core/state: state reader abstraction (#29761)

This pull request introduces a state.Reader interface for state
accessing.

The interface could be implemented in various ways. It can be pure trie
only reader, or the combination of trie and state snapshot. What's more,
this interface allows us to have more flexibility in the future, e.g.
the
archive reader (for accessing archive state).

Additionally, this pull request removes the following metrics

- `chain/snapshot/account/reads`
- `chain/snapshot/storage/reads`

cmd/evm/internal/t8ntool/execution.go

@@ -403,7 +403,7 @@ func (pre *Prestate) Apply(vmConfig vm.Config, chainConfig *params.ChainConfig,
 	}
 	// Re-create statedb instance with new root upon the updated database
 	// for accessing latest states.
-	statedb, err = state.New(root, statedb.Database(), nil)
+	statedb, err = state.New(root, statedb.Database())
 	if err != nil {
 		return nil, nil, nil, NewError(ErrorEVM, fmt.Errorf("could not reopen state: %v", err))
 	}
@@ -412,8 +412,9 @@ func (pre *Prestate) Apply(vmConfig vm.Config, chainConfig *params.ChainConfig,
 }
 
 func MakePreState(db ethdb.Database, accounts types.GenesisAlloc) *state.StateDB {
-	sdb := state.NewDatabaseWithConfig(db, &triedb.Config{Preimages: true})
-	statedb, _ := state.New(types.EmptyRootHash, sdb, nil)
+	tdb := triedb.NewDatabase(db, &triedb.Config{Preimages: true})
+	sdb := state.NewDatabase(tdb, nil)
+	statedb, _ := state.New(types.EmptyRootHash, sdb)
 	for addr, a := range accounts {
 		statedb.SetCode(addr, a.Code)
 		statedb.SetNonce(addr, a.Nonce)
@@ -424,7 +425,7 @@ func MakePreState(db ethdb.Database, accounts types.GenesisAlloc) *state.StateDB
 	}
 	// Commit and re-open to start with a clean state.
 	root, _ := statedb.Commit(0, false)
-	statedb, _ = state.New(root, sdb, nil)
+	statedb, _ = state.New(root, sdb)
 	return statedb
 }
 

cmd/evm/runner.go

@@ -155,8 +155,8 @@ func runCmd(ctx *cli.Context) error {
 	})
 	defer triedb.Close()
 	genesis := genesisConfig.MustCommit(db, triedb)
-	sdb := state.NewDatabaseWithNodeDB(db, triedb)
-	statedb, _ = state.New(genesis.Root(), sdb, nil)
+	sdb := state.NewDatabase(triedb, nil)
+	statedb, _ = state.New(genesis.Root(), sdb)
 	chainConfig = genesisConfig.Config
 
 	if ctx.String(SenderFlag.Name) != "" {
@@ -277,7 +277,7 @@ func runCmd(ctx *cli.Context) error {
 			fmt.Printf("Failed to commit changes %v\n", err)
 			return err
 		}
-		dumpdb, err := state.New(root, sdb, nil)
+		dumpdb, err := state.New(root, sdb)
 		if err != nil {
 			fmt.Printf("Failed to open statedb %v\n", err)
 			return err

cmd/evm/staterunner.go

@@ -107,7 +107,7 @@ func runStateTest(fname string, cfg vm.Config, dump bool) error {
 					result.Root = &root
 					fmt.Fprintf(os.Stderr, "{\"stateRoot\": \"%#x\"}\n", root)
 					if dump { // Dump any state to aid debugging
-						cpy, _ := state.New(root, tstate.StateDB.Database(), nil)
+						cpy, _ := state.New(root, tstate.StateDB.Database())
 						dump := cpy.RawDump(nil)
 						result.State = &dump
 					}

cmd/geth/chaincmd.go

@@ -584,7 +584,7 @@ func dump(ctx *cli.Context) error {
 	triedb := utils.MakeTrieDatabase(ctx, db, true, true, false) // always enable preimage lookup
 	defer triedb.Close()
 
-	state, err := state.New(root, state.NewDatabaseWithNodeDB(db, triedb), nil)
+	state, err := state.New(root, state.NewDatabase(triedb, nil))
 	if err != nil {
 		return err
 	}

core/blockchain.go

@@ -72,14 +72,11 @@ var (
 	storageUpdateTimer = metrics.NewRegisteredResettingTimer("chain/storage/updates", nil)
 	storageCommitTimer = metrics.NewRegisteredResettingTimer("chain/storage/commits", nil)
 
