core/state: fix incorrect contract code state metrics (#33376)

## Description
This PR fixes incorrect contract code state metrics by ensuring
duplicate codes are not counted towards the reported results.

## Rationale
The contract code metrics don't consider database deduplication. The
current implementation assumes that the results are only **slightly
inaccurate**, but this is not true, especially for data collection
efforts that started from the genesis block.

Author: weiihann

core/blockchain.go

@@ -1609,15 +1609,22 @@ func (bc *BlockChain) writeBlockWithState(block *types.Block, receipts []*types.
 	if err := blockBatch.Write(); err != nil {
 		log.Crit("Failed to write block into disk", "err", err)
 	}
-	// Commit all cached state changes into underlying memory database.
-	root, stateUpdate, err := statedb.CommitWithUpdate(block.NumberU64(), bc.chainConfig.IsEIP158(block.Number()), bc.chainConfig.IsCancun(block.Number(), block.Time()))
+
+	var (
+		err      error
+		root     common.Hash
+		isEIP158 = bc.chainConfig.IsEIP158(block.Number())
+		isCancun = bc.chainConfig.IsCancun(block.Number(), block.Time())
+	)
+	if bc.stateSizer == nil {
+		root, err = statedb.Commit(block.NumberU64(), isEIP158, isCancun)
+	} else {
+		root, err = statedb.CommitAndTrack(block.NumberU64(), isEIP158, isCancun, bc.stateSizer)
+	}
 	if err != nil {
 		return err
 	}
-	// Emit the state update to the state sizestats if it's active
-	if bc.stateSizer != nil {
-		bc.stateSizer.Notify(stateUpdate)
-	}
+
 	// If node is running in path mode, skip explicit gc operation
 	// which is unnecessary in this mode.
 	if bc.triedb.Scheme() == rawdb.PathScheme {

core/state/reader.go

@@ -40,6 +40,10 @@ import (
 
 // ContractCodeReader defines the interface for accessing contract code.
 type ContractCodeReader interface {
+	// Has returns the flag indicating whether the contract code with
+	// specified address and hash exists or not.
+	Has(addr common.Address, codeHash common.Hash) bool
+
 	// Code retrieves a particular contract's code.
 	//
 	// - Returns nil code along with nil error if the requested contract code
@@ -170,6 +174,13 @@ func (r *cachingCodeReader) CodeSize(addr common.Address, codeHash common.Hash)
 	return len(code), nil
 }
 
+// Has returns the flag indicating whether the contract code with
+// specified address and hash exists or not.
+func (r *cachingCodeReader) Has(addr common.Address, codeHash common.Hash) bool {
+	code, _ := r.Code(addr, codeHash)
+	return len(code) > 0
+}
+
 // flatReader wraps a database state reader and is safe for concurrent access.
 type flatReader struct {
 	reader database.StateReader

core/state/state_sizer.go

@@ -243,12 +243,14 @@ func calSizeStats(update *stateUpdate) (SizeStats, error) {
 		}
 	}
 
-	// Measure code changes. Note that the reported contract code size may be slightly
-	// inaccurate due to database deduplication (code is stored by its hash). However,
-	// this deviation is negligible and acceptable for measurement purposes.
+	codeExists := make(map[common.Hash]struct{})
 	for _, code := range update.codes {
+		if _, ok := codeExists[code.hash]; ok || code.exists {
+			continue
+		}
 		stats.ContractCodes += 1
 		stats.ContractCodeBytes += codeKeySize + int64(len(code.blob))
+		codeExists[code.hash] = struct{}{}
 	}
 	return stats, nil
 }

core/state/state_sizer_test.go

@@ -58,7 +58,7 @@ func TestSizeTracker(t *testing.T) {
 	state.AddBalance(addr3, uint256.NewInt(3000), tracing.BalanceChangeUnspecified)
 	state.SetNonce(addr3, 3, tracing.NonceChangeUnspecified)
 
-	currentRoot, _, err := state.CommitWithUpdate(1, true, false)
+	currentRoot, err := state.Commit(1, true, false)
 	if err != nil {
 		t.Fatalf("Failed to commit initial state: %v", err)
 	}
@@ -83,7 +83,7 @@ func TestSizeTracker(t *testing.T) {
 		if i%3 == 0 {
 			newState.SetCode(testAddr, []byte{byte(i), 0x60, 0x80, byte(i + 1), 0x52}, tracing.CodeChangeUnspecified)
 		}
-		root, _, err := newState.CommitWithUpdate(blockNum, true, false)
+		root, err := newState.Commit(blockNum, true, false)
 		if err != nil {
 			t.Fatalf("Failed to commit state at block %d: %v", blockNum, err)
 		}
@@ -154,21 +154,22 @@ func TestSizeTracker(t *testing.T) {
 		if i%3 == 0 {
 			newState.SetCode(testAddr, []byte{byte(i), 0x60, 0x80, byte(i + 1), 0x52}, tracing.CodeChangeUnspecified)
 		}
-		root, update, err := newState.CommitWithUpdate(blockNum, true, false)
+		ret, err := newState.commitAndFlush(blockNum, true, false, true)
 		if err != nil {
 			t.Fatalf("Failed to commit state at block %d: %v", blockNum, err)
 		}
-		if err := tdb.Commit(root, false); err != nil {
+		tracker.Notify(ret)
+
+		if err := tdb.Commit(ret.root, false); err != nil {
 			t.Fatalf("Failed to commit trie at block %d: %v", blockNum, err)
 		}
 
