lops.c 19.7 KB
Newer Older
D
David Teigland 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81
/*
 * Copyright (C) Sistina Software, Inc.  1997-2003 All rights reserved.
 * Copyright (C) 2004-2005 Red Hat, Inc.  All rights reserved.
 *
 * This copyrighted material is made available to anyone wishing to use,
 * modify, copy, or redistribute it subject to the terms and conditions
 * of the GNU General Public License v.2.
 */

#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/completion.h>
#include <linux/buffer_head.h>
#include <asm/semaphore.h>

#include "gfs2.h"
#include "glock.h"
#include "log.h"
#include "lops.h"
#include "meta_io.h"
#include "recovery.h"
#include "rgrp.h"
#include "trans.h"

static void glock_lo_add(struct gfs2_sbd *sdp, struct gfs2_log_element *le)
{
	struct gfs2_glock *gl;

	get_transaction->tr_touched = 1;

	if (!list_empty(&le->le_list))
		return;

	gl = container_of(le, struct gfs2_glock, gl_le);
	if (gfs2_assert_withdraw(sdp, gfs2_glock_is_held_excl(gl)))
		return;
	gfs2_glock_hold(gl);
	set_bit(GLF_DIRTY, &gl->gl_flags);

	gfs2_log_lock(sdp);
	sdp->sd_log_num_gl++;
	list_add(&le->le_list, &sdp->sd_log_le_gl);
	gfs2_log_unlock(sdp);
}

static void glock_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_ail *ai)
{
	struct list_head *head = &sdp->sd_log_le_gl;
	struct gfs2_glock *gl;

	while (!list_empty(head)) {
		gl = list_entry(head->next, struct gfs2_glock, gl_le.le_list);
		list_del_init(&gl->gl_le.le_list);
		sdp->sd_log_num_gl--;

		gfs2_assert_withdraw(sdp, gfs2_glock_is_held_excl(gl));
		gfs2_glock_put(gl);
	}
	gfs2_assert_warn(sdp, !sdp->sd_log_num_gl);
}

static void buf_lo_add(struct gfs2_sbd *sdp, struct gfs2_log_element *le)
{
	struct gfs2_bufdata *bd = container_of(le, struct gfs2_bufdata, bd_le);
	struct gfs2_trans *tr;

	if (!list_empty(&bd->bd_list_tr))
		return;

	tr = get_transaction;
	tr->tr_touched = 1;
	tr->tr_num_buf++;
	list_add(&bd->bd_list_tr, &tr->tr_list_buf);

	if (!list_empty(&le->le_list))
		return;

	gfs2_trans_add_gl(bd->bd_gl);

	gfs2_meta_check(sdp, bd->bd_bh);
82
	gfs2_pin(sdp, bd->bd_bh);
D
David Teigland 已提交
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171

	gfs2_log_lock(sdp);
	sdp->sd_log_num_buf++;
	list_add(&le->le_list, &sdp->sd_log_le_buf);
	gfs2_log_unlock(sdp);

	tr->tr_num_buf_new++;
}

static void buf_lo_incore_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
{
	struct list_head *head = &tr->tr_list_buf;
	struct gfs2_bufdata *bd;

	while (!list_empty(head)) {
		bd = list_entry(head->next, struct gfs2_bufdata, bd_list_tr);
		list_del_init(&bd->bd_list_tr);
		tr->tr_num_buf--;
	}
	gfs2_assert_warn(sdp, !tr->tr_num_buf);
}

static void buf_lo_before_commit(struct gfs2_sbd *sdp)
{
	struct buffer_head *bh;
	struct gfs2_log_descriptor *ld;
	struct gfs2_bufdata *bd1 = NULL, *bd2;
	unsigned int total = sdp->sd_log_num_buf;
	unsigned int offset = sizeof(struct gfs2_log_descriptor);
	unsigned int limit;
	unsigned int num;
	unsigned n;
	__be64 *ptr;