-	snapshotAccountReadTimer = metrics.NewRegisteredResettingTimer("chain/snapshot/account/reads", nil)
-	snapshotStorageReadTimer = metrics.NewRegisteredResettingTimer("chain/snapshot/storage/reads", nil)
-	snapshotCommitTimer      = metrics.NewRegisteredResettingTimer("chain/snapshot/commits", nil)
-
 	accountReadSingleTimer = metrics.NewRegisteredResettingTimer("chain/account/single/reads", nil)
 	storageReadSingleTimer = metrics.NewRegisteredResettingTimer("chain/storage/single/reads", nil)
 
-	triedbCommitTimer = metrics.NewRegisteredResettingTimer("chain/triedb/commits", nil)
+	snapshotCommitTimer = metrics.NewRegisteredResettingTimer("chain/snapshot/commits", nil)
+	triedbCommitTimer   = metrics.NewRegisteredResettingTimer("chain/triedb/commits", nil)
 
 	blockInsertTimer     = metrics.NewRegisteredResettingTimer("chain/inserts", nil)
 	blockValidationTimer = metrics.NewRegisteredResettingTimer("chain/validation", nil)
@@ -220,7 +217,7 @@ type BlockChain struct {
 	lastWrite     uint64                           // Last block when the state was flushed
 	flushInterval atomic.Int64                     // Time interval (processing time) after which to flush a state
 	triedb        *triedb.Database                 // The database handler for maintaining trie nodes.
-	stateCache    state.Database                   // State database to reuse between imports (contains state cache)
+	statedb       *state.CachingDB                 // State database to reuse between imports (contains state cache)
 	txIndexer     *txIndexer                       // Transaction indexer, might be nil if not enabled
 
 	hc            *HeaderChain
@@ -311,7 +308,7 @@ func NewBlockChain(db ethdb.Database, cacheConfig *CacheConfig, genesis *Genesis
 		return nil, err
 	}
 	bc.flushInterval.Store(int64(cacheConfig.TrieTimeLimit))
-	bc.stateCache = state.NewDatabaseWithNodeDB(bc.db, bc.triedb)
+	bc.statedb = state.NewDatabase(bc.triedb, nil)
 	bc.validator = NewBlockValidator(chainConfig, bc)
 	bc.prefetcher = newStatePrefetcher(chainConfig, bc.hc)
 	bc.processor = NewStateProcessor(chainConfig, bc.hc)
@@ -449,7 +446,11 @@ func NewBlockChain(db ethdb.Database, cacheConfig *CacheConfig, genesis *Genesis
 			AsyncBuild: !bc.cacheConfig.SnapshotWait,
 		}
 		bc.snaps, _ = snapshot.New(snapconfig, bc.db, bc.triedb, head.Root)
+
+		// Re-initialize the state database with snapshot
+		bc.statedb = state.NewDatabase(bc.triedb, bc.snaps)
 	}
+
 	// Rewind the chain in case of an incompatible config upgrade.
 	if compat, ok := genesisErr.(*params.ConfigCompatError); ok {
 		log.Warn("Rewinding chain to upgrade configuration", "err", compat)
@@ -1767,7 +1768,7 @@ func (bc *BlockChain) insertChain(chain types.Blocks, setHead bool) (int, error)
 		if parent == nil {
 			parent = bc.GetHeader(block.ParentHash(), block.NumberU64()-1)
 		}
-		statedb, err := state.New(parent.Root, bc.stateCache, bc.snaps)
+		statedb, err := state.New(parent.Root, bc.statedb)
 		if err != nil {
 			return it.index, err
 		}
@@ -1793,7 +1794,7 @@ func (bc *BlockChain) insertChain(chain types.Blocks, setHead bool) (int, error)
 		var followupInterrupt atomic.Bool
 		if !bc.cacheConfig.TrieCleanNoPrefetch {
 			if followup, err := it.peek(); followup != nil && err == nil {
-				throwaway, _ := state.New(parent.Root, bc.stateCache, bc.snaps)
+				throwaway, _ := state.New(parent.Root, bc.statedb)
 
 				go func(start time.Time, followup *types.Block, throwaway *state.StateDB) {
 					// Disable tracing for prefetcher executions.
@@ -1914,26 +1915,21 @@ func (bc *BlockChain) processBlock(block *types.Block, statedb *state.StateDB, s
 	proctime := time.Since(start) // processing + validation
 
