rgrp.c 41.3 KB
Newer Older
D
David Teigland 已提交
1 2
/*
 * Copyright (C) Sistina Software, Inc.  1997-2003 All rights reserved.
3
 * Copyright (C) 2004-2008 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 13
 */

#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/completion.h>
#include <linux/buffer_head.h>
14
#include <linux/fs.h>
15
#include <linux/gfs2_ondisk.h>
16
#include <linux/lm_interface.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 26 27
#include "glock.h"
#include "glops.h"
#include "lops.h"
#include "meta_io.h"
#include "quota.h"
#include "rgrp.h"
#include "super.h"
#include "trans.h"
28
#include "util.h"
29
#include "log.h"
30
#include "inode.h"
31
#include "ops_address.h"
D
David Teigland 已提交
32

S
Steven Whitehouse 已提交
33
#define BFITNOENT ((u32)~0)
34
#define NO_BLOCK ((u64)~0)
35 36 37 38

/*
 * These routines are used by the resource group routines (rgrp.c)
 * to keep track of block allocation.  Each block is represented by two
39 40 41 42 43 44
 * bits.  So, each byte represents GFS2_NBBY (i.e. 4) blocks.
 *
 * 0 = Free
 * 1 = Used (not metadata)
 * 2 = Unlinked (still in use) inode
 * 3 = Used (metadata)
45 46 47 48
 */

static const char valid_change[16] = {
	        /* current */
49
	/* n */ 0, 1, 1, 1,
50
	/* e */ 1, 0, 0, 0,
51
	/* w */ 0, 0, 0, 1,
52 53 54
	        1, 0, 0, 0
};

55 56 57
static u32 rgblk_search(struct gfs2_rgrpd *rgd, u32 goal,
                        unsigned char old_state, unsigned char new_state);

58 59 60 61 62 63 64 65 66
/**
 * gfs2_setbit - Set a bit in the bitmaps
 * @buffer: the buffer that holds the bitmaps
 * @buflen: the length (in bytes) of the buffer
 * @block: the block to set
 * @new_state: the new state of the block
 *
 */

67
static void gfs2_setbit(struct gfs2_rgrpd *rgd, unsigned char *buffer,
68
			unsigned int buflen, u32 block,
69
			unsigned char new_state)
70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96
{
	unsigned char *byte, *end, cur_state;
	unsigned int bit;

	byte = buffer + (block / GFS2_NBBY);
	bit = (block % GFS2_NBBY) * GFS2_BIT_SIZE;
	end = buffer + buflen;

	gfs2_assert(rgd->rd_sbd, byte < end);

	cur_state = (*byte >> bit) & GFS2_BIT_MASK;

	if (valid_change[new_state * 4 + cur_state]) {
		*byte ^= cur_state << bit;
		*byte |= new_state << bit;
	} else
		gfs2_consist_rgrpd(rgd);
}

/**
 * gfs2_testbit - test a bit in the bitmaps
 * @buffer: the buffer that holds the bitmaps
 * @buflen: the length (in bytes) of the buffer
 * @block: the block to read
 *
 */

97
static unsigned char gfs2_testbit(struct gfs2_rgrpd *rgd, unsigned char *buffer,
98
				  unsigned int buflen, u32 block)
99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119
{
	unsigned char *byte, *end, cur_state;
	unsigned int bit;

	byte = buffer + (block / GFS2_NBBY);
	bit = (block % GFS2_NBBY) * GFS2_BIT_SIZE;
	end = buffer + buflen;

	gfs2_assert(rgd->rd_sbd, byte < end);

	cur_state = (*byte >> bit) & GFS2_BIT_MASK;

	return cur_state;
}

/**
 * gfs2_bitfit - Search an rgrp's bitmap buffer to find a bit-pair representing
 *       a block in a given allocation state.
 * @buffer: the buffer that holds the bitmaps
 * @buflen: the length (in bytes) of the buffer
 * @goal: start search at this block's bit-pair (within @buffer)
120
 * @old_state: GFS2_BLKST_XXX the state of the block we're looking for.
121 122 123 124 125 126 127 128
 *
 * Scope of @goal and returned block number is only within this bitmap buffer,
 * not entire rgrp or filesystem.  @buffer will be offset from the actual
 * beginning of a bitmap block buffer, skipping any header structures.
 *
 * Return: the block number (bitmap buffer scope) that was found
 */

129 130
static u32 gfs2_bitfit(const u8 *buffer, unsigned int buflen, u32 goal,
		       u8 old_state)
131
{
132
	const u8 *byte;
133
	u32 blk = goal;
134
	unsigned int bit, bitlong;
135 136 137 138 139 140
	const unsigned long *plong;
#if BITS_PER_LONG == 32
	const unsigned long plong55 = 0x55555555;
#else
	const unsigned long plong55 = 0x5555555555555555;
#endif
141 142

	byte = buffer + (goal / GFS2_NBBY);
143
	plong = (const unsigned long *)(buffer + (goal / GFS2_NBBY));
144
	bit = (goal % GFS2_NBBY) * GFS2_BIT_SIZE;
145
	bitlong = bit;
146

147 148 149 150 151 152
	while (byte < buffer + buflen) {

		if (bitlong == 0 && old_state == 0 && *plong == plong55) {
			plong++;
			byte += sizeof(unsigned long);
			blk += sizeof(unsigned long) * GFS2_NBBY;
153 154
			continue;
		}
155
		if (((*byte >> bit) & GFS2_BIT_MASK) == old_state)
156 157 158 159 160 161
			return blk;
		bit += GFS2_BIT_SIZE;
		if (bit >= 8) {
			bit = 0;
			byte++;
		}
162 163 164 165 166
		bitlong += GFS2_BIT_SIZE;
		if (bitlong >= sizeof(unsigned long) * 8) {
			bitlong = 0;
			plong++;
		}
167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182

		blk++;
	}

	return BFITNOENT;
}

/**
 * gfs2_bitcount - count the number of bits in a certain state
 * @buffer: the buffer that holds the bitmaps
 * @buflen: the length (in bytes) of the buffer
 * @state: the state of the block we're looking for
 *
 * Returns: The number of bits
 */

183 184
static u32 gfs2_bitcount(struct gfs2_rgrpd *rgd, const u8 *buffer,
			 unsigned int buflen, u8 state)
185
{
186 187 188 189 190
	const u8 *byte = buffer;
	const u8 *end = buffer + buflen;
	const u8 state1 = state << 2;
	const u8 state2 = state << 4;
	const u8 state3 = state << 6;
191
	u32 count = 0;
192 193 194 195 196 197 198 199 200 201 202 203 204 205 206

	for (; byte < end; byte++) {
		if (((*byte) & 0x03) == state)
			count++;
		if (((*byte) & 0x0C) == state1)
			count++;
		if (((*byte) & 0x30) == state2)
			count++;
		if (((*byte) & 0xC0) == state3)
			count++;
	}

	return count;
}

D
David Teigland 已提交
207 208 209 210 211 212 213 214 215 216 217
/**
 * gfs2_rgrp_verify - Verify that a resource group is consistent
 * @sdp: the filesystem
 * @rgd: the rgrp
 *
 */

void gfs2_rgrp_verify(struct gfs2_rgrpd *rgd)
{
	struct gfs2_sbd *sdp = rgd->rd_sbd;
	struct gfs2_bitmap *bi = NULL;
218
	u32 length = rgd->rd_length;
219
	u32 count[4], tmp;
D
David Teigland 已提交
220 221
	int buf, x;

222
	memset(count, 0, 4 * sizeof(u32));
D
David Teigland 已提交
223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240

	/* Count # blocks in each of 4 possible allocation states */
	for (buf = 0; buf < length; buf++) {
		bi = rgd->rd_bits + buf;
		for (x = 0; x < 4; x++)
			count[x] += gfs2_bitcount(rgd,
						  bi->bi_bh->b_data +
						  bi->bi_offset,
						  bi->bi_len, x);
	}

	if (count[0] != rgd->rd_rg.rg_free) {
		if (gfs2_consist_rgrpd(rgd))
			fs_err(sdp, "free data mismatch:  %u != %u\n",
			       count[0], rgd->rd_rg.rg_free);
		return;
	}

241
	tmp = rgd->rd_data -
D
David Teigland 已提交
242 243
		rgd->rd_rg.rg_free -
		rgd->rd_rg.rg_dinodes;
244
	if (count[1] + count[2] != tmp) {
D
David Teigland 已提交
245 246 247 248 249 250
		if (gfs2_consist_rgrpd(rgd))
			fs_err(sdp, "used data mismatch:  %u != %u\n",
			       count[1], tmp);
		return;
	}