	offset += (sizeof(__be64) - 1);
	offset &= ~(sizeof(__be64) - 1);
	limit = (sdp->sd_sb.sb_bsize - offset)/sizeof(__be64);
	/* for 4k blocks, limit = 503 */

	bd1 = bd2 = list_prepare_entry(bd1, &sdp->sd_log_le_buf, bd_le.le_list);
	while(total) {
		num = total;
		if (total > limit)
			num = limit;
		bh = gfs2_log_get_buf(sdp);
		ld = (struct gfs2_log_descriptor *)bh->b_data;
		ptr = (__be64 *)(bh->b_data + offset);
		ld->ld_header.mh_magic = cpu_to_be32(GFS2_MAGIC);
		ld->ld_header.mh_type = cpu_to_be16(GFS2_METATYPE_LD);
		ld->ld_header.mh_format = cpu_to_be16(GFS2_FORMAT_LD);
		ld->ld_type = cpu_to_be32(GFS2_LOG_DESC_METADATA);
		ld->ld_length = cpu_to_be32(num + 1);
		ld->ld_data1 = cpu_to_be32(num);
		ld->ld_data2 = cpu_to_be32(0);
		memset(ld->ld_reserved, 0, sizeof(ld->ld_reserved));

		n = 0;
		list_for_each_entry_continue(bd1, &sdp->sd_log_le_buf, bd_le.le_list) {
			*ptr++ = cpu_to_be64(bd1->bd_bh->b_blocknr);
			if (++n >= num)
				break;
		}

		set_buffer_dirty(bh);
		ll_rw_block(WRITE, 1, &bh);

		n = 0;
		list_for_each_entry_continue(bd2, &sdp->sd_log_le_buf, bd_le.le_list) {
			bh = gfs2_log_fake_buf(sdp, bd2->bd_bh);
			set_buffer_dirty(bh);
			ll_rw_block(WRITE, 1, &bh);
			if (++n >= num)
				break;
		}

		total -= num;
	}
}

static void buf_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_ail *ai)
{
	struct list_head *head = &sdp->sd_log_le_buf;
	struct gfs2_bufdata *bd;

	while (!list_empty(head)) {
		bd = list_entry(head->next, struct gfs2_bufdata, bd_le.le_list);
		list_del_init(&bd->bd_le.le_list);
		sdp->sd_log_num_buf--;

172
		gfs2_unpin(sdp, bd->bd_bh, ai);
D
David Teigland 已提交
173 174 175 176 177 178 179
	}
	gfs2_assert_warn(sdp, !sdp->sd_log_num_buf);
}

static void buf_lo_before_scan(struct gfs2_jdesc *jd,
			       struct gfs2_log_header *head, int pass)
{
180
	struct gfs2_sbd *sdp = get_v2ip(jd->jd_inode)->i_sbd;
D
David Teigland 已提交
181 182 183 184 185 186 187 188 189 190 191 192

	if (pass != 0)
		return;

	sdp->sd_found_blocks = 0;
	sdp->sd_replayed_blocks = 0;
}

static int buf_lo_scan_elements(struct gfs2_jdesc *jd, unsigned int start,
				struct gfs2_log_descriptor *ld, __be64 *ptr,
				int pass)
{
193 194
	struct gfs2_sbd *sdp = get_v2ip(jd->jd_inode)->i_sbd;
	struct gfs2_glock *gl = get_v2ip(jd->jd_inode)->i_gl;
D
David Teigland 已提交
195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238
	unsigned int blks = be32_to_cpu(ld->ld_data1);
	struct buffer_head *bh_log, *bh_ip;
	uint64_t blkno;
	int error = 0;

	if (pass != 1 || be32_to_cpu(ld->ld_type) != GFS2_LOG_DESC_METADATA)
		return 0;

	gfs2_replay_incr_blk(sdp, &start);

