 102 103 104
	}

	return 0;
}

D
David Teigland 已提交
105 106 107
/**
 * gfs2_unstuff_dinode - Unstuff a dinode when the data has grown too big
 * @ip: The GFS2 inode to unstuff
S
Steven Whitehouse 已提交
108
 * @page: The (optional) page. This is looked up if the @page is NULL
D
David Teigland 已提交
109 110 111 112 113 114 115
 *
 * This routine unstuffs a dinode and returns it to a "normal" state such
 * that the height can be grown in the traditional way.
 *
 * Returns: errno
 */

116
int gfs2_unstuff_dinode(struct gfs2_inode *ip, struct page *page)
D
David Teigland 已提交
117 118
{
	struct buffer_head *bh, *dibh;
119
	struct gfs2_dinode *di;
120
	u64 block = 0;
121
	int isdir = gfs2_is_dir(ip);
D
David Teigland 已提交
122 123 124 125 126 127 128
	int error;

	down_write(&ip->i_rw_mutex);

	error = gfs2_meta_inode_buffer(ip, &dibh);
	if (error)
		goto out;
129

S
Steven Whitehouse 已提交
130
	if (i_size_read(&ip->i_inode)) {
D
David Teigland 已提交
131 132 133
		/* Get a free block, fill it with the stuffed data,
		   and write it out to disk */

134
		unsigned int n = 1;
135
		error = gfs2_alloc_blocks(ip, &block, &n, 0, NULL);
136 137
		if (error)
			goto out_brelse;
138
		if (isdir) {
139
			gfs2_trans_add_unrevoke(GFS2_SB(&ip->i_inode), block, 1);
140
			error = gfs2_dir_get_new_buffer(ip, block, &bh);
D
David Teigland 已提交
141 142
			if (error)
				goto out_brelse;
143
			gfs2_buffer_copy_tail(bh, sizeof(struct gfs2_meta_header),
D
David Teigland 已提交
144 145 146
					      dibh, sizeof(struct gfs2_dinode));
			brelse(bh);
		} else {
147
			error = gfs2_unstuffer_page(ip, dibh, block, page);
D
David Teigland 已提交
148 149 150 151 152 153 154
			if (error)
				goto out_brelse;
		}
	}

	/*  Set up the pointer to the new block  */

155
	gfs2_trans_add_meta(ip->i_gl, dibh);
156
	di = (struct gfs2_dinode *)dibh->b_data;
D
David Teigland 已提交
157 158
	gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode));

S
Steven Whitehouse 已提交
159
	if (i_size_read(&ip->i_inode)) {
160
		*(__be64 *)(di + 1) = cpu_to_be64(block);
161 162
		gfs2_add_inode_blocks(&ip->i_inode, 1);
		di->di_blocks = cpu_to_be64(gfs2_get_inode_blocks(&ip->i_inode));
D
David Teigland 已提交
163 164
	}

165
	ip->i_height = 1;
166
	di->di_height = cpu_to_be16(1);
D
David Teigland 已提交
167

168
out_brelse:
D
David Teigland 已提交
169
	brelse(dibh);
170
out:
D
David Teigland 已提交
171 172 173 174 175 176 177
	up_write(&ip->i_rw_mutex);
	return error;
}


/**
 * find_metapath - Find path through the metadata tree
178
 * @sdp: The superblock
D
David Teigland 已提交
179 180
 * @mp: The metapath to return the result in
 * @block: The disk block to look up
181
 * @height: The pre-calculated height of the metadata tree
D
David Teigland 已提交
182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235
 *
 *   This routine returns a struct metapath structure that defines a path
 *   through the metadata of inode "ip" to get to block "block".
 *
 *   Example:
 *   Given:  "ip" is a height 3 file, "offset" is 101342453, and this is a
 *   filesystem with a blocksize of 4096.
 *
 *   find_metapath() would return a struct metapath structure set to:
 *   mp_offset = 101342453, mp_height = 3, mp_list[0] = 0, mp_list[1] = 48,
 *   and mp_list[2] = 165.
 *
 *   That means that in order to get to the block containing the byte at
 *   offset 101342453, we would load the indirect block pointed to by pointer
 *   0 in the dinode.  We would then load the indirect block pointed to by
 *   pointer 48 in that indirect block.  We would then load the data block
 *   pointed to by pointer 165 in that indirect block.
 *
 *             ----------------------------------------
 *             | Dinode |                             |
 *             |        |                            4|
 *             |        |0 1 2 3 4 5                 9|
 *             |        |                            6|
 *             ----------------------------------------
 *                       |
 *                       |
 *                       V
 *             ----------------------------------------
 *             | Indirect Block                       |
 *             |                                     5|
 *             |            4 4 4 4 4 5 5            1|
 *             |0           5 6 7 8 9 0 1            2|
 *             ----------------------------------------
 *                                |
 *                                |
 *                                V
 *             ----------------------------------------
 *             | Indirect Block                       |
 *             |                         1 1 1 1 1   5|
 *             |                         6 6 6 6 6   1|
 *             |0                        3 4 5 6 7   2|
 *             ----------------------------------------
 *                                           |
 *                                           |
 *                                           V
 *             ----------------------------------------
 *             | Data block containing offset         |
 *             |            101342453                 |
 *             |                                      |
 *             |                                      |
 *             ----------------------------------------
 *
 */

236 237
static void find_metapath(const struct gfs2_sbd *sdp, u64 block,
			  struct metapath *mp, unsigned int height)
D
David Teigland 已提交
238 239 240
{
	unsigned int i;

241
	mp->mp_fheight = height;
242
	for (i = height; i--;)
B
Bob Peterson 已提交
243
		mp->mp_list[i] = do_div(block, sdp->sd_inptrs);
D
David Teigland 已提交
244 245
}

246
static inline unsigned int metapath_branch_start(const struct metapath *mp)
247
{
248 249 250
	if (mp->mp_list[0] == 0)
		return 2;
	return 1;
251 252
}

B
Bob Peterson 已提交
253
/**
254
 * metaptr1 - Return the first possible metadata pointer in a metapath buffer
B
Bob Peterson 已提交
255 256 257 258 259 260 261 262 263 264 265
 * @height: The metadata height (0 = dinode)
 * @mp: The metapath
 */
static inline __be64 *metaptr1(unsigned int height, const struct metapath *mp)
{
	struct buffer_head *bh = mp->mp_bh[height];
	if (height == 0)
		return ((__be64 *)(bh->b_data + sizeof(struct gfs2_dinode)));
	return ((__be64 *)(bh->b_data + sizeof(struct gfs2_meta_header)));
}

D
David Teigland 已提交
266 267 268 269 270 271 272 273 274 275
/**
 * metapointer - Return pointer to start of metadata in a buffer
 * @height: The metadata height (0 = dinode)
 * @mp: The metapath
 *
 * Return a pointer to the block number of the next height of the metadata
 * tree given a buffer containing the pointer to the current height of the
 * metadata tree.
 */

276
static inline __be64 *metapointer(unsigned int height, const struct metapath *mp)
D
David Teigland 已提交
277
{
B
Bob Peterson 已提交
278 279
	__be64 *p = metaptr1(height, mp);
	return p + mp->mp_list[height];
D
David Teigland 已提交
280 281
}

282 283 284 285 286 287 288 289 290 291 292 293 294 295 296
static void gfs2_metapath_ra(struct gfs2_glock *gl,
			     const struct buffer_head *bh, const __be64 *pos)
{
	struct buffer_head *rabh;
	const __be64 *endp = (const __be64 *)(bh->b_data + bh->b_size);
	const __be64 *t;

	for (t = pos; t < endp; t++) {
		if (!*t)
			continue;

		rabh = gfs2_getbuf(gl, be64_to_cpu(*t), CREATE);
		if (trylock_buffer(rabh)) {
			if (!buffer_uptodate(rabh)) {
				rabh->b_end_io = end_buffer_read_sync;
297 298 299
				submit_bh(REQ_OP_READ,
					  REQ_RAHEAD | REQ_META | REQ_PRIO,
					  rabh);
300 301 302 303 304 305 306 307
				continue;
			}
			unlock_buffer(rabh);
		}
		brelse(rabh);
	}
}

B
Bob Peterson 已提交
308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324
/**
 * lookup_mp_height - helper function for lookup_metapath
 * @ip: the inode
 * @mp: the metapath
 * @h: the height which needs looking up
 */
static int lookup_mp_height(struct gfs2_inode *ip, struct metapath *mp, int h)
{
	__be64 *ptr = metapointer(h, mp);
	u64 dblock = be64_to_cpu(*ptr);

	if (!dblock)
		return h + 1;

	return gfs2_meta_indirect_buffer(ip, h + 1, dblock, &mp->mp_bh[h + 1]);
}

D
David Teigland 已提交
325
/**
326 327
 * lookup_metapath - Walk the metadata tree to a specific point
 * @ip: The inode
D
David Teigland 已提交
328 329
 * @mp: The metapath
 *
330 331 332 333 334 335 336 337
 * Assumes that the inode's buffer has already been looked up and
 * hooked onto mp->mp_bh[0] and that the metapath has been initialised
 * by find_metapath().
 *
 * If this function encounters part of the tree which has not been
 * allocated, it returns the current height of the tree at the point
 * at which it found the unallocated block. Blocks which are found are
 * added to the mp->mp_bh[] list.
D
David Teigland 已提交
338
 *
339
 * Returns: error or height of metadata tree
D
David Teigland 已提交
340 341
 */

