From 75cea9860aa6b2350d90a8d78fed114d27c7eca2 Mon Sep 17 00:00:00 2001 From: Michal Pecio Date: Tue, 14 Oct 2025 20:35:28 +0200 Subject: net: usb: rtl8150: Fix frame padding TX frames aren't padded and unknown memory is sent into the ether. Theoretically, it isn't even guaranteed that the extra memory exists and can be sent out, which could cause further problems. In practice, I found that plenty of tailroom exists in the skb itself (in my test with ping at least) and skb_padto() easily succeeds, so use it here. In the event of -ENOMEM drop the frame like other drivers do. The use of one more padding byte instead of a USB zero-length packet is retained to avoid regression. I have a dodgy Etron xHCI controller which doesn't seem to support sending ZLPs at all. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org Signed-off-by: Michal Pecio Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251014203528.3f9783c4.michal.pecio@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/usb/rtl8150.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/usb/rtl8150.c b/drivers/net/usb/rtl8150.c index 92add3daadbb..278e6cb6f4d9 100644 --- a/drivers/net/usb/rtl8150.c +++ b/drivers/net/usb/rtl8150.c @@ -685,9 +685,16 @@ static netdev_tx_t rtl8150_start_xmit(struct sk_buff *skb, rtl8150_t *dev = netdev_priv(netdev); int count, res; + /* pad the frame and ensure terminating USB packet, datasheet 9.2.3 */ + count = max(skb->len, ETH_ZLEN); + if (count % 64 == 0) + count++; + if (skb_padto(skb, count)) { + netdev->stats.tx_dropped++; + return NETDEV_TX_OK; + } + netif_stop_queue(netdev); - count = (skb->len < 60) ? 60 : skb->len; - count = (count & 0x3f) ? count : count + 1; dev->tx_skb = skb; usb_fill_bulk_urb(dev->tx_urb, dev->udev, usb_sndbulkpipe(dev->udev, 2), skb->data, count, write_bulk_callback, dev); -- cgit v1.2.3 From aaf043a5688114703ae2c1482b92e7e0754d684e Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Tue, 14 Oct 2025 13:46:49 -0700 Subject: net/mlx5e: Return 1 instead of 0 in invalid case in mlx5e_mpwrq_umr_entry_size() When building with Clang 20 or newer, there are some objtool warnings from unexpected fallthroughs to other functions: vmlinux.o: warning: objtool: mlx5e_mpwrq_mtts_per_wqe() falls through to next function mlx5e_mpwrq_max_num_entries() vmlinux.o: warning: objtool: mlx5e_mpwrq_max_log_rq_size() falls through to next function mlx5e_get_linear_rq_headroom() LLVM 20 contains an (admittedly problematic [1]) optimization [2] to convert divide by zero into the equivalent of __builtin_unreachable(), which invokes undefined behavior and destroys code generation when it is encountered in a control flow graph. mlx5e_mpwrq_umr_entry_size() returns 0 in the default case of an unrecognized mlx5e_mpwrq_umr_mode value. mlx5e_mpwrq_mtts_per_wqe(), which is inlined into mlx5e_mpwrq_max_log_rq_size(), uses the result of mlx5e_mpwrq_umr_entry_size() in a divide operation without checking for zero, so LLVM is able to infer there will be a divide by zero in this case and invokes undefined behavior. While there is some proposed work to isolate this undefined behavior and avoid the destructive code generation that results in these objtool warnings, code should still be defensive against divide by zero. As the WARN_ONCE() implies that an invalid value should be handled gracefully, return 1 instead of 0 in the default case so that the results of this division operation is always valid. Fixes: 168723c1f8d6 ("net/mlx5e: xsk: Use umr_mode to calculate striding RQ parameters") Link: https://lore.kernel.org/CAGG=3QUk8-Ak7YKnRziO4=0z=1C_7+4jF+6ZeDQ9yF+kuTOHOQ@mail.gmail.com/ [1] Link: https://github.com/llvm/llvm-project/commit/37932643abab699e8bb1def08b7eb4eae7ff1448 [2] Closes: https://github.com/ClangBuiltLinux/linux/issues/2131 Closes: https://github.com/ClangBuiltLinux/linux/issues/2132 Signed-off-by: Nathan Chancellor Reviewed-by: Tariq Toukan Link: https://patch.msgid.link/20251014-mlx5e-avoid-zero-div-from-mlx5e_mpwrq_umr_entry_size-v1-1-dc186b8819ef@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en/params.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c index 3692298e10f2..c9bdee9a8b30 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c @@ -100,7 +100,7 @@ u8 mlx5e_mpwrq_umr_entry_size(enum mlx5e_mpwrq_umr_mode mode) return sizeof(struct mlx5_ksm) * 4; } WARN_ONCE(1, "MPWRQ UMR mode %d is not known\n", mode); - return 0; + return 1; } u8 mlx5e_mpwrq_log_wqe_sz(struct mlx5_core_dev *mdev, u8 page_shift, -- cgit v1.2.3 From 5348d6312446929edabced02bd6438bfe5220e31 Mon Sep 17 00:00:00 2001 From: Cosmin Ratiu Date: Wed, 15 Oct 2025 10:05:23 +0300 Subject: net/mlx5e: psp, avoid 'accel' NULL pointer dereference The 'accel' parameter of mlx5e_txwqe_build_eseg_csum() and the similar 'state' parameter of mlx5e_accel_tx_ids_len() were NULL when called from mlx5i_sq_xmit() and were causing kernel panics from that context. Fix that by passing in a local empty mlx5e_accel_tx_state variable, thus guaranteeing that 'accel' is never NULL. Also remove an unnecessary check from mlx5e_tx_wqe_inline_mode(). Fixes: e5a1861a298e ("net/mlx5e: Implement PSP Tx data path") Signed-off-by: Cosmin Ratiu Reviewed-by: Dragos Tatulea Signed-off-by: Tariq Toukan Reviewed-by: Michal Swiatkowski Link: https://patch.msgid.link/1760511923-890650-1-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en_tx.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index b7227afcb51d..2702b3885f06 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -256,7 +256,7 @@ mlx5e_tx_wqe_inline_mode(struct mlx5e_txqsq *sq, struct sk_buff *skb, u8 mode; #ifdef CONFIG_MLX5_EN_TLS - if (accel && accel->tls.tls_tisn) + if (accel->tls.tls_tisn) return MLX5_INLINE_MODE_TCP_UDP; #endif @@ -982,6 +982,7 @@ void mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, struct mlx5e_tx_attr attr; struct mlx5i_tx_wqe *wqe; + struct mlx5e_accel_tx_state accel = {}; struct mlx5_wqe_datagram_seg *datagram; struct mlx5_wqe_ctrl_seg *cseg; struct mlx5_wqe_eth_seg *eseg; @@ -992,7 +993,7 @@ void mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, int num_dma; u16 pi; - mlx5e_sq_xmit_prepare(sq, skb, NULL, &attr); + mlx5e_sq_xmit_prepare(sq, skb, &accel, &attr); mlx5i_sq_calc_wqe_attr(skb, &attr, &wqe_attr); pi = mlx5e_txqsq_get_next_pi(sq, wqe_attr.num_wqebbs); @@ -1009,7 +1010,7 @@ void mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, mlx5i_txwqe_build_datagram(av, dqpn, dqkey, datagram); - mlx5e_txwqe_build_eseg_csum(sq, skb, NULL, eseg); + mlx5e_txwqe_build_eseg_csum(sq, skb, &accel, eseg); eseg->mss = attr.mss; -- cgit v1.2.3 From 7f864458e9a6d2000b726d14b3d3a706ac92a3b0 Mon Sep 17 00:00:00 2001 From: Sebastian Reichel Date: Tue, 14 Oct 2025 17:49:34 +0200 Subject: net: stmmac: dwmac-rk: Fix disabling set_clock_selection On all platforms set_clock_selection() writes to a GRF register. This requires certain clocks running and thus should happen before the clocks are disabled. This has been noticed on RK3576 Sige5, which hangs during system suspend when trying to suspend the second network interface. Note, that suspending the first interface works, because the second device ensures that the necessary clocks for the GRF are enabled. Cc: stable@vger.kernel.org Fixes: 2f2b60a0ec28 ("net: ethernet: stmmac: dwmac-rk: Add gmac support for rk3588") Signed-off-by: Sebastian Reichel Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251014-rockchip-network-clock-fix-v1-1-c257b4afdf75@collabora.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c index 51ea0caf16c1..0786816e05f0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c @@ -1446,14 +1446,15 @@ static int gmac_clk_enable(struct rk_priv_data *bsp_priv, bool enable) } } else { if (bsp_priv->clk_enabled) { + if (bsp_priv->ops && bsp_priv->ops->set_clock_selection) { + bsp_priv->ops->set_clock_selection(bsp_priv, + bsp_priv->clock_input, false); + } + clk_bulk_disable_unprepare(bsp_priv->num_clks, bsp_priv->clks); clk_disable_unprepare(bsp_priv->clk_phy); - if (bsp_priv->ops && bsp_priv->ops->set_clock_selection) - bsp_priv->ops->set_clock_selection(bsp_priv, - bsp_priv->clock_input, false); - bsp_priv->clk_enabled = false; } } -- cgit v1.2.3 From 50bd33f6b3922a6b760aa30d409cae891cec8fb5 Mon Sep 17 00:00:00 2001 From: Jianpeng Chang Date: Wed, 15 Oct 2025 10:14:27 +0800 Subject: net: enetc: fix the deadlock of enetc_mdio_lock After applying the workaround for err050089, the LS1028A platform experiences RCU stalls on RT kernel. This issue is caused by the recursive acquisition of the read lock enetc_mdio_lock. Here list some of the call stacks identified under the enetc_poll path that may lead to a deadlock: enetc_poll -> enetc_lock_mdio -> enetc_clean_rx_ring OR napi_complete_done -> napi_gro_receive -> enetc_start_xmit -> enetc_lock_mdio -> enetc_map_tx_buffs -> enetc_unlock_mdio -> enetc_unlock_mdio After enetc_poll acquires the read lock, a higher-priority writer attempts to acquire the lock, causing preemption. The writer detects that a read lock is already held and is scheduled out. However, readers under enetc_poll cannot acquire the read lock again because a writer is already waiting, leading to a thread hang. Currently, the deadlock is avoided by adjusting enetc_lock_mdio to prevent recursive lock acquisition. Fixes: 6d36ecdbc441 ("net: enetc: take the MDIO lock only once per NAPI poll cycle") Signed-off-by: Jianpeng Chang Acked-by: Wei Fang Link: https://patch.msgid.link/20251015021427.180757-1-jianpeng.chang.cn@windriver.