octeon_ep_vf: update tx/rx stats locally for persistence

Update tx/rx stats locally, so that ndo_get_stats64()
can use that and not rely on per queue resources to obtain statistics.
The latter used to cause race conditions when the device stopped.

Signed-off-by: Shinas Rasheed <srasheed@marvell.com>
Link: https://patch.msgid.link/20250117094653.2588578-5-srasheed@marvell.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Author: shinas-marvell

drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_ethtool.c

@@ -114,12 +114,9 @@ static void octep_vf_get_ethtool_stats(struct net_device *netdev,
 	iface_tx_stats = &oct->iface_tx_stats;
 	iface_rx_stats = &oct->iface_rx_stats;
 
-	for (q = 0; q < oct->num_oqs; q++) {
-		struct octep_vf_iq *iq = oct->iq[q];
-		struct octep_vf_oq *oq = oct->oq[q];
-
-		tx_busy_errors += iq->stats.tx_busy;
-		rx_alloc_errors += oq->stats.alloc_failures;
+	for (q = 0; q < OCTEP_VF_MAX_QUEUES; q++) {
+		tx_busy_errors += oct->stats_iq[q].tx_busy;
+		rx_alloc_errors += oct->stats_oq[q].alloc_failures;
 	}
 	i = 0;
 	data[i++] = rx_alloc_errors;
@@ -134,22 +131,18 @@ static void octep_vf_get_ethtool_stats(struct net_device *netdev,
 	data[i++] = iface_rx_stats->dropped_octets_fifo_full;
 
 	/* Per Tx Queue stats */
-	for (q = 0; q < oct->num_iqs; q++) {
-		struct octep_vf_iq *iq = oct->iq[q];
-
-		data[i++] = iq->stats.instr_posted;
-		data[i++] = iq->stats.instr_completed;
-		data[i++] = iq->stats.bytes_sent;
-		data[i++] = iq->stats.tx_busy;
+	for (q = 0; q < OCTEP_VF_MAX_QUEUES; q++) {
+		data[i++] = oct->stats_iq[q].instr_posted;
+		data[i++] = oct->stats_iq[q].instr_completed;
+		data[i++] = oct->stats_iq[q].bytes_sent;
+		data[i++] = oct->stats_iq[q].tx_busy;
 	}
 
 	/* Per Rx Queue stats */
 	for (q = 0; q < oct->num_oqs; q++) {
-		struct octep_vf_oq *oq = oct->oq[q];
-
-		data[i++] = oq->stats.packets;
-		data[i++] = oq->stats.bytes;
-		data[i++] = oq->stats.alloc_failures;
+		data[i++] = oct->stats_oq[q].packets;
+		data[i++] = oct->stats_oq[q].bytes;
+		data[i++] = oct->stats_oq[q].alloc_failures;
 	}
 }
 

drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.c

@@ -574,7 +574,7 @@ static int octep_vf_iq_full_check(struct octep_vf_iq *iq)
 		  * caused queues to get re-enabled after
 		  * being stopped
 		  */
-		iq->stats.restart_cnt++;
+		iq->stats->restart_cnt++;
 		fallthrough;
 	case 1: /* Queue left enabled, since IQ is not yet full*/
 		return 0;
@@ -731,7 +731,7 @@ static netdev_tx_t octep_vf_start_xmit(struct sk_buff *skb,
 	/* Flush the hw descriptors before writing to doorbell */
 	smp_wmb();
 	writel(iq->fill_cnt, iq->doorbell_reg);
-	iq->stats.instr_posted += iq->fill_cnt;
+	iq->stats->instr_posted += iq->fill_cnt;
 	iq->fill_cnt = 0;
 	return NETDEV_TX_OK;
 }
@@ -786,14 +786,11 @@ static void octep_vf_get_stats64(struct net_device *netdev,
 	tx_bytes = 0;
 	rx_packets = 0;
 	rx_bytes = 0;
-	for (q = 0; q < oct->num_oqs; q++) {
-		struct octep_vf_iq *iq = oct->iq[q];
-		struct octep_vf_oq *oq = oct->oq[q];
-
-		tx_packets += iq->stats.instr_completed;
-		tx_bytes += iq->stats.bytes_sent;
-		rx_packets += oq->stats.packets;
-		rx_bytes += oq->stats.bytes;
+	for (q = 0; q < OCTEP_VF_MAX_QUEUES; q++) {
+		tx_packets += oct->stats_iq[q].instr_completed;
+		tx_bytes += oct->stats_iq[q].bytes_sent;
+		rx_packets += oct->stats_oq[q].packets;
+		rx_bytes += oct->stats_oq[q].bytes;
 	}
 	stats->tx_packets = tx_packets;
 	stats->tx_bytes = tx_bytes;

drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.h

@@ -246,11 +246,17 @@ struct octep_vf_device {
 	/* Pointers to Octeon Tx queues */
 	struct octep_vf_iq *iq[OCTEP_VF_MAX_IQ];
 
