Skip to contentSkip to conversation

net: xilinx: axienet: Add statistics support

PATCH Applied Sean Anderson 4 participants 8 messages Aug 20, 2024 – Aug 22, 2024

Subsystems (4)
lore
Share & tools
lore
#0 lore raw

v4 cover net: xilinx: axienet: Add statistics support

Add support for hardware statistics counters (if they are enabled) in the AXI Ethernet driver. Unfortunately, the implementation is complicated a bit since the hardware might only support 32-bit counters.

Changes in v4:
- Reduce hw_last_counter to u32 to ensure we use (wrapping) 32-bit arithmetic.
- Implement get_ethtool_stats for nonstandard statistics
- Rebase onto net-next/main

Changes in v3:
- Use explicit mutex_lock/unlock instead of guard() in __axienet_device_reset to make it clear that we need to hold lp->stats_lock for the whole function.

Changes in v2:
- Switch to a seqlock-based implementation to allow fresher updates (rather than always using stale counter values from the previous refresh).
- Take stats_lock unconditionally in __axienet_device_reset
- Fix documentation mismatch

Sean Anderson (2):
net: xilinx: axienet: Report RxRject as rx_dropped
net: xilinx: axienet: Add statistics support

 drivers/net/ethernet/xilinx/xilinx_axienet.h  |  85 +++++
 .../net/ethernet/xilinx/xilinx_axienet_main.c | 327 +++++++++++++++++-
 2 files changed, 408 insertions(+), 4 deletions(-)
Signature · 1 line
2.35.1.1320.gc452695387.dirty
#1 lore raw

PATCH v4 1/2 net: xilinx: axienet: Report RxRject as rx_dropped

The Receive Frame Rejected interrupt is asserted whenever there was a receive error (bad FCS, bad length, etc.) or whenever the frame was dropped due to a mismatched address. So this is really a combination of rx_otherhost_dropped, rx_length_errors, rx_frame_errors, and rx_crc_errors. Mismatched addresses are common and aren't really errors at all (much like how fragments are normal on half-duplex links). To avoid confusion, report these events as rx_dropped. This better reflects what's going on: the packet was received by the MAC but dropped before being processed.

---

(no changes since v1)

 drivers/net/ethernet/xilinx/xilinx_axienet_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/xilinx/xilinx_axienet_main.c +1 -1
index ca04c298daa2..b2d7c396e2e3 100644
@@ -1296,7 +1296,7 @@ static irqreturn_t axienet_eth_irq(int irq, void *_ndev)
 		ndev->stats.rx_missed_errors++;
 
 	if (pending & XAE_INT_RXRJECT_MASK)
-		ndev->stats.rx_frame_errors++;
+		ndev->stats.rx_dropped++;
 
 	axienet_iow(lp, XAE_IS_OFFSET, pending);
 	return IRQ_HANDLED;
Signature · 1 line
2.35.1.1320.gc452695387.dirty
maintainer Reviewed-by on v4 1/2
#2 lore raw

RE: [PATCH net-next v4 1/2] net: xilinx: axienet: Report RxRject as rx_dropped

Quoted 24 lines

-----Original Message-----
From: Sean Anderson <[email protected]>
Sent: Tuesday, August 20, 2024 11:24 PM
To: Andrew Lunn <[email protected]>; Pandey, Radhey Shyam <[email protected]>; [email protected]
Cc: Simek, Michal <[email protected]>; [email protected]; Russell King <[email protected]>; David S . Miller <[email protected]>; Jakub Kicinski <[email protected]>; Paolo Abeni <[email protected]>; Eric Dumazet <[email protected]>; Simon Horman <[email protected]>; [email protected]; Sean Anderson <[email protected]>
Subject: [PATCH net-next v4 1/2] net: xilinx: axienet: Report RxRject as rx_dropped

The Receive Frame Rejected interrupt is asserted whenever there was a receive error (bad FCS, bad length, etc.) or whenever the frame was dropped due to a mismatched address. So this is really a combination of rx_otherhost_dropped, rx_length_errors, rx_frame_errors, and rx_crc_errors. Mismatched addresses are common and aren't really errors at all (much like how fragments are normal on half-duplex links). To avoid confusion, report these events as rx_dropped. This better reflects what's going on: the packet was received by the MAC but dropped before being processed.

Quoted 2 lines
Quoted 24 lines

---

(no changes since v1)

 drivers/net/ethernet/xilinx/xilinx_axienet_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/xilinx/xilinx_axienet_main.c +1 -1
index ca04c298daa2..b2d7c396e2e3 100644
@@ -1296,7 +1296,7 @@ static irqreturn_t axienet_eth_irq(int irq, void
 *_ndev)
 		ndev->stats.rx_missed_errors++;
 
 	if (pending & XAE_INT_RXRJECT_MASK)
-		ndev->stats.rx_frame_errors++;
+		ndev->stats.rx_dropped++;
 
 	axienet_iow(lp, XAE_IS_OFFSET, pending);

return IRQ_HANDLED;
--
2.35.1.1320.gc452695387.dirty

#3 lore raw

PATCH v4 2/2 net: xilinx: axienet: Add statistics support

In reply to Sean Anderson

Add support for reading the statistics counters, if they are enabled. The counters may be 64-bit, but we can't detect this statically as there's no ability bit for it and the counters are read-only. Therefore, we assume the counters are 32-bits by default. To ensure we don't miss an overflow, we read all counters at 13-second intervals. This should be often enough to ensure the bytes counters don't wrap at 2.5 Gbit/s.

Another complication is that the counters may be reset when the device is reset (depending on configuration). To ensure the counters persist across link up/down (including suspend/resume), we maintain our own versions along with the last counter value we saw. Because we might wait up to 100 ms for the reset to complete, we use a mutex to protect writing hw_stats. We can't sleep in ndo_get_stats64, so we use a seqlock to protect readers.

We don't bother disabling the refresh work when we detect 64-bit counters. This is because the reset issue requires us to read hw_stat_base and reset_in_progress anyway, which would still require the seqcount. And I don't think skipping the task is worth the extra bookkeeping.

We can't use the byte counters for either get_stats64 or get_eth_mac_stats. This is because the byte counters include everything in the frame (destination address to FCS, inclusive). But rtnl_link_stats64 wants bytes excluding the FCS, and ethtool_eth_mac_stats wants to exclude the L2 overhead (addresses and length/type). It might be possible to calculate the byte values Linux expects based on the frame counters, but I think it is simpler to use the existing software counters.

get_ethtool_stats is implemented for nonstandard statistics. This includes the aforementioned byte counters, VLAN and PFC frame counters, and user-defined (e.g. with custom RTL) counters.

