xfs_log_cil.c 26.1 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
/*
 * Copyright (c) 2010 Red Hat, Inc. All Rights Reserved.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License as
 * published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it would be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write the Free Software Foundation,
 * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
 */

#include "xfs.h"
#include "xfs_fs.h"
20
#include "xfs_log_format.h"
21
#include "xfs_shared.h"
22
#include "xfs_trans_resv.h"
23 24 25 26 27
#include "xfs_sb.h"
#include "xfs_ag.h"
#include "xfs_mount.h"
#include "xfs_error.h"
#include "xfs_alloc.h"
28
#include "xfs_extent_busy.h"
29
#include "xfs_discard.h"
30 31 32 33
#include "xfs_trans.h"
#include "xfs_trans_priv.h"
#include "xfs_log.h"
#include "xfs_log_priv.h"
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

/*
 * Allocate a new ticket. Failing to get a new ticket makes it really hard to
 * recover, so we don't allow failure here. Also, we allocate in a context that
 * we don't want to be issuing transactions from, so we need to tell the
 * allocation code this as well.
 *
 * We don't reserve any space for the ticket - we are going to steal whatever
 * space we require from transactions as they commit. To ensure we reserve all
 * the space required, we need to set the current reservation of the ticket to
 * zero so that we know to steal the initial transaction overhead from the
 * first transaction commit.
 */
static struct xlog_ticket *
xlog_cil_ticket_alloc(
49
	struct xlog	*log)
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
{
	struct xlog_ticket *tic;

	tic = xlog_ticket_alloc(log, 0, 1, XFS_TRANSACTION, 0,
				KM_SLEEP|KM_NOFS);
	tic->t_trans_type = XFS_TRANS_CHECKPOINT;

	/*
	 * set the current reservation to zero so we know to steal the basic
	 * transaction overhead reservation from the first transaction commit.
	 */
	tic->t_curr_res = 0;
	return tic;
}

/*
 * After the first stage of log recovery is done, we know where the head and
 * tail of the log are. We need this log initialisation done before we can
 * initialise the first CIL checkpoint context.
 *
 * Here we allocate a log ticket to track space usage during a CIL push.  This
 * ticket is passed to xlog_write() directly so that we don't slowly leak log
 * space by failing to account for space used by log headers and additional
 * region headers for split regions.
 */
void
xlog_cil_init_post_recovery(
77
	struct xlog	*log)
78 79 80 81 82 83 84
{
	log->l_cilp->xc_ctx->ticket = xlog_cil_ticket_alloc(log);
	log->l_cilp->xc_ctx->sequence = 1;
	log->l_cilp->xc_ctx->commit_lsn = xlog_assign_lsn(log->l_curr_cycle,
								log->l_curr_block);
}

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
/*
 * Prepare the log item for insertion into the CIL. Calculate the difference in
 * log space and vectors it will consume, and if it is a new item pin it as
 * well.
 */
STATIC void
xfs_cil_prepare_item(
	struct xlog		*log,
	struct xfs_log_vec	*lv,
	struct xfs_log_vec	*old_lv,
	int			*diff_len,
	int			*diff_iovecs)
{
	/* Account for the new LV being passed in */
	if (lv->lv_buf_len != XFS_LOG_VEC_ORDERED) {
		*diff_len += lv->lv_buf_len;
		*diff_iovecs += lv->lv_niovecs;
	}

	/*
	 * If there is no old LV, this is the first time we've seen the item in
	 * this CIL context and so we need to pin it. If we are replacing the
	 * old_lv, then remove the space it accounts for and free it.
	 */
	if (!old_lv)
		lv->lv_item->li_ops->iop_pin(lv->lv_item);
	else if (old_lv != lv) {
		ASSERT(lv->lv_buf_len != XFS_LOG_VEC_ORDERED);

		*diff_len -= old_lv->lv_buf_len;
		*diff_iovecs -= old_lv->lv_niovecs;
		kmem_free(old_lv);
	}

	/* attach new log vector to log item */
	lv->lv_item->li_lv = lv;

	/*
	 * If this is the first time the item is being committed to the
	 * CIL, store the sequence number on the log item so we can
	 * tell in future commits whether this is the first checkpoint
	 * the item is being committed into.
	 */
	if (!lv->lv_item->li_seq)
		lv->lv_item->li_seq = log->l_cilp->xc_ctx->sequence;
}

