[lunar,linux-azure] net: mana: Add page pool for RX buffers

diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
index 146b4f99a5d8..82ccee55b2b7 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -1411,8 +1411,8 @@  static struct sk_buff *mana_build_skb(struct mana_rxq *rxq, void *buf_va,
 	return skb;
 }
 
-static void mana_rx_skb(void *buf_va, struct mana_rxcomp_oob *cqe,
-			struct mana_rxq *rxq)
+static void mana_rx_skb(void *buf_va, bool from_pool,
+			struct mana_rxcomp_oob *cqe, struct mana_rxq *rxq)
 {
 	struct mana_stats_rx *rx_stats = &rxq->stats;
 	struct net_device *ndev = rxq->ndev;
@@ -1445,6 +1445,9 @@  static void mana_rx_skb(void *buf_va, struct mana_rxcomp_oob *cqe,
 	if (!skb)
 		goto drop;
 
+	if (from_pool)
+		skb_mark_for_recycle(skb);
+
 	skb->dev = napi->dev;
 
 	skb->protocol = eth_type_trans(skb, ndev);
@@ -1495,9 +1498,14 @@  static void mana_rx_skb(void *buf_va, struct mana_rxcomp_oob *cqe,
 	u64_stats_update_end(&rx_stats->syncp);
 
 drop:
-	WARN_ON_ONCE(rxq->xdp_save_va);
-	/* Save for reuse */
-	rxq->xdp_save_va = buf_va;
+	if (from_pool) {
+		page_pool_recycle_direct(rxq->page_pool,
+					 virt_to_head_page(buf_va));
+	} else {
+		WARN_ON_ONCE(rxq->xdp_save_va);
+		/* Save for reuse */
+		rxq->xdp_save_va = buf_va;
+	}
 
 	++ndev->stats.rx_dropped;
 
@@ -1505,11 +1513,13 @@  static void mana_rx_skb(void *buf_va, struct mana_rxcomp_oob *cqe,
 }
 
 static void *mana_get_rxfrag(struct mana_rxq *rxq, struct device *dev,
-			     dma_addr_t *da, bool is_napi)
+			     dma_addr_t *da, bool *from_pool, bool is_napi)
 {
 	struct page *page;
 	void *va;
 
+	*from_pool = false;
+
 	/* Reuse XDP dropped page if available */
 	if (rxq->xdp_save_va) {
 		va = rxq->xdp_save_va;
@@ -1530,17 +1540,22 @@  static void *mana_get_rxfrag(struct mana_rxq *rxq, struct device *dev,
 			return NULL;
 		}
 	} else {
-		page = dev_alloc_page();
+		page = page_pool_dev_alloc_pages(rxq->page_pool);
 		if (!page)
 			return NULL;
 
+		*from_pool = true;
 		va = page_to_virt(page);
 	}
 
 	*da = dma_map_single(dev, va + rxq->headroom, rxq->datasize,
 			     DMA_FROM_DEVICE);
 	if (dma_mapping_error(dev, *da)) {
-		put_page(virt_to_head_page(va));
+		if (*from_pool)
+			page_pool_put_full_page(rxq->page_pool, page, false);
+		else
+			put_page(virt_to_head_page(va));
+
 		return NULL;
 	}
 
@@ -1549,21 +1564,25 @@  static void *mana_get_rxfrag(struct mana_rxq *rxq, struct device *dev,
 
 /* Allocate frag for rx buffer, and save the old buf */
 static void mana_refill_rx_oob(struct device *dev, struct mana_rxq *rxq,
-			       struct mana_recv_buf_oob *rxoob, void **old_buf)
+			       struct mana_recv_buf_oob *rxoob, void **old_buf,
+			       bool *old_fp)
 {
+	bool from_pool;
 	dma_addr_t da;
 	void *va;
 
-	va = mana_get_rxfrag(rxq, dev, &da, true);
+	va = mana_get_rxfrag(rxq, dev, &da, &from_pool, true);
 	if (!va)
 		return;
 
 	dma_unmap_single(dev, rxoob->sgl[0].address, rxq->datasize,
 			 DMA_FROM_DEVICE);
 	*old_buf = rxoob->buf_va;
+	*old_fp = rxoob->from_pool;
 
 	rxoob->buf_va = va;
 	rxoob->sgl[0].address = da;
+	rxoob->from_pool = from_pool;
 }
 
 static void mana_process_rx_cqe(struct mana_rxq *rxq, struct mana_cq *cq,
@@ -1577,6 +1596,7 @@  static void mana_process_rx_cqe(struct mana_rxq *rxq, struct mana_cq *cq,
 	struct device *dev = gc->dev;
 	void *old_buf = NULL;
 	u32 curr, pktlen;
+	bool old_fp;
 
 	apc = netdev_priv(ndev);
 
@@ -1619,12 +1639,12 @@  static void mana_process_rx_cqe(struct mana_rxq *rxq, struct mana_cq *cq,
 	rxbuf_oob = &rxq->rx_oobs[curr];
 	WARN_ON_ONCE(rxbuf_oob->wqe_inf.wqe_size_in_bu != 1);
 
-	mana_refill_rx_oob(dev, rxq, rxbuf_oob, &old_buf);
+	mana_refill_rx_oob(dev, rxq, rxbuf_oob, &old_buf, &old_fp);
 