---

Changes in v4:
- Reduce hw_last_counter to u32 to ensure we use (wrapping) 32-bit arithmetic.
- Implement get_ethtool_stats for nonstandard statistics

Changes in v3:
- Use explicit mutex_lock/unlock instead of guard() in __axienet_device_reset to make it clear that we need to hold lp->stats_lock for the whole function.

Changes in v2:
- Switch to a seqlock-based implementation to allow fresher updates (rather than always using stale counter values from the previous refresh).
- Take stats_lock unconditionally in __axienet_device_reset
- Fix documentation mismatch

 drivers/net/ethernet/xilinx/xilinx_axienet.h  |  85 +++++
 .../net/ethernet/xilinx/xilinx_axienet_main.c | 325 +++++++++++++++++-
 2 files changed, 407 insertions(+), 3 deletions(-)
drivers/net/ethernet/xilinx/xilinx_axienet.h +85 -0
index ea4103a635f9..8165f5f271ff 100644
@@ -156,6 +156,7 @@
 #define XAE_TPID0_OFFSET	0x00000028 /* VLAN TPID0 register */
 #define XAE_TPID1_OFFSET	0x0000002C /* VLAN TPID1 register */
 #define XAE_PPST_OFFSET		0x00000030 /* PCS PMA Soft Temac Status Reg */
+#define XAE_STATS_OFFSET	0x00000200 /* Statistics counters */
 #define XAE_RCW0_OFFSET		0x00000400 /* Rx Configuration Word 0 */
 #define XAE_RCW1_OFFSET		0x00000404 /* Rx Configuration Word 1 */
 #define XAE_TC_OFFSET		0x00000408 /* Tx Configuration */
@@ -163,6 +164,7 @@
 #define XAE_EMMC_OFFSET		0x00000410 /* MAC speed configuration */
 #define XAE_PHYC_OFFSET		0x00000414 /* RX Max Frame Configuration */
 #define XAE_ID_OFFSET		0x000004F8 /* Identification register */
+#define XAE_ABILITY_OFFSET	0x000004FC /* Ability Register offset */
 #define XAE_MDIO_MC_OFFSET	0x00000500 /* MDIO Setup */
 #define XAE_MDIO_MCR_OFFSET	0x00000504 /* MDIO Control */
 #define XAE_MDIO_MWD_OFFSET	0x00000508 /* MDIO Write Data */
@@ -283,6 +285,16 @@
 #define XAE_PHYC_SGLINKSPD_100		0x40000000 /* SGMII link 100 Mbit */
 #define XAE_PHYC_SGLINKSPD_1000		0x80000000 /* SGMII link 1000 Mbit */
 
+/* Bit masks for Axi Ethernet ability register */
+#define XAE_ABILITY_PFC			BIT(16)
+#define XAE_ABILITY_FRAME_FILTER	BIT(10)
+#define XAE_ABILITY_HALF_DUPLEX		BIT(9)
+#define XAE_ABILITY_STATS		BIT(8)
+#define XAE_ABILITY_2_5G		BIT(3)
+#define XAE_ABILITY_1G			BIT(2)
+#define XAE_ABILITY_100M		BIT(1)
+#define XAE_ABILITY_10M			BIT(0)
+
 /* Bit masks for Axi Ethernet MDIO interface MC register */
 #define XAE_MDIO_MC_MDIOEN_MASK		0x00000040 /* MII management enable */
 #define XAE_MDIO_MC_CLOCK_DIVIDE_MAX	0x3F	   /* Maximum MDIO divisor */
@@ -331,6 +343,7 @@
 #define XAE_FEATURE_FULL_RX_CSUM	BIT(2)
 #define XAE_FEATURE_FULL_TX_CSUM	BIT(3)
 #define XAE_FEATURE_DMA_64BIT		BIT(4)
+#define XAE_FEATURE_STATS		BIT(5)
 
 #define XAE_NO_CSUM_OFFLOAD		0
 
@@ -344,6 +357,61 @@
 #define XLNX_MII_STD_SELECT_REG		0x11
 #define XLNX_MII_STD_SELECT_SGMII	BIT(0)
 
+/* enum temac_stat - TEMAC statistics counters
+ *
+ * Index of statistics counters within the TEMAC. This must match the
+ * order/offset of hardware registers exactly.
+ */
+enum temac_stat {
+	STAT_RX_BYTES = 0,
+	STAT_TX_BYTES,
+	STAT_UNDERSIZE_FRAMES,
+	STAT_FRAGMENT_FRAMES,
+	STAT_RX_64_BYTE_FRAMES,
+	STAT_RX_65_127_BYTE_FRAMES,
+	STAT_RX_128_255_BYTE_FRAMES,
+	STAT_RX_256_511_BYTE_FRAMES,
+	STAT_RX_512_1023_BYTE_FRAMES,
+	STAT_RX_1024_MAX_BYTE_FRAMES,
+	STAT_RX_OVERSIZE_FRAMES,
+	STAT_TX_64_BYTE_FRAMES,
+	STAT_TX_65_127_BYTE_FRAMES,
+	STAT_TX_128_255_BYTE_FRAMES,
+	STAT_TX_256_511_BYTE_FRAMES,
+	STAT_TX_512_1023_BYTE_FRAMES,
+	STAT_TX_1024_MAX_BYTE_FRAMES,
+	STAT_TX_OVERSIZE_FRAMES,
+	STAT_RX_GOOD_FRAMES,
+	STAT_RX_FCS_ERRORS,
+	STAT_RX_BROADCAST_FRAMES,
+	STAT_RX_MULTICAST_FRAMES,
+	STAT_RX_CONTROL_FRAMES,
+	STAT_RX_LENGTH_ERRORS,
+	STAT_RX_VLAN_FRAMES,
+	STAT_RX_PAUSE_FRAMES,
+	STAT_RX_CONTROL_OPCODE_ERRORS,
+	STAT_TX_GOOD_FRAMES,
+	STAT_TX_BROADCAST_FRAMES,
+	STAT_TX_MULTICAST_FRAMES,
+	STAT_TX_UNDERRUN_ERRORS,
+	STAT_TX_CONTROL_FRAMES,
+	STAT_TX_VLAN_FRAMES,
+	STAT_TX_PAUSE_FRAMES,
+	STAT_TX_SINGLE_COLLISION_FRAMES,
+	STAT_TX_MULTIPLE_COLLISION_FRAMES,
+	STAT_TX_DEFERRED_FRAMES,
+	STAT_TX_LATE_COLLISIONS,
+	STAT_TX_EXCESS_COLLISIONS,
+	STAT_TX_EXCESS_DEFERRAL,
+	STAT_RX_ALIGNMENT_ERRORS,
+	STAT_TX_PFC_FRAMES,
+	STAT_RX_PFC_FRAMES,
+	STAT_USER_DEFINED0,
+	STAT_USER_DEFINED1,
+	STAT_USER_DEFINED2,
+	STAT_COUNT,
+};
+
 /**
  * struct axidma_bd - Axi Dma buffer descriptor layout
  * @next:         MM2S/S2MM Next Descriptor Pointer
@@ -434,6 +502,16 @@ struct skbuf_dma_descriptor {
  * @tx_packets: TX packet count for statistics
  * @tx_bytes:	TX byte count for statistics
  * @tx_stat_sync: Synchronization object for TX stats
+ * @hw_stat_base: Base offset for statistics counters. This may be nonzero if
+ *                the statistics counteres were reset or wrapped around.
+ * @hw_last_counter: Last-seen value of each statistic counter
+ * @reset_in_progress: Set while we are performing a reset and statistics
+ *                     counters may be invalid
+ * @hw_stats_seqcount: Sequence counter for @hw_stat_base, @hw_last_counter,
+ *                     and @reset_in_progress.
+ * @stats_lock: Lock for @hw_stats_seqcount
+ * @stats_work: Work for reading the hardware statistics counters often enough
+ *              to catch overflows.
  * @dma_err_task: Work structure to process Axi DMA errors
  * @tx_irq:	Axidma TX IRQ number
  * @rx_irq:	Axidma RX IRQ number
@@ -505,6 +583,13 @@ struct axienet_local {
 	u64_stats_t tx_bytes;
 	struct u64_stats_sync tx_stat_sync;
 