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
/*
 * Format log item into a flat buffers
 *
 * For delayed logging, we need to hold a formatted buffer containing all the
 * changes on the log item. This enables us to relog the item in memory and
 * write it out asynchronously without needing to relock the object that was
 * modified at the time it gets written into the iclog.
 *
 * This function builds a vector for the changes in each log item in the
 * transaction. It then works out the length of the buffer needed for each log
 * item, allocates them and formats the vector for the item into the buffer.
 * The buffer is then attached to the log item are then inserted into the
 * Committed Item List for tracking until the next checkpoint is written out.
 *
 * We don't set up region headers during this process; we simply copy the
 * regions into the flat buffer. We can do this because we still have to do a
 * formatting step to write the regions into the iclog buffer.  Writing the
 * ophdrs during the iclog write means that we can support splitting large
 * regions across iclog boundares without needing a change in the format of the
 * item/region encapsulation.
 *
 * Hence what we need to do now is change the rewrite the vector array to point
 * to the copied region inside the buffer we just allocated. This allows us to
 * format the regions into the iclog as though they are being formatted
 * directly out of the objects themselves.
 */
158 159 160 161 162 163
static void
xlog_cil_insert_format_items(
	struct xlog		*log,
	struct xfs_trans	*tp,
	int			*diff_len,
	int			*diff_iovecs)
164
{
165
	struct xfs_log_item_desc *lidp;
166

167 168 169 170

	/* Bail out if we didn't find a log item.  */
	if (list_empty(&tp->t_items)) {
		ASSERT(0);
171
		return;
172 173 174
	}

	list_for_each_entry(lidp, &tp->t_items, lid_trans) {
175
		struct xfs_log_item *lip = lidp->lid_item;
176
		struct xfs_log_vec *lv;
177
		struct xfs_log_vec *old_lv;
178 179 180
		int	niovecs = 0;
		int	nbytes = 0;
		int	buf_size;
181
		bool	ordered = false;
182

183 184 185 186
		/* Skip items which aren't dirty in this transaction. */
		if (!(lidp->lid_flags & XFS_LID_DIRTY))
			continue;

187 188 189
		/* get number of vecs and size of data to be stored */
		lip->li_ops->iop_size(lip, &niovecs, &nbytes);

190
		/* Skip items that do not have any vectors for writing */
191
		if (!niovecs)
192 193
			continue;

194 195 196 197 198 199 200 201
		/*
		 * Ordered items need to be tracked but we do not wish to write
		 * them. We need a logvec to track the object, but we do not
		 * need an iovec or buffer to be allocated for copying data.
		 */
		if (niovecs == XFS_LOG_VEC_ORDERED) {
			ordered = true;
			niovecs = 0;
202
			nbytes = 0;
203 204
		}

205 206 207 208 209 210 211
		/*
		 * We 64-bit align the length of each iovec so that the start
		 * of the next one is naturally aligned.  We'll need to
		 * account for that slack space here.
		 */
		nbytes += niovecs * sizeof(uint64_t);

212 213 214
		/* grab the old item if it exists for reservation accounting */
		old_lv = lip->li_lv;

215 216 217
		/* calc buffer size */
		buf_size = sizeof(struct xfs_log_vec) + nbytes +
				niovecs * sizeof(struct xfs_log_iovec);
218

219 220 221 222 223 224 225 226 227
		/* compare to existing item size */
		if (lip->li_lv && buf_size <= lip->li_lv->lv_size) {
			/* same or smaller, optimise common overwrite case */
			lv = lip->li_lv;
			lv->lv_next = NULL;

			if (ordered)
				goto insert;

228 229 230 231 232 233
			/*
			 * set the item up as though it is a new insertion so
			 * that the space reservation accounting is correct.
			 */
			*diff_iovecs -= lv->lv_niovecs;
			*diff_len -= lv->lv_buf_len;
234 235 236 237 238 239 240 241 242 243 244 245
		} else {
			/* allocate new data chunk */
			lv = kmem_zalloc(buf_size, KM_SLEEP|KM_NOFS);
			lv->lv_item = lip;
			lv->lv_size = buf_size;
			if (ordered) {
				/* track as an ordered logvec */
				ASSERT(lip->li_lv == NULL);
				lv->lv_buf_len = XFS_LOG_VEC_ORDERED;
				goto insert;
			}
			lv->lv_iovecp = (struct xfs_log_iovec *)&lv[1];
246 247
		}

