trie: use stacktrie for Derivesha operation (ethereum#21407)

core/types: use stacktrie for derivesha

trie: add stacktrie file

trie: fix linter

core/types: use stacktrie for derivesha

rebased: adapt stacktrie to the newer version of DeriveSha

Co-authored-by: Martin Holst Swende <[email protected]>

More linter fixes

review feedback: no key offset for nodes converted to hashes

trie: use EncodeRLP for full nodes

core/types: insert txs in order in derivesha

trie: tests for derivesha with stacktrie

trie: make stacktrie use pooled hashers

trie: make stacktrie reuse tmp slice space

trie: minor polishes on stacktrie

trie/stacktrie: less rlp dancing

core/types: explain the contorsions in DeriveSha

ci: fix goimport errors

trie: clear mem on subtrie hashing

squashme: linter fix

stracktrie: use pooling, less allocs (ethereum#3)

trie: in-place hex prefix, reduce allocs and add rawNode.EncodeRLP

Reintroduce the `[]node` method, add the missing `EncodeRLP` implementation for `rawNode` and calculate the hex prefix in place.

Co-authored-by: Martin Holst Swende <[email protected]>

Co-authored-by: Martin Holst Swende <[email protected]>

Author: gballet

core/block_validator.go

@@ -62,7 +62,7 @@ func (v *BlockValidator) ValidateBody(block *types.Block) error {
 	if hash := types.CalcUncleHash(block.Uncles()); hash != header.UncleHash {
 		return fmt.Errorf("uncle root hash mismatch: have %x, want %x", hash, header.UncleHash)
 	}
-	if hash := types.DeriveSha(block.Transactions(), new(trie.Trie)); hash != header.TxHash {
+	if hash := types.DeriveSha(block.Transactions(), trie.NewStackTrie(nil)); hash != header.TxHash {
 		return fmt.Errorf("transaction root hash mismatch: have %x, want %x", hash, header.TxHash)
 	}
 	if !v.bc.HasBlockAndState(block.ParentHash(), block.NumberU64()-1) {
@@ -90,7 +90,7 @@ func (v *BlockValidator) ValidateState(block *types.Block, statedb *state.StateD
 		return fmt.Errorf("invalid bloom (remote: %x  local: %x)", header.Bloom, rbloom)
 	}
 	// Tre receipt Trie's root (R = (Tr [[H1, R1], ... [Hn, Rn]]))
-	receiptSha := types.DeriveSha(receipts, new(trie.Trie))
+	receiptSha := types.DeriveSha(receipts, trie.NewStackTrie(nil))
 	if receiptSha != header.ReceiptHash {
 		return fmt.Errorf("invalid receipt root hash (remote: %x local: %x)", header.ReceiptHash, receiptSha)
 	}

core/types/derive_sha.go

@@ -23,7 +23,6 @@ import (
 	"github.com/ethereum/go-ethereum/rlp"
 )
 
-// DerivableList is the interface which can derive the hash.
 type DerivableList interface {
 	Len() int
 	GetRlp(i int) []byte
@@ -39,7 +38,22 @@ type Hasher interface {
 func DeriveSha(list DerivableList, hasher Hasher) common.Hash {
 	hasher.Reset()
 	keybuf := new(bytes.Buffer)
-	for i := 0; i < list.Len(); i++ {
+
+	// StackTrie requires values to be inserted in increasing
+	// hash order, which is not the order that `list` provides
+	// hashes in. This insertion sequence ensures that the
+	// order is correct.
+	for i := 1; i < list.Len() && i <= 0x7f; i++ {
+		keybuf.Reset()
+		rlp.Encode(keybuf, uint(i))
+		hasher.Update(keybuf.Bytes(), list.GetRlp(i))
+	}
+	if list.Len() > 0 {
+		keybuf.Reset()
+		rlp.Encode(keybuf, uint(0))
+		hasher.Update(keybuf.Bytes(), list.GetRlp(0))
+	}
+	for i := 0x80; i < list.Len(); i++ {
 		keybuf.Reset()
 		rlp.Encode(keybuf, uint(i))
 		hasher.Update(keybuf.Bytes(), list.GetRlp(i))

eth/downloader/queue.go

@@ -774,7 +774,7 @@ func (q *queue) DeliverBodies(id string, txLists [][]*types.Transaction, uncleLi
 	q.lock.Lock()
 	defer q.lock.Unlock()
 	validate := func(index int, header *types.Header) error {
-		if types.DeriveSha(types.Transactions(txLists[index]), new(trie.Trie)) != header.TxHash {
+		if types.DeriveSha(types.Transactions(txLists[index]), trie.NewStackTrie(nil)) != header.TxHash {
 			return errInvalidBody
 		}
 		if types.CalcUncleHash(uncleLists[index]) != header.UncleHash {
@@ -799,7 +799,7 @@ func (q *queue) DeliverReceipts(id string, receiptList [][]*types.Receipt) (int,
 	q.lock.Lock()
 	defer q.lock.Unlock()
 	validate := func(index int, header *types.Header) error {
-		if types.DeriveSha(types.Receipts(receiptList[index]), new(trie.Trie)) != header.ReceiptHash {
+		if types.DeriveSha(types.Receipts(receiptList[index]), trie.NewStackTrie(nil)) != header.ReceiptHash {
 			return errInvalidReceipt
 		}
 		return nil

