From 736ef37fd9a44f5966e25319d08ff7ea99ac79e8 Mon Sep 17 00:00:00 2001 From: Coco Li Date: Thu, 23 Dec 2021 22:24:40 +0000 Subject: udp: using datalen to cap ipv6 udp max gso segments The max number of UDP gso segments is intended to cap to UDP_MAX_SEGMENTS, this is checked in udp_send_skb(). skb->len contains network and transport header len here, we should use only data len instead. This is the ipv6 counterpart to the below referenced commit, which missed the ipv6 change Fixes: 158390e45612 ("udp: using datalen to cap max gso segments") Signed-off-by: Coco Li Reviewed-by: Willem de Bruijn Link: https://lore.kernel.org/r/20211223222441.2975883-1-lixiaoyan@google.com Signed-off-by: Jakub Kicinski --- net/ipv6/udp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index a2caca6ccf11..8cde9efd7919 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -1204,7 +1204,7 @@ static int udp_v6_send_skb(struct sk_buff *skb, struct flowi6 *fl6, kfree_skb(skb); return -EINVAL; } - if (skb->len > cork->gso_size * UDP_MAX_SEGMENTS) { + if (datalen > cork->gso_size * UDP_MAX_SEGMENTS) { kfree_skb(skb); return -EINVAL; } -- cgit v1.2.3 From 5ec7d18d1813a5bead0b495045606c93873aecbb Mon Sep 17 00:00:00 2001 From: Xin Long Date: Thu, 23 Dec 2021 13:04:30 -0500 Subject: sctp: use call_rcu to free endpoint This patch is to delay the endpoint free by calling call_rcu() to fix another use-after-free issue in sctp_sock_dump(): BUG: KASAN: use-after-free in __lock_acquire+0x36d9/0x4c20 Call Trace: __lock_acquire+0x36d9/0x4c20 kernel/locking/lockdep.c:3218 lock_acquire+0x1ed/0x520 kernel/locking/lockdep.c:3844 __raw_spin_lock_bh include/linux/spinlock_api_smp.h:135 [inline] _raw_spin_lock_bh+0x31/0x40 kernel/locking/spinlock.c:168 spin_lock_bh include/linux/spinlock.h:334 [inline] __lock_sock+0x203/0x350 net/core/sock.c:2253 lock_sock_nested+0xfe/0x120 net/core/sock.c:2774 lock_sock include/net/sock.h:1492 [inline] sctp_sock_dump+0x122/0xb20 net/sctp/diag.c:324 sctp_for_each_transport+0x2b5/0x370 net/sctp/socket.c:5091 sctp_diag_dump+0x3ac/0x660 net/sctp/diag.c:527 __inet_diag_dump+0xa8/0x140 net/ipv4/inet_diag.c:1049 inet_diag_dump+0x9b/0x110 net/ipv4/inet_diag.c:1065 netlink_dump+0x606/0x1080 net/netlink/af_netlink.c:2244 __netlink_dump_start+0x59a/0x7c0 net/netlink/af_netlink.c:2352 netlink_dump_start include/linux/netlink.h:216 [inline] inet_diag_handler_cmd+0x2ce/0x3f0 net/ipv4/inet_diag.c:1170 __sock_diag_cmd net/core/sock_diag.c:232 [inline] sock_diag_rcv_msg+0x31d/0x410 net/core/sock_diag.c:263 netlink_rcv_skb+0x172/0x440 net/netlink/af_netlink.c:2477 sock_diag_rcv+0x2a/0x40 net/core/sock_diag.c:274 This issue occurs when asoc is peeled off and the old sk is freed after getting it by asoc->base.sk and before calling lock_sock(sk). To prevent the sk free, as a holder of the sk, ep should be alive when calling lock_sock(). This patch uses call_rcu() and moves sock_put and ep free into sctp_endpoint_destroy_rcu(), so that it's safe to try to hold the ep under rcu_read_lock in sctp_transport_traverse_process(). If sctp_endpoint_hold() returns true, it means this ep is still alive and we have held it and can continue to dump it; If it returns false, it means this ep is dead and can be freed after rcu_read_unlock, and we should skip it. In sctp_sock_dump(), after locking the sk, if this ep is different from tsp->asoc->ep, it means during this dumping, this asoc was peeled off before calling lock_sock(), and the sk should be skipped; If this ep is the same with tsp->asoc->ep, it means no peeloff happens on this asoc, and due to lock_sock, no peeloff will happen either until release_sock. Note that delaying endpoint free won't delay the port release, as the port release happens in sctp_endpoint_destroy() before calling call_rcu(). Also, freeing endpoint by call_rcu() makes it safe to access the sk by asoc->base.sk in sctp_assocs_seq_show() and sctp_rcv(). Thanks Jones to bring this issue up. v1->v2: - improve the changelog. - add kfree(ep) into sctp_endpoint_destroy_rcu(), as Jakub noticed. Reported-by: syzbot+9276d76e83e3bcde6c99@syzkaller.appspotmail.com Reported-by: Lee Jones Fixes: d25adbeb0cdb ("sctp: fix an use-after-free issue in sctp_sock_dump") Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/net/sctp/sctp.h | 6 +++--- include/net/sctp/structs.h | 3 ++- net/sctp/diag.c | 12 ++++++------ net/sctp/endpointola.c | 23 +++++++++++++++-------- net/sctp/socket.c | 23 +++++++++++++++-------- 5 files changed, 41 insertions(+), 26 deletions(-) (limited to 'net') diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index 189fdb9db162..d314a180ab93 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -105,6 +105,7 @@ extern struct percpu_counter sctp_sockets_allocated; int sctp_asconf_mgmt(struct sctp_sock *, struct sctp_sockaddr_entry *); struct sk_buff *sctp_skb_recv_datagram(struct sock *, int, int, int *); +typedef int (*sctp_callback_t)(struct sctp_endpoint *, struct sctp_transport *, void *); void sctp_transport_walk_start(struct rhashtable_iter *iter); void sctp_transport_walk_stop(struct rhashtable_iter *iter); struct sctp_transport *sctp_transport_get_next(struct net *net, @@ -115,9 +116,8 @@ int sctp_transport_lookup_process(int (*cb)(struct sctp_transport *, void *), struct net *net, const union sctp_addr *laddr, const union sctp_addr *paddr, void *p); -int sctp_for_each_transport(int (*cb)(struct sctp_transport *, void *), - int (*cb_done)(struct sctp_transport *, void *), - struct net *net, int *pos, void *p); +int sctp_transport_traverse_process(sctp_callback_t cb, sctp_callback_t cb_done, + struct net *net, int *pos, void *p); int sctp_for_each_endpoint(int (*cb)(struct sctp_endpoint *, void *), void *p); int sctp_get_sctp_info(struct sock *sk, struct sctp_association *asoc, struct sctp_info *info); diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 899c29c326ba..8dabd8800006 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1355,6 +1355,7 @@ struct sctp_endpoint { reconf_enable:1; __u8 strreset_enable; + struct rcu_head rcu; }; /* Recover the outter endpoint structure. */ @@ -1370,7 +1371,7 @@ static inline struct sctp_endpoint *sctp_ep(struct sctp_ep_common *base) struct sctp_endpoint *sctp_endpoint_new(struct sock *, gfp_t); void sctp_endpoint_free(struct sctp_endpoint *); void sctp_endpoint_put(struct sctp_endpoint *); -void sctp_endpoint_hold(struct sctp_endpoint *); +int sctp_endpoint_hold(struct sctp_endpoint *ep); void sctp_endpoint_add_asoc(struct sctp_endpoint *, struct