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/enetc/enetc.c | 25 +++++++++++++++++++++---- 1 file changed, 21 insertions(+), 4 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c index aae462a0cf5a..0535e92404e3 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc.c +++ b/drivers/net/ethernet/freescale/enetc/enetc.c @@ -1595,6 +1595,8 @@ static int enetc_clean_rx_ring(struct enetc_bdr *rx_ring, /* next descriptor to process */ i = rx_ring->next_to_clean; + enetc_lock_mdio(); + while (likely(rx_frm_cnt < work_limit)) { union enetc_rx_bd *rxbd; struct sk_buff *skb; @@ -1630,7 +1632,9 @@ static int enetc_clean_rx_ring(struct enetc_bdr *rx_ring, rx_byte_cnt += skb->len + ETH_HLEN; rx_frm_cnt++; + enetc_unlock_mdio(); napi_gro_receive(napi, skb); + enetc_lock_mdio(); } rx_ring->next_to_clean = i; @@ -1638,6 +1642,8 @@ static int enetc_clean_rx_ring(struct enetc_bdr *rx_ring, rx_ring->stats.packets += rx_frm_cnt; rx_ring->stats.bytes += rx_byte_cnt; + enetc_unlock_mdio(); + return rx_frm_cnt; } @@ -1947,6 +1953,8 @@ static int enetc_clean_rx_ring_xdp(struct enetc_bdr *rx_ring, /* next descriptor to process */ i = rx_ring->next_to_clean; + enetc_lock_mdio(); + while (likely(rx_frm_cnt < work_limit)) { union enetc_rx_bd *rxbd, *orig_rxbd; struct xdp_buff xdp_buff; @@ -2010,7 +2018,9 @@ static int enetc_clean_rx_ring_xdp(struct enetc_bdr *rx_ring, */ enetc_bulk_flip_buff(rx_ring, orig_i, i); + enetc_unlock_mdio(); napi_gro_receive(napi, skb); + enetc_lock_mdio(); break; case XDP_TX: tx_ring = priv->xdp_tx_ring[rx_ring->index]; @@ -2045,7 +2055,9 @@ static int enetc_clean_rx_ring_xdp(struct enetc_bdr *rx_ring, } break; case XDP_REDIRECT: + enetc_unlock_mdio(); err = xdp_do_redirect(rx_ring->ndev, &xdp_buff, prog); + enetc_lock_mdio(); if (unlikely(err)) { enetc_xdp_drop(rx_ring, orig_i, i); rx_ring->stats.xdp_redirect_failures++; @@ -2065,8 +2077,11 @@ out: rx_ring->stats.packets += rx_frm_cnt; rx_ring->stats.bytes += rx_byte_cnt; - if (xdp_redirect_frm_cnt) + if (xdp_redirect_frm_cnt) { + enetc_unlock_mdio(); xdp_do_flush(); + enetc_lock_mdio(); + } if (xdp_tx_frm_cnt) enetc_update_tx_ring_tail(tx_ring); @@ -2075,6 +2090,8 @@ out: enetc_refill_rx_ring(rx_ring, enetc_bd_unused(rx_ring) - rx_ring->xdp.xdp_tx_in_flight); + enetc_unlock_mdio(); + return rx_frm_cnt; } @@ -2093,6 +2110,7 @@ static int enetc_poll(struct napi_struct *napi, int budget) for (i = 0; i < v->count_tx_rings; i++) if (!enetc_clean_tx_ring(&v->tx_ring[i], budget)) complete = false; + enetc_unlock_mdio(); prog = rx_ring->xdp.prog; if (prog) @@ -2104,10 +2122,8 @@ static int enetc_poll(struct napi_struct *napi, int budget) if (work_done) v->rx_napi_work = true; - if (!complete) { - enetc_unlock_mdio(); + if (!complete) return budget; - } napi_complete_done(napi, work_done); @@ -2116,6 +2132,7 @@ static int enetc_poll(struct napi_struct *napi, int budget) v->rx_napi_work = false; + enetc_lock_mdio(); /* enable interrupts */ enetc_wr_reg_hot(v->rbier, ENETC_RBIER_RXTIE); -- cgit v1.2.3 From e59bc32df2e989f034623a580e30a2a72af33b3f Mon Sep 17 00:00:00 2001 From: Wei Fang Date: Thu, 16 Oct 2025 16:01:31 +0800 Subject: net: enetc: correct the value of ENETC_RXB_TRUESIZE The ENETC RX ring uses the page halves flipping mechanism, each page is split into two halves for the RX ring to use. And ENETC_RXB_TRUESIZE is defined to 2048 to indicate the size of half a page. However, the page size is configurable, for ARM64 platform, PAGE_SIZE is default to 4K, but it could be configured to 16K or 64K. When PAGE_SIZE is set to 16K or 64K, ENETC_RXB_TRUESIZE is not correct, and the RX ring will always use the first half of the page. This is not consistent with the description in the relevant kernel doc and commit messages. This issue is invisible in most cases, but if users want to increase PAGE_SIZE to receive a Jumbo frame with a single buffer for some use cases, it will not work as expected, because the buffer size of each RX BD is fixed to 2048 bytes. Based on the above two points, we expect to correct ENETC_RXB_TRUESIZE to (PAGE_SIZE >> 1), as described in the comment. Fixes: d4fd0404c1c9 ("enetc: Introduce basic PF and VF ENETC ethernet drivers") Signed-off-by: Wei Fang Reviewed-by: Claudiu Manoil Link: https://patch.msgid.link/20251016080131.3127122-1-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/enetc/enetc.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/freescale/enetc/enetc.h b/drivers/net/ethernet/freescale/enetc/enetc.h index 0ec010a7d640..f279fa597991 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc.h +++ b/drivers/net/ethernet/freescale/enetc/enetc.h @@ -76,7 +76,7 @@ struct enetc_lso_t { #define ENETC_LSO_MAX_DATA_LEN SZ_256K #define ENETC_RX_MAXFRM_SIZE ENETC_MAC_MAXFRM_SIZE -#define ENETC_RXB_TRUESIZE 2048 /* PAGE_SIZE >> 1 */ +#define ENETC_RXB_TRUESIZE (PAGE_SIZE >> 1) #define ENETC_RXB_PAD NET_SKB_PAD /* add extra space if needed */ #define ENETC_RXB_DMA_SIZE \ (SKB_WITH_OVERHEAD(ENETC_RXB_TRUESIZE) - ENETC_RXB_PAD) -- cgit v1.2.3 From e0caeb24f538c3c9c94f471882ceeb43d9dc2739 Mon Sep 17 00:00:00 2001 From: Tonghao Zhang Date: Thu, 16 Oct 2025 20:51:36 +0800 Subject: net: bonding: update the slave array for broadcast mode This patch fixes ce7a381697cb ("net: bonding: add broadcast_neighbor option for 802.3ad"). Before this commit, on the broadcast mode, all devices were traversed using the bond_for_each_slave_rcu. This patch supports traversing devices by using all_slaves. Therefore, we need to update the slave array when enslave or release slave. Fixes: ce7a381697cb ("net: bonding: add broadcast_neighbor option for 802.3ad") Cc: Simon Horman Cc: Jonathan Corbet Cc: Andrew Lunn Cc: Reported-by: Jiri Slaby Tested-by: Jiri Slaby Link: https://lore.kernel.org/all/a97e6e1e-81bc-4a79-8352-9e4794b0d2ca@kernel.org/ Signed-off-by: Tonghao Zhang Reviewed-by: Hangbin Liu Reviewed-by: Nikolay Aleksandrov Acked-by: Jay Vosburgh Link: https://patch.msgid.link/20251016125136.16568-1-tonghao@bamaicloud.com Signed-off-by: Jakub Kicinski --- drivers/net/bonding/bond_main.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 4da619210c1f..67fdcbdd2764 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -2287,7 +2287,9 @@ skip_mac_set: unblock_netpoll_tx(); } - if (bond_mode_can_use_xmit_hash(bond)) + /* broadcast mode uses the all_slaves to loop through slaves. */ + if (bond_mode_can_use_xmit_hash(bond) || + BOND_MODE(bond) == BOND_MODE_BROADCAST) bond_update_slave_arr(bond, NULL); if (!slave_dev->netdev_ops->ndo_bpf || @@ -2463,7 +2465,8 @@ static int __bond_release_one(struct net_device *bond_dev, bond_upper_dev_unlink(bond, slave); - if (bond_mode_can_use_xmit_hash(bond)) + if (bond_mode_can_use_xmit_hash(bond) || + BOND_MODE(bond) == BOND_MODE_BROADCAST) bond_update_slave_arr(bond, slave); slave_info(bond_dev, slave_dev, "Releasing %s interface\n", -- cgit v1.2.3 From 902e81e679d86846a2404630d349709ad9372d0d Mon Sep 17 00:00:00 2001 From: Ioana Ciornei Date: Thu, 16 Oct 2025 16:58:07 +0300 Subject: dpaa2-eth: fix the pointer passed to PTR_ALIGN on Tx path The blamed commit increased the needed headroom to account for alignment. This means that the size required to always align a Tx buffer was added inside the dpaa2_eth_needed_headroom() function. By doing that, a manual adjustment of the pointer passed to PTR_ALIGN() was no longer correct since the 'buffer_start' variable was already pointing to the start of the skb's memory. The behavior of the dpaa2-eth driver without this patch was to drop frames on Tx even when the headroom was matching the 128 bytes necessary. Fix this by removing the manual adjust of 'buffer_start' from the PTR_MODE call. Closes: https://lore.kernel.org/netdev/70f0dcd9-1906-4d13-82df-7bbbbe7194c6@app.fastmail.com/T/#u Fixes: f422abe3f23d ("dpaa2-eth: increase the needed headroom to account for alignment") Signed-off-by: Ioana Ciornei Tested-by: Mathew McBride Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251016135807.360978-1-ioana.ciornei@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c index c96d1d6ba8fe..18d86badd6ea 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c @@ -1077,8 +1077,7 @@ static int dpaa2_eth_build_single_fd(struct dpaa2_eth_priv *priv, dma_addr_t addr; buffer_start = skb->data - dpaa2_eth_needed_headroom(skb); - aligned_start = PTR_ALIGN(buffer_start - DPAA2_ETH_TX_BUF_ALIGN, - DPAA2_ETH_TX_BUF_ALIGN); + aligned_start = PTR_ALIGN(buffer_start, DPAA2_ETH_TX_BUF_ALIGN); if (aligned_start >= skb->head) buffer_start = aligned_start; else -- cgit v1.2.3 From ffff5c8fc2af2218a3332b3d5b97654599d50cde Mon Sep 17 00:00:00 2001 From: Aleksander Jan Bajkowski Date: Thu, 16 Oct 2025 21:22:52 +0200 Subject: net: phy: realtek: fix rtl8221b-vm-cg name When splitting the RTL8221B-VM-CG into C22 and C45 variants, the name was accidentally changed to RTL8221B-VN-CG. This patch brings back the previous part number. Fixes: ad5ce743a6b0 ("net: phy: realtek: Add driver instances for rtl8221b via Clause 45") Signed-off-by: Aleksander Jan Bajkowski Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251016192325.2306757-1-olek2@wp.pl Signed-off-by: Jakub Kicinski --- drivers/net/phy/realtek/realtek_main.