[PATCH net V2] net: xilinx: axienet: Fix RX skb ring management in DMAengine mode

[PATCH net V2] net: xilinx: axienet: Fix RX skb ring management in DMAengine mode

[Suraj Gupta]

Submit multiple descriptors in axienet_rx_cb() to fill Rx skb ring. This
ensures the ring "catches up" on previously missed allocations.

Increment Rx skb ring head pointer after BD is successfully allocated.
Previously, head pointer was incremented before verifying if descriptor is
successfully allocated and has valid entries, which could lead to ring
state inconsistency if descriptor setup failed.

These changes improve reliability by maintaining adequate descriptor
availability and ensuring proper ring buffer state management.

Fixes: 6a91b846af85 ("net: axienet: Introduce dmaengine support")
Signed-off-by: Suraj Gupta <suraj.gupta2@amd.com>
---
Changes in V2:
- Submit multiple descriptors in axienet_rx_cb().
- Modify commit subject and description.
---
 drivers/net/ethernet/xilinx/xilinx_axienet_main.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
index 6011d7eae0c7..0d8a05fe541a 100644
--- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
+++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
@@ -1160,6 +1160,7 @@ static void axienet_dma_rx_cb(void *data, const struct dmaengine_result *result)
 	struct axienet_local *lp = data;
 	struct sk_buff *skb;
 	u32 *app_metadata;
+	int i;
 
 	skbuf_dma = axienet_get_rx_desc(lp, lp->rx_ring_tail++);
 	skb = skbuf_dma->skb;
@@ -1178,7 +1179,10 @@ static void axienet_dma_rx_cb(void *data, const struct dmaengine_result *result)
 	u64_stats_add(&lp->rx_packets, 1);
 	u64_stats_add(&lp->rx_bytes, rx_len);
 	u64_stats_update_end(&lp->rx_stat_sync);
-	axienet_rx_submit_desc(lp->ndev);
+
+	for (i = 0; i < CIRC_SPACE(lp->rx_ring_head, lp->rx_ring_tail,
+				   RX_BUF_NUM_DEFAULT); i++)
+		axienet_rx_submit_desc(lp->ndev);
 	dma_async_issue_pending(lp->rx_chan);
 }
 
@@ -1457,7 +1461,6 @@ static void axienet_rx_submit_desc(struct net_device *ndev)
 	if (!skbuf_dma)
 		return;
 
-	lp->rx_ring_head++;
 	skb = netdev_alloc_skb(ndev, lp->max_frm_size);
 	if (!skb)
 		return;
@@ -1482,6 +1485,7 @@ static void axienet_rx_submit_desc(struct net_device *ndev)
 	skbuf_dma->desc = dma_rx_desc;
 	dma_rx_desc->callback_param = lp;
 	dma_rx_desc->callback_result = axienet_dma_rx_cb;
+	lp->rx_ring_head++;
 	dmaengine_submit(dma_rx_desc);
 
 	return;
-- 
2.25.1

Re: [PATCH net V2] net: xilinx: axienet: Fix RX skb ring management in DMAengine mode

[patchwork-bot+netdevbpf]

Hello:

This patch was applied to netdev/net.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Wed, 13 Aug 2025 19:25:59 +0530 you wrote:
> Submit multiple descriptors in axienet_rx_cb() to fill Rx skb ring. This
> ensures the ring "catches up" on previously missed allocations.
> 
> Increment Rx skb ring head pointer after BD is successfully allocated.
> Previously, head pointer was incremented before verifying if descriptor is
> successfully allocated and has valid entries, which could lead to ring
> state inconsistency if descriptor setup failed.
> 
> [...]

Here is the summary with links:
  - [net,V2] net: xilinx: axienet: Fix RX skb ring management in DMAengine mode
    https://git.kernel.org/netdev/net/c/fd980bf6e9cd

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html