From 23f00807619d15063d676218f36c5dfeda1eb420 Mon Sep 17 00:00:00 2001 From: Mark Zhang Date: Tue, 25 Mar 2025 11:02:26 +0200 Subject: rtnetlink: Allocate vfinfo size for VF GUIDs when supported Commit 30aad41721e0 ("net/core: Add support for getting VF GUIDs") added support for getting VF port and node GUIDs in netlink ifinfo messages, but their size was not taken into consideration in the function that allocates the netlink message, causing the following warning when a netlink message is filled with many VF port and node GUIDs: # echo 64 > /sys/bus/pci/devices/0000\:08\:00.0/sriov_numvfs # ip link show dev ib0 RTNETLINK answers: Message too long Cannot send link get request: Message too long Kernel warning: ------------[ cut here ]------------ WARNING: CPU: 2 PID: 1930 at net/core/rtnetlink.c:4151 rtnl_getlink+0x586/0x5a0 Modules linked in: xt_conntrack xt_MASQUERADE nfnetlink xt_addrtype iptable_nat nf_nat br_netfilter overlay mlx5_ib macsec mlx5_core tls rpcrdma rdma_ucm ib_uverbs ib_iser libiscsi scsi_transport_iscsi ib_umad rdma_cm iw_cm ib_ipoib fuse ib_cm ib_core CPU: 2 UID: 0 PID: 1930 Comm: ip Not tainted 6.14.0-rc2+ #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 RIP: 0010:rtnl_getlink+0x586/0x5a0 Code: cb 82 e8 3d af 0a 00 4d 85 ff 0f 84 08 ff ff ff 4c 89 ff 41 be ea ff ff ff e8 66 63 5b ff 49 c7 07 80 4f cb 82 e9 36 fc ff ff <0f> 0b e9 16 fe ff ff e8 de a0 56 00 66 66 2e 0f 1f 84 00 00 00 00 RSP: 0018:ffff888113557348 EFLAGS: 00010246 RAX: 00000000ffffffa6 RBX: ffff88817e87aa34 RCX: dffffc0000000000 RDX: 0000000000000003 RSI: 0000000000000000 RDI: ffff88817e87afb8 RBP: 0000000000000009 R08: ffffffff821f44aa R09: 0000000000000000 R10: ffff8881260f79a8 R11: ffff88817e87af00 R12: ffff88817e87aa00 R13: ffffffff8563d300 R14: 00000000ffffffa6 R15: 00000000ffffffff FS: 00007f63a5dbf280(0000) GS:ffff88881ee00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f63a5ba4493 CR3: 00000001700fe002 CR4: 0000000000772eb0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: ? __warn+0xa5/0x230 ? rtnl_getlink+0x586/0x5a0 ? report_bug+0x22d/0x240 ? handle_bug+0x53/0xa0 ? exc_invalid_op+0x14/0x50 ? asm_exc_invalid_op+0x16/0x20 ? skb_trim+0x6a/0x80 ? rtnl_getlink+0x586/0x5a0 ? __pfx_rtnl_getlink+0x10/0x10 ? rtnetlink_rcv_msg+0x1e5/0x860 ? __pfx___mutex_lock+0x10/0x10 ? rcu_is_watching+0x34/0x60 ? __pfx_lock_acquire+0x10/0x10 ? stack_trace_save+0x90/0xd0 ? filter_irq_stacks+0x1d/0x70 ? kasan_save_stack+0x30/0x40 ? kasan_save_stack+0x20/0x40 ? kasan_save_track+0x10/0x30 rtnetlink_rcv_msg+0x21c/0x860 ? entry_SYSCALL_64_after_hwframe+0x76/0x7e ? __pfx_rtnetlink_rcv_msg+0x10/0x10 ? arch_stack_walk+0x9e/0xf0 ? rcu_is_watching+0x34/0x60 ? lock_acquire+0xd5/0x410 ? rcu_is_watching+0x34/0x60 netlink_rcv_skb+0xe0/0x210 ? __pfx_rtnetlink_rcv_msg+0x10/0x10 ? __pfx_netlink_rcv_skb+0x10/0x10 ? rcu_is_watching+0x34/0x60 ? __pfx___netlink_lookup+0x10/0x10 ? lock_release+0x62/0x200 ? netlink_deliver_tap+0xfd/0x290 ? rcu_is_watching+0x34/0x60 ? lock_release+0x62/0x200 ? netlink_deliver_tap+0x95/0x290 netlink_unicast+0x31f/0x480 ? __pfx_netlink_unicast+0x10/0x10 ? rcu_is_watching+0x34/0x60 ? lock_acquire+0xd5/0x410 netlink_sendmsg+0x369/0x660 ? lock_release+0x62/0x200 ? __pfx_netlink_sendmsg+0x10/0x10 ? import_ubuf+0xb9/0xf0 ? __import_iovec+0x254/0x2b0 ? lock_release+0x62/0x200 ? __pfx_netlink_sendmsg+0x10/0x10 ____sys_sendmsg+0x559/0x5a0 ? __pfx_____sys_sendmsg+0x10/0x10 ? __pfx_copy_msghdr_from_user+0x10/0x10 ? rcu_is_watching+0x34/0x60 ? do_read_fault+0x213/0x4a0 ? rcu_is_watching+0x34/0x60 ___sys_sendmsg+0xe4/0x150 ? __pfx____sys_sendmsg+0x10/0x10 ? do_fault+0x2cc/0x6f0 ? handle_pte_fault+0x2e3/0x3d0 ? __pfx_handle_pte_fault+0x10/0x10 ? preempt_count_sub+0x14/0xc0 ? __down_read_trylock+0x150/0x270 ? __handle_mm_fault+0x404/0x8e0 ? __pfx___handle_mm_fault+0x10/0x10 ? lock_release+0x62/0x200 ? __rcu_read_unlock+0x65/0x90 ? rcu_is_watching+0x34/0x60 __sys_sendmsg+0xd5/0x150 ? __pfx___sys_sendmsg+0x10/0x10 ? __up_read+0x192/0x480 ? lock_release+0x62/0x200 ? __rcu_read_unlock+0x65/0x90 ? rcu_is_watching+0x34/0x60 do_syscall_64+0x6d/0x140 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x7f63a5b13367 Code: 0e 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b9 0f 1f 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 89 54 24 1c 48 89 74 24 10 RSP: 002b:00007fff8c726bc8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e RAX: ffffffffffffffda RBX: 0000000067b687c2 RCX: 00007f63a5b13367 RDX: 0000000000000000 RSI: 00007fff8c726c30 RDI: 0000000000000004 RBP: 00007fff8c726cb8 R08: 0000000000000000 R09: 0000000000000034 R10: 00007fff8c726c7c R11: 0000000000000246 R12: 0000000000000001 R13: 0000000000000000 R14: 00007fff8c726cd0 R15: 00007fff8c726cd0 irq event stamp: 0 hardirqs last enabled at (0): [<0000000000000000>] 0x0 hardirqs last disabled at (0): [] copy_process+0xd08/0x2830 softirqs last enabled at (0): [] copy_process+0xd08/0x2830 softirqs last disabled at (0): [<0000000000000000>] 0x0 ---[ end trace 0000000000000000 ]--- Thus, when calculating ifinfo message size, take VF GUIDs sizes into account when supported. Fixes: 30aad41721e0 ("net/core: Add support for getting VF GUIDs") Signed-off-by: Mark Zhang Reviewed-by: Maher Sanalla Signed-off-by: Mark Bloch Reviewed-by: Sabrina Dubroca Link: https://patch.msgid.link/20250325090226.749730-1-mbloch@nvidia.com Signed-off-by: Jakub Kicinski --- net/core/rtnetlink.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 5a24a30dfc2d..334db17be37d 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1177,6 +1177,9 @@ static inline int rtnl_vfinfo_size(const struct net_device *dev, /* IFLA_VF_STATS_TX_DROPPED */ nla_total_size_64bit(sizeof(__u64))); } + if (dev->netdev_ops->ndo_get_vf_guid) + size += num_vfs * 2 * + nla_total_size(sizeof(struct ifla_vf_guid)); return size; } else return 0; -- cgit v1.2.3-59-g8ed1b From 67d1a8956d2d62fe6b4c13ebabb57806098511d8 Mon Sep 17 00:00:00 2001 From: Lubomir Rintel Date: Tue, 25 Mar 2025 10:58:41 +0100 Subject: rndis_host: Flag RNDIS modems as WWAN devices Set FLAG_WWAN instead of FLAG_ETHERNET for RNDIS interfaces on Mobile Broadband Modems, as opposed to regular Ethernet adapters. Otherwise NetworkManager gets confused, misjudges the device type, and wouldn't know it should connect a modem to get the device to work. What would be the result depends on ModemManager version -- older ModemManager would end up disconnecting a device after an unsuccessful probe attempt (if it connected without needing to unlock a SIM), while a newer one might spawn a separate PPP connection over a tty interface instead, resulting in a general confusion and no end of chaos. The only way to get this work reliably is to fix the device type and have good enough version ModemManager (or equivalent). Fixes: 63ba395cd7a5 ("rndis_host: support Novatel Verizon USB730L") Signed-off-by: Lubomir Rintel Link: https://patch.msgid.link/20250325095842.1567999-1-lkundrak@v3.sk Signed-off-by: Jakub Kicinski --- drivers/net/usb/rndis_host.