core, miner, trie: relocate witness stats (#34106)

This PR relocates the witness statistics into the witness itself, making
it more self-contained.

core/blockchain.go

@@ -2170,24 +2170,18 @@ func (bc *BlockChain) ProcessBlock(ctx context.Context, parentRoot common.Hash,
 	// If we are past Byzantium, enable prefetching to pull in trie node paths
 	// while processing transactions. Before Byzantium the prefetcher is mostly
 	// useless due to the intermediate root hashing after each transaction.
-	var (
+	var witness *stateless.Witness
-		witness      *stateless.Witness
-		witnessStats *stateless.WitnessStats
-	)
 	if bc.chainConfig.IsByzantium(block.Number()) {
 		// Generate witnesses either if we're self-testing, or if it's the
 		// only block being inserted. A bit crude, but witnesses are huge,
 		// so we refuse to make an entire chain of them.
 		if config.StatelessSelfValidation || config.MakeWitness {
-			witness, err = stateless.NewWitness(block.Header(), bc)
+			witness, err = stateless.NewWitness(block.Header(), bc, config.EnableWitnessStats)
 			if err != nil {
 				return nil, err
 			}
-			if config.EnableWitnessStats {
-				witnessStats = stateless.NewWitnessStats()
-			}
 		}
-		statedb.StartPrefetcher("chain", witness, witnessStats)
+		statedb.StartPrefetcher("chain", witness)
 		defer statedb.StopPrefetcher()
 	}
 
@@ -2306,8 +2300,8 @@ func (bc *BlockChain) ProcessBlock(ctx context.Context, parentRoot common.Hash,
 		stats.BlockWrite = time.Since(wstart) - max(statedb.AccountCommits, statedb.StorageCommits) /* concurrent */ - statedb.DatabaseCommits
 	}
 	// Report the collected witness statistics
-	if witnessStats != nil {
+	if witness != nil {
-		witnessStats.ReportMetrics(block.NumberU64())
+		witness.ReportMetrics(block.NumberU64())
 	}
 	elapsed := time.Since(startTime) + 1 // prevent zero division
 	stats.TotalTime = elapsed

core/state/statedb.go

@@ -135,8 +135,7 @@ type StateDB struct {
 	journal *journal
 
 	// State witness if cross validation is needed
-	witness      *stateless.Witness
+	witness *stateless.Witness
-	witnessStats *stateless.WitnessStats
 
 	// Measurements gathered during execution for debugging purposes
 	AccountReads   time.Duration
@@ -201,13 +200,12 @@ func NewWithReader(root common.Hash, db Database, reader Reader) (*StateDB, erro
 // StartPrefetcher initializes a new trie prefetcher to pull in nodes from the
 // state trie concurrently while the state is mutated so that when we reach the
 // commit phase, most of the needed data is already hot.
-func (s *StateDB) StartPrefetcher(namespace string, witness *stateless.Witness, witnessStats *stateless.WitnessStats) {
+func (s *StateDB) StartPrefetcher(namespace string, witness *stateless.Witness) {
 	// Terminate any previously running prefetcher
 	s.StopPrefetcher()
 
 	// Enable witness collection if requested
 	s.witness = witness
-	s.witnessStats = witnessStats
 
 	// With the switch to the Proof-of-Stake consensus algorithm, block production
 	// rewards are now handled at the consensus layer. Consequently, a block may
@@ -913,7 +911,7 @@ func (s *StateDB) IntermediateRoot(deleteEmptyObjects bool) common.Hash {
 				// If witness building is enabled and the state object has a trie,
 				// gather the witnesses for its specific storage trie
 				if s.witness != nil && obj.trie != nil {
-					s.witness.AddState(obj.trie.Witness())
+					s.witness.AddState(obj.trie.Witness(), obj.addrHash())
 				}
 				return nil
 			})
@@ -930,17 +928,9 @@ func (s *StateDB) IntermediateRoot(deleteEmptyObjects bool) common.Hash {
 				continue
 			}
 			if trie := obj.getPrefetchedTrie(); trie != nil {
-				witness := trie.Witness()
+				s.witness.AddState(trie.Witness(), obj.addrHash())
-				s.witness.AddState(witness)
-				if s.witnessStats != nil {
-					s.witnessStats.Add(witness, obj.addrHash())
-				}
 			} else if obj.trie != nil {
-				witness := obj.trie.Witness()
+				s.witness.AddState(obj.trie.Witness(), obj.addrHash())
-				s.witness.AddState(witness)
-				if s.witnessStats != nil {
-					s.witnessStats.Add(witness, obj.addrHash())
-				}
 			}
 		}
 		// Pull in only-read and non-destructed trie witnesses
@@ -954,17 +944,9 @@ func (s *StateDB) IntermediateRoot(deleteEmptyObjects bool) common.Hash {
 				continue
 			}
 			if trie := obj.getPrefetchedTrie(); trie != nil {
-				witness := trie.Witness()
+				s.witness.AddState(trie.Witness(), obj.addrHash())
-				s.witness.AddState(witness)
-				if s.witnessStats != nil {
-					s.witnessStats.Add(witness, obj.addrHash())
-				}
 			} else if obj.trie != nil {
-				witness := obj.trie.Witness()
+				s.witness.AddState(obj.trie.Witness(), obj.addrHash())
-				s.witness.AddState(witness)
-				if s.witnessStats != nil {
-					s.witnessStats.Add(witness, obj.addrHash())
-				}
 			}
 		}
 	}
@@ -1037,11 +1019,7 @@ func (s *StateDB) IntermediateRoot(deleteEmptyObjects bool) common.Hash {
 