342
static int lookup_metapath(struct gfs2_inode *ip, struct metapath *mp)
343 344 345
{
	unsigned int end_of_metadata = ip->i_height - 1;
	unsigned int x;
346
	int ret;
347 348

	for (x = 0; x < end_of_metadata; x++) {
B
Bob Peterson 已提交
349
		ret = lookup_mp_height(ip, mp, x);
350
		if (ret)
351
			goto out;
352 353
	}

354 355 356 357
	ret = ip->i_height;
out:
	mp->mp_aheight = ret;
	return ret;
358 359
}

B
Bob Peterson 已提交
360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388
/**
 * fillup_metapath - fill up buffers for the metadata path to a specific height
 * @ip: The inode
 * @mp: The metapath
 * @h: The height to which it should be mapped
 *
 * Similar to lookup_metapath, but does lookups for a range of heights
 *
 * Returns: error or height of metadata tree
 */

static int fillup_metapath(struct gfs2_inode *ip, struct metapath *mp, int h)
{
	unsigned int start_h = h - 1;
	int ret;

	if (h) {
		/* find the first buffer we need to look up. */
		while (start_h > 0 && mp->mp_bh[start_h] == NULL)
			start_h--;
		for (; start_h < h; start_h++) {
			ret = lookup_mp_height(ip, mp, start_h);
			if (ret)
				return ret;
		}
	}
	return ip->i_height;
}

389
static inline void release_metapath(struct metapath *mp)
390 391 392
{
	int i;

393 394 395 396 397
	for (i = 0; i < GFS2_MAX_META_HEIGHT; i++) {
		if (mp->mp_bh[i] == NULL)
			break;
		brelse(mp->mp_bh[i]);
	}
398 399
}

400 401 402 403 404 405 406 407 408 409 410 411 412 413 414
/**
 * gfs2_extent_length - Returns length of an extent of blocks
 * @start: Start of the buffer
 * @len: Length of the buffer in bytes
 * @ptr: Current position in the buffer
 * @limit: Max extent length to return (0 = unlimited)
 * @eob: Set to 1 if we hit "end of block"
 *
 * If the first block is zero (unallocated) it will return the number of
 * unallocated blocks in the extent, otherwise it will return the number
 * of contiguous blocks in the extent.
 *
 * Returns: The length of the extent (minimum of one block)
 */

415
static inline unsigned int gfs2_extent_length(void *start, unsigned int len, __be64 *ptr, size_t limit, int *eob)
416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435
{
	const __be64 *end = (start + len);
	const __be64 *first = ptr;
	u64 d = be64_to_cpu(*ptr);

	*eob = 0;
	do {
		ptr++;
		if (ptr >= end)
			break;
		if (limit && --limit == 0)
			break;
		if (d)
			d++;
	} while(be64_to_cpu(*ptr) == d);
	if (ptr >= end)
		*eob = 1;
	return (ptr - first);
}

436
static inline void bmap_lock(struct gfs2_inode *ip, int create)
437 438 439 440 441 442 443
{
	if (create)
		down_write(&ip->i_rw_mutex);
	else
		down_read(&ip->i_rw_mutex);
}

444
static inline void bmap_unlock(struct gfs2_inode *ip, int create)
445 446 447 448 449 450 451
{
	if (create)
		up_write(&ip->i_rw_mutex);
	else
		up_read(&ip->i_rw_mutex);
}

452 453 454 455 456 457 458 459 460 461
static inline __be64 *gfs2_indirect_init(struct metapath *mp,
					 struct gfs2_glock *gl, unsigned int i,
					 unsigned offset, u64 bn)
{
	__be64 *ptr = (__be64 *)(mp->mp_bh[i - 1]->b_data +
		       ((i > 1) ? sizeof(struct gfs2_meta_header) :
				 sizeof(struct gfs2_dinode)));
	BUG_ON(i < 1);
	BUG_ON(mp->mp_bh[i] != NULL);
	mp->mp_bh[i] = gfs2_meta_new(gl, bn);
462
	gfs2_trans_add_meta(gl, mp->mp_bh[i]);
463 464 465 466 467 468 469 470 471 472 473 474 475 476
	gfs2_metatype_set(mp->mp_bh[i], GFS2_METATYPE_IN, GFS2_FORMAT_IN);
	gfs2_buffer_clear_tail(mp->mp_bh[i], sizeof(struct gfs2_meta_header));
	ptr += offset;
	*ptr = cpu_to_be64(bn);
	return ptr;
}

enum alloc_state {
	ALLOC_DATA = 0,
	ALLOC_GROW_DEPTH = 1,
	ALLOC_GROW_HEIGHT = 2,
	/* ALLOC_UNSTUFF = 3,   TBD and rather complicated */
};

B
Bob Peterson 已提交
477 478 479 480 481 482 483
static inline unsigned int hptrs(struct gfs2_sbd *sdp, const unsigned int hgt)
{
	if (hgt)
		return sdp->sd_inptrs;
	return sdp->sd_diptrs;
}

484 485 486 487 488
/**
 * gfs2_bmap_alloc - Build a metadata tree of the requested height
 * @inode: The GFS2 inode
 * @lblock: The logical starting block of the extent
 * @bh_map: This is used to return the mapping details
489 490
 * @zero_new: True if newly allocated blocks should be zeroed
 * @mp: The metapath, with proper height information calculated
491
 * @maxlen: The max number of data blocks to alloc
492 493
 * @dblock: Pointer to return the resulting new block
 * @dblks: Pointer to return the number of blocks allocated
494 495 496 497 498 499 500 501 502 503 504 505 506 507 508
 *
 * In this routine we may have to alloc:
 *   i) Indirect blocks to grow the metadata tree height
 *  ii) Indirect blocks to fill in lower part of the metadata tree
 * iii) Data blocks
 *
 * The function is in two parts. The first part works out the total
 * number of blocks which we need. The second part does the actual
 * allocation asking for an extent at a time (if enough contiguous free
 * blocks are available, there will only be one request per bmap call)
 * and uses the state machine to initialise the blocks in order.
 *
 * Returns: errno on error
 */

B
Bob Peterson 已提交
509 510
static int gfs2_iomap_alloc(struct inode *inode, struct iomap *iomap,
			    unsigned flags, struct metapath *mp)
511 512 513
{
	struct gfs2_inode *ip = GFS2_I(inode);
	struct gfs2_sbd *sdp = GFS2_SB(inode);
514
	struct super_block *sb = sdp->sd_vfs;
515
	struct buffer_head *dibh = mp->mp_bh[0];
516
	u64 bn;
517
	unsigned n, i, blks, alloced = 0, iblks = 0, branch_start = 0;
B
Bob Peterson 已提交
518
	unsigned dblks = 0;
519
	unsigned ptrs_per_blk;
520
	const unsigned end_of_metadata = mp->mp_fheight - 1;
521
	int ret;
522 523 524
	enum alloc_state state;
	__be64 *ptr;
	__be64 zero_bn = 0;
B
Bob Peterson 已提交
525
	size_t maxlen = iomap->length >> inode->i_blkbits;
526

527
	BUG_ON(mp->mp_aheight < 1);
528 529
	BUG_ON(dibh == NULL);

530
	gfs2_trans_add_meta(ip->i_gl, dibh);
531

532
	if (mp->mp_fheight == mp->mp_aheight) {
533
		struct buffer_head *bh;
B
Bob Peterson 已提交
534 535
		int eob;

536 537 538
		/* Bottom indirect block exists, find unalloced extent size */
		ptr = metapointer(end_of_metadata, mp);
		bh = mp->mp_bh[end_of_metadata];
B
Bob Peterson 已提交
539 540 541
		dblks = gfs2_extent_length(bh->b_data, bh->b_size, ptr,
					   maxlen, &eob);
		BUG_ON(dblks < 1);
542 543 544
		state = ALLOC_DATA;
	} else {
		/* Need to allocate indirect blocks */
545 546
		ptrs_per_blk = mp->mp_fheight > 1 ? sdp->sd_inptrs :
			sdp->sd_diptrs;
B
Bob Peterson 已提交
547 548
		dblks = min(maxlen, (size_t)(ptrs_per_blk -
					     mp->mp_list[end_of_metadata]));
549
		if (mp->mp_fheight == ip->i_height) {
550
			/* Writing into existing tree, extend tree down */
551
			iblks = mp->mp_fheight - mp->mp_aheight;
552 553 554 555
			state = ALLOC_GROW_DEPTH;
		} else {
			/* Building up tree height */
			state = ALLOC_GROW_HEIGHT;
556
			iblks = mp->mp_fheight - ip->i_height;
557
			branch_start = metapath_branch_start(mp);
558
			iblks += (mp->mp_fheight - branch_start);
559 560 561 562 563
		}
	}

