[PATCH net V2 2/2] net: xilinx: axienet: Fix BQL accounting for multi-BD TX packets

[PATCH net V2 2/2] net: xilinx: axienet: Fix BQL accounting for multi-BD TX packets

[Suraj Gupta]

When a TX packet spans multiple buffer descriptors (scatter-gather),
the per-BD byte count is accumulated into a local variable that resets
on each NAPI poll. If the BDs for a single packet complete across
different polls, the earlier bytes are lost and never credited to BQL.
This causes BQL to think bytes are permanently in-flight, eventually
stalling the TX queue.

Fix this by replacing the local accumulator with a persistent counter
(tx_compl_bytes) that survives across polls and is reset only after
updating BQL and stats.

Fixes: c900e49d58eb ("net: xilinx: axienet: Implement BQL")
Signed-off-by: Suraj Gupta <suraj.gupta2@amd.com>
---
 drivers/net/ethernet/xilinx/xilinx_axienet.h  |  3 +++
 .../net/ethernet/xilinx/xilinx_axienet_main.c | 20 +++++++++----------
 2 files changed, 13 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet.h b/drivers/net/ethernet/xilinx/xilinx_axienet.h
index 602389843342..a4444c939451 100644
--- a/drivers/net/ethernet/xilinx/xilinx_axienet.h
+++ b/drivers/net/ethernet/xilinx/xilinx_axienet.h
@@ -509,6 +509,8 @@ struct skbuf_dma_descriptor {
  *		complete. Only updated at runtime by TX NAPI poll.
  * @tx_bd_tail:	Stores the index of the next Tx buffer descriptor in the ring
  *              to be populated.
+ * @tx_compl_bytes: Accumulates TX completion length until a full packet is
+ *              reported to the stack.
  * @tx_packets: TX packet count for statistics
  * @tx_bytes:	TX byte count for statistics
  * @tx_stat_sync: Synchronization object for TX stats
@@ -592,6 +594,7 @@ struct axienet_local {
 	u32 tx_bd_num;
 	u32 tx_bd_ci;
 	u32 tx_bd_tail;
+	u32 tx_compl_bytes;
 	u64_stats_t tx_packets;
 	u64_stats_t tx_bytes;
 	struct u64_stats_sync tx_stat_sync;
diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
index b06e4c37ff61..95bf61986cb7 100644
--- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
+++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
@@ -692,6 +692,8 @@ static void axienet_dma_stop(struct axienet_local *lp)
 	axienet_lock_mii(lp);
 	__axienet_device_reset(lp);
 	axienet_unlock_mii(lp);
+
+	lp->tx_compl_bytes = 0;
 }
 
 /**
@@ -770,8 +772,6 @@ static int axienet_device_reset(struct net_device *ndev)
  * @first_bd:	Index of first descriptor to clean up
  * @nr_bds:	Max number of descriptors to clean up
  * @force:	Whether to clean descriptors even if not complete
- * @sizep:	Pointer to a u32 filled with the total sum of all bytes
- *		in all cleaned-up descriptors. Ignored if NULL.
  * @budget:	NAPI budget (use 0 when not called from NAPI poll)
  *
  * Would either be called after a successful transmit operation, or after
@@ -780,7 +780,7 @@ static int axienet_device_reset(struct net_device *ndev)
  * Return: The number of packets handled.
  */
 static int axienet_free_tx_chain(struct axienet_local *lp, u32 first_bd,
-				 int nr_bds, bool force, u32 *sizep, int budget)
+				 int nr_bds, bool force, int budget)
 {
 	struct axidma_bd *cur_p;
 	unsigned int status;
@@ -819,8 +819,8 @@ static int axienet_free_tx_chain(struct axienet_local *lp, u32 first_bd,
 		cur_p->cntrl = 0;
 		cur_p->status = 0;
 