sctp_association *); struct sctp_association *sctp_endpoint_lookup_assoc( const struct sctp_endpoint *ep, diff --git a/net/sctp/diag.c b/net/sctp/diag.c index 760b367644c1..a7d623171501 100644 --- a/net/sctp/diag.c +++ b/net/sctp/diag.c @@ -290,9 +290,8 @@ out: return err; } -static int sctp_sock_dump(struct sctp_transport *tsp, void *p) +static int sctp_sock_dump(struct sctp_endpoint *ep, struct sctp_transport *tsp, void *p) { - struct sctp_endpoint *ep = tsp->asoc->ep; struct sctp_comm_param *commp = p; struct sock *sk = ep->base.sk; struct sk_buff *skb = commp->skb; @@ -302,6 +301,8 @@ static int sctp_sock_dump(struct sctp_transport *tsp, void *p) int err = 0; lock_sock(sk); + if (ep != tsp->asoc->ep) + goto release; list_for_each_entry(assoc, &ep->asocs, asocs) { if (cb->args[4] < cb->args[1]) goto next; @@ -344,9 +345,8 @@ release: return err; } -static int sctp_sock_filter(struct sctp_transport *tsp, void *p) +static int sctp_sock_filter(struct sctp_endpoint *ep, struct sctp_transport *tsp, void *p) { - struct sctp_endpoint *ep = tsp->asoc->ep; struct sctp_comm_param *commp = p; struct sock *sk = ep->base.sk; const struct inet_diag_req_v2 *r = commp->r; @@ -505,8 +505,8 @@ skip: if (!(idiag_states & ~(TCPF_LISTEN | TCPF_CLOSE))) goto done; - sctp_for_each_transport(sctp_sock_filter, sctp_sock_dump, - net, &pos, &commp); + sctp_transport_traverse_process(sctp_sock_filter, sctp_sock_dump, + net, &pos, &commp); cb->args[2] = pos; done: diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c index 48c9c2c7602f..efffde7f2328 100644 --- a/net/sctp/endpointola.c +++ b/net/sctp/endpointola.c @@ -184,6 +184,18 @@ void sctp_endpoint_free(struct sctp_endpoint *ep) } /* Final destructor for endpoint. */ +static void sctp_endpoint_destroy_rcu(struct rcu_head *head) +{ + struct sctp_endpoint *ep = container_of(head, struct sctp_endpoint, rcu); + struct sock *sk = ep->base.sk; + + sctp_sk(sk)->ep = NULL; + sock_put(sk); + + kfree(ep); + SCTP_DBG_OBJCNT_DEC(ep); +} + static void sctp_endpoint_destroy(struct sctp_endpoint *ep) { struct sock *sk; @@ -213,18 +225,13 @@ static void sctp_endpoint_destroy(struct sctp_endpoint *ep) if (sctp_sk(sk)->bind_hash) sctp_put_port(sk); - sctp_sk(sk)->ep = NULL; - /* Give up our hold on the sock */ - sock_put(sk); - - kfree(ep); - SCTP_DBG_OBJCNT_DEC(ep); + call_rcu(&ep->rcu, sctp_endpoint_destroy_rcu); } /* Hold a reference to an endpoint. */ -void sctp_endpoint_hold(struct sctp_endpoint *ep) +int sctp_endpoint_hold(struct sctp_endpoint *ep) { - refcount_inc(&ep->base.refcnt); + return refcount_inc_not_zero(&ep->base.refcnt); } /* Release a reference to an endpoint and clean up if there are diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 33391254fa82..ad5028a07b18 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -5338,11 +5338,12 @@ int sctp_transport_lookup_process(int (*cb)(struct sctp_transport *, void *), } EXPORT_SYMBOL_GPL(sctp_transport_lookup_process); -int sctp_for_each_transport(int (*cb)(struct sctp_transport *, void *), - int (*cb_done)(struct sctp_transport *, void *), - struct net *net, int *pos, void *p) { +int sctp_transport_traverse_process(sctp_callback_t cb, sctp_callback_t cb_done, + struct net *net, int *pos, void *p) +{ struct rhashtable_iter hti; struct sctp_transport *tsp; + struct sctp_endpoint *ep; int ret; again: @@ -5351,26 +5352,32 @@ again: tsp = sctp_transport_get_idx(net, &hti, *pos + 1); for (; !IS_ERR_OR_NULL(tsp); tsp = sctp_transport_get_next(net, &hti)) { - ret = cb(tsp, p); - if (ret) - break; + ep = tsp->asoc->ep; + if (sctp_endpoint_hold(ep)) { /* asoc can be peeled off */ + ret = cb(ep, tsp, p); + if (ret) + break; + sctp_endpoint_put(ep); + } (*pos)++; sctp_transport_put(tsp); } sctp_transport_walk_stop(&hti); if (ret) { - if (cb_done && !cb_done(tsp, p)) { + if (cb_done && !cb_done(ep, tsp, p)) { (*pos)++; + sctp_endpoint_put(ep); sctp_transport_put(tsp); goto again; } + sctp_endpoint_put(ep); sctp_transport_put(tsp); } return ret; } -EXPORT_SYMBOL_GPL(sctp_for_each_transport); +EXPORT_SYMBOL_GPL(sctp_transport_traverse_process); /* 7.2.1 Association Status (SCTP_STATUS) -- cgit v1.2.3 From c1833c3964d5bd8c163bd4e01736a38bc473cb8a Mon Sep 17 00:00:00 2001 From: William Zhao Date: Thu, 23 Dec 2021 12:33:16 -0500 Subject: ip6_vti: initialize __ip6_tnl_parm struct in vti6_siocdevprivate The "__ip6_tnl_parm" struct was left uninitialized causing an invalid load of random data when the "__ip6_tnl_parm" struct was used elsewhere. As an example, in the function "ip6_tnl_xmit_ctl()", it tries to access the "collect_md" member. With "__ip6_tnl_parm" being uninitialized and containing random data, the UBSAN detected that "collect_md" held a non-boolean value. The UBSAN issue is as follows: =============================================================== UBSAN: invalid-load in net/ipv6/ip6_tunnel.c:1025:14 load of value 30 is not a valid value for type '_Bool' CPU: 1 PID: 228 Comm: kworker/1:3 Not tainted 5.16.0-rc4+ #8 Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011 Workqueue: ipv6_addrconf addrconf_dad_work Call Trace: dump_stack_lvl+0x44/0x57 ubsan_epilogue+0x5/0x40 __ubsan_handle_load_invalid_value+0x66/0x70 ? __cpuhp_setup_state+0x1d3/0x210 ip6_tnl_xmit_ctl.cold.52+0x2c/0x6f [ip6_tunnel] vti6_tnl_xmit+0x79c/0x1e96 [ip6_vti] ? lock_is_held_type+0xd9/0x130 ? vti6_rcv+0x100/0x100 [ip6_vti] ? lock_is_held_type+0xd9/0x130 ? rcu_read_lock_bh_held+0xc0/0xc0 ? lock_acquired+0x262/0xb10 dev_hard_start_xmit+0x1e6/0x820 __dev_queue_xmit+0x2079/0x3340 ? mark_lock.part.52+0xf7/0x1050 ? netdev_core_pick_tx+0x290/0x290 ? kvm_clock_read+0x14/0x30 ? kvm_sched_clock_read+0x5/0x10 ? sched_clock_cpu+0x15/0x200 ? find_held_lock+0x3a/0x1c0 ? lock_release+0x42f/0xc90 ? lock_downgrade+0x6b0/0x6b0 ? mark_held_locks+0xb7/0x120 ? neigh_connected_output+0x31f/0x470 ? lockdep_hardirqs_on+0x79/0x100 ? neigh_connected_output+0x31f/0x470 ? ip6_finish_output2+0x9b0/0x1d90 ? rcu_read_lock_bh_held+0x62/0xc0 ? ip6_finish_output2+0x9b0/0x1d90 ip6_finish_output2+0x9b0/0x1d90 ? ip6_append_data+0x330/0x330 ? ip6_mtu+0x166/0x370 ? __ip6_finish_output+0x1ad/0xfb0 ? nf_hook_slow+0xa6/0x170 ip6_output+0x1fb/0x710 ? nf_hook.constprop.32+0x317/0x430 ? ip6_finish_output+0x180/0x180 ? __ip6_finish_output+0xfb0/0xfb0 ? lock_is_held_type+0xd9/0x130 ndisc_send_skb+0xb33/0x1590 ? __sk_mem_raise_allocated+0x11cf/0x1560 ? dst_output+0x4a0/0x4a0 ? ndisc_send_rs+0x432/0x610 addrconf_dad_completed+0x30c/0xbb0 ? addrconf_rs_timer+0x650/0x650 ? addrconf_dad_work+0x73c/0x10e0 addrconf_dad_work+0x73c/0x10e0 ? addrconf_dad_completed+0xbb0/0xbb0 ? rcu_read_lock_sched_held+0xaf/0xe0 ? rcu_read_lock_bh_held+0xc0/0xc0 process_one_work+0x97b/0x1740 ? pwq_dec_nr_in_flight+0x270/0x270 worker_thread+0x87/0xbf0 ? process_one_work+0x1740/0x1740 kthread+0x3ac/0x490 ? set_kthread_struct+0x100/0x100 ret_from_fork+0x22/0x30 =============================================================== The solution is to initialize "__ip6_tnl_parm" struct to zeros in the "vti6_siocdevprivate()" function. Signed-off-by: William Zhao Signed-off-by: David S. Miller --- net/ipv6/ip6_vti.