Merge pull request #15225 from holiman/test_removefrom_dirtyset

Change handling of dirty objects in state

5 files changed, 166 insertions, 91 deletions

diff --git a/core/state/dump.go b/core/state/dump.go
index 46e612850..072dbbf05 100644
--- a/core/state/dump.go
+++ b/core/state/dump.go
@@ -53,7 +53,7 @@ func (self *StateDB) RawDump() Dump {
 			panic(err)
 		}
 
-		obj := newObject(nil, common.BytesToAddress(addr), data, nil)
+		obj := newObject(nil, common.BytesToAddress(addr), data)
 		account := DumpAccount{
 			Balance:  data.Balance.String(),
 			Nonce:    data.Nonce,
diff --git a/core/state/journal.go b/core/state/journal.go
index a89bb3d13..a03ca57db 100644
--- a/core/state/journal.go
+++ b/core/state/journal.go
@@ -22,11 +22,67 @@ import (
 	"github.com/ethereum/go-ethereum/common"
 )
 
+// journalEntry is a modification entry in the state change journal that can be
+// reverted on demand.
 type journalEntry interface {
-	undo(*StateDB)
+	// revert undoes the changes introduced by this journal entry.
+	revert(*StateDB)
+
+	// dirtied returns the Ethereum address modified by this journal entry.
+	dirtied() *common.Address
+}
+
+// journal contains the list of state modifications applied since the last state
+// commit. These are tracked to be able to be reverted in case of an execution
+// exception or revertal request.
+type journal struct {
+	entries []journalEntry         // Current changes tracked by the journal
+	dirties map[common.Address]int // Dirty accounts and the number of changes
+}
+
+// newJournal create a new initialized journal.
+func newJournal() *journal {
+	return &journal{
+		dirties: make(map[common.Address]int),
+	}
+}
+
+// append inserts a new modification entry to the end of the change journal.
+func (j *journal) append(entry journalEntry) {
+	j.entries = append(j.entries, entry)
+	if addr := entry.dirtied(); addr != nil {
+		j.dirties[*addr]++
+	}
+}
+
+// revert undoes a batch of journalled modifications along with any reverted
+// dirty handling too.
+func (j *journal) revert(statedb *StateDB, snapshot int) {
+	for i := len(j.entries) - 1; i >= snapshot; i-- {
+		// Undo the changes made by the operation
+		j.entries[i].revert(statedb)
+
+		// Drop any dirty tracking induced by the change
+		if addr := j.entries[i].dirtied(); addr != nil {
+			if j.dirties[*addr]--; j.dirties[*addr] == 0 {
+				delete(j.dirties, *addr)
+			}
+		}
+	}
+	j.entries = j.entries[:snapshot]
+}
+
+// dirty explicitly sets an address to dirty, even if the change entries would
+// otherwise suggest it as clean. This method is an ugly hack to handle the RIPEMD
+// precompile consensus exception.
+func (j *journal) dirty(addr common.Address) {
+	j.dirties[addr]++
 }
 
-type journal []journalEntry
+// length returns the current number of entries in the journal.
+func (j *journal) length() int {
+	return len(j.entries)
+}
 
 type (
 	// Changes to the account trie.
@@ -77,16 +133,24 @@ type (
 	}
 )
 
-func (ch createObjectChange) undo(s *StateDB) {
+func (ch createObjectChange) revert(s *StateDB) {
 	delete(s.stateObjects, *ch.account)
 	delete(s.stateObjectsDirty, *ch.account)
 }
 
-func (ch resetObjectChange) undo(s *StateDB) {
+func (ch createObjectChange) dirtied() *common.Address {
+	return ch.account
+}
+
+func (ch resetObjectChange) revert(s *StateDB) {
 	s.setStateObject(ch.prev)
 }
 
-func (ch suicideChange) undo(s *StateDB) {
+func (ch resetObjectChange) dirtied() *common.Address {
+	return nil
+}
+
+func (ch suicideChange) revert(s *StateDB) {
 	obj := s.getStateObject(*ch.account)
 	if obj != nil {
 		obj.suicided = ch.prev
@@ -94,38 +158,60 @@ func (ch suicideChange) undo(s *StateDB) {
 	}
 }
 
+func (ch suicideChange) dirtied() *common.Address {
+	return ch.account
+}
+
 var ripemd = common.HexToAddress("0000000000000000000000000000000000000003")
 
