Merge branch 'ena-driver-changes-may-2024'

David Arinzon says:

====================
ENA driver changes May 2024

This patchset contains several misc and minor
changes to the ENA driver.
====================

Link: https://lore.kernel.org/r/20240512134637.25299-1-darinzon@amazon.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>

drivers/net/ethernet/amazon/ena/ena_com.h

@@ -47,7 +47,7 @@
 /* ENA adaptive interrupt moderation settings */
 
 #define ENA_INTR_INITIAL_TX_INTERVAL_USECS 64
-#define ENA_INTR_INITIAL_RX_INTERVAL_USECS 0
+#define ENA_INTR_INITIAL_RX_INTERVAL_USECS 20
 #define ENA_DEFAULT_INTR_DELAY_RESOLUTION 1
 
 #define ENA_HASH_KEY_SIZE 40
@@ -305,6 +305,8 @@ struct ena_com_dev {
 	u16 stats_func; /* Selected function for extended statistic dump */
 	u16 stats_queue; /* Selected queue for extended statistic dump */
 
+	u32 ena_min_poll_delay_us;
+
 	struct ena_com_mmio_read mmio_read;
 
 	struct ena_rss rss;
@@ -325,8 +327,6 @@ struct ena_com_dev {
 	struct ena_intr_moder_entry *intr_moder_tbl;
 
 	struct ena_com_llq_info llq_info;
-
-	u32 ena_min_poll_delay_us;
 };
 
 struct ena_com_dev_get_features_ctx {

drivers/net/ethernet/amazon/ena/ena_eth_com.c

@@ -229,30 +229,43 @@ static struct ena_eth_io_rx_cdesc_base *
 		idx * io_cq->cdesc_entry_size_in_bytes);
 }
 
-static u16 ena_com_cdesc_rx_pkt_get(struct ena_com_io_cq *io_cq,
-					   u16 *first_cdesc_idx)
+static int ena_com_cdesc_rx_pkt_get(struct ena_com_io_cq *io_cq,
+				    u16 *first_cdesc_idx,
+				    u16 *num_descs)
 {
+	u16 count = io_cq->cur_rx_pkt_cdesc_count, head_masked;
 	struct ena_eth_io_rx_cdesc_base *cdesc;
-	u16 count = 0, head_masked;
 	u32 last = 0;
 
 	do {
+		u32 status;
+
 		cdesc = ena_com_get_next_rx_cdesc(io_cq);
 		if (!cdesc)
 			break;
+		status = READ_ONCE(cdesc->status);
 
 		ena_com_cq_inc_head(io_cq);
+		if (unlikely((status & ENA_ETH_IO_RX_CDESC_BASE_FIRST_MASK) >>
+		    ENA_ETH_IO_RX_CDESC_BASE_FIRST_SHIFT && count != 0)) {
+			struct ena_com_dev *dev = ena_com_io_cq_to_ena_dev(io_cq);
+
+			netdev_err(dev->net_device,
+				   "First bit is on in descriptor #%d on q_id: %d, req_id: %u\n",
+				   count, io_cq->qid, cdesc->req_id);
+			return -EFAULT;
+		}
 		count++;
-		last = (READ_ONCE(cdesc->status) & ENA_ETH_IO_RX_CDESC_BASE_LAST_MASK) >>
+		last = (status & ENA_ETH_IO_RX_CDESC_BASE_LAST_MASK) >>
 			ENA_ETH_IO_RX_CDESC_BASE_LAST_SHIFT;
 	} while (!last);
 
 	if (last) {
 		*first_cdesc_idx = io_cq->cur_rx_pkt_cdesc_start_idx;
-		count += io_cq->cur_rx_pkt_cdesc_count;
 
 		head_masked = io_cq->head & (io_cq->q_depth - 1);
 
+		*num_descs = count;
 		io_cq->cur_rx_pkt_cdesc_count = 0;
 		io_cq->cur_rx_pkt_cdesc_start_idx = head_masked;
 
@@ -260,11 +273,11 @@ static u16 ena_com_cdesc_rx_pkt_get(struct ena_com_io_cq *io_cq,
 			   "ENA q_id: %d packets were completed. first desc idx %u descs# %d\n",
 			   io_cq->qid, *first_cdesc_idx, count);
 	} else {
-		io_cq->cur_rx_pkt_cdesc_count += count;
-		count = 0;
+		io_cq->cur_rx_pkt_cdesc_count = count;
+		*num_descs = 0;
 	}
 