+	u64 hw_stat_base[STAT_COUNT];
+	u32 hw_last_counter[STAT_COUNT];
+	seqcount_mutex_t hw_stats_seqcount;
+	struct mutex stats_lock;
+	struct delayed_work stats_work;
+	bool reset_in_progress;
+
 	struct work_struct dma_err_task;
 
 	int tx_irq;
drivers/net/ethernet/xilinx/xilinx_axienet_main.c +322 -3

398 lines not on this page · raw message

Signature · 1 line
2.35.1.1320.gc452695387.dirty
#4 lore raw
Quoted 13 lines

-----Original Message-----
From: Sean Anderson <[email protected]>
Sent: Tuesday, August 20, 2024 11:24 PM
To: Andrew Lunn <[email protected]>; Pandey, Radhey Shyam <[email protected]>; [email protected]
Cc: Simek, Michal <[email protected]>; [email protected]; Russell King <[email protected]>; David S . Miller <[email protected]>; Jakub Kicinski <[email protected]>; Paolo Abeni <[email protected]>; Eric Dumazet <[email protected]>; Simon Horman <[email protected]>; [email protected]; Sean Anderson <[email protected]>
Subject: [PATCH net-next v4 2/2] net: xilinx: axienet: Add statistics support

Quoted 4 lines

Add support for reading the statistics counters, if they are enabled. The counters may be 64-bit, but we can't detect this statically as there's no ability bit for it and the counters are read-only. Therefore, we assume the counters are 32-bits by default. To ensure we don't miss

Any reason why we can't have DT property to detect if stats counter are configured as 32-bit /64bit? The IP export CONFIG.Statistics_Width and device tree generator can read this IP block property and populate stats width property.

Quoted 7 lines

an overflow, we read all counters at 13-second intervals. This should be often enough to ensure the bytes counters don't wrap at 2.5 Gbit/s.

Another complication is that the counters may be reset when the device is reset (depending on configuration). To ensure the counters persist across link up/down (including suspend/resume), we maintain our own versions along with the last counter value we saw. Because we might wait

Is that a standard convention to retain/persist counter values across link up/down?

Quoted 641 lines

up to 100 ms for the reset to complete, we use a mutex to protect writing hw_stats. We can't sleep in ndo_get_stats64, so we use a seqlock to protect readers.

We don't bother disabling the refresh work when we detect 64-bit counters. This is because the reset issue requires us to read hw_stat_base and reset_in_progress anyway, which would still require the seqcount. And I don't think skipping the task is worth the extra bookkeeping.

We can't use the byte counters for either get_stats64 or get_eth_mac_stats. This is because the byte counters include everything in the frame (destination address to FCS, inclusive). But rtnl_link_stats64 wants bytes excluding the FCS, and ethtool_eth_mac_stats wants to exclude the L2 overhead (addresses and length/type). It might be possible to calculate the byte values Linux expects based on the frame counters, but I think it is simpler to use the existing software counters.

get_ethtool_stats is implemented for nonstandard statistics. This includes the aforementioned byte counters, VLAN and PFC frame counters, and user-defined (e.g. with custom RTL) counters.

---

Changes in v4:
- Reduce hw_last_counter to u32 to ensure we use (wrapping) 32-bit arithmetic.
- Implement get_ethtool_stats for nonstandard statistics

Changes in v3:
- Use explicit mutex_lock/unlock instead of guard() in __axienet_device_reset to make it clear that we need to hold lp->stats_lock for the whole function.

Changes in v2:
- Switch to a seqlock-based implementation to allow fresher updates (rather than always using stale counter values from the previous refresh).
- Take stats_lock unconditionally in __axienet_device_reset
- Fix documentation mismatch

 drivers/net/ethernet/xilinx/xilinx_axienet.h  |  85 +++++
 .../net/ethernet/xilinx/xilinx_axienet_main.c | 325 +++++++++++++++++-
 2 files changed, 407 insertions(+), 3 deletions(-)
drivers/net/ethernet/xilinx/xilinx_axienet.h +0 -0
index ea4103a635f9..8165f5f271ff 100644
@@ -156,6 +156,7 @@
 #define XAE_TPID0_OFFSET	0x00000028 /* VLAN TPID0 register */
 #define XAE_TPID1_OFFSET	0x0000002C /* VLAN TPID1 register */
 #define XAE_PPST_OFFSET		0x00000030 /* PCS PMA Soft Temac