	for (; blks; gfs2_replay_incr_blk(sdp, &start), blks--) {
		blkno = be64_to_cpu(*ptr++);

		sdp->sd_found_blocks++;

		if (gfs2_revoke_check(sdp, blkno, start))
			continue;

		error = gfs2_replay_read_block(jd, start, &bh_log);
                if (error)
                        return error;

		bh_ip = gfs2_meta_new(gl, blkno);
		memcpy(bh_ip->b_data, bh_log->b_data, bh_log->b_size);

		if (gfs2_meta_check(sdp, bh_ip))
			error = -EIO;
		else
			mark_buffer_dirty(bh_ip);

		brelse(bh_log);
		brelse(bh_ip);

		if (error)
			break;

		sdp->sd_replayed_blocks++;
	}

	return error;
}

static void buf_lo_after_scan(struct gfs2_jdesc *jd, int error, int pass)
{
239
	struct gfs2_sbd *sdp = get_v2ip(jd->jd_inode)->i_sbd;
D
David Teigland 已提交
240 241

	if (error) {
242
		gfs2_meta_sync(get_v2ip(jd->jd_inode)->i_gl, DIO_START | DIO_WAIT);
D
David Teigland 已提交
243 244 245 246 247
		return;
	}
	if (pass != 1)
		return;

248
	gfs2_meta_sync(get_v2ip(jd->jd_inode)->i_gl, DIO_START | DIO_WAIT);
D
David Teigland 已提交
249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293

	fs_info(sdp, "jid=%u: Replayed %u of %u blocks\n",
	        jd->jd_jid, sdp->sd_replayed_blocks, sdp->sd_found_blocks);
}

static void revoke_lo_add(struct gfs2_sbd *sdp, struct gfs2_log_element *le)
{
	struct gfs2_trans *tr;

	tr = get_transaction;
	tr->tr_touched = 1;
	tr->tr_num_revoke++;

	gfs2_log_lock(sdp);
	sdp->sd_log_num_revoke++;
	list_add(&le->le_list, &sdp->sd_log_le_revoke);
	gfs2_log_unlock(sdp);
}

static void revoke_lo_before_commit(struct gfs2_sbd *sdp)
{
	struct gfs2_log_descriptor *ld;
	struct gfs2_meta_header *mh;
	struct buffer_head *bh;
	unsigned int offset;
	struct list_head *head = &sdp->sd_log_le_revoke;
	struct gfs2_revoke *rv;

	if (!sdp->sd_log_num_revoke)
		return;

	bh = gfs2_log_get_buf(sdp);
	ld = (struct gfs2_log_descriptor *)bh->b_data;
	ld->ld_header.mh_magic = cpu_to_be32(GFS2_MAGIC);
	ld->ld_header.mh_type = cpu_to_be16(GFS2_METATYPE_LD);
	ld->ld_header.mh_format = cpu_to_be16(GFS2_FORMAT_LD);
	ld->ld_type = cpu_to_be32(GFS2_LOG_DESC_REVOKE);
	ld->ld_length = cpu_to_be32(gfs2_struct2blk(sdp, sdp->sd_log_num_revoke, sizeof(uint64_t)));
	ld->ld_data1 = cpu_to_be32(sdp->sd_log_num_revoke);
	ld->ld_data2 = cpu_to_be32(0);
	memset(ld->ld_reserved, 0, sizeof(ld->ld_reserved));
	offset = sizeof(struct gfs2_log_descriptor);

	while (!list_empty(head)) {
		rv = list_entry(head->next, struct gfs2_revoke, rv_le.le_list);
294
		list_del_init(&rv->rv_le.le_list);
D
David Teigland 已提交
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
		sdp->sd_log_num_revoke--;

		if (offset + sizeof(uint64_t) > sdp->sd_sb.sb_bsize) {
			set_buffer_dirty(bh);
			ll_rw_block(WRITE, 1, &bh);

			bh = gfs2_log_get_buf(sdp);
			mh = (struct gfs2_meta_header *)bh->b_data;
			mh->mh_magic = cpu_to_be32(GFS2_MAGIC);
			mh->mh_type = cpu_to_be16(GFS2_METATYPE_LB);
			mh->mh_format = cpu_to_be16(GFS2_FORMAT_LB);
			offset = sizeof(struct gfs2_meta_header);
		}