-	return count;
+	return 0;
 }
 
 static int ena_com_create_meta(struct ena_com_io_sq *io_sq,
@@ -539,10 +552,14 @@ int ena_com_rx_pkt(struct ena_com_io_cq *io_cq,
 	u16 cdesc_idx = 0;
 	u16 nb_hw_desc;
 	u16 i = 0;
+	int rc;
 
 	WARN(io_cq->direction != ENA_COM_IO_QUEUE_DIRECTION_RX, "wrong Q type");
 
-	nb_hw_desc = ena_com_cdesc_rx_pkt_get(io_cq, &cdesc_idx);
+	rc = ena_com_cdesc_rx_pkt_get(io_cq, &cdesc_idx, &nb_hw_desc);
+	if (unlikely(rc != 0))
+		return -EFAULT;
+
 	if (nb_hw_desc == 0) {
 		ena_rx_ctx->descs = nb_hw_desc;
 		return 0;

drivers/net/ethernet/amazon/ena/ena_eth_com.h

@@ -47,7 +47,7 @@ struct ena_com_rx_ctx {
 	bool frag;
 	u32 hash;
 	u16 descs;
-	int max_bufs;
+	u16 max_bufs;
 	u8 pkt_offset;
 };
 

drivers/net/ethernet/amazon/ena/ena_ethtool.c

@@ -49,6 +49,7 @@ static const struct ena_stats ena_stats_global_strings[] = {
 	ENA_STAT_GLOBAL_ENTRY(interface_up),
 	ENA_STAT_GLOBAL_ENTRY(interface_down),
 	ENA_STAT_GLOBAL_ENTRY(admin_q_pause),
+	ENA_STAT_GLOBAL_ENTRY(reset_fail),
 };
 
 static const struct ena_stats ena_stats_eni_strings[] = {
@@ -459,10 +460,18 @@ static void ena_get_drvinfo(struct net_device *dev,
 			    struct ethtool_drvinfo *info)
 {
 	struct ena_adapter *adapter = netdev_priv(dev);
+	ssize_t ret = 0;
 
-	strscpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
-	strscpy(info->bus_info, pci_name(adapter->pdev),
+	ret = strscpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
+	if (ret < 0)
+		netif_dbg(adapter, drv, dev,
+			  "module name will be truncated, status = %zd\n", ret);
+
+	ret = strscpy(info->bus_info, pci_name(adapter->pdev),
 		      sizeof(info->bus_info));
+	if (ret < 0)
+		netif_dbg(adapter, drv, dev,
+			  "bus info will be truncated, status = %zd\n", ret);
 }
 
 static void ena_get_ringparam(struct net_device *netdev,

drivers/net/ethernet/amazon/ena/ena_netdev.c

@@ -42,7 +42,7 @@ MODULE_DEVICE_TABLE(pci, ena_pci_tbl);
 
 static int ena_rss_init_default(struct ena_adapter *adapter);
 static void check_for_admin_com_state(struct ena_adapter *adapter);
-static void ena_destroy_device(struct ena_adapter *adapter, bool graceful);
+static int ena_destroy_device(struct ena_adapter *adapter, bool graceful);
 static int ena_restore_device(struct ena_adapter *adapter);
 
 static void ena_tx_timeout(struct net_device *dev, unsigned int txqueue)
@@ -1347,6 +1347,8 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi,
 	if (rc == -ENOSPC) {
 		ena_increase_stat(&rx_ring->rx_stats.bad_desc_num, 1, &rx_ring->syncp);
 		ena_reset_device(adapter, ENA_REGS_RESET_TOO_MANY_RX_DESCS);
+	} else if (rc == -EFAULT) {
+		ena_reset_device(adapter, ENA_REGS_RESET_RX_DESCRIPTOR_MALFORMED);
 	} else {
 		ena_increase_stat(&rx_ring->rx_stats.bad_req_id, 1,
 				  &rx_ring->syncp);
@@ -2701,6 +2703,7 @@ static void ena_config_host_info(struct ena_com_dev *ena_dev, struct pci_dev *pd
 {
 	struct device *dev = &pdev->dev;
 	struct ena_admin_host_info *host_info;
+	ssize_t ret;
 	int rc;
 
