diff options
Diffstat (limited to 'net/unix')
| -rw-r--r-- | net/unix/Kconfig | 9 | ||||
| -rw-r--r-- | net/unix/Makefile | 1 | ||||
| -rw-r--r-- | net/unix/af_unix.c | 1657 | ||||
| -rw-r--r-- | net/unix/diag.c | 64 | ||||
| -rw-r--r-- | net/unix/garbage.c | 34 | ||||
| -rw-r--r-- | net/unix/scm.c | 7 | ||||
| -rw-r--r-- | net/unix/sysctl_net_unix.c | 21 | ||||
| -rw-r--r-- | net/unix/unix_bpf.c | 175 |
8 files changed, 1497 insertions, 471 deletions
diff --git a/net/unix/Kconfig b/net/unix/Kconfig index a23a5cca9753..b7f811216820 100644 --- a/net/unix/Kconfig +++ b/net/unix/Kconfig @@ -5,7 +5,7 @@ config UNIX tristate "Unix domain sockets" - ---help--- + help If you say Y here, you will include support for Unix domain sockets; sockets are the standard Unix mechanism for establishing and accessing network connections. Many commonly used programs such as @@ -25,10 +25,15 @@ config UNIX_SCM depends on UNIX default y +config AF_UNIX_OOB + bool + depends on UNIX + default y + config UNIX_DIAG tristate "UNIX: socket monitoring interface" depends on UNIX default n - ---help--- + help Support for UNIX socket monitoring interface used by the ss tool. If unsure, say Y. diff --git a/net/unix/Makefile b/net/unix/Makefile index 54e58cc4f945..20491825b4d0 100644 --- a/net/unix/Makefile +++ b/net/unix/Makefile @@ -7,6 +7,7 @@ obj-$(CONFIG_UNIX) += unix.o unix-y := af_unix.o garbage.o unix-$(CONFIG_SYSCTL) += sysctl_net_unix.o +unix-$(CONFIG_BPF_SYSCALL) += unix_bpf.o obj-$(CONFIG_UNIX_DIAG) += unix_diag.o unix_diag-y := diag.o diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 68debcb28fa4..b3545fc68097 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -89,6 +89,7 @@ #include <linux/socket.h> #include <linux/un.h> #include <linux/fcntl.h> +#include <linux/filter.h> #include <linux/termios.h> #include <linux/sockios.h> #include <linux/net.h> @@ -113,27 +114,74 @@ #include <linux/security.h> #include <linux/freezer.h> #include <linux/file.h> +#include <linux/btf_ids.h> #include "scm.h" -struct hlist_head unix_socket_table[2 * UNIX_HASH_SIZE]; -EXPORT_SYMBOL_GPL(unix_socket_table); -DEFINE_SPINLOCK(unix_table_lock); -EXPORT_SYMBOL_GPL(unix_table_lock); static atomic_long_t unix_nr_socks; +static struct hlist_head bsd_socket_buckets[UNIX_HASH_SIZE / 2]; +static spinlock_t bsd_socket_locks[UNIX_HASH_SIZE / 2]; +/* SMP locking strategy: + * hash table is protected with spinlock. + * each socket state is protected by separate spinlock. + */ -static struct hlist_head *unix_sockets_unbound(void *addr) +static unsigned int unix_unbound_hash(struct sock *sk) { - unsigned long hash = (unsigned long)addr; + unsigned long hash = (unsigned long)sk; hash ^= hash >> 16; hash ^= hash >> 8; - hash %= UNIX_HASH_SIZE; - return &unix_socket_table[UNIX_HASH_SIZE + hash]; + hash ^= sk->sk_type; + + return hash & UNIX_HASH_MOD; +} + +static unsigned int unix_bsd_hash(struct inode *i) +{ + return i->i_ino & UNIX_HASH_MOD; } -#define UNIX_ABSTRACT(sk) (unix_sk(sk)->addr->hash < UNIX_HASH_SIZE) +static unsigned int unix_abstract_hash(struct sockaddr_un *sunaddr, + int addr_len, int type) +{ + __wsum csum = csum_partial(sunaddr, addr_len, 0); + unsigned int hash; + + hash = (__force unsigned int)csum_fold(csum); + hash ^= hash >> 8; + hash ^= type; + + return UNIX_HASH_MOD + 1 + (hash & UNIX_HASH_MOD); +} + +static void unix_table_double_lock(struct net *net, + unsigned int hash1, unsigned int hash2) +{ + if (hash1 == hash2) { + spin_lock(&net->unx.table.locks[hash1]); + return; + } + + if (hash1 > hash2) + swap(hash1, hash2); + + spin_lock(&net->unx.table.locks[hash1]); + spin_lock_nested(&net->unx.table.locks[hash2], SINGLE_DEPTH_NESTING); +} + +static void unix_table_double_unlock(struct net *net, + unsigned int hash1, unsigned int hash2) +{ + if (hash1 == hash2) { + spin_unlock(&net->unx.table.locks[hash1]); + return; + } + + spin_unlock(&net->unx.table.locks[hash1]); + spin_unlock(&net->unx.table.locks[hash2]); +} #ifdef CONFIG_SECURITY_NETWORK static void unix_get_secdata(struct scm_cookie *scm, struct sk_buff *skb) @@ -163,20 +211,6 @@ static inline bool unix_secdata_eq(struct scm_cookie *scm, struct sk_buff *skb) } #endif /* CONFIG_SECURITY_NETWORK */ -/* - * SMP locking strategy: - * hash table is protected with spinlock unix_table_lock - * each socket state is protected by separate spin lock. - */ - -static inline unsigned int unix_hash_fold(__wsum n) -{ - unsigned int hash = (__force unsigned int)csum_fold(n); - - hash ^= hash>>8; - return hash&(UNIX_HASH_SIZE-1); -} - #define unix_peer(sk) (unix_sk(sk)->peer) static inline int unix_our_peer(struct sock *sk, struct sock *osk) @@ -213,6 +247,22 @@ struct sock *unix_peer_get(struct sock *s) } EXPORT_SYMBOL_GPL(unix_peer_get); +static struct unix_address *unix_create_addr(struct sockaddr_un *sunaddr, + int addr_len) +{ + struct unix_address *addr; + + addr = kmalloc(sizeof(*addr) + addr_len, GFP_KERNEL); + if (!addr) + return NULL; + + refcount_set(&addr->refcnt, 1); + addr->len = addr_len; + memcpy(addr->name, sunaddr, addr_len); + + return addr; +} + static inline void unix_release_addr(struct unix_address *addr) { if (refcount_dec_and_test(&addr->refcnt)) @@ -226,29 +276,29 @@ static inline void unix_release_addr(struct unix_address *addr) * - if started by zero, it is abstract name. */ -static int unix_mkname(struct sockaddr_un *sunaddr, int len, unsigned int *hashp) +static int unix_validate_addr(struct sockaddr_un *sunaddr, int addr_len) { - *hashp = 0; - - if (len <= sizeof(short) || len > sizeof(*sunaddr)) + if (addr_len <= offsetof(struct sockaddr_un, sun_path) || + addr_len > sizeof(*sunaddr)) return -EINVAL; - if (!sunaddr || sunaddr->sun_family != AF_UNIX) + + if (sunaddr->sun_family != AF_UNIX) return -EINVAL; - if (sunaddr->sun_path[0]) { - /* - * This may look like an off by one error but it is a bit more - * subtle. 108 is the longest valid AF_UNIX path for a binding. - * sun_path[108] doesn't as such exist. However in kernel space - * we are guaranteed that it is a valid memory location in our - * kernel address buffer. - */ - ((char *)sunaddr)[len] = 0; - len = strlen(sunaddr->sun_path)+1+sizeof(short); - return len; - } - *hashp = unix_hash_fold(csum_partial(sunaddr, len, 0)); - return len; + return 0; +} + +static void unix_mkname_bsd(struct sockaddr_un *sunaddr, int addr_len) +{ + /* This may look like an off by one error but it is a bit more + * subtle. 