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index 527e9ead7449..5e9474bc54fc 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -808,6 +808,8 @@ vti6_siocdevprivate(struct net_device *dev, struct ifreq *ifr, void __user *data struct net *net = dev_net(dev); struct vti6_net *ip6n = net_generic(net, vti6_net_id); + memset(&p1, 0, sizeof(p1)); + switch (cmd) { case SIOCGETTUNNEL: if (dev == ip6n->fb_tnl_dev) { -- cgit v1.2.3 From 6d7373dabfd3933ee30c40fc8c09d2a788f6ece1 Mon Sep 17 00:00:00 2001 From: Karsten Graul Date: Mon, 27 Dec 2021 14:35:30 +0100 Subject: net/smc: fix using of uninitialized completions In smc_wr_tx_send_wait() the completion on index specified by pend->idx is initialized and after smc_wr_tx_send() was called the wait for completion starts. pend->idx is used to get the correct index for the wait, but the pend structure could already be cleared in smc_wr_tx_process_cqe(). Introduce pnd_idx to hold and use a local copy of the correct index. Fixes: 09c61d24f96d ("net/smc: wait for departure of an IB message") Signed-off-by: Karsten Graul Signed-off-by: David S. Miller --- net/smc/smc_wr.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/smc/smc_wr.c b/net/smc/smc_wr.c index 600ab5889227..79a7431f534e 100644 --- a/net/smc/smc_wr.c +++ b/net/smc/smc_wr.c @@ -358,18 +358,20 @@ int smc_wr_tx_send_wait(struct smc_link *link, struct smc_wr_tx_pend_priv *priv, unsigned long timeout) { struct smc_wr_tx_pend *pend; + u32 pnd_idx; int rc; pend = container_of(priv, struct smc_wr_tx_pend, priv); pend->compl_requested = 1; - init_completion(&link->wr_tx_compl[pend->idx]); + pnd_idx = pend->idx; + init_completion(&link->wr_tx_compl[pnd_idx]); rc = smc_wr_tx_send(link, priv); if (rc) return rc; /* wait for completion by smc_wr_tx_process_cqe() */ rc = wait_for_completion_interruptible_timeout( - &link->wr_tx_compl[pend->idx], timeout); + &link->wr_tx_compl[pnd_idx], timeout); if (rc <= 0) rc = -ENODATA; if (rc > 0) -- cgit v1.2.3 From 6c25449e1a32c594d743df8e8258e8ef870b6a77 Mon Sep 17 00:00:00 2001 From: yangxingwu Date: Mon, 27 Dec 2021 16:29:51 +0800 Subject: net: udp: fix alignment problem in udp4_seq_show() $ cat /pro/net/udp before: sl local_address rem_address st tx_queue rx_queue tr tm->when 26050: 0100007F:0035 00000000:0000 07 00000000:00000000 00:00000000 26320: 0100007F:0143 00000000:0000 07 00000000:00000000 00:00000000 27135: 00000000:8472 00000000:0000 07 00000000:00000000 00:00000000 after: sl local_address rem_address st tx_queue rx_queue tr tm->when 26050: 0100007F:0035 00000000:0000 07 00000000:00000000 00:00000000 26320: 0100007F:0143 00000000:0000 07 00000000:00000000 00:00000000 27135: 00000000:8472 00000000:0000 07 00000000:00000000 00:00000000 Signed-off-by: yangxingwu Signed-off-by: David S. Miller --- net/ipv4/udp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 15c6b450b8db..0cd6b857e7ec 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -3075,7 +3075,7 @@ int udp4_seq_show(struct seq_file *seq, void *v) { seq_setwidth(seq, 127); if (v == SEQ_START_TOKEN) - seq_puts(seq, " sl local_address rem_address st tx_queue " + seq_puts(seq, " sl local_address rem_address st tx_queue " "rx_queue tr tm->when retrnsmt uid timeout " "inode ref pointer drops"); else { -- cgit v1.2.3 From 90cee52f2e780345d3629e278291aea5ac74f40f Mon Sep 17 00:00:00 2001 From: Dust Li Date: Tue, 28 Dec 2021 17:03:24 +0800 Subject: net/smc: don't send CDC/LLC message if link not ready We found smc_llc_send_link_delete_all() sometimes wait for 2s timeout when testing with RDMA link up/down. It is possible when a smc_link is in ACTIVATING state, the underlaying QP is still in RESET or RTR state, which cannot send any messages out. smc_llc_send_link_delete_all() use smc_link_usable() to checks whether the link is usable, if the QP is still in RESET or RTR state, but the smc_link is in ACTIVATING, this LLC message will always fail without any CQE entering the CQ, and we will always wait 2s before timeout. Since we cannot send any messages through the QP before the QP enter RTS. I add a wrapper smc_link_sendable() which checks the state of QP along with the link state. And replace smc_link_usable() with smc_link_sendable() in all LLC & CDC message sending routine. Fixes: 5f08318f617b ("smc: connection data control (CDC)") Signed-off-by: Dust Li Signed-off-by: David S. Miller --- net/smc/smc_core.c | 2 +- net/smc/smc_core.h | 6 ++++++ net/smc/smc_llc.c | 2 +- net/smc/smc_wr.c | 4 ++-- net/smc/smc_wr.h | 2 +- 5 files changed, 11 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c index 387d28b2f8dd..55ca175e8d57 100644 --- a/net/smc/smc_core.c +++ b/net/smc/smc_core.c @@ -647,7 +647,7 @@ static void smcr_lgr_link_deactivate_all(struct smc_link_group *lgr) for (i = 0; i < SMC_LINKS_PER_LGR_MAX; i++) { struct smc_link *lnk = &lgr->lnk[i]; - if (smc_link_usable(lnk)) + if (smc_link_sendable(lnk)) lnk->state = SMC_LNK_INACTIVE; } wake_up_all(&lgr->llc_msg_waiter); diff --git a/net/smc/smc_core.h b/net/smc/smc_core.h index 59cef3b830d8..d63b08274197 100644 --- a/net/smc/smc_core.h +++ b/net/smc/smc_core.h @@ -415,6 +415,12 @@ static inline bool smc_link_usable(struct smc_link *lnk) return true; } +static inline bool smc_link_sendable(struct smc_link *lnk) +{ + return smc_link_usable(lnk) && + lnk->qp_attr.cur_qp_state == IB_QPS_RTS; +} + static inline bool smc_link_active(struct smc_link *lnk) { return lnk->state == SMC_LNK_ACTIVE; diff --git a/net/smc/smc_llc.c b/net/smc/smc_llc.c index b102680296b8..3e9fd8a3124c 100644 --- a/net/smc/smc_llc.c +++ b/net/smc/smc_llc.c @@ -1630,7 +1630,7 @@ void smc_llc_send_link_delete_all(struct smc_link_group *lgr, bool ord, u32 rsn) delllc.reason = htonl(rsn); for (i = 0; i < SMC_LINKS_PER_LGR_MAX; i++) { - if (!smc_link_usable(&lgr->lnk[i])) + if (!smc_link_sendable(&lgr->lnk[i])) continue; if (!smc_llc_send_message_wait(&lgr->lnk[i], &delllc)) break; diff --git