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) diff --git a/drivers/net/usb/rndis_host.c b/drivers/net/usb/rndis_host.c index 7b3739b29c8f..bb0bf1415872 100644 --- a/drivers/net/usb/rndis_host.c +++ b/drivers/net/usb/rndis_host.c @@ -630,6 +630,16 @@ static const struct driver_info zte_rndis_info = { .tx_fixup = rndis_tx_fixup, }; +static const struct driver_info wwan_rndis_info = { + .description = "Mobile Broadband RNDIS device", + .flags = FLAG_WWAN | FLAG_POINTTOPOINT | FLAG_FRAMING_RN | FLAG_NO_SETINT, + .bind = rndis_bind, + .unbind = rndis_unbind, + .status = rndis_status, + .rx_fixup = rndis_rx_fixup, + .tx_fixup = rndis_tx_fixup, +}; + /*-------------------------------------------------------------------------*/ static const struct usb_device_id products [] = { @@ -666,9 +676,11 @@ static const struct usb_device_id products [] = { USB_INTERFACE_INFO(USB_CLASS_WIRELESS_CONTROLLER, 1, 3), .driver_info = (unsigned long) &rndis_info, }, { - /* Novatel Verizon USB730L */ + /* Mobile Broadband Modem, seen in Novatel Verizon USB730L and + * Telit FN990A (RNDIS) + */ USB_INTERFACE_INFO(USB_CLASS_MISC, 4, 1), - .driver_info = (unsigned long) &rndis_info, + .driver_info = (unsigned long)&wwan_rndis_info, }, { }, // END }; -- cgit v1.2.3-59-g8ed1b From 2eb6c6a34cb1c22b09b219390cdff0f02cd90258 Mon Sep 17 00:00:00 2001 From: Stanislav Fomichev Date: Tue, 25 Mar 2025 10:54:27 -0700 Subject: net: move replay logic to tc_modify_qdisc Eric reports that by the time we call netdev_lock_ops after rtnl_unlock/rtnl_lock, the dev might point to an invalid device. As suggested by Jakub in [0], move rtnl lock/unlock and request_module outside of qdisc_create. This removes extra complexity with relocking the netdev. 0: https://lore.kernel.org/netdev/20250325032803.1542c15e@kernel.org/ Fixes: a0527ee2df3f ("net: hold netdev instance lock during qdisc ndo_setup_tc") Reported-by: Eric Dumazet Link: https://lore.kernel.org/netdev/20250305163732.2766420-1-sdf@fomichev.me/T/#me8dfd778ea4c4463acab55644e3f9836bc608771 Signed-off-by: Stanislav Fomichev Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20250325175427.3818808-1-sdf@fomichev.me Signed-off-by: Jakub Kicinski --- net/sched/sch_api.c | 73 ++++++++++++++++++++--------------------------------- 1 file changed, 27 insertions(+), 46 deletions(-) diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index defb05c1fba4..f74a097f54ae 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1267,38 +1267,8 @@ static struct Qdisc *qdisc_create(struct net_device *dev, struct qdisc_size_table *stab; ops = qdisc_lookup_ops(kind); -#ifdef CONFIG_MODULES - if (ops == NULL && kind != NULL) { - char name[IFNAMSIZ]; - if (nla_strscpy(name, kind, IFNAMSIZ) >= 0) { - /* We dropped the RTNL semaphore in order to - * perform the module load. So, even if we - * succeeded in loading the module we have to - * tell the caller to replay the request. We - * indicate this using -EAGAIN. - * We replay the request because the device may - * go away in the mean time. - */ - netdev_unlock_ops(dev); - rtnl_unlock(); - request_module(NET_SCH_ALIAS_PREFIX "%s", name); - rtnl_lock(); - netdev_lock_ops(dev); - ops = qdisc_lookup_ops(kind); - if (ops != NULL) { - /* We will try again qdisc_lookup_ops, - * so don't keep a reference. - */ - module_put(ops->owner); - err = -EAGAIN; - goto err_out; - } - } - } -#endif - - err = -ENOENT; if (!ops) { + err = -ENOENT; NL_SET_ERR_MSG(extack, "Specified qdisc kind is unknown"); goto err_out; } @@ -1623,8 +1593,7 @@ static int __tc_modify_qdisc(struct sk_buff *skb, struct nlmsghdr *n, struct netlink_ext_ack *extack, struct net_device *dev, struct nlattr *tca[TCA_MAX + 1], - struct tcmsg *tcm, - bool *replay) + struct tcmsg *tcm) { struct Qdisc *q = NULL; struct Qdisc *p = NULL; @@ -1789,13 +1758,8 @@ create_n_graft2: tcm->tcm_parent, tcm->tcm_handle, tca, &err, extack); } - if (q == NULL) { - if (err == -EAGAIN) { - *replay = true; - return 0; - } + if (!q) return err; - } graft: err = qdisc_graft(dev, p, skb, n, clid, q, NULL, extack); @@ -1808,6 +1772,27 @@ graft: return 0; } +static void request_qdisc_module(struct nlattr *kind) +{ + struct Qdisc_ops *ops; + char name[IFNAMSIZ]; + + if (!kind) + return; + + ops = qdisc_lookup_ops(kind); + if (ops) { + module_put(ops->owner); + return; + } + + if (nla_strscpy(name, kind, IFNAMSIZ) >= 0) { + rtnl_unlock(); + request_module(NET_SCH_ALIAS_PREFIX "%s", name); + rtnl_lock(); + } +} + /* * Create/change qdisc. */ @@ -1818,27 +1803,23 @@ static int tc_modify_qdisc(struct sk_buff *skb, struct nlmsghdr *n, struct nlattr *tca[TCA_MAX + 1]; struct net_device *dev; struct tcmsg *tcm; - bool replay; int err; -replay: - /* Reinit, just in case something touches this. */ err = nlmsg_parse_deprecated(n, sizeof(*tcm), tca, TCA_MAX, rtm_tca_policy, extack); if (err < 0) return err; + request_qdisc_module(tca[TCA_KIND]); + tcm = nlmsg_data(n); dev = __dev_get_by_index(net, tcm->tcm_ifindex); if (!dev) return -ENODEV; - replay = false; netdev_lock_ops(dev); - err = __tc_modify_qdisc(skb, n, extack, dev, tca, tcm, &replay); + err = __tc_modify_qdisc(skb, n, extack, dev, tca, tcm); netdev_unlock_ops(dev); - if (replay) - goto replay; return err; } -- cgit v1.2.3-59-g8ed1b From 52c19f901318d32e01a36d975ea2fdd0a26f56e7 Mon Sep 17 00:00:00 2001 From: Maxime Chevallier Date: Thu, 27 Mar 2025 12:00:12 +0100 Subject: MAINTAINERS: Add dedicated entries for phy_link_topology The infrastructure to handle multi-phy devices is fairly standalone. Add myself as maintainer for that part as well as the netlink uAPI that exposes it. Reviewed-by: Andrew Lunn Acked-by: Jakub Kicinski Signed-off-by: Maxime Chevallier Link: https://patch.msgid.link/20250327110013.106865-1-maxime.chevallier@bootlin.com Signed-off-by: Jakub Kicinski --- MAINTAINERS | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 1cd25139cc58..36511ed5bf6a 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -16577,6 +16577,13 @@ F: net/ethtool/mm.c F: tools/testing/selftests/drivers/net/hw/ethtool_mm.sh K: ethtool_mm +NETWORKING [ETHTOOL PHY TOPOLOGY] +M: Maxime Chevallier +F: Documentation/networking/phy-link-topology.rst +F: drivers/net/phy/phy_link_topology.c +F: include/linux/phy_link_topology.h +F: net/ethtool/phy.c + NETWORKING [GENERAL] M: "David S. Miller" M: Eric Dumazet -- cgit v1.2.3-59-g8ed1b From fa37a8849634db2dd3545116873da8cf4b1e67c6 Mon Sep 17 00:00:00 2001 From: Haiyang Zhang Date: Tue, 25 Mar 2025 09:32:37 -0700 Subject: net: mana: Switch to page pool for jumbo frames Frag allocators, such as netdev_alloc_frag(), were not designed to work for fragsz > PAGE_SIZE. So, switch to page pool for jumbo frames instead of using page frag allocators. This driver is using page pool for smaller MTUs already. Cc: stable@vger.kernel.org Fixes: 80f6215b450e ("net: mana: Add support for jumbo frame") Signed-off-by: Haiyang Zhang Reviewed-by: Long Li Reviewed-by: Shradha Gupta Link: https://patch.msgid.link/1742920357-27263-1-git-send-email-haiyangz@microsoft.