108 is the longest valid AF_UNIX path for a binding. + * sun_path[108] doesn't as such exist. However in kernel space + * we are guaranteed that it is a valid memory location in our + * kernel address buffer because syscall functions always pass + * a pointer of struct sockaddr_storage which has a bigger buffer + * than 108. + */ + ((char *)sunaddr)[addr_len] = 0; } static void __unix_remove_socket(struct sock *sk) @@ -256,38 +306,63 @@ static void __unix_remove_socket(struct sock *sk) sk_del_node_init(sk); } -static void __unix_insert_socket(struct hlist_head *list, struct sock *sk) +static void __unix_insert_socket(struct net *net, struct sock *sk) { - WARN_ON(!sk_unhashed(sk)); - sk_add_node(sk, list); + DEBUG_NET_WARN_ON_ONCE(!sk_unhashed(sk)); + sk_add_node(sk, &net->unx.table.buckets[sk->sk_hash]); +} + +static void __unix_set_addr_hash(struct net *net, struct sock *sk, + struct unix_address *addr, unsigned int hash) +{ + __unix_remove_socket(sk); + smp_store_release(&unix_sk(sk)->addr, addr); + + sk->sk_hash = hash; + __unix_insert_socket(net, sk); } -static inline void unix_remove_socket(struct sock *sk) +static void unix_remove_socket(struct net *net, struct sock *sk) { - spin_lock(&unix_table_lock); + spin_lock(&net->unx.table.locks[sk->sk_hash]); __unix_remove_socket(sk); - spin_unlock(&unix_table_lock); + spin_unlock(&net->unx.table.locks[sk->sk_hash]); +} + +static void unix_insert_unbound_socket(struct net *net, struct sock *sk) +{ + spin_lock(&net->unx.table.locks[sk->sk_hash]); + __unix_insert_socket(net, sk); + spin_unlock(&net->unx.table.locks[sk->sk_hash]); +} + +static void unix_insert_bsd_socket(struct sock *sk) +{ + spin_lock(&bsd_socket_locks[sk->sk_hash]); + sk_add_bind_node(sk, &bsd_socket_buckets[sk->sk_hash]); + spin_unlock(&bsd_socket_locks[sk->sk_hash]); } -static inline void unix_insert_socket(struct hlist_head *list, struct sock *sk) +static void unix_remove_bsd_socket(struct sock *sk) { - spin_lock(&unix_table_lock); - __unix_insert_socket(list, sk); - spin_unlock(&unix_table_lock); + if (!hlist_unhashed(&sk->sk_bind_node)) { + spin_lock(&bsd_socket_locks[sk->sk_hash]); + __sk_del_bind_node(sk); + spin_unlock(&bsd_socket_locks[sk->sk_hash]); + + sk_node_init(&sk->sk_bind_node); + } } static struct sock *__unix_find_socket_byname(struct net *net, struct sockaddr_un *sunname, - int len, int type, unsigned int hash) + int len, unsigned int hash) { struct sock *s; - sk_for_each(s, &unix_socket_table[hash ^ type]) { + sk_for_each(s, &net->unx.table.buckets[hash]) { struct unix_sock *u = unix_sk(s); - if (!net_eq(sock_net(s), net)) - continue; - if (u->addr->len == len && !memcmp(u->addr->name, sunname, len)) return s; @@ -297,37 +372,35 @@ static struct sock *__unix_find_socket_byname(struct net *net, static inline struct sock *unix_find_socket_byname(struct net *net, struct sockaddr_un *sunname, - int len, int type, - unsigned int hash) + int len, unsigned int hash) { struct sock *s; - spin_lock(&unix_table_lock); - s = __unix_find_socket_byname(net, sunname, len, type, hash); + spin_lock(&net->unx.table.locks[hash]); + s = __unix_find_socket_byname(net, sunname, len, hash); if (s) sock_hold(s); - spin_unlock(&unix_table_lock); + spin_unlock(&net->unx.table.locks[hash]); return s; } static struct sock *unix_find_socket_byinode(struct inode *i) { + unsigned int hash = unix_bsd_hash(i); struct