248
		/* Ensure the lv is set up according to ->iop_size */
249
		lv->lv_niovecs = niovecs;
250

251
		/* The allocated data region lies beyond the iovec region */
252
		lv->lv_buf_len = 0;
253
		lv->lv_buf = (char *)lv + buf_size - nbytes;
254
		lip->li_ops->iop_format(lip, lv);
255
insert:
256
		ASSERT(lv->lv_buf_len <= nbytes);
257
		xfs_cil_prepare_item(log, lv, old_lv, diff_len, diff_iovecs);
258
	}
259 260 261 262 263 264
}

/*
 * Insert the log items into the CIL and calculate the difference in space
 * consumed by the item. Add the space to the checkpoint ticket and calculate
 * if the change requires additional log metadata. If it does, take that space
265
 * as well. Remove the amount of space we added to the checkpoint ticket from
266 267
 * the current transaction ticket so that the accounting works out correctly.
 */
268 269
static void
xlog_cil_insert_items(
270
	struct xlog		*log,
271
	struct xfs_trans	*tp)
272
{
273 274
	struct xfs_cil		*cil = log->l_cilp;
	struct xfs_cil_ctx	*ctx = cil->xc_ctx;
275
	struct xfs_log_item_desc *lidp;
276 277 278
	int			len = 0;
	int			diff_iovecs = 0;
	int			iclog_space;
279

280
	ASSERT(tp);
281 282 283 284 285

	/*
	 * We can do this safely because the context can't checkpoint until we
	 * are done so it doesn't matter exactly how we update the CIL.
	 */
286 287 288 289 290 291 292
	xlog_cil_insert_format_items(log, tp, &len, &diff_iovecs);

	/*
	 * Now (re-)position everything modified at the tail of the CIL.
	 * We do this here so we only need to take the CIL lock once during
	 * the transaction commit.
	 */
293
	spin_lock(&cil->xc_cil_lock);
294 295
	list_for_each_entry(lidp, &tp->t_items, lid_trans) {
		struct xfs_log_item	*lip = lidp->lid_item;
296

297 298 299
		/* Skip items which aren't dirty in this transaction. */
		if (!(lidp->lid_flags & XFS_LID_DIRTY))
			continue;
300

301
		list_move_tail(&lip->li_cil, &cil->xc_cil);
302
	}
303

304 305
	/* account for space used by new iovec headers  */
	len += diff_iovecs * sizeof(xlog_op_header_t);
306 307
	ctx->nvecs += diff_iovecs;

308 309 310 311
	/* attach the transaction to the CIL if it has any busy extents */
	if (!list_empty(&tp->t_busy))
		list_splice_init(&tp->t_busy, &ctx->busy_extents);

312 313 314 315 316 317 318 319 320
	/*
	 * Now transfer enough transaction reservation to the context ticket
	 * for the checkpoint. The context ticket is special - the unit
	 * reservation has to grow as well as the current reservation as we
	 * steal from tickets so we can correctly determine the space used
	 * during the transaction commit.
	 */
	if (ctx->ticket->t_curr_res == 0) {
		ctx->ticket->t_curr_res = ctx->ticket->t_unit_res;
321
		tp->t_ticket->t_curr_res -= ctx->ticket->t_unit_res;
322 323 324 325 326 327 328 329 330 331 332 333 334
	}

	/* do we need space for more log record headers? */
	iclog_space = log->l_iclog_size - log->l_iclog_hsize;
	if (len > 0 && (ctx->space_used / iclog_space !=
				(ctx->space_used + len) / iclog_space)) {
		int hdrs;

		hdrs = (len + iclog_space - 1) / iclog_space;
		/* need to take into account split region headers, too */
		hdrs *= log->l_iclog_hsize + sizeof(struct xlog_op_header);
		ctx->ticket->t_unit_res += hdrs;
		ctx->ticket->t_curr_res += hdrs;
335 336
		tp->t_ticket->t_curr_res -= hdrs;
		ASSERT(tp->t_ticket->t_curr_res >= len);
337
	}
338
	tp->t_ticket->t_curr_res -= len;
339 340 341
	ctx->space_used += len;

	spin_unlock(&cil->xc_cil_lock);
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
}

static void
xlog_cil_free_logvec(
	struct xfs_log_vec	*log_vector)
{
	struct xfs_log_vec	*lv;

	for (lv = log_vector; lv; ) {
		struct xfs_log_vec *next = lv->lv_next;
		kmem_free(lv);
		lv = next;
	}
}

