diff mbox series

[v1,net-next,1/2] net: ena: Add ENA Express metrics support

Message ID 20240811100711.12921-2-darinzon@amazon.com (mailing list archive)
State Changes Requested
Delegated to: Netdev Maintainers
Headers show
Series ENA driver metrics changes | expand

Checks

Context Check Description
netdev/series_format success Posting correctly formatted
netdev/tree_selection success Clearly marked for net-next
netdev/ynl success Generated files up to date; no warnings/errors; no diff in generated;
netdev/fixes_present success Fixes tag not required for -next series
netdev/header_inline success No static functions without inline keyword in header files
netdev/build_32bit success Errors and warnings before: 29 this patch: 29
netdev/build_tools success No tools touched, skip
netdev/cc_maintainers warning 3 maintainers not CCed: linux-doc@vger.kernel.org horms@kernel.org corbet@lwn.net
netdev/build_clang success Errors and warnings before: 29 this patch: 29
netdev/verify_signedoff success Signed-off-by tag matches author and committer
netdev/deprecated_api success None detected
netdev/check_selftest success No net selftest shell script
netdev/verify_fixes success No Fixes tag
netdev/build_allmodconfig_warn success Errors and warnings before: 29 this patch: 29
netdev/checkpatch warning WARNING: line length of 81 exceeds 80 columns WARNING: line length of 83 exceeds 80 columns WARNING: line length of 85 exceeds 80 columns WARNING: line length of 90 exceeds 80 columns WARNING: line length of 91 exceeds 80 columns WARNING: line length of 93 exceeds 80 columns WARNING: line length of 98 exceeds 80 columns WARNING: line length of 99 exceeds 80 columns
netdev/build_clang_rust success No Rust files in patch. Skipping build
netdev/kdoc success Errors and warnings before: 0 this patch: 0
netdev/source_inline success Was 0 now: 0
netdev/contest success net-next-2024-08-11--21-00 (tests: 707)

Commit Message

Arinzon, David Aug. 11, 2024, 10:07 a.m. UTC
ENA Express metrics, called `ena_srd` are exposed to
customers via ethtool.
The metrics allow customers to check the configuration,
tx/rx counters as well resource usage.

The documentation is also updated to provide a general
explanation and links for further information.

Signed-off-by: Igor Chauskin <igorch@amazon.com>
Signed-off-by: David Arinzon <darinzon@amazon.com>
---
 .../device_drivers/ethernet/amazon/ena.rst    |  5 ++
 .../net/ethernet/amazon/ena/ena_admin_defs.h  | 42 ++++++++++
 drivers/net/ethernet/amazon/ena/ena_com.c     | 21 +++++
 drivers/net/ethernet/amazon/ena/ena_com.h     |  9 ++
 drivers/net/ethernet/amazon/ena/ena_ethtool.c | 84 ++++++++++++++-----
 drivers/net/ethernet/amazon/ena/ena_netdev.c  | 13 ---
 drivers/net/ethernet/amazon/ena/ena_netdev.h  |  2 +-
 7 files changed, 142 insertions(+), 34 deletions(-)

Comments

Jakub Kicinski Aug. 13, 2024, 1:54 a.m. UTC | #1
On Sun, 11 Aug 2024 13:07:10 +0300 David Arinzon wrote:
> +On supported instance types, the statistics will also include the
> +ENA Express data (fields prefixed with `ena_srd`). For a complete
> +documentation of ENA Express data refer to
> +https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ena-express.html#ena-express-monitor

On a quick scan I don't see anything about statistics in this link.
One can probably guess the meaning of most of them from the names.
Arinzon, David Aug. 13, 2024, 11:21 a.m. UTC | #2
> On Sun, 11 Aug 2024 13:07:10 +0300 David Arinzon wrote:
> > +On supported instance types, the statistics will also include the ENA
> > +Express data (fields prefixed with `ena_srd`). For a complete
> > +documentation of ENA Express data refer to
> > +https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ena-
> express.html#
> > +ena-express-monitor
> 
> On a quick scan I don't see anything about statistics in this link.
> One can probably guess the meaning of most of them from the names.

Hi Jakub,

Thanks for looking into this.
There's actually wording about this in the link (in "how ENA Express works", below the note), quoting:

After you've enabled ENA Express for the network interface attachments on both the sending instance and the receiving instance, you can use ENA Express metrics to help ensure that your instances take full advantage of the performance improvements that SRD technology provides. For more information about ENA Express metrics, see Metrics for ENA Express.