Status Reg */
+#define XAE_STATS_OFFSET 0x00000200 /* Statistics counters */
#define XAE_RCW0_OFFSET 0x00000400 /* Rx Configuration
Word 0 */
#define XAE_RCW1_OFFSET 0x00000404 /* Rx Configuration
Word 1 */
#define XAE_TC_OFFSET 0x00000408 /* Tx Configuration */

drivers/net/ethernet/xilinx/xilinx_axienet.h +0 -0
@@ -163,6 +164,7 @@
 #define XAE_EMMC_OFFSET		0x00000410 /* MAC speed

configuration */
#define XAE_PHYC_OFFSET 0x00000414 /* RX Max Frame
Configuration */
#define XAE_ID_OFFSET 0x000004F8 /* Identification register
*/

drivers/net/ethernet/xilinx/xilinx_axienet.h +1 -0
+#define XAE_ABILITY_OFFSET	0x000004FC /* Ability Register offset */
 #define XAE_MDIO_MC_OFFSET	0x00000500 /* MDIO Setup */
 #define XAE_MDIO_MCR_OFFSET	0x00000504 /* MDIO Control */
 #define XAE_MDIO_MWD_OFFSET	0x00000508 /* MDIO Write Data */
@@ -283,6 +285,16 @@
 #define XAE_PHYC_SGLINKSPD_100		0x40000000 /* SGMII link 100

Mbit */
#define XAE_PHYC_SGLINKSPD_1000 0x80000000 /* SGMII link
1000 Mbit */

drivers/net/ethernet/xilinx/xilinx_axienet.h +10 -0
+/* Bit masks for Axi Ethernet ability register */
+#define XAE_ABILITY_PFC			BIT(16)
+#define XAE_ABILITY_FRAME_FILTER	BIT(10)
+#define XAE_ABILITY_HALF_DUPLEX		BIT(9)
+#define XAE_ABILITY_STATS		BIT(8)
+#define XAE_ABILITY_2_5G		BIT(3)
+#define XAE_ABILITY_1G			BIT(2)
+#define XAE_ABILITY_100M		BIT(1)
+#define XAE_ABILITY_10M			BIT(0)
+
 /* Bit masks for Axi Ethernet MDIO interface MC register */
 #define XAE_MDIO_MC_MDIOEN_MASK		0x00000040 /* MII

management enable */
#define XAE_MDIO_MC_CLOCK_DIVIDE_MAX 0x3F /* Maximum MDIO
divisor */

drivers/net/ethernet/xilinx/xilinx_axienet.h +57 -0
@@ -331,6 +343,7 @@
 #define XAE_FEATURE_FULL_RX_CSUM	BIT(2)
 #define XAE_FEATURE_FULL_TX_CSUM	BIT(3)
 #define XAE_FEATURE_DMA_64BIT		BIT(4)
+#define XAE_FEATURE_STATS		BIT(5)
 
 #define XAE_NO_CSUM_OFFLOAD		0
 
@@ -344,6 +357,61 @@
 #define XLNX_MII_STD_SELECT_REG		0x11
 #define XLNX_MII_STD_SELECT_SGMII	BIT(0)
 
+/* enum temac_stat - TEMAC statistics counters
+ *
+ * Index of statistics counters within the TEMAC. This must match the
+ * order/offset of hardware registers exactly.
+ */
+enum temac_stat {
+	STAT_RX_BYTES = 0,
+	STAT_TX_BYTES,
+	STAT_UNDERSIZE_FRAMES,
+	STAT_FRAGMENT_FRAMES,
+	STAT_RX_64_BYTE_FRAMES,
+	STAT_RX_65_127_BYTE_FRAMES,
+	STAT_RX_128_255_BYTE_FRAMES,
+	STAT_RX_256_511_BYTE_FRAMES,
+	STAT_RX_512_1023_BYTE_FRAMES,
+	STAT_RX_1024_MAX_BYTE_FRAMES,
+	STAT_RX_OVERSIZE_FRAMES,
+	STAT_TX_64_BYTE_FRAMES,
+	STAT_TX_65_127_BYTE_FRAMES,
+	STAT_TX_128_255_BYTE_FRAMES,
+	STAT_TX_256_511_BYTE_FRAMES,
+	STAT_TX_512_1023_BYTE_FRAMES,
+	STAT_TX_1024_MAX_BYTE_FRAMES,
+	STAT_TX_OVERSIZE_FRAMES,
+	STAT_RX_GOOD_FRAMES,
+	STAT_RX_FCS_ERRORS,
+	STAT_RX_BROADCAST_FRAMES,
+	STAT_RX_MULTICAST_FRAMES,
+	STAT_RX_CONTROL_FRAMES,
+	STAT_RX_LENGTH_ERRORS,
+	STAT_RX_VLAN_FRAMES,
+	STAT_RX_PAUSE_FRAMES,
+	STAT_RX_CONTROL_OPCODE_ERRORS,
+	STAT_TX_GOOD_FRAMES,
+	STAT_TX_BROADCAST_FRAMES,
+	STAT_TX_MULTICAST_FRAMES,
+	STAT_TX_UNDERRUN_ERRORS,
+	STAT_TX_CONTROL_FRAMES,
+	STAT_TX_VLAN_FRAMES,
+	STAT_TX_PAUSE_FRAMES,
+	STAT_TX_SINGLE_COLLISION_FRAMES,
+	STAT_TX_MULTIPLE_COLLISION_FRAMES,
+	STAT_TX_DEFERRED_FRAMES,
+	STAT_TX_LATE_COLLISIONS,
+	STAT_TX_EXCESS_COLLISIONS,
+	STAT_TX_EXCESS_DEFERRAL,
+	STAT_RX_ALIGNMENT_ERRORS,
+	STAT_TX_PFC_FRAMES,
+	STAT_RX_PFC_FRAMES,
+	STAT_USER_DEFINED0,
+	STAT_USER_DEFINED1,
+	STAT_USER_DEFINED2,
+	STAT_COUNT,
+};
+
 /**
  * struct axidma_bd - Axi Dma buffer descriptor layout
  * @next:         MM2S/S2MM Next Descriptor Pointer
@@ -434,6 +502,16 @@ struct skbuf_dma_descriptor {
  * @tx_packets: TX packet count for statistics
  * @tx_bytes:	TX byte count for statistics
  * @tx_stat_sync: Synchronization object for TX stats
+ * @hw_stat_base: Base offset for statistics counters. This may be nonzero

if
+ * the statistics counteres were reset or wrapped around.
+ * @hw_last_counter: Last-seen value of each statistic counter
+ * @reset_in_progress: Set while we are performing a reset and statistics
+ * counters may be invalid
+ * @hw_stats_seqcount: Sequence counter for @hw_stat_base, @hw_last_counter,
+ * and @reset_in_progress.
+ * @stats_lock: Lock for @hw_stats_seqcount
+ * @stats_work: Work for reading the hardware statistics counters often enough
+ * to catch overflows.
* @dma_err_task: Work structure to process Axi DMA errors
* @tx_irq: Axidma TX IRQ number
* @rx_irq: Axidma RX IRQ number

drivers/net/ethernet/xilinx/xilinx_axienet.h +7 -0
@@ -505,6 +583,13 @@ struct axienet_local {
 	u64_stats_t tx_bytes;
 	struct u64_stats_sync tx_stat_sync;
 
