From 1f0fc3374f3345ff1d150c5c56ac5016e5d3826a Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 18 Feb 2025 19:22:48 +0000 Subject: afs: Give an afs_server object a ref on the afs_cell object it points to Give an afs_server object a ref on the afs_cell object it points to so that the cell doesn't get deleted before the server record. Whilst this is circular (cell -> vol -> server_list -> server -> cell), the ref only pins the memory, not the lifetime as that's controlled by the activity counter. When the volume's activity counter reaches 0, it detaches from the cell and discards its server list; when a cell's activity counter reaches 0, it discards its root volume. At that point, the circularity is cut. Fixes: d2ddc776a458 ("afs: Overhaul volume and server record caching and fileserver rotation") Signed-off-by: David Howells cc: Marc Dionne cc: Simon Horman cc: linux-afs@lists.infradead.org Link: https://patch.msgid.link/20250218192250.296870-6-dhowells@redhat.com Signed-off-by: Jakub Kicinski --- include/trace/events/afs.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/trace/events/afs.h b/include/trace/events/afs.h index b0db89058c91..958a2460330c 100644 --- a/include/trace/events/afs.h +++ b/include/trace/events/afs.h @@ -174,6 +174,7 @@ enum yfs_cm_operation { EM(afs_cell_trace_get_queue_dns, "GET q-dns ") \ EM(afs_cell_trace_get_queue_manage, "GET q-mng ") \ EM(afs_cell_trace_get_queue_new, "GET q-new ") \ + EM(afs_cell_trace_get_server, "GET server") \ EM(afs_cell_trace_get_vol, "GET vol ") \ EM(afs_cell_trace_insert, "INSERT ") \ EM(afs_cell_trace_manage, "MANAGE ") \ @@ -182,6 +183,7 @@ enum yfs_cm_operation { EM(afs_cell_trace_put_destroy, "PUT destry") \ EM(afs_cell_trace_put_queue_work, "PUT q-work") \ EM(afs_cell_trace_put_queue_fail, "PUT q-fail") \ + EM(afs_cell_trace_put_server, "PUT server") \ EM(afs_cell_trace_put_vol, "PUT vol ") \ EM(afs_cell_trace_see_source, "SEE source") \ EM(afs_cell_trace_see_ws, "SEE ws ") \ -- cgit v1.2.3 From 5c70eb5c593d64d93b178905da215a9fd288a4b5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 20 Feb 2025 13:18:54 +0000 Subject: net: better track kernel sockets lifetime While kernel sockets are dismantled during pernet_operations->exit(), their freeing can be delayed by any tx packets still held in qdisc or device queues, due to skb_set_owner_w() prior calls. This then trigger the following warning from ref_tracker_dir_exit() [1] To fix this, make sure that kernel sockets own a reference on net->passive. Add sk_net_refcnt_upgrade() helper, used whenever a kernel socket is converted to a refcounted one. [1] [ 136.263918][ T35] ref_tracker: net notrefcnt@ffff8880638f01e0 has 1/2 users at [ 136.263918][ T35] sk_alloc+0x2b3/0x370 [ 136.263918][ T35] inet6_create+0x6ce/0x10f0 [ 136.263918][ T35] __sock_create+0x4c0/0xa30 [ 136.263918][ T35] inet_ctl_sock_create+0xc2/0x250 [ 136.263918][ T35] igmp6_net_init+0x39/0x390 [ 136.263918][ T35] ops_init+0x31e/0x590 [ 136.263918][ T35] setup_net+0x287/0x9e0 [ 136.263918][ T35] copy_net_ns+0x33f/0x570 [ 136.263918][ T35] create_new_namespaces+0x425/0x7b0 [ 136.263918][ T35] unshare_nsproxy_namespaces+0x124/0x180 [ 136.263918][ T35] ksys_unshare+0x57d/0xa70 [ 136.263918][ T35] __x64_sys_unshare+0x38/0x40 [ 136.263918][ T35] do_syscall_64+0xf3/0x230 [ 136.263918][ T35] entry_SYSCALL_64_after_hwframe+0x77/0x7f [ 136.263918][ T35] [ 136.343488][ T35] ref_tracker: net notrefcnt@ffff8880638f01e0 has 1/2 users at [ 136.343488][ T35] sk_alloc+0x2b3/0x370 [ 136.343488][ T35] inet6_create+0x6ce/0x10f0 [ 136.343488][ T35] __sock_create+0x4c0/0xa30 [ 136.343488][ T35] inet_ctl_sock_create+0xc2/0x250 [ 136.343488][ T35] ndisc_net_init+0xa7/0x2b0 [ 136.343488][ T35] ops_init+0x31e/0x590 [ 136.343488][ T35] setup_net+0x287/0x9e0 [ 136.343488][ T35] copy_net_ns+0x33f/0x570 [ 136.343488][ T35] create_new_namespaces+0x425/0x7b0 [ 136.343488][ T35] unshare_nsproxy_namespaces+0x124/0x180 [ 136.343488][ T35] ksys_unshare+0x57d/0xa70 [ 136.343488][ T35] __x64_sys_unshare+0x38/0x40 [ 136.343488][ T35] do_syscall_64+0xf3/0x230 [ 136.343488][ T35] entry_SYSCALL_64_after_hwframe+0x77/0x7f Fixes: 0cafd77dcd03 ("net: add a refcount tracker for kernel sockets") Reported-by: syzbot+30a19e01a97420719891@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/67b72aeb.050a0220.14d86d.0283.GAE@google.com/T/#u Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250220131854.4048077-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/sock.h | 1 + net/core/sock.c | 27 ++++++++++++++++++++++----- net/mptcp/subflow.c | 5 +---- net/netlink/af_netlink.c | 10 ---------- net/rds/tcp.c | 8 ++------ net/smc/af_smc.c | 5 +---- net/sunrpc/svcsock.c | 5 +---- net/sunrpc/xprtsock.c | 8 ++------ 8 files changed, 30 insertions(+), 39 deletions(-) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index 8036b3b79cd8..7ef728324e4e 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1751,6 +1751,7 @@ static inline bool sock_allow_reclassification(const struct sock *csk) struct sock *sk_alloc(struct net *net, int family, gfp_t priority, struct proto *prot, int kern); void sk_free(struct sock *sk); +void sk_net_refcnt_upgrade(struct sock *sk); void sk_destruct(struct sock *sk); struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority); void sk_free_unlock_clone(struct sock *sk); diff --git a/net/core/sock.c b/net/core/sock.c index eae2ae70a2e0..6c0e87f97fa4 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -2246,6 +2246,7 @@ struct sock *sk_alloc(struct net *net, int family, gfp_t priority, get_net_track(net, &sk->ns_tracker, priority); sock_inuse_add(net, 1); } else { + net_passive_inc(net); __netns_tracker_alloc(net, &sk->ns_tracker, false, priority); } @@ -2270,6 +2271,7 @@ EXPORT_SYMBOL(sk_alloc); static void __sk_destruct(struct rcu_head *head) { struct sock *sk = container_of(head, struct sock, sk_rcu); + struct net *net = sock_net(sk); struct sk_filter *filter; if (sk->sk_destruct) @@ -2301,14 +2303,28 @@ static void __sk_destruct(struct rcu_head *head) put_cred(sk->sk_peer_cred); put_pid(sk->sk_peer_pid); - if (likely(sk->sk_net_refcnt)) - put_net_track(sock_net(sk), &sk->ns_tracker); - else - __netns_tracker_free(sock_net(sk), &sk->ns_tracker, false); - + if (likely(sk->sk_net_refcnt)) { + put_net_track(net, &sk->ns_tracker); + } else { + __netns_tracker_free(net, &sk->ns_tracker, false); + net_passive_dec(net); + } sk_prot_free(sk->sk_prot_creator, sk); } +void sk_net_refcnt_upgrade(struct sock *sk) +{ + struct net *net = sock_net(sk); + + WARN_ON_ONCE(sk->sk_net_refcnt); + __netns_tracker_free(net, &sk->ns_tracker, false); + net_passive_dec(net); + sk->sk_net_refcnt = 1; + get_net_track(net, &sk->ns_tracker, GFP_KERNEL); + sock_inuse_add(net, 1); +} +EXPORT_SYMBOL_GPL(sk_net_refcnt_upgrade); + void sk_destruct(struct sock *sk) { bool use_call_rcu = sock_flag(sk, SOCK_RCU_FREE); @@ -2405,6 +2421,7 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority) * is not properly dismantling its kernel sockets at netns * destroy time. */ + net_passive_inc(sock_net(newsk)); __netns_tracker_alloc(sock_net(newsk), &newsk->ns_tracker, false, priority); } diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index fd021cf8286e..dfcbef9c4624 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -1772,10 +1772,7 @@ int mptcp_subflow_create_socket(struct sock *sk, unsigned short family, * needs it. * Update ns_tracker to current stack trace and refcounted tracker. */ - __netns_tracker_free(net, &sf->sk->ns_tracker, false); - sf->sk->sk_net_refcnt = 1; - get_net_track(net, &sf->sk->ns_tracker, GFP_KERNEL); - sock_inuse_add(net, 1); + sk_net_refcnt_upgrade(sf->sk); err = tcp_set_ulp(sf->sk, "mptcp"); if (err) goto err_free; diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 85311226183a..a53ea60d0a78 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -795,16 +795,6 @@ static int netlink_release(struct socket *sock) sock_prot_inuse_add(sock_net(sk), &netlink_proto, -1); - /* Because struct net might disappear soon, do not keep a pointer. */ - if (!sk->sk_net_refcnt && sock_net(sk) != &init_net) { - __netns_tracker_free(sock_net(sk), &sk->ns_tracker, false); - /* Because of deferred_put_nlk_sk and use of work queue, - * it is possible netns will be freed before this socket. - */ - sock_net_set(sk, &init_net); - __netns_tracker_alloc(&init_net, &sk->ns_tracker, - false, GFP_KERNEL); - } call_rcu(&nlk->rcu, deferred_put_nlk_sk); return 0; } diff --git a/net/rds/tcp.c b/net/rds/tcp.c index 0581c53e6517..3cc2f303bf78 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -504,12 +504,8 @@ bool rds_tcp_tune(struct socket *sock) release_sock(sk); return false; } - /* Update ns_tracker to current stack trace and refcounted tracker */ - __netns_tracker_free(net, &sk->ns_tracker, false); - - sk->sk_net_refcnt = 1; - netns_tracker_alloc(net, &sk->ns_tracker, GFP_KERNEL); - sock_inuse_add(net, 1); + sk_net_refcnt_upgrade(sk); + put_net(net); } rtn = net_generic(net, rds_tcp_netid); if (rtn->sndbuf_size > 0) { diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index ca6984541edb..3e6cb35baf25 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -3337,10 +3337,7 @@ int smc_create_clcsk(struct net *net, struct sock *sk, int family) * which need net ref. */ sk = smc->clcsock->sk; - __netns_tracker_free(net, &sk->ns_tracker, false); - sk->sk_net_refcnt = 1; - get_net_track(net, &sk->ns_tracker, GFP_KERNEL); - sock_inuse_add(net, 1); + sk_net_refcnt_upgrade(sk); return 0; } diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index cb3bd12f5818..72e5a01df3d3 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -1541,10 +1541,7 @@ static struct svc_xprt *svc_create_socket(struct svc_serv *serv, newlen = error; if (protocol == IPPROTO_TCP) { - __netns_tracker_free(net, &sock->sk->ns_tracker, false); - sock->sk->sk_net_refcnt = 1; - get_net_track(net, &sock->sk->ns_tracker, GFP_KERNEL); - sock_inuse_add(net, 1); + sk_net_refcnt_upgrade(sock->sk); if ((error = kernel_listen(sock, 64)) < 0) goto bummer; } diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index c60936d8cef7..940fe65b2a35 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1941,12 +1941,8 @@ static struct socket *xs_create_sock(struct rpc_xprt *xprt, goto out; } - if (protocol == IPPROTO_TCP) { - __netns_tracker_free(xprt->xprt_net, &sock->sk->ns_tracker, false); - sock->sk->sk_net_refcnt = 1; - get_net_track(xprt->xprt_net, &sock->sk->ns_tracker, GFP_KERNEL); - sock_inuse_add(xprt->xprt_net, 1); - } + if (protocol == IPPROTO_TCP) + sk_net_refcnt_upgrade(sock->sk); filp = sock_alloc_file(sock, O_NONBLOCK, NULL); if (IS_ERR(filp)) -- cgit v1.2.3 From 18912c520674ec4d920fe3826e7e4fefeecdf5ae Mon Sep 17 00:00:00 2001 From: Stanislav Fomichev Date: Mon, 24 Feb 2025 09:44:01 -0800 Subject: tcp: devmem: don't write truncated dmabuf CMSGs to userspace Currently, we report -ETOOSMALL (err) only on the first iteration (!sent). When we get put_cmsg error after a bunch of successful put_cmsg calls, we don't signal the error at all. This might be confusing on the userspace side which will see truncated CMSGs but no MSG_CTRUNC signal. Consider the following case: - sizeof(struct cmsghdr) = 16 - sizeof(struct dmabuf_cmsg) = 24 - total cmsg size (CMSG_LEN) = 40 (16+24) When calling recvmsg with msg_controllen=60, the userspace will receive two(!) dmabuf_cmsg(s), the first one will be a valid one and the second one will be silently truncated. There is no easy way to discover the truncation besides doing something like "cm->cmsg_len != CMSG_LEN(sizeof(dmabuf_cmsg))". Introduce new put_devmem_cmsg wrapper that reports an error instead of doing the truncation. Mina suggests that it's the intended way this API should work. Note that we might now report MSG_CTRUNC when the users (incorrectly) call us with msg_control == NULL. Fixes: 8f0b3cc9a4c1 ("tcp: RX path for devmem TCP") Reviewed-by: Mina Almasry Signed-off-by: Stanislav Fomichev Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20250224174401.3582695-1-sdf@fomichev.me Signed-off-by: Jakub Kicinski --- include/linux/socket.h | 2 ++ net/core/scm.c | 10 ++++++++++ net/ipv4/tcp.c | 26 ++++++++++---------------- 3 files changed, 22 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/include/linux/socket.h b/include/linux/socket.h index d18cc47e89bd..c3322eb3d686 100644 --- a/include/linux/socket.h +++ b/include/linux/socket.h @@ -392,6 +392,8 @@ struct ucred { extern int move_addr_to_kernel(void __user *uaddr, int ulen, struct sockaddr_storage *kaddr); extern int put_cmsg(struct msghdr*, int level, int type, int len, void *data); +extern int put_cmsg_notrunc(struct msghdr *msg, int level, int type, int len, + void *data); struct timespec64; struct __kernel_timespec; diff --git a/net/core/scm.c b/net/core/scm.c index 4f6a14babe5a..733c0cbd393d 100644 --- a/net/core/scm.c +++ b/net/core/scm.c @@ -282,6 +282,16 @@ efault: } EXPORT_SYMBOL(put_cmsg); +int put_cmsg_notrunc(struct msghdr *msg, int level, int type, int len, + void *data) +{ + /* Don't produce truncated CMSGs */ + if (!msg->msg_control || msg->msg_controllen < CMSG_LEN(len)) + return -ETOOSMALL; + + return put_cmsg(msg, level, type, len, data); +} + void put_cmsg_scm_timestamping64(struct msghdr *msg, struct scm_timestamping_internal *tss_internal) { struct scm_timestamping64 tss; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 0d704bda6c41..d74281eca14f 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2438,14 +2438,12 @@ static int tcp_recvmsg_dmabuf(struct sock *sk, const struct sk_buff *skb, */ memset(&dmabuf_cmsg, 0, sizeof(dmabuf_cmsg)); dmabuf_cmsg.frag_size = copy; - err = put_cmsg(msg, SOL_SOCKET, SO_DEVMEM_LINEAR, - sizeof(dmabuf_cmsg), &dmabuf_cmsg); - if (err || msg->msg_flags & MSG_CTRUNC) { - msg->msg_flags &= ~MSG_CTRUNC; - if (!err) - err = -ETOOSMALL; + err = put_cmsg_notrunc(msg, SOL_SOCKET, + SO_DEVMEM_LINEAR, + sizeof(dmabuf_cmsg), + &dmabuf_cmsg); + if (err) goto out; - } sent += copy; @@ -2499,16 +2497,12 @@ static int tcp_recvmsg_dmabuf(struct sock *sk, const struct sk_buff *skb, offset += copy; remaining_len -= copy; - err = put_cmsg(msg, SOL_SOCKET, - SO_DEVMEM_DMABUF, - sizeof(dmabuf_cmsg), - &dmabuf_cmsg); - if (err || msg->msg_flags & MSG_CTRUNC) { - msg->msg_flags &= ~MSG_CTRUNC; - if (!err) - err = -ETOOSMALL; + err = put_cmsg_notrunc(msg, SOL_SOCKET, + SO_DEVMEM_DMABUF, + sizeof(dmabuf_cmsg), + &dmabuf_cmsg); + if (err) goto out; - } atomic_long_inc(&niov->pp_ref_count); tcp_xa_pool.netmems[tcp_xa_pool.idx++] = skb_frag_netmem(frag); -- cgit v1.2.3