	/* start of the second part of the function (state machine) */

B
Bob Peterson 已提交
564
	blks = dblks + iblks;
565
	i = mp->mp_aheight;
566
	do {
567
		int error;
568
		n = blks - alloced;
569
		error = gfs2_alloc_blocks(ip, &bn, &n, 0, NULL);
570 571
		if (error)
			return error;
572 573 574 575 576 577 578 579 580 581 582
		alloced += n;
		if (state != ALLOC_DATA || gfs2_is_jdata(ip))
			gfs2_trans_add_unrevoke(sdp, bn, n);
		switch (state) {
		/* Growing height of tree */
		case ALLOC_GROW_HEIGHT:
			if (i == 1) {
				ptr = (__be64 *)(dibh->b_data +
						 sizeof(struct gfs2_dinode));
				zero_bn = *ptr;
			}
583 584
			for (; i - 1 < mp->mp_fheight - ip->i_height && n > 0;
			     i++, n--)
585
				gfs2_indirect_init(mp, ip->i_gl, i, 0, bn++);
586
			if (i - 1 == mp->mp_fheight - ip->i_height) {
587 588 589 590 591 592 593 594 595 596 597
				i--;
				gfs2_buffer_copy_tail(mp->mp_bh[i],
						sizeof(struct gfs2_meta_header),
						dibh, sizeof(struct gfs2_dinode));
				gfs2_buffer_clear_tail(dibh,
						sizeof(struct gfs2_dinode) +
						sizeof(__be64));
				ptr = (__be64 *)(mp->mp_bh[i]->b_data +
					sizeof(struct gfs2_meta_header));
				*ptr = zero_bn;
				state = ALLOC_GROW_DEPTH;
598
				for(i = branch_start; i < mp->mp_fheight; i++) {
599 600 601 602 603
					if (mp->mp_bh[i] == NULL)
						break;
					brelse(mp->mp_bh[i]);
					mp->mp_bh[i] = NULL;
				}
604
				i = branch_start;
605 606 607 608 609
			}
			if (n == 0)
				break;
		/* Branching from existing tree */
		case ALLOC_GROW_DEPTH:
610
			if (i > 1 && i < mp->mp_fheight)
611
				gfs2_trans_add_meta(ip->i_gl, mp->mp_bh[i-1]);
612
			for (; i < mp->mp_fheight && n > 0; i++, n--)
613 614
				gfs2_indirect_init(mp, ip->i_gl, i,
						   mp->mp_list[i-1], bn++);
615
			if (i == mp->mp_fheight)
616 617 618 619 620
				state = ALLOC_DATA;
			if (n == 0)
				break;
		/* Tree complete, adding data blocks */
		case ALLOC_DATA:
B
Bob Peterson 已提交
621
			BUG_ON(n > dblks);
622
			BUG_ON(mp->mp_bh[end_of_metadata] == NULL);
623
			gfs2_trans_add_meta(ip->i_gl, mp->mp_bh[end_of_metadata]);
B
Bob Peterson 已提交
624
			dblks = n;
625
			ptr = metapointer(end_of_metadata, mp);
B
Bob Peterson 已提交
626 627
			iomap->addr = bn << inode->i_blkbits;
			iomap->flags |= IOMAP_F_NEW;
628 629
			while (n-- > 0)
				*ptr++ = cpu_to_be64(bn++);
B
Bob Peterson 已提交
630 631 632
			if (flags & IOMAP_ZERO) {
				ret = sb_issue_zeroout(sb, iomap->addr >> inode->i_blkbits,
						       dblks, GFP_NOFS);
633 634 635
				if (ret) {
					fs_err(sdp,
					       "Failed to zero data buffers\n");
B
Bob Peterson 已提交
636
					flags &= ~IOMAP_ZERO;
637 638
				}
			}
639 640
			break;
		}
B
Bob Peterson 已提交
641
	} while (iomap->addr == IOMAP_NULL_ADDR);
642

B
Bob Peterson 已提交
643
	iomap->length = (u64)dblks << inode->i_blkbits;
644
	ip->i_height = mp->mp_fheight;
645 646 647 648 649
	gfs2_add_inode_blocks(&ip->i_inode, alloced);
	gfs2_dinode_out(ip, mp->mp_bh[0]->b_data);
	return 0;
}

D
David Teigland 已提交
650
/**
B
Bob Peterson 已提交
651
 * hole_size - figure out the size of a hole
S
Steven Whitehouse 已提交
652
 * @inode: The inode
B
Bob Peterson 已提交
653 654
 * @lblock: The logical starting block number
 * @mp: The metapath
D
David Teigland 已提交
655
 *
B
Bob Peterson 已提交
656
 * Returns: The hole size in bytes
D
David Teigland 已提交
657 658
 *
 */
B
Bob Peterson 已提交
659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697
static u64 hole_size(struct inode *inode, sector_t lblock, struct metapath *mp)
{
	struct gfs2_inode *ip = GFS2_I(inode);
	struct gfs2_sbd *sdp = GFS2_SB(inode);
	struct metapath mp_eof;
	u64 factor = 1;
	int hgt;
	u64 holesz = 0;
	const __be64 *first, *end, *ptr;
	const struct buffer_head *bh;
	u64 lblock_stop = (i_size_read(inode) - 1) >> inode->i_blkbits;
	int zeroptrs;
	bool done = false;

	/* Get another metapath, to the very last byte */
	find_metapath(sdp, lblock_stop, &mp_eof, ip->i_height);
	for (hgt = ip->i_height - 1; hgt >= 0 && !done; hgt--) {
		bh = mp->mp_bh[hgt];
		if (bh) {
			zeroptrs = 0;
			first = metapointer(hgt, mp);
			end = (const __be64 *)(bh->b_data + bh->b_size);

			for (ptr = first; ptr < end; ptr++) {
				if (*ptr) {
					done = true;
					break;
				} else {
					zeroptrs++;
				}
			}
		} else {
			zeroptrs = sdp->sd_inptrs;
		}
		if (factor * zeroptrs >= lblock_stop - lblock + 1) {
			holesz = lblock_stop - lblock + 1;
			break;
		}
		holesz += factor * zeroptrs;
D
David Teigland 已提交
698

B
Bob Peterson 已提交
699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729
		factor *= sdp->sd_inptrs;
		if (hgt && (mp->mp_list[hgt - 1] < mp_eof.mp_list[hgt - 1]))
			(mp->mp_list[hgt - 1])++;
	}
	return holesz << inode->i_blkbits;
}

static void gfs2_stuffed_iomap(struct inode *inode, struct iomap *iomap)
{
	struct gfs2_inode *ip = GFS2_I(inode);

	iomap->addr = (ip->i_no_addr << inode->i_blkbits) +
		      sizeof(struct gfs2_dinode);
	iomap->offset = 0;
	iomap->length = i_size_read(inode);
	iomap->type = IOMAP_MAPPED;
	iomap->flags = IOMAP_F_DATA_INLINE;
}

/**
 * gfs2_iomap_begin - Map blocks from an inode to disk blocks
 * @inode: The inode
 * @pos: Starting position in bytes
 * @length: Length to map, in bytes
 * @flags: iomap flags
 * @iomap: The iomap structure
 *
 * Returns: errno
 */
int gfs2_iomap_begin(struct inode *inode, loff_t pos, loff_t length,
		     unsigned flags, struct iomap *iomap)
D
David Teigland 已提交
730
{
731 732
	struct gfs2_inode *ip = GFS2_I(inode);
	struct gfs2_sbd *sdp = GFS2_SB(inode);
B
Bob Peterson 已提交
733
	struct metapath mp = { .mp_aheight = 1, };
734
	unsigned int factor = sdp->sd_sb.sb_bsize;
735
	const u64 *arr = sdp->sd_heightsize;
736
	__be64 *ptr;
B
Bob Peterson 已提交
737 738
	sector_t lblock;
	sector_t lend;
739 740 741 742 743
	int ret;
	int eob;
	unsigned int len;
	struct buffer_head *bh;
	u8 height;
S
Steven Whitehouse 已提交
744

B
Bob Peterson 已提交
745 746 747 748 749
	trace_gfs2_iomap_start(ip, pos, length, flags);
	if (!length) {
		ret = -EINVAL;
		goto out;
	}
D
David Teigland 已提交
750

B
Bob Peterson 已提交
751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767
	if ((flags & IOMAP_REPORT) && gfs2_is_stuffed(ip)) {
		gfs2_stuffed_iomap(inode, iomap);
		if (pos >= iomap->length)
			return -ENOENT;
		ret = 0;
		goto out;
	}

	lblock = pos >> inode->i_blkbits;
	lend = (pos + length + sdp->sd_sb.sb_bsize - 1) >> inode->i_blkbits;

	iomap->offset = lblock << inode->i_blkbits;
	iomap->addr = IOMAP_NULL_ADDR;
	iomap->type = IOMAP_HOLE;
	iomap->length = (u64)(lend - lblock) << inode->i_blkbits;
	iomap->flags = IOMAP_F_MERGED;
	bmap_lock(ip, 0);
768 769 770 771 772 773