251
	if (count[3] != rgd->rd_rg.rg_dinodes) {
D
David Teigland 已提交
252
		if (gfs2_consist_rgrpd(rgd))
253 254
			fs_err(sdp, "used metadata mismatch:  %u != %u\n",
			       count[3], rgd->rd_rg.rg_dinodes);
D
David Teigland 已提交
255 256 257
		return;
	}

258
	if (count[2] > count[3]) {
D
David Teigland 已提交
259
		if (gfs2_consist_rgrpd(rgd))
260 261
			fs_err(sdp, "unlinked inodes > inodes:  %u\n",
			       count[2]);
D
David Teigland 已提交
262 263
		return;
	}
264

D
David Teigland 已提交
265 266
}

267
static inline int rgrp_contains_block(struct gfs2_rgrpd *rgd, u64 block)
D
David Teigland 已提交
268
{
269 270
	u64 first = rgd->rd_data0;
	u64 last = first + rgd->rd_data;
271
	return first <= block && block < last;
D
David Teigland 已提交
272 273 274 275 276 277 278 279 280 281
}

/**
 * gfs2_blk2rgrpd - Find resource group for a given data/meta block number
 * @sdp: The GFS2 superblock
 * @n: The data block number
 *
 * Returns: The resource group, or NULL if not found
 */

282
struct gfs2_rgrpd *gfs2_blk2rgrpd(struct gfs2_sbd *sdp, u64 blk)
D
David Teigland 已提交
283 284 285 286 287 288
{
	struct gfs2_rgrpd *rgd;

	spin_lock(&sdp->sd_rindex_spin);

	list_for_each_entry(rgd, &sdp->sd_rindex_mru_list, rd_list_mru) {
289
		if (rgrp_contains_block(rgd, blk)) {
D
David Teigland 已提交
290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351
			list_move(&rgd->rd_list_mru, &sdp->sd_rindex_mru_list);
			spin_unlock(&sdp->sd_rindex_spin);
			return rgd;
		}
	}

	spin_unlock(&sdp->sd_rindex_spin);

	return NULL;
}

/**
 * gfs2_rgrpd_get_first - get the first Resource Group in the filesystem
 * @sdp: The GFS2 superblock
 *
 * Returns: The first rgrp in the filesystem
 */

struct gfs2_rgrpd *gfs2_rgrpd_get_first(struct gfs2_sbd *sdp)
{
	gfs2_assert(sdp, !list_empty(&sdp->sd_rindex_list));
	return list_entry(sdp->sd_rindex_list.next, struct gfs2_rgrpd, rd_list);
}

/**
 * gfs2_rgrpd_get_next - get the next RG
 * @rgd: A RG
 *
 * Returns: The next rgrp
 */

struct gfs2_rgrpd *gfs2_rgrpd_get_next(struct gfs2_rgrpd *rgd)
{
	if (rgd->rd_list.next == &rgd->rd_sbd->sd_rindex_list)
		return NULL;
	return list_entry(rgd->rd_list.next, struct gfs2_rgrpd, rd_list);
}

static void clear_rgrpdi(struct gfs2_sbd *sdp)
{
	struct list_head *head;
	struct gfs2_rgrpd *rgd;
	struct gfs2_glock *gl;

	spin_lock(&sdp->sd_rindex_spin);
	sdp->sd_rindex_forward = NULL;
	head = &sdp->sd_rindex_recent_list;
	while (!list_empty(head)) {
		rgd = list_entry(head->next, struct gfs2_rgrpd, rd_recent);
		list_del(&rgd->rd_recent);
	}
	spin_unlock(&sdp->sd_rindex_spin);

	head = &sdp->sd_rindex_list;
	while (!list_empty(head)) {
		rgd = list_entry(head->next, struct gfs2_rgrpd, rd_list);
		gl = rgd->rd_gl;

		list_del(&rgd->rd_list);
		list_del(&rgd->rd_list_mru);

		if (gl) {
352
			gl->gl_object = NULL;
D
David Teigland 已提交
353 354 355 356
			gfs2_glock_put(gl);
		}

		kfree(rgd->rd_bits);
357
		kmem_cache_free(gfs2_rgrpd_cachep, rgd);
D
David Teigland 已提交
358 359 360 361 362
	}
}

void gfs2_clear_rgrpd(struct gfs2_sbd *sdp)
{
363
	mutex_lock(&sdp->sd_rindex_mutex);
D
David Teigland 已提交
364
	clear_rgrpdi(sdp);
365
	mutex_unlock(&sdp->sd_rindex_mutex);
D
David Teigland 已提交
366 367
}

368 369 370 371 372 373 374 375 376
static void gfs2_rindex_print(const struct gfs2_rgrpd *rgd)
{
	printk(KERN_INFO "  ri_addr = %llu\n", (unsigned long long)rgd->rd_addr);
	printk(KERN_INFO "  ri_length = %u\n", rgd->rd_length);
	printk(KERN_INFO "  ri_data0 = %llu\n", (unsigned long long)rgd->rd_data0);
	printk(KERN_INFO "  ri_data = %u\n", rgd->rd_data);
	printk(KERN_INFO "  ri_bitbytes = %u\n", rgd->rd_bitbytes);
}

D
David Teigland 已提交
377 378 379 380 381 382 383 384 385 386 387 388 389
/**
 * gfs2_compute_bitstructs - Compute the bitmap sizes
 * @rgd: The resource group descriptor
 *
 * Calculates bitmap descriptors, one for each block that contains bitmap data
 *
 * Returns: errno
 */

static int compute_bitstructs(struct gfs2_rgrpd *rgd)
{
	struct gfs2_sbd *sdp = rgd->rd_sbd;
	struct gfs2_bitmap *bi;
390
	u32 length = rgd->rd_length; /* # blocks in hdr & bitmap */
391
	u32 bytes_left, bytes;
D
David Teigland 已提交
392 393
	int x;

394 395 396
	if (!length)
		return -EINVAL;

397
	rgd->rd_bits = kcalloc(length, sizeof(struct gfs2_bitmap), GFP_NOFS);
D
David Teigland 已提交
398 399 400
	if (!rgd->rd_bits)
		return -ENOMEM;

401
	bytes_left = rgd->rd_bitbytes;
D
David Teigland 已提交
402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421

	for (x = 0; x < length; x++) {
		bi = rgd->rd_bits + x;

		/* small rgrp; bitmap stored completely in header block */
		if (length == 1) {
			bytes = bytes_left;
			bi->bi_offset = sizeof(struct gfs2_rgrp);
			bi->bi_start = 0;
			bi->bi_len = bytes;
		/* header block */
		} else if (x == 0) {
			bytes = sdp->sd_sb.sb_bsize - sizeof(struct gfs2_rgrp);
			bi->bi_offset = sizeof(struct gfs2_rgrp);
			bi->bi_start = 0;
			bi->bi_len = bytes;
		/* last block */
		} else if (x + 1 == length) {
			bytes = bytes_left;
			bi->bi_offset = sizeof(struct gfs2_meta_header);
422
			bi->bi_start = rgd->rd_bitbytes - bytes_left;
D
David Teigland 已提交
423 424 425
			bi->bi_len = bytes;
		/* other blocks */
		} else {
426 427
			bytes = sdp->sd_sb.sb_bsize -
				sizeof(struct gfs2_meta_header);
D
David Teigland 已提交
428
			bi->bi_offset = sizeof(struct gfs2_meta_header);
429
			bi->bi_start = rgd->rd_bitbytes - bytes_left;
D
David Teigland 已提交
430 431 432 433 434 435 436 437 438 439 440
			bi->bi_len = bytes;
		}

		bytes_left -= bytes;
	}

	if (bytes_left) {
		gfs2_consist_rgrpd(rgd);
		return -EIO;
	}
	bi = rgd->rd_bits + (length - 1);
441
	if ((bi->bi_start + bi->bi_len) * GFS2_NBBY != rgd->rd_data) {
D
David Teigland 已提交
442
		if (gfs2_consist_rgrpd(rgd)) {
443
			gfs2_rindex_print(rgd);
D
David Teigland 已提交
444 445 446 447 448 449 450 451 452
			fs_err(sdp, "start=%u len=%u offset=%u\n",
			       bi->bi_start, bi->bi_len, bi->bi_offset);
		}
		return -EIO;
	}

	return 0;
}

453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476
/**
 * gfs2_ri_total - Total up the file system space, according to the rindex.
 *
 */
u64 gfs2_ri_total(struct gfs2_sbd *sdp)
{
	u64 total_data = 0;	
	struct inode *inode = sdp->sd_rindex;
	struct gfs2_inode *ip = GFS2_I(inode);
	char buf[sizeof(struct gfs2_rindex)];
	struct file_ra_state ra_state;
	int error, rgrps;

	mutex_lock(&sdp->sd_rindex_mutex);
	file_ra_state_init(&ra_state, inode->i_mapping);
	for (rgrps = 0;; rgrps++) {
		loff_t pos = rgrps * sizeof(struct gfs2_rindex);

		if (pos + sizeof(struct gfs2_rindex) >= ip->i_di.di_size)
			break;
		error = gfs2_internal_read(ip, &ra_state, buf, &pos,
					   sizeof(struct gfs2_rindex));
		if (error != sizeof(struct gfs2_rindex))
			break;
477
		total_data += be32_to_cpu(((struct gfs2_rindex *)buf)->ri_data);
478 479 480 481 482
	}
	mutex_unlock(&sdp->sd_rindex_mutex);
	return total_data;
}