There's also a link in "see metrics for ENA Express" for further information and examples about the statistics themselves,

Thanks,
David
diff mbox series

Patch

diff --git a/Documentation/networking/device_drivers/ethernet/amazon/ena.rst b/Documentation/networking/device_drivers/ethernet/amazon/ena.rst
index a4c7d0c6..4561e8ab 100644
--- a/Documentation/networking/device_drivers/ethernet/amazon/ena.rst
+++ b/Documentation/networking/device_drivers/ethernet/amazon/ena.rst
@@ -230,6 +230,11 @@  per-queue stats) from the device.
 
 In addition the driver logs the stats to syslog upon device reset.
 
+On supported instance types, the statistics will also include the
+ENA Express data (fields prefixed with `ena_srd`). For a complete
+documentation of ENA Express data refer to
+https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ena-express.html#ena-express-monitor
+
 MTU
 ===
 
diff --git a/drivers/net/ethernet/amazon/ena/ena_admin_defs.h b/drivers/net/ethernet/amazon/ena/ena_admin_defs.h
index 6de0d590..74772b00 100644
--- a/drivers/net/ethernet/amazon/ena/ena_admin_defs.h
+++ b/drivers/net/ethernet/amazon/ena/ena_admin_defs.h
@@ -51,6 +51,8 @@  enum ena_admin_aq_feature_id {
 /* device capabilities */
 enum ena_admin_aq_caps_id {
 	ENA_ADMIN_ENI_STATS                         = 0,
+	/* ENA SRD customer metrics */
+	ENA_ADMIN_ENA_SRD_INFO                      = 1,
 };
 
 enum ena_admin_placement_policy_type {
@@ -99,6 +101,8 @@  enum ena_admin_get_stats_type {
 	ENA_ADMIN_GET_STATS_TYPE_EXTENDED           = 1,
 	/* extra HW stats for specific network interface */
 	ENA_ADMIN_GET_STATS_TYPE_ENI                = 2,
+	/* extra HW stats for ENA SRD */
+	ENA_ADMIN_GET_STATS_TYPE_ENA_SRD            = 3,
 };
 
 enum ena_admin_get_stats_scope {
@@ -106,6 +110,16 @@  enum ena_admin_get_stats_scope {
 	ENA_ADMIN_ETH_TRAFFIC                       = 1,
 };
 
+/* ENA SRD configuration for ENI */
+enum ena_admin_ena_srd_flags {
+	/* Feature enabled */
+	ENA_ADMIN_ENA_SRD_ENABLED                   = BIT(0),
+	/* UDP support enabled */
+	ENA_ADMIN_ENA_SRD_UDP_ENABLED               = BIT(1),
+	/* Bypass Rx UDP ordering */
+	ENA_ADMIN_ENA_SRD_UDP_ORDERING_BYPASS_ENABLED = BIT(2),
+};
+
 struct ena_admin_aq_common_desc {
 	/* 11:0 : command_id
 	 * 15:12 : reserved12
@@ -419,6 +433,32 @@  struct ena_admin_eni_stats {
 	u64 linklocal_allowance_exceeded;
 };
 
+struct ena_admin_ena_srd_stats {
+	/* Number of packets transmitted over ENA SRD */
+	u64 ena_srd_tx_pkts;
+
+	/* Number of packets transmitted or could have been
+	 * transmitted over ENA SRD
+	 */
+	u64 ena_srd_eligible_tx_pkts;
+
+	/* Number of packets received over ENA SRD */
+	u64 ena_srd_rx_pkts;
+
+	/* Percentage of the ENA SRD resources that is in use */
+	u64 ena_srd_resource_utilization;
+};
+
+/* ENA SRD Statistics Command */
+struct ena_admin_ena_srd_info {
+	/* ENA SRD configuration bitmap. See ena_admin_ena_srd_flags for
+	 * details
+	 */
+	u64 flags;
+
+	struct ena_admin_ena_srd_stats ena_srd_stats;
+};
+
 struct ena_admin_acq_get_stats_resp {
 	struct ena_admin_acq_common_desc acq_common_desc;
 