-func (ch touchChange) undo(s *StateDB) {
-	if !ch.prev && *ch.account != ripemd {
-		s.getStateObject(*ch.account).touched = ch.prev
-		if !ch.prevDirty {
-			delete(s.stateObjectsDirty, *ch.account)
-		}
-	}
+func (ch touchChange) revert(s *StateDB) {
+}
+
+func (ch touchChange) dirtied() *common.Address {
+	return ch.account
 }
 
-func (ch balanceChange) undo(s *StateDB) {
+func (ch balanceChange) revert(s *StateDB) {
 	s.getStateObject(*ch.account).setBalance(ch.prev)
 }
 
-func (ch nonceChange) undo(s *StateDB) {
+func (ch balanceChange) dirtied() *common.Address {
+	return ch.account
+}
+
+func (ch nonceChange) revert(s *StateDB) {
 	s.getStateObject(*ch.account).setNonce(ch.prev)
 }
 
-func (ch codeChange) undo(s *StateDB) {
+func (ch nonceChange) dirtied() *common.Address {
+	return ch.account
+}
+
+func (ch codeChange) revert(s *StateDB) {
 	s.getStateObject(*ch.account).setCode(common.BytesToHash(ch.prevhash), ch.prevcode)
 }
 
-func (ch storageChange) undo(s *StateDB) {
+func (ch codeChange) dirtied() *common.Address {
+	return ch.account
+}
+
+func (ch storageChange) revert(s *StateDB) {
 	s.getStateObject(*ch.account).setState(ch.key, ch.prevalue)
 }
 
-func (ch refundChange) undo(s *StateDB) {
+func (ch storageChange) dirtied() *common.Address {
+	return ch.account
+}
+
+func (ch refundChange) revert(s *StateDB) {
 	s.refund = ch.prev
 }
 
-func (ch addLogChange) undo(s *StateDB) {
+func (ch refundChange) dirtied() *common.Address {
+	return nil
+}
+
+func (ch addLogChange) revert(s *StateDB) {
 	logs := s.logs[ch.txhash]
 	if len(logs) == 1 {
 		delete(s.logs, ch.txhash)
@@ -135,6 +221,14 @@ func (ch addLogChange) undo(s *StateDB) {
 	s.logSize--
 }
 
-func (ch addPreimageChange) undo(s *StateDB) {
+func (ch addLogChange) dirtied() *common.Address {
+	return nil
+}
+
+func (ch addPreimageChange) revert(s *StateDB) {
 	delete(s.preimages, ch.hash)
 }
+
+func (ch addPreimageChange) dirtied() *common.Address {
+	return nil
+}
diff --git a/core/state/state_object.go b/core/state/state_object.go
index b2112bfae..3c40c2041 100644
--- a/core/state/state_object.go
+++ b/core/state/state_object.go
@@ -85,9 +85,7 @@ type stateObject struct {
 	// during the "update" phase of the state transition.
 	dirtyCode bool // true if the code was updated
 	suicided  bool
-	touched   bool
 	deleted   bool
-	onDirty   func(addr common.Address) // Callback method to mark a state object newly dirty
 }
 
 // empty returns whether the account is considered empty.
@@ -105,7 +103,7 @@ type Account struct {
 }
 
 // newObject creates a state object.
-func newObject(db *StateDB, address common.Address, data Account, onDirty func(addr common.Address)) *stateObject {
+func newObject(db *StateDB, address common.Address, data Account) *stateObject {
 	if data.Balance == nil {
 		data.Balance = new(big.Int)
 	}
@@ -119,7 +117,6 @@ func newObject(db *StateDB, address common.Address, data Account, onDirty func(a
 		data:          data,
 		cachedStorage: make(Storage),
 		dirtyStorage:  make(Storage),
-		onDirty:       onDirty,
 	}
 }
 
@@ -137,23 +134,17 @@ func (self *stateObject) setError(err error) {
 
 func (self *stateObject) markSuicided() {
 	self.suicided = true
-	if self.onDirty != nil {
-		self.onDirty(self.Address())
-		self.onDirty = nil
-	}
 }
 
 func (c *stateObject) touch() {
-	c.db.journal = append(c.db.journal, touchChange{
-		account:   &c.address,
-		prev:      c.touched,
-		prevDirty: c.onDirty == nil,
+	c.db.journal.append(touchChange{
+		account: &c.address,
 	})
-	if c.onDirty != nil {
-		c.onDirty(c.Address())
-		c.onDirty = nil
+	if c.address == ripemd {
+		// Explicitly put it in the dirty-cache, which is otherwise generated from
+		// flattened journals.
+		c.db.journal.dirty(c.address)
 	}
-	c.touched = true
 }
 
 func (c *stateObject) getTrie(db Database) Trie {
@@ -195,7 +186,7 @@ func (self *stateObject) GetState(db Database, key common.Hash) common.Hash {
 
