From d3153c3b42707d26c81083b426f2ef0951bce545 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 17 Apr 2025 18:53:17 -0700 Subject: net: fix the missing unlock for detached devices The combined condition was left as is when we converted from __dev_get_by_index() to netdev_get_by_index_lock(). There was no need to undo anything with the former, for the latter we need an unlock. Fixes: 1d22d3060b9b ("net: drop rtnl_lock for queue_mgmt operations") Reviewed-by: Mina Almasry Link: https://patch.msgid.link/20250418015317.1954107-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/core/netdev-genl.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c index 5d7af50fe702..230743bdbb14 100644 --- a/net/core/netdev-genl.c +++ b/net/core/netdev-genl.c @@ -861,14 +861,17 @@ int netdev_nl_bind_rx_doit(struct sk_buff *skb, struct genl_info *info) mutex_lock(&priv->lock); + err = 0; netdev = netdev_get_by_index_lock(genl_info_net(info), ifindex); - if (!netdev || !netif_device_present(netdev)) { + if (!netdev) { err = -ENODEV; goto err_unlock_sock; } - - if (!netdev_need_ops_lock(netdev)) { + if (!netif_device_present(netdev)) + err = -ENODEV; + else if (!netdev_need_ops_lock(netdev)) err = -EOPNOTSUPP; + if (err) { NL_SET_BAD_ATTR(info->extack, info->attrs[NETDEV_A_DEV_IFINDEX]); goto err_unlock; -- cgit v1.2.3 From 9e8d1013b0c38910cbc9e60de74dbe883878469d Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Wed, 16 Apr 2025 18:01:25 +0200 Subject: net: selftests: initialize TCP header and skb payload with zero Zero-initialize TCP header via memset() to avoid garbage values that may affect checksum or behavior during test transmission. Also zero-fill allocated payload and padding regions using memset() after skb_put(), ensuring deterministic content for all outgoing test packets. Fixes: 3e1e58d64c3d ("net: add generic selftest support") Signed-off-by: Oleksij Rempel Cc: stable@vger.kernel.org Reviewed-by: Simon Horman Link: https://patch.msgid.link/20250416160125.2914724-1-o.rempel@pengutronix.de Signed-off-by: Paolo Abeni --- net/core/selftests.c | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/core/selftests.c b/net/core/selftests.c index e99ae983fca9..35f807ea9952 100644 --- a/net/core/selftests.c +++ b/net/core/selftests.c @@ -100,10 +100,10 @@ static struct sk_buff *net_test_get_skb(struct net_device *ndev, ehdr->h_proto = htons(ETH_P_IP); if (attr->tcp) { + memset(thdr, 0, sizeof(*thdr)); thdr->source = htons(attr->sport); thdr->dest = htons(attr->dport); thdr->doff = sizeof(struct tcphdr) / 4; - thdr->check = 0; } else { uhdr->source = htons(attr->sport); uhdr->dest = htons(attr->dport); @@ -144,10 +144,18 @@ static struct sk_buff *net_test_get_skb(struct net_device *ndev, attr->id = net_test_next_id; shdr->id = net_test_next_id++; - if (attr->size) - skb_put(skb, attr->size); - if (attr->max_size && attr->max_size > skb->len) - skb_put(skb, attr->max_size - skb->len); + if (attr->size) { + void *payload = skb_put(skb, attr->size); + + memset(payload, 0, attr->size); + } + + if (attr->max_size && attr->max_size > skb->len) { + size_t pad_len = attr->max_size - skb->len; + void *pad = skb_put(skb, pad_len); + + memset(pad, 0, pad_len); + } skb->csum = 0; skb->ip_summed = CHECKSUM_PARTIAL; -- cgit v1.2.3 From c03a49f3093a4903c8a93c8b5c9a297b5343b169 Mon Sep 17 00:00:00 2001 From: Justin Iurman Date: Wed, 16 Apr 2025 18:07:16 +0200 Subject: net: lwtunnel: disable BHs when required In lwtunnel_{output|xmit}(), dev_xmit_recursion() may be called in preemptible scope for PREEMPT kernels. This patch disables BHs before calling dev_xmit_recursion(). BHs are re-enabled only at the end, since we must ensure the same CPU is used for both dev_xmit_recursion_inc() and dev_xmit_recursion_dec() (and any other recursion levels in some cases) in order to maintain valid per-cpu counters. Reported-by: Alexei Starovoitov Closes: https://lore.kernel.org/netdev/CAADnVQJFWn3dBFJtY+ci6oN1pDFL=TzCmNbRgey7MdYxt_AP2g@mail.gmail.com/ Reported-by: Eduard Zingerman Closes: https://lore.kernel.org/netdev/m2h62qwf34.fsf@gmail.com/ Fixes: 986ffb3a57c5 ("net: lwtunnel: fix recursion loops") Signed-off-by: Justin Iurman Reviewed-by: Simon Horman Link: https://patch.msgid.link/20250416160716.8823-1-justin.iurman@uliege.be Signed-off-by: Paolo Abeni --- net/core/lwtunnel.c | 26 ++++++++++++++++++++------ 1 file changed, 20 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/core/lwtunnel.c b/net/core/lwtunnel.c index e39a459540ec..60f27cb4e54f 100644 --- a/net/core/lwtunnel.c +++ b/net/core/lwtunnel.c @@ -333,6 +333,8 @@ int lwtunnel_output(struct net *net, struct sock *sk, struct sk_buff *skb) struct dst_entry *dst; int ret; + local_bh_disable(); + if (dev_xmit_recursion()) { net_crit_ratelimited("%s(): recursion limit reached on datapath\n", __func__); @@ -348,8 +350,10 @@ int lwtunnel_output(struct net *net, struct sock *sk, struct sk_buff *skb) lwtstate = dst->lwtstate; if (lwtstate->type == LWTUNNEL_ENCAP_NONE || - lwtstate->type > LWTUNNEL_ENCAP_MAX) - return 0; + lwtstate->type > LWTUNNEL_ENCAP_MAX) { + ret = 0; + goto out; + } ret = -EOPNOTSUPP; rcu_read_lock(); @@ -364,11 +368,13 @@ int lwtunnel_output(struct net *net, struct sock *sk, struct sk_buff *skb) if (ret == -EOPNOTSUPP) goto drop; - return ret; + goto out; drop: kfree_skb(skb); +out: + local_bh_enable(); return ret; } EXPORT_SYMBOL_GPL(lwtunnel_output); @@ -380,6 +386,8 @@ int lwtunnel_xmit(struct sk_buff *skb) struct dst_entry *dst; int ret; + local_bh_disable(); + if (dev_xmit_recursion()) { net_crit_ratelimited("%s(): recursion limit reached on datapath\n", __func__); @@ -396,8 +404,10 @@ int lwtunnel_xmit(struct sk_buff *skb) lwtstate = dst->lwtstate; if (lwtstate->type == LWTUNNEL_ENCAP_NONE || - lwtstate->type > LWTUNNEL_ENCAP_MAX) - return 0; + lwtstate->type > LWTUNNEL_ENCAP_MAX) { + ret = 0; + goto out; + } ret = -EOPNOTSUPP; rcu_read_lock(); @@ -412,11 +422,13 @@ int lwtunnel_xmit(struct sk_buff *skb) if (ret == -EOPNOTSUPP) goto drop; - return ret; + goto out; drop: kfree_skb(skb); +out: + local_bh_enable(); return ret; } EXPORT_SYMBOL_GPL(lwtunnel_xmit); @@ -428,6 +440,8 @@ int lwtunnel_input(struct sk_buff *skb) struct dst_entry *dst; int ret; + DEBUG_NET_WARN_ON_ONCE(!in_softirq()); + if (dev_xmit_recursion()) { net_crit_ratelimited("%s(): recursion limit reached on datapath\n", __func__); -- cgit v1.2.3 From d63527e109e811ef11abb1c2985048fdb528b4cb Mon Sep 17 00:00:00 2001 From: Tung Nguyen Date: Thu, 17 Apr 2025 14:47:15 +0700 Subject: tipc: fix NULL pointer dereference in tipc_mon_reinit_self() syzbot reported: tipc: Node number set to 1055423674 Oops: general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] SMP KASAN NOPTI KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007] CPU: 3 UID: 0 PID: 6017 Comm: kworker/3:5 Not tainted 6.15.0-rc1-syzkaller-00246-g900241a5cc15 #0 PREEMPT(full) Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014 Workqueue: events tipc_net_finalize_work RIP: 0010:tipc_mon_reinit_self+0x11c/0x210 net/tipc/monitor.c:719 ... RSP: 0018:ffffc9000356fb68 EFLAGS: 00010246 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 000000003ee87cba RDX: 0000000000000000 RSI: ffffffff8dbc56a7 RDI: ffff88804c2cc010 RBP: dffffc0000000000 R08: 0000000000000001 R09: 0000000000000000 R10: 0000000000000001 R11: 0000000000000000 R12: 0000000000000007 R13: fffffbfff2111097 R14: ffff88804ead8000 R15: ffff88804ead9010 FS: 0000000000000000(0000) GS:ffff888097ab9000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00000000f720eb00 CR3: 000000000e182000 CR4: 0000000000352ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: tipc_net_finalize+0x10b/0x180 net/tipc/net.c:140 process_one_work+0x9cc/0x1b70 kernel/workqueue.c:3238 process_scheduled_works kernel/workqueue.c:3319 [inline] worker_thread+0x6c8/0xf10 kernel/workqueue.c:3400 kthread+0x3c2/0x780 kernel/kthread.c:464 ret_from_fork+0x45/0x80 arch/x86/kernel/process.c:153 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245 ... RIP: 0010:tipc_mon_reinit_self+0x11c/0x210 net/tipc/monitor.c:719 ... RSP: 0018:ffffc9000356fb68 EFLAGS: 00010246 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 000000003ee87cba RDX: 0000000000000000 RSI: ffffffff8dbc56a7 