483 484 485 486 487 488 489 490 491 492 493
static void gfs2_rindex_in(struct gfs2_rgrpd *rgd, const void *buf)
{
	const struct gfs2_rindex *str = buf;

	rgd->rd_addr = be64_to_cpu(str->ri_addr);
	rgd->rd_length = be32_to_cpu(str->ri_length);
	rgd->rd_data0 = be64_to_cpu(str->ri_data0);
	rgd->rd_data = be32_to_cpu(str->ri_data);
	rgd->rd_bitbytes = be32_to_cpu(str->ri_bitbytes);
}

D
David Teigland 已提交
494
/**
495
 * read_rindex_entry - Pull in a new resource index entry from the disk
D
David Teigland 已提交
496 497
 * @gl: The glock covering the rindex inode
 *
498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519
 * Returns: 0 on success, error code otherwise
 */

static int read_rindex_entry(struct gfs2_inode *ip,
			     struct file_ra_state *ra_state)
{
	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
	loff_t pos = sdp->sd_rgrps * sizeof(struct gfs2_rindex);
	char buf[sizeof(struct gfs2_rindex)];
	int error;
	struct gfs2_rgrpd *rgd;

	error = gfs2_internal_read(ip, ra_state, buf, &pos,
				   sizeof(struct gfs2_rindex));
	if (!error)
		return 0;
	if (error != sizeof(struct gfs2_rindex)) {
		if (error > 0)
			error = -EIO;
		return error;
	}

520
	rgd = kmem_cache_zalloc(gfs2_rgrpd_cachep, GFP_NOFS);
521 522 523 524 525 526 527 528 529 530 531
	error = -ENOMEM;
	if (!rgd)
		return error;

	mutex_init(&rgd->rd_mutex);
	lops_init_le(&rgd->rd_le, &gfs2_rg_lops);
	rgd->rd_sbd = sdp;

	list_add_tail(&rgd->rd_list, &sdp->sd_rindex_list);
	list_add_tail(&rgd->rd_list_mru, &sdp->sd_rindex_mru_list);

532
	gfs2_rindex_in(rgd, buf);
533 534 535 536
	error = compute_bitstructs(rgd);
	if (error)
		return error;

537
	error = gfs2_glock_get(sdp, rgd->rd_addr,
538 539 540 541 542
			       &gfs2_rgrp_glops, CREATE, &rgd->rd_gl);
	if (error)
		return error;

	rgd->rd_gl->gl_object = rgd;
543
	rgd->rd_flags &= ~GFS2_RDF_UPTODATE;
544
	rgd->rd_flags |= GFS2_RDF_CHECK;
545 546 547 548 549 550 551
	return error;
}

/**
 * gfs2_ri_update - Pull in a new resource index from the disk
 * @ip: pointer to the rindex inode
 *
D
David Teigland 已提交
552 553 554 555 556
 * Returns: 0 on successful update, error code otherwise
 */

static int gfs2_ri_update(struct gfs2_inode *ip)
{
557 558
	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
	struct inode *inode = &ip->i_inode;
559
	struct file_ra_state ra_state;
560
	u64 rgrp_count = ip->i_di.di_size;
D
David Teigland 已提交
561 562
	int error;

563
	if (do_div(rgrp_count, sizeof(struct gfs2_rindex))) {
D
David Teigland 已提交
564 565 566 567 568 569
		gfs2_consist_inode(ip);
		return -EIO;
	}

	clear_rgrpdi(sdp);

570
	file_ra_state_init(&ra_state, inode->i_mapping);
571
	for (sdp->sd_rgrps = 0; sdp->sd_rgrps < rgrp_count; sdp->sd_rgrps++) {
572 573 574 575
		error = read_rindex_entry(ip, &ra_state);
		if (error) {
			clear_rgrpdi(sdp);
			return error;
D
David Teigland 已提交
576
		}
577
	}
D
David Teigland 已提交
578

579
	sdp->sd_rindex_uptodate = 1;
580 581
	return 0;
}
D
David Teigland 已提交
582

583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598
/**
 * gfs2_ri_update_special - Pull in a new resource index from the disk
 *
 * This is a special version that's safe to call from gfs2_inplace_reserve_i.
 * In this case we know that we don't have any resource groups in memory yet.
 *
 * @ip: pointer to the rindex inode
 *
 * Returns: 0 on successful update, error code otherwise
 */
static int gfs2_ri_update_special(struct gfs2_inode *ip)
{
	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
	struct inode *inode = &ip->i_inode;
	struct file_ra_state ra_state;
	int error;
D
David Teigland 已提交
599

600 601 602 603 604 605 606 607 608 609 610
	file_ra_state_init(&ra_state, inode->i_mapping);
	for (sdp->sd_rgrps = 0;; sdp->sd_rgrps++) {
		/* Ignore partials */
		if ((sdp->sd_rgrps + 1) * sizeof(struct gfs2_rindex) >
		    ip->i_di.di_size)
			break;
		error = read_rindex_entry(ip, &ra_state);
		if (error) {
			clear_rgrpdi(sdp);
			return error;
		}
D
David Teigland 已提交
611 612
	}

613
	sdp->sd_rindex_uptodate = 1;
D
David Teigland 已提交
614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636
	return 0;
}

/**
 * gfs2_rindex_hold - Grab a lock on the rindex
 * @sdp: The GFS2 superblock
 * @ri_gh: the glock holder
 *
 * We grab a lock on the rindex inode to make sure that it doesn't
 * change whilst we are performing an operation. We keep this lock
 * for quite long periods of time compared to other locks. This
 * doesn't matter, since it is shared and it is very, very rarely
 * accessed in the exclusive mode (i.e. only when expanding the filesystem).
 *
 * This makes sure that we're using the latest copy of the resource index
 * special file, which might have been updated if someone expanded the
 * filesystem (via gfs2_grow utility), which adds new resource groups.
 *
 * Returns: 0 on success, error code otherwise
 */

int gfs2_rindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ri_gh)
{
637
	struct gfs2_inode *ip = GFS2_I(sdp->sd_rindex);
D
David Teigland 已提交
638 639 640 641 642 643 644 645
	struct gfs2_glock *gl = ip->i_gl;
	int error;

	error = gfs2_glock_nq_init(gl, LM_ST_SHARED, 0, ri_gh);
	if (error)
		return error;

	/* Read new copy from disk if we don't have the latest */
646
	if (!sdp->sd_rindex_uptodate) {
647
		mutex_lock(&sdp->sd_rindex_mutex);
648
		if (!sdp->sd_rindex_uptodate) {
D
David Teigland 已提交
649 650 651 652
			error = gfs2_ri_update(ip);
			if (error)
				gfs2_glock_dq_uninit(ri_gh);
		}
653
		mutex_unlock(&sdp->sd_rindex_mutex);
D
David Teigland 已提交
654 655 656 657 658
	}

	return error;
}

659
static void gfs2_rgrp_in(struct gfs2_rgrpd *rgd, const void *buf)
660 661
{
	const struct gfs2_rgrp *str = buf;
662 663
	struct gfs2_rgrp_host *rg = &rgd->rd_rg;
	u32 rg_flags;
664

665 666 667 668 669
	rg_flags = be32_to_cpu(str->rg_flags);
	if (rg_flags & GFS2_RGF_NOALLOC)
		rgd->rd_flags |= GFS2_RDF_NOALLOC;
	else
		rgd->rd_flags &= ~GFS2_RDF_NOALLOC;
670 671 672 673 674
	rg->rg_free = be32_to_cpu(str->rg_free);
	rg->rg_dinodes = be32_to_cpu(str->rg_dinodes);
	rg->rg_igeneration = be64_to_cpu(str->rg_igeneration);
}