/*
 * Mark all items committed and clear busy extents. We free the log vector
 * chains in a separate pass so that we unpin the log items as quickly as
 * possible.
 */
static void
xlog_cil_committed(
	void	*args,
	int	abort)
{
	struct xfs_cil_ctx	*ctx = args;
368
	struct xfs_mount	*mp = ctx->cil->xc_log->l_mp;
369

370 371
	xfs_trans_committed_bulk(ctx->cil->xc_log->l_ailp, ctx->lv_chain,
					ctx->start_lsn, abort);
372

373 374
	xfs_extent_busy_sort(&ctx->busy_extents);
	xfs_extent_busy_clear(mp, &ctx->busy_extents,
375
			     (mp->m_flags & XFS_MOUNT_DISCARD) && !abort);
376

Dave Chinner's avatar
Dave Chinner committed
377
	spin_lock(&ctx->cil->xc_push_lock);
378
	list_del(&ctx->committing);
Dave Chinner's avatar
Dave Chinner committed
379
	spin_unlock(&ctx->cil->xc_push_lock);
380 381

	xlog_cil_free_logvec(ctx->lv_chain);
382 383 384 385 386

	if (!list_empty(&ctx->busy_extents)) {
		ASSERT(mp->m_flags & XFS_MOUNT_DISCARD);

		xfs_discard_extents(mp, &ctx->busy_extents);
387
		xfs_extent_busy_clear(mp, &ctx->busy_extents, false);
388 389
	}

390 391 392 393
	kmem_free(ctx);
}

/*
394 395 396 397 398 399 400 401 402 403 404 405
 * Push the Committed Item List to the log. If @push_seq flag is zero, then it
 * is a background flush and so we can chose to ignore it. Otherwise, if the
 * current sequence is the same as @push_seq we need to do a flush. If
 * @push_seq is less than the current sequence, then it has already been
 * flushed and we don't need to do anything - the caller will wait for it to
 * complete if necessary.
 *
 * @push_seq is a value rather than a flag because that allows us to do an
 * unlocked check of the sequence number for a match. Hence we can allows log
 * forces to run racily and not issue pushes for the same sequence twice. If we
 * get a race between multiple pushes for the same sequence they will block on
 * the first one and then abort, hence avoiding needless pushes.
406
 */
407
STATIC int
408
xlog_cil_push(
409
	struct xlog		*log)
410 411 412 413 414 415 416 417 418 419 420 421 422
{
	struct xfs_cil		*cil = log->l_cilp;
	struct xfs_log_vec	*lv;
	struct xfs_cil_ctx	*ctx;
	struct xfs_cil_ctx	*new_ctx;
	struct xlog_in_core	*commit_iclog;
	struct xlog_ticket	*tic;
	int			num_iovecs;
	int			error = 0;
	struct xfs_trans_header thdr;
	struct xfs_log_iovec	lhdr;
	struct xfs_log_vec	lvhdr = { NULL };
	xfs_lsn_t		commit_lsn;
423
	xfs_lsn_t		push_seq;
424 425 426 427 428 429 430

	if (!cil)
		return 0;

	new_ctx = kmem_zalloc(sizeof(*new_ctx), KM_SLEEP|KM_NOFS);
	new_ctx->ticket = xlog_cil_ticket_alloc(log);

431
	down_write(&cil->xc_ctx_lock);
432 433
	ctx = cil->xc_ctx;

Dave Chinner's avatar
Dave Chinner committed
434
	spin_lock(&cil->xc_push_lock);
435 436
	push_seq = cil->xc_push_seq;
	ASSERT(push_seq <= ctx->sequence);
437

438 439 440 441 442 443 444
	/*
	 * Check if we've anything to push. If there is nothing, then we don't
	 * move on to a new sequence number and so we have to be able to push
	 * this sequence again later.
	 */
	if (list_empty(&cil->xc_cil)) {
		cil->xc_push_seq = 0;
Dave Chinner's avatar
Dave Chinner committed
445
		spin_unlock(&cil->xc_push_lock);
446
		goto out_skip;
447
	}
Dave Chinner's avatar
Dave Chinner committed
448
	spin_unlock(&cil->xc_push_lock);
449

450 451

	/* check for a previously pushed seqeunce */
452
	if (push_seq < cil->xc_ctx->sequence)
453 454
		goto out_skip;