 	// Update the metrics touched during block processing and validation
-	accountReadTimer.Update(statedb.AccountReads)                 // Account reads are complete(in processing)
-	storageReadTimer.Update(statedb.StorageReads)                 // Storage reads are complete(in processing)
-	snapshotAccountReadTimer.Update(statedb.SnapshotAccountReads) // Account reads are complete(in processing)
-	snapshotStorageReadTimer.Update(statedb.SnapshotStorageReads) // Storage reads are complete(in processing)
-
-	accountRead := statedb.SnapshotAccountReads + statedb.AccountReads // The time spent on account read
-	storageRead := statedb.SnapshotStorageReads + statedb.StorageReads // The time spent on storage read
+	accountReadTimer.Update(statedb.AccountReads) // Account reads are complete(in processing)
+	storageReadTimer.Update(statedb.StorageReads) // Storage reads are complete(in processing)
 	if statedb.AccountLoaded != 0 {
-		accountReadSingleTimer.Update(accountRead / time.Duration(statedb.AccountLoaded))
+		accountReadSingleTimer.Update(statedb.AccountReads / time.Duration(statedb.AccountLoaded))
 	}
 	if statedb.StorageLoaded != 0 {
-		storageReadSingleTimer.Update(storageRead / time.Duration(statedb.StorageLoaded))
-	}
-	accountUpdateTimer.Update(statedb.AccountUpdates)               // Account updates are complete(in validation)
-	storageUpdateTimer.Update(statedb.StorageUpdates)               // Storage updates are complete(in validation)
-	accountHashTimer.Update(statedb.AccountHashes)                  // Account hashes are complete(in validation)
-	triehash := statedb.AccountHashes                               // The time spent on tries hashing
-	trieUpdate := statedb.AccountUpdates + statedb.StorageUpdates   // The time spent on tries update
-	blockExecutionTimer.Update(ptime - (accountRead + storageRead)) // The time spent on EVM processing
-	blockValidationTimer.Update(vtime - (triehash + trieUpdate))    // The time spent on block validation
+		storageReadSingleTimer.Update(statedb.StorageReads / time.Duration(statedb.StorageLoaded))
+	}
+	accountUpdateTimer.Update(statedb.AccountUpdates)                                 // Account updates are complete(in validation)
+	storageUpdateTimer.Update(statedb.StorageUpdates)                                 // Storage updates are complete(in validation)
+	accountHashTimer.Update(statedb.AccountHashes)                                    // Account hashes are complete(in validation)
+	triehash := statedb.AccountHashes                                                 // The time spent on tries hashing
+	trieUpdate := statedb.AccountUpdates + statedb.StorageUpdates                     // The time spent on tries update
+	blockExecutionTimer.Update(ptime - (statedb.AccountReads + statedb.StorageReads)) // The time spent on EVM processing
+	blockValidationTimer.Update(vtime - (triehash + trieUpdate))                      // The time spent on block validation
 
 	// Write the block to the chain and get the status.
 	var (

core/blockchain_reader.go

@@ -308,7 +308,7 @@ func (bc *BlockChain) GetTd(hash common.Hash, number uint64) *big.Int {
 
 // HasState checks if state trie is fully present in the database or not.
 func (bc *BlockChain) HasState(hash common.Hash) bool {
-	_, err := bc.stateCache.OpenTrie(hash)
+	_, err := bc.statedb.OpenTrie(hash)
 	return err == nil
 }
 
@@ -341,12 +341,9 @@ func (bc *BlockChain) stateRecoverable(root common.Hash) bool {
 // If the code doesn't exist in the in-memory cache, check the storage with
 // new code scheme.
 func (bc *BlockChain) ContractCodeWithPrefix(hash common.Hash) ([]byte, error) {
-	type codeReader interface {
-		ContractCodeWithPrefix(address common.Address, codeHash common.Hash) ([]byte, error)
-	}
 	// TODO(rjl493456442) The associated account address is also required
 	// in Verkle scheme. Fix it once snap-sync is supported for Verkle.
-	return bc.stateCache.(codeReader).ContractCodeWithPrefix(common.Address{}, hash)
+	return bc.statedb.ContractCodeWithPrefix(common.Address{}, hash)
 }
 
 // State returns a new mutable state based on the current HEAD block.
@@ -356,7 +353,7 @@ func (bc *BlockChain) State() (*state.StateDB, error) {
 