675
static void gfs2_rgrp_out(struct gfs2_rgrpd *rgd, void *buf)
676 677
{
	struct gfs2_rgrp *str = buf;
678 679
	struct gfs2_rgrp_host *rg = &rgd->rd_rg;
	u32 rg_flags = 0;
680

681 682 683
	if (rgd->rd_flags & GFS2_RDF_NOALLOC)
		rg_flags |= GFS2_RGF_NOALLOC;
	str->rg_flags = cpu_to_be32(rg_flags);
684 685 686 687 688 689 690
	str->rg_free = cpu_to_be32(rg->rg_free);
	str->rg_dinodes = cpu_to_be32(rg->rg_dinodes);
	str->__pad = cpu_to_be32(0);
	str->rg_igeneration = cpu_to_be64(rg->rg_igeneration);
	memset(&str->rg_reserved, 0, sizeof(str->rg_reserved));
}

D
David Teigland 已提交
691 692 693 694 695 696 697 698 699 700 701 702 703 704
/**
 * gfs2_rgrp_bh_get - Read in a RG's header and bitmaps
 * @rgd: the struct gfs2_rgrpd describing the RG to read in
 *
 * Read in all of a Resource Group's header and bitmap blocks.
 * Caller must eventually call gfs2_rgrp_relse() to free the bitmaps.
 *
 * Returns: errno
 */

int gfs2_rgrp_bh_get(struct gfs2_rgrpd *rgd)
{
	struct gfs2_sbd *sdp = rgd->rd_sbd;
	struct gfs2_glock *gl = rgd->rd_gl;
705
	unsigned int length = rgd->rd_length;
D
David Teigland 已提交
706 707 708 709
	struct gfs2_bitmap *bi;
	unsigned int x, y;
	int error;

710
	mutex_lock(&rgd->rd_mutex);
D
David Teigland 已提交
711 712 713 714 715

	spin_lock(&sdp->sd_rindex_spin);
	if (rgd->rd_bh_count) {
		rgd->rd_bh_count++;
		spin_unlock(&sdp->sd_rindex_spin);
716
		mutex_unlock(&rgd->rd_mutex);
D
David Teigland 已提交
717 718 719 720 721 722
		return 0;
	}
	spin_unlock(&sdp->sd_rindex_spin);

	for (x = 0; x < length; x++) {
		bi = rgd->rd_bits + x;
723
		error = gfs2_meta_read(gl, rgd->rd_addr + x, 0, &bi->bi_bh);
D
David Teigland 已提交
724 725 726 727 728 729
		if (error)
			goto fail;
	}

	for (y = length; y--;) {
		bi = rgd->rd_bits + y;
S
Steven Whitehouse 已提交
730
		error = gfs2_meta_wait(sdp, bi->bi_bh);
D
David Teigland 已提交
731 732
		if (error)
			goto fail;
733
		if (gfs2_metatype_check(sdp, bi->bi_bh, y ? GFS2_METATYPE_RB :
D
David Teigland 已提交
734 735 736 737 738 739
					      GFS2_METATYPE_RG)) {
			error = -EIO;
			goto fail;
		}
	}

740
	if (!(rgd->rd_flags & GFS2_RDF_UPTODATE)) {
741
		gfs2_rgrp_in(rgd, (rgd->rd_bits[0].bi_bh)->b_data);
742
		rgd->rd_flags |= GFS2_RDF_UPTODATE;
D
David Teigland 已提交
743 744 745 746 747 748 749
	}

	spin_lock(&sdp->sd_rindex_spin);
	rgd->rd_free_clone = rgd->rd_rg.rg_free;
	rgd->rd_bh_count++;
	spin_unlock(&sdp->sd_rindex_spin);

750
	mutex_unlock(&rgd->rd_mutex);
D
David Teigland 已提交
751 752 753

	return 0;

754
fail:
D
David Teigland 已提交
755 756 757 758 759 760
	while (x--) {
		bi = rgd->rd_bits + x;
		brelse(bi->bi_bh);
		bi->bi_bh = NULL;
		gfs2_assert_warn(sdp, !bi->bi_clone);
	}
761
	mutex_unlock(&rgd->rd_mutex);
D
David Teigland 已提交
762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784

	return error;
}

void gfs2_rgrp_bh_hold(struct gfs2_rgrpd *rgd)
{
	struct gfs2_sbd *sdp = rgd->rd_sbd;

	spin_lock(&sdp->sd_rindex_spin);
	gfs2_assert_warn(rgd->rd_sbd, rgd->rd_bh_count);
	rgd->rd_bh_count++;
	spin_unlock(&sdp->sd_rindex_spin);
}

/**
 * gfs2_rgrp_bh_put - Release RG bitmaps read in with gfs2_rgrp_bh_get()
 * @rgd: the struct gfs2_rgrpd describing the RG to read in
 *
 */

void gfs2_rgrp_bh_put(struct gfs2_rgrpd *rgd)
{
	struct gfs2_sbd *sdp = rgd->rd_sbd;
785
	int x, length = rgd->rd_length;
D
David Teigland 已提交
786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807

	spin_lock(&sdp->sd_rindex_spin);
	gfs2_assert_warn(rgd->rd_sbd, rgd->rd_bh_count);
	if (--rgd->rd_bh_count) {
		spin_unlock(&sdp->sd_rindex_spin);
		return;
	}

	for (x = 0; x < length; x++) {
		struct gfs2_bitmap *bi = rgd->rd_bits + x;
		kfree(bi->bi_clone);
		bi->bi_clone = NULL;
		brelse(bi->bi_bh);
		bi->bi_bh = NULL;
	}

	spin_unlock(&sdp->sd_rindex_spin);
}

void gfs2_rgrp_repolish_clones(struct gfs2_rgrpd *rgd)
{
	struct gfs2_sbd *sdp = rgd->rd_sbd;
808
	unsigned int length = rgd->rd_length;
D
David Teigland 已提交
809 810 811 812 813 814 815
	unsigned int x;

	for (x = 0; x < length; x++) {
		struct gfs2_bitmap *bi = rgd->rd_bits + x;
		if (!bi->bi_clone)
			continue;
		memcpy(bi->bi_clone + bi->bi_offset,
816
		       bi->bi_bh->b_data + bi->bi_offset, bi->bi_len);
D
David Teigland 已提交
817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832
	}

	spin_lock(&sdp->sd_rindex_spin);
	rgd->rd_free_clone = rgd->rd_rg.rg_free;
	spin_unlock(&sdp->sd_rindex_spin);
}

/**
 * gfs2_alloc_get - get the struct gfs2_alloc structure for an inode
 * @ip: the incore GFS2 inode structure
 *
 * Returns: the struct gfs2_alloc
 */

struct gfs2_alloc *gfs2_alloc_get(struct gfs2_inode *ip)
{
833 834 835
	BUG_ON(ip->i_alloc != NULL);
	ip->i_alloc = kzalloc(sizeof(struct gfs2_alloc), GFP_KERNEL);
	return ip->i_alloc;
D
David Teigland 已提交
836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853
}

/**
 * try_rgrp_fit - See if a given reservation will fit in a given RG
 * @rgd: the RG data
 * @al: the struct gfs2_alloc structure describing the reservation
 *
 * If there's room for the requested blocks to be allocated from the RG:
 *   Sets the $al_rgd field in @al.
 *
 * Returns: 1 on success (it fits), 0 on failure (it doesn't fit)
 */

static int try_rgrp_fit(struct gfs2_rgrpd *rgd, struct gfs2_alloc *al)
{
	struct gfs2_sbd *sdp = rgd->rd_sbd;
	int ret = 0;

854
	if (rgd->rd_flags & GFS2_RDF_NOALLOC)
855 856
		return 0;