	/*
	 * Directory data blocks have a struct gfs2_meta_header header, so the
	 * remaining size is smaller than the filesystem block size.  Logical
	 * block numbers for directories are in units of this remaining size!
	 */
774
	if (gfs2_is_dir(ip)) {
775
		factor = sdp->sd_jbsize;
776 777
		arr = sdp->sd_jheightsize;
	}
778

779 780
	ret = gfs2_meta_inode_buffer(ip, &mp.mp_bh[0]);
	if (ret)
B
Bob Peterson 已提交
781
		goto out_release;
D
David Teigland 已提交
782

783
	height = ip->i_height;
B
Bob Peterson 已提交
784
	while ((lblock + 1) * factor > arr[height])
785 786 787 788
		height++;
	find_metapath(sdp, lblock, &mp, height);
	if (height > ip->i_height || gfs2_is_stuffed(ip))
		goto do_alloc;
B
Bob Peterson 已提交
789

790 791
	ret = lookup_metapath(ip, &mp);
	if (ret < 0)
B
Bob Peterson 已提交
792 793
		goto out_release;

794
	if (mp.mp_aheight != ip->i_height)
795
		goto do_alloc;
B
Bob Peterson 已提交
796

797 798 799
	ptr = metapointer(ip->i_height - 1, &mp);
	if (*ptr == 0)
		goto do_alloc;
B
Bob Peterson 已提交
800 801 802 803

	iomap->type = IOMAP_MAPPED;
	iomap->addr = be64_to_cpu(*ptr) << inode->i_blkbits;

804
	bh = mp.mp_bh[ip->i_height - 1];
B
Bob Peterson 已提交
805
	len = gfs2_extent_length(bh->b_data, bh->b_size, ptr, lend - lblock, &eob);
806
	if (eob)
B
Bob Peterson 已提交
807 808 809
		iomap->flags |= IOMAP_F_BOUNDARY;
	iomap->length = (u64)len << inode->i_blkbits;

810
	ret = 0;
B
Bob Peterson 已提交
811 812

out_release:
813
	release_metapath(&mp);
B
Bob Peterson 已提交
814 815 816
	bmap_unlock(ip, 0);
out:
	trace_gfs2_iomap_end(ip, iomap, ret);
817
	return ret;
818

819
do_alloc:
B
Bob Peterson 已提交
820 821 822 823 824
	if (!(flags & IOMAP_WRITE)) {
		if (pos >= i_size_read(inode)) {
			ret = -ENOENT;
			goto out_release;
		}
825
		ret = 0;
B
Bob Peterson 已提交
826 827
		iomap->length = hole_size(inode, lblock, &mp);
		goto out_release;
D
David Teigland 已提交
828
	}
829

B
Bob Peterson 已提交
830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861
	ret = gfs2_iomap_alloc(inode, iomap, flags, &mp);
	goto out_release;
}

/**
 * gfs2_block_map - Map a block from an inode to a disk block
 * @inode: The inode
 * @lblock: The logical block number
 * @bh_map: The bh to be mapped
 * @create: True if its ok to alloc blocks to satify the request
 *
 * Sets buffer_mapped() if successful, sets buffer_boundary() if a
 * read of metadata will be required before the next block can be
 * mapped. Sets buffer_new() if new blocks were allocated.
 *
 * Returns: errno
 */

int gfs2_block_map(struct inode *inode, sector_t lblock,
		   struct buffer_head *bh_map, int create)
{
	struct gfs2_inode *ip = GFS2_I(inode);
	struct iomap iomap;
	int ret, flags = 0;

	clear_buffer_mapped(bh_map);
	clear_buffer_new(bh_map);
	clear_buffer_boundary(bh_map);
	trace_gfs2_bmap(ip, bh_map, lblock, create, 1);

	if (create)
		flags |= IOMAP_WRITE;
862
	if (buffer_zeronew(bh_map))
B
Bob Peterson 已提交
863 864 865 866 867 868 869 870 871 872 873 874 875 876
		flags |= IOMAP_ZERO;
	ret = gfs2_iomap_begin(inode, (loff_t)lblock << inode->i_blkbits,
			       bh_map->b_size, flags, &iomap);
	if (ret) {
		if (!create && ret == -ENOENT) {
			/* Return unmapped buffer beyond the end of file.  */
			ret = 0;
		}
		goto out;
	}

	if (iomap.length > bh_map->b_size) {
		iomap.length = bh_map->b_size;
		iomap.flags &= ~IOMAP_F_BOUNDARY;
877
	}
B
Bob Peterson 已提交
878 879 880 881 882 883 884 885 886 887 888
	if (iomap.addr != IOMAP_NULL_ADDR)
		map_bh(bh_map, inode->i_sb, iomap.addr >> inode->i_blkbits);
	bh_map->b_size = iomap.length;
	if (iomap.flags & IOMAP_F_BOUNDARY)
		set_buffer_boundary(bh_map);
	if (iomap.flags & IOMAP_F_NEW)
		set_buffer_new(bh_map);

out:
	trace_gfs2_bmap(ip, bh_map, lblock, create, ret);
	return ret;
S
Steven Whitehouse 已提交
889 890
}

891 892 893
/*
 * Deprecated: do not use in new code
 */
S
Steven Whitehouse 已提交
894 895
int gfs2_extent_map(struct inode *inode, u64 lblock, int *new, u64 *dblock, unsigned *extlen)
{
896
	struct buffer_head bh = { .b_state = 0, .b_blocknr = 0 };
897
	int ret;
S
Steven Whitehouse 已提交
898 899 900 901 902 903
	int create = *new;

	BUG_ON(!extlen);
	BUG_ON(!dblock);
	BUG_ON(!new);

F
Fabian Frederick 已提交
904
	bh.b_size = BIT(inode->i_blkbits + (create ? 0 : 5));
905
	ret = gfs2_block_map(inode, lblock, &bh, create);
906 907 908 909 910 911 912
	*extlen = bh.b_size >> inode->i_blkbits;
	*dblock = bh.b_blocknr;
	if (buffer_new(&bh))
		*new = 1;
	else
		*new = 0;
	return ret;
D
David Teigland 已提交
913 914
}

S
Steven Whitehouse 已提交
915 916 917 918 919
/**
 * gfs2_block_truncate_page - Deal with zeroing out data for truncate
 *
 * This is partly borrowed from ext3.
 */
S
Steven Whitehouse 已提交
920
static int gfs2_block_truncate_page(struct address_space *mapping, loff_t from)
S
Steven Whitehouse 已提交
921 922 923
{
	struct inode *inode = mapping->host;
	struct gfs2_inode *ip = GFS2_I(inode);
924 925
	unsigned long index = from >> PAGE_SHIFT;
	unsigned offset = from & (PAGE_SIZE-1);
S
Steven Whitehouse 已提交
926 927 928 929 930
	unsigned blocksize, iblock, length, pos;
	struct buffer_head *bh;
	struct page *page;
	int err;

931
	page = find_or_create_page(mapping, index, GFP_NOFS);
S
Steven Whitehouse 已提交
932 933 934 935 936
	if (!page)
		return 0;

	blocksize = inode->i_sb->s_blocksize;
	length = blocksize - (offset & (blocksize - 1));
937
	iblock = index << (PAGE_SHIFT - inode->i_sb->s_blocksize_bits);
S
Steven Whitehouse 已提交
938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953

	if (!page_has_buffers(page))
		create_empty_buffers(page, blocksize, 0);

	/* Find the buffer that contains "offset" */
	bh = page_buffers(page);
	pos = blocksize;
	while (offset >= pos) {
		bh = bh->b_this_page;
		iblock++;
		pos += blocksize;
	}

	err = 0;

	if (!buffer_mapped(bh)) {
954
		gfs2_block_map(inode, iblock, bh, 0);
S
Steven Whitehouse 已提交
955 956 957 958 959 960 961 962 963 964 965
		/* unmapped? It's a hole - nothing to do */
		if (!buffer_mapped(bh))
			goto unlock;
	}

	/* Ok, it's mapped. Make sure it's up-to-date */
	if (PageUptodate(page))
		set_buffer_uptodate(bh);

	if (!buffer_uptodate(bh)) {
		err = -EIO;
966
		ll_rw_block(REQ_OP_READ, 0, 1, &bh);
S
Steven Whitehouse 已提交
967 968 969 970
		wait_on_buffer(bh);
		/* Uhhuh. Read error. Complain and punt. */
		if (!buffer_uptodate(bh))
			goto unlock;
971
		err = 0;
S
Steven Whitehouse 已提交
972 973
	}

974
	if (!gfs2_is_writeback(ip))
975
		gfs2_trans_add_data(ip->i_gl, bh);
S
Steven Whitehouse 已提交
976

