Switch to github.com/etcd-io/bbolt

github.com/boltdb/bolt is no longer maintained.

Signed-off-by: Valentin Rothberg <rothberg@redhat.com>

vendor.conf

@@ -2,7 +2,7 @@
 github.com/urfave/cli v1.20.0
 github.com/kr/pretty v0.1.0
 github.com/kr/text v0.1.0
-github.com/containers/image be45336d5996db292aade0a0e3dbaf59a28a26b8
+github.com/containers/image e9c3d17ddb8cb5d48450bc0588525b17b181c049
 github.com/containers/buildah 25b7c1164a262934a6526ded1115dfacfc51f8ec
 github.com/vbauerster/mpb v3.3.4
 github.com/mattn/go-isatty v0.0.4
@@ -18,7 +18,7 @@ github.com/pmezard/go-difflib 5d4384ee4fb2527b0a1256a821ebfc92f91efefc
 github.com/pkg/errors v0.8.1
 golang.org/x/crypto a4c6cb3142f211c99e4bf4cd769535b29a9b616f
 github.com/ulikunitz/xz v0.5.4
-github.com/boltdb/bolt v1.3.1
+github.com/etcd-io/bbolt v1.3.2
 # docker deps from https://github.com/docker/docker/blob/v1.11.2/hack/vendor.sh
 github.com/docker/docker da99009bbb1165d1ac5688b5c81d2f589d418341
 github.com/docker/go-connections 7beb39f0b969b075d1325fecb092faf27fd357b6

vendor/github.com/boltdb/bolt/freelist.go

@@ -1,252 +0,0 @@
-package bolt
-
-import (
-	"fmt"
-	"sort"
-	"unsafe"
-)
-
-// freelist represents a list of all pages that are available for allocation.
-// It also tracks pages that have been freed but are still in use by open transactions.
-type freelist struct {
-	ids     []pgid          // all free and available free page ids.
-	pending map[txid][]pgid // mapping of soon-to-be free page ids by tx.
-	cache   map[pgid]bool   // fast lookup of all free and pending page ids.
-}
-
-// newFreelist returns an empty, initialized freelist.
-func newFreelist() *freelist {
-	return &freelist{
-		pending: make(map[txid][]pgid),
-		cache:   make(map[pgid]bool),
-	}
-}
-
-// size returns the size of the page after serialization.
-func (f *freelist) size() int {
-	n := f.count()
-	if n >= 0xFFFF {
-		// The first element will be used to store the count. See freelist.write.
-		n++
-	}
-	return pageHeaderSize + (int(unsafe.Sizeof(pgid(0))) * n)
-}
-
-// count returns count of pages on the freelist
-func (f *freelist) count() int {
-	return f.free_count() + f.pending_count()
-}
-
-// free_count returns count of free pages
-func (f *freelist) free_count() int {
-	return len(f.ids)
-}
-
-// pending_count returns count of pending pages
-func (f *freelist) pending_count() int {
-	var count int
-	for _, list := range f.pending {
-		count += len(list)
-	}
-	return count
-}
-
-// copyall copies into dst a list of all free ids and all pending ids in one sorted list.
-// f.count returns the minimum length required for dst.
-func (f *freelist) copyall(dst []pgid) {
-	m := make(pgids, 0, f.pending_count())
-	for _, list := range f.pending {
-		m = append(m, list...)
-	}
-	sort.Sort(m)
-	mergepgids(dst, f.ids, m)
-}
-
-// allocate returns the starting page id of a contiguous list of pages of a given size.
-// If a contiguous block cannot be found then 0 is returned.
-func (f *freelist) allocate(n int) pgid {
-	if len(f.ids) == 0 {
-		return 0
-	}
-
-	var initial, previd pgid
-	for i, id := range f.ids {
-		if id <= 1 {
-			panic(fmt.Sprintf("invalid page allocation: %d", id))
-		}
-
-		// Reset initial page if this is not contiguous.
-		if previd == 0 || id-previd != 1 {
-			initial = id
-		}
-
-		// If we found a contiguous block then remove it and return it.
-		if (id-initial)+1 == pgid(n) {
-			// If we're allocating off the beginning then take the fast path
-			// and just adjust the existing slice. This will use extra memory
-			// temporarily but the append() in free() will realloc the slice
-			// as is necessary.
-			if (i + 1) == n {
-				f.ids = f.ids[i+1:]
-			} else {
-				copy(f.ids[i-n+1:], f.ids[i+1:])
-				f.ids = f.ids[:len(f.ids)-n]
-			}
-
-			// Remove from the free cache.
-			for i := pgid(0); i < pgid(n); i++ {
-				delete(f.cache, initial+i)
-			}
-
-			return initial
-		}
-
-		previd = id
-	}
-	return 0
-}
-
-// free releases a page and its overflow for a given transaction id.
-// If the page is already free then a panic will occur.
-func (f *freelist) free(txid txid, p *page) {
-	if p.id <= 1 {
-		panic(fmt.Sprintf("cannot free page 0 or 1: %d", p.id))
-	}
-
-	// Free page and all its overflow pages.
-	var ids = f.pending[txid]
-	for id := p.id; id <= p.id+pgid(p.overflow); id++ {
-		// Verify that page is not already free.
-		if f.cache[id] {
-			panic(fmt.Sprintf("page %d already freed", id))
-		}
-
-		// Add to the freelist and cache.
-		ids = append(ids, id)
-		f.cache[id] = true
-	}
-	f.pending[txid] = ids
-}
-
-// release moves all page ids for a transaction id (or older) to the freelist.
-func (f *freelist) release(txid txid) {
-	m := make(pgids, 0)
-	for tid, ids := range f.pending {
-		if tid <= txid {
-			// Move transaction's pending pages to the available freelist.
-			// Don't remove from the cache since the page is still free.
-			m = append(m, ids...)
-			delete(f.pending, tid)
-		}
-	}
-	sort.Sort(m)
-	f.ids = pgids(f.ids).merge(m)
-}
-
-// rollback removes the pages from a given pending tx.
-func (f *freelist) rollback(txid txid) {
-	// Remove page ids from cache.
-	for _, id := range f.pending[txid] {
-		delete(f.cache, id)
-	}
-
-	// Remove pages from pending list.
-	delete(f.pending, txid)
-}
-
-// freed returns whether a given page is in the free list.
-func (f *freelist) freed(pgid pgid) bool {
-	return f.cache[pgid]
-}
-
-// read initializes the freelist from a freelist page.
-func (f *freelist) read(p *page) {
-	// If the page.count is at the max uint16 value (64k) then it's considered
-	// an overflow and the size of the freelist is stored as the first element.
-	idx, count := 0, int(p.count)
-	if count == 0xFFFF {
-		idx = 1
-		count = int(((*[maxAllocSize]pgid)(unsafe.Pointer(&p.ptr)))[0])
-	}
-
-	// Copy the list of page ids from the freelist.
-	if count == 0 {
-		f.ids = nil
-	} else {
-		ids := ((*[maxAllocSize]pgid)(unsafe.Pointer(&p.ptr)))[idx:count]
-		f.ids = make([]pgid, len(ids))
-		copy(f.ids, ids)
-
-		// Make sure they're sorted.
-		sort.Sort(pgids(f.ids))
-	}
-
-	// Rebuild the page cache.
-	f.reindex()
-}
-
-// write writes the page ids onto a freelist page. All free and pending ids are
-// saved to disk since in the event of a program crash, all pending ids will
-// become free.
-func (f *freelist) write(p *page) error {
-	// Combine the old free pgids and pgids waiting on an open transaction.
-
-	// Update the header flag.
-	p.flags |= freelistPageFlag
-
-	// The page.count can only hold up to 64k elements so if we overflow that
-	// number then we handle it by putting the size in the first element.
-	lenids := f.count()
-	if lenids == 0 {
-		p.count = uint16(lenids)
-	} else if lenids < 0xFFFF {
-		p.count = uint16(lenids)
-		f.copyall(((*[maxAllocSize]pgid)(unsafe.Pointer(&p.ptr)))[:])
-	} else {
-		p.count = 0xFFFF
-		((*[maxAllocSize]pgid)(unsafe.Pointer(&p.ptr)))[0] = pgid(lenids)
-		f.copyall(((*[maxAllocSize]pgid)(unsafe.Pointer(&p.ptr)))[1:])
-	}
-
-	return nil
-}
-
-// reload reads the freelist from a page and filters out pending items.
-func (f *freelist) reload(p *page) {
-	f.read(p)
-
-	// Build a cache of only pending pages.
-	pcache := make(map[pgid]bool)
-	for _, pendingIDs := range f.pending {
-		for _, pendingID := range pendingIDs {
-			pcache[pendingID] = true
-		}
-	}
-
-	// Check each page in the freelist and build a new available freelist
-	// with any pages not in the pending lists.
-	var a []pgid
-	for _, id := range f.ids {
-		if !pcache[id] {
-			a = append(a, id)
-		}
-	}
-	f.ids = a
-
-	// Once the available list is rebuilt then rebuild the free cache so that
-	// it includes the available and pending free pages.
-	f.reindex()
-}
-
-// reindex rebuilds the free cache based on available and pending free lists.
-func (f *freelist) reindex() {
-	f.cache = make(map[pgid]bool, len(f.ids))
-	for _, id := range f.ids {
-		f.cache[id] = true
-	}
-	for _, pendingIDs := range f.pending {
-		for _, pendingID := range pendingIDs {
-			f.cache[pendingID] = true
-		}
-	}
-}

vendor/github.com/containers/image/pkg/blobinfocache/boltdb/boltdb.go

@@ -7,9 +7,9 @@ import (
 	"sync"
 	"time"
 
-	"github.com/boltdb/bolt"
 	"github.com/containers/image/pkg/blobinfocache/internal/prioritize"
 	"github.com/containers/image/types"
+	bolt "github.com/etcd-io/bbolt"
 	"github.com/opencontainers/go-digest"
 	"github.com/sirupsen/logrus"
 )

vendor/github.com/containers/image/pkg/docker/config/config.go

@@ -85,21 +85,6 @@ func GetAuthentication(sys *types.SystemContext, registry string) (string, strin
 	return "", "", nil
 }
 
