aboutsummaryrefslogtreecommitdiff
path: root/vendor/github.com/etcd-io/bbolt/tx.go
blob: f50864142762cadcd3a323d95177156c332a3f61 (plain)
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
82
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
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
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
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
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
382
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
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
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
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
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
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
package bbolt

import (
	"fmt"
	"io"
	"os"
	"sort"
	"strings"
	"time"
	"unsafe"
)

// txid represents the internal transaction identifier.
type txid uint64

// Tx represents a read-only or read/write transaction on the database.
// Read-only transactions can be used for retrieving values for keys and creating cursors.
// Read/write transactions can create and remove buckets and create and remove keys.
//
// IMPORTANT: You must commit or rollback transactions when you are done with
// them. Pages can not be reclaimed by the writer until no more transactions
// are using them. A long running read transaction can cause the database to
// quickly grow.
type Tx struct {
	writable       bool
	managed        bool
	db             *DB
	meta           *meta
	root           Bucket
	pages          map[pgid]*page
	stats          TxStats
	commitHandlers []func()

	// WriteFlag specifies the flag for write-related methods like WriteTo().
	// Tx opens the database file with the specified flag to copy the data.
	//
	// By default, the flag is unset, which works well for mostly in-memory
	// workloads. For databases that are much larger than available RAM,
	// set the flag to syscall.O_DIRECT to avoid trashing the page cache.
	WriteFlag int
}

// init initializes the transaction.
func (tx *Tx) init(db *DB) {
	tx.db = db
	tx.pages = nil

	// Copy the meta page since it can be changed by the writer.
	tx.meta = &meta{}
	db.meta().copy(tx.meta)

	// Copy over the root bucket.
	tx.root = newBucket(tx)
	tx.root.bucket = &bucket{}
	*tx.root.bucket = tx.meta.root

	// Increment the transaction id and add a page cache for writable transactions.
	if tx.writable {
		tx.pages = make(map[pgid]*page)
		tx.meta.txid += txid(1)
	}
}

// ID returns the transaction id.
func (tx *Tx) ID() int {
	return int(tx.meta.txid)
}

// DB returns a reference to the database that created the transaction.
func (tx *Tx) DB() *DB {
	return tx.db
}

// Size returns current database size in bytes as seen by this transaction.
func (tx *Tx) Size() int64 {
	return int64(tx.meta.pgid) * int64(tx.db.pageSize)
}

// Writable returns whether the transaction can perform write operations.
func (tx *Tx) Writable() bool {
	return tx.writable
}

// Cursor creates a cursor associated with the root bucket.
// All items in the cursor will return a nil value because all root bucket keys point to buckets.
// The cursor is only valid as long as the transaction is open.
// Do not use a cursor after the transaction is closed.
func (tx *Tx) Cursor() *Cursor {
	return tx.root.Cursor()
}

// Stats retrieves a copy of the current transaction statistics.
func (tx *Tx) Stats() TxStats {
	return tx.stats
}

// Bucket retrieves a bucket by name.
// Returns nil if the bucket does not exist.
// The bucket instance is only valid for the lifetime of the transaction.
func (tx *Tx) Bucket(name []byte) *Bucket {
	return tx.root.Bucket(name)
}

// CreateBucket creates a new bucket.
// Returns an error if the bucket already exists, if the bucket name is blank, or if the bucket name is too long.
// The bucket instance is only valid for the lifetime of the transaction.
func (tx *Tx) CreateBucket(name []byte) (*Bucket, error) {
	return tx.root.CreateBucket(name)
}

// CreateBucketIfNotExists creates a new bucket if it doesn't already exist.
// Returns an error if the bucket name is blank, or if the bucket name is too long.
// The bucket instance is only valid for the lifetime of the transaction.
func (tx *Tx) CreateBucketIfNotExists(name []byte) (*Bucket, error) {
	return tx.root.CreateBucketIfNotExists(name)
}

// DeleteBucket deletes a bucket.
// Returns an error if the bucket cannot be found or if the key represents a non-bucket value.
func (tx *Tx) DeleteBucket(name []byte) error {
	return tx.root.DeleteBucket(name)
}

// ForEach executes a function for each bucket in the root.
// If the provided function returns an error then the iteration is stopped and
// the error is returned to the caller.
func (tx *Tx) ForEach(fn func(name []byte, b *Bucket) error) error {
	return tx.root.ForEach(func(k, v []byte) error {
		return fn(k, tx.root.Bucket(k))
	})
}