RDI: ffff88804c2cc010 RBP: dffffc0000000000 R08: 0000000000000001 R09: 0000000000000000 R10: 0000000000000001 R11: 0000000000000000 R12: 0000000000000007 R13: fffffbfff2111097 R14: ffff88804ead8000 R15: ffff88804ead9010 FS: 0000000000000000(0000) GS:ffff888097ab9000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00000000f720eb00 CR3: 000000000e182000 CR4: 0000000000352ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 There is a racing condition between workqueue created when enabling bearer and another thread created when disabling bearer right after that as follow: enabling_bearer | disabling_bearer --------------- | ---------------- tipc_disc_timeout() | { | bearer_disable() ... | { schedule_work(&tn->work); | tipc_mon_delete() ... | { } | ... | write_lock_bh(&mon->lock); | mon->self = NULL; | write_unlock_bh(&mon->lock); | ... | } tipc_net_finalize_work() | } { | ... | tipc_net_finalize() | { | ... | tipc_mon_reinit_self() | { | ... | write_lock_bh(&mon->lock); | mon->self->addr = tipc_own_addr(net); | write_unlock_bh(&mon->lock); | ... | } | ... | } | ... | } | 'mon->self' is set to NULL in disabling_bearer thread and dereferenced later in enabling_bearer thread. This commit fixes this issue by validating 'mon->self' before assigning node address to it. Reported-by: syzbot+ed60da8d686dc709164c@syzkaller.appspotmail.com Fixes: 46cb01eeeb86 ("tipc: update mon's self addr when node addr generated") Signed-off-by: Tung Nguyen Reviewed-by: Simon Horman Link: https://patch.msgid.link/20250417074826.578115-1-tung.quang.nguyen@est.tech Signed-off-by: Jakub Kicinski --- net/tipc/monitor.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/tipc/monitor.c b/net/tipc/monitor.c index e2f19627e43d..b45c5b91bc7a 100644 --- a/net/tipc/monitor.c +++ b/net/tipc/monitor.c @@ -716,7 +716,8 @@ void tipc_mon_reinit_self(struct net *net) if (!mon) continue; write_lock_bh(&mon->lock); - mon->self->addr = tipc_own_addr(net); + if (mon->self) + mon->self->addr = tipc_own_addr(net); write_unlock_bh(&mon->lock); } } -- cgit v1.2.3 From abf078c0a322159f5ebe2adaa0cd69dc45b1e710 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 22 Apr 2025 21:32:51 +0200 Subject: wifi: mac80211: restore monitor for outgoing frames This code was accidentally dropped during the cooked monitor removal, but really should've been simplified instead. Add the simple version back. Fixes: 286e69677065 ("wifi: mac80211: Drop cooked monitor support") Link: https://patch.msgid.link/20250422213251.b3d65fd0f323.Id2a6901583f7af86bbe94deb355968b238f350c6@changeid Signed-off-by: Johannes Berg --- net/mac80211/status.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/status.c b/net/mac80211/status.c index b17b3cc7fb90..a362254b310c 100644 --- a/net/mac80211/status.c +++ b/net/mac80211/status.c @@ -1085,7 +1085,13 @@ static void __ieee80211_tx_status(struct ieee80211_hw *hw, ieee80211_report_used_skb(local, skb, false, status->ack_hwtstamp); - if (status->free_list) + /* + * This is a bit racy but we can avoid a lot of work + * with this test... + */ + if (local->tx_mntrs) + ieee80211_tx_monitor(local, skb, retry_count, status); + else if (status->free_list) list_add_tail(&skb->list, status->free_list); else dev_kfree_skb(skb); -- cgit v1.2.3 From 13b4ece33cf9def67966bb8716783c42cec20617 Mon Sep 17 00:00:00 2001 From: Mat Martineau Date: Mon, 21 Apr 2025 19:07:13 +0200 Subject: mptcp: pm: Defer freeing of MPTCP userspace path manager entries When path manager entries are deleted from the local address list, they are first unlinked from the address list using list_del_rcu(). The entries must not be freed until after the RCU grace period, but the existing code immediately frees the entry. Use kfree_rcu_mightsleep() and adjust sk_omem_alloc in open code instead of using the sock_kfree_s() helper. This code path is only called in a netlink handler, so the "might sleep" function is preferable to adding a rarely-used rcu_head member to struct mptcp_pm_addr_entry. Fixes: 88d097316371 ("mptcp: drop free_list for deleting entries") Cc: stable@vger.kernel.org Signed-off-by: Mat Martineau Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20250421-net-mptcp-pm-defer-freeing-v1-1-e731dc6e86b9@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm_userspace.