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/microsoft/mana/mana_en.c | 46 ++++++--------------------- 1 file changed, 9 insertions(+), 37 deletions(-) diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c index e190d5ee5154..3ac73d97337e 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_en.c +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c @@ -661,30 +661,16 @@ int mana_pre_alloc_rxbufs(struct mana_port_context *mpc, int new_mtu, int num_qu mpc->rxbpre_total = 0; for (i = 0; i < num_rxb; i++) { - if (mpc->rxbpre_alloc_size > PAGE_SIZE) { - va = netdev_alloc_frag(mpc->rxbpre_alloc_size); - if (!va) - goto error; - - page = virt_to_head_page(va); - /* Check if the frag falls back to single page */ - if (compound_order(page) < - get_order(mpc->rxbpre_alloc_size)) { - put_page(page); - goto error; - } - } else { - page = dev_alloc_page(); - if (!page) - goto error; + page = dev_alloc_pages(get_order(mpc->rxbpre_alloc_size)); + if (!page) + goto error; - va = page_to_virt(page); - } + va = page_to_virt(page); da = dma_map_single(dev, va + mpc->rxbpre_headroom, mpc->rxbpre_datasize, DMA_FROM_DEVICE); if (dma_mapping_error(dev, da)) { - put_page(virt_to_head_page(va)); + put_page(page); goto error; } @@ -1676,7 +1662,7 @@ drop: } static void *mana_get_rxfrag(struct mana_rxq *rxq, struct device *dev, - dma_addr_t *da, bool *from_pool, bool is_napi) + dma_addr_t *da, bool *from_pool) { struct page *page; void *va; @@ -1687,21 +1673,6 @@ static void *mana_get_rxfrag(struct mana_rxq *rxq, struct device *dev, if (rxq->xdp_save_va) { va = rxq->xdp_save_va; rxq->xdp_save_va = NULL; - } else if (rxq->alloc_size > PAGE_SIZE) { - if (is_napi) - va = napi_alloc_frag(rxq->alloc_size); - else - va = netdev_alloc_frag(rxq->alloc_size); - - if (!va) - return NULL; - - page = virt_to_head_page(va); - /* Check if the frag falls back to single page */ - if (compound_order(page) < get_order(rxq->alloc_size)) { - put_page(page); - return NULL; - } } else { page = page_pool_dev_alloc_pages(rxq->page_pool); if (!page) @@ -1734,7 +1705,7 @@ static void mana_refill_rx_oob(struct device *dev, struct mana_rxq *rxq, dma_addr_t da; void *va; - va = mana_get_rxfrag(rxq, dev, &da, &from_pool, true); + va = mana_get_rxfrag(rxq, dev, &da, &from_pool); if (!va) return; @@ -2176,7 +2147,7 @@ static int mana_fill_rx_oob(struct mana_recv_buf_oob *rx_oob, u32 mem_key, if (mpc->rxbufs_pre) va = mana_get_rxbuf_pre(rxq, &da); else - va = mana_get_rxfrag(rxq, dev, &da, &from_pool, false); + va = mana_get_rxfrag(rxq, dev, &da, &from_pool); if (!va) return -ENOMEM; @@ -2262,6 +2233,7 @@ static int mana_create_page_pool(struct mana_rxq *rxq, struct gdma_context *gc) pprm.nid = gc->numa_node; pprm.napi = &rxq->rx_cq.napi; pprm.netdev = rxq->ndev; + pprm.order = get_order(rxq->alloc_size); rxq->page_pool = page_pool_create(&pprm); -- cgit v1.2.3-59-g8ed1b From fab05835688526f9de123d1e98e4d1f838da4e22 Mon Sep 17 00:00:00 2001 From: Lama Kayal Date: Sun, 23 Mar 2025 14:28:26 +0200 Subject: net/mlx5e: SHAMPO, Make reserved size independent of page size When hw-gro is enabled, the maximum number of header entries that are needed per wqe (hd_per_wqe) is calculated based on the size of the reservations among other parameters. Miscalculation of the size of reservations leads to incorrect calculation of hd_per_wqe as 0, particularly in the case of large page size like in aarch64, this prevents the SHAMPO header from being correctly initialized in the device, ultimately causing the following cqe err that indicates a violation of PD. mlx5_core 0000:00:08.0 eth2: ERR CQE on RQ: 0x1180 mlx5_core 0000:00:08.0 eth2: Error cqe on cqn 0x510, ci 0x0, qn 0x1180, opcode 0xe, syndrome 0x4, vendor syndrome 0x32 00000000: 00 00 00 00 04 4a 00 00 00 00 00 00 20 00 93 32 00000010: 55 00 00 00 fb cc 00 00 00 00 00 00 07 18 00 00 00000020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 4a 00000030: 00 00 00 9a 93 00 32 04 00 00 00 00 00 00 da e1 Use the correct formula for calculating the size of reservations, precisely it shouldn't be dependent on page size, instead use the correct multiply of MLX5E_SHAMPO_WQ_BASE_RESRV_SIZE. Fixes: e5ca8fb08ab2 ("net/mlx5e: Add control path for SHAMPO feature") Signed-off-by: Lama Kayal Reviewed-by: Dragos Tatulea Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1742732906-166564-1-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/en/params.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c index aa36670d9a36..58ec5e44aa7a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c @@ -430,7 +430,7 @@ u8 mlx5e_shampo_get_log_pkt_per_rsrv(struct mlx5_core_dev *mdev, struct mlx5e_params *params) { u32 resrv_size = BIT(mlx5e_shampo_get_log_rsrv_size(mdev, params)) * - PAGE_SIZE; + MLX5E_SHAMPO_WQ_BASE_RESRV_SIZE; return order_base_2(DIV_ROUND_UP(resrv_size, params->sw_mtu)); } @@ -834,7 +834,8 @@ static u32 mlx5e_shampo_get_log_cq_size(struct mlx5_core_dev *mdev, struct mlx5e_params *params, struct mlx5e_xsk_param *xsk) { - int rsrv_size = BIT(mlx5e_shampo_get_log_rsrv_size(mdev, params)) * PAGE_SIZE; + int rsrv_size = BIT(mlx5e_shampo_get_log_rsrv_size(mdev, params)) * + MLX5E_SHAMPO_WQ_BASE_RESRV_SIZE; u16 num_strides = BIT(mlx5e_mpwqe_get_log_num_strides(mdev, params, xsk)); int pkt_per_rsrv = BIT(mlx5e_shampo_get_log_pkt_per_rsrv(mdev, params)); u8 log_stride_sz = mlx5e_mpwqe_get_log_stride_size(mdev, params, xsk); @@ -1043,7 +1044,8 @@ u32 mlx5e_shampo_hd_per_wqe(struct mlx5_core_dev *mdev, struct mlx5e_params *params, struct mlx5e_rq_param *rq_param) { - int resv_size = BIT(mlx5e_shampo_get_log_rsrv_size(mdev, params)) * PAGE_SIZE; + int resv_size = BIT(mlx5e_shampo_get_log_rsrv_size(mdev, params)) * + MLX5E_SHAMPO_WQ_BASE_RESRV_SIZE; u16 num_strides = BIT(mlx5e_mpwqe_get_log_num_strides(mdev, params, NULL)); int pkt_per_resv = BIT(mlx5e_shampo_get_log_pkt_per_rsrv(mdev, params)); u8 log_stride_sz = mlx5e_mpwqe_get_log_stride_size(mdev, params, NULL); -- cgit v1.2.3-59-g8ed1b From 2ea396448f26d0d7d66224cb56500a6789c7ed07 Mon Sep 17 00:00:00 2001 From: Dominique Martinet Date: Wed, 26 Mar 2025 17:32:36 +0900 Subject: net: usb: usbnet: restore usb%d name exception for local mac addresses commit 8a7d12d674ac ("net: usb: usbnet: fix name regression") assumed that local addresses always came from the kernel, but some devices hand out local mac addresses so we ended up with point-to-point devices with a mac set by the driver, renaming to eth%d when they used to be named usb%d. Userspace should not rely on device name, but for the sake of stability restore the local mac address check portion of the naming exception: point to point devices which either have no mac set by the driver or have a local mac handed out by the driver will keep the usb%d name. (some USB LTE modems are known to hand out a stable mac from the locally administered range; that mac appears to be random (different for mulitple devices) and can be reset with device-specific commands, so while such devices would benefit from getting a OUI reserved, we have to deal with these and might as well preserve the existing behavior to avoid breaking fragile openwrt configurations and such on upgrade.) Link: https://lkml.kernel.org/r/20241203130457.904325-1-asmadeus@codewreck.org Fixes: 8a7d12d674ac ("net: usb: usbnet: fix name regression") Cc: stable@vger.kernel.org Tested-by: Ahmed Naseef Signed-off-by: Dominique Martinet Acked-by: Oliver Neukum Link: https://patch.msgid.link/20250326-usbnet_rename-v2-1-57eb21fcff26@atmark-techno.com Signed-off-by: Jakub Kicinski --- drivers/net/usb/usbnet.c | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-) diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c index 44179f4e807f..aeab2308b150 100644 --- a/drivers/net/usb/usbnet.c +++ b/drivers/net/usb/usbnet.c @@ -178,6 +178,17 @@ int usbnet_get_ethernet_addr(struct usbnet *dev, int iMACAddress) } EXPORT_SYMBOL_GPL(usbnet_get_ethernet_addr); +static bool usbnet_needs_usb_name_format(struct usbnet *dev, struct net_device *net) +{ + /* Point to point devices which don't have a real MAC address + * (or report a fake local one) have historically used the usb%d + * naming. Preserve this.. + */ + return (dev->driver_info->flags & FLAG_POINTTOPOINT) != 0 && + (is_zero_ether_addr(net->dev_addr) || + is_local_ether_addr(net->dev_addr)); +} + static void intr_complete (struct urb *urb) { struct usbnet *dev = urb->context; @@ -1762,13 +1773,11 @@ usbnet_probe (struct usb_interface *udev, const struct usb_device_id *prod) if (status < 0) goto out1; - // heuristic: "usb%d" for links we know are two-host, - // else "eth%d" when there's reasonable doubt. userspace - // can rename the link if it knows better. + /* heuristic: rename to "eth%d" if we are not sure this link + * is two-host (these links keep "usb%d") + */ if ((dev->driver_info->flags & FLAG_ETHER) != 0 && - ((dev->driver_info->flags & FLAG_POINTTOPOINT) == 0 || - /* somebody touched it*/ - !is_zero_ether_addr(net->dev_addr))) + !usbnet_needs_usb_name_format(dev, net)) strscpy(net->name, "eth%d", sizeof(net->name)); /* WLAN devices should always be named "wlan%d" */ if ((dev->driver_info->flags & FLAG_WLAN) != 0) -- cgit v1.2.3-59-g8ed1b From 4f1eaabb4b66a1f7473f584e14e15b2ac19dfaf3 Mon Sep 17 00:00:00 2001 From: Jim Liu Date: Thu, 27 Mar 2025 14:29:42 +0800 Subject: net: phy: broadcom: Correct BCM5221 PHY model detection Correct detect condition is applied to the entire 5221 family of PHYs. Fixes: 3abbd0699b67 ("net: phy: broadcom: add support for BCM5221 phy") Signed-off-by: Jim Liu Reviewed-by: Michal Swiatkowski Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/broadcom.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/phy/broadcom.c b/drivers/net/phy/broadcom.c index 13e43fee1906..9b1de54fd483 100644 --- a/drivers/net/phy/broadcom.c +++ b/drivers/net/phy/broadcom.c @@ -859,7 +859,7 @@ static int brcm_fet_config_init(struct phy_device *phydev) return reg; /* Unmask events we are interested in and mask interrupts globally. */ - if (phydev->phy_id == PHY_ID_BCM5221) + if (phydev->drv->phy_id == PHY_ID_BCM5221) reg = MII_BRCM_FET_IR_ENABLE | MII_BRCM_FET_IR_MASK; else @@ -888,7 +888,7 @@ static int brcm_fet_config_init(struct phy_device *phydev) return err; } - if (phydev->phy_id != PHY_ID_BCM5221) { + if (phydev->drv->phy_id != PHY_ID_BCM5221) { /* Set the LED mode */ reg = __phy_read(phydev, MII_BRCM_FET_SHDW_AUXMODE4); if (reg < 0) { @@ -1009,7 +1009,7 @@ static int brcm_fet_suspend(struct phy_device *phydev) return err; } - if (phydev->phy_id == PHY_ID_BCM5221) + if (phydev->drv->phy_id == PHY_ID_BCM5221) /* Force Low Power Mode with clock enabled */ reg = BCM5221_SHDW_AM4_EN_CLK_LPM | BCM5221_SHDW_AM4_FORCE_LPM; else -- cgit v1.2.3-59-g8ed1b From 7220e8f4d4eec0b2f682eef45e2d36c092738413 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 27 Mar 2025 14:44:39 +0000 Subject: net: lapbether: use netdev_lockdep_set_classes() helper drivers/net/wan/lapbether.c uses stacked devices. Like similar drivers, it must use netdev_lockdep_set_classes() to avoid LOCKDEP splats. This is similar to commit 9bfc9d65a1dc ("hamradio: use netdev_lockdep_set_classes() helper") Fixes: 7e4d784f5810 ("net: hold netdev instance lock during rtnetlink operations") Reported-by: syzbot+377b71db585c9c705f8e@syzkaller.appspotmail.com Closes: https://lore.kernel.org/lkml/67cd611c.050a0220.14db68.0073.GAE@google.com/T/#u Signed-off-by: Eric Dumazet Acked-by: Stanislav Fomichev Link: https://patch.msgid.link/20250327144439.2463509-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- drivers/net/wan/lapbether.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/wan/lapbether.c b/drivers/net/wan/lapbether.c index 56326f38fe8a..995a7207bdf8 100644 --- a/drivers/net/wan/lapbether.c +++ b/drivers/net/wan/lapbether.c @@ -39,6 +39,7 @@ #include #include +#include #include static const u8 bcast_addr[6] = { 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF }; @@ -366,6 +367,7 @@ static const struct net_device_ops lapbeth_netdev_ops = { static void lapbeth_setup(struct net_device *dev) { + netdev_lockdep_set_classes(dev); dev->netdev_ops = &lapbeth_netdev_ops; dev->needs_free_netdev = true; dev->type = ARPHRD_X25; -- cgit v1.2.3-59-g8ed1b From e514d77334a63f1dcb9a3b47d5aee8f51d66cb1d Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 27 Mar 2025 15:23:13 -0700 Subject: selftests: drv-net: replace the rpath helper with Path objects The single letter + "path" helpers do not have many fans (see Link). Use a Path object with a better name. test_dir is the replacement for rpath(), net_lib_dir is a new path of the $ksft/net/lib directory. The Path() class overloads the "/" operator and can be cast to string automatically, so to get a path to a file tests can do: path = env.test_dir / "binary" Link: https://lore.kernel.org/CA+FuTSemTNVZ5MxXkq8T9P=DYm=nSXcJnL7CJBPZNAT_9UFisQ@mail.gmail.com Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20250327222315.1098596-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/hds.py | 2 +- tools/testing/selftests/drivers/net/hw/csum.py | 2 +- tools/testing/selftests/drivers/net/hw/irq.py | 2 +- tools/testing/selftests/drivers/net/lib/py/env.py | 21 ++++++++------------- tools/testing/selftests/drivers/net/queues.py | 4 ++-- 5 files changed, 13 insertions(+), 18 deletions(-) diff --git a/tools/testing/selftests/drivers/net/hds.py b/tools/testing/selftests/drivers/net/hds.py index 7cc74faed743..8b7f6acad15f 100755 --- a/tools/testing/selftests/drivers/net/hds.py +++ b/tools/testing/selftests/drivers/net/hds.py @@ -20,7 +20,7 @@ def _get_hds_mode(cfg, netnl) -> str: def _xdp_onoff(cfg): - prog = cfg.rpath("../../net/lib/xdp_dummy.bpf.o") + prog = cfg.net_lib_dir / "xdp_dummy.bpf.o" ip("link set dev %s xdp obj %s sec xdp" % (cfg.ifname, prog)) ip("link set dev %s xdp off" % cfg.ifname) diff --git a/tools/testing/selftests/drivers/net/hw/csum.py b/tools/testing/selftests/drivers/net/hw/csum.py index 701aca1361e0..cd23af875317 100755 --- a/tools/testing/selftests/drivers/net/hw/csum.py +++ b/tools/testing/selftests/drivers/net/hw/csum.py @@ -88,7 +88,7 @@ def main() -> None: with NetDrvEpEnv(__file__, nsim_test=False) as cfg: check_nic_features(cfg) - cfg.bin_local = cfg.rpath("../../../net/lib/csum") + cfg.bin_local = cfg.net_lib_dir / "csum" cfg.bin_remote = cfg.remote.deploy(cfg.bin_local) cases = [] diff --git a/tools/testing/selftests/drivers/net/hw/irq.py b/tools/testing/selftests/drivers/net/hw/irq.py index 42ab98370245..d772a18d8a1b 100755 --- a/tools/testing/selftests/drivers/net/hw/irq.py +++ b/tools/testing/selftests/drivers/net/hw/irq.py @@ -69,7 +69,7 @@ def check_reconfig_queues(cfg) -> None: def check_reconfig_xdp(cfg) -> None: def reconfig(cfg) -> None: ip(f"link set dev %s xdp obj %s sec xdp" % - (cfg.ifname, cfg.rpath("xdp_dummy.bpf.o"))) + (cfg.ifname, cfg.test_dir / "xdp_dummy.bpf.o")) ip(f"link set dev %s xdp off" % cfg.ifname) _check_reconfig(cfg, reconfig) diff --git a/tools/testing/selftests/drivers/net/lib/py/env.py b/tools/testing/selftests/drivers/net/lib/py/env.py index fd4d674e6c72..ad5ff645183a 100644 --- a/tools/testing/selftests/drivers/net/lib/py/env.py +++ b/tools/testing/selftests/drivers/net/lib/py/env.py @@ -13,22 +13,17 @@ from .remote import Remote class NetDrvEnvBase: """ Base class for a NIC / host envirnoments + + Attributes: + test_dir: Path to the source directory of the test + net_lib_dir: Path to the net/lib directory """ def __init__(self, src_path): - self.src_path = src_path - self.env = self._load_env_file() - - def rpath(self, path): - """ - Get an absolute path to a file based on a path relative to the directory - containing the test which constructed env. + self.src_path = Path(src_path) + self.test_dir = self.src_path.parent.resolve() + self.net_lib_dir = (Path(__file__).parent / "../../../../net/lib").resolve() - For example, if the test.py is in the same directory as - a binary (built from helper.c), the test can use env.rpath("helper") - to get the absolute path to the binary - """ - src_dir = Path(self.src_path).parent.resolve() - return (src_dir / path).as_posix() + self.env = self._load_env_file() def _load_env_file(self): env = os.environ.copy() diff --git a/tools/testing/selftests/drivers/net/queues.py b/tools/testing/selftests/drivers/net/queues.py index cae923f84f69..06abd3f233e1 100755 --- a/tools/testing/selftests/drivers/net/queues.py +++ b/tools/testing/selftests/drivers/net/queues.py @@ -26,13 +26,13 @@ def nl_get_queues(cfg, nl, qtype='rx'): def check_xsk(cfg, nl, xdp_queue_id=0) -> None: # Probe for support - xdp = cmd(cfg.rpath("xdp_helper") + ' - -', fail=False) + xdp = cmd(f'{cfg.test_dir / "xdp_helper"} - -', fail=False) if xdp.ret == 255: raise KsftSkipEx('AF_XDP unsupported') elif xdp.ret > 0: raise KsftFailEx('unable to create AF_XDP socket') - with bkg(f'{cfg.rpath("xdp_helper")} {cfg.ifindex} {xdp_queue_id}', + with bkg(f'{cfg.test_dir / "xdp_helper"} {cfg.ifindex} {xdp_queue_id}', ksft_wait=3): rx = tx = False -- cgit v1.2.3-59-g8ed1b From c231e12ecd45fcb34ff3b52d6557d614ba49b699 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 27 Mar 2025 15:23:14 -0700 Subject: selftests: net: use the dummy bpf from net/lib Commit 29b036be1b0b ("selftests: drv-net: test XDP, HDS auto and the ioctl path") added an sample XDP_PASS prog in net/lib, so that we can reuse it in various sub-directories. Delete the old sample and use the one from the lib in existing tests. Acked-by: Stanislav Fomichev Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20250327222315.1098596-3-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/hw/irq.py | 2 +- tools/testing/selftests/drivers/net/hw/xdp_dummy.bpf.c | 13 ------------- tools/testing/selftests/net/udpgro_bench.sh | 2 +- tools/testing/selftests/net/udpgro_frglist.sh | 2 +- tools/testing/selftests/net/udpgro_fwd.sh | 2 +- tools/testing/selftests/net/veth.sh | 2 +- tools/testing/selftests/net/xdp_dummy.bpf.c | 13 ------------- 7 files changed, 5 insertions(+), 31 deletions(-) delete mode 100644 tools/testing/selftests/drivers/net/hw/xdp_dummy.bpf.c delete mode 100644 tools/testing/selftests/net/xdp_dummy.bpf.c diff --git a/tools/testing/selftests/drivers/net/hw/irq.py b/tools/testing/selftests/drivers/net/hw/irq.py index d772a18d8a1b..0699d6a8b4e2 100755 --- a/tools/testing/selftests/drivers/net/hw/irq.py +++ b/tools/testing/selftests/drivers/net/hw/irq.py @@ -69,7 +69,7 @@ def check_reconfig_queues(cfg) -> None: def check_reconfig_xdp(cfg) -> None: def reconfig(cfg) -> None: ip(f"link set dev %s xdp obj %s sec xdp" % - (cfg.ifname, cfg.test_dir / "xdp_dummy.bpf.o")) + (cfg.ifname, cfg.net_lib_dir / "xdp_dummy.bpf.o")) ip(f"link set dev %s xdp off" % cfg.ifname) _check_reconfig(cfg, reconfig) diff --git a/tools/testing/selftests/drivers/net/hw/xdp_dummy.bpf.c b/tools/testing/selftests/drivers/net/hw/xdp_dummy.bpf.c deleted file mode 100644 index d988b2e0cee8..000000000000 --- a/tools/testing/selftests/drivers/net/hw/xdp_dummy.bpf.c +++ /dev/null @@ -1,13 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 - -#define KBUILD_MODNAME "xdp_dummy" -#include -#include - -SEC("xdp") -int xdp_dummy_prog(struct xdp_md *ctx) -{ - return XDP_PASS; -} - -char _license[] SEC("license") = "GPL"; diff --git a/tools/testing/selftests/net/udpgro_bench.sh b/tools/testing/selftests/net/udpgro_bench.sh index c51ea90a1395..815fad8c53a8 100755 --- a/tools/testing/selftests/net/udpgro_bench.sh +++ b/tools/testing/selftests/net/udpgro_bench.sh @@ -7,7 +7,7 @@ source net_helper.sh readonly PEER_NS="ns-peer-$(mktemp -u XXXXXX)" -BPF_FILE="xdp_dummy.bpf.o" +BPF_FILE="lib/xdp_dummy.bpf.o" cleanup() { local -r jobs="$(jobs -p)" diff --git a/tools/testing/selftests/net/udpgro_frglist.sh b/tools/testing/selftests/net/udpgro_frglist.sh index 17404f49cdb6..5f3d1a110d11 100755 --- a/tools/testing/selftests/net/udpgro_frglist.sh +++ b/tools/testing/selftests/net/udpgro_frglist.sh @@ -7,7 +7,7 @@ source net_helper.sh readonly PEER_NS="ns-peer-$(mktemp -u XXXXXX)" -BPF_FILE="xdp_dummy.bpf.o" +BPF_FILE="lib/xdp_dummy.bpf.o" cleanup() { local -r jobs="$(jobs -p)" diff --git a/tools/testing/selftests/net/udpgro_fwd.sh b/tools/testing/selftests/net/udpgro_fwd.sh index 550d8eb3e224..f22f6c66997e 100755 --- a/tools/testing/selftests/net/udpgro_fwd.sh +++ b/tools/testing/selftests/net/udpgro_fwd.sh @@ -3,7 +3,7 @@ source net_helper.sh -BPF_FILE="xdp_dummy.bpf.o" +BPF_FILE="lib/xdp_dummy.bpf.o" readonly BASE="ns-$(mktemp -u XXXXXX)" readonly SRC=2 readonly DST=1 diff --git a/tools/testing/selftests/net/veth.sh b/tools/testing/selftests/net/veth.sh index 6bb7dfaa30b6..9709dd067c72 100755 --- a/tools/testing/selftests/net/veth.sh +++ b/tools/testing/selftests/net/veth.sh @@ -1,7 +1,7 @@ #!/bin/sh # SPDX-License-Identifier: GPL-2.0 -BPF_FILE="xdp_dummy.bpf.o" +BPF_FILE="lib/xdp_dummy.bpf.o" readonly STATS="$(mktemp -p /tmp ns-XXXXXX)" readonly BASE=`basename $STATS` readonly SRC=2 diff --git a/tools/testing/selftests/net/xdp_dummy.bpf.c b/tools/testing/selftests/net/xdp_dummy.bpf.c deleted file mode 100644 index d988b2e0cee8..000000000000 --- a/tools/testing/selftests/net/xdp_dummy.bpf.c +++ /dev/null @@ -1,13 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 - -#define KBUILD_MODNAME "xdp_dummy" -#include -#include - -SEC("xdp") -int xdp_dummy_prog(struct xdp_md *ctx) -{ - return XDP_PASS; -} - -char _license[] SEC("license") = "GPL"; -- cgit v1.2.3-59-g8ed1b From 88dec030dfcd72fa4322181eb64db06c514f33b1 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 27 Mar 2025 15:23:15 -0700 Subject: selftests: net: use Path helpers in ping Now that net and net-next have converged we can use the Path helpers in the ping test without conflicts. Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20250327222315.1098596-4-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/ping.py | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-) diff --git a/tools/testing/selftests/drivers/net/ping.py b/tools/testing/selftests/drivers/net/ping.py index 93120e86e102..4b6822866066 100755 --- a/tools/testing/selftests/drivers/net/ping.py +++ b/tools/testing/selftests/drivers/net/ping.py @@ -56,8 +56,7 @@ def _set_offload_checksum(cfg, netnl, on) -> None: return def _set_xdp_generic_sb_on(cfg) -> None: - test_dir = os.path.dirname(os.path.realpath(__file__)) - prog = test_dir + "/../../net/lib/xdp_dummy.bpf.o" + prog = cfg.net_lib_dir / "xdp_dummy.bpf.o" cmd(f"ip link set dev {remote_ifname} mtu 1500", shell=True, host=cfg.remote) cmd(f"ip link set dev {cfg.ifname} mtu 1500 xdpgeneric obj {prog} sec xdp", shell=True) defer(cmd, f"ip link set dev {cfg.ifname} xdpgeneric off") @@ -66,8 +65,7 @@ def _set_xdp_generic_sb_on(cfg) -> None: time.sleep(10) def _set_xdp_generic_mb_on(cfg) -> None: - test_dir = os.path.dirname(os.path.realpath(__file__)) - prog = test_dir + "/../../net/lib/xdp_dummy.bpf.o" + prog = cfg.net_lib_dir / "xdp_dummy.bpf.o" cmd(f"ip link set dev {remote_ifname} mtu 9000", shell=True, host=cfg.remote) defer(ip, f"link set dev {remote_ifname} mtu 1500", host=cfg.remote) ip("link set dev %s mtu 9000 xdpgeneric obj %s sec xdp.frags" % (cfg.ifname, prog)) @@ -77,8 +75,7 @@ def _set_xdp_generic_mb_on(cfg) -> None: time.sleep(10) def _set_xdp_native_sb_on(cfg) -> None: - test_dir = os.path.dirname(os.path.realpath(__file__)) - prog = test_dir + "/../../net/lib/xdp_dummy.bpf.o" + prog = cfg.net_lib_dir / "xdp_dummy.bpf.o" cmd(f"ip link set dev {remote_ifname} mtu 1500", shell=True, host=cfg.remote) cmd(f"ip -j link set dev {cfg.ifname} mtu 1500 xdp obj {prog} sec xdp", shell=True) defer(ip, f"link set dev {cfg.ifname} mtu 1500 xdp off") @@ -95,8 +92,7 @@ def _set_xdp_native_sb_on(cfg) -> None: time.sleep(10) def _set_xdp_native_mb_on(cfg) -> None: - test_dir = os.path.dirname(os.path.realpath(__file__)) - prog = test_dir + "/../../net/lib/xdp_dummy.bpf.o" + prog = cfg.net_lib_dir / "xdp_dummy.bpf.o" cmd(f"ip link set dev {remote_ifname} mtu 9000", shell=True, host=cfg.remote) defer(ip, f"link set dev {remote_ifname} mtu 1500", host=cfg.remote) try: @@ -109,8 +105,7 @@ def _set_xdp_native_mb_on(cfg) -> None: time.sleep(10) def _set_xdp_offload_on(cfg) -> None: - test_dir = os.path.dirname(os.path.realpath(__file__)) - prog = test_dir + "/../../net/lib/xdp_dummy.bpf.o" + prog = cfg.net_lib_dir / "xdp_dummy.bpf.o" cmd(f"ip link set dev {cfg.ifname} mtu 1500", shell=True) try: cmd(f"ip link set dev {cfg.ifname} xdpoffload obj {prog} sec xdp", shell=True) -- cgit v1.2.3-59-g8ed1b From 42f342387841891bbbd15e25d33eb510a0cf7a9a Mon Sep 17 00:00:00 2001 From: Taehee Yoo Date: Fri, 28 Mar 2025 06:22:37 +0000 Subject: net: fix use-after-free in the netdev_nl_sock_priv_destroy() In the netdev_nl_sock_priv_destroy(), an instance lock is acquired before calling net_devmem_unbind_dmabuf(), then releasing an instance lock(netdev_unlock(binding->dev)). However, a binding is freed in the net_devmem_unbind_dmabuf(). So using a binding after net_devmem_unbind_dmabuf() occurs UAF. To fix this UAF, it needs to use temporary variable. Fixes: ba6f418fbf64 ("net: bubble up taking netdev instance lock to callers of net_devmem_unbind_dmabuf()") Signed-off-by: Taehee Yoo Reviewed-by: Jakub Kicinski Reviewed-by: Mina Almasry Reviewed-by: Xuan Zhuo Reviewed-by: Simon Horman Link: https://patch.msgid.link/20250328062237.3746875-1-ap420073@gmail.com Signed-off-by: Jakub Kicinski --- net/core/netdev-genl.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c index fd1cfa9707dc..3afeaa8c5dc5 100644 --- a/net/core/netdev-genl.c +++ b/net/core/netdev-genl.c @@ -951,12 +951,14 @@ void netdev_nl_sock_priv_destroy(struct netdev_nl_sock *priv) { struct net_devmem_dmabuf_binding *binding; struct net_devmem_dmabuf_binding *temp; + struct net_device *dev; mutex_lock(&priv->lock); list_for_each_entry_safe(binding, temp, &priv->bindings, list) { - netdev_lock(binding->dev); + dev = binding->dev; + netdev_lock(dev); net_devmem_unbind_dmabuf(binding); - netdev_unlock(binding->dev); + netdev_unlock(dev); } mutex_unlock(&priv->lock); } -- cgit v1.2.3-59-g8ed1b From 0fdba88a211508984eb5df62008c29688692b134 Mon Sep 17 00:00:00 2001 From: Geetha sowjanya Date: Thu, 27 Mar 2025 14:44:41 +0530 Subject: octeontx2-af: Fix mbox INTR handler when num VFs > 64 When number of RVU VFs > 64, the vfs value passed to "rvu_queue_work" function is incorrect. Due to which mbox workqueue entries for VFs 0 to 63 never gets added to workqueue. Fixes: 9bdc47a6e328 ("octeontx2-af: Mbox communication support btw AF and it's VFs") Signed-off-by: Geetha sowjanya Reviewed-by: Simon Horman Link: https://patch.msgid.link/20250327091441.1284-1-gakula@marvell.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/marvell/octeontx2/af/rvu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c index cd0d7b7774f1..6575c422635b 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c @@ -2634,7 +2634,7 @@ static irqreturn_t rvu_mbox_intr_handler(int irq, void *rvu_irq) rvupf_write64(rvu, RVU_PF_VFPF_MBOX_INTX(1), intr); rvu_queue_work(&rvu->afvf_wq_info, 64, vfs, intr); - vfs -= 64; + vfs = 64; } intr = rvupf_read64(rvu, RVU_PF_VFPF_MBOX_INTX(0)); -- cgit v1.2.3-59-g8ed1b From 323d6db6dc7decb06f2545efb9496259ddacd4f4 Mon Sep 17 00:00:00 2001 From: Geetha sowjanya Date: Thu, 27 Mar 2025 15:10:54 +0530 Subject: octeontx2-af: Free NIX_AF_INT_VEC_GEN irq Due to the incorrect initial vector number in rvu_nix_unregister_interrupts(), NIX_AF_INT_VEC_GEN is not geeting free. Fix the vector number to include NIX_AF_INT_VEC_GEN irq. Fixes: 5ed66306eab6 ("octeontx2-af: Add devlink health reporters for NIX") Signed-off-by: Geetha sowjanya Reviewed-by: Simon Horman Link: https://patch.msgid.link/20250327094054.2312-1-gakula@marvell.