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/phy/realtek/realtek_main.c b/drivers/net/phy/realtek/realtek_main.c index a724b21b4fe7..16a347084293 100644 --- a/drivers/net/phy/realtek/realtek_main.c +++ b/drivers/net/phy/realtek/realtek_main.c @@ -154,7 +154,7 @@ #define RTL_8211FVD_PHYID 0x001cc878 #define RTL_8221B 0x001cc840 #define RTL_8221B_VB_CG 0x001cc849 -#define RTL_8221B_VN_CG 0x001cc84a +#define RTL_8221B_VM_CG 0x001cc84a #define RTL_8251B 0x001cc862 #define RTL_8261C 0x001cc890 @@ -1523,16 +1523,16 @@ static int rtl8221b_vb_cg_c45_match_phy_device(struct phy_device *phydev, return rtlgen_is_c45_match(phydev, RTL_8221B_VB_CG, true); } -static int rtl8221b_vn_cg_c22_match_phy_device(struct phy_device *phydev, +static int rtl8221b_vm_cg_c22_match_phy_device(struct phy_device *phydev, const struct phy_driver *phydrv) { - return rtlgen_is_c45_match(phydev, RTL_8221B_VN_CG, false); + return rtlgen_is_c45_match(phydev, RTL_8221B_VM_CG, false); } -static int rtl8221b_vn_cg_c45_match_phy_device(struct phy_device *phydev, +static int rtl8221b_vm_cg_c45_match_phy_device(struct phy_device *phydev, const struct phy_driver *phydrv) { - return rtlgen_is_c45_match(phydev, RTL_8221B_VN_CG, true); + return rtlgen_is_c45_match(phydev, RTL_8221B_VM_CG, true); } static int rtl_internal_nbaset_match_phy_device(struct phy_device *phydev, @@ -1879,7 +1879,7 @@ static struct phy_driver realtek_drvs[] = { .suspend = genphy_c45_pma_suspend, .resume = rtlgen_c45_resume, }, { - .match_phy_device = rtl8221b_vn_cg_c22_match_phy_device, + .match_phy_device = rtl8221b_vm_cg_c22_match_phy_device, .name = "RTL8221B-VM-CG 2.5Gbps PHY (C22)", .probe = rtl822x_probe, .get_features = rtl822x_get_features, @@ -1892,8 +1892,8 @@ static struct phy_driver realtek_drvs[] = { .read_page = rtl821x_read_page, .write_page = rtl821x_write_page, }, { - .match_phy_device = rtl8221b_vn_cg_c45_match_phy_device, - .name = "RTL8221B-VN-CG 2.5Gbps PHY (C45)", + .match_phy_device = rtl8221b_vm_cg_c45_match_phy_device, + .name = "RTL8221B-VM-CG 2.5Gbps PHY (C45)", .probe = rtl822x_probe, .config_init = rtl822xb_config_init, .get_rate_matching = rtl822xb_get_rate_matching, -- cgit v1.2.3 From 3a20c444cd123e820e10ae22eeaf00e189315aa1 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Fri, 17 Oct 2025 16:28:49 +0200 Subject: can: bxcan: bxcan_start_xmit(): use can_dev_dropped_skb() instead of can_dropped_invalid_skb() In addition to can_dropped_invalid_skb(), the helper function can_dev_dropped_skb() checks whether the device is in listen-only mode and discards the skb accordingly. Replace can_dropped_invalid_skb() by can_dev_dropped_skb() to also drop skbs in for listen-only mode. Reported-by: Marc Kleine-Budde Closes: https://lore.kernel.org/all/20251017-bizarre-enchanted-quokka-f3c704-mkl@pengutronix.de/ Fixes: f00647d8127b ("can: bxcan: add support for ST bxCAN controller") Link: https://patch.msgid.link/20251017-fix-skb-drop-check-v1-1-556665793fa4@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/bxcan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net') diff --git a/drivers/net/can/bxcan.c b/drivers/net/can/bxcan.c index bfc60eb33dc3..333ad42ea73b 100644 --- a/drivers/net/can/bxcan.c +++ b/drivers/net/can/bxcan.c @@ -842,7 +842,7 @@ static netdev_tx_t bxcan_start_xmit(struct sk_buff *skb, u32 id; int i, j; - if (can_dropped_invalid_skb(ndev, skb)) + if (can_dev_dropped_skb(ndev, skb)) return NETDEV_TX_OK; if (bxcan_tx_busy(priv)) -- cgit v1.2.3 From 0bee15a5caf36fe513fdeee07fd4f0331e61c064 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Fri, 17 Oct 2025 16:28:49 +0200 Subject: can: esd: acc_start_xmit(): use can_dev_dropped_skb() instead of can_dropped_invalid_skb() In addition to can_dropped_invalid_skb(), the helper function can_dev_dropped_skb() checks whether the device is in listen-only mode and discards the skb accordingly. Replace can_dropped_invalid_skb() by can_dev_dropped_skb() to also drop skbs in for listen-only mode. Reported-by: Marc Kleine-Budde Closes: https://lore.kernel.org/all/20251017-bizarre-enchanted-quokka-f3c704-mkl@pengutronix.de/ Fixes: 9721866f07e1 ("can: esd: add support for esd GmbH PCIe/402 CAN interface family") Link: https://patch.msgid.link/20251017-fix-skb-drop-check-v1-2-556665793fa4@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/esd/esdacc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net') diff --git a/drivers/net/can/esd/esdacc.c b/drivers/net/can/esd/esdacc.c index c80032bc1a52..73e66f9a3781 100644 --- a/drivers/net/can/esd/esdacc.c +++ b/drivers/net/can/esd/esdacc.c @@ -254,7 +254,7 @@ netdev_tx_t acc_start_xmit(struct sk_buff *skb, struct net_device *netdev) u32 acc_id; u32 acc_dlc; - if (can_dropped_invalid_skb(netdev, skb)) + if (can_dev_dropped_skb(netdev, skb)) return NETDEV_TX_OK; /* Access core->tx_fifo_tail only once because it may be changed -- cgit v1.2.3 From 3a3bc9bbb3a0287164a595787df0c70d91e77cfd Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Fri, 17 Oct 2025 16:28:49 +0200 Subject: can: rockchip-canfd: rkcanfd_start_xmit(): use can_dev_dropped_skb() instead of can_dropped_invalid_skb() In addition to can_dropped_invalid_skb(), the helper function can_dev_dropped_skb() checks whether the device is in listen-only mode and discards the skb accordingly. Replace can_dropped_invalid_skb() by can_dev_dropped_skb() to also drop skbs in for listen-only mode. Reported-by: Marc Kleine-Budde Closes: https://lore.kernel.org/all/20251017-bizarre-enchanted-quokka-f3c704-mkl@pengutronix.de/ Fixes: ff60bfbaf67f ("can: rockchip_canfd: add driver for Rockchip CAN-FD controller") Link: https://patch.msgid.link/20251017-fix-skb-drop-check-v1-3-556665793fa4@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/rockchip/rockchip_canfd-tx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net') diff --git a/drivers/net/can/rockchip/rockchip_canfd-tx.c b/drivers/net/can/rockchip/rockchip_canfd-tx.c index 865a15e033a9..12200dcfd338 100644 --- a/drivers/net/can/rockchip/rockchip_canfd-tx.c +++ b/drivers/net/can/rockchip/rockchip_canfd-tx.c @@ -72,7 +72,7 @@ netdev_tx_t rkcanfd_start_xmit(struct sk_buff *skb, struct net_device *ndev) int err; u8 i; - if (can_dropped_invalid_skb(ndev, skb)) + if (can_dev_dropped_skb(ndev, skb)) return NETDEV_TX_OK; if (!netif_subqueue_maybe_stop(priv->ndev, 0, -- cgit v1.2.3 From 8e93ac51e4c6dc399fad59ec21f55f2cfb46d27c Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Mon, 20 Oct 2025 11:51:03 +0200 Subject: can: netlink: can_changelink(): allow disabling of automatic restart Since the commit c1f3f9797c1f ("can: netlink: can_changelink(): fix NULL pointer deref of struct can_priv::do_set_mode"), the automatic restart delay can only be set for devices that implement the restart handler struct can_priv::do_set_mode. As it makes no sense to configure a automatic restart for devices that doesn't support it. However, since systemd commit 13ce5d4632e3 ("network/can: properly handle CAN.RestartSec=0") [1], systemd-networkd correctly handles a restart delay of "0" (i.e. the restart is disabled). Which means that a disabled restart is always configured in the kernel. On systems with both changes active this causes that CAN interfaces that don't implement a restart handler cannot be brought up by systemd-networkd. Solve this problem by allowing a delay of "0" to be configured, even if the device does not implement a restart handler. [1] https://github.com/systemd/systemd/commit/13ce5d4632e395521e6205c954493c7fc1c4c6e0 Cc: stable@vger.kernel.org Cc: Andrei Lalaev Reported-by: Marc Kleine-Budde Closes: https://lore.kernel.org/all/20251020-certain-arrogant-vole-of-sunshine-141841-mkl@pengutronix.de Fixes: c1f3f9797c1f ("can: netlink: can_changelink(): fix NULL pointer deref of struct can_priv::do_set_mode") Link: https://patch.msgid.link/20251020-netlink-fix-restart-v1-1-3f53c7f8520b@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/dev/netlink.