// OnCommit adds a handler function to be executed after the transaction successfully commits.
func (tx *Tx) OnCommit(fn func()) {
	tx.commitHandlers = append(tx.commitHandlers, fn)
}

// Commit writes all changes to disk and updates the meta page.
// Returns an error if a disk write error occurs, or if Commit is
// called on a read-only transaction.
func (tx *Tx) Commit() error {
	_assert(!tx.managed, "managed tx commit not allowed")
	if tx.db == nil {
		return ErrTxClosed
	} else if !tx.writable {
		return ErrTxNotWritable
	}

	// TODO(benbjohnson): Use vectorized I/O to write out dirty pages.

	// Rebalance nodes which have had deletions.
	var startTime = time.Now()
	tx.root.rebalance()
	if tx.stats.Rebalance > 0 {
		tx.stats.RebalanceTime += time.Since(startTime)
	}

	// spill data onto dirty pages.
	startTime = time.Now()
	if err := tx.root.spill(); err != nil {
		tx.rollback()
		return err
	}
	tx.stats.SpillTime += time.Since(startTime)

	// Free the old root bucket.
	tx.meta.root.root = tx.root.root

	// Free the old freelist because commit writes out a fresh freelist.
	if tx.meta.freelist != pgidNoFreelist {
		tx.db.freelist.free(tx.meta.txid, tx.db.page(tx.meta.freelist))
	}

	if !tx.db.NoFreelistSync {
		err := tx.commitFreelist()
		if err != nil {
			return err
		}
	} else {
		tx.meta.freelist = pgidNoFreelist
	}

	// Write dirty pages to disk.
	startTime = time.Now()
	if err := tx.write(); err != nil {
		tx.rollback()
		return err
	}

	// If strict mode is enabled then perform a consistency check.
	// Only the first consistency error is reported in the panic.
	if tx.db.StrictMode {
		ch := tx.Check()
		var errs []string
		for {
			err, ok := <-ch
			if !ok {
				break
			}
			errs = append(errs, err.Error())
		}
		if len(errs) > 0 {
			panic("check fail: " + strings.Join(errs, "\n"))
		}
	}

	// Write meta to disk.
	if err := tx.writeMeta(); err != nil {
		tx.rollback()
		return err
	}
	tx.stats.WriteTime += time.Since(startTime)

	// Finalize the transaction.
	tx.close()

	// Execute commit handlers now that the locks have been removed.
	for _, fn := range tx.commitHandlers {
		fn()
	}

	return nil
}

func (tx *Tx) commitFreelist() error {
	// Allocate new pages for the new free list. This will overestimate
	// the size of the freelist but not underestimate the size (which would be bad).
	opgid := tx.meta.pgid
	p, err := tx.allocate((tx.db.freelist.size() / tx.db.pageSize) + 1)
	if err != nil {
		tx.rollback()
		return err
	}
	if err := tx.db.freelist.write(p); err != nil {
		tx.rollback()
		return err
	}
	tx.meta.freelist = p.id
	// If the high water mark has moved up then attempt to grow the database.
	if tx.meta.pgid > opgid {
		if err := tx.db.grow(int(tx.meta.pgid+1) * tx.db.pageSize); err != nil {
			tx.rollback()
			return err
		}
	}

	return nil
}

// Rollback closes the transaction and ignores all previous updates. Read-only
// transactions must be rolled back and not committed.
func (tx *Tx) Rollback() error {
	_assert(!tx.managed, "managed tx rollback not allowed")
	if tx.db == nil {
		return ErrTxClosed
	}
	tx.rollback()
	return nil
}

func (tx *Tx) rollback() {
	if tx.db == nil {
		return
	}
	if tx.writable {
		tx.db.freelist.rollback(tx.meta.txid)
		tx.db.freelist.reload(tx.db.page(tx.db.meta().freelist))
	}
	tx.close()
}

func (tx *Tx) close() {
	if tx.db == nil {
		return
	}
	if tx.writable {
		// Grab freelist stats.
		var freelistFreeN = tx.db.freelist.free_count()
		var freelistPendingN = tx.db.freelist.pending_count()
		var freelistAlloc = tx.db.freelist.size()

		// Remove transaction ref & writer lock.
		tx.db.rwtx = nil
		tx.db.rwlock.Unlock()