455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489
	/*
	 * pull all the log vectors off the items in the CIL, and
	 * remove the items from the CIL. We don't need the CIL lock
	 * here because it's only needed on the transaction commit
	 * side which is currently locked out by the flush lock.
	 */
	lv = NULL;
	num_iovecs = 0;
	while (!list_empty(&cil->xc_cil)) {
		struct xfs_log_item	*item;

		item = list_first_entry(&cil->xc_cil,
					struct xfs_log_item, li_cil);
		list_del_init(&item->li_cil);
		if (!ctx->lv_chain)
			ctx->lv_chain = item->li_lv;
		else
			lv->lv_next = item->li_lv;
		lv = item->li_lv;
		item->li_lv = NULL;
		num_iovecs += lv->lv_niovecs;
	}

	/*
	 * initialise the new context and attach it to the CIL. Then attach
	 * the current context to the CIL committing lsit so it can be found
	 * during log forces to extract the commit lsn of the sequence that
	 * needs to be forced.
	 */
	INIT_LIST_HEAD(&new_ctx->committing);
	INIT_LIST_HEAD(&new_ctx->busy_extents);
	new_ctx->sequence = ctx->sequence + 1;
	new_ctx->cil = cil;
	cil->xc_ctx = new_ctx;

490 491 492 493 494 495 496
	/*
	 * mirror the new sequence into the cil structure so that we can do
	 * unlocked checks against the current sequence in log forces without
	 * risking deferencing a freed context pointer.
	 */
	cil->xc_current_sequence = new_ctx->sequence;

497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515
	/*
	 * The switch is now done, so we can drop the context lock and move out
	 * of a shared context. We can't just go straight to the commit record,
	 * though - we need to synchronise with previous and future commits so
	 * that the commit records are correctly ordered in the log to ensure
	 * that we process items during log IO completion in the correct order.
	 *
	 * For example, if we get an EFI in one checkpoint and the EFD in the
	 * next (e.g. due to log forces), we do not want the checkpoint with
	 * the EFD to be committed before the checkpoint with the EFI.  Hence
	 * we must strictly order the commit records of the checkpoints so
	 * that: a) the checkpoint callbacks are attached to the iclogs in the
	 * correct order; and b) the checkpoints are replayed in correct order
	 * in log recovery.
	 *
	 * Hence we need to add this context to the committing context list so
	 * that higher sequences will wait for us to write out a commit record
	 * before they do.
	 */
Dave Chinner's avatar
Dave Chinner committed
516
	spin_lock(&cil->xc_push_lock);
517
	list_add(&ctx->committing, &cil->xc_committing);
Dave Chinner's avatar
Dave Chinner committed
518
	spin_unlock(&cil->xc_push_lock);
519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534
	up_write(&cil->xc_ctx_lock);

	/*
	 * Build a checkpoint transaction header and write it to the log to
	 * begin the transaction. We need to account for the space used by the
	 * transaction header here as it is not accounted for in xlog_write().
	 *
	 * The LSN we need to pass to the log items on transaction commit is
	 * the LSN reported by the first log vector write. If we use the commit
	 * record lsn then we can move the tail beyond the grant write head.
	 */
	tic = ctx->ticket;
	thdr.th_magic = XFS_TRANS_HEADER_MAGIC;
	thdr.th_type = XFS_TRANS_CHECKPOINT;
	thdr.th_tid = tic->t_tid;
	thdr.th_num_items = num_iovecs;
535
	lhdr.i_addr = &thdr;
536 537 538 539 540 541 542 543 544 545
	lhdr.i_len = sizeof(xfs_trans_header_t);
	lhdr.i_type = XLOG_REG_TYPE_TRANSHDR;
	tic->t_curr_res -= lhdr.i_len + sizeof(xlog_op_header_t);

	lvhdr.lv_niovecs = 1;
	lvhdr.lv_iovecp = &lhdr;
	lvhdr.lv_next = ctx->lv_chain;

	error = xlog_write(log, &lvhdr, tic, &ctx->start_lsn, NULL, 0);
	if (error)
546
		goto out_abort_free_ticket;
547 548 549 550 551 552