+	u64 hw_stat_base[STAT_COUNT];
+	u32 hw_last_counter[STAT_COUNT];
+	seqcount_mutex_t hw_stats_seqcount;
+	struct mutex stats_lock;
+	struct delayed_work stats_work;
+	bool reset_in_progress;
+
 	struct work_struct dma_err_task;
 
 	int tx_irq;
drivers/net/ethernet/xilinx/xilinx_axienet_main.c +44 -0
index b2d7c396e2e3..38f7b764fe66 100644
@@ -519,11 +519,55 @@ static void axienet_setoptions(struct net_device
 *ndev, u32 options)
 	lp->options |= options;
 }
 
+static u64 axienet_stat(struct axienet_local *lp, enum temac_stat stat)
+{
+	u32 counter;
+
+	if (lp->reset_in_progress)
+		return lp->hw_stat_base[stat];
+
+	counter = axienet_ior(lp, XAE_STATS_OFFSET + stat * 8);
+	return lp->hw_stat_base[stat] + (counter - lp-
 >hw_last_counter[stat]);
+}
+
+static void axienet_stats_update(struct axienet_local *lp, bool reset)
+{
+	enum temac_stat stat;
+
+	write_seqcount_begin(&lp->hw_stats_seqcount);
+	lp->reset_in_progress = reset;
+	for (stat = 0; stat < STAT_COUNT; stat++) {
+		u32 counter = axienet_ior(lp, XAE_STATS_OFFSET + stat * 8);
+
+		lp->hw_stat_base[stat] += counter - lp-
 >hw_last_counter[stat];
+		lp->hw_last_counter[stat] = counter;
+	}
+	write_seqcount_end(&lp->hw_stats_seqcount);
+}
+
+static void axienet_refresh_stats(struct work_struct *work)
+{
+	struct axienet_local *lp = container_of(work, struct axienet_local,
+						stats_work.work);
+
+	mutex_lock(&lp->stats_lock);
+	axienet_stats_update(lp, false);
+	mutex_unlock(&lp->stats_lock);
+
+	/* Just less than 2^32 bytes at 2.5 GBit/s */
+	schedule_delayed_work(&lp->stats_work, 13 * HZ);
+}
+
 static int __axienet_device_reset(struct axienet_local *lp)
 {
 	u32 value;
 	int ret;
 
+	/* Save statistics counters in case they will be reset */
+	mutex_lock(&lp->stats_lock);
+	if (lp->features & XAE_FEATURE_STATS)
+		axienet_stats_update(lp, true);
+

/* Reset Axi DMA. This would reset Axi Ethernet core as well. The
reset

drivers/net/ethernet/xilinx/xilinx_axienet_main.c +1 -1
 	 * process of Axi DMA takes a while to complete as all pending
 	 * commands/transfers will be flushed or completed during this
@@ -538,7 +582,7 @@ static int __axienet_device_reset(struct axienet_local
 *lp)
 				XAXIDMA_TX_CR_OFFSET);
 	if (ret) {
 		dev_err(lp->dev, "%s: DMA reset timeout!\n", __func__);
-		return ret;
+		goto out;
 	}
 

/* Wait for PhyRstCmplt bit to be set, indicating the PHY reset has
finished */

drivers/net/ethernet/xilinx/xilinx_axienet_main.c +21 -2
@@ -548,10 +592,29 @@ static int __axienet_device_reset(struct
 axienet_local *lp)
 				XAE_IS_OFFSET);
 	if (ret) {
 		dev_err(lp->dev, "%s: timeout waiting for PhyRstCmplt\n",
 __func__);
-		return ret;
+		goto out;
 	}
 
-	return 0;
+	/* Update statistics counters with new values */
+	if (lp->features & XAE_FEATURE_STATS) {
+		enum temac_stat stat;
+
+		write_seqcount_begin(&lp->hw_stats_seqcount);
+		lp->reset_in_progress = false;
+		for (stat = 0; stat < STAT_COUNT; stat++) {
+			u32 counter =
+				axienet_ior(lp, XAE_STATS_OFFSET + stat * 8);
+
+			lp->hw_stat_base[stat] +=
+				lp->hw_last_counter[stat] - counter;
+			lp->hw_last_counter[stat] = counter;
+		}
+		write_seqcount_end(&lp->hw_stats_seqcount);
+	}
+
+out:
+	mutex_unlock(&lp->stats_lock);
+	return ret;
 }

/**

drivers/net/ethernet/xilinx/xilinx_axienet_main.c +18 -0
@@ -1530,6 +1593,9 @@ static int axienet_open(struct net_device *ndev)
 
 	phylink_start(lp->phylink);
 
+	/* Start the statistics refresh work */
+	schedule_delayed_work(&lp->stats_work, 0);
+
 	if (lp->use_dmaengine) {
 		/* Enable interrupts for Axi Ethernet core (if defined) */
 		if (lp->eth_irq > 0) {
@@ -1554,6 +1620,7 @@ static int axienet_open(struct net_device *ndev)
 	if (lp->eth_irq > 0)
 		free_irq(lp->eth_irq, ndev);
 err_phy:
+	cancel_delayed_work_sync(&lp->stats_work);
 	phylink_stop(lp->phylink);
 	phylink_disconnect_phy(lp->phylink);
 	return ret;
@@ -1579,6 +1646,8 @@ static int axienet_stop(struct net_device *ndev)
 		napi_disable(&lp->napi_rx);
 	}
 
+	cancel_delayed_work_sync(&lp->stats_work);
+
 	phylink_stop(lp->phylink);
 	phylink_disconnect_phy(lp->phylink);
 
@@ -1692,6 +1761,35 @@ axienet_get_stats64(struct net_device *dev,
 struct rtnl_link_stats64 *stats)
 		stats->tx_packets = u64_stats_read(&lp->tx_packets);
 		stats->tx_bytes = u64_stats_read(&lp->tx_bytes);
 	} while (u64_stats_fetch_retry(&lp->tx_stat_sync, start));
+
+	if (!(lp->features & XAE_FEATURE_STATS))
+		return;
+
+	do {
+		start = read_seqcount_begin(&lp->hw_stats_seqcount);
+		stats->rx_length_errors =
+			axienet_stat(lp, STAT_RX_LENGTH_ERRORS);
+		stats->rx_crc_errors = axienet_stat(lp,
 STAT_RX_FCS_ERRORS);
+		stats->rx_frame_errors =
+			axienet_stat(lp, STAT_RX_ALIGNMENT_ERRORS);
+		stats->rx_errors = axienet_stat(lp,