@@ -428,6 +468,8 @@  struct ena_admin_acq_get_stats_resp {
 		struct ena_admin_basic_stats basic_stats;
 
 		struct ena_admin_eni_stats eni_stats;
+
+		struct ena_admin_ena_srd_info ena_srd_info;
 	} u;
 };
 
diff --git a/drivers/net/ethernet/amazon/ena/ena_com.c b/drivers/net/ethernet/amazon/ena/ena_com.c
index 713a5953..3cc3830d 100644
--- a/drivers/net/ethernet/amazon/ena/ena_com.c
+++ b/drivers/net/ethernet/amazon/ena/ena_com.c
@@ -2152,6 +2152,27 @@  int ena_com_get_eni_stats(struct ena_com_dev *ena_dev,
 	return ret;
 }
 
+int ena_com_get_ena_srd_info(struct ena_com_dev *ena_dev,
+			     struct ena_admin_ena_srd_info *info)
+{
+	struct ena_com_stats_ctx ctx;
+	int ret;
+
+	if (!ena_com_get_cap(ena_dev, ENA_ADMIN_ENA_SRD_INFO)) {
+		netdev_err(ena_dev->net_device, "Capability %d isn't supported\n",
+			   ENA_ADMIN_ENA_SRD_INFO);
+		return -EOPNOTSUPP;
+	}
+
+	memset(&ctx, 0x0, sizeof(ctx));
+	ret = ena_get_dev_stats(ena_dev, &ctx, ENA_ADMIN_GET_STATS_TYPE_ENA_SRD);
+	if (likely(ret == 0))
+		memcpy(info, &ctx.get_resp.u.ena_srd_info,
+		       sizeof(ctx.get_resp.u.ena_srd_info));
+
+	return ret;
+}
+
 int ena_com_get_dev_basic_stats(struct ena_com_dev *ena_dev,
 				struct ena_admin_basic_stats *stats)
 {
diff --git a/drivers/net/ethernet/amazon/ena/ena_com.h b/drivers/net/ethernet/amazon/ena/ena_com.h
index 924f03f5..372066e0 100644
--- a/drivers/net/ethernet/amazon/ena/ena_com.h
+++ b/drivers/net/ethernet/amazon/ena/ena_com.h
@@ -595,6 +595,15 @@  int ena_com_get_dev_basic_stats(struct ena_com_dev *ena_dev,
 int ena_com_get_eni_stats(struct ena_com_dev *ena_dev,
 			  struct ena_admin_eni_stats *stats);
 
+/* ena_com_get_ena_srd_info - Get ENA SRD network interface statistics
+ * @ena_dev: ENA communication layer struct
+ * @info: ena srd stats and flags
+ *
+ * @return: 0 on Success and negative value otherwise.
+ */
+int ena_com_get_ena_srd_info(struct ena_com_dev *ena_dev,
+			     struct ena_admin_ena_srd_info *info);
+
 /* ena_com_set_dev_mtu - Configure the device mtu.
  * @ena_dev: ENA communication layer struct
  * @mtu: mtu value
diff --git a/drivers/net/ethernet/amazon/ena/ena_ethtool.c b/drivers/net/ethernet/amazon/ena/ena_ethtool.c
index b24cc3f0..5efd3e43 100644
--- a/drivers/net/ethernet/amazon/ena/ena_ethtool.c
+++ b/drivers/net/ethernet/amazon/ena/ena_ethtool.c
@@ -41,6 +41,14 @@  struct ena_stats {
 #define ENA_STAT_ENI_ENTRY(stat) \
 	ENA_STAT_HW_ENTRY(stat, eni_stats)
 
+#define ENA_STAT_ENA_SRD_ENTRY(stat) \
+	ENA_STAT_HW_ENTRY(stat, ena_srd_stats)
+
+#define ENA_STAT_ENA_SRD_MODE_ENTRY(stat) { \
+	.name = #stat, \
+	.stat_offset = offsetof(struct ena_admin_ena_srd_info, flags) / sizeof(u64) \
+}
+
 static const struct ena_stats ena_stats_global_strings[] = {
 	ENA_STAT_GLOBAL_ENTRY(tx_timeout),
 	ENA_STAT_GLOBAL_ENTRY(suspend),
@@ -60,6 +68,14 @@  static const struct ena_stats ena_stats_eni_strings[] = {
 	ENA_STAT_ENI_ENTRY(linklocal_allowance_exceeded),
 };
 