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c index 2cb62f026b1f..a715dcbe0146 100644 --- a/net/mptcp/pm_userspace.c +++ b/net/mptcp/pm_userspace.c @@ -337,7 +337,11 @@ int mptcp_pm_nl_remove_doit(struct sk_buff *skb, struct genl_info *info) release_sock(sk); - sock_kfree_s(sk, match, sizeof(*match)); + kfree_rcu_mightsleep(match); + /* Adjust sk_omem_alloc like sock_kfree_s() does, to match + * with allocation of this memory by sock_kmemdup() + */ + atomic_sub(sizeof(*match), &sk->sk_omem_alloc); err = 0; out: -- cgit v1.2.3 From 3df275ef0a6ae181e8428a6589ef5d5231e58b5c Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Thu, 17 Apr 2025 11:47:30 -0700 Subject: net_sched: hfsc: Fix a UAF vulnerability in class handling This patch fixes a Use-After-Free vulnerability in the HFSC qdisc class handling. The issue occurs due to a time-of-check/time-of-use condition in hfsc_change_class() when working with certain child qdiscs like netem or codel. The vulnerability works as follows: 1. hfsc_change_class() checks if a class has packets (q.qlen != 0) 2. It then calls qdisc_peek_len(), which for certain qdiscs (e.g., codel, netem) might drop packets and empty the queue 3. The code continues assuming the queue is still non-empty, adding the class to vttree 4. This breaks HFSC scheduler assumptions that only non-empty classes are in vttree 5. Later, when the class is destroyed, this can lead to a Use-After-Free The fix adds a second queue length check after qdisc_peek_len() to verify the queue wasn't emptied. Fixes: 21f4d5cc25ec ("net_sched/hfsc: fix curve activation in hfsc_change_class()") Reported-by: Gerrard Tai Reviewed-by: Konstantin Khlebnikov Signed-off-by: Cong Wang Reviewed-by: Jamal Hadi Salim Link: https://patch.msgid.link/20250417184732.943057-2-xiyou.wangcong@gmail.com Signed-off-by: Jakub Kicinski --- net/sched/sch_hfsc.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index ce5045eea065..b368ac0595d5 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -961,6 +961,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid, if (cl != NULL) { int old_flags; + int len = 0; if (parentid) { if (cl->cl_parent && @@ -991,9 +992,13 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid, if (usc != NULL) hfsc_change_usc(cl, usc, cur_time); + if (cl->qdisc->q.qlen != 0) + len = qdisc_peek_len(cl->qdisc); + /* Check queue length again since some qdisc implementations + * (e.g., netem/codel) might empty the queue during the peek + * operation. + */ if (cl->qdisc->q.qlen != 0) { - int len = qdisc_peek_len(cl->qdisc); - if (cl->cl_flags & HFSC_RSC) { if (old_flags & HFSC_RSC) update_ed(cl, len); -- cgit v1.2.3 From 6ccbda44e2cc3d26fd22af54c650d6d5d801addf Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Thu, 17 Apr 2025 11:47:31 -0700 Subject: net_sched: hfsc: Fix a potential UAF in hfsc_dequeue() too Similarly to the previous patch, we need to safe guard hfsc_dequeue() too. But for this one, we don't have a reliable reproducer. Fixes: 1da177e4c3f41524e886b7f1b8a0c1fc7321cac2 ("Linux-2.6.12-rc2") Reported-by: Gerrard Tai Signed-off-by: Cong Wang Reviewed-by: Jamal Hadi Salim Link: https://patch.msgid.link/20250417184732.943057-3-xiyou.wangcong@gmail.com Signed-off-by: Jakub Kicinski --- net/sched/sch_hfsc.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index b368ac0595d5..6c8ef826cec0 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -1641,10 +1641,16 @@ hfsc_dequeue(struct Qdisc *sch) if (cl->qdisc->q.qlen != 0) { /* update ed */ next_len = qdisc_peek_len(cl->qdisc); - if (realtime) - update_ed(cl, next_len); - else - update_d(cl, next_len); + /* Check queue length again since some qdisc implementations + * (e.g., netem/codel) might empty the queue during the peek + * operation. + */ + if (cl->qdisc->q.qlen != 0) { + if (realtime) + update_ed(cl, next_len); + else + update_d(cl, next_len); + } } else { /* the class becomes passive */ eltree_remove(cl); -- cgit v1.2.3