STAT_UNDERSIZE_FRAMES) +

drivers/net/ethernet/xilinx/xilinx_axienet_main.c +6 -0
+				   axienet_stat(lp, STAT_FRAGMENT_FRAMES)
+
+				   stats->rx_length_errors +
+				   stats->rx_crc_errors +
+				   stats->rx_frame_errors;
+		stats->multicast = axienet_stat(lp,

STAT_RX_MULTICAST_FRAMES);

drivers/net/ethernet/xilinx/xilinx_axienet_main.c +8 -0
+
+		stats->tx_aborted_errors =
+			axienet_stat(lp, STAT_TX_EXCESS_COLLISIONS);
+		stats->tx_fifo_errors =
+			axienet_stat(lp, STAT_TX_UNDERRUN_ERRORS);
+		stats->tx_window_errors =
+			axienet_stat(lp, STAT_TX_LATE_COLLISIONS);
+		stats->tx_errors = axienet_stat(lp,

STAT_TX_EXCESS_DEFERRAL) +

drivers/net/ethernet/xilinx/xilinx_axienet_main.c +98 -0
+				   stats->tx_aborted_errors +
+				   stats->tx_fifo_errors +
+				   stats->tx_window_errors;
+	} while (read_seqcount_retry(&lp->hw_stats_seqcount, start));
 }
 
 static const struct net_device_ops axienet_netdev_ops = {
@@ -1984,6 +2082,213 @@ static int axienet_ethtools_nway_reset(struct
 net_device *dev)
 	return phylink_ethtool_nway_reset(lp->phylink);
 }
 
+static void axienet_ethtools_get_ethtool_stats(struct net_device *dev,
+					       struct ethtool_stats *stats,
+					       u64 *data)
+{
+	struct axienet_local *lp = netdev_priv(dev);
+	unsigned int start;
+
+	do {
+		start = read_seqcount_begin(&lp->hw_stats_seqcount);
+		data[0] = axienet_stat(lp, STAT_RX_BYTES);
+		data[1] = axienet_stat(lp, STAT_TX_BYTES);
+		data[2] = axienet_stat(lp, STAT_RX_VLAN_FRAMES);
+		data[3] = axienet_stat(lp, STAT_TX_VLAN_FRAMES);
+		data[6] = axienet_stat(lp, STAT_TX_PFC_FRAMES);
+		data[7] = axienet_stat(lp, STAT_RX_PFC_FRAMES);
+		data[8] = axienet_stat(lp, STAT_USER_DEFINED0);
+		data[9] = axienet_stat(lp, STAT_USER_DEFINED1);
+		data[10] = axienet_stat(lp, STAT_USER_DEFINED2);
+	} while (read_seqcount_retry(&lp->hw_stats_seqcount, start));
+}
+
+static const char axienet_ethtool_stats_strings[][ETH_GSTRING_LEN] = {
+	"Received bytes",
+	"Transmitted bytes",
+	"RX Good VLAN Tagged Frames",
+	"TX Good VLAN Tagged Frames",
+	"TX Good PFC Frames",
+	"RX Good PFC Frames",
+	"User Defined Counter 0",
+	"User Defined Counter 1",
+	"User Defined Counter 2",
+};
+
+static void axienet_ethtools_get_strings(struct net_device *dev, u32
 stringset, u8 *data)
+{
+	switch (stringset) {
+	case ETH_SS_STATS:
+		memcpy(data, axienet_ethtool_stats_strings,
+		       sizeof(axienet_ethtool_stats_strings));
+		break;
+	}
+}
+
+static int axienet_ethtools_get_sset_count(struct net_device *dev, int sset)
+{
+	struct axienet_local *lp = netdev_priv(dev);
+
+	switch (sset) {
+	case ETH_SS_STATS:
+		if (lp->features & XAE_FEATURE_STATS)
+			return ARRAY_SIZE(axienet_ethtool_stats_strings);
+		fallthrough;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static void
+axienet_ethtools_get_pause_stats(struct net_device *dev,
+				 struct ethtool_pause_stats *pause_stats)
+{
+	struct axienet_local *lp = netdev_priv(dev);
+	unsigned int start;
+
+	if (!(lp->features & XAE_FEATURE_STATS))
+		return;
+
+	do {
+		start = read_seqcount_begin(&lp->hw_stats_seqcount);
+		pause_stats->tx_pause_frames =
+			axienet_stat(lp, STAT_TX_PAUSE_FRAMES);
+		pause_stats->rx_pause_frames =
+			axienet_stat(lp, STAT_RX_PAUSE_FRAMES);
+	} while (read_seqcount_retry(&lp->hw_stats_seqcount, start));
+}
+
+static void
+axienet_ethtool_get_eth_mac_stats(struct net_device *dev,
+				  struct ethtool_eth_mac_stats *mac_stats)
+{
+	struct axienet_local *lp = netdev_priv(dev);
+	unsigned int start;
+
+	if (!(lp->features & XAE_FEATURE_STATS))
+		return;
+
+	do {
+		start = read_seqcount_begin(&lp->hw_stats_seqcount);
+		mac_stats->FramesTransmittedOK =
+			axienet_stat(lp, STAT_TX_GOOD_FRAMES);
+		mac_stats->SingleCollisionFrames =
+			axienet_stat(lp,
 STAT_TX_SINGLE_COLLISION_FRAMES);
+		mac_stats->MultipleCollisionFrames =
+			axienet_stat(lp,

STAT_TX_MULTIPLE_COLLISION_FRAMES);

