Use channels to trigger flushes in a dedicated goroutine

blockstore/autobatch.go

@@ -18,7 +18,10 @@ type AutobatchBlockstore struct {
 	bufferedBlks   []block.Block
 	addedCids      map[cid.Cid]struct{}
 	bufferedBlksLk sync.Mutex
-	flushLk        sync.Mutex
+	flushCh        chan struct{}
+	flushErr       error
+	shutdownCh     chan struct{}
+	flushCtx       context.Context
 	backingBs      Blockstore
 	bufferCapacity int
 	bufferSize     int
@@ -29,8 +32,11 @@ func NewAutobatch(ctx context.Context, backingBs Blockstore, bufferCapacity int)
 		backingBs:      backingBs,
 		bufferCapacity: bufferCapacity,
 		addedCids:      make(map[cid.Cid]struct{}),
+		flushCtx:       ctx,
 	}
 
+	go bs.flushWorker()
+
 	return bs
 }
 
@@ -42,24 +48,54 @@ func (bs *AutobatchBlockstore) Put(ctx context.Context, blk block.Block) error {
 		bs.addedCids[blk.Cid()] = struct{}{}
 		bs.bufferSize += len(blk.RawData())
 		if bs.bufferSize >= bs.bufferCapacity {
-			// time to flush
-			go bs.Flush(ctx)
+			// signal that a flush is appropriate, may be ignored
+			select {
+			case bs.flushCh <- struct{}{}:
+			default:
+				// do nothing
+			}
 		}
 	}
 	bs.bufferedBlksLk.Unlock()
 	return nil
 }
 
-func (bs *AutobatchBlockstore) Flush(ctx context.Context) {
-	bs.flushLk.Lock()
-	defer bs.flushLk.Unlock()
+func (bs *AutobatchBlockstore) flushWorker() {
+	for {
+		select {
+		case <-bs.flushCh:
+			putErr := bs.doFlush(bs.flushCtx)
+			if putErr != nil {
+				autolog.Errorf("FLUSH ERRORED: %w", putErr)
+				bs.flushErr = xerrors.Errorf("%w, put error: %w", bs.flushErr, putErr)
+			}
+		case <-bs.shutdownCh:
+			return
+		}
+	}
+}
+
+func (bs *AutobatchBlockstore) doFlush(ctx context.Context) error {
 	bs.bufferedBlksLk.Lock()
 	// We do NOT clear addedCids here, because its purpose is to expedite Puts
 	toFlush := bs.bufferedBlks
 	bs.bufferedBlks = []block.Block{}
 	bs.bufferedBlksLk.Unlock()
-	err := bs.backingBs.PutMany(ctx, toFlush)
-	autolog.Errorf("FLUSH ERRORED, maybe async: %w", err)
+	return bs.backingBs.PutMany(ctx, toFlush)
+}
+
+func (bs *AutobatchBlockstore) Flush(ctx context.Context) error {
+	return bs.doFlush(ctx)
+}
+
+func (bs *AutobatchBlockstore) Shutdown(ctx context.Context) error {
+	bs.shutdownCh <- struct{}{}
+	if bs.flushErr != nil {
+		return xerrors.Errorf("flushWorker errored: %w", bs.flushErr)
+	}
+
+	// one last flush in case it's needed
+	return bs.doFlush(ctx)
 }
 
 // May be very slow if the cid queried wasn't in the backingBs at the time of creation of this AutobatchBlockstore

chain/consensus/filcns/upgrades.go

@@ -1298,7 +1298,13 @@ func upgradeActorsV7Common(
 	}
 
 	// Persist the new tree. Blocks until the entire writeStore is in the state blockstore.
-	writeStore.Flush(ctx)
+	if err := writeStore.Flush(ctx); err != nil {
+		return cid.Undef, xerrors.Errorf("failed to flush writestore: %w", err)
+	}
+
+	if err := writeStore.Shutdown(ctx); err != nil {
+		return cid.Undef, xerrors.Errorf("writeStore failed: %w", err)
+	}
 
 	return newRoot, nil
 }