diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index 7bada4e8460b..9f0561b67c12 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -355,16 +355,15 @@ static u32 xsk_tx_peek_release_fallback(struct xsk_buff_pool *pool, u32 max_entr
 	return nb_pkts;
 }
 
-u32 xsk_tx_peek_release_desc_batch(struct xsk_buff_pool *pool, u32 max_entries)
+u32 xsk_tx_peek_release_desc_batch(struct xsk_buff_pool *pool, u32 nb_pkts)
 {
 	struct xdp_sock *xs;
-	u32 nb_pkts;
 
 	rcu_read_lock();
 	if (!list_is_singular(&pool->xsk_tx_list)) {
 		/* Fallback to the non-batched version */
 		rcu_read_unlock();
-		return xsk_tx_peek_release_fallback(pool, max_entries);
+		return xsk_tx_peek_release_fallback(pool, nb_pkts);
 	}
 
 	xs = list_first_or_null_rcu(&pool->xsk_tx_list, struct xdp_sock, tx_list);
@@ -373,12 +372,7 @@ u32 xsk_tx_peek_release_desc_batch(struct xsk_buff_pool *pool, u32 max_entries)
 		goto out;
 	}
 
-	max_entries = xskq_cons_nb_entries(xs->tx, max_entries);
-	nb_pkts = xskq_cons_read_desc_batch(xs->tx, pool, max_entries);
-	if (!nb_pkts) {
-		xs->tx->queue_empty_descs++;
-		goto out;
-	}
+	nb_pkts = xskq_cons_nb_entries(xs->tx, nb_pkts);
 
 	/* This is the backpressure mechanism for the Tx path. Try to
 	 * reserve space in the completion queue for all packets, but
@@ -386,12 +380,18 @@ u32 xsk_tx_peek_release_desc_batch(struct xsk_buff_pool *pool, u32 max_entries)
 	 * packets. This avoids having to implement any buffering in
 	 * the Tx path.
 	 */
-	nb_pkts = xskq_prod_reserve_addr_batch(pool->cq, pool->tx_descs, nb_pkts);
+	nb_pkts = xskq_prod_nb_free(pool->cq, nb_pkts);
 	if (!nb_pkts)
 		goto out;
 
-	xskq_cons_release_n(xs->tx, max_entries);
+	nb_pkts = xskq_cons_read_desc_batch(xs->tx, pool, nb_pkts);
+	if (!nb_pkts) {
+		xs->tx->queue_empty_descs++;
+		goto out;
+	}
+
 	__xskq_cons_release(xs->tx);
+	xskq_prod_write_addr_batch(pool->cq, pool->tx_descs, nb_pkts);
 	xs->sk.sk_write_space(&xs->sk);
 
 out:
diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index fb20bf7207cf..c6fb6b763658 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -205,6 +205,11 @@ static inline bool xskq_cons_read_desc(struct xsk_queue *q,
 	return false;
 }
 
+static inline void xskq_cons_release_n(struct xsk_queue *q, u32 cnt)
+{
+	q->cached_cons += cnt;
+}
+
 static inline u32 xskq_cons_read_desc_batch(struct xsk_queue *q, struct xsk_buff_pool *pool,
 					    u32 max)
 {
@@ -226,6 +231,8 @@ static inline u32 xskq_cons_read_desc_batch(struct xsk_queue *q, struct xsk_buff
 		cached_cons++;
 	}
 
+	/* Release valid plus any invalid entries */
+	xskq_cons_release_n(q, cached_cons - q->cached_cons);
 	return nb_entries;
 }
 
@@ -291,11 +298,6 @@ static inline void xskq_cons_release(struct xsk_queue *q)
 	q->cached_cons++;
 }
 
-static inline void xskq_cons_release_n(struct xsk_queue *q, u32 cnt)
-{
-	q->cached_cons += cnt;
-}
-
 static inline u32 xskq_cons_present_entries(struct xsk_queue *q)
 {
 	/* No barriers needed since data is not accessed */
@@ -350,21 +352,17 @@ static inline int xskq_prod_reserve_addr(struct xsk_queue *q, u64 addr)
 	return 0;
 }
 
-static inline u32 xskq_prod_reserve_addr_batch(struct xsk_queue *q, struct xdp_desc *descs,
-					       u32 max)
+static inline void xskq_prod_write_addr_batch(struct xsk_queue *q, struct xdp_desc *descs,
+					      u32 nb_entries)
 {
 	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
-	u32 nb_entries, i, cached_prod;
-
-	nb_entries = xskq_prod_nb_free(q, max);
+	u32 i, cached_prod;
 
 	/* A, matches D */
 	cached_prod = q->cached_prod;
 	for (i = 0; i < nb_entries; i++)
 		ring->desc[cached_prod++ & q->ring_mask] = descs[i].addr;
 	q->cached_prod = cached_prod;
-
-	return nb_entries;
 }
 
 static inline int xskq_prod_reserve_desc(struct xsk_queue *q,