[PATCH] net/mlx5e: don't assume psp tx skbs are ipv6 csum handling
From: Daniel Zahka <daniel.zahka@gmail.com>
Date: 2026-01-23 18:11:17
Also in:
linux-rdma, lkml
Subsystem:
mellanox ethernet driver (mlx5e), mellanox ethernet innova drivers, mellanox mlx5 core vpi driver, networking drivers, the rest · Maintainers:
Saeed Mahameed, Tariq Toukan, Mark Bloch, Leon Romanovsky, Andrew Lunn, "David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, Linus Torvalds
mlx5e_psp_handle_tx_skb() assumes skbs are ipv6 when doing a partial
TCP checksum with tso. Make correctly mlx5e_psp_handle_tx_skb() handle
ipv4 packets.
Fixes: e5a1861a298e ("net/mlx5e: Implement PSP Tx data path")
Signed-off-by: Daniel Zahka <daniel.zahka@gmail.com>
---
This is a bug when an ipv4 tx skb passes through
mlx5e_psp_handle_tx_skb() and tso is requested. It was previously
undetected in my testing because my setup involves cx7's on both ends,
and mlx5e_handle_csum() marks PSP rx skb's with csum_unnecessary.
To reproduce the problem just turn off NETIF_F_RXCSUM and observe:
nstat -a | grep TcpInCsumErrors
---
.../net/ethernet/mellanox/mlx5/core/en_accel/psp_rxtx.c | 15 +++++++++++----
1 file changed, 11 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp_rxtx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp_rxtx.c
index c17ea0fcd8ef..15a344ad471d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp_rxtx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/psp_rxtx.c@@ -178,7 +178,9 @@ bool mlx5e_psp_handle_tx_skb(struct net_device *netdev, struct mlx5e_priv *priv = netdev_priv(netdev); struct net *net = sock_net(skb->sk); const struct ipv6hdr *ip6; + const struct iphdr *ip; struct tcphdr *th; + int len; if (!mlx5e_psp_set_state(priv, skb, psp_st)) return true;
@@ -190,11 +192,16 @@ bool mlx5e_psp_handle_tx_skb(struct net_device *netdev, return false; } if (skb_is_gso(skb)) { - ip6 = ipv6_hdr(skb); th = inner_tcp_hdr(skb); - - th->check = ~tcp_v6_check(skb_shinfo(skb)->gso_size + inner_tcp_hdrlen(skb), &ip6->saddr, - &ip6->daddr, 0); + len = skb_shinfo(skb)->gso_size + inner_tcp_hdrlen(skb); + + if (skb->protocol == htons(ETH_P_IP)) { + ip = ip_hdr(skb); + th->check = ~tcp_v4_check(len, ip->saddr, ip->daddr, 0); + } else { + ip6 = ipv6_hdr(skb); + th->check = ~tcp_v6_check(len, &ip6->saddr, &ip6->daddr, 0); + } } return true;
--- base-commit: 4a3dba48188208e4f66822800e042686784d29d1 change-id: 20260122-dzahka-fix-tx-csum-partial-952e8dc28375 Best regards, -- Daniel Zahka [off-list ref]