D
David Teigland 已提交
857 858 859 860 861 862 863 864 865 866
	spin_lock(&sdp->sd_rindex_spin);
	if (rgd->rd_free_clone >= al->al_requested) {
		al->al_rgd = rgd;
		ret = 1;
	}
	spin_unlock(&sdp->sd_rindex_spin);

	return ret;
}

867 868 869 870 871 872 873 874 875 876
/**
 * try_rgrp_unlink - Look for any unlinked, allocated, but unused inodes
 * @rgd: The rgrp
 *
 * Returns: The inode, if one has been found
 */

static struct inode *try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked)
{
	struct inode *inode;
877
	u32 goal = 0, block;
878
	u64 no_addr;
879
	struct gfs2_sbd *sdp = rgd->rd_sbd;
880 881

	for(;;) {
882 883
		if (goal >= rgd->rd_data)
			break;
884
		down_write(&sdp->sd_log_flush_lock);
885 886
		block = rgblk_search(rgd, goal, GFS2_BLKST_UNLINKED,
				     GFS2_BLKST_UNLINKED);
887
		up_write(&sdp->sd_log_flush_lock);
888
		if (block == BFITNOENT)
889
			break;
890 891 892
		/* rgblk_search can return a block < goal, so we need to
		   keep it marching forward. */
		no_addr = block + rgd->rd_data0;
893
		goal++;
894
		if (*last_unlinked != NO_BLOCK && no_addr <= *last_unlinked)
895
			continue;
896 897
		*last_unlinked = no_addr;
		inode = gfs2_inode_lookup(rgd->rd_sbd->sd_vfs, DT_UNKNOWN,
898
					  no_addr, -1, 1);
899 900 901 902 903 904 905 906
		if (!IS_ERR(inode))
			return inode;
	}

	rgd->rd_flags &= ~GFS2_RDF_CHECK;
	return NULL;
}

D
David Teigland 已提交
907 908 909 910 911 912 913 914 915
/**
 * recent_rgrp_first - get first RG from "recent" list
 * @sdp: The GFS2 superblock
 * @rglast: address of the rgrp used last
 *
 * Returns: The first rgrp in the recent list
 */

static struct gfs2_rgrpd *recent_rgrp_first(struct gfs2_sbd *sdp,
916
					    u64 rglast)
D
David Teigland 已提交
917
{
918
	struct gfs2_rgrpd *rgd;
D
David Teigland 已提交
919 920 921

	spin_lock(&sdp->sd_rindex_spin);

922 923 924 925 926
	if (rglast) {
		list_for_each_entry(rgd, &sdp->sd_rindex_recent_list, rd_recent) {
			if (rgrp_contains_block(rgd, rglast))
				goto out;
		}
D
David Teigland 已提交
927
	}
928 929 930 931
	rgd = NULL;
	if (!list_empty(&sdp->sd_rindex_recent_list))
		rgd = list_entry(sdp->sd_rindex_recent_list.next,
				 struct gfs2_rgrpd, rd_recent);
932
out:
D
David Teigland 已提交
933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974
	spin_unlock(&sdp->sd_rindex_spin);
	return rgd;
}

/**
 * recent_rgrp_next - get next RG from "recent" list
 * @cur_rgd: current rgrp
 * @remove:
 *
 * Returns: The next rgrp in the recent list
 */

static struct gfs2_rgrpd *recent_rgrp_next(struct gfs2_rgrpd *cur_rgd,
					   int remove)
{
	struct gfs2_sbd *sdp = cur_rgd->rd_sbd;
	struct list_head *head;
	struct gfs2_rgrpd *rgd;

	spin_lock(&sdp->sd_rindex_spin);

	head = &sdp->sd_rindex_recent_list;

	list_for_each_entry(rgd, head, rd_recent) {
		if (rgd == cur_rgd) {
			if (cur_rgd->rd_recent.next != head)
				rgd = list_entry(cur_rgd->rd_recent.next,
						 struct gfs2_rgrpd, rd_recent);
			else
				rgd = NULL;

			if (remove)
				list_del(&cur_rgd->rd_recent);

			goto out;
		}
	}

	rgd = NULL;
	if (!list_empty(head))
		rgd = list_entry(head->next, struct gfs2_rgrpd, rd_recent);

975
out:
D
David Teigland 已提交
976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003
	spin_unlock(&sdp->sd_rindex_spin);
	return rgd;
}

/**
 * recent_rgrp_add - add an RG to tail of "recent" list
 * @new_rgd: The rgrp to add
 *
 */

static void recent_rgrp_add(struct gfs2_rgrpd *new_rgd)
{
	struct gfs2_sbd *sdp = new_rgd->rd_sbd;
	struct gfs2_rgrpd *rgd;
	unsigned int count = 0;
	unsigned int max = sdp->sd_rgrps / gfs2_jindex_size(sdp);

	spin_lock(&sdp->sd_rindex_spin);

	list_for_each_entry(rgd, &sdp->sd_rindex_recent_list, rd_recent) {
		if (rgd == new_rgd)
			goto out;

		if (++count >= max)
			goto out;
	}
	list_add_tail(&new_rgd->rd_recent, &sdp->sd_rindex_recent_list);

1004
out:
D
David Teigland 已提交
1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027
	spin_unlock(&sdp->sd_rindex_spin);
}

/**
 * forward_rgrp_get - get an rgrp to try next from full list
 * @sdp: The GFS2 superblock
 *
 * Returns: The rgrp to try next
 */

static struct gfs2_rgrpd *forward_rgrp_get(struct gfs2_sbd *sdp)
{
	struct gfs2_rgrpd *rgd;
	unsigned int journals = gfs2_jindex_size(sdp);
	unsigned int rg = 0, x;

	spin_lock(&sdp->sd_rindex_spin);

	rgd = sdp->sd_rindex_forward;
	if (!rgd) {
		if (sdp->sd_rgrps >= journals)
			rg = sdp->sd_rgrps * sdp->sd_jdesc->jd_jid / journals;

S
Steven Whitehouse 已提交
1028
		for (x = 0, rgd = gfs2_rgrpd_get_first(sdp); x < rg;
D
David Teigland 已提交
1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063
		     x++, rgd = gfs2_rgrpd_get_next(rgd))
			/* Do Nothing */;

		sdp->sd_rindex_forward = rgd;
	}

	spin_unlock(&sdp->sd_rindex_spin);

	return rgd;
}

/**
 * forward_rgrp_set - set the forward rgrp pointer
 * @sdp: the filesystem
 * @rgd: The new forward rgrp
 *
 */

static void forward_rgrp_set(struct gfs2_sbd *sdp, struct gfs2_rgrpd *rgd)
{
	spin_lock(&sdp->sd_rindex_spin);
	sdp->sd_rindex_forward = rgd;
	spin_unlock(&sdp->sd_rindex_spin);
}

/**
 * get_local_rgrp - Choose and lock a rgrp for allocation
 * @ip: the inode to reserve space for
 * @rgp: the chosen and locked rgrp
 *
 * Try to acquire rgrp in way which avoids contending with others.
 *
 * Returns: errno
 */

1064
static struct inode *get_local_rgrp(struct gfs2_inode *ip, u64 *last_unlinked)
D
David Teigland 已提交
1065
{
1066
	struct inode *inode = NULL;
1067
	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
D
David Teigland 已提交
1068
	struct gfs2_rgrpd *rgd, *begin = NULL;
1069
	struct gfs2_alloc *al = ip->i_alloc;
D
David Teigland 已提交
1070 1071 1072
	int flags = LM_FLAG_TRY;
	int skipped = 0;
	int loops = 0;
1073
	int error, rg_locked;
D
David Teigland 已提交
1074 1075 1076

	/* Try recently successful rgrps */

1077
	rgd = recent_rgrp_first(sdp, ip->i_goal);
D
David Teigland 已提交
1078 1079

	while (rgd) {
1080 1081 1082 1083 1084 1085 1086 1087 1088
		rg_locked = 0;

		if (gfs2_glock_is_locked_by_me(rgd->rd_gl)) {
			rg_locked = 1;
			error = 0;
		} else {
			error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE,
						   LM_FLAG_TRY, &al->al_rgd_gh);
		}
D
David Teigland 已提交
1089 1090 1091 1092
		switch (error) {
		case 0:
			if (try_rgrp_fit(rgd, al))
				goto out;
1093 1094
			if (rgd->rd_flags & GFS2_RDF_CHECK)
				inode = try_rgrp_unlink(rgd, last_unlinked);
1095 1096
			if (!rg_locked)
				gfs2_glock_dq_uninit(&al->al_rgd_gh);
1097 1098
			if (inode)
				return inode;
D
David Teigland 已提交
1099 1100 1101 1102 1103 1104 1105 1106
			rgd = recent_rgrp_next(rgd, 1);
			break;

		case GLR_TRYFAILED:
			rgd = recent_rgrp_next(rgd, 0);
			break;

		default:
1107
			return ERR_PTR(error);
D
David Teigland 已提交
1108 1109 1110 1111 1112 1113 1114 1115
		}
	}