	/*
	 * now that we've written the checkpoint into the log, strictly
	 * order the commit records so replay will get them in the right order.
	 */
restart:
Dave Chinner's avatar
Dave Chinner committed
553
	spin_lock(&cil->xc_push_lock);
554 555 556 557 558 559 560 561 562 563 564 565
	list_for_each_entry(new_ctx, &cil->xc_committing, committing) {
		/*
		 * Higher sequences will wait for this one so skip them.
		 * Don't wait for own own sequence, either.
		 */
		if (new_ctx->sequence >= ctx->sequence)
			continue;
		if (!new_ctx->commit_lsn) {
			/*
			 * It is still being pushed! Wait for the push to
			 * complete, then start again from the beginning.
			 */
Dave Chinner's avatar
Dave Chinner committed
566
			xlog_wait(&cil->xc_commit_wait, &cil->xc_push_lock);
567 568 569
			goto restart;
		}
	}
Dave Chinner's avatar
Dave Chinner committed
570
	spin_unlock(&cil->xc_push_lock);
571

572
	/* xfs_log_done always frees the ticket on error. */
573
	commit_lsn = xfs_log_done(log->l_mp, tic, &commit_iclog, 0);
574
	if (commit_lsn == -1)
575 576 577 578 579 580 581 582 583 584 585 586 587 588
		goto out_abort;

	/* attach all the transactions w/ busy extents to iclog */
	ctx->log_cb.cb_func = xlog_cil_committed;
	ctx->log_cb.cb_arg = ctx;
	error = xfs_log_notify(log->l_mp, commit_iclog, &ctx->log_cb);
	if (error)
		goto out_abort;

	/*
	 * now the checkpoint commit is complete and we've attached the
	 * callbacks to the iclog we can assign the commit LSN to the context
	 * and wake up anyone who is waiting for the commit to complete.
	 */
Dave Chinner's avatar
Dave Chinner committed
589
	spin_lock(&cil->xc_push_lock);
590
	ctx->commit_lsn = commit_lsn;
591
	wake_up_all(&cil->xc_commit_wait);
Dave Chinner's avatar
Dave Chinner committed
592
	spin_unlock(&cil->xc_push_lock);
593 594 595 596 597 598 599 600 601 602

	/* release the hounds! */
	return xfs_log_release_iclog(log->l_mp, commit_iclog);

out_skip:
	up_write(&cil->xc_ctx_lock);
	xfs_log_ticket_put(new_ctx->ticket);
	kmem_free(new_ctx);
	return 0;

603 604
out_abort_free_ticket:
	xfs_log_ticket_put(tic);
605 606 607 608 609
out_abort:
	xlog_cil_committed(ctx, XFS_LI_ABORTED);
	return XFS_ERROR(EIO);
}

610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627
static void
xlog_cil_push_work(
	struct work_struct	*work)
{
	struct xfs_cil		*cil = container_of(work, struct xfs_cil,
							xc_push_work);
	xlog_cil_push(cil->xc_log);
}

/*
 * We need to push CIL every so often so we don't cache more than we can fit in
 * the log. The limit really is that a checkpoint can't be more than half the
 * log (the current checkpoint is not allowed to overwrite the previous
 * checkpoint), but commit latency and memory usage limit this to a smaller
 * size.
 */
static void
xlog_cil_push_background(
628
	struct xlog	*log)
629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644
{
	struct xfs_cil	*cil = log->l_cilp;

	/*
	 * The cil won't be empty because we are called while holding the
	 * context lock so whatever we added to the CIL will still be there
	 */
	ASSERT(!list_empty(&cil->xc_cil));

	/*
	 * don't do a background push if we haven't used up all the
	 * space available yet.
	 */
	if (cil->xc_ctx->space_used < XLOG_CIL_SPACE_LIMIT(log))
		return;

Dave Chinner's avatar
Dave Chinner committed
645
	spin_lock(&cil->xc_push_lock);
646 647 648 649
	if (cil->xc_push_seq < cil->xc_current_sequence) {
		cil->xc_push_seq = cil->xc_current_sequence;
		queue_work(log->l_mp->m_cil_workqueue, &cil->xc_push_work);
	}
Dave Chinner's avatar
Dave Chinner committed
650
	spin_unlock(&cil->xc_push_lock);
651 652 653 654 655

}

static void
xlog_cil_push_foreground(
656
	struct xlog	*log,
657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672
	xfs_lsn_t	push_seq)
{
	struct xfs_cil	*cil = log->l_cilp;

	if (!cil)
		return;

	ASSERT(push_seq && push_seq <= cil->xc_current_sequence);

	/* start on any pending background push to minimise wait time on it */
	flush_work(&cil->xc_push_work);