-		diff, err := calSizeStats(update)
+		diff, err := calSizeStats(ret)
 		if err != nil {
 			t.Fatalf("Failed to calculate size stats for block %d: %v", blockNum, err)
 		}
 		trackedUpdates = append(trackedUpdates, diff)
-		tracker.Notify(update)
-		currentRoot = root
+		currentRoot = ret.root
 	}
 	finalRoot := rawdb.ReadSnapshotRoot(db)
 

core/state/statedb.go

@@ -1317,11 +1317,16 @@ func (s *StateDB) commit(deleteEmptyObjects bool, noStorageWiping bool, blockNum
 
 // commitAndFlush is a wrapper of commit which also commits the state mutations
 // to the configured data stores.
-func (s *StateDB) commitAndFlush(block uint64, deleteEmptyObjects bool, noStorageWiping bool) (*stateUpdate, error) {
+func (s *StateDB) commitAndFlush(block uint64, deleteEmptyObjects bool, noStorageWiping bool, dedupCode bool) (*stateUpdate, error) {
 	ret, err := s.commit(deleteEmptyObjects, noStorageWiping, block)
 	if err != nil {
 		return nil, err
 	}
+
+	if dedupCode {
+		ret.markCodeExistence(s.reader)
+	}
+
 	// Commit dirty contract code if any exists
 	if db := s.db.TrieDB().Disk(); db != nil && len(ret.codes) > 0 {
 		batch := db.NewBatch()
@@ -1376,21 +1381,21 @@ func (s *StateDB) commitAndFlush(block uint64, deleteEmptyObjects bool, noStorag
 // no empty accounts left that could be deleted by EIP-158, storage wiping
 // should not occur.
 func (s *StateDB) Commit(block uint64, deleteEmptyObjects bool, noStorageWiping bool) (common.Hash, error) {
-	ret, err := s.commitAndFlush(block, deleteEmptyObjects, noStorageWiping)
+	ret, err := s.commitAndFlush(block, deleteEmptyObjects, noStorageWiping, false)
 	if err != nil {
 		return common.Hash{}, err
 	}
 	return ret.root, nil
 }
 
-// CommitWithUpdate writes the state mutations and returns both the root hash and the state update.
-// This is useful for tracking state changes at the blockchain level.
-func (s *StateDB) CommitWithUpdate(block uint64, deleteEmptyObjects bool, noStorageWiping bool) (common.Hash, *stateUpdate, error) {
-	ret, err := s.commitAndFlush(block, deleteEmptyObjects, noStorageWiping)
+// CommitAndTrack writes the state mutations and notifies the size tracker of the state changes.
+func (s *StateDB) CommitAndTrack(block uint64, deleteEmptyObjects bool, noStorageWiping bool, sizer *SizeTracker) (common.Hash, error) {
+	ret, err := s.commitAndFlush(block, deleteEmptyObjects, noStorageWiping, true)
 	if err != nil {
-		return common.Hash{}, nil, err
+		return common.Hash{}, err
 	}
-	return ret.root, ret, nil
+	sizer.Notify(ret)
+	return ret.root, nil
 }
 
 // Prepare handles the preparatory steps for executing a state transition with.

core/state/statedb_fuzz_test.go

@@ -228,7 +228,7 @@ func (test *stateTest) run() bool {
 		} else {
 			state.IntermediateRoot(true) // call intermediateRoot at the transaction boundary
 		}
-		ret, err := state.commitAndFlush(0, true, false) // call commit at the block boundary
+		ret, err := state.commitAndFlush(0, true, false, false) // call commit at the block boundary
 		if err != nil {
 			panic(err)
 		}

core/state/stateupdate.go

@@ -26,8 +26,9 @@ import (
 
 // contractCode represents a contract code with associated metadata.
 type contractCode struct {
-	hash common.Hash // hash is the cryptographic hash of the contract code.
-	blob []byte      // blob is the binary representation of the contract code.
+	hash   common.Hash // hash is the cryptographic hash of the contract code.
+	blob   []byte      // blob is the binary representation of the contract code.
+	exists bool        // flag whether the code has been existent
 }
 
 // accountDelete represents an operation for deleting an Ethereum account.
@@ -190,3 +191,22 @@ func (sc *stateUpdate) stateSet() *triedb.StateSet {
 		RawStorageKey:  sc.rawStorageKey,
 	}
 }
+
+// markCodeExistence determines whether each piece of contract code referenced
+// in this state update actually exists.
+//
+// Note: This operation is expensive and not needed during normal state transitions.
+// It is only required when SizeTracker is enabled to produce accurate state
+// statistics.
+func (sc *stateUpdate) markCodeExistence(reader ContractCodeReader) {
+	cache := make(map[common.Hash]bool)
+	for addr, code := range sc.codes {
+		if exists, ok := cache[code.hash]; ok {
+			code.exists = exists
+			continue
+		}
+		res := reader.Has(addr, code.hash)
+		cache[code.hash] = res
+		code.exists = res
+	}
+}