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/can/dev/netlink.c b/drivers/net/can/dev/netlink.c index 0591406b6f32..6f83b87d54fc 100644 --- a/drivers/net/can/dev/netlink.c +++ b/drivers/net/can/dev/netlink.c @@ -452,7 +452,9 @@ static int can_changelink(struct net_device *dev, struct nlattr *tb[], } if (data[IFLA_CAN_RESTART_MS]) { - if (!priv->do_set_mode) { + unsigned int restart_ms = nla_get_u32(data[IFLA_CAN_RESTART_MS]); + + if (restart_ms != 0 && !priv->do_set_mode) { NL_SET_ERR_MSG(extack, "Device doesn't support restart from Bus Off"); return -EOPNOTSUPP; @@ -461,7 +463,7 @@ static int can_changelink(struct net_device *dev, struct nlattr *tb[], /* Do not allow changing restart delay while running */ if (dev->flags & IFF_UP) return -EBUSY; - priv->restart_ms = nla_get_u32(data[IFLA_CAN_RESTART_MS]); + priv->restart_ms = restart_ms; } if (data[IFLA_CAN_RESTART]) { -- cgit v1.2.3 From afd5ba577c10639f62e8120df67dc70ea4b61176 Mon Sep 17 00:00:00 2001 From: Amery Hung Date: Thu, 16 Oct 2025 22:55:39 +0300 Subject: net/mlx5e: RX, Fix generating skb from non-linear xdp_buff for legacy RQ XDP programs can release xdp_buff fragments when calling bpf_xdp_adjust_tail(). The driver currently assumes the number of fragments to be unchanged and may generate skb with wrong truesize or containing invalid frags. Fix the bug by generating skb according to xdp_buff after the XDP program runs. Fixes: ea5d49bdae8b ("net/mlx5e: Add XDP multi buffer support to the non-linear legacy RQ") Reviewed-by: Dragos Tatulea Signed-off-by: Amery Hung Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1760644540-899148-2-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index 263d5628ee44..17cab14b328b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -1794,14 +1794,27 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi } prog = rcu_dereference(rq->xdp_prog); - if (prog && mlx5e_xdp_handle(rq, prog, mxbuf)) { - if (__test_and_clear_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags)) { - struct mlx5e_wqe_frag_info *pwi; + if (prog) { + u8 nr_frags_free, old_nr_frags = sinfo->nr_frags; + + if (mlx5e_xdp_handle(rq, prog, mxbuf)) { + if (__test_and_clear_bit(MLX5E_RQ_FLAG_XDP_XMIT, + rq->flags)) { + struct mlx5e_wqe_frag_info *pwi; + + wi -= old_nr_frags - sinfo->nr_frags; + + for (pwi = head_wi; pwi < wi; pwi++) + pwi->frag_page->frags++; + } + return NULL; /* page/packet was consumed by XDP */ + } - for (pwi = head_wi; pwi < wi; pwi++) - pwi->frag_page->frags++; + nr_frags_free = old_nr_frags - sinfo->nr_frags; + if (unlikely(nr_frags_free)) { + wi -= nr_frags_free; + truesize -= nr_frags_free * frag_info->frag_stride; } - return NULL; /* page/packet was consumed by XDP */ } skb = mlx5e_build_linear_skb( -- cgit v1.2.3 From 87bcef158ac1faca1bd7e0104588e8e2956d10be Mon Sep 17 00:00:00 2001 From: Amery Hung Date: Thu, 16 Oct 2025 22:55:40 +0300 Subject: net/mlx5e: RX, Fix generating skb from non-linear xdp_buff for striding RQ XDP programs can change the layout of an xdp_buff through bpf_xdp_adjust_tail() and bpf_xdp_adjust_head(). Therefore, the driver cannot assume the size of the linear data area nor fragments. Fix the bug in mlx5 by generating skb according to xdp_buff after XDP programs run. Currently, when handling multi-buf XDP, the mlx5 driver assumes the layout of an xdp_buff to be unchanged. That is, the linear data area continues to be empty and fragments remain the same. This may cause the driver to generate erroneous skb or triggering a kernel warning. When an XDP program added linear data through bpf_xdp_adjust_head(), the linear data will be ignored as mlx5e_build_linear_skb() builds an skb without linear data and then pull data from fragments to fill the linear data area. When an XDP program has shrunk the non-linear data through bpf_xdp_adjust_tail(), the delta passed to __pskb_pull_tail() may exceed the actual nonlinear data size and trigger the BUG_ON in it. To fix the issue, first record the original number of fragments. If the number of fragments changes after the XDP program runs, rewind the end fragment pointer by the difference and recalculate the truesize. Then, build the skb with the linear data area matching the xdp_buff. Finally, only pull data in if there is non-linear data and fill the linear part up to 256 bytes. Fixes: f52ac7028bec ("net/mlx5e: RX, Add XDP multi-buffer support in Striding RQ") Signed-off-by: Amery Hung Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1760644540-899148-3-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 26 ++++++++++++++++++++++--- 1 file changed, 23 insertions(+), 3 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index 17cab14b328b..1c79adc51a04 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -2040,6 +2040,7 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w u32 byte_cnt = cqe_bcnt; struct skb_shared_info *sinfo; unsigned int truesize = 0; + u32 pg_consumed_bytes; struct bpf_prog *prog; struct sk_buff *skb; u32 linear_frame_sz; @@ -2093,7 +2094,8 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w while (byte_cnt) { /* Non-linear mode, hence non-XSK, which always uses PAGE_SIZE. */ - u32 pg_consumed_bytes = min_t(u32, PAGE_SIZE - frag_offset, byte_cnt); + pg_consumed_bytes = + min_t(u32, PAGE_SIZE - frag_offset, byte_cnt); if (test_bit(MLX5E_RQ_STATE_SHAMPO, &rq->state)) truesize += pg_consumed_bytes; @@ -2109,10 +2111,15 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w } if (prog) { + u8 nr_frags_free, old_nr_frags = sinfo->nr_frags; + u32 len; + if (mlx5e_xdp_handle(rq, prog, mxbuf)) { if (__test_and_clear_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags)) { struct mlx5e_frag_page *pfp; + frag_page -= old_nr_frags - sinfo->nr_frags; + for (pfp = head_page; pfp < frag_page; pfp++) pfp->frags++; @@ -2123,9 +2130,19 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w return NULL; /* page/packet was consumed by XDP */ } + nr_frags_free = old_nr_frags - sinfo->nr_frags; + if (unlikely(nr_frags_free)) { + frag_page -= nr_frags_free; + truesize -= (nr_frags_free - 1) * PAGE_SIZE + + ALIGN(pg_consumed_bytes, + BIT(rq->mpwqe.log_stride_sz)); + } + + len = mxbuf->xdp.data_end - mxbuf->xdp.data; + skb = mlx5e_build_linear_skb( rq, mxbuf->xdp.data_hard_start, linear_frame_sz, - mxbuf->xdp.data - mxbuf->xdp.data_hard_start, 0, + mxbuf->xdp.data - mxbuf->xdp.data_hard_start, len, mxbuf->xdp.data - mxbuf->xdp.data_meta); if (unlikely(!skb)) { mlx5e_page_release_fragmented(rq->page_pool, @@ -2150,8 +2167,11 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w do pagep->frags++; while (++pagep < frag_page); + + headlen = min_t(u16, MLX5E_RX_MAX_HEAD - len, + skb->data_len); + __pskb_pull_tail(skb, headlen); } - __pskb_pull_tail(skb, headlen); } else { dma_addr_t addr; -- cgit v1.2.3 From 49d34f3dd8519581030547eb7543a62f9ab5fa08 Mon Sep 17 00:00:00 2001 From: Aksh Garg Date: Thu, 16 Oct 2025 17:27:55 +0530 Subject: net: ethernet: ti: am65-cpts: fix timestamp loss due to race conditions Resolve race conditions in timestamp events list handling between TX and RX paths causing missed timestamps. The current implementation uses a single events list for both TX and RX timestamps. The am65_cpts_find_ts() function acquires the lock, splices all events (TX as well as RX events) to a temporary list, and releases the lock. This function performs matching of timestamps for TX packets only. Before it acquires the lock again to put the non-TX events back to the main events list, a concurrent RX processing thread could acquire the lock (as observed in practice), find an empty events list, and fail to attach timestamp to it, even though a relevant event exists in the spliced list which is yet to be restored to the main list. Fix this by creating separate events lists to handle TX and RX timestamps independently. Fixes: c459f606f66df ("net: ethernet: ti: am65-cpts: Enable RX HW timestamp for PTP packets using CPTS FIFO") Signed-off-by: Aksh Garg Reviewed-by: Siddharth Vadapalli Link: https://patch.msgid.link/20251016115755.1123646-1-a-garg7@ti.