-// GetUserLoggedIn returns the username logged in to registry from either
-// auth.json or XDG_RUNTIME_DIR
-// Used to tell the user if someone is logged in to the registry when logging in
-func GetUserLoggedIn(sys *types.SystemContext, registry string) (string, error) {
-	path, err := getPathToAuth(sys)
-	if err != nil {
-		return "", err
-	}
-	username, _, _ := findAuthentication(registry, path, false)
-	if username != "" {
-		return username, nil
-	}
-	return "", nil
-}
-
 // RemoveAuthentication deletes the credentials stored in auth.json
 func RemoveAuthentication(sys *types.SystemContext, registry string) error {
 	return modifyJSON(sys, func(auths *dockerConfigFile) (bool, error) {

vendor/github.com/containers/image/vendor.conf

@@ -13,7 +13,6 @@ github.com/containerd/continuity d8fb8589b0e8e85b8c8bbaa8840226d0dfeb7371
 github.com/ghodss/yaml 04f313413ffd65ce25f2541bfd2b2ceec5c0908c
 github.com/gorilla/mux 94e7d24fd285520f3d12ae998f7fdd6b5393d453
 github.com/imdario/mergo 50d4dbd4eb0e84778abe37cefef140271d96fade
-github.com/mattn/go-runewidth 14207d285c6c197daabb5c9793d63e7af9ab2d50
 github.com/mistifyio/go-zfs c0224de804d438efd11ea6e52ada8014537d6062
 github.com/mtrmac/gpgme b2432428689ca58c2b8e8dea9449d3295cf96fc9
 github.com/opencontainers/go-digest c9281466c8b2f606084ac71339773efd177436e7
@@ -43,7 +42,7 @@ github.com/syndtr/gocapability master
 github.com/Microsoft/go-winio ab35fc04b6365e8fcb18e6e9e41ea4a02b10b175
 github.com/Microsoft/hcsshim eca7177590cdcbd25bbc5df27e3b693a54b53a6a
 github.com/ulikunitz/xz v0.5.4
-github.com/boltdb/bolt master
+github.com/etcd-io/bbolt v1.3.2
 github.com/klauspost/pgzip v1.2.1
 github.com/klauspost/compress v1.4.1
 github.com/klauspost/cpuid v1.2.0

vendor/github.com/etcd-io/bbolt/README.md

@@ -1,5 +1,18 @@
-Bolt [![Coverage Status](https://coveralls.io/repos/boltdb/bolt/badge.svg?branch=master)](https://coveralls.io/r/boltdb/bolt?branch=master) [![GoDoc](https://godoc.org/github.com/boltdb/bolt?status.svg)](https://godoc.org/github.com/boltdb/bolt) ![Version](https://img.shields.io/badge/version-1.2.1-green.svg)
+bbolt
-====
+=====
+
+[![Go Report Card](https://goreportcard.com/badge/github.com/etcd-io/bbolt?style=flat-square)](https://goreportcard.com/report/github.com/etcd-io/bbolt)
+[![Coverage](https://codecov.io/gh/etcd-io/bbolt/branch/master/graph/badge.svg)](https://codecov.io/gh/etcd-io/bbolt)
+[![Build Status Travis](https://img.shields.io/travis/etcd-io/bboltlabs.svg?style=flat-square&&branch=master)](https://travis-ci.com/etcd-io/bbolt)
+[![Godoc](http://img.shields.io/badge/go-documentation-blue.svg?style=flat-square)](https://godoc.org/github.com/etcd-io/bbolt)
+[![Releases](https://img.shields.io/github/release/etcd-io/bbolt/all.svg?style=flat-square)](https://github.com/etcd-io/bbolt/releases)
+[![LICENSE](https://img.shields.io/github/license/etcd-io/bbolt.svg?style=flat-square)](https://github.com/etcd-io/bbolt/blob/master/LICENSE)
+
+bbolt is a fork of [Ben Johnson's][gh_ben] [Bolt][bolt] key/value
+store. The purpose of this fork is to provide the Go community with an active
+maintenance and development target for Bolt; the goal is improved reliability
+and stability. bbolt includes bug fixes, performance enhancements, and features
+not found in Bolt while preserving backwards compatibility with the Bolt API.
 
 Bolt is a pure Go key/value store inspired by [Howard Chu's][hyc_symas]
 [LMDB project][lmdb]. The goal of the project is to provide a simple,
@@ -10,6 +23,8 @@ Since Bolt is meant to be used as such a low-level piece of functionality,
 simplicity is key. The API will be small and only focus on getting values
 and setting values. That's it.
 
+[gh_ben]: https://github.com/benbjohnson
+[bolt]: https://github.com/boltdb/bolt
 [hyc_symas]: https://twitter.com/hyc_symas
 [lmdb]: http://symas.com/mdb/
 
@@ -21,9 +36,15 @@ consistency and thread safety. Bolt is currently used in high-load production
 environments serving databases as large as 1TB. Many companies such as
 Shopify and Heroku use Bolt-backed services every day.
 
+## Project versioning
+
+bbolt uses [semantic versioning](http://semver.org).
+API should not change between patch and minor releases.
+New minor versions may add additional features to the API.
+
 ## Table of Contents
 
-- [Getting Started](#getting-started)
+  - [Getting Started](#getting-started)
     - [Installing](#installing)
     - [Opening a database](#opening-a-database)
     - [Transactions](#transactions)
@@ -43,14 +64,14 @@ Shopify and Heroku use Bolt-backed services every day.
     - [Statistics](#statistics)
     - [Read-Only Mode](#read-only-mode)
     - [Mobile Use (iOS/Android)](#mobile-use-iosandroid)
-- [Resources](#resources)
+  - [Resources](#resources)
-- [Comparison with other databases](#comparison-with-other-databases)
+  - [Comparison with other databases](#comparison-with-other-databases)
     - [Postgres, MySQL, & other relational databases](#postgres-mysql--other-relational-databases)
     - [LevelDB, RocksDB](#leveldb-rocksdb)
     - [LMDB](#lmdb)
-- [Caveats & Limitations](#caveats--limitations)
+  - [Caveats & Limitations](#caveats--limitations)
-- [Reading the Source](#reading-the-source)
+  - [Reading the Source](#reading-the-source)
-- [Other Projects Using Bolt](#other-projects-using-bolt)
+  - [Other Projects Using Bolt](#other-projects-using-bolt)
 
 ## Getting Started
 
@@ -59,13 +80,28 @@ Shopify and Heroku use Bolt-backed services every day.
 To start using Bolt, install Go and run `go get`:
 
 ```sh
-$ go get github.com/boltdb/bolt/...
+$ go get go.etcd.io/bbolt/...
 ```
 
 This will retrieve the library and install the `bolt` command line utility into
 your `$GOBIN` path.
 
 
+### Importing bbolt
+
+To use bbolt as an embedded key-value store, import as:
+
+```go
+import bolt "go.etcd.io/bbolt"
+
+db, err := bolt.Open(path, 0666, nil)
+if err != nil {
+  return err
+}
+defer db.Close()
+```
+
+
 ### Opening a database
 
 The top-level object in Bolt is a `DB`. It is represented as a single file on
@@ -79,7 +115,7 @@ package main
 import (
 	"log"
 
-	"github.com/boltdb/bolt"
+	bolt "go.etcd.io/bbolt"
 )
 
 func main() {
@@ -522,7 +558,7 @@ this from a read-only transaction, it will perform a hot backup and not block
 your other database reads and writes.
 
 By default, it will use a regular file handle which will utilize the operating
-system's page cache. See the [`Tx`](https://godoc.org/github.com/boltdb/bolt#Tx)
+system's page cache. See the [`Tx`](https://godoc.org/go.etcd.io/bbolt#Tx)
 documentation for information about optimizing for larger-than-RAM datasets.
 
 One common use case is to backup over HTTP so you can use tools like `cURL` to
@@ -811,7 +847,7 @@ Here are a few things to note when evaluating and using Bolt:
 
 ## Reading the Source
 
-Bolt is a relatively small code base (<3KLOC) for an embedded, serializable,
+Bolt is a relatively small code base (<5KLOC) for an embedded, serializable,
 transactional key/value database so it can be a good starting point for people
 interested in how databases work.
 
@@ -863,54 +899,56 @@ them via pull request.
 
 Below is a list of public, open source projects that use Bolt:
 