drivers/net/ethernet/xilinx/xilinx_axienet_main.c +45 -0
+		mac_stats->FramesReceivedOK =
+			axienet_stat(lp, STAT_RX_GOOD_FRAMES);
+		mac_stats->FrameCheckSequenceErrors =
+			axienet_stat(lp, STAT_RX_FCS_ERRORS);
+		mac_stats->AlignmentErrors =
+			axienet_stat(lp, STAT_RX_ALIGNMENT_ERRORS);
+		mac_stats->FramesWithDeferredXmissions =
+			axienet_stat(lp, STAT_TX_DEFERRED_FRAMES);
+		mac_stats->LateCollisions =
+			axienet_stat(lp, STAT_TX_LATE_COLLISIONS);
+		mac_stats->FramesAbortedDueToXSColls =
+			axienet_stat(lp, STAT_TX_EXCESS_COLLISIONS);
+		mac_stats->MulticastFramesXmittedOK =
+			axienet_stat(lp, STAT_TX_MULTICAST_FRAMES);
+		mac_stats->BroadcastFramesXmittedOK =
+			axienet_stat(lp, STAT_TX_BROADCAST_FRAMES);
+		mac_stats->FramesWithExcessiveDeferral =
+			axienet_stat(lp, STAT_TX_EXCESS_DEFERRAL);
+		mac_stats->MulticastFramesReceivedOK =
+			axienet_stat(lp, STAT_RX_MULTICAST_FRAMES);
+		mac_stats->BroadcastFramesReceivedOK =
+			axienet_stat(lp, STAT_RX_BROADCAST_FRAMES);
+		mac_stats->InRangeLengthErrors =
+			axienet_stat(lp, STAT_RX_LENGTH_ERRORS);
+	} while (read_seqcount_retry(&lp->hw_stats_seqcount, start));
+}
+
+static void
+axienet_ethtool_get_eth_ctrl_stats(struct net_device *dev,
+				   struct ethtool_eth_ctrl_stats *ctrl_stats)
+{
+	struct axienet_local *lp = netdev_priv(dev);
+	unsigned int start;
+
+	if (!(lp->features & XAE_FEATURE_STATS))
+		return;
+
+	do {
+		start = read_seqcount_begin(&lp->hw_stats_seqcount);
+		ctrl_stats->MACControlFramesTransmitted =
+			axienet_stat(lp, STAT_TX_CONTROL_FRAMES);
+		ctrl_stats->MACControlFramesReceived =
+			axienet_stat(lp, STAT_RX_CONTROL_FRAMES);
+		ctrl_stats->UnsupportedOpcodesReceived =
+			axienet_stat(lp,

STAT_RX_CONTROL_OPCODE_ERRORS);

drivers/net/ethernet/xilinx/xilinx_axienet_main.c +46 -0
+	} while (read_seqcount_retry(&lp->hw_stats_seqcount, start));
+}
+
+static const struct ethtool_rmon_hist_range axienet_rmon_ranges[] = {
+	{   64,    64 },
+	{   65,   127 },
+	{  128,   255 },
+	{  256,   511 },
+	{  512,  1023 },
+	{ 1024,  1518 },
+	{ 1519, 16384 },
+	{ },
+};
+
+static void
+axienet_ethtool_get_rmon_stats(struct net_device *dev,
+			       struct ethtool_rmon_stats *rmon_stats,
+			       const struct ethtool_rmon_hist_range **ranges)
+{
+	struct axienet_local *lp = netdev_priv(dev);
+	unsigned int start;
+
+	if (!(lp->features & XAE_FEATURE_STATS))
+		return;
+
+	do {
+		start = read_seqcount_begin(&lp->hw_stats_seqcount);
+		rmon_stats->undersize_pkts =
+			axienet_stat(lp, STAT_UNDERSIZE_FRAMES);
+		rmon_stats->oversize_pkts =
+			axienet_stat(lp, STAT_RX_OVERSIZE_FRAMES);
+		rmon_stats->fragments =
+			axienet_stat(lp, STAT_FRAGMENT_FRAMES);
+
+		rmon_stats->hist[0] =
+			axienet_stat(lp, STAT_RX_64_BYTE_FRAMES);
+		rmon_stats->hist[1] =
+			axienet_stat(lp, STAT_RX_65_127_BYTE_FRAMES);
+		rmon_stats->hist[2] =
+			axienet_stat(lp, STAT_RX_128_255_BYTE_FRAMES);
+		rmon_stats->hist[3] =
+			axienet_stat(lp, STAT_RX_256_511_BYTE_FRAMES);
+		rmon_stats->hist[4] =
+			axienet_stat(lp, STAT_RX_512_1023_BYTE_FRAMES);
+		rmon_stats->hist[5] =
+			axienet_stat(lp,

STAT_RX_1024_MAX_BYTE_FRAMES);

drivers/net/ethernet/xilinx/xilinx_axienet_main.c +22 -0
+		rmon_stats->hist[6] =
+			rmon_stats->oversize_pkts;
+
+		rmon_stats->hist_tx[0] =
+			axienet_stat(lp, STAT_TX_64_BYTE_FRAMES);
+		rmon_stats->hist_tx[1] =
+			axienet_stat(lp, STAT_TX_65_127_BYTE_FRAMES);
+		rmon_stats->hist_tx[2] =
+			axienet_stat(lp, STAT_TX_128_255_BYTE_FRAMES);
+		rmon_stats->hist_tx[3] =
+			axienet_stat(lp, STAT_TX_256_511_BYTE_FRAMES);
+		rmon_stats->hist_tx[4] =
+			axienet_stat(lp, STAT_TX_512_1023_BYTE_FRAMES);
+		rmon_stats->hist_tx[5] =
+			axienet_stat(lp, STAT_TX_1024_MAX_BYTE_FRAMES);
+		rmon_stats->hist_tx[6] =
+			axienet_stat(lp, STAT_TX_OVERSIZE_FRAMES);
+	} while (read_seqcount_retry(&lp->hw_stats_seqcount, start));
+
+	*ranges = axienet_rmon_ranges;
+}
+
 static const struct ethtool_ops axienet_ethtool_ops = {
 	.supported_coalesce_params = ETHTOOL_COALESCE_MAX_FRAMES

|

drivers/net/ethernet/xilinx/xilinx_axienet_main.c +7 -0
 				     ETHTOOL_COALESCE_USECS,
@@ -2000,6 +2305,13 @@ static const struct ethtool_ops
 axienet_ethtool_ops = {
 	.get_link_ksettings = axienet_ethtools_get_link_ksettings,
 	.set_link_ksettings = axienet_ethtools_set_link_ksettings,
 	.nway_reset	= axienet_ethtools_nway_reset,
+	.get_ethtool_stats = axienet_ethtools_get_ethtool_stats,
+	.get_strings    = axienet_ethtools_get_strings,
+	.get_sset_count = axienet_ethtools_get_sset_count,
+	.get_pause_stats = axienet_ethtools_get_pause_stats,
+	.get_eth_mac_stats = axienet_ethtool_get_eth_mac_stats,
+	.get_eth_ctrl_stats = axienet_ethtool_get_eth_ctrl_stats,
+	.get_rmon_stats = axienet_ethtool_get_rmon_stats,
 };
 

static struct axienet_local *pcs_to_axienet_local(struct phylink_pcs *pcs)

drivers/net/ethernet/xilinx/xilinx_axienet_main.c +4 -0
@@ -2268,6 +2580,10 @@ static int axienet_probe(struct platform_device
 *pdev)
 	u64_stats_init(&lp->rx_stat_sync);
 	u64_stats_init(&lp->tx_stat_sync);
 