	/*
	 * If the CIL is empty or we've already pushed the sequence then
	 * there's no work we need to do.
	 */
Dave Chinner's avatar
Dave Chinner committed
673
	spin_lock(&cil->xc_push_lock);
674
	if (list_empty(&cil->xc_cil) || push_seq <= cil->xc_push_seq) {
Dave Chinner's avatar
Dave Chinner committed
675
		spin_unlock(&cil->xc_push_lock);
676 677 678 679
		return;
	}

	cil->xc_push_seq = push_seq;
Dave Chinner's avatar
Dave Chinner committed
680
	spin_unlock(&cil->xc_push_lock);
681 682 683 684 685

	/* do the push now */
	xlog_cil_push(log);
}

686 687 688 689 690 691 692 693 694 695 696 697 698 699
bool
xlog_cil_empty(
	struct xlog	*log)
{
	struct xfs_cil	*cil = log->l_cilp;
	bool		empty = false;

	spin_lock(&cil->xc_push_lock);
	if (list_empty(&cil->xc_cil))
		empty = true;
	spin_unlock(&cil->xc_push_lock);
	return empty;
}

700 701 702 703 704 705 706 707 708 709 710 711 712
/*
 * Commit a transaction with the given vector to the Committed Item List.
 *
 * To do this, we need to format the item, pin it in memory if required and
 * account for the space used by the transaction. Once we have done that we
 * need to release the unused reservation for the transaction, attach the
 * transaction to the checkpoint context so we carry the busy extents through
 * to checkpoint completion, and then unlock all the items in the transaction.
 *
 * Called with the context lock already held in read mode to lock out
 * background commit, returns without it held once background commits are
 * allowed again.
 */
713
void
714 715 716 717 718 719
xfs_log_commit_cil(
	struct xfs_mount	*mp,
	struct xfs_trans	*tp,
	xfs_lsn_t		*commit_lsn,
	int			flags)
{
720
	struct xlog		*log = mp->m_log;
721
	struct xfs_cil		*cil = log->l_cilp;
722 723 724 725 726
	int			log_flags = 0;

	if (flags & XFS_TRANS_RELEASE_LOG_RES)
		log_flags = XFS_LOG_REL_PERM_RESERV;

727
	/* lock out background commit */
728
	down_read(&cil->xc_ctx_lock);
729

730
	xlog_cil_insert_items(log, tp);
731 732 733

	/* check we didn't blow the reservation */
	if (tp->t_ticket->t_curr_res < 0)
734
		xlog_print_tic_res(mp, tp->t_ticket);
735

736 737 738
	tp->t_commit_lsn = cil->xc_ctx->sequence;
	if (commit_lsn)
		*commit_lsn = tp->t_commit_lsn;
739 740 741 742 743 744 745 746 747 748 749 750 751 752 753

	xfs_log_done(mp, tp->t_ticket, NULL, log_flags);
	xfs_trans_unreserve_and_mod_sb(tp);

	/*
	 * Once all the items of the transaction have been copied to the CIL,
	 * the items can be unlocked and freed.
	 *
	 * This needs to be done before we drop the CIL context lock because we
	 * have to update state in the log items and unlock them before they go
	 * to disk. If we don't, then the CIL checkpoint can race with us and
	 * we can run checkpoint completion before we've updated and unlocked
	 * the log items. This affects (at least) processing of stale buffers,
	 * inodes and EFIs.
	 */
754
	xfs_trans_free_items(tp, tp->t_commit_lsn, 0);
755

756
	xlog_cil_push_background(log);
757

758
	up_read(&cil->xc_ctx_lock);
759 760
}

761 762 763 764 765 766 767 768 769 770 771
/*
 * Conditionally push the CIL based on the sequence passed in.
 *
 * We only need to push if we haven't already pushed the sequence
 * number given. Hence the only time we will trigger a push here is
 * if the push sequence is the same as the current context.
 *
 * We return the current commit lsn to allow the callers to determine if a
 * iclog flush is necessary following this call.
 */
xfs_lsn_t
772
xlog_cil_force_lsn(
773
	struct xlog	*log,
774
	xfs_lsn_t	sequence)
775 776 777 778 779
{
	struct xfs_cil		*cil = log->l_cilp;
	struct xfs_cil_ctx	*ctx;
	xfs_lsn_t		commit_lsn = NULLCOMMITLSN;