		*(__be64 *)(bh->b_data + offset) = cpu_to_be64(rv->rv_blkno);
		kfree(rv);

		offset += sizeof(uint64_t);
	}
	gfs2_assert_withdraw(sdp, !sdp->sd_log_num_revoke);

	set_buffer_dirty(bh);
	ll_rw_block(WRITE, 1, &bh);
}

static void revoke_lo_before_scan(struct gfs2_jdesc *jd,
				  struct gfs2_log_header *head, int pass)
{
323
	struct gfs2_sbd *sdp = get_v2ip(jd->jd_inode)->i_sbd;
D
David Teigland 已提交
324 325 326 327 328 329 330 331 332 333 334 335

	if (pass != 0)
		return;

	sdp->sd_found_revokes = 0;
	sdp->sd_replay_tail = head->lh_tail;
}

static int revoke_lo_scan_elements(struct gfs2_jdesc *jd, unsigned int start,
				   struct gfs2_log_descriptor *ld, __be64 *ptr,
				   int pass)
{
336
	struct gfs2_sbd *sdp = get_v2ip(jd->jd_inode)->i_sbd;
D
David Teigland 已提交
337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381
	unsigned int blks = be32_to_cpu(ld->ld_length);
	unsigned int revokes = be32_to_cpu(ld->ld_data1);
	struct buffer_head *bh;
	unsigned int offset;
	uint64_t blkno;
	int first = 1;
	int error;

	if (pass != 0 || be32_to_cpu(ld->ld_type) != GFS2_LOG_DESC_REVOKE)
		return 0;

	offset = sizeof(struct gfs2_log_descriptor);

	for (; blks; gfs2_replay_incr_blk(sdp, &start), blks--) {
		error = gfs2_replay_read_block(jd, start, &bh);
		if (error)
			return error;

		if (!first)
			gfs2_metatype_check(sdp, bh, GFS2_METATYPE_LB);

		while (offset + sizeof(uint64_t) <= sdp->sd_sb.sb_bsize) {
			blkno = be64_to_cpu(*(__be64 *)(bh->b_data + offset));

			error = gfs2_revoke_add(sdp, blkno, start);
			if (error < 0)
				return error;
			else if (error)
				sdp->sd_found_revokes++;

			if (!--revokes)
				break;
			offset += sizeof(uint64_t);
		}

		brelse(bh);
		offset = sizeof(struct gfs2_meta_header);
		first = 0;
	}

	return 0;
}

static void revoke_lo_after_scan(struct gfs2_jdesc *jd, int error, int pass)
{
382
	struct gfs2_sbd *sdp = get_v2ip(jd->jd_inode)->i_sbd;
D
David Teigland 已提交
383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430

	if (error) {
		gfs2_revoke_clean(sdp);
		return;
	}
	if (pass != 1)
		return;

	fs_info(sdp, "jid=%u: Found %u revoke tags\n",
	        jd->jd_jid, sdp->sd_found_revokes);

	gfs2_revoke_clean(sdp);
}

static void rg_lo_add(struct gfs2_sbd *sdp, struct gfs2_log_element *le)
{
	struct gfs2_rgrpd *rgd;

	get_transaction->tr_touched = 1;

	if (!list_empty(&le->le_list))
		return;

	rgd = container_of(le, struct gfs2_rgrpd, rd_le);
	gfs2_rgrp_bh_hold(rgd);

	gfs2_log_lock(sdp);
	sdp->sd_log_num_rg++;
	list_add(&le->le_list, &sdp->sd_log_le_rg);
	gfs2_log_unlock(sdp);	
}