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/ti/am65-cpts.c | 63 +++++++++++++++++++++++++------------ 1 file changed, 43 insertions(+), 20 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/ti/am65-cpts.c b/drivers/net/ethernet/ti/am65-cpts.c index 59d6ab989c55..8ffbfaa3ab18 100644 --- a/drivers/net/ethernet/ti/am65-cpts.c +++ b/drivers/net/ethernet/ti/am65-cpts.c @@ -163,7 +163,9 @@ struct am65_cpts { struct device_node *clk_mux_np; struct clk *refclk; u32 refclk_freq; - struct list_head events; + /* separate lists to handle TX and RX timestamp independently */ + struct list_head events_tx; + struct list_head events_rx; struct list_head pool; struct am65_cpts_event pool_data[AM65_CPTS_MAX_EVENTS]; spinlock_t lock; /* protects events lists*/ @@ -227,6 +229,24 @@ static void am65_cpts_disable(struct am65_cpts *cpts) am65_cpts_write32(cpts, 0, int_enable); } +static int am65_cpts_purge_event_list(struct am65_cpts *cpts, + struct list_head *events) +{ + struct list_head *this, *next; + struct am65_cpts_event *event; + int removed = 0; + + list_for_each_safe(this, next, events) { + event = list_entry(this, struct am65_cpts_event, list); + if (time_after(jiffies, event->tmo)) { + list_del_init(&event->list); + list_add(&event->list, &cpts->pool); + ++removed; + } + } + return removed; +} + static int am65_cpts_event_get_port(struct am65_cpts_event *event) { return (event->event1 & AM65_CPTS_EVENT_1_PORT_NUMBER_MASK) >> @@ -239,20 +259,12 @@ static int am65_cpts_event_get_type(struct am65_cpts_event *event) AM65_CPTS_EVENT_1_EVENT_TYPE_SHIFT; } -static int am65_cpts_cpts_purge_events(struct am65_cpts *cpts) +static int am65_cpts_purge_events(struct am65_cpts *cpts) { - struct list_head *this, *next; - struct am65_cpts_event *event; int removed = 0; - list_for_each_safe(this, next, &cpts->events) { - event = list_entry(this, struct am65_cpts_event, list); - if (time_after(jiffies, event->tmo)) { - list_del_init(&event->list); - list_add(&event->list, &cpts->pool); - ++removed; - } - } + removed += am65_cpts_purge_event_list(cpts, &cpts->events_tx); + removed += am65_cpts_purge_event_list(cpts, &cpts->events_rx); if (removed) dev_dbg(cpts->dev, "event pool cleaned up %d\n", removed); @@ -287,7 +299,7 @@ static int __am65_cpts_fifo_read(struct am65_cpts *cpts) struct am65_cpts_event, list); if (!event) { - if (am65_cpts_cpts_purge_events(cpts)) { + if (am65_cpts_purge_events(cpts)) { dev_err(cpts->dev, "cpts: event pool empty\n"); ret = -1; goto out; @@ -306,11 +318,21 @@ static int __am65_cpts_fifo_read(struct am65_cpts *cpts) cpts->timestamp); break; case AM65_CPTS_EV_RX: + event->tmo = jiffies + + msecs_to_jiffies(AM65_CPTS_EVENT_RX_TX_TIMEOUT); + + list_move_tail(&event->list, &cpts->events_rx); + + dev_dbg(cpts->dev, + "AM65_CPTS_EV_RX e1:%08x e2:%08x t:%lld\n", + event->event1, event->event2, + event->timestamp); + break; case AM65_CPTS_EV_TX: event->tmo = jiffies + msecs_to_jiffies(AM65_CPTS_EVENT_RX_TX_TIMEOUT); - list_move_tail(&event->list, &cpts->events); + list_move_tail(&event->list, &cpts->events_tx); dev_dbg(cpts->dev, "AM65_CPTS_EV_TX e1:%08x e2:%08x t:%lld\n", @@ -828,7 +850,7 @@ static bool am65_cpts_match_tx_ts(struct am65_cpts *cpts, return found; } -static void am65_cpts_find_ts(struct am65_cpts *cpts) +static void am65_cpts_find_tx_ts(struct am65_cpts *cpts) { struct am65_cpts_event *event; struct list_head *this, *next; @@ -837,7 +859,7 @@ static void am65_cpts_find_ts(struct am65_cpts *cpts) LIST_HEAD(events); spin_lock_irqsave(&cpts->lock, flags); - list_splice_init(&cpts->events, &events); + list_splice_init(&cpts->events_tx, &events); spin_unlock_irqrestore(&cpts->lock, flags); list_for_each_safe(this, next, &events) { @@ -850,7 +872,7 @@ static void am65_cpts_find_ts(struct am65_cpts *cpts) } spin_lock_irqsave(&cpts->lock, flags); - list_splice_tail(&events, &cpts->events); + list_splice_tail(&events, &cpts->events_tx); list_splice_tail(&events_free, &cpts->pool); spin_unlock_irqrestore(&cpts->lock, flags); } @@ -861,7 +883,7 @@ static long am65_cpts_ts_work(struct ptp_clock_info *ptp) unsigned long flags; long delay = -1; - am65_cpts_find_ts(cpts); + am65_cpts_find_tx_ts(cpts); spin_lock_irqsave(&cpts->txq.lock, flags); if (!skb_queue_empty(&cpts->txq)) @@ -905,7 +927,7 @@ static u64 am65_cpts_find_rx_ts(struct am65_cpts *cpts, u32 skb_mtype_seqid) spin_lock_irqsave(&cpts->lock, flags); __am65_cpts_fifo_read(cpts); - list_for_each_safe(this, next, &cpts->events) { + list_for_each_safe(this, next, &cpts->events_rx) { event = list_entry(this, struct am65_cpts_event, list); if (time_after(jiffies, event->tmo)) { list_move(&event->list, &cpts->pool); @@ -1155,7 +1177,8 @@ struct am65_cpts *am65_cpts_create(struct device *dev, void __iomem *regs, return ERR_PTR(ret); mutex_init(&cpts->ptp_clk_lock); - INIT_LIST_HEAD(&cpts->events); + INIT_LIST_HEAD(&cpts->events_tx); + INIT_LIST_HEAD(&cpts->events_rx); INIT_LIST_HEAD(&cpts->pool); spin_lock_init(&cpts->lock); skb_queue_head_init(&cpts->txq); -- cgit v1.2.3 From 5523508258d390fd55b25c1c0f99c092f23a39e9 Mon Sep 17 00:00:00 2001 From: Yeounsu Moon Date: Sun, 19 Oct 2025 16:55:40 +0900 Subject: net: dlink: use dev_kfree_skb_any instead of dev_kfree_skb Replace `dev_kfree_skb()` with `dev_kfree_skb_any()` in `start_xmit()` which can be called from netpoll (hard IRQ) and from other contexts. Also, `np->link_status` can be changed at any time by interrupt handler. -0 [011] dNh4. 4541.754603: start_xmit <-netpoll_start_xmit -0 [011] dNh4. 4541.754622: => [FTRACE TRAMPOLINE] => start_xmit => netpoll_start_xmit => netpoll_send_skb => write_msg => console_flush_all => console_unlock => vprintk_emit => _printk => rio_interrupt => __handle_irq_event_percpu => handle_irq_event => handle_fasteoi_irq => __common_interrupt => common_interrupt => asm_common_interrupt => mwait_idle => default_idle_call => do_idle => cpu_startup_entry => start_secondary => common_startup_64 This issue can occur when the link state changes from off to on (e.g., plugging or unplugging the LAN cable) while transmitting a packet. If the skb has a destructor, a warning message may be printed in this situation. -> consume_skb (dev_kfree_skb()) -> __kfree_skb() -> skb_release_all() -> skb_release_head_state(skb) if (skb->destructor) { DEBUG_NET_WARN_ON_ONCE(in_hardirq()); skb->destructor(skb); } Found by inspection. Signed-off-by: Yeounsu Moon Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Tested-on: D-Link DGE-550T Rev-A3 Reviewed-by: Simon Horman Link: https://patch.msgid.link/20251019075540.55697-1-yyyynoom@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/dlink/dl2k.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/dlink/dl2k.c b/drivers/net/ethernet/dlink/dl2k.c index 7077d705e471..6e4f17142519 100644 --- a/drivers/net/ethernet/dlink/dl2k.c +++ b/drivers/net/ethernet/dlink/dl2k.c @@ -733,7 +733,7 @@ start_xmit (struct sk_buff *skb, struct net_device *dev) u64 tfc_vlan_tag = 0; if (np->link_status == 0) { /* Link Down */ - dev_kfree_skb(skb); + dev_kfree_skb_any(skb); return NETDEV_TX_OK; } entry = np->cur_tx % TX_RING_SIZE; -- cgit v1.2.3 From d63f0391d6c7b75e1a847e1a26349fa8cad0004d Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Mon, 20 Oct 2025 08:54:54 +0200 Subject: net: hibmcge: select FIXED_PHY hibmcge uses fixed_phy_register() et al, but doesn't cater for the case that hibmcge is built-in and fixed_phy is a module. To solve this select FIXED_PHY. Fixes: 1d7cd7a9c69c ("net: hibmcge: support scenario without PHY") Signed-off-by: Heiner Kallweit Reviewed-by: Jijie Shao Link: https://patch.msgid.link/c4fc061f-b6d5-418b-a0dc-6b238cdbedce@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/hisilicon/Kconfig | 1 + 1 file changed, 1 insertion(+) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/hisilicon/Kconfig b/drivers/net/ethernet/hisilicon/Kconfig index 65302c41bfb1..38875c196cb6 100644 --- a/drivers/net/ethernet/hisilicon/Kconfig +++ b/drivers/net/ethernet/hisilicon/Kconfig @@ -148,6 +148,7 @@ config HIBMCGE tristate "Hisilicon BMC Gigabit Ethernet Device Support" depends on PCI && PCI_MSI select PHYLIB + select FIXED_PHY select MOTORCOMM_PHY select REALTEK_PHY help -- cgit v1.2.3 From 5370c31e84b0e0999c7b5ff949f4e104def35584 Mon Sep 17 00:00:00 2001 From: Lad Prabhakar Date: Fri, 17 Oct 2025 16:18:29 +0100 Subject: net: ravb: Enforce descriptor type ordering MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Ensure the TX descriptor type fields are published in a safe order so the DMA engine never begins processing a descriptor chain before all descriptor fields are fully initialised. For multi-descriptor transmits the driver writes DT_FEND into the last descriptor and DT_FSTART into the first. The DMA engine begins processing when it observes DT_FSTART. Move the dma_wmb() barrier so it executes immediately after DT_FEND and immediately before writing DT_FSTART (and before DT_FSINGLE in the single-descriptor case). This guarantees that all prior CPU writes to the descriptor memory are visible to the device before DT_FSTART is seen. This avoids a situation where compiler/CPU reordering could publish DT_FSTART ahead of DT_FEND or other descriptor fields, allowing the DMA to start on a partially initialised chain and causing corrupted transmissions or TX timeouts. Such a failure was observed on RZ/G2L with an RT kernel as transmit queue timeouts and device resets. Fixes: 2f45d1902acf ("ravb: minimize TX data copying") Cc: stable@vger.kernel.org Co-developed-by: Fabrizio Castro Signed-off-by: Fabrizio Castro Signed-off-by: Lad Prabhakar Reviewed-by: Niklas Söderlund Link: https://patch.msgid.link/20251017151830.171062-4-prabhakar.mahadev-lad.rj@bp.renesas.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/renesas/ravb_main.