-* [BoltDbWeb](https://github.com/evnix/boltdbweb) - A web based GUI for BoltDB files.
+* [Algernon](https://github.com/xyproto/algernon) - A HTTP/2 web server with built-in support for Lua. Uses BoltDB as the default database backend.
-* [Operation Go: A Routine Mission](http://gocode.io) - An online programming game for Golang using Bolt for user accounts and a leaderboard.
 * [Bazil](https://bazil.org/) - A file system that lets your data reside where it is most convenient for it to reside.
-* [DVID](https://github.com/janelia-flyem/dvid) - Added Bolt as optional storage engine and testing it against Basho-tuned leveldb.
+* [bolter](https://github.com/hasit/bolter) - Command-line app for viewing BoltDB file in your terminal.
-* [Skybox Analytics](https://github.com/skybox/skybox) - A standalone funnel analysis tool for web analytics.
+* [boltcli](https://github.com/spacewander/boltcli) - the redis-cli for boltdb with Lua script support.
-* [Scuttlebutt](https://github.com/benbjohnson/scuttlebutt) - Uses Bolt to store and process all Twitter mentions of GitHub projects.
+* [BoltHold](https://github.com/timshannon/bolthold) - An embeddable NoSQL store for Go types built on BoltDB
-* [Wiki](https://github.com/peterhellberg/wiki) - A tiny wiki using Goji, BoltDB and Blackfriday.
-* [ChainStore](https://github.com/pressly/chainstore) - Simple key-value interface to a variety of storage engines organized as a chain of operations.
-* [MetricBase](https://github.com/msiebuhr/MetricBase) - Single-binary version of Graphite.
-* [Gitchain](https://github.com/gitchain/gitchain) - Decentralized, peer-to-peer Git repositories aka "Git meets Bitcoin".
-* [event-shuttle](https://github.com/sclasen/event-shuttle) - A Unix system service to collect and reliably deliver messages to Kafka.
-* [ipxed](https://github.com/kelseyhightower/ipxed) - Web interface and api for ipxed.
 * [BoltStore](https://github.com/yosssi/boltstore) - Session store using Bolt.
-* [photosite/session](https://godoc.org/bitbucket.org/kardianos/photosite/session) - Sessions for a photo viewing site.
+* [Boltdb Boilerplate](https://github.com/bobintornado/boltdb-boilerplate) - Boilerplate wrapper around bolt aiming to make simple calls one-liners.
-* [LedisDB](https://github.com/siddontang/ledisdb) - A high performance NoSQL, using Bolt as optional storage.
+* [BoltDbWeb](https://github.com/evnix/boltdbweb) - A web based GUI for BoltDB files.
-* [ipLocator](https://github.com/AndreasBriese/ipLocator) - A fast ip-geo-location-server using bolt with bloom filters.
-* [cayley](https://github.com/google/cayley) - Cayley is an open-source graph database using Bolt as optional backend.
 * [bleve](http://www.blevesearch.com/) - A pure Go search engine similar to ElasticSearch that uses Bolt as the default storage backend.
-* [tentacool](https://github.com/optiflows/tentacool) - REST api server to manage system stuff (IP, DNS, Gateway...) on a linux server.
+* [btcwallet](https://github.com/btcsuite/btcwallet) - A bitcoin wallet.
-* [Seaweed File System](https://github.com/chrislusf/seaweedfs) - Highly scalable distributed key~file system with O(1) disk read.
-* [InfluxDB](https://influxdata.com) - Scalable datastore for metrics, events, and real-time analytics.
-* [Freehold](http://tshannon.bitbucket.org/freehold/) - An open, secure, and lightweight platform for your files and data.
-* [Prometheus Annotation Server](https://github.com/oliver006/prom_annotation_server) - Annotation server for PromDash & Prometheus service monitoring system.
-* [Consul](https://github.com/hashicorp/consul) - Consul is service discovery and configuration made easy. Distributed, highly available, and datacenter-aware.
-* [Kala](https://github.com/ajvb/kala) - Kala is a modern job scheduler optimized to run on a single node. It is persistent, JSON over HTTP API, ISO 8601 duration notation, and dependent jobs.
-* [drive](https://github.com/odeke-em/drive) - drive is an unofficial Google Drive command line client for \*NIX operating systems.
-* [stow](https://github.com/djherbis/stow) -  a persistence manager for objects
-  backed by boltdb.
 * [buckets](https://github.com/joyrexus/buckets) - a bolt wrapper streamlining
   simple tx and key scans.
-* [mbuckets](https://github.com/abhigupta912/mbuckets) - A Bolt wrapper that allows easy operations on multi level (nested) buckets.
+* [cayley](https://github.com/google/cayley) - Cayley is an open-source graph database using Bolt as optional backend.
-* [Request Baskets](https://github.com/darklynx/request-baskets) - A web service to collect arbitrary HTTP requests and inspect them via REST API or simple web UI, similar to [RequestBin](http://requestb.in/) service
+* [ChainStore](https://github.com/pressly/chainstore) - Simple key-value interface to a variety of storage engines organized as a chain of operations.
-* [Go Report Card](https://goreportcard.com/) - Go code quality report cards as a (free and open source) service.
+* [Consul](https://github.com/hashicorp/consul) - Consul is service discovery and configuration made easy. Distributed, highly available, and datacenter-aware.
-* [Boltdb Boilerplate](https://github.com/bobintornado/boltdb-boilerplate) - Boilerplate wrapper around bolt aiming to make simple calls one-liners.
+* [DVID](https://github.com/janelia-flyem/dvid) - Added Bolt as optional storage engine and testing it against Basho-tuned leveldb.
-* [lru](https://github.com/crowdriff/lru) - Easy to use Bolt-backed Least-Recently-Used (LRU) read-through cache with chainable remote stores.
-* [Storm](https://github.com/asdine/storm) - Simple and powerful ORM for BoltDB.
-* [GoWebApp](https://github.com/josephspurrier/gowebapp) - A basic MVC web application in Go using BoltDB.
-* [SimpleBolt](https://github.com/xyproto/simplebolt) - A simple way to use BoltDB. Deals mainly with strings.
-* [Algernon](https://github.com/xyproto/algernon) - A HTTP/2 web server with built-in support for Lua. Uses BoltDB as the default database backend.
-* [MuLiFS](https://github.com/dankomiocevic/mulifs) - Music Library Filesystem creates a filesystem to organise your music files.
-* [GoShort](https://github.com/pankajkhairnar/goShort) - GoShort is a URL shortener written in Golang and BoltDB for persistent key/value storage and for routing it's using high performent HTTPRouter.
-* [torrent](https://github.com/anacrolix/torrent) - Full-featured BitTorrent client package and utilities in Go. BoltDB is a storage backend in development.
-* [gopherpit](https://github.com/gopherpit/gopherpit) - A web service to manage Go remote import paths with custom domains
-* [bolter](https://github.com/hasit/bolter) - Command-line app for viewing BoltDB file in your terminal.
-* [btcwallet](https://github.com/btcsuite/btcwallet) - A bitcoin wallet.
 * [dcrwallet](https://github.com/decred/dcrwallet) - A wallet for the Decred cryptocurrency.
+* [drive](https://github.com/odeke-em/drive) - drive is an unofficial Google Drive command line client for \*NIX operating systems.
+* [event-shuttle](https://github.com/sclasen/event-shuttle) - A Unix system service to collect and reliably deliver messages to Kafka.
+* [Freehold](http://tshannon.bitbucket.org/freehold/) - An open, secure, and lightweight platform for your files and data.
+* [Go Report Card](https://goreportcard.com/) - Go code quality report cards as a (free and open source) service.
+* [GoWebApp](https://github.com/josephspurrier/gowebapp) - A basic MVC web application in Go using BoltDB.
+* [GoShort](https://github.com/pankajkhairnar/goShort) - GoShort is a URL shortener written in Golang and BoltDB for persistent key/value storage and for routing it's using high performent HTTPRouter.
+* [gopherpit](https://github.com/gopherpit/gopherpit) - A web service to manage Go remote import paths with custom domains
+* [Gitchain](https://github.com/gitchain/gitchain) - Decentralized, peer-to-peer Git repositories aka "Git meets Bitcoin".
+* [InfluxDB](https://influxdata.com) - Scalable datastore for metrics, events, and real-time analytics.
+* [ipLocator](https://github.com/AndreasBriese/ipLocator) - A fast ip-geo-location-server using bolt with bloom filters.
+* [ipxed](https://github.com/kelseyhightower/ipxed) - Web interface and api for ipxed.
 * [Ironsmith](https://github.com/timshannon/ironsmith) - A simple, script-driven continuous integration (build - > test -> release) tool, with no external dependencies
-* [BoltHold](https://github.com/timshannon/bolthold) - An embeddable NoSQL store for Go types built on BoltDB
+* [Kala](https://github.com/ajvb/kala) - Kala is a modern job scheduler optimized to run on a single node. It is persistent, JSON over HTTP API, ISO 8601 duration notation, and dependent jobs.
-* [Ponzu CMS](https://ponzu-cms.org) - Headless CMS + automatic JSON API with auto-HTTPS, HTTP/2 Server Push, and flexible server framework.
+* [Key Value Access Langusge (KVAL)](https://github.com/kval-access-language) - A proposed grammar for key-value datastores offering a bbolt binding.
+* [LedisDB](https://github.com/siddontang/ledisdb) - A high performance NoSQL, using Bolt as optional storage.
+* [lru](https://github.com/crowdriff/lru) - Easy to use Bolt-backed Least-Recently-Used (LRU) read-through cache with chainable remote stores.
+* [mbuckets](https://github.com/abhigupta912/mbuckets) - A Bolt wrapper that allows easy operations on multi level (nested) buckets.
+* [MetricBase](https://github.com/msiebuhr/MetricBase) - Single-binary version of Graphite.
+* [MuLiFS](https://github.com/dankomiocevic/mulifs) - Music Library Filesystem creates a filesystem to organise your music files.
+* [Operation Go: A Routine Mission](http://gocode.io) - An online programming game for Golang using Bolt for user accounts and a leaderboard.
+* [photosite/session](https://godoc.org/bitbucket.org/kardianos/photosite/session) - Sessions for a photo viewing site.
+* [Prometheus Annotation Server](https://github.com/oliver006/prom_annotation_server) - Annotation server for PromDash & Prometheus service monitoring system.
+* [reef-pi](https://github.com/reef-pi/reef-pi) - reef-pi is an award winning, modular, DIY reef tank controller using easy to learn electronics based on a Raspberry Pi.
+* [Request Baskets](https://github.com/darklynx/request-baskets) - A web service to collect arbitrary HTTP requests and inspect them via REST API or simple web UI, similar to [RequestBin](http://requestb.in/) service
+* [Seaweed File System](https://github.com/chrislusf/seaweedfs) - Highly scalable distributed key~file system with O(1) disk read.
+* [stow](https://github.com/djherbis/stow) -  a persistence manager for objects
+  backed by boltdb.
+* [Storm](https://github.com/asdine/storm) - Simple and powerful ORM for BoltDB.
+* [SimpleBolt](https://github.com/xyproto/simplebolt) - A simple way to use BoltDB. Deals mainly with strings.
+* [Skybox Analytics](https://github.com/skybox/skybox) - A standalone funnel analysis tool for web analytics.
+* [Scuttlebutt](https://github.com/benbjohnson/scuttlebutt) - Uses Bolt to store and process all Twitter mentions of GitHub projects.
+* [tentacool](https://github.com/optiflows/tentacool) - REST api server to manage system stuff (IP, DNS, Gateway...) on a linux server.
+* [torrent](https://github.com/anacrolix/torrent) - Full-featured BitTorrent client package and utilities in Go. BoltDB is a storage backend in development.
+* [Wiki](https://github.com/peterhellberg/wiki) - A tiny wiki using Goji, BoltDB and Blackfriday.
 