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c index dab4deca893f..27c3a2daaaa9 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c @@ -207,7 +207,7 @@ static void rvu_nix_unregister_interrupts(struct rvu *rvu) rvu->irq_allocated[offs + NIX_AF_INT_VEC_RVU] = false; } - for (i = NIX_AF_INT_VEC_AF_ERR; i < NIX_AF_INT_VEC_CNT; i++) + for (i = NIX_AF_INT_VEC_GEN; i < NIX_AF_INT_VEC_CNT; i++) if (rvu->irq_allocated[offs + i]) { free_irq(pci_irq_vector(rvu->pdev, offs + i), rvu_dl); rvu->irq_allocated[offs + i] = false; -- cgit v1.2.3-59-g8ed1b From 443041deb5ef6a1289a99ed95015ec7442f141dc Mon Sep 17 00:00:00 2001 From: Gang Yan Date: Fri, 28 Mar 2025 15:27:16 +0100 Subject: mptcp: fix NULL pointer in can_accept_new_subflow When testing valkey benchmark tool with MPTCP, the kernel panics in 'mptcp_can_accept_new_subflow' because subflow_req->msk is NULL. Call trace: mptcp_can_accept_new_subflow (./net/mptcp/subflow.c:63 (discriminator 4)) (P) subflow_syn_recv_sock (./net/mptcp/subflow.c:854) tcp_check_req (./net/ipv4/tcp_minisocks.c:863) tcp_v4_rcv (./net/ipv4/tcp_ipv4.c:2268) ip_protocol_deliver_rcu (./net/ipv4/ip_input.c:207) ip_local_deliver_finish (./net/ipv4/ip_input.c:234) ip_local_deliver (./net/ipv4/ip_input.c:254) ip_rcv_finish (./net/ipv4/ip_input.c:449) ... According to the debug log, the same req received two SYN-ACK in a very short time, very likely because the client retransmits the syn ack due to multiple reasons. Even if the packets are transmitted with a relevant time interval, they can be processed by the server on different CPUs concurrently). The 'subflow_req->msk' ownership is transferred to the subflow the first, and there will be a risk of a null pointer dereference here. This patch fixes this issue by moving the 'subflow_req->msk' under the `own_req == true` conditional. Note that the !msk check in subflow_hmac_valid() can be dropped, because the same check already exists under the own_req mpj branch where the code has been moved to. Fixes: 9466a1ccebbe ("mptcp: enable JOIN requests even if cookies are in use") Cc: stable@vger.kernel.org Suggested-by: Paolo Abeni Signed-off-by: Gang Yan Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20250328-net-mptcp-misc-fixes-6-15-v1-1-34161a482a7f@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/subflow.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index efe8d86496db..409bd415ef1d 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -754,8 +754,6 @@ static bool subflow_hmac_valid(const struct request_sock *req, subflow_req = mptcp_subflow_rsk(req); msk = subflow_req->msk; - if (!msk) - return false; subflow_generate_hmac(READ_ONCE(msk->remote_key), READ_ONCE(msk->local_key), @@ -850,12 +848,8 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk, } else if (subflow_req->mp_join) { mptcp_get_options(skb, &mp_opt); - if (!(mp_opt.suboptions & OPTION_MPTCP_MPJ_ACK) || - !subflow_hmac_valid(req, &mp_opt) || - !mptcp_can_accept_new_subflow(subflow_req->msk)) { - SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINACKMAC); + if (!(mp_opt.suboptions & OPTION_MPTCP_MPJ_ACK)) fallback = true; - } } create_child: @@ -905,6 +899,13 @@ create_child: goto dispose_child; } + if (!subflow_hmac_valid(req, &mp_opt) || + !mptcp_can_accept_new_subflow(subflow_req->msk)) { + SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINACKMAC); + subflow_add_reset_reason(skb, MPTCP_RST_EPROHIBIT); + goto dispose_child; + } + /* move the msk reference ownership to the subflow */ subflow_req->msk = NULL; ctx->conn = (struct sock *)owner; -- cgit v1.2.3-59-g8ed1b From 7335d4ac812917c16e04958775826d12d481c92d Mon Sep 17 00:00:00 2001 From: Cong Liu Date: Fri, 28 Mar 2025 15:27:17 +0100 Subject: selftests: mptcp: fix incorrect fd checks in main_loop Fix a bug where the code was checking the wrong file descriptors when opening the input files. The code was checking 'fd' instead of 'fd_in', which could lead to incorrect error handling. Fixes: 05be5e273c84 ("selftests: mptcp: add disconnect tests") Cc: stable@vger.kernel.org Fixes: ca7ae8916043 ("selftests: mptcp: mptfo Initiator/Listener") Co-developed-by: Geliang Tang Signed-off-by: Geliang Tang Signed-off-by: Cong Liu Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20250328-net-mptcp-misc-fixes-6-15-v1-2-34161a482a7f@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/mptcp/mptcp_connect.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.c b/tools/testing/selftests/net/mptcp/mptcp_connect.c index d240d02fa443..893dc36b12f6 100644 --- a/tools/testing/selftests/net/mptcp/mptcp_connect.c +++ b/tools/testing/selftests/net/mptcp/mptcp_connect.c @@ -1270,7 +1270,7 @@ int main_loop(void) if (cfg_input && cfg_sockopt_types.mptfo) { fd_in = open(cfg_input, O_RDONLY); - if (fd < 0) + if (fd_in < 0) xerror("can't open %s:%d", cfg_input, errno); } @@ -1293,7 +1293,7 @@ again: if (cfg_input && !cfg_sockopt_types.mptfo) { fd_in = open(cfg_input, O_RDONLY); - if (fd < 0) + if (fd_in < 0) xerror("can't open %s:%d", cfg_input, errno); } -- cgit v1.2.3-59-g8ed1b From c183165f87a486d5879f782c05a23c179c3794ab Mon Sep 17 00:00:00 2001 From: Geliang Tang Date: Fri, 28 Mar 2025 15:27:18 +0100 Subject: selftests: mptcp: close fd_in before returning in main_loop The file descriptor 'fd_in' is opened when cfg_input is configured, but not closed in main_loop(), this patch fixes it. Fixes: 05be5e273c84 ("selftests: mptcp: add disconnect tests") Cc: stable@vger.kernel.org Co-developed-by: Cong Liu Signed-off-by: Cong Liu Signed-off-by: Geliang Tang Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20250328-net-mptcp-misc-fixes-6-15-v1-3-34161a482a7f@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/mptcp/mptcp_connect.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.c b/tools/testing/selftests/net/mptcp/mptcp_connect.c index 893dc36b12f6..c83a8b47bbdf 100644 --- a/tools/testing/selftests/net/mptcp/mptcp_connect.c +++ b/tools/testing/selftests/net/mptcp/mptcp_connect.c @@ -1299,7 +1299,7 @@ again: ret = copyfd_io(fd_in, fd, 1, 0, &winfo); if (ret) - return ret; + goto out; if (cfg_truncate > 0) { shutdown(fd, SHUT_WR); @@ -1320,7 +1320,10 @@ again: close(fd); } - return 0; +out: + if (cfg_input) + close(fd_in); + return ret; } int parse_proto(const char *proto) -- cgit v1.2.3-59-g8ed1b From b44a4c28228fc50b0af05b5d15b44c2172f112a0 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Fri, 28 Mar 2025 15:27:19 +0100 Subject: selftests: mptcp: ignore mptcp_diag binary A new binary is now generated by the MPTCP selftests: mptcp_diag. Like the other binaries from this directory, there is no need to track this in Git, it should then be ignored. Fixes: 00f5e338cf7e ("selftests: mptcp: Add a tool to get specific msk_info") Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20250328-net-mptcp-misc-fixes-6-15-v1-4-34161a482a7f@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/mptcp/.gitignore | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/testing/selftests/net/mptcp/.gitignore b/tools/testing/selftests/net/mptcp/.gitignore index 49daae73c41e..833279fb34e2 100644 --- a/tools/testing/selftests/net/mptcp/.gitignore +++ b/tools/testing/selftests/net/mptcp/.gitignore @@ -1,5 +1,6 @@ # SPDX-License-Identifier: GPL-2.0-only mptcp_connect +mptcp_diag mptcp_inq mptcp_sockopt pm_nl_ctl -- cgit v1.2.3-59-g8ed1b From 9e3267cf02c240065fddfbe1a58cdb99d0b00531 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 28 Mar 2025 09:47:42 -0700 Subject: eth: gve: add missing netdev locks on reset and shutdown paths All the misc entry points end up calling into either gve_open() or gve_close(), they take rtnl_lock today but since the recent instance locking changes should also take the instance lock. Found by code inspection and untested. Fixes: cae03e5bdd9e ("net: hold netdev instance lock during queue operations") Acked-by: Stanislav Fomichev Reviewed-by: Harshitha Ramamurthy Link: https://patch.msgid.link/20250328164742.1268069-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/google/gve/gve_main.