+static const struct ena_stats ena_srd_info_strings[] = {
+	ENA_STAT_ENA_SRD_MODE_ENTRY(ena_srd_mode),
+	ENA_STAT_ENA_SRD_ENTRY(ena_srd_tx_pkts),
+	ENA_STAT_ENA_SRD_ENTRY(ena_srd_eligible_tx_pkts),
+	ENA_STAT_ENA_SRD_ENTRY(ena_srd_rx_pkts),
+	ENA_STAT_ENA_SRD_ENTRY(ena_srd_resource_utilization)
+};
+
 static const struct ena_stats ena_stats_tx_strings[] = {
 	ENA_STAT_TX_ENTRY(cnt),
 	ENA_STAT_TX_ENTRY(bytes),
@@ -112,7 +128,8 @@  static const struct ena_stats ena_stats_ena_com_strings[] = {
 #define ENA_STATS_ARRAY_TX		ARRAY_SIZE(ena_stats_tx_strings)
 #define ENA_STATS_ARRAY_RX		ARRAY_SIZE(ena_stats_rx_strings)
 #define ENA_STATS_ARRAY_ENA_COM		ARRAY_SIZE(ena_stats_ena_com_strings)
-#define ENA_STATS_ARRAY_ENI(adapter)	ARRAY_SIZE(ena_stats_eni_strings)
+#define ENA_STATS_ARRAY_ENI		ARRAY_SIZE(ena_stats_eni_strings)
+#define ENA_STATS_ARRAY_ENA_SRD		ARRAY_SIZE(ena_srd_info_strings)
 
 static void ena_safe_update_stat(u64 *src, u64 *dst,
 				 struct u64_stats_sync *syncp)
@@ -179,7 +196,7 @@  static void ena_dev_admin_queue_stats(struct ena_adapter *adapter, u64 **data)
 
 static void ena_get_stats(struct ena_adapter *adapter,
 			  u64 *data,
-			  bool eni_stats_needed)
+			  bool hw_stats_needed)
 {
 	const struct ena_stats *ena_stats;
 	u64 *ptr;
@@ -193,15 +210,37 @@  static void ena_get_stats(struct ena_adapter *adapter,
 		ena_safe_update_stat(ptr, data++, &adapter->syncp);
 	}
 
-	if (eni_stats_needed) {
-		ena_update_hw_stats(adapter);
-		for (i = 0; i < ENA_STATS_ARRAY_ENI(adapter); i++) {
-			ena_stats = &ena_stats_eni_strings[i];
+	if (hw_stats_needed) {
+		if (ena_com_get_cap(adapter->ena_dev, ENA_ADMIN_ENI_STATS)) {
+			ena_com_get_eni_stats(adapter->ena_dev, &adapter->eni_stats);
+			/* Updating regardless of rc - once we told ethtool how many stats we have
+			 * it will print that much stats. We can't leave holes in the stats
+			 */
+			for (i = 0; i < ENA_STATS_ARRAY_ENI; i++) {
+				ena_stats = &ena_stats_eni_strings[i];
 
-			ptr = (u64 *)&adapter->eni_stats +
-				ena_stats->stat_offset;
+				ptr = (u64 *)&adapter->eni_stats +
+					ena_stats->stat_offset;
 
+				ena_safe_update_stat(ptr, data++, &adapter->syncp);
+			}
+		}
+
+		if (ena_com_get_cap(adapter->ena_dev, ENA_ADMIN_ENA_SRD_INFO)) {
+			ena_com_get_ena_srd_info(adapter->ena_dev, &adapter->ena_srd_info);
+			/* Get ENA SRD mode */
+			ptr = (u64 *)&adapter->ena_srd_info;
 			ena_safe_update_stat(ptr, data++, &adapter->syncp);
+			for (i = 1; i < ENA_STATS_ARRAY_ENA_SRD; i++) {
+				ena_stats = &ena_srd_info_strings[i];
+				/* Wrapped within an outer struct - need to accommodate an
+				 * additional offset of the ENA SRD mode that was already processed
+				 */
+				ptr = (u64 *)&adapter->ena_srd_info +
+					ena_stats->stat_offset + 1;
+
+				ena_safe_update_stat(ptr, data++, &adapter->syncp);
+			}
 		}
 	}
 
@@ -214,9 +253,8 @@  static void ena_get_ethtool_stats(struct net_device *netdev,
 				  u64 *data)
 {
 	struct ena_adapter *adapter = netdev_priv(netdev);
-	struct ena_com_dev *dev = adapter->ena_dev;
 