 	/* Allocate only the host info */
@@ -2715,11 +2718,19 @@ static void ena_config_host_info(struct ena_com_dev *ena_dev, struct pci_dev *pd
 	host_info->bdf = pci_dev_id(pdev);
 	host_info->os_type = ENA_ADMIN_OS_LINUX;
 	host_info->kernel_ver = LINUX_VERSION_CODE;
-	strscpy(host_info->kernel_ver_str, utsname()->version,
-		sizeof(host_info->kernel_ver_str) - 1);
+	ret = strscpy(host_info->kernel_ver_str, utsname()->version,
+		      sizeof(host_info->kernel_ver_str));
+	if (ret < 0)
+		dev_dbg(dev,
+			"kernel version string will be truncated, status = %zd\n", ret);
+
 	host_info->os_dist = 0;
-	strscpy(host_info->os_dist_str, utsname()->release,
+	ret = strscpy(host_info->os_dist_str, utsname()->release,
 		      sizeof(host_info->os_dist_str));
+	if (ret < 0)
+		dev_dbg(dev,
+			"OS distribution string will be truncated, status = %zd\n", ret);
+
 	host_info->driver_version =
 		(DRV_MODULE_GEN_MAJOR) |
 		(DRV_MODULE_GEN_MINOR << ENA_ADMIN_HOST_INFO_MINOR_SHIFT) |
@@ -3235,14 +3246,15 @@ static int ena_enable_msix_and_set_admin_interrupts(struct ena_adapter *adapter)
 	return rc;
 }
 
-static void ena_destroy_device(struct ena_adapter *adapter, bool graceful)
+static int ena_destroy_device(struct ena_adapter *adapter, bool graceful)
 {
 	struct net_device *netdev = adapter->netdev;
 	struct ena_com_dev *ena_dev = adapter->ena_dev;
 	bool dev_up;
+	int rc = 0;
 
 	if (!test_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags))
-		return;
+		return 0;
 
 	netif_carrier_off(netdev);
 
@@ -3260,7 +3272,7 @@ static void ena_destroy_device(struct ena_adapter *adapter, bool graceful)
 	 *  and device is up, ena_down() already reset the device.
 	 */
 	if (!(test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags) && dev_up))
-		ena_com_dev_reset(adapter->ena_dev, adapter->reset_reason);
+		rc = ena_com_dev_reset(adapter->ena_dev, adapter->reset_reason);
 
 	ena_free_mgmnt_irq(adapter);
 
@@ -3279,6 +3291,8 @@ static void ena_destroy_device(struct ena_adapter *adapter, bool graceful)
 
 	clear_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags);
 	clear_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags);
+
+	return rc;
 }
 
 static int ena_restore_device(struct ena_adapter *adapter)
@@ -3355,14 +3369,17 @@ static int ena_restore_device(struct ena_adapter *adapter)
 
 static void ena_fw_reset_device(struct work_struct *work)
 {
+	int rc = 0;
+
 	struct ena_adapter *adapter =
 		container_of(work, struct ena_adapter, reset_task);
 
 	rtnl_lock();
 
 	if (likely(test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags))) {
-		ena_destroy_device(adapter, false);
-		ena_restore_device(adapter);
+		rc |= ena_destroy_device(adapter, false);
+		rc |= ena_restore_device(adapter);
+		adapter->dev_stats.reset_fail += !!rc;
 
 		dev_err(&adapter->pdev->dev, "Device reset completed successfully\n");
 	}

drivers/net/ethernet/amazon/ena/ena_netdev.h

@@ -290,6 +290,7 @@ struct ena_stats_dev {
 	u64 admin_q_pause;
 	u64 rx_drops;
 	u64 tx_drops;
+	u64 reset_fail;
 };
 
 enum ena_flags_t {

drivers/net/ethernet/amazon/ena/ena_regs_defs.h

@@ -22,6 +22,7 @@ enum ena_regs_reset_reason_types {
 	ENA_REGS_RESET_GENERIC                      = 13,
 	ENA_REGS_RESET_MISS_INTERRUPT               = 14,
 	ENA_REGS_RESET_SUSPECTED_POLL_STARVATION    = 15,
+	ENA_REGS_RESET_RX_DESCRIPTOR_MALFORMED	    = 16,
 };
 
 /* ena_registers offsets */