eth, eth/downloader: make synchronize thread safe

This commit is contained in:
Péter Szilágyi 2015-05-07 21:07:20 +03:00
parent 43901c9282
commit 9d188f73b5
5 changed files with 22 additions and 82 deletions

@ -68,8 +68,7 @@ type Downloader struct {
getBlock getBlockFn getBlock getBlockFn
// Status // Status
fetchingHashes int32 synchronizing int32
downloadingBlocks int32
// Channels // Channels
newPeerCh chan *peer newPeerCh chan *peer
@ -120,43 +119,26 @@ func (d *Downloader) UnregisterPeer(id string) {
delete(d.peers, id) delete(d.peers, id)
} }
// SynchroniseWithPeer will select the peer and use it for synchronizing. If an empty string is given // Synchronize will select the peer and use it for synchronizing. If an empty string is given
// it will use the best peer possible and synchronize if it's TD is higher than our own. If any of the // it will use the best peer possible and synchronize if it's TD is higher than our own. If any of the
// checks fail an error will be returned. This method is synchronous // checks fail an error will be returned. This method is synchronous
func (d *Downloader) Synchronise(id string, hash common.Hash) error { func (d *Downloader) Synchronize(id string, hash common.Hash) error {
// Make sure it's doing neither. Once done we can restart the // Make sure only one goroutine is ever allowed past this point at once
// downloading process if the TD is higher. For now just get on if !atomic.CompareAndSwapInt32(&d.synchronizing, 0, 1) {
// with whatever is going on. This prevents unnecessary switching. return nil
if d.isBusy() {
return errBusy
} }
defer atomic.StoreInt32(&d.synchronizing, 0)
// When a synchronization attempt is made while the queue still // Abort if the queue still contains some leftover data
// contains items we abort the sync attempt if _, cached := d.queue.Size(); cached > 0 {
if done, pend := d.queue.Size(); done+pend > 0 {
return errPendingQueue return errPendingQueue
} }
// Retrieve the origin peer and initiate the downloading process
// Fetch the peer using the id or throw an error if the peer couldn't be found
p := d.peers[id] p := d.peers[id]
if p == nil { if p == nil {
return errUnknownPeer return errUnknownPeer
} }
return d.getFromPeer(p, hash, false)
// Get the hash from the peer and initiate the downloading progress.
err := d.getFromPeer(p, hash, false)
if err != nil {
return err
}
return nil
}
// Done lets the downloader know that whatever previous hashes were taken
// are processed. If the block count reaches zero and done is called
// we reset the queue for the next batch of incoming hashes and blocks.
func (d *Downloader) Done() {
d.queue.Done()
} }
// TakeBlocks takes blocks from the queue and yields them to the blockTaker handler // TakeBlocks takes blocks from the queue and yields them to the blockTaker handler
@ -176,6 +158,7 @@ func (d *Downloader) Has(hash common.Hash) bool {