977
	zero_user(page, offset, length);
978
	mark_buffer_dirty(bh);
S
Steven Whitehouse 已提交
979 980
unlock:
	unlock_page(page);
981
	put_page(page);
S
Steven Whitehouse 已提交
982 983 984
	return err;
}

985 986
#define GFS2_JTRUNC_REVOKES 8192

987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008
/**
 * gfs2_journaled_truncate - Wrapper for truncate_pagecache for jdata files
 * @inode: The inode being truncated
 * @oldsize: The original (larger) size
 * @newsize: The new smaller size
 *
 * With jdata files, we have to journal a revoke for each block which is
 * truncated. As a result, we need to split this into separate transactions
 * if the number of pages being truncated gets too large.
 */

static int gfs2_journaled_truncate(struct inode *inode, u64 oldsize, u64 newsize)
{
	struct gfs2_sbd *sdp = GFS2_SB(inode);
	u64 max_chunk = GFS2_JTRUNC_REVOKES * sdp->sd_vfs->s_blocksize;
	u64 chunk;
	int error;

	while (oldsize != newsize) {
		chunk = oldsize - newsize;
		if (chunk > max_chunk)
			chunk = max_chunk;
1009
		truncate_pagecache(inode, oldsize - chunk);
1010 1011 1012 1013 1014 1015 1016 1017 1018 1019
		oldsize -= chunk;
		gfs2_trans_end(sdp);
		error = gfs2_trans_begin(sdp, RES_DINODE, GFS2_JTRUNC_REVOKES);
		if (error)
			return error;
	}

	return 0;
}

S
Steven Whitehouse 已提交
1020
static int trunc_start(struct inode *inode, u64 oldsize, u64 newsize)
D
David Teigland 已提交
1021
{
S
Steven Whitehouse 已提交
1022 1023 1024
	struct gfs2_inode *ip = GFS2_I(inode);
	struct gfs2_sbd *sdp = GFS2_SB(inode);
	struct address_space *mapping = inode->i_mapping;
D
David Teigland 已提交
1025 1026 1027 1028
	struct buffer_head *dibh;
	int journaled = gfs2_is_jdata(ip);
	int error;

1029 1030 1031 1032
	if (journaled)
		error = gfs2_trans_begin(sdp, RES_DINODE + RES_JDATA, GFS2_JTRUNC_REVOKES);
	else
		error = gfs2_trans_begin(sdp, RES_DINODE, 0);
D
David Teigland 已提交
1033 1034 1035 1036 1037 1038 1039
	if (error)
		return error;

	error = gfs2_meta_inode_buffer(ip, &dibh);
	if (error)
		goto out;

1040
	gfs2_trans_add_meta(ip->i_gl, dibh);
S
Steven Whitehouse 已提交
1041

D
David Teigland 已提交
1042
	if (gfs2_is_stuffed(ip)) {
S
Steven Whitehouse 已提交
1043
		gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode) + newsize);
D
David Teigland 已提交
1044
	} else {
S
Steven Whitehouse 已提交
1045 1046 1047 1048
		if (newsize & (u64)(sdp->sd_sb.sb_bsize - 1)) {
			error = gfs2_block_truncate_page(mapping, newsize);
			if (error)
				goto out_brelse;
D
David Teigland 已提交
1049
		}
S
Steven Whitehouse 已提交
1050
		ip->i_diskflags |= GFS2_DIF_TRUNC_IN_PROG;
D
David Teigland 已提交
1051 1052
	}

S
Steven Whitehouse 已提交
1053
	i_size_write(inode, newsize);
1054
	ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
S
Steven Whitehouse 已提交
1055
	gfs2_dinode_out(ip, dibh->b_data);
D
David Teigland 已提交
1056

1057 1058 1059
	if (journaled)
		error = gfs2_journaled_truncate(inode, oldsize, newsize);
	else
1060
		truncate_pagecache(inode, newsize);
1061 1062 1063 1064 1065 1066

	if (error) {
		brelse(dibh);
		return error;
	}

S
Steven Whitehouse 已提交
1067 1068
out_brelse:
	brelse(dibh);
1069
out:
D
David Teigland 已提交
1070 1071 1072 1073
	gfs2_trans_end(sdp);
	return error;
}

B
Bob Peterson 已提交
1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097
/**
 * sweep_bh_for_rgrps - find an rgrp in a meta buffer and free blocks therein
 * @ip: inode
 * @rg_gh: holder of resource group glock
 * @mp: current metapath fully populated with buffers
 * @btotal: place to keep count of total blocks freed
 * @hgt: height we're processing
 * @first: true if this is the first call to this function for this height
 *
 * We sweep a metadata buffer (provided by the metapath) for blocks we need to
 * free, and free them all. However, we do it one rgrp at a time. If this
 * block has references to multiple rgrps, we break it into individual
 * transactions. This allows other processes to use the rgrps while we're
 * focused on a single one, for better concurrency / performance.
 * At every transaction boundary, we rewrite the inode into the journal.
 * That way the bitmaps are kept consistent with the inode and we can recover
 * if we're interrupted by power-outages.
 *
 * Returns: 0, or return code if an error occurred.
 *          *btotal has the total number of blocks freed
 */
static int sweep_bh_for_rgrps(struct gfs2_inode *ip, struct gfs2_holder *rd_gh,
			      const struct metapath *mp, u32 *btotal, int hgt,
			      bool preserve1)
D
David Teigland 已提交
1098
{
1099
	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
B
Bob Peterson 已提交
1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131
	struct gfs2_rgrpd *rgd;
	struct gfs2_trans *tr;
	struct buffer_head *bh = mp->mp_bh[hgt];
	__be64 *top, *bottom, *p;
	int blks_outside_rgrp;
	u64 bn, bstart, isize_blks;
	s64 blen; /* needs to be s64 or gfs2_add_inode_blocks breaks */
	int meta = ((hgt != ip->i_height - 1) ? 1 : 0);
	int ret = 0;
	bool buf_in_tr = false; /* buffer was added to transaction */

	if (gfs2_metatype_check(sdp, bh,
				(hgt ? GFS2_METATYPE_IN : GFS2_METATYPE_DI)))
		return -EIO;

more_rgrps:
	blks_outside_rgrp = 0;
	bstart = 0;
	blen = 0;
	top = metapointer(hgt, mp); /* first ptr from metapath */
	/* If we're keeping some data at the truncation point, we've got to
	   preserve the metadata tree by adding 1 to the starting metapath. */
	if (preserve1)
		top++;

	bottom = (__be64 *)(bh->b_data + bh->b_size);

	for (p = top; p < bottom; p++) {
		if (!*p)
			continue;
		bn = be64_to_cpu(*p);
		if (gfs2_holder_initialized(rd_gh)) {
1132
			rgd = gfs2_glock2rgrp(rd_gh->gh_gl);
B
Bob Peterson 已提交
1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233
			gfs2_assert_withdraw(sdp,
				     gfs2_glock_is_locked_by_me(rd_gh->gh_gl));
		} else {
			rgd = gfs2_blk2rgrpd(sdp, bn, false);
			ret = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE,
						 0, rd_gh);
			if (ret)
				goto out;

			/* Must be done with the rgrp glock held: */
			if (gfs2_rs_active(&ip->i_res) &&
			    rgd == ip->i_res.rs_rbm.rgd)
				gfs2_rs_deltree(&ip->i_res);
		}

		if (!rgrp_contains_block(rgd, bn)) {
			blks_outside_rgrp++;
			continue;
		}

		/* The size of our transactions will be unknown until we
		   actually process all the metadata blocks that relate to
		   the rgrp. So we estimate. We know it can't be more than
		   the dinode's i_blocks and we don't want to exceed the
		   journal flush threshold, sd_log_thresh2. */
		if (current->journal_info == NULL) {
			unsigned int jblocks_rqsted, revokes;

			jblocks_rqsted = rgd->rd_length + RES_DINODE +
				RES_INDIRECT;
			isize_blks = gfs2_get_inode_blocks(&ip->i_inode);
			if (isize_blks > atomic_read(&sdp->sd_log_thresh2))
				jblocks_rqsted +=
					atomic_read(&sdp->sd_log_thresh2);
			else
				jblocks_rqsted += isize_blks;
			revokes = jblocks_rqsted;
			if (meta)
				revokes += hptrs(sdp, hgt);
			else if (ip->i_depth)
				revokes += sdp->sd_inptrs;
			ret = gfs2_trans_begin(sdp, jblocks_rqsted, revokes);
			if (ret)
				goto out_unlock;
			down_write(&ip->i_rw_mutex);
		}
		/* check if we will exceed the transaction blocks requested */
		tr = current->journal_info;
		if (tr->tr_num_buf_new + RES_STATFS +
		    RES_QUOTA >= atomic_read(&sdp->sd_log_thresh2)) {
			/* We set blks_outside_rgrp to ensure the loop will
			   be repeated for the same rgrp, but with a new
			   transaction. */
			blks_outside_rgrp++;
			/* This next part is tricky. If the buffer was added
			   to the transaction, we've already set some block
			   pointers to 0, so we better follow through and free
			   them, or we will introduce corruption (so break).
			   This may be impossible, or at least rare, but I
			   decided to cover the case regardless.