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 9d3bd65b85ff..044ee83c63bb 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -2211,13 +2211,25 @@ static netdev_tx_t ravb_start_xmit(struct sk_buff *skb, struct net_device *ndev) skb_tx_timestamp(skb); } - /* Descriptor type must be set after all the above writes */ - dma_wmb(); + if (num_tx_desc > 1) { desc->die_dt = DT_FEND; desc--; + /* When using multi-descriptors, DT_FEND needs to get written + * before DT_FSTART, but the compiler may reorder the memory + * writes in an attempt to optimize the code. + * Use a dma_wmb() barrier to make sure DT_FEND and DT_FSTART + * are written exactly in the order shown in the code. + * This is particularly important for cases where the DMA engine + * is already running when we are running this code. If the DMA + * sees DT_FSTART without the corresponding DT_FEND it will enter + * an error condition. + */ + dma_wmb(); desc->die_dt = DT_FSTART; } else { + /* Descriptor type must be set after all the above writes */ + dma_wmb(); desc->die_dt = DT_FSINGLE; } ravb_modify(ndev, TCCR, TCCR_TSRQ0 << q, TCCR_TSRQ0 << q); -- cgit v1.2.3 From 706136c5723626fcde8dd8f598a4dcd251e24927 Mon Sep 17 00:00:00 2001 From: Lad Prabhakar Date: Fri, 17 Oct 2025 16:18:30 +0100 Subject: net: ravb: Ensure memory write completes before ringing TX doorbell MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add a final dma_wmb() barrier before triggering the transmit request (TCCR_TSRQ) to ensure all descriptor and buffer writes are visible to the DMA engine. According to the hardware manual, a read-back operation is required before writing to the doorbell register to guarantee completion of previous writes. Instead of performing a dummy read, a dma_wmb() is used to both enforce the same ordering semantics on the CPU side and also to ensure completion of writes. Fixes: c156633f1353 ("Renesas Ethernet AVB driver proper") Cc: stable@vger.kernel.org Co-developed-by: Fabrizio Castro Signed-off-by: Fabrizio Castro Signed-off-by: Lad Prabhakar Reviewed-by: Niklas Söderlund Link: https://patch.msgid.link/20251017151830.171062-5-prabhakar.mahadev-lad.rj@bp.renesas.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/renesas/ravb_main.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 044ee83c63bb..e2d7ce1a85e8 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -2232,6 +2232,14 @@ static netdev_tx_t ravb_start_xmit(struct sk_buff *skb, struct net_device *ndev) dma_wmb(); desc->die_dt = DT_FSINGLE; } + + /* Before ringing the doorbell we need to make sure that the latest + * writes have been committed to memory, otherwise it could delay + * things until the doorbell is rang again. + * This is in replacement of the read operation mentioned in the HW + * manuals. + */ + dma_wmb(); ravb_modify(ndev, TCCR, TCCR_TSRQ0 << q, TCCR_TSRQ0 << q); priv->cur_tx[q] += num_tx_desc; -- cgit v1.2.3 From 10843e1492e474c02b91314963161731fa92af91 Mon Sep 17 00:00:00 2001 From: Tonghao Zhang Date: Tue, 21 Oct 2025 13:09:33 +0800 Subject: net: bonding: fix possible peer notify event loss or dup issue If the send_peer_notif counter and the peer event notify are not synchronized. It may cause problems such as the loss or dup of peer notify event. Before this patch: - If should_notify_peers is true and the lock for send_peer_notif-- fails, peer event may be sent again in next mii_monitor loop, because should_notify_peers is still true. - If should_notify_peers is true and the lock for send_peer_notif-- succeeded, but the lock for peer event fails, the peer event will be lost. This patch locks the RTNL for send_peer_notif, events, and commit simultaneously. Fixes: 07a4ddec3ce9 ("bonding: add an option to specify a delay between peer notifications") Cc: Jay Vosburgh Cc: Andrew Lunn Cc: Eric Dumazet Cc: Jakub Kicinski Cc: Paolo Abeni Cc: Hangbin Liu Cc: Nikolay Aleksandrov Cc: Vincent Bernat Cc: Signed-off-by: Tonghao Zhang Acked-by: Jay Vosburgh Link: https://patch.msgid.link/20251021050933.46412-1-tonghao@bamaicloud.com Signed-off-by: Paolo Abeni --- drivers/net/bonding/bond_main.c | 40 ++++++++++++++++++---------------------- 1 file changed, 18 insertions(+), 22 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 67fdcbdd2764..e95e593cd12d 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -2874,7 +2874,7 @@ static void bond_mii_monitor(struct work_struct *work) { struct bonding *bond = container_of(work, struct bonding, mii_work.work); - bool should_notify_peers = false; + bool should_notify_peers; bool commit; unsigned long delay; struct slave *slave; @@ -2886,30 +2886,33 @@ static void bond_mii_monitor(struct work_struct *work) goto re_arm; rcu_read_lock(); + should_notify_peers = bond_should_notify_peers(bond); commit = !!bond_miimon_inspect(bond); - if (bond->send_peer_notif) { - rcu_read_unlock(); - if (rtnl_trylock()) { - bond->send_peer_notif--; - rtnl_unlock(); - } - } else { - rcu_read_unlock(); - } - if (commit) { + rcu_read_unlock(); + + if (commit || bond->send_peer_notif) { /* Race avoidance with bond_close cancel of workqueue */ if (!rtnl_trylock()) { delay = 1; - should_notify_peers = false; goto re_arm; } - bond_for_each_slave(bond, slave, iter) { - bond_commit_link_state(slave, BOND_SLAVE_NOTIFY_LATER); + if (commit) { + bond_for_each_slave(bond, slave, iter) { + bond_commit_link_state(slave, + BOND_SLAVE_NOTIFY_LATER); + } + bond_miimon_commit(bond); + } + + if (bond->send_peer_notif) { + bond->send_peer_notif--; + if (should_notify_peers) + call_netdevice_notifiers(NETDEV_NOTIFY_PEERS, + bond->dev); } - bond_miimon_commit(bond); rtnl_unlock(); /* might sleep, hold no other locks */ } @@ -2917,13 +2920,6 @@ static void bond_mii_monitor(struct work_struct *work) re_arm: if (bond->params.miimon) queue_delayed_work(bond->wq, &bond->mii_work, delay); - - if (should_notify_peers) { - if (!rtnl_trylock()) - return; - call_netdevice_notifiers(NETDEV_NOTIFY_PEERS, bond->dev); - rtnl_unlock(); - } } static int bond_upper_dev_walk(struct net_device *upper, -- cgit v1.2.3 From efd729408bc7d57e0c8d027b9ff514187fc1a05b Mon Sep 17 00:00:00 2001 From: Ralf Lici Date: Tue, 21 Oct 2025 12:09:42 +0200 Subject: ovpn: use datagram_poll_queue for socket readiness in TCP openvpn TCP encapsulation uses a custom queue to deliver packets to userspace. Currently it relies on datagram_poll, which checks sk_receive_queue, leading to false readiness signals when that queue contains non-userspace packets. Switch ovpn_tcp_poll to use datagram_poll_queue with the peer's user_queue, ensuring poll only signals readiness when userspace data is actually available. Also refactor ovpn_tcp_poll in order to enforce the assumption we can make on the lifetime of ovpn_sock and peer. Fixes: 11851cbd60ea ("ovpn: implement TCP transport") Signed-off-by: Antonio Quartulli Signed-off-by: Ralf Lici Reviewed-by: Sabrina Dubroca Link: https://patch.msgid.link/20251021100942.195010-4-ralf@mandelbit.com Signed-off-by: Paolo Abeni --- drivers/net/ovpn/tcp.c | 26 ++++++++++++++++++++++---- 1 file changed, 22 insertions(+), 4 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/ovpn/tcp.c b/drivers/net/ovpn/tcp.c index 289f62c5d2c7..0d7f30360d87 100644 --- a/drivers/net/ovpn/tcp.c +++ b/drivers/net/ovpn/tcp.c @@ -560,16 +560,34 @@ static void ovpn_tcp_close(struct sock *sk, long timeout) static __poll_t ovpn_tcp_poll(struct file *file, struct socket *sock, poll_table *wait) { - __poll_t mask = datagram_poll(file, sock, wait); + struct sk_buff_head *queue = &sock->sk->sk_receive_queue; struct ovpn_socket *ovpn_sock; + struct ovpn_peer *peer = NULL; + __poll_t mask; rcu_read_lock(); ovpn_sock = rcu_dereference_sk_user_data(sock->sk); - if (ovpn_sock && ovpn_sock->peer && - !skb_queue_empty(&ovpn_sock->peer->tcp.user_queue)) - mask |= EPOLLIN | EPOLLRDNORM; + /* if we landed in this callback, we expect to have a + * meaningful state. The ovpn_socket lifecycle would + * prevent it otherwise. + */ + if (WARN(!ovpn_sock || !ovpn_sock->peer, + "ovpn: null state in ovpn_tcp_poll!")) { + rcu_read_unlock(); + return 0; + } + + if (ovpn_peer_hold(ovpn_sock->peer)) { + peer = ovpn_sock->peer; + queue = &peer->tcp.user_queue; + } rcu_read_unlock(); + mask = datagram_poll_queue(file, sock, wait, queue); + + if (peer) + ovpn_peer_put(peer); + return mask; } -- cgit v1.2.3 From 399d10934740ae8cdaa4e3245f7c5f6c332da844 Mon Sep 17 00:00:00 2001 From: Robert Marko Date: Tue, 21 Oct 2025 15:20:26 +0200 Subject: net: phy: micrel: always set shared->phydev for LAN8814 Currently, during the LAN8814 PTP probe shared->phydev is only set if PTP clock gets actually set, otherwise the function will return before setting it. This is an issue as shared->phydev is unconditionally being used when IRQ is being handled, especially in lan8814_gpio_process_cap and since it was not set it will cause a NULL pointer exception and crash the kernel. So, simply always set shared->phydev to avoid the NULL pointer exception. Fixes: b3f1a08fcf0d ("net: phy: micrel: Add support for PTP_PF_EXTTS for lan8814") Signed-off-by: Robert Marko Tested-by: Horatiu Vultur Link: https://patch.msgid.link/20251021132034.983936-1-robert.marko@sartura.hr Signed-off-by: Jakub Kicinski --- drivers/net/phy/micrel.