 // SetState updates a value in account storage.
 func (self *stateObject) SetState(db Database, key, value common.Hash) {
-	self.db.journal = append(self.db.journal, storageChange{
+	self.db.journal.append(storageChange{
 		account:  &self.address,
 		key:      key,
 		prevalue: self.GetState(db, key),
@@ -207,10 +198,6 @@ func (self *stateObject) setState(key, value common.Hash) {
 	self.cachedStorage[key] = value
 	self.dirtyStorage[key] = value
 
-	if self.onDirty != nil {
-		self.onDirty(self.Address())
-		self.onDirty = nil
-	}
 }
 
 // updateTrie writes cached storage modifications into the object's storage trie.
@@ -274,7 +261,7 @@ func (c *stateObject) SubBalance(amount *big.Int) {
 }
 
 func (self *stateObject) SetBalance(amount *big.Int) {
-	self.db.journal = append(self.db.journal, balanceChange{
+	self.db.journal.append(balanceChange{
 		account: &self.address,
 		prev:    new(big.Int).Set(self.data.Balance),
 	})
@@ -283,17 +270,13 @@ func (self *stateObject) SetBalance(amount *big.Int) {
 
 func (self *stateObject) setBalance(amount *big.Int) {
 	self.data.Balance = amount
-	if self.onDirty != nil {
-		self.onDirty(self.Address())
-		self.onDirty = nil
-	}
 }
 
 // Return the gas back to the origin. Used by the Virtual machine or Closures
 func (c *stateObject) ReturnGas(gas *big.Int) {}
 
-func (self *stateObject) deepCopy(db *StateDB, onDirty func(addr common.Address)) *stateObject {
-	stateObject := newObject(db, self.address, self.data, onDirty)
+func (self *stateObject) deepCopy(db *StateDB) *stateObject {
+	stateObject := newObject(db, self.address, self.data)
 	if self.trie != nil {
 		stateObject.trie = db.db.CopyTrie(self.trie)
 	}
@@ -333,7 +316,7 @@ func (self *stateObject) Code(db Database) []byte {
 
 func (self *stateObject) SetCode(codeHash common.Hash, code []byte) {
 	prevcode := self.Code(self.db.db)
-	self.db.journal = append(self.db.journal, codeChange{
+	self.db.journal.append(codeChange{
 		account:  &self.address,
 		prevhash: self.CodeHash(),
 		prevcode: prevcode,
@@ -345,14 +328,10 @@ func (self *stateObject) setCode(codeHash common.Hash, code []byte) {
 	self.code = code
 	self.data.CodeHash = codeHash[:]
 	self.dirtyCode = true
-	if self.onDirty != nil {
-		self.onDirty(self.Address())
-		self.onDirty = nil
-	}
 }
 
 func (self *stateObject) SetNonce(nonce uint64) {
-	self.db.journal = append(self.db.journal, nonceChange{
+	self.db.journal.append(nonceChange{
 		account: &self.address,
 		prev:    self.data.Nonce,
 	})
@@ -361,10 +340,6 @@ func (self *stateObject) SetNonce(nonce uint64) {
 
 func (self *stateObject) setNonce(nonce uint64) {
 	self.data.Nonce = nonce
-	if self.onDirty != nil {
-		self.onDirty(self.Address())
-		self.onDirty = nil
-	}
 }
 
 func (self *stateObject) CodeHash() []byte {
diff --git a/core/state/statedb.go b/core/state/statedb.go
index bd67e789d..08f18c2d2 100644
--- a/core/state/statedb.go
+++ b/core/state/statedb.go
@@ -76,7 +76,7 @@ type StateDB struct {
 
 	// Journal of state modifications. This is the backbone of
 	// Snapshot and RevertToSnapshot.
-	journal        journal
+	journal        *journal
 	validRevisions []revision
 	nextRevisionId int
 
@@ -96,6 +96,7 @@ func New(root common.Hash, db Database) (*StateDB, error) {
 		stateObjectsDirty: make(map[common.Address]struct{}),
 		logs:              make(map[common.Hash][]*types.Log),
 		preimages:         make(map[common.Hash][]byte),
+		journal:           newJournal(),
 	}, nil
 }
 