a/net/smc/smc_wr.c b/net/smc/smc_wr.c index 79a7431f534e..df1dc225cbab 100644 --- a/net/smc/smc_wr.c +++ b/net/smc/smc_wr.c @@ -188,7 +188,7 @@ void smc_wr_tx_cq_handler(struct ib_cq *ib_cq, void *cq_context) static inline int smc_wr_tx_get_free_slot_index(struct smc_link *link, u32 *idx) { *idx = link->wr_tx_cnt; - if (!smc_link_usable(link)) + if (!smc_link_sendable(link)) return -ENOLINK; for_each_clear_bit(*idx, link->wr_tx_mask, link->wr_tx_cnt) { if (!test_and_set_bit(*idx, link->wr_tx_mask)) @@ -231,7 +231,7 @@ int smc_wr_tx_get_free_slot(struct smc_link *link, } else { rc = wait_event_interruptible_timeout( link->wr_tx_wait, - !smc_link_usable(link) || + !smc_link_sendable(link) || lgr->terminating || (smc_wr_tx_get_free_slot_index(link, &idx) != -EBUSY), SMC_WR_TX_WAIT_FREE_SLOT_TIME); diff --git a/net/smc/smc_wr.h b/net/smc/smc_wr.h index f353311e6f84..48ed9b08ac7a 100644 --- a/net/smc/smc_wr.h +++ b/net/smc/smc_wr.h @@ -62,7 +62,7 @@ static inline void smc_wr_tx_set_wr_id(atomic_long_t *wr_tx_id, long val) static inline bool smc_wr_tx_link_hold(struct smc_link *link) { - if (!smc_link_usable(link)) + if (!smc_link_sendable(link)) return false; atomic_inc(&link->wr_tx_refcnt); return true; -- cgit v1.2.3 From 349d43127dac00c15231e8ffbcaabd70f7b0e544 Mon Sep 17 00:00:00 2001 From: Dust Li Date: Tue, 28 Dec 2021 17:03:25 +0800 Subject: net/smc: fix kernel panic caused by race of smc_sock A crash occurs when smc_cdc_tx_handler() tries to access smc_sock but smc_release() has already freed it. [ 4570.695099] BUG: unable to handle page fault for address: 000000002eae9e88 [ 4570.696048] #PF: supervisor write access in kernel mode [ 4570.696728] #PF: error_code(0x0002) - not-present page [ 4570.697401] PGD 0 P4D 0 [ 4570.697716] Oops: 0002 [#1] PREEMPT SMP NOPTI [ 4570.698228] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 5.16.0-rc4+ #111 [ 4570.699013] Hardware name: Alibaba Cloud Alibaba Cloud ECS, BIOS 8c24b4c 04/0 [ 4570.699933] RIP: 0010:_raw_spin_lock+0x1a/0x30 <...> [ 4570.711446] Call Trace: [ 4570.711746] [ 4570.711992] smc_cdc_tx_handler+0x41/0xc0 [ 4570.712470] smc_wr_tx_tasklet_fn+0x213/0x560 [ 4570.712981] ? smc_cdc_tx_dismisser+0x10/0x10 [ 4570.713489] tasklet_action_common.isra.17+0x66/0x140 [ 4570.714083] __do_softirq+0x123/0x2f4 [ 4570.714521] irq_exit_rcu+0xc4/0xf0 [ 4570.714934] common_interrupt+0xba/0xe0 Though smc_cdc_tx_handler() checked the existence of smc connection, smc_release() may have already dismissed and released the smc socket before smc_cdc_tx_handler() further visits it. smc_cdc_tx_handler() |smc_release() if (!conn) | | |smc_cdc_tx_dismiss_slots() | smc_cdc_tx_dismisser() | |sock_put(&smc->sk) <- last sock_put, | smc_sock freed bh_lock_sock(&smc->sk) (panic) | To make sure we won't receive any CDC messages after we free the smc_sock, add a refcount on the smc_connection for inflight CDC message(posted to the QP but haven't received related CQE), and don't release the smc_connection until all the inflight CDC messages haven been done, for both success or failed ones. Using refcount on CDC messages brings another problem: when the link is going to be destroyed, smcr_link_clear() will reset the QP, which then remove all the pending CQEs related to the QP in the CQ. To make sure all the CQEs will always come back so the refcount on the smc_connection can always reach 0, smc_ib_modify_qp_reset() was replaced by smc_ib_modify_qp_error(). And remove the timeout in smc_wr_tx_wait_no_pending_sends() since we need to wait for all pending WQEs done, or we may encounter use-after- free when handling CQEs. For IB device removal routine, we need to wait for all the QPs on that device been destroyed before we can destroy CQs on the device, or the refcount on smc_connection won't reach 0 and smc_sock cannot be released. Fixes: 5f08318f617b ("smc: connection data control (CDC)") Reported-by: Wen Gu Signed-off-by: Dust Li Signed-off-by: David S. Miller --- net/smc/smc.h | 5 +++++ net/smc/smc_cdc.c | 52 ++++++++++++++++++++++++---------------------------- net/smc/smc_cdc.h | 2 +- net/smc/smc_core.c | 25 ++++++++++++++++++++----- net/smc/smc_ib.c | 4 ++-- net/smc/smc_ib.h | 1 + net/smc/smc_wr.c | 41 +++-------------------------------------- net/smc/smc_wr.h | 3 +-- 8 files changed, 57 insertions(+), 76 deletions(-) (limited to 'net') diff --git a/net/smc/smc.h b/net/smc/smc.h index f4286ca1f228..1a4fc1c6c4ab 100644 --- a/net/smc/smc.h +++ b/net/smc/smc.h @@ -180,6 +180,11 @@ struct smc_connection { u16 tx_cdc_seq; /* sequence # for CDC send */ u16 tx_cdc_seq_fin; /* sequence # - tx completed */ spinlock_t send_lock; /* protect wr_sends */ + atomic_t cdc_pend_tx_wr; /* number of pending tx CDC wqe + * - inc when post wqe, + * - dec on polled tx cqe + */ + wait_queue_head_t cdc_pend_tx_wq; /* wakeup on no cdc_pend_tx_wr*/ struct delayed_work tx_work; /* retry of smc_cdc_msg_send */ u32 tx_off; /* base offset in peer rmb */ diff --git a/net/smc/smc_cdc.c b/net/smc/smc_cdc.c index 99acd337ba90..84c8a4374fdd 100644 --- a/net/smc/smc_cdc.c +++ b/net/smc/smc_cdc.c @@ -31,10 +31,6 @@ static void smc_cdc_tx_handler(struct smc_wr_tx_pend_priv *pnd_snd, struct smc_sock *smc; int diff; - if (!conn) - /* already dismissed */ - return; - smc = container_of(conn, struct smc_sock, conn); bh_lock_sock(&smc->sk); if (!wc_status) { @@ -51,6 +47,12 @@ static void smc_cdc_tx_handler(struct smc_wr_tx_pend_priv *pnd_snd, conn); conn->tx_cdc_seq_fin = cdcpend->ctrl_seq; } + + if (atomic_dec_and_test(&conn->cdc_pend_tx_wr) && + unlikely(wq_has_sleeper(&conn->cdc_pend_tx_wq))) + wake_up(&conn->cdc_pend_tx_wq); + WARN_ON(atomic_read(&conn->cdc_pend_tx_wr) < 0); + smc_tx_sndbuf_nonfull(smc); bh_unlock_sock(&smc->sk); } @@ -107,6 +109,10 @@ int smc_cdc_msg_send(struct smc_connection *conn, conn->tx_cdc_seq++; conn->local_tx_ctrl.seqno = conn->tx_cdc_seq; smc_host_msg_to_cdc((struct smc_cdc_msg *)wr_buf, conn, &cfed); + + atomic_inc(&conn->cdc_pend_tx_wr); + smp_mb__after_atomic(); /* Make sure cdc_pend_tx_wr added before post */ + rc = smc_wr_tx_send(link, (struct smc_wr_tx_pend_priv *)pend); if (!rc) { smc_curs_copy(&conn->rx_curs_confirmed, &cfed, conn); @@ -114,6 +120,7 @@ int smc_cdc_msg_send(struct smc_connection *conn, } else { conn->tx_cdc_seq--; conn->local_tx_ctrl.seqno = conn->tx_cdc_seq; + atomic_dec(&conn->cdc_pend_tx_wr); } return rc; @@ -136,7 +143,18 @@ int smcr_cdc_msg_send_validation(struct smc_connection *conn, peer->token = htonl(local->token); peer->prod_flags.failover_validation = 1; + /* We need to set pend->conn here to make sure smc_cdc_tx_handler() + * can handle properly + */ + smc_cdc_add_pending_send(conn, pend); + + atomic_inc(&conn->cdc_pend_tx_wr); + smp_mb__after_atomic(); /* Make sure cdc_pend_tx_wr added before post */ + rc = smc_wr_tx_send(link, (struct smc_wr_tx_pend_priv *)pend); + if (unlikely(rc)) + atomic_dec(&conn->cdc_pend_tx_wr); + return rc; } @@ -193,31 +211,9 @@ int smc_cdc_get_slot_and_msg_send(struct smc_connection *conn) return rc; } -static bool smc_cdc_tx_filter(struct smc_wr_tx_pend_priv *tx_pend, - unsigned long data) +void smc_cdc_wait_pend_tx_wr(struct smc_connection *conn) { - struct smc_connection *conn = (struct smc_connection *)data; - struct smc_cdc_tx_pend *cdc_pend = - (struct smc_cdc_tx_pend *)tx_pend; - - return cdc_pend->conn == conn; -} - -static void smc_cdc_tx_dismisser(struct smc_wr_tx_pend_priv *tx_pend) -{ - struct smc_cdc_tx_pend *cdc_pend = - (struct smc_cdc_tx_pend *)tx_pend; - - cdc_pend->conn = NULL; -} - -void smc_cdc_tx_dismiss_slots(struct smc_connection *conn) -{ - struct smc_link *link = conn->lnk; - - smc_wr_tx_dismiss_slots(link, SMC_CDC_MSG_TYPE, - smc_cdc_tx_filter, smc_cdc_tx_dismisser, - (unsigned long)conn); + wait_event(conn->cdc_pend_tx_wq, !atomic_read(&conn->cdc_pend_tx_wr)); } /* Send a SMC-D CDC header. diff --git a/net/smc/smc_cdc.h b/net/smc/smc_cdc.h index 0a0a89abd38b..696cc11f2303 100644 --- a/net/smc/smc_cdc.h +++ b/net/smc/smc_cdc.h @@ -291,7 +291,7 @@ int smc_cdc_get_free_slot(struct smc_connection *conn, struct smc_wr_buf **wr_buf, struct smc_rdma_wr **wr_rdma_buf, struct smc_cdc_tx_pend **pend); -void smc_cdc_tx_dismiss_slots(struct smc_connection *conn); +void smc_cdc_wait_pend_tx_wr(struct smc_connection *conn); int smc_cdc_msg_send(struct smc_connection *conn, struct smc_wr_buf *wr_buf, struct smc_cdc_tx_pend *pend); int smc_cdc_get_slot_and_msg_send(struct smc_connection *conn); diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c index 55ca175e8d57..a6849362f4dd 100644 --- a/net/smc/smc_core.c +++ b/net/smc/smc_core.c @@ -1127,7 +1127,7 @@ void smc_conn_free(struct smc_connection *conn) smc_ism_unset_conn(conn); tasklet_kill(&conn->rx_tsklet); } else { - smc_cdc_tx_dismiss_slots(conn); + smc_cdc_wait_pend_tx_wr(conn); if (current_work() != &conn->abort_work) cancel_work_sync(&conn->abort_work); } @@ -1204,7 +1204,7 @@ void smcr_link_clear(struct smc_link *lnk, bool log) smc_llc_link_clear(lnk, log); smcr_buf_unmap_lgr(lnk); smcr_rtoken_clear_link(lnk); - smc_ib_modify_qp_reset(lnk); + smc_ib_modify_qp_error(lnk); smc_wr_free_link(lnk); smc_ib_destroy_queue_pair(lnk); smc_ib_dealloc_protection_domain(lnk); @@ -1336,7 +1336,7 @@ static void smc_conn_kill(struct smc_connection *conn, bool soft) else tasklet_unlock_wait(&conn->rx_tsklet); } else { - smc_cdc_tx_dismiss_slots(conn); + smc_cdc_wait_pend_tx_wr(conn); } smc_lgr_unregister_conn(conn); smc_close_active_abort(smc); @@ -1459,11 +1459,16 @@ void smc_smcd_terminate_all(struct smcd_dev *smcd) /* Called when an SMCR device is removed or the smc module is unloaded. * If smcibdev is given, all SMCR link groups using this device are terminated. * If smcibdev is NULL, all SMCR link groups are terminated. + * + * We must wait here for QPs been destroyed before we destroy the CQs, + * or we won't received any CQEs and cdc_pend_tx_wr cannot reach 0 thus + * smc_sock cannot be released. */ void smc_smcr_terminate_all(struct smc_ib_device *smcibdev) { struct smc_link_group *lgr, *lg; LIST_HEAD(lgr_free_list); + LIST_HEAD(lgr_linkdown_list); int i; spin_lock_bh(&smc_lgr_list.lock); @@ -1475,7 +1480,7 @@ void smc_smcr_terminate_all(struct smc_ib_device *smcibdev) list_for_each_entry_safe(lgr, lg, &smc_lgr_list.list, list) { for (i = 0; i < SMC_LINKS_PER_LGR_MAX; i++) { if (lgr->lnk[i].smcibdev == smcibdev) - smcr_link_down_cond_sched(&lgr->lnk[i]); + list_move_tail(&lgr->list, &lgr_linkdown_list); } } } @@ -1487,6 +1492,16 @@ void smc_smcr_terminate_all(struct smc_ib_device *smcibdev) __smc_lgr_terminate(lgr, false); } + list_for_each_entry_safe(lgr, lg, &lgr_linkdown_list, list) { + for (i = 0; i < SMC_LINKS_PER_LGR_MAX; i++) { + if (lgr->lnk[i].smcibdev == smcibdev) { + mutex_lock(&lgr->llc_conf_mutex); + smcr_link_down_cond(&lgr->lnk[i]); + mutex_unlock(&lgr->llc_conf_mutex); + } + } + } + if (smcibdev) { if (atomic_read(&smcibdev->lnk_cnt)) wait_event(smcibdev->lnks_deleted, @@ -1586,7 +1601,6 @@ static void smcr_link_down(struct smc_link *lnk) if (!lgr || lnk->state == SMC_LNK_UNUSED || list_empty(&lgr->list)) return; - smc_ib_modify_qp_reset(lnk); to_lnk = smc_switch_conns(lgr, lnk, true); if (!to_lnk) { /* no backup link available */ smcr_link_clear(lnk, true); @@ -1824,6 +1838,7 @@ create: conn->local_tx_ctrl.common.type = SMC_CDC_MSG_TYPE; conn->local_tx_ctrl.len = SMC_WR_TX_SIZE; conn->urg_state = SMC_URG_READ; + init_waitqueue_head(&conn->cdc_pend_tx_wq); INIT_WORK(&smc->conn.abort_work, smc_conn_abort_work); if (ini->is_smcd) { conn->rx_off = sizeof(struct smcd_cdc_msg); diff --git a/net/smc/smc_ib.c b/net/smc/smc_ib.c index d93055ec17ae..fe5d5399c4e8 100644 --- a/net/smc/smc_ib.c +++ b/net/smc/smc_ib.c @@ -109,12 +109,12 @@ int smc_ib_modify_qp_rts(struct smc_link *lnk) IB_QP_MAX_QP_RD_ATOMIC); } -int smc_ib_modify_qp_reset(struct smc_link *lnk) +int smc_ib_modify_qp_error(struct smc_link *lnk) { struct ib_qp_attr qp_attr; memset(&qp_attr, 0, sizeof(qp_attr)); - qp_attr.qp_state = IB_QPS_RESET; + qp_attr.qp_state = IB_QPS_ERR; return ib_modify_qp(lnk->roce_qp, &qp_attr, IB_QP_STATE); } diff --git a/net/smc/smc_ib.h b/net/smc/smc_ib.h index 07585937370e..bfa1c6bf6313 100644 --- a/net/smc/smc_ib.h +++ b/net/smc/smc_ib.h @@ -90,6 +90,7 @@ int smc_ib_create_queue_pair(struct smc_link *lnk); int smc_ib_ready_link(struct smc_link *lnk); int smc_ib_modify_qp_rts(struct smc_link *lnk); int smc_ib_modify_qp_reset(struct smc_link *lnk); +int smc_ib_modify_qp_error(struct smc_link *lnk); long smc_ib_setup_per_ibdev(struct smc_ib_device *smcibdev); int smc_ib_get_memory_region(struct ib_pd *pd, int access_flags, struct smc_buf_desc *buf_slot, u8 link_idx); diff --git a/net/smc/smc_wr.c b/net/smc/smc_wr.c index df1dc225cbab..c6cfdea8b71b 100644 --- a/net/smc/smc_wr.c +++ b/net/smc/smc_wr.c @@ -62,13 +62,9 @@ static inline bool smc_wr_is_tx_pend(struct smc_link *link) } /* wait till all pending tx work requests on the given link are completed */ -int smc_wr_tx_wait_no_pending_sends(struct smc_link *link) +void