			   If the buffer was not added to the transaction
			   (this call), doing so would exceed our transaction
			   size, so we need to end the transaction and start a
			   new one (so goto). */

			if (buf_in_tr)
				break;
			goto out_unlock;
		}

		gfs2_trans_add_meta(ip->i_gl, bh);
		buf_in_tr = true;
		*p = 0;
		if (bstart + blen == bn) {
			blen++;
			continue;
		}
		if (bstart) {
			__gfs2_free_blocks(ip, bstart, (u32)blen, meta);
			(*btotal) += blen;
			gfs2_add_inode_blocks(&ip->i_inode, -blen);
		}
		bstart = bn;
		blen = 1;
	}
	if (bstart) {
		__gfs2_free_blocks(ip, bstart, (u32)blen, meta);
		(*btotal) += blen;
		gfs2_add_inode_blocks(&ip->i_inode, -blen);
	}
out_unlock:
	if (!ret && blks_outside_rgrp) { /* If buffer still has non-zero blocks
					    outside the rgrp we just processed,
					    do it all over again. */
		if (current->journal_info) {
			struct buffer_head *dibh = mp->mp_bh[0];

			/* Every transaction boundary, we rewrite the dinode
			   to keep its di_blocks current in case of failure. */
			ip->i_inode.i_mtime = ip->i_inode.i_ctime =
1234
				current_time(&ip->i_inode);
B
Bob Peterson 已提交
1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264
			gfs2_trans_add_meta(ip->i_gl, dibh);
			gfs2_dinode_out(ip, dibh->b_data);
			up_write(&ip->i_rw_mutex);
			gfs2_trans_end(sdp);
		}
		gfs2_glock_dq_uninit(rd_gh);
		cond_resched();
		goto more_rgrps;
	}
out:
	return ret;
}

/**
 * find_nonnull_ptr - find a non-null pointer given a metapath and height
 * assumes the metapath is valid (with buffers) out to height h
 * @mp: starting metapath
 * @h: desired height to search
 *
 * Returns: true if a non-null pointer was found in the metapath buffer
 *          false if all remaining pointers are NULL in the buffer
 */
static bool find_nonnull_ptr(struct gfs2_sbd *sdp, struct metapath *mp,
			     unsigned int h)
{
	__be64 *ptr;
	unsigned int ptrs = hptrs(sdp, h) - 1;

	while (true) {
		ptr = metapointer(h, mp);
1265 1266 1267 1268 1269 1270 1271
		if (*ptr) { /* if we have a non-null pointer */
			/* Now zero the metapath after the current height. */
			h++;
			if (h < GFS2_MAX_META_HEIGHT)
				memset(&mp->mp_list[h], 0,
				       (GFS2_MAX_META_HEIGHT - h) *
				       sizeof(mp->mp_list[0]));
B
Bob Peterson 已提交
1272
			return true;
1273
		}
B
Bob Peterson 已提交
1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287

		if (mp->mp_list[h] < ptrs)
			mp->mp_list[h]++;
		else
			return false; /* no more pointers in this buffer */
	}
}

enum dealloc_states {
	DEALLOC_MP_FULL = 0,    /* Strip a metapath with all buffers read in */
	DEALLOC_MP_LOWER = 1,   /* lower the metapath strip height */
	DEALLOC_FILL_MP = 2,  /* Fill in the metapath to the given height. */
	DEALLOC_DONE = 3,       /* process complete */
};
D
David Teigland 已提交
1288

1289 1290 1291 1292 1293 1294 1295
static bool mp_eq_to_hgt(struct metapath *mp, __u16 *nbof, unsigned int h)
{
	if (memcmp(mp->mp_list, nbof, h * sizeof(mp->mp_list[0])))
		return false;
	return true;
}

B
Bob Peterson 已提交
1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323
/**
 * trunc_dealloc - truncate a file down to a desired size
 * @ip: inode to truncate
 * @newsize: The desired size of the file
 *
 * This function truncates a file to newsize. It works from the
 * bottom up, and from the right to the left. In other words, it strips off
 * the highest layer (data) before stripping any of the metadata. Doing it
 * this way is best in case the operation is interrupted by power failure, etc.
 * The dinode is rewritten in every transaction to guarantee integrity.
 */
static int trunc_dealloc(struct gfs2_inode *ip, u64 newsize)
{
	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
	struct metapath mp;
	struct buffer_head *dibh, *bh;
	struct gfs2_holder rd_gh;
	u64 lblock;
	__u16 nbof[GFS2_MAX_META_HEIGHT]; /* new beginning of truncation */
	unsigned int strip_h = ip->i_height - 1;
	u32 btotal = 0;
	int ret, state;
	int mp_h; /* metapath buffers are read in to this height */
	sector_t last_ra = 0;
	u64 prev_bnr = 0;
	bool preserve1; /* need to preserve the first meta pointer? */

	if (!newsize)
D
David Teigland 已提交
1324
		lblock = 0;
1325
	else
B
Bob Peterson 已提交
1326
		lblock = (newsize - 1) >> sdp->sd_sb.sb_bsize_shift;
D
David Teigland 已提交
1327

B
Bob Peterson 已提交
1328
	memset(&mp, 0, sizeof(mp));
1329
	find_metapath(sdp, lblock, &mp, ip->i_height);
D
David Teigland 已提交
1330

B
Bob Peterson 已提交
1331 1332 1333 1334 1335
	memcpy(&nbof, &mp.mp_list, sizeof(nbof));

	ret = gfs2_meta_inode_buffer(ip, &dibh);
	if (ret)
		return ret;
D
David Teigland 已提交
1336

B
Bob Peterson 已提交
1337 1338 1339 1340 1341 1342
	mp.mp_bh[0] = dibh;
	ret = lookup_metapath(ip, &mp);
	if (ret == ip->i_height)
		state = DEALLOC_MP_FULL; /* We have a complete metapath */
	else
		state = DEALLOC_FILL_MP; /* deal with partial metapath */
D
David Teigland 已提交
1343

B
Bob Peterson 已提交
1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372
	ret = gfs2_rindex_update(sdp);
	if (ret)
		goto out_metapath;

	ret = gfs2_quota_hold(ip, NO_UID_QUOTA_CHANGE, NO_GID_QUOTA_CHANGE);
	if (ret)
		goto out_metapath;
	gfs2_holder_mark_uninitialized(&rd_gh);

	mp_h = strip_h;

	while (state != DEALLOC_DONE) {
		switch (state) {
		/* Truncate a full metapath at the given strip height.
		 * Note that strip_h == mp_h in order to be in this state. */
		case DEALLOC_MP_FULL:
			if (mp_h > 0) { /* issue read-ahead on metadata */
				__be64 *top;

				bh = mp.mp_bh[mp_h - 1];
				if (bh->b_blocknr != last_ra) {
					last_ra = bh->b_blocknr;
					top = metaptr1(mp_h - 1, &mp);
					gfs2_metapath_ra(ip->i_gl, bh, top);
				}
			}
			/* If we're truncating to a non-zero size and the mp is
			   at the beginning of file for the strip height, we
			   need to preserve the first metadata pointer. */
1373
			preserve1 = (newsize && mp_eq_to_hgt(&mp, nbof, mp_h));
B
Bob Peterson 已提交
1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452
			bh = mp.mp_bh[mp_h];
			gfs2_assert_withdraw(sdp, bh);
			if (gfs2_assert_withdraw(sdp,
						 prev_bnr != bh->b_blocknr)) {
				printk(KERN_EMERG "GFS2: fsid=%s:inode %llu, "
				       "block:%llu, i_h:%u, s_h:%u, mp_h:%u\n",
				       sdp->sd_fsname,
				       (unsigned long long)ip->i_no_addr,
				       prev_bnr, ip->i_height, strip_h, mp_h);
			}
			prev_bnr = bh->b_blocknr;
			ret = sweep_bh_for_rgrps(ip, &rd_gh, &mp, &btotal,
						 mp_h, preserve1);
			/* If we hit an error or just swept dinode buffer,
			   just exit. */
			if (ret || !mp_h) {
				state = DEALLOC_DONE;
				break;
			}
			state = DEALLOC_MP_LOWER;
			break;

		/* lower the metapath strip height */
		case DEALLOC_MP_LOWER:
			/* We're done with the current buffer, so release it,
			   unless it's the dinode buffer. Then back up to the
			   previous pointer. */
			if (mp_h) {
				brelse(mp.mp_bh[mp_h]);
				mp.mp_bh[mp_h] = NULL;
			}
			/* If we can't get any lower in height, we've stripped
			   off all we can. Next step is to back up and start
			   stripping the previous level of metadata. */
			if (mp_h == 0) {
				strip_h--;
				memcpy(&mp.mp_list, &nbof, sizeof(nbof));
				mp_h = strip_h;
				state = DEALLOC_FILL_MP;
				break;
			}
			mp.mp_list[mp_h] = 0;
			mp_h--; /* search one metadata height down */
			if (mp.mp_list[mp_h] >= hptrs(sdp, mp_h) - 1)
				break; /* loop around in the same state */
			mp.mp_list[mp_h]++;
			/* Here we've found a part of the metapath that is not
			 * allocated. We need to search at that height for the
			 * next non-null pointer. */
			if (find_nonnull_ptr(sdp, &mp, mp_h)) {
				state = DEALLOC_FILL_MP;
				mp_h++;
			}
			/* No more non-null pointers at this height. Back up
			   to the previous height and try again. */
			break; /* loop around in the same state */