 	// If witness building is enabled, gather the account trie witness
 	if s.witness != nil {
-		witness := s.trie.Witness()
+		s.witness.AddState(s.trie.Witness(), common.Hash{})
-		s.witness.AddState(witness)
-		if s.witnessStats != nil {
-			s.witnessStats.Add(witness, common.Hash{})
-		}
 	}
 	return hash
 }

core/stateless/stats.go

@@ -54,6 +54,13 @@ func NewWitnessStats() *WitnessStats {
 	}
 }
 
+func (s *WitnessStats) copy() *WitnessStats {
+	return &WitnessStats{
+		accountTrie: s.accountTrie.Copy(),
+		storageTrie: s.storageTrie.Copy(),
+	}
+}
+
 func (s *WitnessStats) init() {
 	if s.accountTrie == nil {
 		s.accountTrie = trie.NewLevelStats()

core/stateless/witness.go

@@ -42,12 +42,13 @@ type Witness struct {
 	Codes   map[string]struct{} // Set of bytecodes ran or accessed
 	State   map[string]struct{} // Set of MPT state trie nodes (account and storage together)
 
-	chain HeaderReader // Chain reader to convert block hash ops to header proofs
+	chain HeaderReader  // Chain reader to convert block hash ops to header proofs
-	lock  sync.Mutex   // Lock to allow concurrent state insertions
+	stats *WitnessStats // Optional statistics collector
+	lock  sync.Mutex    // Lock to allow concurrent state insertions
 }
 
 // NewWitness creates an empty witness ready for population.
-func NewWitness(context *types.Header, chain HeaderReader) (*Witness, error) {
+func NewWitness(context *types.Header, chain HeaderReader, enableStats bool) (*Witness, error) {
 	// When building witnesses, retrieve the parent header, which will *always*
 	// be included to act as a trustless pre-root hash container
 	var headers []*types.Header
@@ -59,13 +60,17 @@ func NewWitness(context *types.Header, chain HeaderReader) (*Witness, error) {
 		headers = append(headers, parent)
 	}
 	// Create the witness with a reconstructed gutted out block
-	return &Witness{
+	w := &Witness{
 		context: context,
 		Headers: headers,
 		Codes:   make(map[string]struct{}),
 		State:   make(map[string]struct{}),
 		chain:   chain,
-	}, nil
+	}
+	if enableStats {
+		w.stats = NewWitnessStats()
+	}
+	return w, nil
 }
 
 // AddBlockHash adds a "blockhash" to the witness with the designated offset from
@@ -87,8 +92,11 @@ func (w *Witness) AddCode(code []byte) {
 	w.Codes[string(code)] = struct{}{}
 }
 
-// AddState inserts a batch of MPT trie nodes into the witness.
+// AddState inserts a batch of MPT trie nodes into the witness. The owner
-func (w *Witness) AddState(nodes map[string][]byte) {
+// identifies which trie the nodes belong to: the zero hash for the account
+// trie, or the hashed address for a storage trie. This is used for optional
+// statistics collection.
+func (w *Witness) AddState(nodes map[string][]byte, owner common.Hash) {
 	if len(nodes) == 0 {
 		return
 	}
@@ -98,6 +106,17 @@ func (w *Witness) AddState(nodes map[string][]byte) {
 	for _, value := range nodes {
 		w.State[string(value)] = struct{}{}
 	}
+	if w.stats != nil {
+		w.stats.Add(nodes, owner)
+	}
+}
+
+// ReportMetrics reports the collected statistics to the global metrics registry.
+func (w *Witness) ReportMetrics(blockNumber uint64) {
+	if w.stats == nil {
+		return
+	}
+	w.stats.ReportMetrics(blockNumber)
 }
 
 func (w *Witness) AddKey() {
@@ -113,6 +132,9 @@ func (w *Witness) Copy() *Witness {
 		State:   maps.Clone(w.State),
 		chain:   w.chain,
 	}
+	if w.stats != nil {
+		cpy.stats = w.stats.copy()
+	}
 	if w.context != nil {
 		cpy.context = types.CopyHeader(w.context)
 	}

miner/worker.go

@@ -330,12 +330,12 @@ func (miner *Miner) makeEnv(parent *types.Header, header *types.Header, coinbase
 	}
 	var bundle *stateless.Witness
 	if witness {
-		bundle, err = stateless.NewWitness(header, miner.chain)
+		bundle, err = stateless.NewWitness(header, miner.chain, false)
 		if err != nil {
 			return nil, err
 		}
 	}
-	state.StartPrefetcher("miner", bundle, nil)
+	state.StartPrefetcher("miner", bundle)
 	// Note the passed coinbase may be different with header.Coinbase.
 	return &environment{
 		signer:   types.MakeSigner(miner.chainConfig, header.Number, header.Time),

trie/levelstats.go

@@ -36,6 +36,18 @@ func NewLevelStats() *LevelStats {
 	return &LevelStats{}
 }
 
+// Copy returns a deep copy of the statistics.
+func (s *LevelStats) Copy() *LevelStats {
+	cpy := NewLevelStats()
+	for i := range s.level {
+		cpy.level[i].short.Store(s.level[i].short.Load())
+		cpy.level[i].full.Store(s.level[i].full.Load())
+		cpy.level[i].value.Store(s.level[i].value.Load())
+		cpy.level[i].size.Store(s.level[i].size.Load())
+	}
+	return cpy
+}
+
 // MaxDepth iterates each level and finds the deepest level with at least one
 // trie node.
 func (s *LevelStats) MaxDepth() int {