static void rg_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_ail *ai)
{
	struct list_head *head = &sdp->sd_log_le_rg;
	struct gfs2_rgrpd *rgd;

	while (!list_empty(head)) {
		rgd = list_entry(head->next, struct gfs2_rgrpd, rd_le.le_list);
		list_del_init(&rgd->rd_le.le_list);
		sdp->sd_log_num_rg--;

		gfs2_rgrp_repolish_clones(rgd);
		gfs2_rgrp_bh_put(rgd);
	}
	gfs2_assert_warn(sdp, !sdp->sd_log_num_rg);
}

431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446
/**
 * databuf_lo_add - Add a databuf to the transaction.
 *
 * This is used in two distinct cases:
 * i) In ordered write mode
 *    We put the data buffer on a list so that we can ensure that its
 *    synced to disk at the right time
 * ii) In journaled data mode
 *    We need to journal the data block in the same way as metadata in
 *    the functions above. The difference is that here we have a tag
 *    which is two __be64's being the block number (as per meta data)
 *    and a flag which says whether the data block needs escaping or
 *    not. This means we need a new log entry for each 251 or so data
 *    blocks, which isn't an enormous overhead but twice as much as
 *    for normal metadata blocks.
 */
D
David Teigland 已提交
447 448
static void databuf_lo_add(struct gfs2_sbd *sdp, struct gfs2_log_element *le)
{
449 450 451 452
	struct gfs2_bufdata *bd = container_of(le, struct gfs2_bufdata, bd_le);
	struct gfs2_trans *tr = get_transaction;
	struct address_space *mapping = bd->bd_bh->b_page->mapping;
	struct gfs2_inode *ip = get_v2ip(mapping->host);
D
David Teigland 已提交
453

454 455 456 457 458 459 460 461
	tr->tr_touched = 1;
	if (!list_empty(&bd->bd_list_tr) &&
	    (ip->i_di.di_flags & GFS2_DIF_JDATA)) {
		tr->tr_num_buf++;
		gfs2_trans_add_gl(bd->bd_gl);
		list_add(&bd->bd_list_tr, &tr->tr_list_buf);
		gfs2_pin(sdp, bd->bd_bh);
	}
D
David Teigland 已提交
462
	gfs2_log_lock(sdp);
463 464 465 466 467 468
	if (!list_empty(&le->le_list)) {
		if (ip->i_di.di_flags & GFS2_DIF_JDATA)
			sdp->sd_log_num_jdata++;
		sdp->sd_log_num_databuf++;
		list_add(&le->le_list, &sdp->sd_log_le_databuf);
	}
D
David Teigland 已提交
469 470 471
	gfs2_log_unlock(sdp);
}

472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495
static int gfs2_check_magic(struct buffer_head *bh)
{
	struct page *page = bh->b_page;
	void *kaddr;
	__be32 *ptr;
	int rv = 0;

	kaddr = kmap_atomic(page, KM_USER0);
	ptr = kaddr + bh_offset(bh);
	if (*ptr == cpu_to_be32(GFS2_MAGIC))
		rv = 1;
	kunmap_atomic(page, KM_USER0);

	return rv;
}

/**
 * databuf_lo_before_commit - Scan the data buffers, writing as we go
 *
 * Here we scan through the lists of buffers and make the assumption
 * that any buffer thats been pinned is being journaled, and that
 * any unpinned buffer is an ordered write data buffer and therefore
 * will be written back rather than journaled.
 */
D
David Teigland 已提交
496 497 498
static void databuf_lo_before_commit(struct gfs2_sbd *sdp)
{
	LIST_HEAD(started);
499 500 501 502 503 504 505 506
	struct gfs2_bufdata *bd1 = NULL, *bd2, *bdt;
	struct buffer_head *bh = NULL;
	unsigned int offset = sizeof(struct gfs2_log_descriptor);
	struct gfs2_log_descriptor *ld;
	unsigned int limit;
	unsigned int total_dbuf = sdp->sd_log_num_databuf;
	unsigned int total_jdata = sdp->sd_log_num_jdata;
	unsigned int num, n;
507
	__be64 *ptr = NULL;
D
David Teigland 已提交
508