-		if (sizep)
-			*sizep += status & XAXIDMA_BD_STS_ACTUAL_LEN_MASK;
+		if (!force)
+			lp->tx_compl_bytes += status & XAXIDMA_BD_STS_ACTUAL_LEN_MASK;
 	}
 
 	if (!force) {
@@ -999,18 +999,18 @@ static int axienet_tx_poll(struct napi_struct *napi, int budget)
 {
 	struct axienet_local *lp = container_of(napi, struct axienet_local, napi_tx);
 	struct net_device *ndev = lp->ndev;
-	u32 size = 0;
 	int packets;
 
 	packets = axienet_free_tx_chain(lp, lp->tx_bd_ci, lp->tx_bd_num, false,
-					&size, budget);
+					budget);
 
 	if (packets) {
-		netdev_completed_queue(ndev, packets, size);
+		netdev_completed_queue(ndev, packets, lp->tx_compl_bytes);
 		u64_stats_update_begin(&lp->tx_stat_sync);
 		u64_stats_add(&lp->tx_packets, packets);
-		u64_stats_add(&lp->tx_bytes, size);
+		u64_stats_add(&lp->tx_bytes, lp->tx_compl_bytes);
 		u64_stats_update_end(&lp->tx_stat_sync);
+		lp->tx_compl_bytes = 0;
 
 		/* Matches barrier in axienet_start_xmit */
 		smp_mb();
@@ -1115,7 +1115,7 @@ axienet_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 				netdev_err(ndev, "TX DMA mapping error\n");
 			ndev->stats.tx_dropped++;
 			axienet_free_tx_chain(lp, orig_tail_ptr, ii + 1,
-					      true, NULL, 0);
+					      true, 0);
 			dev_kfree_skb_any(skb);
 			return NETDEV_TX_OK;
 		}
-- 
2.49.1

[PATCH net V2 0/2] Correct BD length masks and BQL accounting for multi-BD TX packets

[Suraj Gupta]

This patch series fixes two issues in the Xilinx AXI Ethernet driver:
 1. Corrects the BD length masks to match the AXIDMA IP spec.
 2. Fixes BQL accounting for multi-BD TX packets.

---
Changes in V2:
- Define BD length masks with GENMASK(25, 0) per PG021.
- Credit BQL using skb->len on packet completion instead of summing
  per-BD actual lengths from descriptor status.
---

Suraj Gupta (2):
  net: xilinx: axienet: Correct BD length masks to match AXIDMA IP spec
  net: xilinx: axienet: Fix BQL accounting for multi-BD TX packets

 drivers/net/ethernet/xilinx/xilinx_axienet.h      | 4 ++--
 drivers/net/ethernet/xilinx/xilinx_axienet_main.c | 9 ++++-----
 2 files changed, 6 insertions(+), 7 deletions(-)

-- 
2.49.1

[PATCH net V2 1/2] net: xilinx: axienet: Correct BD length masks to match AXIDMA IP spec

[Suraj Gupta]

The XAXIDMA_BD_CTRL_LENGTH_MASK and XAXIDMA_BD_STS_ACTUAL_LEN_MASK
macros were defined as 0x007FFFFF (23 bits), but the AXI DMA IP
product guide (PG021) specifies the buffer length field as bits 25:0
(26 bits). Update both masks to match the IP documentation.

In practice this had no functional impact, since Ethernet frames are
far smaller than 2^23 bytes and the extra bits were always zero, but
the masks should still reflect the hardware specification.

Fixes: 8a3b7a252dca ("drivers/net/ethernet/xilinx: added Xilinx AXI Ethernet driver")
Signed-off-by: Suraj Gupta <suraj.gupta2@amd.com>
---
 drivers/net/ethernet/xilinx/xilinx_axienet.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet.h b/drivers/net/ethernet/xilinx/xilinx_axienet.h
index 5ff742103beb..fcd3aaef27fc 100644
--- a/drivers/net/ethernet/xilinx/xilinx_axienet.h
+++ b/drivers/net/ethernet/xilinx/xilinx_axienet.h
@@ -105,7 +105,7 @@
 #define XAXIDMA_BD_HAS_DRE_MASK		0xF00 /* Whether has DRE mask */
 #define XAXIDMA_BD_WORDLEN_MASK		0xFF /* Whether has DRE mask */
 