 If you are using Bolt in a project please send a pull request to add it to the list.

vendor/github.com/etcd-io/bbolt/bolt_386.go

@@ -1,4 +1,4 @@
-package bolt
+package bbolt
 
 // maxMapSize represents the largest mmap size supported by Bolt.
 const maxMapSize = 0x7FFFFFFF // 2GB

vendor/github.com/etcd-io/bbolt/bolt_amd64.go

@@ -1,4 +1,4 @@
-package bolt
+package bbolt
 
 // maxMapSize represents the largest mmap size supported by Bolt.
 const maxMapSize = 0xFFFFFFFFFFFF // 256TB

vendor/github.com/etcd-io/bbolt/bolt_arm.go

@@ -1,4 +1,4 @@
-package bolt
+package bbolt
 
 import "unsafe"
 

vendor/github.com/etcd-io/bbolt/bolt_arm64.go

@@ -1,6 +1,6 @@
 // +build arm64
 
-package bolt
+package bbolt
 
 // maxMapSize represents the largest mmap size supported by Bolt.
 const maxMapSize = 0xFFFFFFFFFFFF // 256TB

vendor/github.com/etcd-io/bbolt/bolt_linux.go

@@ -1,4 +1,4 @@
-package bolt
+package bbolt
 
 import (
 	"syscall"

vendor/github.com/etcd-io/bbolt/bolt_mips64x.go

@@ -0,0 +1,12 @@
+// +build mips64 mips64le
+
+package bbolt
+
+// maxMapSize represents the largest mmap size supported by Bolt.
+const maxMapSize = 0x8000000000 // 512GB
+
+// maxAllocSize is the size used when creating array pointers.
+const maxAllocSize = 0x7FFFFFFF
+
+// Are unaligned load/stores broken on this arch?
+var brokenUnaligned = false

vendor/github.com/etcd-io/bbolt/bolt_mipsx.go

@@ -0,0 +1,12 @@
+// +build mips mipsle
+
+package bbolt
+
+// maxMapSize represents the largest mmap size supported by Bolt.
+const maxMapSize = 0x40000000 // 1GB
+
+// maxAllocSize is the size used when creating array pointers.
+const maxAllocSize = 0xFFFFFFF
+
+// Are unaligned load/stores broken on this arch?
+var brokenUnaligned = false

vendor/github.com/etcd-io/bbolt/bolt_openbsd.go

@@ -1,4 +1,4 @@
-package bolt
+package bbolt
 
 import (
 	"syscall"

vendor/github.com/etcd-io/bbolt/bolt_ppc.go

@@ -1,9 +1,12 @@
 // +build ppc
 
-package bolt
+package bbolt
 
 // maxMapSize represents the largest mmap size supported by Bolt.
 const maxMapSize = 0x7FFFFFFF // 2GB
 
 // maxAllocSize is the size used when creating array pointers.
 const maxAllocSize = 0xFFFFFFF
+
+// Are unaligned load/stores broken on this arch?
+var brokenUnaligned = false

vendor/github.com/etcd-io/bbolt/bolt_ppc64.go

@@ -1,6 +1,6 @@
 // +build ppc64
 
-package bolt
+package bbolt
 
 // maxMapSize represents the largest mmap size supported by Bolt.
 const maxMapSize = 0xFFFFFFFFFFFF // 256TB

vendor/github.com/etcd-io/bbolt/bolt_ppc64le.go

@@ -1,6 +1,6 @@
 // +build ppc64le
 
-package bolt
+package bbolt
 
 // maxMapSize represents the largest mmap size supported by Bolt.
 const maxMapSize = 0xFFFFFFFFFFFF // 256TB

vendor/github.com/etcd-io/bbolt/bolt_s390x.go

@@ -1,6 +1,6 @@
 // +build s390x
 
-package bolt
+package bbolt
 
 // maxMapSize represents the largest mmap size supported by Bolt.
 const maxMapSize = 0xFFFFFFFFFFFF // 256TB

vendor/github.com/etcd-io/bbolt/bolt_unix.go

@@ -1,41 +1,43 @@
 // +build !windows,!plan9,!solaris
 
-package bolt
+package bbolt
 
 import (
 	"fmt"
-	"os"
 	"syscall"
 	"time"
 	"unsafe"
 )
 
 // flock acquires an advisory lock on a file descriptor.
-func flock(db *DB, mode os.FileMode, exclusive bool, timeout time.Duration) error {
+func flock(db *DB, exclusive bool, timeout time.Duration) error {
 	var t time.Time
-	for {
+	if timeout != 0 {
-		// If we're beyond our timeout then return an error.
-		// This can only occur after we've attempted a flock once.
-		if t.IsZero() {
 		t = time.Now()
-		} else if timeout > 0 && time.Since(t) > timeout {
-			return ErrTimeout
 	}
-		flag := syscall.LOCK_SH
+	fd := db.file.Fd()
+	flag := syscall.LOCK_NB
 	if exclusive {
-			flag = syscall.LOCK_EX
+		flag |= syscall.LOCK_EX
+	} else {
+		flag |= syscall.LOCK_SH
 	}
-
+	for {
-		// Otherwise attempt to obtain an exclusive lock.
+		// Attempt to obtain an exclusive lock.
-		err := syscall.Flock(int(db.file.Fd()), flag|syscall.LOCK_NB)
+		err := syscall.Flock(int(fd), flag)
 		if err == nil {
 			return nil
 		} else if err != syscall.EWOULDBLOCK {
 			return err
 		}
 
+		// If we timed out then return an error.
+		if timeout != 0 && time.Since(t) > timeout-flockRetryTimeout {
+			return ErrTimeout
+		}
+
 		// Wait for a bit and try again.
-		time.Sleep(50 * time.Millisecond)
+		time.Sleep(flockRetryTimeout)
 	}
 }
 
@@ -53,7 +55,9 @@ func mmap(db *DB, sz int) error {
 	}
 
 	// Advise the kernel that the mmap is accessed randomly.
-	if err := madvise(b, syscall.MADV_RANDOM); err != nil {
+	err = madvise(b, syscall.MADV_RANDOM)
+	if err != nil && err != syscall.ENOSYS {
+		// Ignore not implemented error in kernel because it still works.
 		return fmt.Errorf("madvise: %s", err)
 	}
 

vendor/github.com/etcd-io/bbolt/bolt_unix_solaris.go

@@ -1,8 +1,7 @@
-package bolt
+package bbolt
 
 import (
 	"fmt"
-	"os"
 	"syscall"
 	"time"
 	"unsafe"
@@ -11,36 +10,35 @@ import (
 )
 
 // flock acquires an advisory lock on a file descriptor.
-func flock(db *DB, mode os.FileMode, exclusive bool, timeout time.Duration) error {
+func flock(db *DB, exclusive bool, timeout time.Duration) error {
 	var t time.Time
-	for {
+	if timeout != 0 {
-		// If we're beyond our timeout then return an error.
-		// This can only occur after we've attempted a flock once.
-		if t.IsZero() {
 		t = time.Now()
-		} else if timeout > 0 && time.Since(t) > timeout {
-			return ErrTimeout
 	}
-		var lock syscall.Flock_t
+	fd := db.file.Fd()
-		lock.Start = 0
+	var lockType int16
-		lock.Len = 0
-		lock.Pid = 0
-		lock.Whence = 0
-		lock.Pid = 0
 	if exclusive {
-			lock.Type = syscall.F_WRLCK
+		lockType = syscall.F_WRLCK
 	} else {
-			lock.Type = syscall.F_RDLCK
+		lockType = syscall.F_RDLCK
 	}
-		err := syscall.FcntlFlock(db.file.Fd(), syscall.F_SETLK, &lock)
+	for {
+		// Attempt to obtain an exclusive lock.
+		lock := syscall.Flock_t{Type: lockType}
+		err := syscall.FcntlFlock(fd, syscall.F_SETLK, &lock)
 		if err == nil {
 			return nil
 		} else if err != syscall.EAGAIN {
 			return err
 		}
 
+		// If we timed out then return an error.
+		if timeout != 0 && time.Since(t) > timeout-flockRetryTimeout {
+			return ErrTimeout
+		}
+
 		// Wait for a bit and try again.
-		time.Sleep(50 * time.Millisecond)
+		time.Sleep(flockRetryTimeout)
 	}
 }
 

vendor/github.com/etcd-io/bbolt/bolt_windows.go

@@ -1,4 +1,4 @@
-package bolt
+package bbolt
 
 import (
 	"fmt"
@@ -16,8 +16,6 @@ var (
 )
 
 const (
-	lockExt = ".lock"
-
 	// see https://msdn.microsoft.com/en-us/library/windows/desktop/aa365203(v=vs.85).aspx
 	flagLockExclusive       = 2
 	flagLockFailImmediately = 1
@@ -48,48 +46,47 @@ func fdatasync(db *DB) error {
 }
 
 // flock acquires an advisory lock on a file descriptor.
-func flock(db *DB, mode os.FileMode, exclusive bool, timeout time.Duration) error {
+func flock(db *DB, exclusive bool, timeout time.Duration) error {
-	// Create a separate lock file on windows because a process
-	// cannot share an exclusive lock on the same file. This is
-	// needed during Tx.WriteTo().
-	f, err := os.OpenFile(db.path+lockExt, os.O_CREATE, mode)
-	if err != nil {
-		return err
-	}
-	db.lockfile = f
-
 	var t time.Time
-	for {
+	if timeout != 0 {
-		// If we're beyond our timeout then return an error.
-		// This can only occur after we've attempted a flock once.
-		if t.IsZero() {
 		t = time.Now()
-		} else if timeout > 0 && time.Since(t) > timeout {
-			return ErrTimeout
 	}
-
 	var flag uint32 = flagLockFailImmediately
 	if exclusive {
 		flag |= flagLockExclusive
 	}
+	for {
+		// Fix for https://github.com/etcd-io/bbolt/issues/121. Use byte-range
+		// -1..0 as the lock on the database file.
+		var m1 uint32 = (1 << 32) - 1 // -1 in a uint32
+		err := lockFileEx(syscall.Handle(db.file.Fd()), flag, 0, 1, 0, &syscall.Overlapped{
+			Offset:     m1,
+			OffsetHigh: m1,
+		})
 
-		err := lockFileEx(syscall.Handle(db.lockfile.Fd()), flag, 0, 1, 0, &syscall.Overlapped{})
 		if err == nil {
 			return nil
 		} else if err != errLockViolation {
 			return err
 		}
 
+		// If we timed oumercit then return an error.
+		if timeout != 0 && time.Since(t) > timeout-flockRetryTimeout {
+			return ErrTimeout
+		}
+
 		// Wait for a bit and try again.
-		time.Sleep(50 * time.Millisecond)
+		time.Sleep(flockRetryTimeout)
 	}
 }
 
 // funlock releases an advisory lock on a file descriptor.
 func funlock(db *DB) error {
-	err := unlockFileEx(syscall.Handle(db.lockfile.Fd()), 0, 1, 0, &syscall.Overlapped{})
+	var m1 uint32 = (1 << 32) - 1 // -1 in a uint32
-	db.lockfile.Close()
+	err := unlockFileEx(syscall.Handle(db.file.Fd()), 0, 1, 0, &syscall.Overlapped{
-	os.Remove(db.path + lockExt)
+		Offset:     m1,
+		OffsetHigh: m1,
+	})
 	return err
 }
 

vendor/github.com/etcd-io/bbolt/boltsync_unix.go

@@ -1,6 +1,6 @@
 // +build !windows,!plan9,!linux,!openbsd
 
-package bolt
+package bbolt
 
 // fdatasync flushes written data to a file descriptor.
 func fdatasync(db *DB) error {

vendor/github.com/etcd-io/bbolt/bucket.go

@@ -1,4 +1,4 @@
-package bolt
+package bbolt
 
 import (
 	"bytes"
@@ -14,13 +14,6 @@ const (
 	MaxValueSize = (1 << 31) - 2
 )
 