@@ -131,7 +132,7 @@ func (self *StateDB) Reset(root common.Hash) error {
 }
 
 func (self *StateDB) AddLog(log *types.Log) {
-	self.journal = append(self.journal, addLogChange{txhash: self.thash})
+	self.journal.append(addLogChange{txhash: self.thash})
 
 	log.TxHash = self.thash
 	log.BlockHash = self.bhash
@@ -156,7 +157,7 @@ func (self *StateDB) Logs() []*types.Log {
 // AddPreimage records a SHA3 preimage seen by the VM.
 func (self *StateDB) AddPreimage(hash common.Hash, preimage []byte) {
 	if _, ok := self.preimages[hash]; !ok {
-		self.journal = append(self.journal, addPreimageChange{hash: hash})
+		self.journal.append(addPreimageChange{hash: hash})
 		pi := make([]byte, len(preimage))
 		copy(pi, preimage)
 		self.preimages[hash] = pi
@@ -169,7 +170,7 @@ func (self *StateDB) Preimages() map[common.Hash][]byte {
 }
 
 func (self *StateDB) AddRefund(gas uint64) {
-	self.journal = append(self.journal, refundChange{prev: self.refund})
+	self.journal.append(refundChange{prev: self.refund})
 	self.refund += gas
 }
 
@@ -255,7 +256,7 @@ func (self *StateDB) StorageTrie(addr common.Address) Trie {
 	if stateObject == nil {
 		return nil
 	}
-	cpy := stateObject.deepCopy(self, nil)
+	cpy := stateObject.deepCopy(self)
 	return cpy.updateTrie(self.db)
 }
 
@@ -325,7 +326,7 @@ func (self *StateDB) Suicide(addr common.Address) bool {
 	if stateObject == nil {
 		return false
 	}
-	self.journal = append(self.journal, suicideChange{
+	self.journal.append(suicideChange{
 		account:     &addr,
 		prev:        stateObject.suicided,
 		prevbalance: new(big.Int).Set(stateObject.Balance()),
@@ -379,7 +380,7 @@ func (self *StateDB) getStateObject(addr common.Address) (stateObject *stateObje
 		return nil
 	}
 	// Insert into the live set.
-	obj := newObject(self, addr, data, self.MarkStateObjectDirty)
+	obj := newObject(self, addr, data)
 	self.setStateObject(obj)
 	return obj
 }
@@ -397,22 +398,16 @@ func (self *StateDB) GetOrNewStateObject(addr common.Address) *stateObject {
 	return stateObject
 }
 
-// MarkStateObjectDirty adds the specified object to the dirty map to avoid costly
-// state object cache iteration to find a handful of modified ones.
-func (self *StateDB) MarkStateObjectDirty(addr common.Address) {
-	self.stateObjectsDirty[addr] = struct{}{}
-}
-
 // createObject creates a new state object. If there is an existing account with
 // the given address, it is overwritten and returned as the second return value.
 func (self *StateDB) createObject(addr common.Address) (newobj, prev *stateObject) {
 	prev = self.getStateObject(addr)
-	newobj = newObject(self, addr, Account{}, self.MarkStateObjectDirty)
+	newobj = newObject(self, addr, Account{})
 	newobj.setNonce(0) // sets the object to dirty
 	if prev == nil {
-		self.journal = append(self.journal, createObjectChange{account: &addr})
+		self.journal.append(createObjectChange{account: &addr})
 	} else {
-		self.journal = append(self.journal, resetObjectChange{prev: prev})
+		self.journal.append(resetObjectChange{prev: prev})
 	}
 	self.setStateObject(newobj)
 	return newobj, prev
@@ -466,16 +461,17 @@ func (self *StateDB) Copy() *StateDB {
 	state := &StateDB{
 		db:                self.db,
 		trie:              self.db.CopyTrie(self.trie),
-		stateObjects:      make(map[common.Address]*stateObject, len(self.stateObjectsDirty)),
-		stateObjectsDirty: make(map[common.Address]struct{}, len(self.stateObjectsDirty)),
+		stateObjects:      make(map[common.Address]*stateObject, len(self.journal.dirties)),
+		stateObjectsDirty: make(map[common.Address]struct{}, len(self.journal.dirties)),
 		refund:            self.refund,
 		logs:              make(map[common.Hash][]*types.Log, len(self.logs)),
 		logSize:           self.logSize,
 		preimages:         make(map[common.Hash][]byte),
+		journal:           newJournal(),
 	}
 	// Copy the dirty states, logs, and preimages
-	for addr := range self.stateObjectsDirty {
-		state.stateObjects[addr] = self.stateObjects[addr].deepCopy(state, state.MarkStateObjectDirty)
+	for addr := range self.journal.dirties {
+		state.stateObjects[addr] = self.stateObjects[addr].deepCopy(state)
 		state.stateObjectsDirty[addr] = struct{}{}
 	}
 	for hash, logs := range self.logs {
@@ -492,7 +488,7 @@ func (self *StateDB) Copy() *StateDB {
 func (self *StateDB) Snapshot() int {
 	id := self.nextRevisionId
 	self.nextRevisionId++
-	self.validRevisions = append(self.validRevisions, revision{id, len(self.journal)})
+	self.validRevisions = append(self.validRevisions, revision{id, self.journal.length()})
 	return id
 }
 