780 781 782 783 784 785 786
	ASSERT(sequence <= cil->xc_current_sequence);

	/*
	 * check to see if we need to force out the current context.
	 * xlog_cil_push() handles racing pushes for the same sequence,
	 * so no need to deal with it here.
	 */
787
	xlog_cil_push_foreground(log, sequence);
788 789 790 791 792 793 794

	/*
	 * See if we can find a previous sequence still committing.
	 * We need to wait for all previous sequence commits to complete
	 * before allowing the force of push_seq to go ahead. Hence block
	 * on commits for those as well.
	 */
795
restart:
Dave Chinner's avatar
Dave Chinner committed
796
	spin_lock(&cil->xc_push_lock);
797
	list_for_each_entry(ctx, &cil->xc_committing, committing) {
798
		if (ctx->sequence > sequence)
799 800 801 802 803 804
			continue;
		if (!ctx->commit_lsn) {
			/*
			 * It is still being pushed! Wait for the push to
			 * complete, then start again from the beginning.
			 */
Dave Chinner's avatar
Dave Chinner committed
805
			xlog_wait(&cil->xc_commit_wait, &cil->xc_push_lock);
806 807
			goto restart;
		}
808
		if (ctx->sequence != sequence)
809 810 811 812
			continue;
		/* found it! */
		commit_lsn = ctx->commit_lsn;
	}
Dave Chinner's avatar
Dave Chinner committed
813
	spin_unlock(&cil->xc_push_lock);
814 815
	return commit_lsn;
}
816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845

/*
 * Check if the current log item was first committed in this sequence.
 * We can't rely on just the log item being in the CIL, we have to check
 * the recorded commit sequence number.
 *
 * Note: for this to be used in a non-racy manner, it has to be called with
 * CIL flushing locked out. As a result, it should only be used during the
 * transaction commit process when deciding what to format into the item.
 */
bool
xfs_log_item_in_current_chkpt(
	struct xfs_log_item *lip)
{
	struct xfs_cil_ctx *ctx;

	if (list_empty(&lip->li_cil))
		return false;

	ctx = lip->li_mountp->m_log->l_cilp->xc_ctx;

	/*
	 * li_seq is written on the first commit of a log item to record the
	 * first checkpoint it is written to. Hence if it is different to the
	 * current sequence, we're in a new checkpoint.
	 */
	if (XFS_LSN_CMP(lip->li_seq, ctx->sequence) != 0)
		return false;
	return true;
}
846 847 848 849 850 851

/*
 * Perform initial CIL structure initialisation.
 */
int
xlog_cil_init(
852
	struct xlog	*log)
853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870
{
	struct xfs_cil	*cil;
	struct xfs_cil_ctx *ctx;

	cil = kmem_zalloc(sizeof(*cil), KM_SLEEP|KM_MAYFAIL);
	if (!cil)
		return ENOMEM;

	ctx = kmem_zalloc(sizeof(*ctx), KM_SLEEP|KM_MAYFAIL);
	if (!ctx) {
		kmem_free(cil);
		return ENOMEM;
	}

	INIT_WORK(&cil->xc_push_work, xlog_cil_push_work);
	INIT_LIST_HEAD(&cil->xc_cil);
	INIT_LIST_HEAD(&cil->xc_committing);
	spin_lock_init(&cil->xc_cil_lock);
Dave Chinner's avatar
Dave Chinner committed
871
	spin_lock_init(&cil->xc_push_lock);
872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888
	init_rwsem(&cil->xc_ctx_lock);
	init_waitqueue_head(&cil->xc_commit_wait);

	INIT_LIST_HEAD(&ctx->committing);
	INIT_LIST_HEAD(&ctx->busy_extents);
	ctx->sequence = 1;
	ctx->cil = cil;
	cil->xc_ctx = ctx;
	cil->xc_current_sequence = ctx->sequence;

	cil->xc_log = log;
	log->l_cilp = cil;
	return 0;
}

void
xlog_cil_destroy(
889
	struct xlog	*log)
890 891 892 893 894 895 896 897 898 899 900
{
	if (log->l_cilp->xc_ctx) {
		if (log->l_cilp->xc_ctx->ticket)
			xfs_log_ticket_put(log->l_cilp->xc_ctx->ticket);
		kmem_free(log->l_cilp->xc_ctx);
	}

	ASSERT(list_empty(&log->l_cilp->xc_cil));
	kmem_free(log->l_cilp);
}