gve: DQO: Add ring allocation and initialization

Allocate the buffer and completion ring structures. Do not populate the
rings yet. That will happen in the respective rx and tx datapath
follow-on patches

Signed-off-by: Bailey Forrest <bcf@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Catherine Sullivan <csully@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/ethernet/google/gve/gve.h

@@ -204,6 +204,10 @@ struct gve_rx_ring {
 	struct gve_queue_resources *q_resources; /* head and tail pointer idx */
 	dma_addr_t q_resources_bus; /* dma address for the queue resources */
 	struct u64_stats_sync statss; /* sync stats for 32bit archs */
+
+	/* head and tail of skb chain for the current packet or NULL if none */
+	struct sk_buff *skb_head;
+	struct sk_buff *skb_tail;
 };
 
 /* A TX desc ring entry */
@@ -816,14 +820,14 @@ void gve_free_page(struct device *dev, struct page *page, dma_addr_t dma,
 netdev_tx_t gve_tx(struct sk_buff *skb, struct net_device *dev);
 bool gve_tx_poll(struct gve_notify_block *block, int budget);
 int gve_tx_alloc_rings(struct gve_priv *priv);
-void gve_tx_free_rings(struct gve_priv *priv);
+void gve_tx_free_rings_gqi(struct gve_priv *priv);
 __be32 gve_tx_load_event_counter(struct gve_priv *priv,
 				 struct gve_tx_ring *tx);
 /* rx handling */
 void gve_rx_write_doorbell(struct gve_priv *priv, struct gve_rx_ring *rx);
 bool gve_rx_poll(struct gve_notify_block *block, int budget);
 int gve_rx_alloc_rings(struct gve_priv *priv);
-void gve_rx_free_rings(struct gve_priv *priv);
+void gve_rx_free_rings_gqi(struct gve_priv *priv);
 bool gve_clean_rx_done(struct gve_rx_ring *rx, int budget,
 		       netdev_features_t feat);
 /* Reset */

drivers/net/ethernet/google/gve/gve_dqo.h

@@ -19,6 +19,24 @@
 netdev_tx_t gve_tx_dqo(struct sk_buff *skb, struct net_device *dev);
 bool gve_tx_poll_dqo(struct gve_notify_block *block, bool do_clean);
 int gve_rx_poll_dqo(struct gve_notify_block *block, int budget);
+int gve_tx_alloc_rings_dqo(struct gve_priv *priv);
+void gve_tx_free_rings_dqo(struct gve_priv *priv);
+int gve_rx_alloc_rings_dqo(struct gve_priv *priv);
+void gve_rx_free_rings_dqo(struct gve_priv *priv);
+int gve_clean_tx_done_dqo(struct gve_priv *priv, struct gve_tx_ring *tx,
+			  struct napi_struct *napi);
+void gve_rx_post_buffers_dqo(struct gve_rx_ring *rx);
+void gve_rx_write_doorbell_dqo(const struct gve_priv *priv, int queue_idx);
+
+static inline void
+gve_tx_put_doorbell_dqo(const struct gve_priv *priv,
+			const struct gve_queue_resources *q_resources, u32 val)
+{
+	u64 index;
+
+	index = be32_to_cpu(q_resources->db_index);
+	iowrite32(val, &priv->db_bar2[index]);
+}
 
 static inline void
 gve_write_irq_doorbell_dqo(const struct gve_priv *priv,

drivers/net/ethernet/google/gve/gve_main.c

@@ -571,13 +571,21 @@ static int gve_create_rings(struct gve_priv *priv)
 	netif_dbg(priv, drv, priv->dev, "created %d rx queues\n",
 		  priv->rx_cfg.num_queues);
 
-	/* Rx data ring has been prefilled with packet buffers at queue
-	 * allocation time.
-	 * Write the doorbell to provide descriptor slots and packet buffers
-	 * to the NIC.
-	 */
-	for (i = 0; i < priv->rx_cfg.num_queues; i++)
-		gve_rx_write_doorbell(priv, &priv->rx[i]);
+	if (gve_is_gqi(priv)) {
+		/* Rx data ring has been prefilled with packet buffers at queue
+		 * allocation time.
+		 *
+		 * Write the doorbell to provide descriptor slots and packet
+		 * buffers to the NIC.
+		 */
+		for (i = 0; i < priv->rx_cfg.num_queues; i++)
+			gve_rx_write_doorbell(priv, &priv->rx[i]);
+	} else {
+		for (i = 0; i < priv->rx_cfg.num_queues; i++) {
+			/* Post buffers and ring doorbell. */
+			gve_rx_post_buffers_dqo(&priv->rx[i]);
+		}
+	}
 
 	return 0;
 }
@@ -606,6 +614,15 @@ static void add_napi_init_sync_stats(struct gve_priv *priv,
 	}
 }
 