 // StateAt returns a new mutable state based on a particular point in time.
 func (bc *BlockChain) StateAt(root common.Hash) (*state.StateDB, error) {
-	return state.New(root, bc.stateCache, bc.snaps)
+	return state.New(root, bc.statedb)
 }
 
 // Config retrieves the chain's fork configuration.
@@ -382,7 +379,7 @@ func (bc *BlockChain) Processor() Processor {
 
 // StateCache returns the caching database underpinning the blockchain instance.
 func (bc *BlockChain) StateCache() state.Database {
-	return bc.stateCache
+	return bc.statedb
 }
 
 // GasLimit returns the gas limit of the current HEAD block.

core/blockchain_sethead_test.go

@@ -2040,7 +2040,7 @@ func testSetHeadWithScheme(t *testing.T, tt *rewindTest, snapshots bool, scheme
 		dbconfig.HashDB = hashdb.Defaults
 	}
 	chain.triedb = triedb.NewDatabase(chain.db, dbconfig)
-	chain.stateCache = state.NewDatabaseWithNodeDB(chain.db, chain.triedb)
+	chain.statedb = state.NewDatabase(chain.triedb, chain.snaps)
 
 	// Force run a freeze cycle
 	type freezer interface {

core/blockchain_test.go

@@ -160,7 +160,7 @@ func testBlockChainImport(chain types.Blocks, blockchain *BlockChain) error {
 			}
 			return err
 		}
-		statedb, err := state.New(blockchain.GetBlockByHash(block.ParentHash()).Root(), blockchain.stateCache, nil)
+		statedb, err := state.New(blockchain.GetBlockByHash(block.ParentHash()).Root(), blockchain.statedb)
 		if err != nil {
 			return err
 		}

core/chain_makers.go

@@ -379,7 +379,7 @@ func GenerateChain(config *params.ChainConfig, parent *types.Block, engine conse
 	defer triedb.Close()
 
 	for i := 0; i < n; i++ {
-		statedb, err := state.New(parent.Root(), state.NewDatabaseWithNodeDB(db, triedb), nil)
+		statedb, err := state.New(parent.Root(), state.NewDatabase(triedb, nil))
 		if err != nil {
 			panic(err)
 		}
@@ -485,7 +485,7 @@ func GenerateVerkleChain(config *params.ChainConfig, parent *types.Block, engine
 	}
 
 	for i := 0; i < n; i++ {
-		statedb, err := state.New(parent.Root(), state.NewDatabaseWithNodeDB(db, trdb), nil)
+		statedb, err := state.New(parent.Root(), state.NewDatabase(trdb, nil))
 		if err != nil {
 			panic(err)
 		}

core/genesis.go

@@ -127,8 +127,8 @@ func hashAlloc(ga *types.GenesisAlloc, isVerkle bool) (common.Hash, error) {
 	}
 	// Create an ephemeral in-memory database for computing hash,
 	// all the derived states will be discarded to not pollute disk.
-	db := state.NewDatabaseWithConfig(rawdb.NewMemoryDatabase(), config)
-	statedb, err := state.New(types.EmptyRootHash, db, nil)
+	db := rawdb.NewMemoryDatabase()
+	statedb, err := state.New(types.EmptyRootHash, state.NewDatabase(triedb.NewDatabase(db, config), nil))
 	if err != nil {
 		return common.Hash{}, err
 	}
@@ -147,8 +147,8 @@ func hashAlloc(ga *types.GenesisAlloc, isVerkle bool) (common.Hash, error) {
 
 // flushAlloc is very similar with hash, but the main difference is all the
 // generated states will be persisted into the given database.
-func flushAlloc(ga *types.GenesisAlloc, db ethdb.Database, triedb *triedb.Database) (common.Hash, error) {
-	statedb, err := state.New(types.EmptyRootHash, state.NewDatabaseWithNodeDB(db, triedb), nil)
+func flushAlloc(ga *types.GenesisAlloc, triedb *triedb.Database) (common.Hash, error) {
+	statedb, err := state.New(types.EmptyRootHash, state.NewDatabase(triedb, nil))
 	if err != nil {
 		return common.Hash{}, err
 	}
@@ -497,7 +497,7 @@ func (g *Genesis) Commit(db ethdb.Database, triedb *triedb.Database) (*types.Blo
 		return nil, errors.New("can't start clique chain without signers")
 	}
 	// flush the data to disk and compute the state root
-	root, err := flushAlloc(&g.Alloc, db, triedb)
+	root, err := flushAlloc(&g.Alloc, triedb)
 	if err != nil {
 		return nil, err
 	}