+	/* Per iq stats */
+	struct octep_vf_iq_stats stats_iq[OCTEP_VF_MAX_IQ];
+
 	/* Rx queues (OQ: Output Queue) */
 	u16 num_oqs;
 	/* Pointers to Octeon Rx queues */
 	struct octep_vf_oq *oq[OCTEP_VF_MAX_OQ];
 
+	/* Per oq stats */
+	struct octep_vf_oq_stats stats_oq[OCTEP_VF_MAX_OQ];
+
 	/* Hardware port number of the PCIe interface */
 	u16 pcie_port;
 

drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_rx.c

@@ -87,7 +87,7 @@ static int octep_vf_oq_refill(struct octep_vf_device *oct, struct octep_vf_oq *o
 		page = dev_alloc_page();
 		if (unlikely(!page)) {
 			dev_err(oq->dev, "refill: rx buffer alloc failed\n");
-			oq->stats.alloc_failures++;
+			oq->stats->alloc_failures++;
 			break;
 		}
 
@@ -98,7 +98,7 @@ static int octep_vf_oq_refill(struct octep_vf_device *oct, struct octep_vf_oq *o
 				"OQ-%d buffer refill: DMA mapping error!\n",
 				oq->q_no);
 			put_page(page);
-			oq->stats.alloc_failures++;
+			oq->stats->alloc_failures++;
 			break;
 		}
 		oq->buff_info[refill_idx].page = page;
@@ -134,6 +134,7 @@ static int octep_vf_setup_oq(struct octep_vf_device *oct, int q_no)
 	oq->netdev = oct->netdev;
 	oq->dev = &oct->pdev->dev;
 	oq->q_no = q_no;
+	oq->stats = &oct->stats_oq[q_no];
 	oq->max_count = CFG_GET_OQ_NUM_DESC(oct->conf);
 	oq->ring_size_mask = oq->max_count - 1;
 	oq->buffer_size = CFG_GET_OQ_BUF_SIZE(oct->conf);
@@ -458,8 +459,8 @@ static int __octep_vf_oq_process_rx(struct octep_vf_device *oct,
 
 	oq->host_read_idx = read_idx;
 	oq->refill_count += desc_used;
-	oq->stats.packets += pkt;
-	oq->stats.bytes += rx_bytes;
+	oq->stats->packets += pkt;
+	oq->stats->bytes += rx_bytes;
 
 	return pkt;
 }

drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_rx.h

@@ -187,7 +187,7 @@ struct octep_vf_oq {
 	u8 __iomem *pkts_sent_reg;
 
 	/* Statistics for this OQ. */
-	struct octep_vf_oq_stats stats;
+	struct octep_vf_oq_stats *stats;
 
 	/* Packets pending to be processed */
 	u32 pkts_pending;

drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_tx.c

@@ -82,9 +82,9 @@ int octep_vf_iq_process_completions(struct octep_vf_iq *iq, u16 budget)
 	}
 
 	iq->pkts_processed += compl_pkts;
-	iq->stats.instr_completed += compl_pkts;
-	iq->stats.bytes_sent += compl_bytes;
-	iq->stats.sgentry_sent += compl_sg;
+	iq->stats->instr_completed += compl_pkts;
+	iq->stats->bytes_sent += compl_bytes;
+	iq->stats->sgentry_sent += compl_sg;
 	iq->flush_index = fi;
 
 	netif_subqueue_completed_wake(iq->netdev, iq->q_no, compl_pkts,
@@ -186,6 +186,7 @@ static int octep_vf_setup_iq(struct octep_vf_device *oct, int q_no)
 	iq->netdev = oct->netdev;
 	iq->dev = &oct->pdev->dev;
 	iq->q_no = q_no;
+	iq->stats = &oct->stats_iq[q_no];
 	iq->max_count = CFG_GET_IQ_NUM_DESC(oct->conf);
 	iq->ring_size_mask = iq->max_count - 1;
 	iq->fill_threshold = CFG_GET_IQ_DB_MIN(oct->conf);

drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_tx.h

@@ -129,7 +129,7 @@ struct octep_vf_iq {
 	u16 flush_index;
 
 	/* Statistics for this input queue. */
-	struct octep_vf_iq_stats stats;
+	struct octep_vf_iq_stats *stats;
 
 	/* Pointer to the Virtual Base addr of the input ring. */
 	struct octep_vf_tx_desc_hw *desc_ring;