		/* Fill the metapath with buffers to the given height. */
		case DEALLOC_FILL_MP:
			/* Fill the buffers out to the current height. */
			ret = fillup_metapath(ip, &mp, mp_h);
			if (ret < 0)
				goto out;

			/* If buffers found for the entire strip height */
			if ((ret == ip->i_height) && (mp_h == strip_h)) {
				state = DEALLOC_MP_FULL;
				break;
			}
			if (ret < ip->i_height) /* We have a partial height */
				mp_h = ret - 1;

			/* If we find a non-null block pointer, crawl a bit
			   higher up in the metapath and try again, otherwise
			   we need to look lower for a new starting point. */
			if (find_nonnull_ptr(sdp, &mp, mp_h))
				mp_h++;
			else
				state = DEALLOC_MP_LOWER;
D
David Teigland 已提交
1453
			break;
B
Bob Peterson 已提交
1454
		}
D
David Teigland 已提交
1455 1456
	}

B
Bob Peterson 已提交
1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467
	if (btotal) {
		if (current->journal_info == NULL) {
			ret = gfs2_trans_begin(sdp, RES_DINODE + RES_STATFS +
					       RES_QUOTA, 0);
			if (ret)
				goto out;
			down_write(&ip->i_rw_mutex);
		}
		gfs2_statfs_change(sdp, 0, +btotal, 0);
		gfs2_quota_change(ip, -(s64)btotal, ip->i_inode.i_uid,
				  ip->i_inode.i_gid);
1468
		ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
B
Bob Peterson 已提交
1469 1470 1471 1472 1473
		gfs2_trans_add_meta(ip->i_gl, dibh);
		gfs2_dinode_out(ip, dibh->b_data);
		up_write(&ip->i_rw_mutex);
		gfs2_trans_end(sdp);
	}
D
David Teigland 已提交
1474

B
Bob Peterson 已提交
1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486
out:
	if (gfs2_holder_initialized(&rd_gh))
		gfs2_glock_dq_uninit(&rd_gh);
	if (current->journal_info) {
		up_write(&ip->i_rw_mutex);
		gfs2_trans_end(sdp);
		cond_resched();
	}
	gfs2_quota_unhold(ip);
out_metapath:
	release_metapath(&mp);
	return ret;
D
David Teigland 已提交
1487 1488 1489 1490
}

static int trunc_end(struct gfs2_inode *ip)
{
1491
	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
D
David Teigland 已提交
1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504
	struct buffer_head *dibh;
	int error;

	error = gfs2_trans_begin(sdp, RES_DINODE, 0);
	if (error)
		return error;

	down_write(&ip->i_rw_mutex);

	error = gfs2_meta_inode_buffer(ip, &dibh);
	if (error)
		goto out;

S
Steven Whitehouse 已提交
1505
	if (!i_size_read(&ip->i_inode)) {
1506
		ip->i_height = 0;
1507
		ip->i_goal = ip->i_no_addr;
D
David Teigland 已提交
1508
		gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode));
1509
		gfs2_ordered_del_inode(ip);
D
David Teigland 已提交
1510
	}
1511
	ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
1512
	ip->i_diskflags &= ~GFS2_DIF_TRUNC_IN_PROG;
D
David Teigland 已提交
1513

1514
	gfs2_trans_add_meta(ip->i_gl, dibh);
1515
	gfs2_dinode_out(ip, dibh->b_data);
D
David Teigland 已提交
1516 1517
	brelse(dibh);

1518
out:
D
David Teigland 已提交
1519 1520 1521 1522 1523 1524 1525
	up_write(&ip->i_rw_mutex);
	gfs2_trans_end(sdp);
	return error;
}

/**
 * do_shrink - make a file smaller
S
Steven Whitehouse 已提交
1526 1527 1528
 * @inode: the inode
 * @oldsize: the current inode size
 * @newsize: the size to make the file
D
David Teigland 已提交
1529
 *
S
Steven Whitehouse 已提交
1530 1531
 * Called with an exclusive lock on @inode. The @size must
 * be equal to or smaller than the current inode size.
D
David Teigland 已提交
1532 1533 1534 1535
 *
 * Returns: errno
 */

S
Steven Whitehouse 已提交
1536
static int do_shrink(struct inode *inode, u64 oldsize, u64 newsize)
D
David Teigland 已提交
1537
{
S
Steven Whitehouse 已提交
1538
	struct gfs2_inode *ip = GFS2_I(inode);
D
David Teigland 已提交
1539 1540
	int error;

S
Steven Whitehouse 已提交
1541
	error = trunc_start(inode, oldsize, newsize);
D
David Teigland 已提交
1542 1543
	if (error < 0)
		return error;
S
Steven Whitehouse 已提交
1544
	if (gfs2_is_stuffed(ip))
D
David Teigland 已提交
1545 1546
		return 0;

S
Steven Whitehouse 已提交
1547 1548
	error = trunc_dealloc(ip, newsize);
	if (error == 0)
D
David Teigland 已提交
1549 1550 1551 1552 1553
		error = trunc_end(ip);

	return error;
}

S
Steven Whitehouse 已提交
1554
void gfs2_trim_blocks(struct inode *inode)
W
Wendy Cheng 已提交
1555
{
S
Steven Whitehouse 已提交
1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576
	u64 size = inode->i_size;
	int ret;

	ret = do_shrink(inode, size, size);
	WARN_ON(ret != 0);
}

/**
 * do_grow - Touch and update inode size
 * @inode: The inode
 * @size: The new size
 *
 * This function updates the timestamps on the inode and
 * may also increase the size of the inode. This function
 * must not be called with @size any smaller than the current
 * inode size.
 *
 * Although it is not strictly required to unstuff files here,
 * earlier versions of GFS2 have a bug in the stuffed file reading
 * code which will result in a buffer overrun if the size is larger
 * than the max stuffed file size. In order to prevent this from
L
Lucas De Marchi 已提交
1577
 * occurring, such files are unstuffed, but in other cases we can
S
Steven Whitehouse 已提交
1578 1579 1580 1581 1582 1583 1584 1585 1586
 * just update the inode size directly.
 *
 * Returns: 0 on success, or -ve on error
 */

static int do_grow(struct inode *inode, u64 size)
{
	struct gfs2_inode *ip = GFS2_I(inode);
	struct gfs2_sbd *sdp = GFS2_SB(inode);
1587
	struct gfs2_alloc_parms ap = { .target = 1, };
W
Wendy Cheng 已提交
1588 1589
	struct buffer_head *dibh;
	int error;
1590
	int unstuff = 0;
W
Wendy Cheng 已提交
1591

S
Steven Whitehouse 已提交
1592 1593
	if (gfs2_is_stuffed(ip) &&
	    (size > (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode)))) {
1594
		error = gfs2_quota_lock_check(ip, &ap);
S
Steven Whitehouse 已提交
1595
		if (error)
1596
			return error;
S
Steven Whitehouse 已提交
1597

1598
		error = gfs2_inplace_reserve(ip, &ap);
S
Steven Whitehouse 已提交
1599 1600
		if (error)
			goto do_grow_qunlock;
1601
		unstuff = 1;
S
Steven Whitehouse 已提交
1602 1603
	}

1604 1605 1606
	error = gfs2_trans_begin(sdp, RES_DINODE + RES_STATFS + RES_RG_BIT +
				 (sdp->sd_args.ar_quota == GFS2_QUOTA_OFF ?
				  0 : RES_QUOTA), 0);
W
Wendy Cheng 已提交
1607
	if (error)
S
Steven Whitehouse 已提交
1608
		goto do_grow_release;
W
Wendy Cheng 已提交
1609

1610
	if (unstuff) {
S
Steven Whitehouse 已提交
1611 1612 1613 1614
		error = gfs2_unstuff_dinode(ip, NULL);
		if (error)
			goto do_end_trans;
	}
W
Wendy Cheng 已提交
1615 1616 1617

	error = gfs2_meta_inode_buffer(ip, &dibh);
	if (error)
S
Steven Whitehouse 已提交
1618
		goto do_end_trans;
W
Wendy Cheng 已提交
1619

S
Steven Whitehouse 已提交
1620
	i_size_write(inode, size);
1621
	ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
1622
	gfs2_trans_add_meta(ip->i_gl, dibh);
W
Wendy Cheng 已提交
1623 1624 1625
	gfs2_dinode_out(ip, dibh->b_data);
	brelse(dibh);