} }
func (d *Downloader) getFromPeer(p *peer, hash common.Hash, ignoreInitial bool) (err error) { func (d *Downloader) getFromPeer(p *peer, hash common.Hash, ignoreInitial bool) (err error) {
d.activePeer = p.id d.activePeer = p.id
defer func() { defer func() {
// reset on error // reset on error
@ -184,7 +167,7 @@ func (d *Downloader) getFromPeer(p *peer, hash common.Hash, ignoreInitial bool)
} }
}() }()
glog.V(logger.Detail).Infoln("Synchronising with the network using:", p.id) glog.V(logger.Debug).Infoln("Synchronizing with the network using:", p.id)
// Start the fetcher. This will block the update entirely // Start the fetcher. This will block the update entirely
// interupts need to be send to the appropriate channels // interupts need to be send to the appropriate channels
// respectively. // respectively.
@ -200,20 +183,13 @@ func (d *Downloader) getFromPeer(p *peer, hash common.Hash, ignoreInitial bool)
return err return err
} }
glog.V(logger.Detail).Infoln("Sync completed") glog.V(logger.Debug).Infoln("Synchronization completed")
return nil return nil
} }
// XXX Make synchronous // XXX Make synchronous
func (d *Downloader) startFetchingHashes(p *peer, h common.Hash, ignoreInitial bool) error { func (d *Downloader) startFetchingHashes(p *peer, h common.Hash, ignoreInitial bool) error {
atomic.StoreInt32(&d.fetchingHashes, 1)
defer atomic.StoreInt32(&d.fetchingHashes, 0)
if d.queue.Has(h) { // TODO: Is this possible? Shouldn't queue be empty for startFetchingHashes to be even called?
return errAlreadyInPool
}
glog.V(logger.Debug).Infof("Downloading hashes (%x) from %s", h[:4], p.id) glog.V(logger.Debug).Infof("Downloading hashes (%x) from %s", h[:4], p.id)
start := time.Now() start := time.Now()
@ -312,10 +288,8 @@ out:
} }
func (d *Downloader) startFetchingBlocks(p *peer) error { func (d *Downloader) startFetchingBlocks(p *peer) error {
glog.V(logger.Detail).Infoln("Downloading", d.queue.Pending(), "block(s)") glog.V(logger.Debug).Infoln("Downloading", d.queue.Pending(), "block(s)")
atomic.StoreInt32(&d.downloadingBlocks, 1)
defer atomic.StoreInt32(&d.downloadingBlocks, 0)
// Defer the peer reset. This will empty the peer requested set // Defer the peer reset. This will empty the peer requested set
// and makes sure there are no lingering peers with an incorrect // and makes sure there are no lingering peers with an incorrect
// state // state
@ -439,19 +413,3 @@ func (d *Downloader) AddHashes(id string, hashes []common.Hash) error {
return nil return nil
} }
func (d *Downloader) isFetchingHashes() bool {
return atomic.LoadInt32(&d.fetchingHashes) == 1
}
func (d *Downloader) isDownloadingBlocks() bool {
return atomic.LoadInt32(&d.downloadingBlocks) == 1
}
func (d *Downloader) isBusy() bool {
return d.isFetchingHashes() || d.isDownloadingBlocks()
}
func (d *Downloader) IsBusy() bool {
return d.isBusy()
}