+static void gve_tx_free_rings(struct gve_priv *priv)
+{
+	if (gve_is_gqi(priv)) {
+		gve_tx_free_rings_gqi(priv);
+	} else {
+		gve_tx_free_rings_dqo(priv);
+	}
+}
+
 static int gve_alloc_rings(struct gve_priv *priv)
 {
 	int err;
@@ -615,17 +632,26 @@ static int gve_alloc_rings(struct gve_priv *priv)
 			    GFP_KERNEL);
 	if (!priv->tx)
 		return -ENOMEM;
-	err = gve_tx_alloc_rings(priv);
+
+	if (gve_is_gqi(priv))
+		err = gve_tx_alloc_rings(priv);
+	else
+		err = gve_tx_alloc_rings_dqo(priv);
 	if (err)
 		goto free_tx;
+
 	/* Setup rx rings */
 	priv->rx = kvzalloc(priv->rx_cfg.num_queues * sizeof(*priv->rx),
 			    GFP_KERNEL);
 	if (!priv->rx) {
 		err = -ENOMEM;
 		goto free_tx_queue;
 	}
-	err = gve_rx_alloc_rings(priv);
+
+	if (gve_is_gqi(priv))
+		err = gve_rx_alloc_rings(priv);
+	else
+		err = gve_rx_alloc_rings_dqo(priv);
 	if (err)
 		goto free_rx;
 
@@ -670,6 +696,14 @@ static int gve_destroy_rings(struct gve_priv *priv)
 	return 0;
 }
 
+static inline void gve_rx_free_rings(struct gve_priv *priv)
+{
+	if (gve_is_gqi(priv))
+		gve_rx_free_rings_gqi(priv);
+	else
+		gve_rx_free_rings_dqo(priv);
+}
+
 static void gve_free_rings(struct gve_priv *priv)
 {
 	int ntfy_idx;
@@ -869,6 +903,7 @@ static int gve_open(struct net_device *dev)
 	err = gve_alloc_qpls(priv);
 	if (err)
 		return err;
+
 	err = gve_alloc_rings(priv);
 	if (err)
 		goto free_qpls;

drivers/net/ethernet/google/gve/gve_rx.c

@@ -238,7 +238,7 @@ int gve_rx_alloc_rings(struct gve_priv *priv)
 	return err;
 }
 
-void gve_rx_free_rings(struct gve_priv *priv)
+void gve_rx_free_rings_gqi(struct gve_priv *priv)
 {
 	int i;
 