smc_wr_tx_wait_no_pending_sends(struct smc_link *link) { - if (wait_event_timeout(link->wr_tx_wait, !smc_wr_is_tx_pend(link), - SMC_WR_TX_WAIT_PENDING_TIME)) - return 0; - else /* timeout */ - return -EPIPE; + wait_event(link->wr_tx_wait, !smc_wr_is_tx_pend(link)); } static inline int smc_wr_tx_find_pending_index(struct smc_link *link, u64 wr_id) @@ -87,7 +83,6 @@ static inline void smc_wr_tx_process_cqe(struct ib_wc *wc) struct smc_wr_tx_pend pnd_snd; struct smc_link *link; u32 pnd_snd_idx; - int i; link = wc->qp->qp_context; @@ -128,14 +123,6 @@ static inline void smc_wr_tx_process_cqe(struct ib_wc *wc) } if (wc->status) { - for_each_set_bit(i, link->wr_tx_mask, link->wr_tx_cnt) { - /* clear full struct smc_wr_tx_pend including .priv */ - memset(&link->wr_tx_pends[i], 0, - sizeof(link->wr_tx_pends[i])); - memset(&link->wr_tx_bufs[i], 0, - sizeof(link->wr_tx_bufs[i])); - clear_bit(i, link->wr_tx_mask); - } if (link->lgr->smc_version == SMC_V2) { memset(link->wr_tx_v2_pend, 0, sizeof(*link->wr_tx_v2_pend)); @@ -421,25 +408,6 @@ int smc_wr_reg_send(struct smc_link *link, struct ib_mr *mr) return rc; } -void smc_wr_tx_dismiss_slots(struct smc_link *link, u8 wr_tx_hdr_type, - smc_wr_tx_filter filter, - smc_wr_tx_dismisser dismisser, - unsigned long data) -{ - struct smc_wr_tx_pend_priv *tx_pend; - struct smc_wr_rx_hdr *wr_tx; - int i; - - for_each_set_bit(i, link->wr_tx_mask, link->wr_tx_cnt) { - wr_tx = (struct smc_wr_rx_hdr *)&link->wr_tx_bufs[i]; - if (wr_tx->type != wr_tx_hdr_type) - continue; - tx_pend = &link->wr_tx_pends[i].priv; - if (filter(tx_pend, data)) - dismisser(tx_pend); - } -} - /****************************** receive queue ********************************/ int smc_wr_rx_register_handler(struct smc_wr_rx_handler *handler) @@ -675,10 +643,7 @@ void smc_wr_free_link(struct smc_link *lnk) smc_wr_wakeup_reg_wait(lnk); smc_wr_wakeup_tx_wait(lnk); - if (smc_wr_tx_wait_no_pending_sends(lnk)) - memset(lnk->wr_tx_mask, 0, - BITS_TO_LONGS(SMC_WR_BUF_CNT) * - sizeof(*lnk->wr_tx_mask)); + smc_wr_tx_wait_no_pending_sends(lnk); wait_event(lnk->wr_reg_wait, (!atomic_read(&lnk->wr_reg_refcnt))); wait_event(lnk->wr_tx_wait, (!atomic_read(&lnk->wr_tx_refcnt))); diff --git a/net/smc/smc_wr.h b/net/smc/smc_wr.h index 48ed9b08ac7a..47512ccce5ef 100644 --- a/net/smc/smc_wr.h +++ b/net/smc/smc_wr.h @@ -22,7 +22,6 @@ #define SMC_WR_BUF_CNT 16 /* # of ctrl buffers per link */ #define SMC_WR_TX_WAIT_FREE_SLOT_TIME (10 * HZ) -#define SMC_WR_TX_WAIT_PENDING_TIME (5 * HZ) #define SMC_WR_TX_SIZE 44 /* actual size of wr_send data (<=SMC_WR_BUF_SIZE) */ @@ -130,7 +129,7 @@ void smc_wr_tx_dismiss_slots(struct smc_link *lnk, u8 wr_rx_hdr_type, smc_wr_tx_filter filter, smc_wr_tx_dismisser dismisser, unsigned long data); -int smc_wr_tx_wait_no_pending_sends(struct smc_link *link); +void smc_wr_tx_wait_no_pending_sends(struct smc_link *link); int smc_wr_rx_register_handler(struct smc_wr_rx_handler *handler); int smc_wr_rx_post_init(struct smc_link *link); -- cgit v1.2.3 From 5bec7ca2be6955ca1aa0d7bae2b981de9b1c9844 Mon Sep 17 00:00:00 2001 From: Ciara Loftus Date: Mon, 20 Dec 2021 15:52:50 +0000 Subject: xsk: Initialise xskb free_list_node This commit initialises the xskb's free_list_node when the xskb is allocated. This prevents a potential false negative returned from a call to list_empty for that node, such as the one introduced in commit 199d983bc015 ("xsk: Fix crash on double free in buffer pool") In my environment this issue caused packets to not be received by the xdpsock application if the traffic was running prior to application launch. This happened when the first batch of packets failed the xskmap lookup and XDP_PASS was returned from the bpf program. This action is handled in the i40e zc driver (and others) by allocating an skbuff, freeing the xdp_buff and adding the associated xskb to the xsk_buff_pool's free_list if it hadn't been added already. Without this fix, the xskb is not added to the free_list because the check to determine if it was added already returns an invalid positive result. Later, this caused allocation errors in the driver and the failure to receive packets. Fixes: 199d983bc015 ("xsk: Fix crash on double free in buffer pool") Fixes: 2b43470add8c ("xsk: Introduce AF_XDP buffer allocation API") Signed-off-by: Ciara Loftus Acked-by: Magnus Karlsson Link: https://lore.kernel.org/r/20211220155250.2746-1-ciara.loftus@intel.com Signed-off-by: Jakub Kicinski --- net/xdp/xsk_buff_pool.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/xdp/xsk_buff_pool.c b/net/xdp/xsk_buff_pool.c index bc4ad48ea4f0..fd39bb660ebc 100644 --- a/net/xdp/xsk_buff_pool.c +++ b/net/xdp/xsk_buff_pool.c @@ -83,6 +83,7 @@ struct xsk_buff_pool *xp_create_and_assign_umem(struct xdp_sock *xs, xskb = &pool->heads[i]; xskb->pool = pool; xskb->xdp.frame_sz = umem->chunk_size - umem->headroom; + INIT_LIST_HEAD(&xskb->free_list_node); if (pool->unaligned) pool->free_heads[i] = xskb; else -- cgit v1.2.3 From fb7bc9204095090731430c8921f9e629740c110a Mon Sep 17 00:00:00 2001 From: Tamir Duberstein Date: Wed, 29 Dec 2021 15:09:47 -0500 Subject: ipv6: raw: check passed optlen before reading Add a check that the user-provided option is at least as long as the number of bytes we intend to read. Before this patch we would blindly read sizeof(int) bytes even in cases where the user passed optlen Signed-off-by: Willem de Bruijn Link: https://lore.kernel.org/r/20211229200947.2862255-1-willemdebruijn.kernel@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv6/raw.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 60f1e4f5be5a..c51d5ce3711c 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -1020,6 +1020,9 @@ static int do_rawv6_setsockopt(struct sock *sk, int level, int optname, struct raw6_sock *rp = raw6_sk(sk); int val; + if (optlen < sizeof(val)) + return -EINVAL; + if (copy_from_sockptr(&val, optval, sizeof(val))) return -EFAULT; -- cgit v1.2.3 From 99b40610956a8a8755653a67392e2a8b772453be Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Mon, 27 Dec 2021 19:21:15 +0200 Subject: net: bridge: mcast: add and enforce query interval minimum As reported[1] if query interval is set too low and we have multiple bridges or even a single bridge with multiple querier vlans configured we can crash the machine. Add a 1 second minimum which must be enforced by overwriting the value if set lower (i.e. without returning an error) to avoid breaking user-space. If that happens a log message is emitted to let the administrator know that the interval