@ -61,7 +61,7 @@ func newTester(t *testing.T, hashes []common.Hash, blocks map[common.Hash]*types
func (dl *downloadTester) sync(peerId string, hash common.Hash) error { func (dl *downloadTester) sync(peerId string, hash common.Hash) error {
dl.activePeerId = peerId dl.activePeerId = peerId
return dl.downloader.Synchronise(peerId, hash) return dl.downloader.Synchronize(peerId, hash)
} }
func (dl *downloadTester) hasBlock(hash common.Hash) bool { func (dl *downloadTester) hasBlock(hash common.Hash) bool {

@ -63,16 +63,6 @@ func (q *queue) Reset() {
q.blockCache = nil q.blockCache = nil
} }
// Done checks if all the downloads have been retrieved, wiping the queue.
func (q *queue) Done() {
q.lock.Lock()
defer q.lock.Unlock()
if len(q.blockCache) == 0 {
q.Reset()
}
}
// Size retrieves the number of hashes in the queue, returning separately for // Size retrieves the number of hashes in the queue, returning separately for
// pending and already downloaded. // pending and already downloaded.
func (q *queue) Size() (int, int) { func (q *queue) Size() (int, int) {

@ -307,7 +307,7 @@ func (self *ProtocolManager) handleMsg(p *peer) error {
// Attempt to insert the newly received by checking if the parent exists. // Attempt to insert the newly received by checking if the parent exists.
// if the parent exists we process the block and propagate to our peers // if the parent exists we process the block and propagate to our peers
// otherwise synchronise with the peer // otherwise synchronize with the peer
if self.chainman.HasBlock(request.Block.ParentHash()) { if self.chainman.HasBlock(request.Block.ParentHash()) {
if _, err := self.chainman.InsertChain(types.Blocks{request.Block}); err != nil { if _, err := self.chainman.InsertChain(types.Blocks{request.Block}); err != nil {
glog.V(logger.Error).Infoln("removed peer (", p.id, ") due to block error") glog.V(logger.Error).Infoln("removed peer (", p.id, ") due to block error")
@ -324,7 +324,7 @@ func (self *ProtocolManager) handleMsg(p *peer) error {
} }
self.BroadcastBlock(hash, request.Block) self.BroadcastBlock(hash, request.Block)
} else { } else {
go self.synchronise(p) go self.synchronize(p)
} }
default: default:
return errResp(ErrInvalidMsgCode, "%v", msg.Code) return errResp(ErrInvalidMsgCode, "%v", msg.Code)

@ -32,14 +32,14 @@ func (pm *ProtocolManager) update() {
} }
itimer.Stop() itimer.Stop()
go pm.synchronise(peer) go pm.synchronize(peer)
case <-itimer.C: case <-itimer.C:
// The timer will make sure that the downloader keeps an active state // The timer will make sure that the downloader keeps an active state
// in which it attempts to always check the network for highest td peers // in which it attempts to always check the network for highest td peers
// Either select the peer or restart the timer if no peers could // Either select the peer or restart the timer if no peers could
// be selected. // be selected.
if peer := getBestPeer(pm.peers); peer != nil { if peer := getBestPeer(pm.peers); peer != nil {
go pm.synchronise(peer) go pm.synchronize(peer)
} else { } else {
itimer.Reset(5 * time.Second) itimer.Reset(5 * time.Second)
} }
@ -63,7 +63,6 @@ func (pm *ProtocolManager) processBlocks() error {
if len(blocks) == 0 { if len(blocks) == 0 {
return nil return nil
} }
defer pm.downloader.Done()
glog.V(logger.Debug).Infof("Inserting chain with %d blocks (#%v - #%v)\n", len(blocks), blocks[0].Number(), blocks[len(blocks)-1].Number()) glog.V(logger.Debug).Infof("Inserting chain with %d blocks (#%v - #%v)\n", len(blocks), blocks[0].Number(), blocks[len(blocks)-1].Number())
@ -78,26 +77,19 @@ func (pm *ProtocolManager) processBlocks() error {
return nil return nil
} }
func (pm *ProtocolManager) synchronise(peer *peer) { func (pm *ProtocolManager) synchronize(peer *peer) {
// Make sure the peer's TD is higher than our own. If not drop. // Make sure the peer's TD is higher than our own. If not drop.
if peer.td.Cmp(pm.chainman.Td()) <= 0 { if peer.td.Cmp(pm.chainman.Td()) <= 0 {
return return
} }
// Check downloader if it's busy so it doesn't show the sync message
// for every attempty
if pm.downloader.IsBusy() {
return
}
// FIXME if we have the hash in our chain and the TD of the peer is // FIXME if we have the hash in our chain and the TD of the peer is
// much higher than ours, something is wrong with us or the peer. // much higher than ours, something is wrong with us or the peer.
// Check if the hash is on our own chain // Check if the hash is on our own chain
if pm.chainman.HasBlock(peer.recentHash) { if pm.chainman.HasBlock(peer.recentHash) {
return return
} }
// Get the hashes from the peer (synchronously) // Get the hashes from the peer (synchronously)
err := pm.downloader.Synchronise(peer.id, peer.recentHash) err := pm.downloader.Synchronize(peer.id, peer.recentHash)
if err != nil && err == downloader.ErrBadPeer { if err != nil && err == downloader.ErrBadPeer {
glog.V(logger.Debug).Infoln("removed peer from peer set due to bad action") glog.V(logger.Debug).Infoln("removed peer from peer set due to bad action")
pm.removePeer(peer) pm.removePeer(peer)