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 79ce3eb6752b..604b5de0c158 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -4262,6 +4262,8 @@ static int __lan8814_ptp_probe_once(struct phy_device *phydev, char *pin_name, { struct lan8814_shared_priv *shared = phy_package_get_priv(phydev); + shared->phydev = phydev; + /* Initialise shared lock for clock*/ mutex_init(&shared->shared_lock); @@ -4317,8 +4319,6 @@ static int __lan8814_ptp_probe_once(struct phy_device *phydev, char *pin_name, phydev_dbg(phydev, "successfully registered ptp clock\n"); - shared->phydev = phydev; - /* The EP.4 is shared between all the PHYs in the package and also it * can be accessed by any of the PHYs */ -- cgit v1.2.3 From d58a9a917aa39b10250ec16cb9c24e942cbc77d3 Mon Sep 17 00:00:00 2001 From: Alexei Lazar Date: Wed, 22 Oct 2025 15:29:40 +0300 Subject: net/mlx5e: Skip PPHCR register query if not supported by the device Check the PCAM supported registers mask before querying the PPHCR register, as it is not supported in older devices. Fixes: 44907e7c8fd0 ("net/mlx5e: Add logic to read RS-FEC histogram bin ranges from PPHCR") Signed-off-by: Alexei Lazar Reviewed-by: Yael Chemla Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761136182-918470-3-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c index 7c029a7d0fd7..a2802cfc9b98 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c @@ -1614,7 +1614,9 @@ void mlx5e_stats_fec_get(struct mlx5e_priv *priv, fec_set_corrected_bits_total(priv, fec_stats); fec_set_block_stats(priv, mode, fec_stats); - fec_set_histograms_stats(priv, mode, hist); + + if (MLX5_CAP_PCAM_REG(priv->mdev, pphcr)) + fec_set_histograms_stats(priv, mode, hist); } #define PPORT_ETH_EXT_OFF(c) \ -- cgit v1.2.3 From 8f82f89550daafc8ca3ba74c389ae1b4afdd75c8 Mon Sep 17 00:00:00 2001 From: Patrisious Haddad Date: Wed, 22 Oct 2025 15:29:41 +0300 Subject: net/mlx5: Refactor devcom to return NULL on failure Devcom device and component registration isn't always critical to the functionality of the caller, hence the registration can fail and we can continue working with an ERR_PTR value saved inside a variable. In order to avoid that make sure all devcom failures return NULL. Signed-off-by: Patrisious Haddad Reviewed-by: Leon Romanovsky Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761136182-918470-4-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 6 +-- .../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 4 +- drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c | 7 ++- .../net/ethernet/mellanox/mlx5/core/lib/clock.c | 2 +- .../net/ethernet/mellanox/mlx5/core/lib/devcom.c | 53 +++++++++++----------- drivers/net/ethernet/mellanox/mlx5/core/lib/sd.c | 4 +- drivers/net/ethernet/mellanox/mlx5/core/main.c | 5 +- 7 files changed, 39 insertions(+), 42 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index a56825921c23..41fd5eee6306 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -242,8 +242,8 @@ static int mlx5e_devcom_init_mpv(struct mlx5e_priv *priv, u64 *data) &attr, mlx5e_devcom_event_mpv, priv); - if (IS_ERR(priv->devcom)) - return PTR_ERR(priv->devcom); + if (!priv->devcom) + return -EINVAL; if (mlx5_core_is_mp_master(priv->mdev)) { mlx5_devcom_send_event(priv->devcom, MPV_DEVCOM_MASTER_UP, @@ -256,7 +256,7 @@ static int mlx5e_devcom_init_mpv(struct mlx5e_priv *priv, u64 *data) static void mlx5e_devcom_cleanup_mpv(struct mlx5e_priv *priv) { - if (IS_ERR_OR_NULL(priv->devcom)) + if (!priv->devcom) return; if (mlx5_core_is_mp_master(priv->mdev)) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 4cf995be127d..34749814f19b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -3129,7 +3129,7 @@ void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw, attr, mlx5_esw_offloads_devcom_event, esw); - if (IS_ERR(esw->devcom)) + if (!esw->devcom) return; mlx5_devcom_send_event(esw->devcom, @@ -3140,7 +3140,7 @@ void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw, void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw) { - if (IS_ERR_OR_NULL(esw->devcom)) + if (!esw->devcom) return; mlx5_devcom_send_event(esw->devcom, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c index 59c00c911275..3db0387bf6dc 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c @@ -1430,11 +1430,10 @@ static int mlx5_lag_register_hca_devcom_comp(struct mlx5_core_dev *dev) mlx5_devcom_register_component(dev->priv.devc, MLX5_DEVCOM_HCA_PORTS, &attr, NULL, dev); - if (IS_ERR(dev->priv.hca_devcom_comp)) { + if (!dev->priv.hca_devcom_comp) { mlx5_core_err(dev, - "Failed to register devcom HCA component, err: %ld\n", - PTR_ERR(dev->priv.hca_devcom_comp)); - return PTR_ERR(dev->priv.hca_devcom_comp); + "Failed to register devcom HCA component."); + return -EINVAL; } return 0; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c index d0ba83d77cd1..29e7fa09c32c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c @@ -1444,7 +1444,7 @@ static void mlx5_shared_clock_register(struct mlx5_core_dev *mdev, u64 key) compd = mlx5_devcom_register_component(mdev->priv.devc, MLX5_DEVCOM_SHARED_CLOCK, &attr, NULL, mdev); - if (IS_ERR(compd)) + if (!compd) return; mdev->clock_state->compdev = compd; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c index faa2833602c8..e749618229bc 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c @@ -76,20 +76,18 @@ mlx5_devcom_dev_alloc(struct mlx5_core_dev *dev) struct mlx5_devcom_dev * mlx5_devcom_register_device(struct mlx5_core_dev *dev) { - struct mlx5_devcom_dev *devc; + struct mlx5_devcom_dev *devc = NULL; mutex_lock(&dev_list_lock); if (devcom_dev_exists(dev)) { - devc = ERR_PTR(-EEXIST); + mlx5_core_err(dev, "devcom device already exists"); goto out; } devc = mlx5_devcom_dev_alloc(dev); - if (!devc) { - devc = ERR_PTR(-ENOMEM); + if (!devc) goto out; - } list_add_tail(&devc->list, &devcom_dev_list); out: @@ -110,8 +108,10 @@ mlx5_devcom_dev_release(struct kref *ref) void mlx5_devcom_unregister_device(struct mlx5_devcom_dev *devc) { - if (!IS_ERR_OR_NULL(devc)) - kref_put(&devc->ref, mlx5_devcom_dev_release); + if (!devc) + return; + + kref_put(&devc->ref, mlx5_devcom_dev_release); } static struct mlx5_devcom_comp * @@ -122,7 +122,7 @@ mlx5_devcom_comp_alloc(u64 id, const struct mlx5_devcom_match_attr *attr, comp = kzalloc(sizeof(*comp), GFP_KERNEL); if (!comp) - return ERR_PTR(-ENOMEM); + return NULL; comp->id = id; comp->key.key = attr->key; @@ -160,7 +160,7 @@ devcom_alloc_comp_dev(struct mlx5_devcom_dev *devc, devcom = kzalloc(sizeof(*devcom), GFP_KERNEL); if (!devcom) - return ERR_PTR(-ENOMEM); + return NULL; kref_get(&devc->ref); devcom->devc = devc; @@ -240,31 +240,28 @@ mlx5_devcom_register_component(struct mlx5_devcom_dev *devc, mlx5_devcom_event_handler_t handler, void *data) { - struct mlx5_devcom_comp_dev *devcom; + struct mlx5_devcom_comp_dev *devcom = NULL; struct mlx5_devcom_comp *comp; - if (IS_ERR_OR_NULL(devc)) - return ERR_PTR(-EINVAL); + if (!devc) + return NULL; mutex_lock(&comp_list_lock); comp = devcom_component_get(devc, id, attr, handler); - if (IS_ERR(comp)) { - devcom = ERR_PTR(-EINVAL); + if (IS_ERR(comp)) goto out_unlock; - } if (!comp) { comp = mlx5_devcom_comp_alloc(id, attr, handler); - if (IS_ERR(comp)) { - devcom = ERR_CAST(comp); + if (!comp) goto out_unlock; - } + list_add_tail(&comp->comp_list, &devcom_comp_list); } mutex_unlock(&comp_list_lock); devcom = devcom_alloc_comp_dev(devc, comp, data); - if (IS_ERR(devcom)) + if (!devcom) kref_put(&comp->ref, mlx5_devcom_comp_release); return devcom; @@ -276,8 +273,10 @@ out_unlock: void mlx5_devcom_unregister_component(struct mlx5_devcom_comp_dev *devcom) { - if (!IS_ERR_OR_NULL(devcom)) - devcom_free_comp_dev(devcom); + if (!devcom) + return; + + devcom_free_comp_dev(devcom); } int mlx5_devcom_comp_get_size(struct mlx5_devcom_comp_dev *devcom) @@ -296,7 +295,7 @@ int mlx5_devcom_send_event(struct mlx5_devcom_comp_dev *devcom, int err = 0; void *data; - if (IS_ERR_OR_NULL(devcom)) + if (!devcom) return -ENODEV; comp = devcom->comp; @@ -338,7 +337,7 @@ void mlx5_devcom_comp_set_ready(struct mlx5_devcom_comp_dev *devcom, bool ready) bool mlx5_devcom_comp_is_ready(struct mlx5_devcom_comp_dev *devcom) { - if (IS_ERR_OR_NULL(devcom)) + if (!devcom) return false; return READ_ONCE(devcom->comp->ready); @@ -348,7 +347,7 @@ bool mlx5_devcom_for_each_peer_begin(struct mlx5_devcom_comp_dev *devcom) { struct mlx5_devcom_comp *comp; - if (IS_ERR_OR_NULL(devcom)) + if (!devcom) return false; comp = devcom->comp; @@ -421,21 +420,21 @@ void *mlx5_devcom_get_next_peer_data_rcu(struct mlx5_devcom_comp_dev *devcom, void mlx5_devcom_comp_lock(struct mlx5_devcom_comp_dev *devcom) { - if (IS_ERR_OR_NULL(devcom)) + if (!devcom) return; down_write(&devcom->comp->sem); } void mlx5_devcom_comp_unlock(struct mlx5_devcom_comp_dev *devcom) { - if (IS_ERR_OR_NULL(devcom)) + if (!devcom) return; up_write(&devcom->comp->sem); } int mlx5_devcom_comp_trylock(struct mlx5_devcom_comp_dev *devcom) { - if (IS_ERR_OR_NULL(devcom)) + if (!devcom) return 0; return down_write_trylock(&devcom->comp->sem); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/sd.