509 510 511
	offset += (2*sizeof(__be64) - 1);
	offset &= ~(2*sizeof(__be64) - 1);
	limit = (sdp->sd_sb.sb_bsize - offset)/sizeof(__be64);
D
David Teigland 已提交
512

513 514 515 516
	/*
	 * Start writing ordered buffers, write journaled buffers
	 * into the log along with a header
	 */
517
	gfs2_log_lock(sdp);
518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535
	bd2 = bd1 = list_prepare_entry(bd1, &sdp->sd_log_le_databuf, bd_le.le_list);
	while(total_dbuf) {
		num = total_jdata;
		if (num > limit)
			num = limit;
		n = 0;
		list_for_each_entry_safe_continue(bd1, bdt, &sdp->sd_log_le_databuf, bd_le.le_list) {
			/* An ordered write buffer */
			if (bd1->bd_bh && !buffer_pinned(bd1->bd_bh)) {
				list_move(&bd1->bd_le.le_list, &started);
				if (bd1 == bd2) {
					bd2 = NULL;
					bd2 = list_prepare_entry(bd2, &sdp->sd_log_le_databuf, bd_le.le_list);
				}
				total_dbuf--;
				if (bd1->bd_bh) {
					get_bh(bd1->bd_bh);
					if (buffer_dirty(bd1->bd_bh)) {
536
						gfs2_log_unlock(sdp);
537 538
						wait_on_buffer(bd1->bd_bh);
						ll_rw_block(WRITE, 1, &bd1->bd_bh);
539
						gfs2_log_lock(sdp);
540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566
					}
					brelse(bd1->bd_bh);
					continue;
				}
				continue;
			} else if (bd1->bd_bh) { /* A journaled buffer */
				int magic;
				gfs2_log_unlock(sdp);
				if (!bh) {
					bh = gfs2_log_get_buf(sdp);
					ld = (struct gfs2_log_descriptor *)bh->b_data;
					ptr = (__be64 *)(bh->b_data + offset);
					ld->ld_header.mh_magic = cpu_to_be32(GFS2_MAGIC);
					ld->ld_header.mh_type = cpu_to_be16(GFS2_METATYPE_LD);
					ld->ld_header.mh_format = cpu_to_be16(GFS2_FORMAT_LD);
					ld->ld_type = cpu_to_be32(GFS2_LOG_DESC_JDATA);
					ld->ld_length = cpu_to_be32(num + 1);
					ld->ld_data1 = cpu_to_be32(num);
					ld->ld_data2 = cpu_to_be32(0);
					memset(ld->ld_reserved, 0, sizeof(ld->ld_reserved));
				}
				magic = gfs2_check_magic(bd1->bd_bh);
				*ptr++ = cpu_to_be64(bd1->bd_bh->b_blocknr);
				*ptr++ = cpu_to_be64((__u64)magic);
				clear_buffer_escaped(bd1->bd_bh);
				if (unlikely(magic != 0))
					set_buffer_escaped(bd1->bd_bh);
567
				gfs2_log_lock(sdp);
568 569 570 571
				if (n++ > num)
					break;
			}
		}
572
		gfs2_log_unlock(sdp);
D
David Teigland 已提交
573
		if (bh) {
574 575 576 577 578
			set_buffer_dirty(bh);
			ll_rw_block(WRITE, 1, &bh);
			bh = NULL;
		}
		n = 0;
579
		gfs2_log_lock(sdp);
580 581 582 583
		list_for_each_entry_continue(bd2, &sdp->sd_log_le_databuf, bd_le.le_list) {
			if (!bd2->bd_bh)
				continue;
			/* copy buffer if it needs escaping */
584
			gfs2_log_unlock(sdp);
585 586 587 588 589 590 591 592 593 594
			if (unlikely(buffer_escaped(bd2->bd_bh))) {
				void *kaddr;
				struct page *page = bd2->bd_bh->b_page;
				bh = gfs2_log_get_buf(sdp);
				kaddr = kmap_atomic(page, KM_USER0);
				memcpy(bh->b_data, kaddr + bh_offset(bd2->bd_bh), sdp->sd_sb.sb_bsize);
				kunmap_atomic(page, KM_USER0);
				*(__be32 *)bh->b_data = 0;
			} else {
				bh = gfs2_log_fake_buf(sdp, bd2->bd_bh);
D
David Teigland 已提交
595
			}
596 597
			set_buffer_dirty(bh);
			ll_rw_block(WRITE, 1, &bh);
598
			gfs2_log_lock(sdp);
599 600 601 602 603 604
			if (++n >= num)
				break;
		}
		bh = NULL;
		total_dbuf -= num;
		total_jdata -= num;
D
David Teigland 已提交
605
	}
606 607
	gfs2_log_unlock(sdp);