	/* Go through full list of rgrps */

	begin = rgd = forward_rgrp_get(sdp);

	for (;;) {
1116 1117 1118 1119 1120 1121 1122 1123 1124
		rg_locked = 0;

		if (gfs2_glock_is_locked_by_me(rgd->rd_gl)) {
			rg_locked = 1;
			error = 0;
		} else {
			error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, flags,
						   &al->al_rgd_gh);
		}
D
David Teigland 已提交
1125 1126 1127 1128
		switch (error) {
		case 0:
			if (try_rgrp_fit(rgd, al))
				goto out;
1129 1130
			if (rgd->rd_flags & GFS2_RDF_CHECK)
				inode = try_rgrp_unlink(rgd, last_unlinked);
1131 1132
			if (!rg_locked)
				gfs2_glock_dq_uninit(&al->al_rgd_gh);
1133 1134
			if (inode)
				return inode;
D
David Teigland 已提交
1135 1136 1137 1138 1139 1140 1141
			break;

		case GLR_TRYFAILED:
			skipped++;
			break;

		default:
1142
			return ERR_PTR(error);
D
David Teigland 已提交
1143 1144 1145 1146 1147 1148 1149
		}

		rgd = gfs2_rgrpd_get_next(rgd);
		if (!rgd)
			rgd = gfs2_rgrpd_get_first(sdp);

		if (rgd == begin) {
1150
			if (++loops >= 3)
1151
				return ERR_PTR(-ENOSPC);
1152 1153
			if (!skipped)
				loops++;
D
David Teigland 已提交
1154
			flags = 0;
1155 1156
			if (loops == 2)
				gfs2_log_flush(sdp, NULL);
D
David Teigland 已提交
1157 1158 1159
		}
	}

1160
out:
D
David Teigland 已提交
1161 1162 1163 1164 1165 1166 1167 1168
	if (begin) {
		recent_rgrp_add(rgd);
		rgd = gfs2_rgrpd_get_next(rgd);
		if (!rgd)
			rgd = gfs2_rgrpd_get_first(sdp);
		forward_rgrp_set(sdp, rgd);
	}

1169
	return NULL;
D
David Teigland 已提交
1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180
}

/**
 * gfs2_inplace_reserve_i - Reserve space in the filesystem
 * @ip: the inode to reserve space for
 *
 * Returns: errno
 */

int gfs2_inplace_reserve_i(struct gfs2_inode *ip, char *file, unsigned int line)
{
1181
	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
1182
	struct gfs2_alloc *al = ip->i_alloc;
1183
	struct inode *inode;
1184
	int error = 0;
1185
	u64 last_unlinked = NO_BLOCK;
D
David Teigland 已提交
1186 1187 1188 1189

	if (gfs2_assert_warn(sdp, al->al_requested))
		return -EINVAL;

1190
try_again:
1191 1192 1193 1194 1195
	/* We need to hold the rindex unless the inode we're using is
	   the rindex itself, in which case it's already held. */
	if (ip != GFS2_I(sdp->sd_rindex))
		error = gfs2_rindex_hold(sdp, &al->al_ri_gh);
	else if (!sdp->sd_rgrps) /* We may not have the rindex read in, so: */
1196
		error = gfs2_ri_update_special(ip);
1197

D
David Teigland 已提交
1198 1199 1200
	if (error)
		return error;

1201 1202
	inode = get_local_rgrp(ip, &last_unlinked);
	if (inode) {
1203 1204
		if (ip != GFS2_I(sdp->sd_rindex))
			gfs2_glock_dq_uninit(&al->al_ri_gh);
1205 1206 1207 1208 1209
		if (IS_ERR(inode))
			return PTR_ERR(inode);
		iput(inode);
		gfs2_log_flush(sdp, NULL);
		goto try_again;
D
David Teigland 已提交
1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226
	}

	al->al_file = file;
	al->al_line = line;

	return 0;
}

/**
 * gfs2_inplace_release - release an inplace reservation
 * @ip: the inode the reservation was taken out on
 *
 * Release a reservation made by gfs2_inplace_reserve().
 */

void gfs2_inplace_release(struct gfs2_inode *ip)
{
1227
	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
1228
	struct gfs2_alloc *al = ip->i_alloc;
D
David Teigland 已提交
1229 1230 1231 1232 1233 1234 1235 1236

	if (gfs2_assert_warn(sdp, al->al_alloced <= al->al_requested) == -1)
		fs_warn(sdp, "al_alloced = %u, al_requested = %u "
			     "al_file = %s, al_line = %u\n",
		             al->al_alloced, al->al_requested, al->al_file,
			     al->al_line);

	al->al_rgd = NULL;
1237 1238
	if (al->al_rgd_gh.gh_gl)
		gfs2_glock_dq_uninit(&al->al_rgd_gh);
1239 1240
	if (ip != GFS2_I(sdp->sd_rindex))
		gfs2_glock_dq_uninit(&al->al_ri_gh);
D
David Teigland 已提交
1241 1242 1243 1244 1245 1246 1247 1248 1249 1250
}

/**
 * gfs2_get_block_type - Check a block in a RG is of given type
 * @rgd: the resource group holding the block
 * @block: the block number
 *
 * Returns: The block type (GFS2_BLKST_*)
 */

1251
unsigned char gfs2_get_block_type(struct gfs2_rgrpd *rgd, u64 block)
D
David Teigland 已提交
1252 1253
{
	struct gfs2_bitmap *bi = NULL;
1254
	u32 length, rgrp_block, buf_block;
D
David Teigland 已提交
1255 1256 1257
	unsigned int buf;
	unsigned char type;

1258 1259
	length = rgd->rd_length;
	rgrp_block = block - rgd->rd_data0;
D
David Teigland 已提交
1260 1261 1262 1263 1264 1265 1266 1267 1268 1269

	for (buf = 0; buf < length; buf++) {
		bi = rgd->rd_bits + buf;
		if (rgrp_block < (bi->bi_start + bi->bi_len) * GFS2_NBBY)
			break;
	}

	gfs2_assert(rgd->rd_sbd, buf < length);
	buf_block = rgrp_block - bi->bi_start * GFS2_NBBY;

1270
	type = gfs2_testbit(rgd, bi->bi_bh->b_data + bi->bi_offset,
D
David Teigland 已提交
1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296
			   bi->bi_len, buf_block);

	return type;
}

/**
 * rgblk_search - find a block in @old_state, change allocation
 *           state to @new_state
 * @rgd: the resource group descriptor
 * @goal: the goal block within the RG (start here to search for avail block)
 * @old_state: GFS2_BLKST_XXX the before-allocation state to find
 * @new_state: GFS2_BLKST_XXX the after-allocation block state
 *
 * Walk rgrp's bitmap to find bits that represent a block in @old_state.
 * Add the found bitmap buffer to the transaction.
 * Set the found bits to @new_state to change block's allocation state.
 *
 * This function never fails, because we wouldn't call it unless we
 * know (from reservation results, etc.) that a block is available.
 *
 * Scope of @goal and returned block is just within rgrp, not the whole
 * filesystem.
 *
 * Returns:  the block number allocated
 */

1297
static u32 rgblk_search(struct gfs2_rgrpd *rgd, u32 goal,
1298
			unsigned char old_state, unsigned char new_state)
D
David Teigland 已提交
1299 1300
{
	struct gfs2_bitmap *bi = NULL;
1301
	u32 length = rgd->rd_length;
1302
	u32 blk = 0;
D
David Teigland 已提交
1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322
	unsigned int buf, x;

	/* Find bitmap block that contains bits for goal block */
	for (buf = 0; buf < length; buf++) {
		bi = rgd->rd_bits + buf;
		if (goal < (bi->bi_start + bi->bi_len) * GFS2_NBBY)
			break;
	}

	gfs2_assert(rgd->rd_sbd, buf < length);

	/* Convert scope of "goal" from rgrp-wide to within found bit block */
	goal -= bi->bi_start * GFS2_NBBY;

	/* Search (up to entire) bitmap in this rgrp for allocatable block.
	   "x <= length", instead of "x < length", because we typically start
	   the search in the middle of a bit block, but if we can't find an
	   allocatable block anywhere else, we want to be able wrap around and
	   search in the first part of our first-searched bit block.  */
	for (x = 0; x <= length; x++) {
1323 1324
		/* The GFS2_BLKST_UNLINKED state doesn't apply to the clone
		   bitmaps, so we must search the originals for that. */
1325
		const u8 *buffer = bi->bi_bh->b_data + bi->bi_offset;
1326
		if (old_state != GFS2_BLKST_UNLINKED && bi->bi_clone)
1327 1328 1329
			buffer = bi->bi_clone + bi->bi_offset;

		blk = gfs2_bitfit(buffer, bi->bi_len, goal, old_state);
D
David Teigland 已提交
1330 1331 1332 1333 1334 1335 1336 1337 1338
		if (blk != BFITNOENT)
			break;