		// Merge statistics.
		tx.db.statlock.Lock()
		tx.db.stats.FreePageN = freelistFreeN
		tx.db.stats.PendingPageN = freelistPendingN
		tx.db.stats.FreeAlloc = (freelistFreeN + freelistPendingN) * tx.db.pageSize
		tx.db.stats.FreelistInuse = freelistAlloc
		tx.db.stats.TxStats.add(&tx.stats)
		tx.db.statlock.Unlock()
	} else {
		tx.db.removeTx(tx)
	}

	// Clear all references.
	tx.db = nil
	tx.meta = nil
	tx.root = Bucket{tx: tx}
	tx.pages = nil
}

// Copy writes the entire database to a writer.
// This function exists for backwards compatibility.
//
// Deprecated; Use WriteTo() instead.
func (tx *Tx) Copy(w io.Writer) error {
	_, err := tx.WriteTo(w)
	return err
}

// WriteTo writes the entire database to a writer.
// If err == nil then exactly tx.Size() bytes will be written into the writer.
func (tx *Tx) WriteTo(w io.Writer) (n int64, err error) {
	// Attempt to open reader with WriteFlag
	f, err := os.OpenFile(tx.db.path, os.O_RDONLY|tx.WriteFlag, 0)
	if err != nil {
		return 0, err
	}
	defer func() {
		if cerr := f.Close(); err == nil {
			err = cerr
		}
	}()

	// Generate a meta page. We use the same page data for both meta pages.
	buf := make([]byte, tx.db.pageSize)
	page := (*page)(unsafe.Pointer(&buf[0]))
	page.flags = metaPageFlag
	*page.meta() = *tx.meta

	// Write meta 0.
	page.id = 0
	page.meta().checksum = page.meta().sum64()
	nn, err := w.Write(buf)
	n += int64(nn)
	if err != nil {
		return n, fmt.Errorf("meta 0 copy: %s", err)
	}

	// Write meta 1 with a lower transaction id.
	page.id = 1
	page.meta().txid -= 1
	page.meta().checksum = page.meta().sum64()
	nn, err = w.Write(buf)
	n += int64(nn)
	if err != nil {
		return n, fmt.Errorf("meta 1 copy: %s", err)
	}

	// Move past the meta pages in the file.
	if _, err := f.Seek(int64(tx.db.pageSize*2), io.SeekStart); err != nil {
		return n, fmt.Errorf("seek: %s", err)
	}

	// Copy data pages.
	wn, err := io.CopyN(w, f, tx.Size()-int64(tx.db.pageSize*2))
	n += wn
	if err != nil {
		return n, err
	}

	return n, nil
}

// CopyFile copies the entire database to file at the given path.
// A reader transaction is maintained during the copy so it is safe to continue
// using the database while a copy is in progress.
func (tx *Tx) CopyFile(path string, mode os.FileMode) error {
	f, err := os.OpenFile(path, os.O_RDWR|os.O_CREATE|os.O_TRUNC, mode)
	if err != nil {
		return err
	}

	err = tx.Copy(f)
	if err != nil {
		_ = f.Close()
		return err
	}
	return f.Close()
}

// Check performs several consistency checks on the database for this transaction.
// An error is returned if any inconsistency is found.
//
// It can be safely run concurrently on a writable transaction. However, this
// incurs a high cost for large databases and databases with a lot of subbuckets
// because of caching. This overhead can be removed if running on a read-only
// transaction, however, it is not safe to execute other writer transactions at
// the same time.
func (tx *Tx) Check() <-chan error {
	ch := make(chan error)
	go tx.check(ch)
	return ch
}

func (tx *Tx) check(ch chan error) {
	// Force loading free list if opened in ReadOnly mode.
	tx.db.loadFreelist()

	// Check if any pages are double freed.
	freed := make(map[pgid]bool)
	all := make([]pgid, tx.db.freelist.count())
	tx.db.freelist.copyall(all)
	for _, id := range all {
		if freed[id] {
			ch <- fmt.Errorf("page %d: already freed", id)
		}
		freed[id] = true
	}

	// Track every reachable page.
	reachable := make(map[pgid]*page)
	reachable[0] = tx.page(0) // meta0
	reachable[1] = tx.page(1) // meta1
	if tx.meta.freelist != pgidNoFreelist {
		for i := uint32(0); i <= tx.page(tx.meta.freelist).overflow; i++ {
			reachable[tx.meta.freelist+pgid(i)] = tx.page(tx.meta.freelist)
		}
	}