has been set to the minimum. The issue has been present since these intervals could be user-controlled. [1] https://lore.kernel.org/netdev/e8b9ce41-57b9-b6e2-a46a-ff9c791cf0ba@gmail.com/ Fixes: d902eee43f19 ("bridge: Add multicast count/interval sysfs entries") Reported-by: Eric Dumazet Signed-off-by: Nikolay Aleksandrov Signed-off-by: Jakub Kicinski --- net/bridge/br_multicast.c | 16 ++++++++++++++++ net/bridge/br_netlink.c | 2 +- net/bridge/br_private.h | 3 +++ net/bridge/br_sysfs_br.c | 2 +- net/bridge/br_vlan_options.c | 2 +- 5 files changed, 22 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index f3d751105343..998da4a2d209 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -4522,6 +4522,22 @@ int br_multicast_set_mld_version(struct net_bridge_mcast *brmctx, } #endif +void br_multicast_set_query_intvl(struct net_bridge_mcast *brmctx, + unsigned long val) +{ + unsigned long intvl_jiffies = clock_t_to_jiffies(val); + + if (intvl_jiffies < BR_MULTICAST_QUERY_INTVL_MIN) { + br_info(brmctx->br, + "trying to set multicast query interval below minimum, setting to %lu (%ums)\n", + jiffies_to_clock_t(BR_MULTICAST_QUERY_INTVL_MIN), + jiffies_to_msecs(BR_MULTICAST_QUERY_INTVL_MIN)); + intvl_jiffies = BR_MULTICAST_QUERY_INTVL_MIN; + } + + brmctx->multicast_query_interval = intvl_jiffies; +} + /** * br_multicast_list_adjacent - Returns snooped multicast addresses * @dev: The bridge port adjacent to which to retrieve addresses diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index 0c8b5f1a15bc..701dd8b8455e 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -1357,7 +1357,7 @@ static int br_changelink(struct net_device *brdev, struct nlattr *tb[], if (data[IFLA_BR_MCAST_QUERY_INTVL]) { u64 val = nla_get_u64(data[IFLA_BR_MCAST_QUERY_INTVL]); - br->multicast_ctx.multicast_query_interval = clock_t_to_jiffies(val); + br_multicast_set_query_intvl(&br->multicast_ctx, val); } if (data[IFLA_BR_MCAST_QUERY_RESPONSE_INTVL]) { diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index c0efd697865a..4ed7f11042e8 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -28,6 +28,7 @@ #define BR_MAX_PORTS (1<multicast_ctx.multicast_query_interval = clock_t_to_jiffies(val); + br_multicast_set_query_intvl(&br->multicast_ctx, val); return 0; } diff --git a/net/bridge/br_vlan_options.c b/net/bridge/br_vlan_options.c index 8ffd4ed2563c..bf1ac0874279 100644 --- a/net/bridge/br_vlan_options.c +++ b/net/bridge/br_vlan_options.c @@ -521,7 +521,7 @@ static int br_vlan_process_global_one_opts(const struct net_bridge *br, u64 val; val = nla_get_u64(tb[BRIDGE_VLANDB_GOPTS_MCAST_QUERY_INTVL]); - v->br_mcast_ctx.multicast_query_interval = clock_t_to_jiffies(val); + br_multicast_set_query_intvl(&v->br_mcast_ctx, val); *changed = true; } if (tb[BRIDGE_VLANDB_GOPTS_MCAST_QUERY_RESPONSE_INTVL]) { -- cgit v1.2.3 From f83a112bd91a494cdee671aec74e777470fb4a07 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Mon, 27 Dec 2021 19:21:16 +0200 Subject: net: bridge: mcast: add and enforce startup query interval minimum As reported[1] if startup query interval is set too low in combination with large number of startup queries and we have multiple bridges or even a single bridge with multiple querier vlans configured we can crash the machine. Add a 1 second minimum which must be enforced by overwriting the value if set lower (i.e. without returning an error) to avoid breaking user-space. If that happens a log message is emitted to let the admin know that the startup interval has been set to the minimum. It doesn't make sense to make the startup interval lower than the normal query interval so use the same value of 1 second. The issue has been present since these intervals could be user-controlled. [1] https://lore.kernel.org/netdev/e8b9ce41-57b9-b6e2-a46a-ff9c791cf0ba@gmail.com/ Fixes: d902eee43f19 ("bridge: Add multicast count/interval sysfs entries") Reported-by: Eric Dumazet Signed-off-by: Nikolay Aleksandrov Signed-off-by: Jakub Kicinski --- net/bridge/br_multicast.c | 16 ++++++++++++++++ net/bridge/br_netlink.c | 2 +- net/bridge/br_private.h | 3 +++ net/bridge/br_sysfs_br.c | 2 +- net/bridge/br_vlan_options.c | 2 +- 5 files changed, 22 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 998da4a2d209..de2409889489 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -4538,6 +4538,22 @@ void br_multicast_set_query_intvl(struct net_bridge_mcast *brmctx, brmctx->multicast_query_interval = intvl_jiffies; } +void br_multicast_set_startup_query_intvl(struct net_bridge_mcast *brmctx, + unsigned long val) +{ + unsigned long intvl_jiffies = clock_t_to_jiffies(val); + + if (intvl_jiffies < BR_MULTICAST_STARTUP_QUERY_INTVL_MIN) { + br_info(brmctx->br, + "trying to set multicast startup query interval below minimum, setting to %lu (%ums)\n", + jiffies_to_clock_t(BR_MULTICAST_STARTUP_QUERY_INTVL_MIN), + jiffies_to_msecs(BR_MULTICAST_STARTUP_QUERY_INTVL_MIN)); + intvl_jiffies = BR_MULTICAST_STARTUP_QUERY_INTVL_MIN; + } + + brmctx->multicast_startup_query_interval = intvl_jiffies; +} + /** * br_multicast_list_adjacent - Returns snooped multicast addresses * @dev: The bridge port adjacent to which to retrieve addresses diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index 701dd8b8455e..2ff83d84230d 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -1369,7 +1369,7 @@ static int br_changelink(struct net_device *brdev, struct nlattr *tb[], if (data[IFLA_BR_MCAST_STARTUP_QUERY_INTVL]) { u64 val = nla_get_u64(data[IFLA_BR_MCAST_STARTUP_QUERY_INTVL]); - br->multicast_ctx.multicast_startup_query_interval = clock_t_to_jiffies(val); + br_multicast_set_startup_query_intvl(&br->multicast_ctx, val); } if (data[IFLA_BR_MCAST_STATS_ENABLED]) { diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 4ed7f11042e8..2187a0c3fd22 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -29,6 +29,7 @@ #define BR_MULTICAST_DEFAULT_HASH_MAX 4096 #define BR_MULTICAST_QUERY_INTVL_MIN msecs_to_jiffies(1000) +#define BR_MULTICAST_STARTUP_QUERY_INTVL_MIN BR_MULTICAST_QUERY_INTVL_MIN #define BR_HWDOM_MAX BITS_PER_LONG @@ -966,6 +967,8 @@ size_t br_multicast_querier_state_size(void); size_t br_rports_size(const struct net_bridge_mcast *brmctx); void br_multicast_set_query_intvl(struct net_bridge_mcast *brmctx, unsigned long val); +void br_multicast_set_startup_query_intvl(struct net_bridge_mcast *brmctx, + unsigned long val); static inline bool br_group_is_l2(const struct br_ip *group) { diff --git a/net/bridge/br_sysfs_br.c b/net/bridge/br_sysfs_br.c index