		/* Try next bitmap block (wrap back to rgrp header if at end) */
		buf = (buf + 1) % length;
		bi = rgd->rd_bits + buf;
		goal = 0;
	}

1339
	if (blk != BFITNOENT && old_state != new_state) {
1340 1341
		gfs2_trans_add_bh(rgd->rd_gl, bi->bi_bh, 1);
		gfs2_setbit(rgd, bi->bi_bh->b_data + bi->bi_offset,
D
David Teigland 已提交
1342
			    bi->bi_len, blk, new_state);
1343 1344 1345 1346
		if (bi->bi_clone)
			gfs2_setbit(rgd, bi->bi_clone + bi->bi_offset,
				    bi->bi_len, blk, new_state);
	}
D
David Teigland 已提交
1347

1348
	return (blk == BFITNOENT) ? blk : (bi->bi_start * GFS2_NBBY) + blk;
D
David Teigland 已提交
1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360
}

/**
 * rgblk_free - Change alloc state of given block(s)
 * @sdp: the filesystem
 * @bstart: the start of a run of blocks to free
 * @blen: the length of the block run (all must lie within ONE RG!)
 * @new_state: GFS2_BLKST_XXX the after-allocation block state
 *
 * Returns:  Resource group containing the block(s)
 */

1361 1362
static struct gfs2_rgrpd *rgblk_free(struct gfs2_sbd *sdp, u64 bstart,
				     u32 blen, unsigned char new_state)
D
David Teigland 已提交
1363 1364 1365
{
	struct gfs2_rgrpd *rgd;
	struct gfs2_bitmap *bi = NULL;
1366
	u32 length, rgrp_blk, buf_blk;
D
David Teigland 已提交
1367 1368 1369 1370 1371
	unsigned int buf;

	rgd = gfs2_blk2rgrpd(sdp, bstart);
	if (!rgd) {
		if (gfs2_consist(sdp))
1372
			fs_err(sdp, "block = %llu\n", (unsigned long long)bstart);
D
David Teigland 已提交
1373 1374 1375
		return NULL;
	}

1376
	length = rgd->rd_length;
D
David Teigland 已提交
1377

1378
	rgrp_blk = bstart - rgd->rd_data0;
D
David Teigland 已提交
1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393

	while (blen--) {
		for (buf = 0; buf < length; buf++) {
			bi = rgd->rd_bits + buf;
			if (rgrp_blk < (bi->bi_start + bi->bi_len) * GFS2_NBBY)
				break;
		}

		gfs2_assert(rgd->rd_sbd, buf < length);

		buf_blk = rgrp_blk - bi->bi_start * GFS2_NBBY;
		rgrp_blk++;

		if (!bi->bi_clone) {
			bi->bi_clone = kmalloc(bi->bi_bh->b_size,
1394
					       GFP_NOFS | __GFP_NOFAIL);
D
David Teigland 已提交
1395 1396 1397 1398
			memcpy(bi->bi_clone + bi->bi_offset,
			       bi->bi_bh->b_data + bi->bi_offset,
			       bi->bi_len);
		}
1399
		gfs2_trans_add_bh(rgd->rd_gl, bi->bi_bh, 1);
1400
		gfs2_setbit(rgd, bi->bi_bh->b_data + bi->bi_offset,
D
David Teigland 已提交
1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413
			    bi->bi_len, buf_blk, new_state);
	}

	return rgd;
}

/**
 * gfs2_alloc_data - Allocate a data block
 * @ip: the inode to allocate the data block for
 *
 * Returns: the allocated block
 */

S
Steven Whitehouse 已提交
1414
u64 gfs2_alloc_data(struct gfs2_inode *ip)
D
David Teigland 已提交
1415
{
1416
	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
1417
	struct gfs2_alloc *al = ip->i_alloc;
D
David Teigland 已提交
1418
	struct gfs2_rgrpd *rgd = al->al_rgd;
1419 1420
	u32 goal, blk;
	u64 block;
D
David Teigland 已提交
1421

1422 1423
	if (rgrp_contains_block(rgd, ip->i_goal))
		goal = ip->i_goal - rgd->rd_data0;
D
David Teigland 已提交
1424
	else
1425
		goal = rgd->rd_last_alloc;
D
David Teigland 已提交
1426

S
Steven Whitehouse 已提交
1427
	blk = rgblk_search(rgd, goal, GFS2_BLKST_FREE, GFS2_BLKST_USED);
1428
	BUG_ON(blk == BFITNOENT);
1429
	rgd->rd_last_alloc = blk;
D
David Teigland 已提交
1430

1431
	block = rgd->rd_data0 + blk;
1432
	ip->i_goal = block;
D
David Teigland 已提交
1433 1434 1435 1436

	gfs2_assert_withdraw(sdp, rgd->rd_rg.rg_free);
	rgd->rd_rg.rg_free--;

1437
	gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
1438
	gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data);
D
David Teigland 已提交
1439 1440 1441 1442

	al->al_alloced++;

	gfs2_statfs_change(sdp, 0, -1, 0);
1443
	gfs2_quota_change(ip, +1, ip->i_inode.i_uid, ip->i_inode.i_gid);
D
David Teigland 已提交
1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458

	spin_lock(&sdp->sd_rindex_spin);
	rgd->rd_free_clone--;
	spin_unlock(&sdp->sd_rindex_spin);

	return block;
}

/**
 * gfs2_alloc_meta - Allocate a metadata block
 * @ip: the inode to allocate the metadata block for
 *
 * Returns: the allocated block
 */

S
Steven Whitehouse 已提交
1459
u64 gfs2_alloc_meta(struct gfs2_inode *ip)
D
David Teigland 已提交
1460
{
1461
	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
1462
	struct gfs2_alloc *al = ip->i_alloc;
D
David Teigland 已提交
1463
	struct gfs2_rgrpd *rgd = al->al_rgd;
1464 1465
	u32 goal, blk;
	u64 block;
D
David Teigland 已提交
1466

1467 1468
	if (rgrp_contains_block(rgd, ip->i_goal))
		goal = ip->i_goal - rgd->rd_data0;
D
David Teigland 已提交
1469
	else
1470
		goal = rgd->rd_last_alloc;
D
David Teigland 已提交
1471

S
Steven Whitehouse 已提交
1472
	blk = rgblk_search(rgd, goal, GFS2_BLKST_FREE, GFS2_BLKST_USED);
1473
	BUG_ON(blk == BFITNOENT);
1474
	rgd->rd_last_alloc = blk;
D
David Teigland 已提交
1475

1476
	block = rgd->rd_data0 + blk;
1477
	ip->i_goal = block;
D
David Teigland 已提交
1478 1479 1480 1481

	gfs2_assert_withdraw(sdp, rgd->rd_rg.rg_free);
	rgd->rd_rg.rg_free--;

1482
	gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
1483
	gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data);
D
David Teigland 已提交
1484 1485 1486 1487

	al->al_alloced++;

	gfs2_statfs_change(sdp, 0, -1, 0);
1488
	gfs2_quota_change(ip, +1, ip->i_inode.i_uid, ip->i_inode.i_gid);
D
David Teigland 已提交
1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503

	spin_lock(&sdp->sd_rindex_spin);
	rgd->rd_free_clone--;
	spin_unlock(&sdp->sd_rindex_spin);

	return block;
}

/**
 * gfs2_alloc_di - Allocate a dinode
 * @dip: the directory that the inode is going in
 *
 * Returns: the block allocated
 */

S
Steven Whitehouse 已提交
1504
u64 gfs2_alloc_di(struct gfs2_inode *dip, u64 *generation)
D
David Teigland 已提交
1505
{
1506
	struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode);
1507
	struct gfs2_alloc *al = dip->i_alloc;
D
David Teigland 已提交
1508
	struct gfs2_rgrpd *rgd = al->al_rgd;
S
Steven Whitehouse 已提交
1509 1510
	u32 blk;
	u64 block;
D
David Teigland 已提交
1511

1512
	blk = rgblk_search(rgd, rgd->rd_last_alloc,
D
David Teigland 已提交
1513
			   GFS2_BLKST_FREE, GFS2_BLKST_DINODE);
1514
	BUG_ON(blk == BFITNOENT);
D
David Teigland 已提交
1515