 	/* Unsuccessful refill will have old_buf == NULL.
 	 * In this case, mana_rx_skb() will drop the packet.
 	 */
-	mana_rx_skb(old_buf, oob, rxq);
+	mana_rx_skb(old_buf, old_fp, oob, rxq);
 
 drop:
 	mana_move_wq_tail(rxq->gdma_rq, rxbuf_oob->wqe_inf.wqe_size_in_bu);
@@ -1884,6 +1904,7 @@  static void mana_destroy_rxq(struct mana_port_context *apc,
 	struct mana_recv_buf_oob *rx_oob;
 	struct device *dev = gc->dev;
 	struct napi_struct *napi;
+	struct page *page;
 	int i;
 
 	if (!rxq)
@@ -1916,10 +1937,18 @@  static void mana_destroy_rxq(struct mana_port_context *apc,
 		dma_unmap_single(dev, rx_oob->sgl[0].address,
 				 rx_oob->sgl[0].size, DMA_FROM_DEVICE);
 
-		put_page(virt_to_head_page(rx_oob->buf_va));
+		page = virt_to_head_page(rx_oob->buf_va);
+
+		if (rx_oob->from_pool)
+			page_pool_put_full_page(rxq->page_pool, page, false);
+		else
+			put_page(page);
+
 		rx_oob->buf_va = NULL;
 	}
 
+	page_pool_destroy(rxq->page_pool);
+
 	if (rxq->gdma_rq)
 		mana_gd_destroy_queue(gc, rxq->gdma_rq);
 
@@ -1930,18 +1959,20 @@  static int mana_fill_rx_oob(struct mana_recv_buf_oob *rx_oob, u32 mem_key,
 			    struct mana_rxq *rxq, struct device *dev)
 {
 	struct mana_port_context *mpc = netdev_priv(rxq->ndev);
+	bool from_pool = false;
 	dma_addr_t da;
 	void *va;
 
 	if (mpc->rxbufs_pre)
 		va = mana_get_rxbuf_pre(rxq, &da);
 	else
-		va = mana_get_rxfrag(rxq, dev, &da, false);
+		va = mana_get_rxfrag(rxq, dev, &da, &from_pool, false);
 
 	if (!va)
 		return -ENOMEM;
 
 	rx_oob->buf_va = va;
+	rx_oob->from_pool = from_pool;
 
 	rx_oob->sgl[0].address = da;
 	rx_oob->sgl[0].size = rxq->datasize;
@@ -2011,6 +2042,25 @@  static int mana_push_wqe(struct mana_rxq *rxq)
 	return 0;
 }
 
+static int mana_create_page_pool(struct mana_rxq *rxq, struct gdma_context *gc)
+{
+	struct page_pool_params pprm = {};
+	int ret;
+
+	pprm.pool_size = RX_BUFFERS_PER_QUEUE;
+	pprm.nid = gc->numa_node;
+
+	rxq->page_pool = page_pool_create(&pprm);
+
+	if (IS_ERR(rxq->page_pool)) {
+		ret = PTR_ERR(rxq->page_pool);
+		rxq->page_pool = NULL;
+		return ret;
+	}
+
+	return 0;
+}
+
 static struct mana_rxq *mana_create_rxq(struct mana_port_context *apc,
 					u32 rxq_idx, struct mana_eq *eq,
 					struct net_device *ndev)
@@ -2040,6 +2090,13 @@  static struct mana_rxq *mana_create_rxq(struct mana_port_context *apc,
 	mana_get_rxbuf_cfg(ndev->mtu, &rxq->datasize, &rxq->alloc_size,
 			   &rxq->headroom);
 
+	/* Create page pool for RX queue */
+	err = mana_create_page_pool(rxq, gc);
+	if (err) {
+		netdev_err(ndev, "Create page pool err:%d\n", err);
+		goto out;
+	}
+
 	err = mana_alloc_rx_wqe(apc, rxq, &rq_size, &cq_size);
 	if (err)
 		goto out;
@@ -2111,8 +2168,8 @@  static struct mana_rxq *mana_create_rxq(struct mana_port_context *apc,
 
 	WARN_ON(xdp_rxq_info_reg(&rxq->xdp_rxq, ndev, rxq_idx,
 				 cq->napi.napi_id));
-	WARN_ON(xdp_rxq_info_reg_mem_model(&rxq->xdp_rxq,
-					   MEM_TYPE_PAGE_SHARED, NULL));
+	WARN_ON(xdp_rxq_info_reg_mem_model(&rxq->xdp_rxq, MEM_TYPE_PAGE_POOL,
+					   rxq->page_pool));
 
 	napi_enable(&cq->napi);
 
diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h
index 9eef19972845..4a7fc0d6bbc9 100644
--- a/include/net/mana/mana.h
+++ b/include/net/mana/mana.h
@@ -280,6 +280,7 @@  struct mana_recv_buf_oob {
 	struct gdma_wqe_request wqe_req;
 
 	void *buf_va;
+	bool from_pool; /* allocated from a page pool */
 
 	/* SGL of the buffer going to be sent has part of the work request. */
 	u32 num_sge;
@@ -330,6 +331,8 @@  struct mana_rxq {
 	bool xdp_flush;
 	int xdp_rc; /* XDP redirect return code */
 
+	struct page_pool *page_pool;
+
 	/* MUST BE THE LAST MEMBER:
 	 * Each receive buffer has an associated mana_recv_buf_oob.
 	 */