eth/handler.go

@@ -715,7 +715,7 @@ func (pm *ProtocolManager) handleMsg(p *peer) error {
 			log.Warn("Propagated block has invalid uncles", "have", hash, "exp", request.Block.UncleHash())
 			break // TODO(karalabe): return error eventually, but wait a few releases
 		}
-		if hash := types.DeriveSha(request.Block.Transactions(), new(trie.Trie)); hash != request.Block.TxHash() {
+		if hash := types.DeriveSha(request.Block.Transactions(), trie.NewStackTrie(nil)); hash != request.Block.TxHash() {
 			log.Warn("Propagated block has invalid body", "have", hash, "exp", request.Block.TxHash())
 			break // TODO(karalabe): return error eventually, but wait a few releases
 		}

trie/database.go

@@ -99,6 +99,11 @@ type rawNode []byte
 func (n rawNode) cache() (hashNode, bool)   { panic("this should never end up in a live trie") }
 func (n rawNode) fstring(ind string) string { panic("this should never end up in a live trie") }
 
+func (n rawNode) EncodeRLP(w io.Writer) error {
+	_, err := w.Write([]byte(n))
+	return err
+}
+
 // rawFullNode represents only the useful data content of a full node, with the
 // caches and flags stripped out to minimize its data storage. This type honors
 // the same RLP encoding as the original parent.
@@ -199,7 +204,7 @@ func forGatherChildren(n node, onChild func(hash common.Hash)) {
 		}
 	case hashNode:
 		onChild(common.BytesToHash(n))
-	case valueNode, nil:
+	case valueNode, nil, rawNode:
 	default:
 		panic(fmt.Sprintf("unknown node type: %T", n))
 	}

trie/encoding.go

@@ -51,6 +51,35 @@ func hexToCompact(hex []byte) []byte {
 	return buf
 }
 
+// hexToCompactInPlace places the compact key in input buffer, returning the length
+// needed for the representation
+func hexToCompactInPlace(hex []byte) int {
+	var (
+		hexLen    = len(hex) // length of the hex input
+		firstByte = byte(0)
+	)
+	// Check if we have a terminator there
+	if hexLen > 0 && hex[hexLen-1] == 16 {
+		firstByte = 1 << 5
+		hexLen-- // last part was the terminator, ignore that
+	}
+	var (
+		binLen = hexLen/2 + 1
+		ni     = 0 // index in hex
+		bi     = 1 // index in bin (compact)
+	)
+	if hexLen&1 == 1 {
+		firstByte |= 1 << 4 // odd flag
+		firstByte |= hex[0] // first nibble is contained in the first byte
+		ni++
+	}
+	for ; ni < hexLen; bi, ni = bi+1, ni+2 {
+		hex[bi] = hex[ni]<<4 | hex[ni+1]
+	}
+	hex[0] = firstByte
+	return binLen
+}
+
 func compactToHex(compact []byte) []byte {
 	if len(compact) == 0 {
 		return compact

trie/encoding_test.go

@@ -18,6 +18,8 @@ package trie
 
 import (
 	"bytes"
+	"encoding/hex"
+	"math/rand"
 	"testing"
 )
 
@@ -75,6 +77,40 @@ func TestHexKeybytes(t *testing.T) {
 	}
 }
 
+func TestHexToCompactInPlace(t *testing.T) {
+	for i, keyS := range []string{
+		"00",
+		"060a040c0f000a090b040803010801010900080d090a0a0d0903000b10",
+		"10",
+	} {
+		hexBytes, _ := hex.DecodeString(keyS)
+		exp := hexToCompact(hexBytes)
+		sz := hexToCompactInPlace(hexBytes)
+		got := hexBytes[:sz]
+		if !bytes.Equal(exp, got) {
+			t.Fatalf("test %d: encoding err\ninp %v\ngot %x\nexp %x\n", i, keyS, got, exp)
+		}
+	}
+}
+
+func TestHexToCompactInPlaceRandom(t *testing.T) {
+	for i := 0; i < 10000; i++ {
+		l := rand.Intn(128)
+		key := make([]byte, l)
+		rand.Read(key)
+		hexBytes := keybytesToHex(key)
+		hexOrig := []byte(string(hexBytes))
+		exp := hexToCompact(hexBytes)
+		sz := hexToCompactInPlace(hexBytes)
+		got := hexBytes[:sz]
+
+		if !bytes.Equal(exp, got) {
+			t.Fatalf("encoding err \ncpt %x\nhex %x\ngot %x\nexp %x\n",
+				key, hexOrig, got, exp)
+		}
+	}
+}
+
 func BenchmarkHexToCompact(b *testing.B) {
 	testBytes := []byte{0, 15, 1, 12, 11, 8, 16 /*term*/}
 	for i := 0; i < b.N; i++ {