Merge pull request #3858 from gyuho/godep_bolt_20151112

Godeps: update boltdb/bolt for MAP_POPULATE

Godeps/Godeps.json

@@ -24,8 +24,8 @@
 		},
 		{
 			"ImportPath": "github.com/boltdb/bolt",
-			"Comment": "v1.0-158-g81db894",
-			"Rev": "81db89446cb805bc352f803151f47fea849241e2"
+			"Comment": "v1.1.0-19-g0b00eff",
+			"Rev": "0b00effdd7a8270ebd91c24297e51643e370dd52"
 		},
 		{
 			"ImportPath": "github.com/bradfitz/http2",

Godeps/_workspace/src/github.com/boltdb/bolt/bolt_unix.go

@@ -58,7 +58,7 @@ func mmap(db *DB, sz int) error {
 	}
 
 	// Map the data file to memory.
-	b, err := syscall.Mmap(int(db.file.Fd()), 0, sz, syscall.PROT_READ, syscall.MAP_SHARED)
+	b, err := syscall.Mmap(int(db.file.Fd()), 0, sz, syscall.PROT_READ, syscall.MAP_SHARED|db.MmapFlags)
 	if err != nil {
 		return err
 	}

Godeps/_workspace/src/github.com/boltdb/bolt/bolt_unix_solaris.go

@@ -2,11 +2,12 @@ package bolt
 
 import (
 	"fmt"
-	"github.com/coreos/etcd/Godeps/_workspace/src/golang.org/x/sys/unix"
 	"os"
 	"syscall"
 	"time"
 	"unsafe"
+
+	"github.com/coreos/etcd/Godeps/_workspace/src/golang.org/x/sys/unix"
 )
 
 // flock acquires an advisory lock on a file descriptor.
@@ -67,7 +68,7 @@ func mmap(db *DB, sz int) error {
 	}
 
 	// Map the data file to memory.
-	b, err := unix.Mmap(int(db.file.Fd()), 0, sz, syscall.PROT_READ, syscall.MAP_SHARED)
+	b, err := unix.Mmap(int(db.file.Fd()), 0, sz, syscall.PROT_READ, syscall.MAP_SHARED|db.MmapFlags)
 	if err != nil {
 		return err
 	}

Godeps/_workspace/src/github.com/boltdb/bolt/cursor.go

@@ -34,6 +34,13 @@ func (c *Cursor) First() (key []byte, value []byte) {
 	p, n := c.bucket.pageNode(c.bucket.root)
 	c.stack = append(c.stack, elemRef{page: p, node: n, index: 0})
 	c.first()
+
+	// If we land on an empty page then move to the next value.
+	// https://github.com/boltdb/bolt/issues/450
+	if c.stack[len(c.stack)-1].count() == 0 {
+		c.next()
+	}
+
 	k, v, flags := c.keyValue()
 	if (flags & uint32(bucketLeafFlag)) != 0 {
 		return k, nil
@@ -209,28 +216,37 @@ func (c *Cursor) last() {
 // next moves to the next leaf element and returns the key and value.
 // If the cursor is at the last leaf element then it stays there and returns nil.
 func (c *Cursor) next() (key []byte, value []byte, flags uint32) {
-	// Attempt to move over one element until we're successful.
-	// Move up the stack as we hit the end of each page in our stack.
-	var i int
-	for i = len(c.stack) - 1; i >= 0; i-- {
-		elem := &c.stack[i]
-		if elem.index < elem.count()-1 {
-			elem.index++
-			break
+	for {
+		// Attempt to move over one element until we're successful.
+		// Move up the stack as we hit the end of each page in our stack.
+		var i int
+		for i = len(c.stack) - 1; i >= 0; i-- {
+			elem := &c.stack[i]
+			if elem.index < elem.count()-1 {
+				elem.index++
+				break
+			}
 		}
-	}
 
