diff --git a/dma.c b/dma.c
index 07065816..09dc37bb 100644
--- a/dma.c
+++ b/dma.c
@@ -233,9 +233,7 @@ static void
 mt76_dma_tx_cleanup(struct mt76_dev *dev, struct mt76_queue *q, bool flush)
 {
 	struct mt76_queue_entry entry;
-	int done = 0;
 	int last;
-	int tail;
 
 	if (!q || !q->ndesc)
 		return;
@@ -246,39 +244,21 @@ mt76_dma_tx_cleanup(struct mt76_dev *dev, struct mt76_queue *q, bool flush)
 	else
 		last = readl(&q->regs->dma_idx);
 
-	tail = q->tail;
-	while (q->queued - done > 0 && tail != last) {
+	while (q->queued > 0 && q->tail != last) {
 		mt76_dma_tx_cleanup_idx(dev, q, q->tail, &entry);
-		if (entry.skb)
-			dev->drv->tx_complete_skb(dev, &entry);
-		tail = (tail + 1) % q->ndesc;
-		done++;
+		mt76_queue_tx_complete(dev, q, &entry);
 
 		if (entry.txwi) {
 			if (!(dev->drv->drv_flags & MT_DRV_TXWI_NO_FREE))
 				mt76_put_txwi(dev, entry.txwi);
 		}
 
-		if (!flush && tail == last)
+		if (!flush && q->tail == last)
 			last = readl(&q->regs->dma_idx);
 
-		if (done > 16) {
-			spin_lock_bh(&q->lock);
-			q->queued -= done;
-			q->tail = tail;
-			spin_unlock_bh(&q->lock);
-			done = 0;
-		}
 	}
 	spin_unlock_bh(&q->cleanup_lock);
 
-	if (done) {
-		spin_lock_bh(&q->lock);
-		q->queued -= done;
-		q->tail = tail;
-		spin_unlock_bh(&q->lock);
-	}
-
 	if (flush) {
 		spin_lock_bh(&q->lock);
 		mt76_dma_sync_idx(dev, q);