@@ -507,13 +503,8 @@ func (self *StateDB) RevertToSnapshot(revid int) {
 	}
 	snapshot := self.validRevisions[idx].journalIndex
 
-	// Replay the journal to undo changes.
-	for i := len(self.journal) - 1; i >= snapshot; i-- {
-		self.journal[i].undo(self)
-	}
-	self.journal = self.journal[:snapshot]
-
-	// Remove invalidated snapshots from the stack.
+	// Replay the journal to undo changes and remove invalidated snapshots
+	self.journal.revert(self, snapshot)
 	self.validRevisions = self.validRevisions[:idx]
 }
 
@@ -525,14 +516,25 @@ func (self *StateDB) GetRefund() uint64 {
 // Finalise finalises the state by removing the self destructed objects
 // and clears the journal as well as the refunds.
 func (s *StateDB) Finalise(deleteEmptyObjects bool) {
-	for addr := range s.stateObjectsDirty {
-		stateObject := s.stateObjects[addr]
+	for addr := range s.journal.dirties {
+		stateObject, exist := s.stateObjects[addr]
+		if !exist {
+			// ripeMD is 'touched' at block 1714175, in tx 0x1237f737031e40bcde4a8b7e717b2d15e3ecadfe49bb1bbc71ee9deb09c6fcf2
+			// That tx goes out of gas, and although the notion of 'touched' does not exist there, the
+			// touch-event will still be recorded in the journal. Since ripeMD is a special snowflake,
+			// it will persist in the journal even though the journal is reverted. In this special circumstance,
+			// it may exist in `s.journal.dirties` but not in `s.stateObjects`.
+			// Thus, we can safely ignore it here
+			continue
+		}
+
 		if stateObject.suicided || (deleteEmptyObjects && stateObject.empty()) {
 			s.deleteStateObject(stateObject)
 		} else {
 			stateObject.updateRoot(s.db)
 			s.updateStateObject(stateObject)
 		}
+		s.stateObjectsDirty[addr] = struct{}{}
 	}
 	// Invalidate journal because reverting across transactions is not allowed.
 	s.clearJournalAndRefund()
@@ -576,7 +578,7 @@ func (s *StateDB) DeleteSuicides() {
 }
 
 func (s *StateDB) clearJournalAndRefund() {
-	s.journal = nil
+	s.journal = newJournal()
 	s.validRevisions = s.validRevisions[:0]
 	s.refund = 0
 }
@@ -585,6 +587,9 @@ func (s *StateDB) clearJournalAndRefund() {
 func (s *StateDB) Commit(deleteEmptyObjects bool) (root common.Hash, err error) {
 	defer s.clearJournalAndRefund()
 
+	for addr := range s.journal.dirties {
+		s.stateObjectsDirty[addr] = struct{}{}
+	}
 	// Commit objects to the trie.
 	for addr, stateObject := range s.stateObjects {
 		_, isDirty := s.stateObjectsDirty[addr]
diff --git a/core/state/statedb_test.go b/core/state/statedb_test.go
index d9e3d9b79..340c840f1 100644
--- a/core/state/statedb_test.go
+++ b/core/state/statedb_test.go
@@ -413,11 +413,12 @@ func (s *StateSuite) TestTouchDelete(c *check.C) {
 
 	snapshot := s.state.Snapshot()
 	s.state.AddBalance(common.Address{}, new(big.Int))
-	if len(s.state.stateObjectsDirty) != 1 {
+
+	if len(s.state.journal.dirties) != 1 {
 		c.Fatal("expected one dirty state object")
 	}
 	s.state.RevertToSnapshot(snapshot)
-	if len(s.state.stateObjectsDirty) != 0 {
+	if len(s.state.journal.dirties) != 0 {
 		c.Fatal("expected no dirty state object")
 	}
 }