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c index cb2f9978f45e..f9a73c956861 100644 --- a/drivers/net/ethernet/google/gve/gve_main.c +++ b/drivers/net/ethernet/google/gve/gve_main.c @@ -2077,7 +2077,9 @@ static void gve_handle_reset(struct gve_priv *priv) if (gve_get_do_reset(priv)) { rtnl_lock(); + netdev_lock(priv->dev); gve_reset(priv, false); + netdev_unlock(priv->dev); rtnl_unlock(); } } @@ -2714,6 +2716,7 @@ static void gve_shutdown(struct pci_dev *pdev) bool was_up = netif_running(priv->dev); rtnl_lock(); + netdev_lock(netdev); if (was_up && gve_close(priv->dev)) { /* If the dev was up, attempt to close, if close fails, reset */ gve_reset_and_teardown(priv, was_up); @@ -2721,6 +2724,7 @@ static void gve_shutdown(struct pci_dev *pdev) /* If the dev wasn't up or close worked, finish tearing down */ gve_teardown_priv_resources(priv); } + netdev_unlock(netdev); rtnl_unlock(); } -- cgit v1.2.3-59-g8ed1b From dd07df9ff3d148aee87fcbab99ff14f0727752f4 Mon Sep 17 00:00:00 2001 From: Stanislav Fomichev Date: Fri, 28 Mar 2025 10:42:16 -0700 Subject: bnxt_en: bring back rtnl lock in bnxt_shutdown Taehee reports missing rtnl from bnxt_shutdown path: inetdev_event (./include/linux/inetdevice.h:256 net/ipv4/devinet.c:1585) notifier_call_chain (kernel/notifier.c:85) __dev_close_many (net/core/dev.c:1732 (discriminator 3)) kernel/locking/mutex.c:713 kernel/locking/mutex.c:732) dev_close_many (net/core/dev.c:1786) netif_close (./include/linux/list.h:124 ./include/linux/list.h:215 bnxt_shutdown (drivers/net/ethernet/broadcom/bnxt/bnxt.c:16707) bnxt_en pci_device_shutdown (drivers/pci/pci-driver.c:511) device_shutdown (drivers/base/core.c:4820) kernel_restart (kernel/reboot.c:271 kernel/reboot.c:285) Bring back the rtnl lock. Link: https://lore.kernel.org/netdev/CAMArcTV4P8PFsc6O2tSgzRno050DzafgqkLA2b7t=Fv_SY=brw@mail.gmail.com/ Fixes: 004b5008016a ("eth: bnxt: remove most dependencies on RTNL") Reported-by: Taehee Yoo Signed-off-by: Stanislav Fomichev Tested-by: Taehee Yoo Tested-by: Breno Leitao Link: https://patch.msgid.link/20250328174216.3513079-1-sdf@fomichev.me Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 934ba9425857..1a70605fad38 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -16698,6 +16698,7 @@ static void bnxt_shutdown(struct pci_dev *pdev) if (!dev) return; + rtnl_lock(); netdev_lock(dev); bp = netdev_priv(dev); if (!bp) @@ -16717,6 +16718,7 @@ static void bnxt_shutdown(struct pci_dev *pdev) shutdown_exit: netdev_unlock(dev); + rtnl_unlock(); } #ifdef CONFIG_PM_SLEEP -- cgit v1.2.3-59-g8ed1b From f278b6d5bb465c7fd66f3d103812947e55b376ed Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 31 Mar 2025 07:59:46 +0000 Subject: Revert "tcp: avoid atomic operations on sk->sk_rmem_alloc" This reverts commit 0de2a5c4b824da2205658ebebb99a55c43cdf60f. I forgot that a TCP socket could receive messages in its error queue. sock_queue_err_skb() can be called without socket lock being held, and changes sk->sk_rmem_alloc. The fact that skbs in error queue are limited by sk->sk_rcvbuf means that error messages can be dropped if socket receive queues are full, which is an orthogonal issue. In future kernels, we could use a separate sk->sk_error_mem_alloc counter specifically for the error queue. Fixes: 0de2a5c4b824 ("tcp: avoid atomic operations on sk->sk_rmem_alloc") Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20250331075946.31960-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/tcp.h | 15 --------------- net/ipv4/tcp.c | 18 ++---------------- net/ipv4/tcp_fastopen.c | 2 +- net/ipv4/tcp_input.c | 6 +++--- 4 files changed, 6 insertions(+), 35 deletions(-) diff --git a/include/net/tcp.h b/include/net/tcp.h index df04dc09c519..4450c384ef17 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -779,7 +779,6 @@ static inline int tcp_bound_to_half_wnd(struct tcp_sock *tp, int pktsize) /* tcp.c */ void tcp_get_info(struct sock *, struct tcp_info *); -void tcp_sock_rfree(struct sk_buff *skb); /* Read 'sendfile()'-style from a TCP socket */ int tcp_read_sock(struct sock *sk, read_descriptor_t *desc, @@ -2899,18 +2898,4 @@ enum skb_drop_reason tcp_inbound_hash(struct sock *sk, const void *saddr, const void *daddr, int family, int dif, int sdif); -/* version of skb_set_owner_r() avoiding one atomic_add() */ -static inline void tcp_skb_set_owner_r(struct sk_buff *skb, struct sock *sk) -{ - skb_orphan(skb); - skb->sk = sk; - skb->destructor = tcp_sock_rfree; - - sock_owned_by_me(sk); - atomic_set(&sk->sk_rmem_alloc, - atomic_read(&sk->sk_rmem_alloc) + skb->truesize); - - sk_forward_alloc_add(sk, -skb->truesize); -} - #endif /* _TCP_H */ diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index ea8de00f669d..6edc441b3702 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1525,25 +1525,11 @@ void tcp_cleanup_rbuf(struct sock *sk, int copied) __tcp_cleanup_rbuf(sk, copied); } -/* private version of sock_rfree() avoiding one atomic_sub() */ -void tcp_sock_rfree(struct sk_buff *skb) -{ - struct sock *sk = skb->sk; - unsigned int len = skb->truesize; - - sock_owned_by_me(sk); - atomic_set(&sk->sk_rmem_alloc, - atomic_read(&sk->sk_rmem_alloc) - len); - - sk_forward_alloc_add(sk, len); - sk_mem_reclaim(sk); -} - static void tcp_eat_recv_skb(struct sock *sk, struct sk_buff *skb) { __skb_unlink(skb, &sk->sk_receive_queue); - if (likely(skb->destructor == tcp_sock_rfree)) { - tcp_sock_rfree(skb); + if (likely(skb->destructor == sock_rfree)) { + sock_rfree(skb); skb->destructor = NULL; skb->sk = NULL; return skb_attempt_defer_free(skb); diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index ca40665145c6..1a6b1bc54245 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -189,7 +189,7 @@ void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb) tcp_segs_in(tp, skb); __skb_pull(skb, tcp_hdrlen(skb)); sk_forced_mem_schedule(sk, skb->truesize); - tcp_skb_set_owner_r(skb, sk); + skb_set_owner_r(skb, sk); TCP_SKB_CB(skb)->seq++; TCP_SKB_CB(skb)->tcp_flags &= ~TCPHDR_SYN; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index e1f952fbac48..a35018e2d0ba 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5171,7 +5171,7 @@ end: if (tcp_is_sack(tp)) tcp_grow_window(sk, skb, false); skb_condense(skb); - tcp_skb_set_owner_r(skb, sk); + skb_set_owner_r(skb, sk); } } @@ -5187,7 +5187,7 @@ static int __must_check tcp_queue_rcv(struct sock *sk, struct sk_buff *skb, tcp_rcv_nxt_update(tcp_sk(sk), TCP_SKB_CB(skb)->end_seq); if (!eaten) { tcp_add_receive_queue(sk, skb); - tcp_skb_set_owner_r(skb, sk); + skb_set_owner_r(skb, sk); } return eaten; } @@ -5504,7 +5504,7 @@ skip_this: __skb_queue_before(list, skb, nskb); else __skb_queue_tail(&tmp, nskb); /* defer rbtree insertion */ - tcp_skb_set_owner_r(nskb, sk); + skb_set_owner_r(nskb, sk); mptcp_skb_ext_move(nskb, skb); /* Copy data, releasing collapsed skbs. */ -- cgit v1.2.3-59-g8ed1b