-#define XAXIDMA_BD_CTRL_LENGTH_MASK	0x007FFFFF /* Requested len */
+#define XAXIDMA_BD_CTRL_LENGTH_MASK	GENMASK(25, 0) /* Requested len */
 #define XAXIDMA_BD_CTRL_TXSOF_MASK	0x08000000 /* First tx packet */
 #define XAXIDMA_BD_CTRL_TXEOF_MASK	0x04000000 /* Last tx packet */
 #define XAXIDMA_BD_CTRL_ALL_MASK	0x0C000000 /* All control bits */
@@ -130,7 +130,7 @@
 #define XAXIDMA_BD_CTRL_TXEOF_MASK	0x04000000 /* Last tx packet */
 #define XAXIDMA_BD_CTRL_ALL_MASK	0x0C000000 /* All control bits */
 
-#define XAXIDMA_BD_STS_ACTUAL_LEN_MASK	0x007FFFFF /* Actual len */
+#define XAXIDMA_BD_STS_ACTUAL_LEN_MASK	GENMASK(25, 0) /* Actual len */
 #define XAXIDMA_BD_STS_COMPLETE_MASK	0x80000000 /* Completed */
 #define XAXIDMA_BD_STS_DEC_ERR_MASK	0x40000000 /* Decode error */
 #define XAXIDMA_BD_STS_SLV_ERR_MASK	0x20000000 /* Slave error */
-- 
2.49.1

[PATCH net V2 2/2] net: xilinx: axienet: Fix BQL accounting for multi-BD TX packets

[Suraj Gupta]

When a TX packet spans multiple buffer descriptors (scatter-gather),
axienet_free_tx_chain sums the per-BD actual length from descriptor
status into a caller-provided accumulator. That sum is reset on each
NAPI poll. If the BDs for a single packet complete across different
polls, the earlier bytes are lost and never credited to BQL. This
causes BQL to think bytes are permanently in-flight, eventually
stalling the TX queue.

The SKB pointer is stored only on the last BD of a packet. When that
BD completes, use skb->len for the byte count instead of summing
per-BD status lengths. This matches netdev_sent_queue(), which debits
skb->len, and naturally survives across polls because no partial
packet contributes to the accumulator.

Fixes: c900e49d58eb ("net: xilinx: axienet: Implement BQL")
Signed-off-by: Suraj Gupta <suraj.gupta2@amd.com>
---
 drivers/net/ethernet/xilinx/xilinx_axienet_main.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
index b06e4c37ff61..263c4b67fd5a 100644
--- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
+++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
@@ -770,8 +770,8 @@ static int axienet_device_reset(struct net_device *ndev)
  * @first_bd:	Index of first descriptor to clean up
  * @nr_bds:	Max number of descriptors to clean up
  * @force:	Whether to clean descriptors even if not complete
- * @sizep:	Pointer to a u32 filled with the total sum of all bytes
- *		in all cleaned-up descriptors. Ignored if NULL.
+ * @sizep:	Pointer to a u32 accumulating the total byte count of
+ *		completed packets (using skb->len). Ignored if NULL.
  * @budget:	NAPI budget (use 0 when not called from NAPI poll)
  *
  * Would either be called after a successful transmit operation, or after
@@ -805,6 +805,8 @@ static int axienet_free_tx_chain(struct axienet_local *lp, u32 first_bd,
 				 DMA_TO_DEVICE);
 
 		if (cur_p->skb && (status & XAXIDMA_BD_STS_COMPLETE_MASK)) {
+			if (sizep)
+				*sizep += cur_p->skb->len;
 			napi_consume_skb(cur_p->skb, budget);
 			packets++;
 		}
@@ -818,9 +820,6 @@ static int axienet_free_tx_chain(struct axienet_local *lp, u32 first_bd,
 		wmb();
 		cur_p->cntrl = 0;
 		cur_p->status = 0;
-
-		if (sizep)
-			*sizep += status & XAXIDMA_BD_STS_ACTUAL_LEN_MASK;
 	}
 
 	if (!force) {
-- 
2.49.1