608
	/* Wait on all ordered buffers */
D
David Teigland 已提交
609
	while (!list_empty(&started)) {
610
		gfs2_log_lock(sdp);
611 612
		bd1 = list_entry(started.next, struct gfs2_bufdata, bd_le.le_list);
		list_del(&bd1->bd_le.le_list);
D
David Teigland 已提交
613 614
		sdp->sd_log_num_databuf--;

615
		bh = bd1->bd_bh;
D
David Teigland 已提交
616
		if (bh) {
617
			set_v2bd(bh, NULL);
D
David Teigland 已提交
618 619 620 621 622 623
			gfs2_log_unlock(sdp);
			wait_on_buffer(bh);
			brelse(bh);
		} else
			gfs2_log_unlock(sdp);

624
		kfree(bd1);
D
David Teigland 已提交
625 626
	}

627 628 629 630 631 632 633 634
	/* We've removed all the ordered write bufs here, so only jdata left */
	gfs2_assert_warn(sdp, sdp->sd_log_num_databuf == sdp->sd_log_num_jdata);
}

static int databuf_lo_scan_elements(struct gfs2_jdesc *jd, unsigned int start,
				    struct gfs2_log_descriptor *ld,
				    __be64 *ptr, int pass)
{
635 636
	struct gfs2_sbd *sdp = get_v2ip(jd->jd_inode)->i_sbd;
	struct gfs2_glock *gl = get_v2ip(jd->jd_inode)->i_gl;
637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684
	unsigned int blks = be32_to_cpu(ld->ld_data1);
	struct buffer_head *bh_log, *bh_ip;
	uint64_t blkno;
	uint64_t esc;
	int error = 0;

	if (pass != 1 || be32_to_cpu(ld->ld_type) != GFS2_LOG_DESC_JDATA)
		return 0;

	gfs2_replay_incr_blk(sdp, &start);
	for (; blks; gfs2_replay_incr_blk(sdp, &start), blks--) {
		blkno = be64_to_cpu(*ptr++);
		esc = be64_to_cpu(*ptr++);

		sdp->sd_found_blocks++;

		if (gfs2_revoke_check(sdp, blkno, start))
			continue;

		error = gfs2_replay_read_block(jd, start, &bh_log);
		if (error)
			return error;

		bh_ip = gfs2_meta_new(gl, blkno);
		memcpy(bh_ip->b_data, bh_log->b_data, bh_log->b_size);