-const (
-	maxUint = ^uint(0)
-	minUint = 0
-	maxInt  = int(^uint(0) >> 1)
-	minInt  = -maxInt - 1
-)
-
 const bucketHeaderSize = int(unsafe.Sizeof(bucket{}))
 
 const (
@@ -323,7 +316,12 @@ func (b *Bucket) Delete(key []byte) error {
 
 	// Move cursor to correct position.
 	c := b.Cursor()
-	_, _, flags := c.seek(key)
+	k, _, flags := c.seek(key)
+
+	// Return nil if the key doesn't exist.
+	if !bytes.Equal(key, k) {
+		return nil
+	}
 
 	// Return an error if there is already existing bucket value.
 	if (flags & bucketLeafFlag) != 0 {

vendor/github.com/etcd-io/bbolt/cursor.go

@@ -1,4 +1,4 @@
-package bolt
+package bbolt
 
 import (
 	"bytes"
@@ -157,12 +157,6 @@ func (c *Cursor) seek(seek []byte) (key []byte, value []byte, flags uint32) {
 	// Start from root page/node and traverse to correct page.
 	c.stack = c.stack[:0]
 	c.search(seek, c.bucket.root)
-	ref := &c.stack[len(c.stack)-1]
-
-	// If the cursor is pointing to the end of page/node then return nil.
-	if ref.index >= ref.count() {
-		return nil, nil, 0
-	}
 
 	// If this is a bucket then return a nil value.
 	return c.keyValue()
@@ -339,6 +333,8 @@ func (c *Cursor) nsearch(key []byte) {
 // keyValue returns the key and value of the current leaf element.
 func (c *Cursor) keyValue() ([]byte, []byte, uint32) {
 	ref := &c.stack[len(c.stack)-1]
+
+	// If the cursor is pointing to the end of page/node then return nil.
 	if ref.count() == 0 || ref.index >= ref.count() {
 		return nil, nil, 0
 	}

vendor/github.com/etcd-io/bbolt/db.go

@@ -1,4 +1,4 @@
-package bolt
+package bbolt
 
 import (
 	"errors"
@@ -7,8 +7,7 @@ import (
 	"log"
 	"os"
 	"runtime"
-	"runtime/debug"
+	"sort"
-	"strings"
 	"sync"
 	"time"
 	"unsafe"
@@ -23,6 +22,8 @@ const version = 2
 // Represents a marker value to indicate that a file is a Bolt DB.
 const magic uint32 = 0xED0CDAED
 
+const pgidNoFreelist pgid = 0xffffffffffffffff
+
 // IgnoreNoSync specifies whether the NoSync field of a DB is ignored when
 // syncing changes to a file.  This is required as some operating systems,
 // such as OpenBSD, do not have a unified buffer cache (UBC) and writes
@@ -39,6 +40,19 @@ const (
 // default page size for db is set to the OS page size.
 var defaultPageSize = os.Getpagesize()
 
+// The time elapsed between consecutive file locking attempts.
+const flockRetryTimeout = 50 * time.Millisecond
+
+// FreelistType is the type of the freelist backend
+type FreelistType string
+
+const (
+	// FreelistArrayType indicates backend freelist type is array
+	FreelistArrayType = FreelistType("array")
+	// FreelistMapType indicates backend freelist type is hashmap
+	FreelistMapType = FreelistType("hashmap")
+)
+
 // DB represents a collection of buckets persisted to a file on disk.
 // All data access is performed through transactions which can be obtained through the DB.
 // All the functions on DB will return a ErrDatabaseNotOpen if accessed before Open() is called.
@@ -61,6 +75,18 @@ type DB struct {
 	// THIS IS UNSAFE. PLEASE USE WITH CAUTION.
 	NoSync bool
 
+	// When true, skips syncing freelist to disk. This improves the database
+	// write performance under normal operation, but requires a full database
+	// re-sync during recovery.
+	NoFreelistSync bool
+
+	// FreelistType sets the backend freelist type. There are two options. Array which is simple but endures
+	// dramatic performance degradation if database is large and framentation in freelist is common.
+	// The alternative one is using hashmap, it is faster in almost all circumstances
+	// but it doesn't guarantee that it offers the smallest page id available. In normal case it is safe.
+	// The default type is array
+	FreelistType FreelistType
+
 	// When true, skips the truncate call when growing the database.
 	// Setting this to true is only safe on non-ext3/ext4 systems.
 	// Skipping truncation avoids preallocation of hard drive space and
@@ -96,7 +122,6 @@ type DB struct {
 
 	path     string
 	file     *os.File
-	lockfile *os.File // windows only
 	dataref  []byte // mmap'ed readonly, write throws SEGV
 	data     *[maxMapSize]byte
 	datasz   int
@@ -107,9 +132,11 @@ type DB struct {
 	opened   bool
 	rwtx     *Tx
 	txs      []*Tx
-	freelist *freelist
 	stats    Stats
 
+	freelist     *freelist
+	freelistLoad sync.Once
+
 	pagePool sync.Pool
 
 	batchMu sync.Mutex
@@ -148,14 +175,18 @@ func (db *DB) String() string {
 // If the file does not exist then it will be created automatically.
 // Passing in nil options will cause Bolt to open the database with the default options.
 func Open(path string, mode os.FileMode, options *Options) (*DB, error) {
-	var db = &DB{opened: true}
+	db := &DB{
-
+		opened: true,
+	}
 	// Set default options if no options are provided.
 	if options == nil {
 		options = DefaultOptions
 	}
+	db.NoSync = options.NoSync
 	db.NoGrowSync = options.NoGrowSync
 	db.MmapFlags = options.MmapFlags
+	db.NoFreelistSync = options.NoFreelistSync
+	db.FreelistType = options.FreelistType
 
 	// Set default values for later DB operations.
 	db.MaxBatchSize = DefaultMaxBatchSize
@@ -183,7 +214,7 @@ func Open(path string, mode os.FileMode, options *Options) (*DB, error) {
 	// if !options.ReadOnly.
 	// The database file is locked using the shared lock (more than one process may
 	// hold a lock at the same time) otherwise (options.ReadOnly is set).
-	if err := flock(db, mode, !db.readOnly, options.Timeout); err != nil {
+	if err := flock(db, !db.readOnly, options.Timeout); err != nil {
 		_ = db.close()
 		return nil, err
 	}
@@ -191,31 +222,41 @@ func Open(path string, mode os.FileMode, options *Options) (*DB, error) {
 	// Default values for test hooks
 	db.ops.writeAt = db.file.WriteAt
 
+	if db.pageSize = options.PageSize; db.pageSize == 0 {
+		// Set the default page size to the OS page size.
+		db.pageSize = defaultPageSize
+	}
+
 	// Initialize the database if it doesn't exist.
 	if info, err := db.file.Stat(); err != nil {
+		_ = db.close()
 		return nil, err
 	} else if info.Size() == 0 {
 		// Initialize new files with meta pages.
 		if err := db.init(); err != nil {
+			// clean up file descriptor on initialization fail
+			_ = db.close()
 			return nil, err
 		}
 	} else {
 		// Read the first meta page to determine the page size.
 		var buf [0x1000]byte
-		if _, err := db.file.ReadAt(buf[:], 0); err == nil {
+		// If we can't read the page size, but can read a page, assume
-			m := db.pageInBuffer(buf[:], 0).meta()
+		// it's the same as the OS or one given -- since that's how the
-			if err := m.validate(); err != nil {
+		// page size was chosen in the first place.
-				// If we can't read the page size, we can assume it's the same
-				// as the OS -- since that's how the page size was chosen in the
-				// first place.
 		//
 		// If the first page is invalid and this OS uses a different
 		// page size than what the database was created with then we
 		// are out of luck and cannot access the database.
-				db.pageSize = os.Getpagesize()
+		//
-			} else {
+		// TODO: scan for next page
+		if bw, err := db.file.ReadAt(buf[:], 0); err == nil && bw == len(buf) {
+			if m := db.pageInBuffer(buf[:], 0).meta(); m.validate() == nil {
 				db.pageSize = int(m.pageSize)
 			}
+		} else {
+			_ = db.close()
+			return nil, ErrInvalid
 		}
 	}
 
@@ -232,14 +273,50 @@ func Open(path string, mode os.FileMode, options *Options) (*DB, error) {
 		return nil, err
 	}
 
-	// Read in the freelist.
+	if db.readOnly {
-	db.freelist = newFreelist()
+		return db, nil
-	db.freelist.read(db.page(db.meta().freelist))
+	}
+
+	db.loadFreelist()
+
+	// Flush freelist when transitioning from no sync to sync so
+	// NoFreelistSync unaware boltdb can open the db later.
+	if !db.NoFreelistSync && !db.hasSyncedFreelist() {
+		tx, err := db.Begin(true)
+		if tx != nil {
+			err = tx.Commit()
+		}
+		if err != nil {
+			_ = db.close()
+			return nil, err
+		}
+	}
 
 	// Mark the database as opened and return.
 	return db, nil
 }
 
+// loadFreelist reads the freelist if it is synced, or reconstructs it
+// by scanning the DB if it is not synced. It assumes there are no
+// concurrent accesses being made to the freelist.
+func (db *DB) loadFreelist() {
+	db.freelistLoad.Do(func() {
+		db.freelist = newFreelist(db.FreelistType)
+		if !db.hasSyncedFreelist() {
+			// Reconstruct free list by scanning the DB.
+			db.freelist.readIDs(db.freepages())
+		} else {
+			// Read free list from freelist page.
+			db.freelist.read(db.page(db.meta().freelist))
+		}
+		db.stats.FreePageN = db.freelist.free_count()
+	})
+}
+
+func (db *DB) hasSyncedFreelist() bool {
+	return db.meta().freelist != pgidNoFreelist
+}
+
 // mmap opens the underlying memory-mapped file and initializes the meta references.
 // minsz is the minimum size that the new mmap can be.
 func (db *DB) mmap(minsz int) error {
@@ -341,9 +418,6 @@ func (db *DB) mmapSize(size int) (int, error) {
 