+	mutex_init(&lp->stats_lock);
+	seqcount_mutex_init(&lp->hw_stats_seqcount, &lp->stats_lock);
+	INIT_DEFERRABLE_WORK(&lp->stats_work, axienet_refresh_stats);
+
 	lp->axi_clk = devm_clk_get_optional(&pdev->dev, "s_axi_lite_clk");
 	if (!lp->axi_clk) {

/* For backward compatibility, if named AXI clock is not
present,

drivers/net/ethernet/xilinx/xilinx_axienet_main.c +3 -0
@@ -2308,6 +2624,9 @@ static int axienet_probe(struct platform_device
 *pdev)
 	/* Setup checksum offload, but default to off if not specified */
 	lp->features = 0;
 
+	if (axienet_ior(lp, XAE_ABILITY_OFFSET) & XAE_ABILITY_STATS)
+		lp->features |= XAE_FEATURE_STATS;
+
 	ret = of_property_read_u32(pdev->dev.of_node, "xlnx,txcsum",
 &value);
drivers/net/ethernet/xilinx/xilinx_axienet_main.c +0 -1
 	if (!ret) {
 		switch (value) {
--

2.35.1.1320.gc452695387.dirty

#5 lore raw
Quoted 18 lines · Radhey Shyam
Quoted 13 lines

-----Original Message-----
From: Sean Anderson <[email protected]>
Sent: Tuesday, August 20, 2024 11:24 PM
To: Andrew Lunn <[email protected]>; Pandey, Radhey Shyam <[email protected]>; [email protected]
Cc: Simek, Michal <[email protected]>; [email protected]; Russell King <[email protected]>; David S . Miller <[email protected]>; Jakub Kicinski <[email protected]>; Paolo Abeni <[email protected]>; Eric Dumazet <[email protected]>; Simon Horman <[email protected]>; [email protected]; Sean Anderson <[email protected]>
Subject: [PATCH net-next v4 2/2] net: xilinx: axienet: Add statistics support

Quoted 4 lines

Add support for reading the statistics counters, if they are enabled. The counters may be 64-bit, but we can't detect this statically as there's no ability bit for it and the counters are read-only. Therefore, we assume the counters are 32-bits by default. To ensure we don't miss

Quoted 4 lines

Any reason why we can't have DT property to detect if stats counter are configured as 32-bit /64bit? The IP export CONFIG.Statistics_Width and device tree generator can read this IP block property and populate stats width property.

Mainly simplicity:

- We need the functions to work with 32-bit counters anyway
- We can always treat 64-bit counters are 32-bit counters
- The reset issue below necessitates keeping track of a "base" anyway.

And for my devicetrees (generated with 2022.2) all I get is

xlnx,stats = <0x1>;

regardless of whether I select 32- or 64-bit counters. So this wouldn't be something we could reuse from existing devictrees.

Quoted 10 lines
Quoted 7 lines

an overflow, we read all counters at 13-second intervals. This should be often enough to ensure the bytes counters don't wrap at 2.5 Gbit/s.

Another complication is that the counters may be reset when the device is reset (depending on configuration). To ensure the counters persist across link up/down (including suspend/resume), we maintain our own versions along with the last counter value we saw. Because we might wait

Is that a standard convention to retain/persist counter values across link up/down?

IEEE 802.3 section 30.2.1 says

| All counters defined in this specification are assumed to be
| wrap-around counters. Wrap-around counters are those that
| automatically go from their maximum value (or final value) to zero and
| continue to operate. These unsigned counters do not provide for any
| explicit means to return them to their minimum (zero), i.e., reset.

And get_eth_mac_stats implements these counters for Linux. So I would say that resetting the counters on link up/down would be non-conformant.

Other drivers also preserve stats across link up/down. For example, MACB/GEM doesn't reset it stats either. And keeping the stats is also more friendly for users and monitoring tools.

---

If you happen to have an ear with the RTL designers, I would say that saturating, clear-on-read counters would be much easier to work with in software.

--Sean

Quoted 19 lines
Quoted 19 lines

up to 100 ms for the reset to complete, we use a mutex to protect writing hw_stats. We can't sleep in ndo_get_stats64, so we use a seqlock to protect readers.

We don't bother disabling the refresh work when we detect 64-bit counters. This is because the reset issue requires us to read hw_stat_base and reset_in_progress anyway, which would still require the seqcount. And I don't think skipping the task is worth the extra bookkeeping.

We can't use the byte counters for either get_stats64 or get_eth_mac_stats. This is because the byte counters include everything in the frame (destination address to FCS, inclusive). But rtnl_link_stats64 wants bytes excluding the FCS, and ethtool_eth_mac_stats wants to exclude the L2 overhead (addresses and length/type). It might be possible to calculate the byte values Linux expects based on the frame counters, but I think it is simpler to use the existing software counters.

Quoted 5 lines
Quoted 5 lines

get_ethtool_stats is implemented for nonstandard statistics. This includes the aforementioned byte counters, VLAN and PFC frame counters, and user-defined (e.g. with custom RTL) counters.

--Sean

#6 lore raw
Quoted 11 lines
Quoted 2 lines

Is that a standard convention to retain/persist counter values across link up/down?

IEEE 802.3 section 30.2.1 says

| All counters defined in this specification are assumed to be
| wrap-around counters. Wrap-around counters are those that
| automatically go from their maximum value (or final value) to zero and
| continue to operate. These unsigned counters do not provide for any
| explicit means to return them to their minimum (zero), i.e., reset.
Quoted 6 lines

And get_eth_mac_stats implements these counters for Linux. So I would say that resetting the counters on link up/down would be non-conformant.

Other drivers also preserve stats across link up/down. For example, MACB/GEM doesn't reset it stats either. And keeping the stats is also more friendly for users and monitoring tools.

Agreed.

Some driver get this wrong, and clear them. But as a reviewer, it try to spot this.

Andrew

#7 lore raw
In reply to Sean Anderson
This series was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:

Hello:

This series was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:

Quoted 11 lines · On Tue, 20 Aug 2024 13:53:40 -0400 you wrote:

Add support for hardware statistics counters (if they are enabled) in the AXI Ethernet driver. Unfortunately, the implementation is complicated a bit since the hardware might only support 32-bit counters.

Changes in v4:
- Reduce hw_last_counter to u32 to ensure we use (wrapping) 32-bit arithmetic.
- Implement get_ethtool_stats for nonstandard statistics
- Rebase onto net-next/main

[...]

Here is the summary with links:
- [net-next,v4,1/2] net: xilinx: axienet: Report RxRject as rx_dropped

    https://git.kernel.org/netdev/net-next/c/d70e3788da1d
  - [net-next,v4,2/2] net: xilinx: axienet: Add statistics support
    https://git.kernel.org/netdev/net-next/c/76abb5d675c4

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html