-	// If we've hit the root page then stop and return. This will leave the
-	// cursor on the last element of the last page.
-	if i == -1 {
-		return nil, nil, 0
-	}
+		// If we've hit the root page then stop and return. This will leave the
+		// cursor on the last element of the last page.
+		if i == -1 {
+			return nil, nil, 0
+		}
 
-	// Otherwise start from where we left off in the stack and find the
-	// first element of the first leaf page.
-	c.stack = c.stack[:i+1]
-	c.first()
-	return c.keyValue()
+		// Otherwise start from where we left off in the stack and find the
+		// first element of the first leaf page.
+		c.stack = c.stack[:i+1]
+		c.first()
+
+		// If this is an empty page then restart and move back up the stack.
+		// https://github.com/boltdb/bolt/issues/450
+		if c.stack[len(c.stack)-1].count() == 0 {
+			continue
+		}
+
+		return c.keyValue()
+	}
 }
 
 // search recursively performs a binary search against a given page/node until it finds a given key.

Godeps/_workspace/src/github.com/boltdb/bolt/cursor_test.go

@@ -303,6 +303,49 @@ func TestCursor_Restart(t *testing.T) {
 	tx.Rollback()
 }
 
+// Ensure that a cursor can skip over empty pages that have been deleted.
+func TestCursor_First_EmptyPages(t *testing.T) {
+	db := NewTestDB()
+	defer db.Close()
+
+	// Create 1000 keys in the "widgets" bucket.
+	db.Update(func(tx *bolt.Tx) error {
+		b, err := tx.CreateBucket([]byte("widgets"))
+		if err != nil {
+			t.Fatal(err)
+		}
+
+		for i := 0; i < 1000; i++ {
+			if err := b.Put(u64tob(uint64(i)), []byte{}); err != nil {
+				t.Fatal(err)
+			}
+		}
+
+		return nil
+	})
+
+	// Delete half the keys and then try to iterate.
+	db.Update(func(tx *bolt.Tx) error {
+		b := tx.Bucket([]byte("widgets"))
+		for i := 0; i < 600; i++ {
+			if err := b.Delete(u64tob(uint64(i))); err != nil {
+				t.Fatal(err)
+			}
+		}
+
+		c := b.Cursor()
+		var n int
+		for k, _ := c.First(); k != nil; k, _ = c.Next() {
+			n++
+		}
+		if n != 400 {
+			t.Fatalf("unexpected key count: %d", n)
+		}
+
+		return nil
+	})
+}
+
 // Ensure that a Tx can iterate over all elements in a bucket.
 func TestCursor_QuickCheck(t *testing.T) {
 	f := func(items testdata) bool {

Godeps/_workspace/src/github.com/boltdb/bolt/db.go

@@ -63,6 +63,10 @@ type DB struct {
 	// https://github.com/boltdb/bolt/issues/284
 	NoGrowSync bool
 
+	// If you want to read the entire database fast, you can set MmapFlag to
+	// syscall.MAP_POPULATE on Linux 2.6.23+ for sequential read-ahead.
+	MmapFlags int
+
 	// MaxBatchSize is the maximum size of a batch. Default value is
 	// copied from DefaultMaxBatchSize in Open.
 	//
@@ -136,6 +140,7 @@ func Open(path string, mode os.FileMode, options *Options) (*DB, error) {
 		options = DefaultOptions
 	}
 	db.NoGrowSync = options.NoGrowSync
+	db.MmapFlags = options.MmapFlags
 
 	// Set default values for later DB operations.
 	db.MaxBatchSize = DefaultMaxBatchSize
@@ -672,6 +677,9 @@ type Options struct {
 	// Open database in read-only mode. Uses flock(..., LOCK_SH |LOCK_NB) to
 	// grab a shared lock (UNIX).
 	ReadOnly bool
+
+	// Sets the DB.MmapFlags flag before memory mapping the file.
+	MmapFlags int
 }
 
 // DefaultOptions represent the options used if nil options are passed into Open().