 // init creates a new database file and initializes its meta pages.
 func (db *DB) init() error {
-	// Set the page size to the OS page size.
-	db.pageSize = os.Getpagesize()
-
 	// Create two meta pages on a buffer.
 	buf := make([]byte, db.pageSize*4)
 	for i := 0; i < 2; i++ {
@@ -387,7 +461,8 @@ func (db *DB) init() error {
 }
 
 // Close releases all database resources.
-// All transactions must be closed before closing the database.
+// It will block waiting for any open transactions to finish
+// before closing the database and returning.
 func (db *DB) Close() error {
 	db.rwlock.Lock()
 	defer db.rwlock.Unlock()
@@ -395,8 +470,8 @@ func (db *DB) Close() error {
 	db.metalock.Lock()
 	defer db.metalock.Unlock()
 
-	db.mmaplock.RLock()
+	db.mmaplock.Lock()
-	defer db.mmaplock.RUnlock()
+	defer db.mmaplock.Unlock()
 
 	return db.close()
 }
@@ -526,21 +601,36 @@ func (db *DB) beginRWTx() (*Tx, error) {
 	t := &Tx{writable: true}
 	t.init(db)
 	db.rwtx = t
+	db.freePages()
+	return t, nil
+}
 
-	// Free any pages associated with closed read-only transactions.
+// freePages releases any pages associated with closed read-only transactions.
-	var minid txid = 0xFFFFFFFFFFFFFFFF
+func (db *DB) freePages() {
-	for _, t := range db.txs {
+	// Free all pending pages prior to earliest open transaction.
-		if t.meta.txid < minid {
+	sort.Sort(txsById(db.txs))
-			minid = t.meta.txid
+	minid := txid(0xFFFFFFFFFFFFFFFF)
-		}
+	if len(db.txs) > 0 {
+		minid = db.txs[0].meta.txid
 	}
 	if minid > 0 {
 		db.freelist.release(minid - 1)
 	}
-
+	// Release unused txid extents.
-	return t, nil
+	for _, t := range db.txs {
+		db.freelist.releaseRange(minid, t.meta.txid-1)
+		minid = t.meta.txid + 1
+	}
+	db.freelist.releaseRange(minid, txid(0xFFFFFFFFFFFFFFFF))
+	// Any page both allocated and freed in an extent is safe to release.
 }
 
+type txsById []*Tx
+
+func (t txsById) Len() int           { return len(t) }
+func (t txsById) Swap(i, j int)      { t[i], t[j] = t[j], t[i] }
+func (t txsById) Less(i, j int) bool { return t[i].meta.txid < t[j].meta.txid }
+
 // removeTx removes a transaction from the database.
 func (db *DB) removeTx(tx *Tx) {
 	// Release the read lock on the mmap.
@@ -633,11 +723,7 @@ func (db *DB) View(fn func(*Tx) error) error {
 		return err
 	}
 
-	if err := t.Rollback(); err != nil {
+	return t.Rollback()
-		return err
-	}
-
-	return nil
 }
 
 // Batch calls fn as part of a batch. It behaves similar to Update,
@@ -737,10 +823,8 @@ retry:
 
 		// pass success, or bolt internal errors, to all callers
 		for _, c := range b.calls {
-			if c.err != nil {
 			c.err <- err
 		}
-		}
 		break retry
 	}
 }
@@ -826,7 +910,7 @@ func (db *DB) meta() *meta {
 }
 
 // allocate returns a contiguous block of memory starting at a given page.
-func (db *DB) allocate(count int) (*page, error) {
+func (db *DB) allocate(txid txid, count int) (*page, error) {
 	// Allocate a temporary buffer for the page.
 	var buf []byte
 	if count == 1 {
@@ -838,7 +922,7 @@ func (db *DB) allocate(count int) (*page, error) {
 	p.overflow = uint32(count - 1)
 
 	// Use pages from the freelist if they are available.
-	if p.id = db.freelist.allocate(count); p.id != 0 {
+	if p.id = db.freelist.allocate(txid, count); p.id != 0 {
 		return p, nil
 	}
 
@@ -893,6 +977,38 @@ func (db *DB) IsReadOnly() bool {
 	return db.readOnly
 }
 
+func (db *DB) freepages() []pgid {
+	tx, err := db.beginTx()
+	defer func() {
+		err = tx.Rollback()
+		if err != nil {
+			panic("freepages: failed to rollback tx")
+		}
+	}()
+	if err != nil {
+		panic("freepages: failed to open read only tx")
+	}
+
+	reachable := make(map[pgid]*page)
+	nofreed := make(map[pgid]bool)
+	ech := make(chan error)
+	go func() {
+		for e := range ech {
+			panic(fmt.Sprintf("freepages: failed to get all reachable pages (%v)", e))
+		}
+	}()
+	tx.checkBucket(&tx.root, reachable, nofreed, ech)
+	close(ech)
+
+	var fids []pgid
+	for i := pgid(2); i < db.meta().pgid; i++ {
+		if _, ok := reachable[i]; !ok {
+			fids = append(fids, i)
+		}
+	}
+	return fids
+}
+
 // Options represents the options that can be set when opening a database.
 type Options struct {
 	// Timeout is the amount of time to wait to obtain a file lock.
@@ -903,6 +1019,17 @@ type Options struct {
 	// Sets the DB.NoGrowSync flag before memory mapping the file.
 	NoGrowSync bool
 
+	// Do not sync freelist to disk. This improves the database write performance
+	// under normal operation, but requires a full database re-sync during recovery.
+	NoFreelistSync bool
+
+	// FreelistType sets the backend freelist type. There are two options. Array which is simple but endures
+	// dramatic performance degradation if database is large and framentation in freelist is common.
+	// The alternative one is using hashmap, it is faster in almost all circumstances
+	// but it doesn't guarantee that it offers the smallest page id available. In normal case it is safe.
+	// The default type is array
+	FreelistType FreelistType
+
 	// Open database in read-only mode. Uses flock(..., LOCK_SH |LOCK_NB) to
 	// grab a shared lock (UNIX).
 	ReadOnly bool
@@ -919,6 +1046,14 @@ type Options struct {
 	// If initialMmapSize is smaller than the previous database size,
 	// it takes no effect.
 	InitialMmapSize int
+
+	// PageSize overrides the default OS page size.
+	PageSize int
+
+	// NoSync sets the initial value of DB.NoSync. Normally this can just be
+	// set directly on the DB itself when returned from Open(), but this option
+	// is useful in APIs which expose Options but not the underlying DB.
+	NoSync bool
 }
 
 // DefaultOptions represent the options used if nil options are passed into Open().
@@ -926,6 +1061,7 @@ type Options struct {
 var DefaultOptions = &Options{
 	Timeout:      0,
 	NoGrowSync:   false,
+	FreelistType: FreelistArrayType,
 }
 
 // Stats represents statistics about the database.
@@ -960,10 +1096,6 @@ func (s *Stats) Sub(other *Stats) Stats {
 	return diff
 }
 
-func (s *Stats) add(other *Stats) {
-	s.TxStats.add(&other.TxStats)
-}
-
 type Info struct {
 	Data     uintptr
 	PageSize int
@@ -1002,7 +1134,8 @@ func (m *meta) copy(dest *meta) {
 func (m *meta) write(p *page) {
 	if m.root.root >= m.pgid {
 		panic(fmt.Sprintf("root bucket pgid (%d) above high water mark (%d)", m.root.root, m.pgid))
-	} else if m.freelist >= m.pgid {
+	} else if m.freelist >= m.pgid && m.freelist != pgidNoFreelist {
+		// TODO: reject pgidNoFreeList if !NoFreelistSync
 		panic(fmt.Sprintf("freelist pgid (%d) above high water mark (%d)", m.freelist, m.pgid))
 	}
 
@@ -1029,11 +1162,3 @@ func _assert(condition bool, msg string, v ...interface{}) {
 		panic(fmt.Sprintf("assertion failed: "+msg, v...))
 	}
 }
-
-func warn(v ...interface{})              { fmt.Fprintln(os.Stderr, v...) }
-func warnf(msg string, v ...interface{}) { fmt.Fprintf(os.Stderr, msg+"\n", v...) }
-
-func printstack() {
-	stack := strings.Join(strings.Split(string(debug.Stack()), "\n")[2:], "\n")
-	fmt.Fprintln(os.Stderr, stack)
-}

vendor/github.com/etcd-io/bbolt/doc.go

@@ -1,5 +1,5 @@
 /*
-Package bolt implements a low-level key/value store in pure Go. It supports
+package bbolt implements a low-level key/value store in pure Go. It supports
 fully serializable transactions, ACID semantics, and lock-free MVCC with
 multiple readers and a single writer. Bolt can be used for projects that
 want a simple data store without the need to add large dependencies such as
@@ -41,4 +41,4 @@ point to different data or can point to invalid memory which will cause a panic.
 
 
 */
-package bolt
+package bbolt

vendor/github.com/etcd-io/bbolt/errors.go

@@ -1,4 +1,4 @@
-package bolt
+package bbolt
 
 import "errors"
 