f5bd1114a434..7b0c19772111 100644 --- a/net/bridge/br_sysfs_br.c +++ b/net/bridge/br_sysfs_br.c @@ -706,7 +706,7 @@ static ssize_t multicast_startup_query_interval_show( static int set_startup_query_interval(struct net_bridge *br, unsigned long val, struct netlink_ext_ack *extack) { - br->multicast_ctx.multicast_startup_query_interval = clock_t_to_jiffies(val); + br_multicast_set_startup_query_intvl(&br->multicast_ctx, val); return 0; } diff --git a/net/bridge/br_vlan_options.c b/net/bridge/br_vlan_options.c index bf1ac0874279..a6382973b3e7 100644 --- a/net/bridge/br_vlan_options.c +++ b/net/bridge/br_vlan_options.c @@ -535,7 +535,7 @@ static int br_vlan_process_global_one_opts(const struct net_bridge *br, u64 val; val = nla_get_u64(tb[BRIDGE_VLANDB_GOPTS_MCAST_STARTUP_QUERY_INTVL]); - v->br_mcast_ctx.multicast_startup_query_interval = clock_t_to_jiffies(val); + br_multicast_set_startup_query_intvl(&v->br_mcast_ctx, val); *changed = true; } if (tb[BRIDGE_VLANDB_GOPTS_MCAST_QUERIER]) { -- cgit v1.2.3 From e22e45fc9e41bf9fcc1e92cfb78eb92786728ef0 Mon Sep 17 00:00:00 2001 From: Muchun Song Date: Tue, 28 Dec 2021 18:41:45 +0800 Subject: net: fix use-after-free in tw_timer_handler A real world panic issue was found as follow in Linux 5.4. BUG: unable to handle page fault for address: ffffde49a863de28 PGD 7e6fe62067 P4D 7e6fe62067 PUD 7e6fe63067 PMD f51e064067 PTE 0 RIP: 0010:tw_timer_handler+0x20/0x40 Call Trace: call_timer_fn+0x2b/0x120 run_timer_softirq+0x1ef/0x450 __do_softirq+0x10d/0x2b8 irq_exit+0xc7/0xd0 smp_apic_timer_interrupt+0x68/0x120 apic_timer_interrupt+0xf/0x20 This issue was also reported since 2017 in the thread [1], unfortunately, the issue was still can be reproduced after fixing DCCP. The ipv4_mib_exit_net is called before tcp_sk_exit_batch when a net namespace is destroyed since tcp_sk_ops is registered befrore ipv4_mib_ops, which means tcp_sk_ops is in the front of ipv4_mib_ops in the list of pernet_list. There will be a use-after-free on net->mib.net_statistics in tw_timer_handler after ipv4_mib_exit_net if there are some inflight time-wait timers. This bug is not introduced by commit f2bf415cfed7 ("mib: add net to NET_ADD_STATS_BH") since the net_statistics is a global variable instead of dynamic allocation and freeing. Actually, commit 61a7e26028b9 ("mib: put net statistics on struct net") introduces the bug since it put net statistics on struct net and free it when net namespace is destroyed. Moving init_ipv4_mibs() to the front of tcp_init() to fix this bug and replace pr_crit() with panic() since continuing is meaningless when init_ipv4_mibs() fails. [1] https://groups.google.com/g/syzkaller/c/p1tn-_Kc6l4/m/smuL_FMAAgAJ?pli=1 Fixes: 61a7e26028b9 ("mib: put net statistics on struct net") Signed-off-by: Muchun Song Cc: Cong Wang Cc: Fam Zheng Cc: Link: https://lore.kernel.org/r/20211228104145.9426-1-songmuchun@bytedance.com Signed-off-by: Jakub Kicinski --- net/ipv4/af_inet.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 6b5956500436..5f70ffdae1b5 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1994,6 +1994,10 @@ static int __init inet_init(void) ip_init(); + /* Initialise per-cpu ipv4 mibs */ + if (init_ipv4_mibs()) + panic("%s: Cannot init ipv4 mibs\n", __func__); + /* Setup TCP slab cache for open requests. */ tcp_init(); @@ -2024,12 +2028,6 @@ static int __init inet_init(void) if (init_inet_pernet_ops()) pr_crit("%s: Cannot init ipv4 inet pernet ops\n", __func__); - /* - * Initialise per-cpu ipv4 mibs - */ - - if (init_ipv4_mibs()) - pr_crit("%s: Cannot init ipv4 mibs\n", __func__); ipv4_proc_init(); -- cgit v1.2.3 From 168fed986b3a7ec7b98cab1fe84e2f282b9e6a8f Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Tue, 28 Dec 2021 17:31:42 +0200 Subject: net: bridge: mcast: fix br_multicast_ctx_vlan_global_disabled helper We need to first check if the context is a vlan one, then we need to check the global bridge multicast vlan snooping flag, and finally the vlan's multicast flag, otherwise we will unnecessarily enable vlan mcast processing (e.g. querier timers). Fixes: 7b54aaaf53cb ("net: bridge: multicast: add vlan state initialization and control") Signed-off-by: Nikolay Aleksandrov Link: https://lore.kernel.org/r/20211228153142.536969-1-nikolay@nvidia.com Signed-off-by: Jakub Kicinski --- net/bridge/br_private.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 2187a0c3fd22..e8c6ee322c71 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -1153,9 +1153,9 @@ br_multicast_port_ctx_get_global(const struct net_bridge_mcast_port *pmctx) static inline bool br_multicast_ctx_vlan_global_disabled(const struct net_bridge_mcast *brmctx) { - return br_opt_get(brmctx->br, BROPT_MCAST_VLAN_SNOOPING_ENABLED) && - br_multicast_ctx_is_vlan(brmctx) && - !(brmctx->vlan->priv_flags & BR_VLFLAG_GLOBAL_MCAST_ENABLED); + return br_multicast_ctx_is_vlan(brmctx) && + (!br_opt_get(brmctx->br, BROPT_MCAST_VLAN_SNOOPING_ENABLED) || + !(brmctx->vlan->priv_flags & BR_VLFLAG_GLOBAL_MCAST_ENABLED)); } static inline bool -- cgit v1.2.3 From 92a34ab169f9eefe29cd420ce96b0a0a2a1da853 Mon Sep 17 00:00:00 2001 From: Jiasheng Jiang Date: Wed, 29 Dec 2021 11:21:18 +0800 Subject: net/ncsi: check for error return from call to nla_put_u32 As we can see from the comment of the nla_put() that it could return -EMSGSIZE if the tailroom of the skb is insufficient. Therefore, it should be better to check the return value of the nla_put_u32 and return the error code if error accurs. Also, there are many other functions have the same problem, and if this patch is correct, I will commit a new version to fix all. Fixes: 955dc68cb9b2 ("net/ncsi: Add generic netlink family") Signed-off-by: Jiasheng Jiang Link: https://lore.kernel.org/r/20211229032118.1706294-1-jiasheng@iscas.ac.cn Signed-off-by: Jakub Kicinski --- net/ncsi/ncsi-netlink.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ncsi/ncsi-netlink.c b/net/ncsi/ncsi-netlink.c index bb5f1650f11c..c189b4c8a182 100644 --- a/net/ncsi/ncsi-netlink.c +++ b/net/ncsi/ncsi-netlink.c @@ -112,7 +112,11 @@ static int ncsi_write_package_info(struct sk_buff *skb, pnest = nla_nest_start_noflag(skb, NCSI_PKG_ATTR); if (!pnest) return -ENOMEM; - nla_put_u32(skb, NCSI_PKG_ATTR_ID, np->id); + rc = nla_put_u32(skb, NCSI_PKG_ATTR_ID, np->id); + if (rc) { + nla_nest_cancel(skb, pnest); + return rc; + } if ((0x1 << np->id) == ndp->package_whitelist) nla_put_flag(skb, NCSI_PKG_ATTR_FORCED); cnest = nla_nest_start_noflag(skb, NCSI_PKG_ATTR_CHANNEL_LIST); -- cgit v1.2.3