		/* Unescape */
		if (esc) {
			__be32 *eptr = (__be32 *)bh_ip->b_data;
			*eptr = cpu_to_be32(GFS2_MAGIC);
		}
		mark_buffer_dirty(bh_ip);

		brelse(bh_log);
		brelse(bh_ip);
		if (error)
			break;

		sdp->sd_replayed_blocks++;
	}

	return error;
}

/* FIXME: sort out accounting for log blocks etc. */

static void databuf_lo_after_scan(struct gfs2_jdesc *jd, int error, int pass)
{
685
	struct gfs2_sbd *sdp = get_v2ip(jd->jd_inode)->i_sbd;
686 687

	if (error) {
688
		gfs2_meta_sync(get_v2ip(jd->jd_inode)->i_gl, DIO_START | DIO_WAIT);
689 690 691 692 693 694
		return;
	}
	if (pass != 1)
		return;

	/* data sync? */
695
	gfs2_meta_sync(get_v2ip(jd->jd_inode)->i_gl, DIO_START | DIO_WAIT);
696 697 698 699 700 701 702 703 704 705 706 707

	fs_info(sdp, "jid=%u: Replayed %u of %u data blocks\n",
		jd->jd_jid, sdp->sd_replayed_blocks, sdp->sd_found_blocks);
}

static void databuf_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_ail *ai)
{
	struct list_head *head = &sdp->sd_log_le_databuf;
	struct gfs2_bufdata *bd;

	while (!list_empty(head)) {
		bd = list_entry(head->next, struct gfs2_bufdata, bd_le.le_list);
708
		list_del(&bd->bd_le.le_list);
709 710 711 712
		sdp->sd_log_num_databuf--;
		sdp->sd_log_num_jdata--;
		gfs2_unpin(sdp, bd->bd_bh, ai);
	}
D
David Teigland 已提交
713
	gfs2_assert_warn(sdp, !sdp->sd_log_num_databuf);
714
	gfs2_assert_warn(sdp, !sdp->sd_log_num_jdata);
D
David Teigland 已提交
715 716
}

717

D
David Teigland 已提交
718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751
struct gfs2_log_operations gfs2_glock_lops = {
	.lo_add = glock_lo_add,
	.lo_after_commit = glock_lo_after_commit,
	.lo_name = "glock"
};

struct gfs2_log_operations gfs2_buf_lops = {
	.lo_add = buf_lo_add,
	.lo_incore_commit = buf_lo_incore_commit,
	.lo_before_commit = buf_lo_before_commit,
	.lo_after_commit = buf_lo_after_commit,
	.lo_before_scan = buf_lo_before_scan,
	.lo_scan_elements = buf_lo_scan_elements,
	.lo_after_scan = buf_lo_after_scan,
	.lo_name = "buf"
};

struct gfs2_log_operations gfs2_revoke_lops = {
	.lo_add = revoke_lo_add,
	.lo_before_commit = revoke_lo_before_commit,
	.lo_before_scan = revoke_lo_before_scan,
	.lo_scan_elements = revoke_lo_scan_elements,
	.lo_after_scan = revoke_lo_after_scan,
	.lo_name = "revoke"
};

struct gfs2_log_operations gfs2_rg_lops = {
	.lo_add = rg_lo_add,
	.lo_after_commit = rg_lo_after_commit,
	.lo_name = "rg"
};

struct gfs2_log_operations gfs2_databuf_lops = {
	.lo_add = databuf_lo_add,
752
	.lo_incore_commit = buf_lo_incore_commit,
D
David Teigland 已提交
753
	.lo_before_commit = databuf_lo_before_commit,
754 755 756
	.lo_after_commit = databuf_lo_after_commit,
	.lo_scan_elements = databuf_lo_scan_elements,
	.lo_after_scan = databuf_lo_after_scan,
D
David Teigland 已提交
757 758 759 760 761 762 763 764 765 766 767 768
	.lo_name = "databuf"
};

struct gfs2_log_operations *gfs2_log_ops[] = {
	&gfs2_glock_lops,
	&gfs2_buf_lops,
	&gfs2_revoke_lops,
	&gfs2_rg_lops,
	&gfs2_databuf_lops,
	NULL
};