	// Recursively check buckets.
	tx.checkBucket(&tx.root, reachable, freed, ch)

	// Ensure all pages below high water mark are either reachable or freed.
	for i := pgid(0); i < tx.meta.pgid; i++ {
		_, isReachable := reachable[i]
		if !isReachable && !freed[i] {
			ch <- fmt.Errorf("page %d: unreachable unfreed", int(i))
		}
	}

	// Close the channel to signal completion.
	close(ch)
}

func (tx *Tx) checkBucket(b *Bucket, reachable map[pgid]*page, freed map[pgid]bool, ch chan error) {
	// Ignore inline buckets.
	if b.root == 0 {
		return
	}

	// Check every page used by this bucket.
	b.tx.forEachPage(b.root, 0, func(p *page, _ int) {
		if p.id > tx.meta.pgid {
			ch <- fmt.Errorf("page %d: out of bounds: %d", int(p.id), int(b.tx.meta.pgid))
		}

		// Ensure each page is only referenced once.
		for i := pgid(0); i <= pgid(p.overflow); i++ {
			var id = p.id + i
			if _, ok := reachable[id]; ok {
				ch <- fmt.Errorf("page %d: multiple references", int(id))
			}
			reachable[id] = p
		}

		// We should only encounter un-freed leaf and branch pages.
		if freed[p.id] {
			ch <- fmt.Errorf("page %d: reachable freed", int(p.id))
		} else if (p.flags&branchPageFlag) == 0 && (p.flags&leafPageFlag) == 0 {
			ch <- fmt.Errorf("page %d: invalid type: %s", int(p.id), p.typ())
		}
	})

	// Check each bucket within this bucket.
	_ = b.ForEach(func(k, v []byte) error {
		if child := b.Bucket(k); child != nil {
			tx.checkBucket(child, reachable, freed, ch)
		}
		return nil
	})
}

// allocate returns a contiguous block of memory starting at a given page.
func (tx *Tx) allocate(count int) (*page, error) {
	p, err := tx.db.allocate(tx.meta.txid, count)
	if err != nil {
		return nil, err
	}

	// Save to our page cache.
	tx.pages[p.id] = p

	// Update statistics.
	tx.stats.PageCount += count
	tx.stats.PageAlloc += count * tx.db.pageSize

	return p, nil
}

// write writes any dirty pages to disk.
func (tx *Tx) write() error {
	// Sort pages by id.
	pages := make(pages, 0, len(tx.pages))
	for _, p := range tx.pages {
		pages = append(pages, p)
	}
	// Clear out page cache early.
	tx.pages = make(map[pgid]*page)
	sort.Sort(pages)

	// Write pages to disk in order.
	for _, p := range pages {
		size := (int(p.overflow) + 1) * tx.db.pageSize
		offset := int64(p.id) * int64(tx.db.pageSize)

		// Write out page in "max allocation" sized chunks.
		ptr := (*[maxAllocSize]byte)(unsafe.Pointer(p))
		for {
			// Limit our write to our max allocation size.
			sz := size
			if sz > maxAllocSize-1 {
				sz = maxAllocSize - 1
			}

			// Write chunk to disk.
			buf := ptr[:sz]
			if _, err := tx.db.ops.writeAt(buf, offset); err != nil {
				return err
			}

			// Update statistics.
			tx.stats.Write++

			// Exit inner for loop if we've written all the chunks.
			size -= sz
			if size == 0 {
				break
			}

			// Otherwise move offset forward and move pointer to next chunk.
			offset += int64(sz)
			ptr = (*[maxAllocSize]byte)(unsafe.Pointer(&ptr[sz]))
		}
	}

	// Ignore file sync if flag is set on DB.
	if !tx.db.NoSync || IgnoreNoSync {
		if err := fdatasync(tx.db); err != nil {
			return err
		}
	}

	// Put small pages back to page pool.
	for _, p := range pages {
		// Ignore page sizes over 1 page.
		// These are allocated using make() instead of the page pool.
		if int(p.overflow) != 0 {
			continue
		}

		buf := (*[maxAllocSize]byte)(unsafe.Pointer(p))[:tx.db.pageSize]

		// See https://go.googlesource.com/go/+/f03c9202c43e0abb130669852082117ca50aa9b1
		for i := range buf {
			buf[i] = 0
		}
		tx.db.pagePool.Put(buf)
	}

	return nil
}

// writeMeta writes the meta to the disk.
func (tx *Tx) writeMeta() error {
	// Create a temporary buffer for the meta page.
	buf := make([]byte, tx.db.pageSize)
	p := tx.db.pageInBuffer(buf, 0)
	tx.meta.write(p)