drivers/net/ethernet/google/gve/gve_rx_dqo.c

@@ -16,6 +16,163 @@
 #include <net/ipv6.h>
 #include <net/tcp.h>
 
+static void gve_free_page_dqo(struct gve_priv *priv,
+			      struct gve_rx_buf_state_dqo *bs)
+{
+}
+
+static void gve_rx_free_ring_dqo(struct gve_priv *priv, int idx)
+{
+	struct gve_rx_ring *rx = &priv->rx[idx];
+	struct device *hdev = &priv->pdev->dev;
+	size_t completion_queue_slots;
+	size_t buffer_queue_slots;
+	size_t size;
+	int i;
+
+	completion_queue_slots = rx->dqo.complq.mask + 1;
+	buffer_queue_slots = rx->dqo.bufq.mask + 1;
+
+	gve_rx_remove_from_block(priv, idx);
+
+	if (rx->q_resources) {
+		dma_free_coherent(hdev, sizeof(*rx->q_resources),
+				  rx->q_resources, rx->q_resources_bus);
+		rx->q_resources = NULL;
+	}
+
+	for (i = 0; i < rx->dqo.num_buf_states; i++) {
+		struct gve_rx_buf_state_dqo *bs = &rx->dqo.buf_states[i];
+
+		if (bs->page_info.page)
+			gve_free_page_dqo(priv, bs);
+	}
+
+	if (rx->dqo.bufq.desc_ring) {
+		size = sizeof(rx->dqo.bufq.desc_ring[0]) * buffer_queue_slots;
+		dma_free_coherent(hdev, size, rx->dqo.bufq.desc_ring,
+				  rx->dqo.bufq.bus);
+		rx->dqo.bufq.desc_ring = NULL;
+	}
+
+	if (rx->dqo.complq.desc_ring) {
+		size = sizeof(rx->dqo.complq.desc_ring[0]) *
+			completion_queue_slots;
+		dma_free_coherent(hdev, size, rx->dqo.complq.desc_ring,
+				  rx->dqo.complq.bus);
+		rx->dqo.complq.desc_ring = NULL;
+	}
+
+	kvfree(rx->dqo.buf_states);
+	rx->dqo.buf_states = NULL;
+
+	netif_dbg(priv, drv, priv->dev, "freed rx ring %d\n", idx);
+}
+
+static int gve_rx_alloc_ring_dqo(struct gve_priv *priv, int idx)
+{
+	struct gve_rx_ring *rx = &priv->rx[idx];
+	struct device *hdev = &priv->pdev->dev;
+	size_t size;
+	int i;
+
+	const u32 buffer_queue_slots =
+		priv->options_dqo_rda.rx_buff_ring_entries;
+	const u32 completion_queue_slots = priv->rx_desc_cnt;
+
+	netif_dbg(priv, drv, priv->dev, "allocating rx ring DQO\n");
+
+	memset(rx, 0, sizeof(*rx));
+	rx->gve = priv;
+	rx->q_num = idx;
+	rx->dqo.bufq.mask = buffer_queue_slots - 1;
+	rx->dqo.complq.num_free_slots = completion_queue_slots;
+	rx->dqo.complq.mask = completion_queue_slots - 1;
+	rx->skb_head = NULL;
+	rx->skb_tail = NULL;
+
+	rx->dqo.num_buf_states = min_t(s16, S16_MAX, buffer_queue_slots * 4);
+	rx->dqo.buf_states = kvcalloc(rx->dqo.num_buf_states,
+				      sizeof(rx->dqo.buf_states[0]),
+				      GFP_KERNEL);
+	if (!rx->dqo.buf_states)
+		return -ENOMEM;
+
+	/* Set up linked list of buffer IDs */
+	for (i = 0; i < rx->dqo.num_buf_states - 1; i++)
+		rx->dqo.buf_states[i].next = i + 1;
+
+	rx->dqo.buf_states[rx->dqo.num_buf_states - 1].next = -1;
+	rx->dqo.recycled_buf_states.head = -1;
+	rx->dqo.recycled_buf_states.tail = -1;
+	rx->dqo.used_buf_states.head = -1;
+	rx->dqo.used_buf_states.tail = -1;
+
+	/* Allocate RX completion queue */
+	size = sizeof(rx->dqo.complq.desc_ring[0]) *
+		completion_queue_slots;
+	rx->dqo.complq.desc_ring =
+		dma_alloc_coherent(hdev, size, &rx->dqo.complq.bus, GFP_KERNEL);
+	if (!rx->dqo.complq.desc_ring)
+		goto err;
+
+	/* Allocate RX buffer queue */
+	size = sizeof(rx->dqo.bufq.desc_ring[0]) * buffer_queue_slots;
+	rx->dqo.bufq.desc_ring =
+		dma_alloc_coherent(hdev, size, &rx->dqo.bufq.bus, GFP_KERNEL);
+	if (!rx->dqo.bufq.desc_ring)
+		goto err;
+
+	rx->q_resources = dma_alloc_coherent(hdev, sizeof(*rx->q_resources),
+					     &rx->q_resources_bus, GFP_KERNEL);
+	if (!rx->q_resources)
+		goto err;
+
+	gve_rx_add_to_block(priv, idx);
+
+	return 0;
+
+err:
+	gve_rx_free_ring_dqo(priv, idx);
+	return -ENOMEM;
+}
+
+int gve_rx_alloc_rings_dqo(struct gve_priv *priv)
+{
+	int err = 0;
+	int i;
+
+	for (i = 0; i < priv->rx_cfg.num_queues; i++) {
+		err = gve_rx_alloc_ring_dqo(priv, i);
+		if (err) {
+			netif_err(priv, drv, priv->dev,
+				  "Failed to alloc rx ring=%d: err=%d\n",
+				  i, err);
+			goto err;
+		}
+	}
+
+	return 0;
+
+err:
+	for (i--; i >= 0; i--)
+		gve_rx_free_ring_dqo(priv, i);
+
+	return err;
+}
+
+void gve_rx_free_rings_dqo(struct gve_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->rx_cfg.num_queues; i++)
+		gve_rx_free_ring_dqo(priv, i);
+}
+
+void gve_rx_post_buffers_dqo(struct gve_rx_ring *rx)
+{
+}
+
 int gve_rx_poll_dqo(struct gve_notify_block *block, int budget)
 {
 	u32 work_done = 0;

drivers/net/ethernet/google/gve/gve_tx.c

@@ -256,7 +256,7 @@ int gve_tx_alloc_rings(struct gve_priv *priv)
 	return err;
 }
 
-void gve_tx_free_rings(struct gve_priv *priv)
+void gve_tx_free_rings_gqi(struct gve_priv *priv)
 {
 	int i;