S
Steven Whitehouse 已提交
1626
do_end_trans:
W
Wendy Cheng 已提交
1627
	gfs2_trans_end(sdp);
S
Steven Whitehouse 已提交
1628
do_grow_release:
1629
	if (unstuff) {
S
Steven Whitehouse 已提交
1630 1631 1632 1633
		gfs2_inplace_release(ip);
do_grow_qunlock:
		gfs2_quota_unlock(ip);
	}
W
Wendy Cheng 已提交
1634 1635 1636
	return error;
}

D
David Teigland 已提交
1637
/**
S
Steven Whitehouse 已提交
1638 1639 1640
 * gfs2_setattr_size - make a file a given size
 * @inode: the inode
 * @newsize: the size to make the file
D
David Teigland 已提交
1641
 *
S
Steven Whitehouse 已提交
1642 1643 1644
 * The file size can grow, shrink, or stay the same size. This
 * is called holding i_mutex and an exclusive glock on the inode
 * in question.
D
David Teigland 已提交
1645 1646 1647 1648
 *
 * Returns: errno
 */

S
Steven Whitehouse 已提交
1649
int gfs2_setattr_size(struct inode *inode, u64 newsize)
D
David Teigland 已提交
1650
{
1651
	struct gfs2_inode *ip = GFS2_I(inode);
S
Steven Whitehouse 已提交
1652 1653
	int ret;
	u64 oldsize;
D
David Teigland 已提交
1654

S
Steven Whitehouse 已提交
1655
	BUG_ON(!S_ISREG(inode->i_mode));
D
David Teigland 已提交
1656

S
Steven Whitehouse 已提交
1657 1658 1659
	ret = inode_newsize_ok(inode, newsize);
	if (ret)
		return ret;
D
David Teigland 已提交
1660

1661 1662
	inode_dio_wait(inode);

1663
	ret = gfs2_rsqa_alloc(ip);
1664
	if (ret)
1665
		goto out;
1666

S
Steven Whitehouse 已提交
1667
	oldsize = inode->i_size;
1668 1669 1670 1671
	if (newsize >= oldsize) {
		ret = do_grow(inode, newsize);
		goto out;
	}
S
Steven Whitehouse 已提交
1672

1673 1674
	ret = do_shrink(inode, oldsize, newsize);
out:
1675
	gfs2_rsqa_delete(ip, NULL);
1676
	return ret;
D
David Teigland 已提交
1677 1678 1679 1680 1681
}

int gfs2_truncatei_resume(struct gfs2_inode *ip)
{
	int error;
S
Steven Whitehouse 已提交
1682
	error = trunc_dealloc(ip, i_size_read(&ip->i_inode));
D
David Teigland 已提交
1683 1684 1685 1686 1687 1688 1689 1690 1691 1692
	if (!error)
		error = trunc_end(ip);
	return error;
}

int gfs2_file_dealloc(struct gfs2_inode *ip)
{
	return trunc_dealloc(ip, 0);
}

1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713
/**
 * gfs2_free_journal_extents - Free cached journal bmap info
 * @jd: The journal
 *
 */

void gfs2_free_journal_extents(struct gfs2_jdesc *jd)
{
	struct gfs2_journal_extent *jext;

	while(!list_empty(&jd->extent_list)) {
		jext = list_entry(jd->extent_list.next, struct gfs2_journal_extent, list);
		list_del(&jext->list);
		kfree(jext);
	}
}

/**
 * gfs2_add_jextent - Add or merge a new extent to extent cache
 * @jd: The journal descriptor
 * @lblock: The logical block at start of new extent
1714
 * @dblock: The physical block at start of new extent
1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807
 * @blocks: Size of extent in fs blocks
 *
 * Returns: 0 on success or -ENOMEM
 */

static int gfs2_add_jextent(struct gfs2_jdesc *jd, u64 lblock, u64 dblock, u64 blocks)
{
	struct gfs2_journal_extent *jext;

	if (!list_empty(&jd->extent_list)) {
		jext = list_entry(jd->extent_list.prev, struct gfs2_journal_extent, list);
		if ((jext->dblock + jext->blocks) == dblock) {
			jext->blocks += blocks;
			return 0;
		}
	}

	jext = kzalloc(sizeof(struct gfs2_journal_extent), GFP_NOFS);
	if (jext == NULL)
		return -ENOMEM;
	jext->dblock = dblock;
	jext->lblock = lblock;
	jext->blocks = blocks;
	list_add_tail(&jext->list, &jd->extent_list);
	jd->nr_extents++;
	return 0;
}

/**
 * gfs2_map_journal_extents - Cache journal bmap info
 * @sdp: The super block
 * @jd: The journal to map
 *
 * Create a reusable "extent" mapping from all logical
 * blocks to all physical blocks for the given journal.  This will save
 * us time when writing journal blocks.  Most journals will have only one
 * extent that maps all their logical blocks.  That's because gfs2.mkfs
 * arranges the journal blocks sequentially to maximize performance.
 * So the extent would map the first block for the entire file length.
 * However, gfs2_jadd can happen while file activity is happening, so
 * those journals may not be sequential.  Less likely is the case where
 * the users created their own journals by mounting the metafs and
 * laying it out.  But it's still possible.  These journals might have
 * several extents.
 *
 * Returns: 0 on success, or error on failure
 */

int gfs2_map_journal_extents(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd)
{
	u64 lblock = 0;
	u64 lblock_stop;
	struct gfs2_inode *ip = GFS2_I(jd->jd_inode);
	struct buffer_head bh;
	unsigned int shift = sdp->sd_sb.sb_bsize_shift;
	u64 size;
	int rc;

	lblock_stop = i_size_read(jd->jd_inode) >> shift;
	size = (lblock_stop - lblock) << shift;
	jd->nr_extents = 0;
	WARN_ON(!list_empty(&jd->extent_list));

	do {
		bh.b_state = 0;
		bh.b_blocknr = 0;
		bh.b_size = size;
		rc = gfs2_block_map(jd->jd_inode, lblock, &bh, 0);
		if (rc || !buffer_mapped(&bh))
			goto fail;
		rc = gfs2_add_jextent(jd, lblock, bh.b_blocknr, bh.b_size >> shift);
		if (rc)
			goto fail;
		size -= bh.b_size;
		lblock += (bh.b_size >> ip->i_inode.i_blkbits);
	} while(size > 0);

	fs_info(sdp, "journal %d mapped with %u extents\n", jd->jd_jid,
		jd->nr_extents);
	return 0;

fail:
	fs_warn(sdp, "error %d mapping journal %u at offset %llu (extent %u)\n",
		rc, jd->jd_jid,
		(unsigned long long)(i_size_read(jd->jd_inode) - size),
		jd->nr_extents);
	fs_warn(sdp, "bmap=%d lblock=%llu block=%llu, state=0x%08lx, size=%llu\n",
		rc, (unsigned long long)lblock, (unsigned long long)bh.b_blocknr,
		bh.b_state, (unsigned long long)bh.b_size);
	gfs2_free_journal_extents(jd);
	return rc;
}

D
David Teigland 已提交
1808 1809 1810 1811 1812 1813
/**
 * gfs2_write_alloc_required - figure out if a write will require an allocation
 * @ip: the file being written to
 * @offset: the offset to write to
 * @len: the number of bytes being written
 *
1814
 * Returns: 1 if an alloc is required, 0 otherwise
D
David Teigland 已提交
1815 1816
 */

1817
int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset,
1818
			      unsigned int len)
D
David Teigland 已提交
1819
{
1820
	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
1821 1822 1823
	struct buffer_head bh;
	unsigned int shift;
	u64 lblock, lblock_stop, size;
1824
	u64 end_of_file;
D
David Teigland 已提交
1825 1826 1827 1828 1829 1830 1831

	if (!len)
		return 0;

	if (gfs2_is_stuffed(ip)) {
		if (offset + len >
		    sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode))
1832
			return 1;
D
David Teigland 已提交
1833 1834 1835
		return 0;
	}

1836
	shift = sdp->sd_sb.sb_bsize_shift;
1837
	BUG_ON(gfs2_is_dir(ip));
S
Steven Whitehouse 已提交
1838
	end_of_file = (i_size_read(&ip->i_inode) + sdp->sd_sb.sb_bsize - 1) >> shift;
1839 1840 1841
	lblock = offset >> shift;
	lblock_stop = (offset + len + sdp->sd_sb.sb_bsize - 1) >> shift;
	if (lblock_stop > end_of_file)
1842
		return 1;
D
David Teigland 已提交
1843

1844 1845 1846 1847 1848 1849
	size = (lblock_stop - lblock) << shift;
	do {
		bh.b_state = 0;
		bh.b_size = size;
		gfs2_block_map(&ip->i_inode, lblock, &bh, 0);
		if (!buffer_mapped(&bh))
1850
			return 1;
1851 1852 1853
		size -= bh.b_size;
		lblock += (bh.b_size >> ip->i_inode.i_blkbits);
	} while(size > 0);
D
David Teigland 已提交
1854 1855 1856 1857

	return 0;
}