	// Write the meta page to file.
	if _, err := tx.db.ops.writeAt(buf, int64(p.id)*int64(tx.db.pageSize)); err != nil {
		return err
	}
	if !tx.db.NoSync || IgnoreNoSync {
		if err := fdatasync(tx.db); err != nil {
			return err
		}
	}

	// Update statistics.
	tx.stats.Write++

	return nil
}

// page returns a reference to the page with a given id.
// If page has been written to then a temporary buffered page is returned.
func (tx *Tx) page(id pgid) *page {
	// Check the dirty pages first.
	if tx.pages != nil {
		if p, ok := tx.pages[id]; ok {
			return p
		}
	}

	// Otherwise return directly from the mmap.
	return tx.db.page(id)
}

// forEachPage iterates over every page within a given page and executes a function.
func (tx *Tx) forEachPage(pgid pgid, depth int, fn func(*page, int)) {
	p := tx.page(pgid)

	// Execute function.
	fn(p, depth)

	// Recursively loop over children.
	if (p.flags & branchPageFlag) != 0 {
		for i := 0; i < int(p.count); i++ {
			elem := p.branchPageElement(uint16(i))
			tx.forEachPage(elem.pgid, depth+1, fn)
		}
	}
}

// Page returns page information for a given page number.
// This is only safe for concurrent use when used by a writable transaction.
func (tx *Tx) Page(id int) (*PageInfo, error) {
	if tx.db == nil {
		return nil, ErrTxClosed
	} else if pgid(id) >= tx.meta.pgid {
		return nil, nil
	}

	// Build the page info.
	p := tx.db.page(pgid(id))
	info := &PageInfo{
		ID:            id,
		Count:         int(p.count),
		OverflowCount: int(p.overflow),
	}

	// Determine the type (or if it's free).
	if tx.db.freelist.freed(pgid(id)) {
		info.Type = "free"
	} else {
		info.Type = p.typ()
	}

	return info, nil
}

// TxStats represents statistics about the actions performed by the transaction.
type TxStats struct {
	// Page statistics.
	PageCount int // number of page allocations
	PageAlloc int // total bytes allocated

	// Cursor statistics.
	CursorCount int // number of cursors created

	// Node statistics
	NodeCount int // number of node allocations
	NodeDeref int // number of node dereferences

	// Rebalance statistics.
	Rebalance     int           // number of node rebalances
	RebalanceTime time.Duration // total time spent rebalancing

	// Split/Spill statistics.
	Split     int           // number of nodes split
	Spill     int           // number of nodes spilled
	SpillTime time.Duration // total time spent spilling

	// Write statistics.
	Write     int           // number of writes performed
	WriteTime time.Duration // total time spent writing to disk
}

func (s *TxStats) add(other *TxStats) {
	s.PageCount += other.PageCount
	s.PageAlloc += other.PageAlloc
	s.CursorCount += other.CursorCount
	s.NodeCount += other.NodeCount
	s.NodeDeref += other.NodeDeref
	s.Rebalance += other.Rebalance
	s.RebalanceTime += other.RebalanceTime
	s.Split += other.Split
	s.Spill += other.Spill
	s.SpillTime += other.SpillTime
	s.Write += other.Write
	s.WriteTime += other.WriteTime
}

// Sub calculates and returns the difference between two sets of transaction stats.
// This is useful when obtaining stats at two different points and time and
// you need the performance counters that occurred within that time span.
func (s *TxStats) Sub(other *TxStats) TxStats {
	var diff TxStats
	diff.PageCount = s.PageCount - other.PageCount
	diff.PageAlloc = s.PageAlloc - other.PageAlloc
	diff.CursorCount = s.CursorCount - other.CursorCount
	diff.NodeCount = s.NodeCount - other.NodeCount
	diff.NodeDeref = s.NodeDeref - other.NodeDeref
	diff.Rebalance = s.Rebalance - other.Rebalance
	diff.RebalanceTime = s.RebalanceTime - other.RebalanceTime
	diff.Split = s.Split - other.Split
	diff.Spill = s.Spill - other.Spill
	diff.SpillTime = s.SpillTime - other.SpillTime
	diff.Write = s.Write - other.Write
	diff.WriteTime = s.WriteTime - other.WriteTime
	return diff
}