1516
	rgd->rd_last_alloc = blk;
D
David Teigland 已提交
1517

1518
	block = rgd->rd_data0 + blk;
D
David Teigland 已提交
1519 1520 1521 1522

	gfs2_assert_withdraw(sdp, rgd->rd_rg.rg_free);
	rgd->rd_rg.rg_free--;
	rgd->rd_rg.rg_dinodes++;
S
Steven Whitehouse 已提交
1523
	*generation = rgd->rd_rg.rg_igeneration++;
1524
	gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
1525
	gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data);
D
David Teigland 已提交
1526 1527 1528 1529

	al->al_alloced++;

	gfs2_statfs_change(sdp, 0, -1, +1);
1530
	gfs2_trans_add_unrevoke(sdp, block, 1);
D
David Teigland 已提交
1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546

	spin_lock(&sdp->sd_rindex_spin);
	rgd->rd_free_clone--;
	spin_unlock(&sdp->sd_rindex_spin);

	return block;
}

/**
 * gfs2_free_data - free a contiguous run of data block(s)
 * @ip: the inode these blocks are being freed from
 * @bstart: first block of a run of contiguous blocks
 * @blen: the length of the block run
 *
 */

1547
void gfs2_free_data(struct gfs2_inode *ip, u64 bstart, u32 blen)
D
David Teigland 已提交
1548
{
1549
	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
D
David Teigland 已提交
1550 1551 1552 1553 1554 1555 1556 1557
	struct gfs2_rgrpd *rgd;

	rgd = rgblk_free(sdp, bstart, blen, GFS2_BLKST_FREE);
	if (!rgd)
		return;

	rgd->rd_rg.rg_free += blen;

1558
	gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
1559
	gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data);
D
David Teigland 已提交
1560 1561 1562 1563

	gfs2_trans_add_rg(rgd);

	gfs2_statfs_change(sdp, 0, +blen, 0);
1564
	gfs2_quota_change(ip, -(s64)blen, ip->i_inode.i_uid, ip->i_inode.i_gid);
D
David Teigland 已提交
1565 1566 1567 1568 1569 1570 1571 1572 1573 1574
}

/**
 * gfs2_free_meta - free a contiguous run of data block(s)
 * @ip: the inode these blocks are being freed from
 * @bstart: first block of a run of contiguous blocks
 * @blen: the length of the block run
 *
 */

1575
void gfs2_free_meta(struct gfs2_inode *ip, u64 bstart, u32 blen)
D
David Teigland 已提交
1576
{
1577
	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
D
David Teigland 已提交
1578 1579 1580 1581 1582 1583 1584 1585
	struct gfs2_rgrpd *rgd;

	rgd = rgblk_free(sdp, bstart, blen, GFS2_BLKST_FREE);
	if (!rgd)
		return;

	rgd->rd_rg.rg_free += blen;

1586
	gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
1587
	gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data);
D
David Teigland 已提交
1588 1589 1590 1591

	gfs2_trans_add_rg(rgd);

	gfs2_statfs_change(sdp, 0, +blen, 0);
1592
	gfs2_quota_change(ip, -(s64)blen, ip->i_inode.i_uid, ip->i_inode.i_gid);
D
David Teigland 已提交
1593 1594 1595
	gfs2_meta_wipe(ip, bstart, blen);
}

1596 1597 1598 1599 1600
void gfs2_unlink_di(struct inode *inode)
{
	struct gfs2_inode *ip = GFS2_I(inode);
	struct gfs2_sbd *sdp = GFS2_SB(inode);
	struct gfs2_rgrpd *rgd;
1601
	u64 blkno = ip->i_no_addr;
1602 1603 1604 1605 1606

	rgd = rgblk_free(sdp, blkno, 1, GFS2_BLKST_UNLINKED);
	if (!rgd)
		return;
	gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
1607
	gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data);
1608 1609 1610
	gfs2_trans_add_rg(rgd);
}

1611
static void gfs2_free_uninit_di(struct gfs2_rgrpd *rgd, u64 blkno)
D
David Teigland 已提交
1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625
{
	struct gfs2_sbd *sdp = rgd->rd_sbd;
	struct gfs2_rgrpd *tmp_rgd;

	tmp_rgd = rgblk_free(sdp, blkno, 1, GFS2_BLKST_FREE);
	if (!tmp_rgd)
		return;
	gfs2_assert_withdraw(sdp, rgd == tmp_rgd);

	if (!rgd->rd_rg.rg_dinodes)
		gfs2_consist_rgrpd(rgd);
	rgd->rd_rg.rg_dinodes--;
	rgd->rd_rg.rg_free++;

1626
	gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
1627
	gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data);
D
David Teigland 已提交
1628 1629 1630 1631 1632 1633 1634 1635

	gfs2_statfs_change(sdp, 0, +1, -1);
	gfs2_trans_add_rg(rgd);
}


void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip)
{
1636
	gfs2_free_uninit_di(rgd, ip->i_no_addr);
1637
	gfs2_quota_change(ip, -1, ip->i_inode.i_uid, ip->i_inode.i_gid);
1638
	gfs2_meta_wipe(ip, ip->i_no_addr, 1);
D
David Teigland 已提交
1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653
}

/**
 * gfs2_rlist_add - add a RG to a list of RGs
 * @sdp: the filesystem
 * @rlist: the list of resource groups
 * @block: the block
 *
 * Figure out what RG a block belongs to and add that RG to the list
 *
 * FIXME: Don't use NOFAIL
 *
 */

void gfs2_rlist_add(struct gfs2_sbd *sdp, struct gfs2_rgrp_list *rlist,
1654
		    u64 block)
D
David Teigland 已提交
1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666
{
	struct gfs2_rgrpd *rgd;
	struct gfs2_rgrpd **tmp;
	unsigned int new_space;
	unsigned int x;

	if (gfs2_assert_warn(sdp, !rlist->rl_ghs))
		return;

	rgd = gfs2_blk2rgrpd(sdp, block);
	if (!rgd) {
		if (gfs2_consist(sdp))
1667
			fs_err(sdp, "block = %llu\n", (unsigned long long)block);
D
David Teigland 已提交
1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678
		return;
	}

	for (x = 0; x < rlist->rl_rgrps; x++)
		if (rlist->rl_rgd[x] == rgd)
			return;

	if (rlist->rl_rgrps == rlist->rl_space) {
		new_space = rlist->rl_space + 10;

		tmp = kcalloc(new_space, sizeof(struct gfs2_rgrpd *),
1679
			      GFP_NOFS | __GFP_NOFAIL);
D
David Teigland 已提交
1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704

		if (rlist->rl_rgd) {
			memcpy(tmp, rlist->rl_rgd,
			       rlist->rl_space * sizeof(struct gfs2_rgrpd *));
			kfree(rlist->rl_rgd);
		}

		rlist->rl_space = new_space;
		rlist->rl_rgd = tmp;
	}

	rlist->rl_rgd[rlist->rl_rgrps++] = rgd;
}

/**
 * gfs2_rlist_alloc - all RGs have been added to the rlist, now allocate
 *      and initialize an array of glock holders for them
 * @rlist: the list of resource groups
 * @state: the lock state to acquire the RG lock in
 * @flags: the modifier flags for the holder structures
 *
 * FIXME: Don't use NOFAIL
 *
 */

1705
void gfs2_rlist_alloc(struct gfs2_rgrp_list *rlist, unsigned int state)
D
David Teigland 已提交
1706 1707 1708 1709
{
	unsigned int x;

	rlist->rl_ghs = kcalloc(rlist->rl_rgrps, sizeof(struct gfs2_holder),
1710
				GFP_NOFS | __GFP_NOFAIL);
D
David Teigland 已提交
1711 1712
	for (x = 0; x < rlist->rl_rgrps; x++)
		gfs2_holder_init(rlist->rl_rgd[x]->rd_gl,
1713
				state, 0,
D
David Teigland 已提交
1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735
				&rlist->rl_ghs[x]);
}

/**
 * gfs2_rlist_free - free a resource group list
 * @list: the list of resource groups
 *
 */

void gfs2_rlist_free(struct gfs2_rgrp_list *rlist)
{
	unsigned int x;

	kfree(rlist->rl_rgd);

	if (rlist->rl_ghs) {
		for (x = 0; x < rlist->rl_rgrps; x++)
			gfs2_holder_uninit(&rlist->rl_ghs[x]);
		kfree(rlist->rl_ghs);
	}
}