les: historical data garbage collection (#19570)

This change introduces garbage collection for the light client. Historical
chain data is deleted periodically. If you want to disable the GC, use
the --light.nopruning flag.

trie/database.go

@@ -693,7 +693,7 @@ func (db *Database) Cap(limit common.StorageSize) error {
 //
 // Note, this method is a non-synchronized mutator. It is unsafe to call this
 // concurrently with other mutators.
-func (db *Database) Commit(node common.Hash, report bool) error {
+func (db *Database) Commit(node common.Hash, report bool, callback func(common.Hash)) error {
 	// Create a database batch to flush persistent data out. It is important that
 	// outside code doesn't see an inconsistent state (referenced data removed from
 	// memory cache during commit but not yet in persistent storage). This is ensured
@@ -732,7 +732,7 @@ func (db *Database) Commit(node common.Hash, report bool) error {
 	nodes, storage := len(db.dirties), db.dirtiesSize
 
 	uncacher := &cleaner{db}
-	if err := db.commit(node, batch, uncacher); err != nil {
+	if err := db.commit(node, batch, uncacher, callback); err != nil {
 		log.Error("Failed to commit trie from trie database", "err", err)
 		return err
 	}
@@ -771,7 +771,7 @@ func (db *Database) Commit(node common.Hash, report bool) error {
 }
 
 // commit is the private locked version of Commit.
-func (db *Database) commit(hash common.Hash, batch ethdb.Batch, uncacher *cleaner) error {
+func (db *Database) commit(hash common.Hash, batch ethdb.Batch, uncacher *cleaner, callback func(common.Hash)) error {
 	// If the node does not exist, it's a previously committed node
 	node, ok := db.dirties[hash]
 	if !ok {
@@ -780,7 +780,7 @@ func (db *Database) commit(hash common.Hash, batch ethdb.Batch, uncacher *cleane
 	var err error
 	node.forChilds(func(child common.Hash) {
 		if err == nil {
-			err = db.commit(child, batch, uncacher)
+			err = db.commit(child, batch, uncacher, callback)
 		}
 	})
 	if err != nil {
@@ -789,6 +789,9 @@ func (db *Database) commit(hash common.Hash, batch ethdb.Batch, uncacher *cleane
 	if err := batch.Put(hash[:], node.rlp()); err != nil {
 		return err
 	}
+	if callback != nil {
+		callback(hash)
+	}
 	// If we've reached an optimal batch size, commit and start over
 	if batch.ValueSize() >= ethdb.IdealBatchSize {
 		if err := batch.Write(); err != nil {

trie/iterator_test.go

@@ -301,7 +301,7 @@ func testIteratorContinueAfterError(t *testing.T, memonly bool) {
 	}
 	tr.Commit(nil)
 	if !memonly {
-		triedb.Commit(tr.Hash(), true)
+		triedb.Commit(tr.Hash(), true, nil)
 	}
 	wantNodeCount := checkIteratorNoDups(t, tr.NodeIterator(nil), nil)
 
@@ -392,7 +392,7 @@ func testIteratorContinueAfterSeekError(t *testing.T, memonly bool) {
 	}
 	root, _ := ctr.Commit(nil)
 	if !memonly {
-		triedb.Commit(root, true)
+		triedb.Commit(root, true, nil)
 	}
 	barNodeHash := common.HexToHash("05041990364eb72fcb1127652ce40d8bab765f2bfe53225b1170d276cc101c2e")
 	var (

trie/trie_test.go

@@ -88,7 +88,7 @@ func testMissingNode(t *testing.T, memonly bool) {
 	updateString(trie, "123456", "asdfasdfasdfasdfasdfasdfasdfasdf")
 	root, _ := trie.Commit(nil)
 	if !memonly {
-		triedb.Commit(root, true)
+		triedb.Commit(root, true, nil)
 	}
 
 	trie, _ = New(root, triedb)