vendor/github.com/etcd-io/bbolt/freelist.go

@@ -0,0 +1,370 @@
+package bbolt
+
+import (
+	"fmt"
+	"sort"
+	"unsafe"
+)
+
+// txPending holds a list of pgids and corresponding allocation txns
+// that are pending to be freed.
+type txPending struct {
+	ids              []pgid
+	alloctx          []txid // txids allocating the ids
+	lastReleaseBegin txid   // beginning txid of last matching releaseRange
+}
+
+// pidSet holds the set of starting pgids which have the same span size
+type pidSet map[pgid]struct{}
+
+// freelist represents a list of all pages that are available for allocation.
+// It also tracks pages that have been freed but are still in use by open transactions.
+type freelist struct {
+	freelistType   FreelistType                // freelist type
+	ids            []pgid                      // all free and available free page ids.
+	allocs         map[pgid]txid               // mapping of txid that allocated a pgid.
+	pending        map[txid]*txPending         // mapping of soon-to-be free page ids by tx.
+	cache          map[pgid]bool               // fast lookup of all free and pending page ids.
+	freemaps       map[uint64]pidSet           // key is the size of continuous pages(span), value is a set which contains the starting pgids of same size
+	forwardMap     map[pgid]uint64             // key is start pgid, value is its span size
+	backwardMap    map[pgid]uint64             // key is end pgid, value is its span size
+	allocate       func(txid txid, n int) pgid // the freelist allocate func
+	free_count     func() int                  // the function which gives you free page number
+	mergeSpans     func(ids pgids)             // the mergeSpan func
+	getFreePageIDs func() []pgid               // get free pgids func
+	readIDs        func(pgids []pgid)          // readIDs func reads list of pages and init the freelist
+}
+
+// newFreelist returns an empty, initialized freelist.
+func newFreelist(freelistType FreelistType) *freelist {
+	f := &freelist{
+		freelistType: freelistType,
+		allocs:       make(map[pgid]txid),
+		pending:      make(map[txid]*txPending),
+		cache:        make(map[pgid]bool),
+		freemaps:     make(map[uint64]pidSet),
+		forwardMap:   make(map[pgid]uint64),
+		backwardMap:  make(map[pgid]uint64),
+	}
+
+	if freelistType == FreelistMapType {
+		f.allocate = f.hashmapAllocate
+		f.free_count = f.hashmapFreeCount
+		f.mergeSpans = f.hashmapMergeSpans
+		f.getFreePageIDs = f.hashmapGetFreePageIDs
+		f.readIDs = f.hashmapReadIDs
+	} else {
+		f.allocate = f.arrayAllocate
+		f.free_count = f.arrayFreeCount
+		f.mergeSpans = f.arrayMergeSpans
+		f.getFreePageIDs = f.arrayGetFreePageIDs
+		f.readIDs = f.arrayReadIDs
+	}
+
+	return f
+}
+
+// size returns the size of the page after serialization.
+func (f *freelist) size() int {
+	n := f.count()
+	if n >= 0xFFFF {
+		// The first element will be used to store the count. See freelist.write.
+		n++
+	}
+	return pageHeaderSize + (int(unsafe.Sizeof(pgid(0))) * n)
+}
+
+// count returns count of pages on the freelist
+func (f *freelist) count() int {
+	return f.free_count() + f.pending_count()
+}
+
+// arrayFreeCount returns count of free pages(array version)
+func (f *freelist) arrayFreeCount() int {
+	return len(f.ids)
+}
+
+// pending_count returns count of pending pages
+func (f *freelist) pending_count() int {
+	var count int
+	for _, txp := range f.pending {
+		count += len(txp.ids)
+	}
+	return count
+}
+
+// copyall copies into dst a list of all free ids and all pending ids in one sorted list.
+// f.count returns the minimum length required for dst.
+func (f *freelist) copyall(dst []pgid) {
+	m := make(pgids, 0, f.pending_count())
+	for _, txp := range f.pending {
+		m = append(m, txp.ids...)
+	}
+	sort.Sort(m)
+	mergepgids(dst, f.getFreePageIDs(), m)
+}
+
+// arrayAllocate returns the starting page id of a contiguous list of pages of a given size.
+// If a contiguous block cannot be found then 0 is returned.
+func (f *freelist) arrayAllocate(txid txid, n int) pgid {
+	if len(f.ids) == 0 {
+		return 0
+	}
+
+	var initial, previd pgid
+	for i, id := range f.ids {
+		if id <= 1 {
+			panic(fmt.Sprintf("invalid page allocation: %d", id))
+		}
+
+		// Reset initial page if this is not contiguous.
+		if previd == 0 || id-previd != 1 {
+			initial = id
+		}
+
+		// If we found a contiguous block then remove it and return it.
+		if (id-initial)+1 == pgid(n) {
+			// If we're allocating off the beginning then take the fast path
+			// and just adjust the existing slice. This will use extra memory
+			// temporarily but the append() in free() will realloc the slice
+			// as is necessary.
+			if (i + 1) == n {
+				f.ids = f.ids[i+1:]
+			} else {
+				copy(f.ids[i-n+1:], f.ids[i+1:])
+				f.ids = f.ids[:len(f.ids)-n]
+			}
+
+			// Remove from the free cache.
+			for i := pgid(0); i < pgid(n); i++ {
+				delete(f.cache, initial+i)
+			}
+			f.allocs[initial] = txid
+			return initial
+		}
+
+		previd = id
+	}
+	return 0
+}
+
+// free releases a page and its overflow for a given transaction id.
+// If the page is already free then a panic will occur.
+func (f *freelist) free(txid txid, p *page) {
+	if p.id <= 1 {
+		panic(fmt.Sprintf("cannot free page 0 or 1: %d", p.id))
+	}
+
+	// Free page and all its overflow pages.
+	txp := f.pending[txid]
+	if txp == nil {
+		txp = &txPending{}
+		f.pending[txid] = txp
+	}
+	allocTxid, ok := f.allocs[p.id]
+	if ok {
+		delete(f.allocs, p.id)
+	} else if (p.flags & freelistPageFlag) != 0 {
+		// Freelist is always allocated by prior tx.
+		allocTxid = txid - 1
+	}
+
+	for id := p.id; id <= p.id+pgid(p.overflow); id++ {
+		// Verify that page is not already free.
+		if f.cache[id] {
+			panic(fmt.Sprintf("page %d already freed", id))
+		}
+		// Add to the freelist and cache.
+		txp.ids = append(txp.ids, id)
+		txp.alloctx = append(txp.alloctx, allocTxid)
+		f.cache[id] = true
+	}
+}
+
+// release moves all page ids for a transaction id (or older) to the freelist.
+func (f *freelist) release(txid txid) {
+	m := make(pgids, 0)
+	for tid, txp := range f.pending {
+		if tid <= txid {
+			// Move transaction's pending pages to the available freelist.
+			// Don't remove from the cache since the page is still free.
+			m = append(m, txp.ids...)
+			delete(f.pending, tid)
+		}
+	}
+	f.mergeSpans(m)
+}
+
+// releaseRange moves pending pages allocated within an extent [begin,end] to the free list.
+func (f *freelist) releaseRange(begin, end txid) {
+	if begin > end {
+		return
+	}
+	var m pgids
+	for tid, txp := range f.pending {
+		if tid < begin || tid > end {
+			continue
+		}
+		// Don't recompute freed pages if ranges haven't updated.
+		if txp.lastReleaseBegin == begin {
+			continue
+		}
+		for i := 0; i < len(txp.ids); i++ {
+			if atx := txp.alloctx[i]; atx < begin || atx > end {
+				continue
+			}
+			m = append(m, txp.ids[i])
+			txp.ids[i] = txp.ids[len(txp.ids)-1]
+			txp.ids = txp.ids[:len(txp.ids)-1]
+			txp.alloctx[i] = txp.alloctx[len(txp.alloctx)-1]
+			txp.alloctx = txp.alloctx[:len(txp.alloctx)-1]
+			i--
+		}
+		txp.lastReleaseBegin = begin
+		if len(txp.ids) == 0 {
+			delete(f.pending, tid)
+		}
+	}
+	f.mergeSpans(m)
+}
+
+// rollback removes the pages from a given pending tx.
+func (f *freelist) rollback(txid txid) {
+	// Remove page ids from cache.
+	txp := f.pending[txid]
+	if txp == nil {
+		return
+	}
+	var m pgids
+	for i, pgid := range txp.ids {
+		delete(f.cache, pgid)
+		tx := txp.alloctx[i]
+		if tx == 0 {
+			continue
+		}
+		if tx != txid {
+			// Pending free aborted; restore page back to alloc list.
+			f.allocs[pgid] = tx
+		} else {
+			// Freed page was allocated by this txn; OK to throw away.
+			m = append(m, pgid)
+		}
+	}
+	// Remove pages from pending list and mark as free if allocated by txid.
+	delete(f.pending, txid)
+	f.mergeSpans(m)
+}
+
+// freed returns whether a given page is in the free list.
+func (f *freelist) freed(pgid pgid) bool {
+	return f.cache[pgid]
+}
+
+// read initializes the freelist from a freelist page.
+func (f *freelist) read(p *page) {
+	if (p.flags & freelistPageFlag) == 0 {
+		panic(fmt.Sprintf("invalid freelist page: %d, page type is %s", p.id, p.typ()))
+	}
+	// If the page.count is at the max uint16 value (64k) then it's considered
+	// an overflow and the size of the freelist is stored as the first element.
+	idx, count := 0, int(p.count)
+	if count == 0xFFFF {
+		idx = 1
+		count = int(((*[maxAllocSize]pgid)(unsafe.Pointer(&p.ptr)))[0])
+	}
+
+	// Copy the list of page ids from the freelist.
+	if count == 0 {
+		f.ids = nil
+	} else {
+		ids := ((*[maxAllocSize]pgid)(unsafe.Pointer(&p.ptr)))[idx : idx+count]
+
+		// copy the ids, so we don't modify on the freelist page directly
+		idsCopy := make([]pgid, count)
+		copy(idsCopy, ids)
+		// Make sure they're sorted.
+		sort.Sort(pgids(idsCopy))
+
+		f.readIDs(idsCopy)
+	}
+}
+
+// arrayReadIDs initializes the freelist from a given list of ids.
+func (f *freelist) arrayReadIDs(ids []pgid) {
+	f.ids = ids
+	f.reindex()
+}
+
+func (f *freelist) arrayGetFreePageIDs() []pgid {
+	return f.ids
+}
+
+// write writes the page ids onto a freelist page. All free and pending ids are
+// saved to disk since in the event of a program crash, all pending ids will
+// become free.
+func (f *freelist) write(p *page) error {
+	// Combine the old free pgids and pgids waiting on an open transaction.
+
+	// Update the header flag.
+	p.flags |= freelistPageFlag
+
+	// The page.count can only hold up to 64k elements so if we overflow that
+	// number then we handle it by putting the size in the first element.
+	lenids := f.count()
+	if lenids == 0 {
+		p.count = uint16(lenids)
+	} else if lenids < 0xFFFF {
+		p.count = uint16(lenids)
+		f.copyall(((*[maxAllocSize]pgid)(unsafe.Pointer(&p.ptr)))[:])
+	} else {
+		p.count = 0xFFFF
+		((*[maxAllocSize]pgid)(unsafe.Pointer(&p.ptr)))[0] = pgid(lenids)
+		f.copyall(((*[maxAllocSize]pgid)(unsafe.Pointer(&p.ptr)))[1:])
+	}
+
+	return nil
+}
+
+// reload reads the freelist from a page and filters out pending items.
+func (f *freelist) reload(p *page) {
+	f.read(p)
+
+	// Build a cache of only pending pages.
+	pcache := make(map[pgid]bool)
+	for _, txp := range f.pending {
+		for _, pendingID := range txp.ids {
+			pcache[pendingID] = true
+		}
+	}
+
+	// Check each page in the freelist and build a new available freelist
+	// with any pages not in the pending lists.
+	var a []pgid
+	for _, id := range f.getFreePageIDs() {
+		if !pcache[id] {
+			a = append(a, id)
+		}
+	}
+
+	f.readIDs(a)
+}
+
+// reindex rebuilds the free cache based on available and pending free lists.
+func (f *freelist) reindex() {
+	ids := f.getFreePageIDs()
+	f.cache = make(map[pgid]bool, len(ids))
+	for _, id := range ids {
+		f.cache[id] = true
+	}
+	for _, txp := range f.pending {
+		for _, pendingID := range txp.ids {
+			f.cache[pendingID] = true
+		}
+	}
+}
+
+// arrayMergeSpans try to merge list of pages(represented by pgids) with existing spans but using array
+func (f *freelist) arrayMergeSpans(ids pgids) {
+	sort.Sort(ids)
+	f.ids = pgids(f.ids).merge(ids)
+}