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/sd.c index f5c2701f6e87..8e17daae48af 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/sd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/sd.c @@ -221,8 +221,8 @@ static int sd_register(struct mlx5_core_dev *dev) attr.net = mlx5_core_net(dev); devcom = mlx5_devcom_register_component(dev->priv.devc, MLX5_DEVCOM_SD_GROUP, &attr, NULL, dev); - if (IS_ERR(devcom)) - return PTR_ERR(devcom); + if (!devcom) + return -EINVAL; sd->devcom = devcom; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index df93625c9dfa..70c156591b0b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -978,9 +978,8 @@ static int mlx5_init_once(struct mlx5_core_dev *dev) int err; dev->priv.devc = mlx5_devcom_register_device(dev); - if (IS_ERR(dev->priv.devc)) - mlx5_core_warn(dev, "failed to register devcom device %pe\n", - dev->priv.devc); + if (!dev->priv.devc) + mlx5_core_warn(dev, "failed to register devcom device\n"); err = mlx5_query_board_id(dev); if (err) { -- cgit v1.2.3 From 664f76be38a18c61151d0ef248c7e2f3afb4f3c7 Mon Sep 17 00:00:00 2001 From: Patrisious Haddad Date: Wed, 22 Oct 2025 15:29:42 +0300 Subject: net/mlx5: Fix IPsec cleanup over MPV device When we do mlx5e_detach_netdev() we eventually disable blocking events notifier, among those events are IPsec MPV events from IB to core. So before disabling those blocking events, make sure to also unregister the devcom device and mark all this device operations as complete, in order to prevent the other device from using invalid netdev during future devcom events which could cause the trace below. BUG: kernel NULL pointer dereference, address: 0000000000000010 PGD 146427067 P4D 146427067 PUD 146488067 PMD 0 Oops: Oops: 0000 [#1] SMP CPU: 1 UID: 0 PID: 7735 Comm: devlink Tainted: GW 6.12.0-rc6_for_upstream_min_debug_2024_11_08_00_46 #1 Tainted: [W]=WARN Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 RIP: 0010:mlx5_devcom_comp_set_ready+0x5/0x40 [mlx5_core] Code: 00 01 48 83 05 23 32 1e 00 01 41 b8 ed ff ff ff e9 60 ff ff ff 48 83 05 00 32 1e 00 01 eb e3 66 0f 1f 44 00 00 0f 1f 44 00 00 <48> 8b 47 10 48 83 05 5f 32 1e 00 01 48 8b 50 40 48 85 d2 74 05 40 RSP: 0018:ffff88811a5c35f8 EFLAGS: 00010206 RAX: ffff888106e8ab80 RBX: ffff888107d7e200 RCX: ffff88810d6f0a00 RDX: ffff88810d6f0a00 RSI: 0000000000000001 RDI: 0000000000000000 RBP: ffff88811a17e620 R08: 0000000000000040 R09: 0000000000000000 R10: ffff88811a5c3618 R11: 0000000de85d51bd R12: ffff88811a17e600 R13: ffff88810d6f0a00 R14: 0000000000000000 R15: ffff8881034bda80 FS: 00007f27bdf89180(0000) GS:ffff88852c880000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000010 CR3: 000000010f159005 CR4: 0000000000372eb0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: ? __die+0x20/0x60 ? page_fault_oops+0x150/0x3e0 ? exc_page_fault+0x74/0x130 ? asm_exc_page_fault+0x22/0x30 ? mlx5_devcom_comp_set_ready+0x5/0x40 [mlx5_core] mlx5e_devcom_event_mpv+0x42/0x60 [mlx5_core] mlx5_devcom_send_event+0x8c/0x170 [mlx5_core] blocking_event+0x17b/0x230 [mlx5_core] notifier_call_chain+0x35/0xa0 blocking_notifier_call_chain+0x3d/0x60 mlx5_blocking_notifier_call_chain+0x22/0x30 [mlx5_core] mlx5_core_mp_event_replay+0x12/0x20 [mlx5_core] mlx5_ib_bind_slave_port+0x228/0x2c0 [mlx5_ib] mlx5_ib_stage_init_init+0x664/0x9d0 [mlx5_ib] ? idr_alloc_cyclic+0x50/0xb0 ? __kmalloc_cache_noprof+0x167/0x340 ? __kmalloc_noprof+0x1a7/0x430 __mlx5_ib_add+0x34/0xd0 [mlx5_ib] mlx5r_probe+0xe9/0x310 [mlx5_ib] ? kernfs_add_one+0x107/0x150 ? __mlx5_ib_add+0xd0/0xd0 [mlx5_ib] auxiliary_bus_probe+0x3e/0x90 really_probe+0xc5/0x3a0 ? driver_probe_device+0x90/0x90 __driver_probe_device+0x80/0x160 driver_probe_device+0x1e/0x90 __device_attach_driver+0x7d/0x100 bus_for_each_drv+0x80/0xd0 __device_attach+0xbc/0x1f0 bus_probe_device+0x86/0xa0 device_add+0x62d/0x830 __auxiliary_device_add+0x3b/0xa0 ? auxiliary_device_init+0x41/0x90 add_adev+0xd1/0x150 [mlx5_core] mlx5_rescan_drivers_locked+0x21c/0x300 [mlx5_core] esw_mode_change+0x6c/0xc0 [mlx5_core] mlx5_devlink_eswitch_mode_set+0x21e/0x640 [mlx5_core] devlink_nl_eswitch_set_doit+0x60/0xe0 genl_family_rcv_msg_doit+0xd0/0x120 genl_rcv_msg+0x180/0x2b0 ? devlink_get_from_attrs_lock+0x170/0x170 ? devlink_nl_eswitch_get_doit+0x290/0x290 ? devlink_nl_pre_doit_port_optional+0x50/0x50 ? genl_family_rcv_msg_dumpit+0xf0/0xf0 netlink_rcv_skb+0x54/0x100 genl_rcv+0x24/0x40 netlink_unicast+0x1fc/0x2d0 netlink_sendmsg+0x1e4/0x410 __sock_sendmsg+0x38/0x60 ? sockfd_lookup_light+0x12/0x60 __sys_sendto+0x105/0x160 ? __sys_recvmsg+0x4e/0x90 __x64_sys_sendto+0x20/0x30 do_syscall_64+0x4c/0x100 entry_SYSCALL_64_after_hwframe+0x4b/0x53 RIP: 0033:0x7f27bc91b13a Code: bb 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 8b 05 fa 96 2c 00 45 89 c9 4c 63 d1 48 63 ff 85 c0 75 15 b8 2c 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 76 f3 c3 0f 1f 40 00 41 55 41 54 4d 89 c5 55 RSP: 002b:00007fff369557e8 EFLAGS: 00000246 ORIG_RAX: 000000000000002c RAX: ffffffffffffffda RBX: 0000000009c54b10 RCX: 00007f27bc91b13a RDX: 0000000000000038 RSI: 0000000009c54b10 RDI: 0000000000000006 RBP: 0000000009c54920 R08: 00007f27bd0030e0 R09: 000000000000000c R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000001 Modules linked in: mlx5_vdpa vringh vhost_iotlb vdpa xt_MASQUERADE nf_conntrack_netlink nfnetlink iptable_nat xt_addrtype xt_conntrack nf_nat br_netfilter rpcsec_gss_krb5 auth_rpcgss oid_registry overlay rpcrdma rdma_ucm ib_iser libiscsi ib_umad scsi_transport_iscsi ib_ipoib rdma_cm iw_cm ib_cm mlx5_fwctl mlx5_ib ib_uverbs ib_core mlx5_core CR2: 0000000000000010 Fixes: 82f9378c443c ("net/mlx5: Handle IPsec steering upon master unbind/bind") Signed-off-by: Patrisious Haddad Reviewed-by: Leon Romanovsky Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761136182-918470-5-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- .../ethernet/mellanox/mlx5/core/en_accel/ipsec.h | 5 +++++ .../mellanox/mlx5/core/en_accel/ipsec_fs.c | 25 ++++++++++++++++++++-- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 2 ++ 3 files changed, 30 insertions(+), 2 deletions(-) (limited to 'drivers/net') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h index 5d7c15abfcaf..f8eaaf37963b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h @@ -342,6 +342,7 @@ void mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry, void mlx5e_ipsec_handle_mpv_event(int event, struct mlx5e_priv *slave_priv, struct mlx5e_priv *master_priv); void mlx5e_ipsec_send_event(struct mlx5e_priv *priv, int event); +void mlx5e_ipsec_disable_events(struct mlx5e_priv *priv); static inline struct mlx5_core_dev * mlx5e_ipsec_sa2dev(struct mlx5e_ipsec_sa_entry *sa_entry) @@ -387,6 +388,10 @@ static inline void mlx5e_ipsec_handle_mpv_event(int event, struct mlx5e_priv *sl static inline void mlx5e_ipsec_send_event(struct mlx5e_priv *priv, int event) { } + +static inline void mlx5e_ipsec_disable_events(struct mlx5e_priv *priv) +{ +} #endif #endif /* __MLX5E_IPSEC_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c index bf1d2769d4f1..feef86fff4bf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c @@ -2893,9 +2893,30 @@ void mlx5e_ipsec_handle_mpv_event(int event, struct mlx5e_priv *slave_priv, void mlx5e_ipsec_send_event(struct mlx5e_priv *priv, int event) { - if (!priv->ipsec) - return; /* IPsec not supported */ + if (!priv->ipsec || mlx5_devcom_comp_get_size(priv->devcom) < 2) + return; /* IPsec not supported or no peers */ mlx5_devcom_send_event(priv->devcom, event, event, priv); wait_for_completion(&priv->ipsec->comp); } + +void mlx5e_ipsec_disable_events(struct mlx5e_priv *priv) +{ + struct mlx5_devcom_comp_dev *tmp = NULL; + struct mlx5e_priv *peer_priv; + + if (!priv->devcom) + return; + + if (!mlx5_devcom_for_each_peer_begin(priv->devcom)) + goto out; + + peer_priv = mlx5_devcom_get_next_peer_data(priv->devcom, &tmp); + if (peer_priv) + complete_all(&peer_priv->ipsec->comp); + + mlx5_devcom_for_each_peer_end(priv->devcom); +out: + mlx5_devcom_unregister_component(priv->devcom); + priv->devcom = NULL; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 41fd5eee6306..9c46511e7b43 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -266,6 +266,7 @@ static void mlx5e_devcom_cleanup_mpv(struct mlx5e_priv *priv) } mlx5_devcom_unregister_component(priv->devcom); + priv->devcom = NULL; } static int blocking_event(struct notifier_block *nb, unsigned long event, void *data) @@ -6120,6 +6121,7 @@ static void mlx5e_nic_disable(struct mlx5e_priv *priv) if (mlx5e_monitor_counter_supported(priv)) mlx5e_monitor_counter_cleanup(priv); + mlx5e_ipsec_disable_events(priv); mlx5e_disable_blocking_events(priv); mlx5e_disable_async_events(priv); mlx5_lag_remove_netdev(mdev, priv->netdev); -- cgit v1.2.3