-	ena_get_stats(adapter, data, ena_com_get_cap(dev, ENA_ADMIN_ENI_STATS));
+	ena_get_stats(adapter, data, true);
 }
 
 static int ena_get_sw_stats_count(struct ena_adapter *adapter)
@@ -228,9 +266,8 @@  static int ena_get_sw_stats_count(struct ena_adapter *adapter)
 
 static int ena_get_hw_stats_count(struct ena_adapter *adapter)
 {
-	bool supported = ena_com_get_cap(adapter->ena_dev, ENA_ADMIN_ENI_STATS);
-
-	return ENA_STATS_ARRAY_ENI(adapter) * supported;
+	return ENA_STATS_ARRAY_ENI * ena_com_get_cap(adapter->ena_dev, ENA_ADMIN_ENI_STATS) +
+	       ENA_STATS_ARRAY_ENA_SRD * ena_com_get_cap(adapter->ena_dev, ENA_ADMIN_ENA_SRD_INFO);
 }
 
 int ena_get_sset_count(struct net_device *netdev, int sset)
@@ -291,7 +328,7 @@  static void ena_com_dev_strings(u8 **data)
 
 static void ena_get_strings(struct ena_adapter *adapter,
 			    u8 *data,
-			    bool eni_stats_needed)
+			    bool hw_stats_needed)
 {
 	const struct ena_stats *ena_stats;
 	int i;
@@ -301,10 +338,18 @@  static void ena_get_strings(struct ena_adapter *adapter,
 		ethtool_puts(&data, ena_stats->name);
 	}
 
-	if (eni_stats_needed) {
-		for (i = 0; i < ENA_STATS_ARRAY_ENI(adapter); i++) {
-			ena_stats = &ena_stats_eni_strings[i];
-			ethtool_puts(&data, ena_stats->name);
+	if (hw_stats_needed) {
+		if (ena_com_get_cap(adapter->ena_dev, ENA_ADMIN_ENI_STATS)) {
+			for (i = 0; i < ENA_STATS_ARRAY_ENI; i++) {
+				ena_stats = &ena_stats_eni_strings[i];
+				ethtool_puts(&data, ena_stats->name);
+			}
+		}
+		if (ena_com_get_cap(adapter->ena_dev, ENA_ADMIN_ENA_SRD_INFO)) {
+			for (i = 0; i < ENA_STATS_ARRAY_ENA_SRD; i++) {
+				ena_stats = &ena_srd_info_strings[i];
+				ethtool_puts(&data, ena_stats->name);
+			}
 		}
 	}
 
@@ -317,11 +362,10 @@  static void ena_get_ethtool_strings(struct net_device *netdev,
 				    u8 *data)
 {
 	struct ena_adapter *adapter = netdev_priv(netdev);
-	struct ena_com_dev *dev = adapter->ena_dev;
 
 	switch (sset) {
 	case ETH_SS_STATS:
-		ena_get_strings(adapter, data, ena_com_get_cap(dev, ENA_ADMIN_ENI_STATS));
+		ena_get_strings(adapter, data, true);
 		break;
 	}
 }
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index 184b6e6c..0883c9a2 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -2798,19 +2798,6 @@  err:
 	ena_com_delete_debug_area(adapter->ena_dev);
 }
 
-int ena_update_hw_stats(struct ena_adapter *adapter)
-{
-	int rc;
-
-	rc = ena_com_get_eni_stats(adapter->ena_dev, &adapter->eni_stats);
-	if (rc) {
-		netdev_err(adapter->netdev, "Failed to get ENI stats\n");
-		return rc;
-	}
-
-	return 0;
-}
-
 static void ena_get_stats64(struct net_device *netdev,
 			    struct rtnl_link_stats64 *stats)
 {
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h
index d5950974..6e12ae3b 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.h
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h
@@ -373,6 +373,7 @@  struct ena_adapter {
 	struct u64_stats_sync syncp;
 	struct ena_stats_dev dev_stats;
 	struct ena_admin_eni_stats eni_stats;
+	struct ena_admin_ena_srd_info ena_srd_info;
 
 	/* last queue index that was checked for uncompleted tx packets */
 	u32 last_monitored_tx_qid;
@@ -390,7 +391,6 @@  void ena_dump_stats_to_dmesg(struct ena_adapter *adapter);
 
 void ena_dump_stats_to_buf(struct ena_adapter *adapter, u8 *buf);
 
-int ena_update_hw_stats(struct ena_adapter *adapter);
 
 int ena_update_queue_params(struct ena_adapter *adapter,
 			    u32 new_tx_size,