vendor/github.com/etcd-io/bbolt/freelist_hmap.go

@@ -0,0 +1,178 @@
+package bbolt
+
+import "sort"
+
+// hashmapFreeCount returns count of free pages(hashmap version)
+func (f *freelist) hashmapFreeCount() int {
+	// use the forwardmap to get the total count
+	count := 0
+	for _, size := range f.forwardMap {
+		count += int(size)
+	}
+	return count
+}
+
+// hashmapAllocate serves the same purpose as arrayAllocate, but use hashmap as backend
+func (f *freelist) hashmapAllocate(txid txid, n int) pgid {
+	if n == 0 {
+		return 0
+	}
+
+	// if we have a exact size match just return short path
+	if bm, ok := f.freemaps[uint64(n)]; ok {
+		for pid := range bm {
+			// remove the span
+			f.delSpan(pid, uint64(n))
+
+			f.allocs[pid] = txid
+
+			for i := pgid(0); i < pgid(n); i++ {
+				delete(f.cache, pid+pgid(i))
+			}
+			return pid
+		}
+	}
+
+	// lookup the map to find larger span
+	for size, bm := range f.freemaps {
+		if size < uint64(n) {
+			continue
+		}
+
+		for pid := range bm {
+			// remove the initial
+			f.delSpan(pid, uint64(size))
+
+			f.allocs[pid] = txid
+
+			remain := size - uint64(n)
+
+			// add remain span
+			f.addSpan(pid+pgid(n), remain)
+
+			for i := pgid(0); i < pgid(n); i++ {
+				delete(f.cache, pid+pgid(i))
+			}
+			return pid
+		}
+	}
+
+	return 0
+}
+
+// hashmapReadIDs reads pgids as input an initial the freelist(hashmap version)
+func (f *freelist) hashmapReadIDs(pgids []pgid) {
+	f.init(pgids)
+
+	// Rebuild the page cache.
+	f.reindex()
+}
+
+// hashmapGetFreePageIDs returns the sorted free page ids
+func (f *freelist) hashmapGetFreePageIDs() []pgid {
+	count := f.free_count()
+	if count == 0 {
+		return nil
+	}
+
+	m := make([]pgid, 0, count)
+	for start, size := range f.forwardMap {
+		for i := 0; i < int(size); i++ {
+			m = append(m, start+pgid(i))
+		}
+	}
+	sort.Sort(pgids(m))
+
+	return m
+}
+
+// hashmapMergeSpans try to merge list of pages(represented by pgids) with existing spans
+func (f *freelist) hashmapMergeSpans(ids pgids) {
+	for _, id := range ids {
+		// try to see if we can merge and update
+		f.mergeWithExistingSpan(id)
+	}
+}
+
+// mergeWithExistingSpan merges pid to the existing free spans, try to merge it backward and forward
+func (f *freelist) mergeWithExistingSpan(pid pgid) {
+	prev := pid - 1
+	next := pid + 1
+
+	preSize, mergeWithPrev := f.backwardMap[prev]
+	nextSize, mergeWithNext := f.forwardMap[next]
+	newStart := pid
+	newSize := uint64(1)
+
+	if mergeWithPrev {
+		//merge with previous span
+		start := prev + 1 - pgid(preSize)
+		f.delSpan(start, preSize)
+
+		newStart -= pgid(preSize)
+		newSize += preSize
+	}
+
+	if mergeWithNext {
+		// merge with next span
+		f.delSpan(next, nextSize)
+		newSize += nextSize
+	}
+
+	f.addSpan(newStart, newSize)
+}
+
+func (f *freelist) addSpan(start pgid, size uint64) {
+	f.backwardMap[start-1+pgid(size)] = size
+	f.forwardMap[start] = size
+	if _, ok := f.freemaps[size]; !ok {
+		f.freemaps[size] = make(map[pgid]struct{})
+	}
+
+	f.freemaps[size][start] = struct{}{}
+}
+
+func (f *freelist) delSpan(start pgid, size uint64) {
+	delete(f.forwardMap, start)
+	delete(f.backwardMap, start+pgid(size-1))
+	delete(f.freemaps[size], start)
+	if len(f.freemaps[size]) == 0 {
+		delete(f.freemaps, size)
+	}
+}
+
+// initial from pgids using when use hashmap version
+// pgids must be sorted
+func (f *freelist) init(pgids []pgid) {
+	if len(pgids) == 0 {
+		return
+	}
+
+	size := uint64(1)
+	start := pgids[0]
+
+	if !sort.SliceIsSorted([]pgid(pgids), func(i, j int) bool { return pgids[i] < pgids[j] }) {
+		panic("pgids not sorted")
+	}
+
+	f.freemaps = make(map[uint64]pidSet)
+	f.forwardMap = make(map[pgid]uint64)
+	f.backwardMap = make(map[pgid]uint64)
+
+	for i := 1; i < len(pgids); i++ {
+		// continuous page
+		if pgids[i] == pgids[i-1]+1 {
+			size++
+		} else {
+			f.addSpan(start, size)
+
+			size = 1
+			start = pgids[i]
+		}
+	}
+
+	// init the tail
+	if size != 0 && start != 0 {
+		f.addSpan(start, size)
+	}
+}

vendor/github.com/etcd-io/bbolt/node.go

@@ -1,4 +1,4 @@
-package bolt
+package bbolt
 
 import (
 	"bytes"
@@ -365,7 +365,7 @@ func (n *node) spill() error {
 		}
 
 		// Allocate contiguous space for the node.
-		p, err := tx.allocate((node.size() / tx.db.pageSize) + 1)
+		p, err := tx.allocate((node.size() + tx.db.pageSize - 1) / tx.db.pageSize)
 		if err != nil {
 			return err
 		}

vendor/github.com/etcd-io/bbolt/page.go

@@ -1,4 +1,4 @@
-package bolt
+package bbolt
 
 import (
 	"fmt"

vendor/github.com/etcd-io/bbolt/tx.go

@@ -1,4 +1,4 @@
-package bolt
+package bbolt
 
 import (
 	"fmt"
@@ -126,10 +126,7 @@ func (tx *Tx) DeleteBucket(name []byte) error {
 // 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 {
-		if err := fn(k, tx.root.Bucket(k)); err != nil {
+		return fn(k, tx.root.Bucket(k))
-			return err
-		}
-		return nil
 	})
 }
 
@@ -169,28 +166,18 @@ func (tx *Tx) Commit() error {
 	// Free the old root bucket.
 	tx.meta.root.root = tx.root.root
 
-	opgid := tx.meta.pgid
+	// Free the old freelist because commit writes out a fresh freelist.
-
+	if tx.meta.freelist != pgidNoFreelist {
-	// Free the freelist and allocate new pages for it. This will overestimate
-	// the size of the freelist but not underestimate the size (which would be bad).
 		tx.db.freelist.free(tx.meta.txid, tx.db.page(tx.meta.freelist))
-	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.db.NoFreelistSync {
-	if tx.meta.pgid > opgid {
+		err := tx.commitFreelist()
-		if err := tx.db.grow(int(tx.meta.pgid+1) * tx.db.pageSize); err != nil {
+		if err != nil {
-			tx.rollback()
 			return err
 		}
+	} else {
+		tx.meta.freelist = pgidNoFreelist
 	}
 
 	// Write dirty pages to disk.
@@ -235,6 +222,31 @@ func (tx *Tx) Commit() error {
 	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 {
@@ -291,7 +303,9 @@ func (tx *Tx) close() {
 }
 
 // Copy writes the entire database to a writer.
-// This function exists for backwards compatibility. Use WriteTo() instead.
+// This function exists for backwards compatibility.
+//
+// Deprecated; Use WriteTo() instead.
 func (tx *Tx) Copy(w io.Writer) error {
 	_, err := tx.WriteTo(w)
 	return err
@@ -305,7 +319,11 @@ func (tx *Tx) WriteTo(w io.Writer) (n int64, err error) {
 	if err != nil {
 		return 0, err
 	}
-	defer func() { _ = f.Close() }()
+	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)
@@ -333,7 +351,7 @@ func (tx *Tx) WriteTo(w io.Writer) (n int64, err error) {
 	}
 
 	// Move past the meta pages in the file.
-	if _, err := f.Seek(int64(tx.db.pageSize*2), os.SEEK_SET); err != nil {
+	if _, err := f.Seek(int64(tx.db.pageSize*2), io.SeekStart); err != nil {
 		return n, fmt.Errorf("seek: %s", err)
 	}
 
@@ -344,7 +362,7 @@ func (tx *Tx) WriteTo(w io.Writer) (n int64, err error) {
 		return n, err
 	}
 
-	return n, f.Close()
+	return n, nil
 }
 
 // CopyFile copies the entire database to file at the given path.
@@ -379,6 +397,9 @@ func (tx *Tx) Check() <-chan error {
 }
 
 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())
@@ -394,9 +415,11 @@ func (tx *Tx) check(ch chan error) {
 	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)
@@ -453,7 +476,7 @@ func (tx *Tx) checkBucket(b *Bucket, reachable map[pgid]*page, freed map[pgid]bo
 
 // 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(count)
+	p, err := tx.db.allocate(tx.meta.txid, count)
 	if err != nil {
 		return nil, err
 	}
@@ -462,7 +485,7 @@ func (tx *Tx) allocate(count int) (*page, error) {
 	tx.pages[p.id] = p
 
 	// Update statistics.
-	tx.stats.PageCount++
+	tx.stats.PageCount += count
 	tx.stats.PageAlloc += count * tx.db.pageSize
 
 	return p, nil