sock *s; - spin_lock(&unix_table_lock); - sk_for_each(s, - &unix_socket_table[i->i_ino & (UNIX_HASH_SIZE - 1)]) { + spin_lock(&bsd_socket_locks[hash]); + sk_for_each_bound(s, &bsd_socket_buckets[hash]) { struct dentry *dentry = unix_sk(s)->path.dentry; if (dentry && d_backing_inode(dentry) == i) { sock_hold(s); - goto found; + spin_unlock(&bsd_socket_locks[hash]); + return s; } } - s = NULL; -found: - spin_unlock(&unix_table_lock); - return s; + spin_unlock(&bsd_socket_locks[hash]); + return NULL; } /* Support code for asymmetrically connected dgram sockets @@ -438,7 +511,7 @@ static int unix_dgram_peer_wake_me(struct sock *sk, struct sock *other) * -ECONNREFUSED. Otherwise, if we haven't queued any skbs * to other and its full, we will hang waiting for POLLOUT. */ - if (unix_recvq_full(other) && !sock_flag(other, SOCK_DEAD)) + if (unix_recvq_full_lockless(other) && !sock_flag(other, SOCK_DEAD)) return 1; if (connected) @@ -484,9 +557,10 @@ static void unix_dgram_disconnected(struct sock *sk, struct sock *other) */ if (!sock_flag(other, SOCK_DEAD) && unix_peer(other) == sk) { other->sk_err = ECONNRESET; - other->sk_error_report(other); + sk_error_report(other); } } + other->sk_state = TCP_CLOSE; } static void unix_sock_destructor(struct sock *sk) @@ -495,9 +569,9 @@ static void unix_sock_destructor(struct sock *sk) skb_queue_purge(&sk->sk_receive_queue); - WARN_ON(refcount_read(&sk->sk_wmem_alloc)); - WARN_ON(!sk_unhashed(sk)); - WARN_ON(sk->sk_socket); + DEBUG_NET_WARN_ON_ONCE(refcount_read(&sk->sk_wmem_alloc)); + DEBUG_NET_WARN_ON_ONCE(!sk_unhashed(sk)); + DEBUG_NET_WARN_ON_ONCE(sk->sk_socket); if (!sock_flag(sk, SOCK_DEAD)) { pr_info("Attempt to release alive unix socket: %p\n", sk); return; @@ -507,9 +581,7 @@ static void unix_sock_destructor(struct sock *sk) unix_release_addr(u->addr); atomic_long_dec(&unix_nr_socks); - local_bh_disable(); sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); - local_bh_enable(); #ifdef UNIX_REFCNT_DEBUG pr_debug("UNIX %p is destroyed, %ld are still alive.\n", sk, atomic_long_read(&unix_nr_socks)); @@ -519,12 +591,13 @@ static void unix_sock_destructor(struct sock *sk) static void unix_release_sock(struct sock *sk, int embrion) { struct unix_sock *u = unix_sk(sk); - struct path path; struct sock *skpair; struct sk_buff *skb; + struct path path; int state; - unix_remove_socket(sk); + unix_remove_socket(sock_net(sk), sk); + unix_remove_bsd_socket(sk); /* Clear state */ unix_state_lock(sk); @@ -535,11 +608,20 @@ static void unix_release_sock(struct sock *sk, int embrion) u->path.mnt = NULL; state = sk->sk_state; sk->sk_state = TCP_CLOSE; + + skpair = unix_peer(sk); + unix_peer(sk) = NULL; + unix_state_unlock(sk); - wake_up_interruptible_all(&u->peer_wait); +#if IS_ENABLED(CONFIG_AF_UNIX_OOB) + if (u->oob_skb) { + kfree_skb(u->oob_skb); + u->oob_skb = NULL; + } +#endif - skpair = unix_peer(sk); + wake_up_interruptible_all(&u->peer_wait); if (skpair != NULL) { if (sk->sk_type == SOCK_STREAM || sk->sk_type == SOCK_SEQPACKET) { @@ -555,7 +637,6 @@ static void unix_release_sock(struct sock *sk, int embrion) unix_dgram_peer_wake_disconnect(sk, skpair); sock_put(skpair); /* It may now die */ - unix_peer(sk) = NULL; } /* Try to flush out this socket. Throw out buffers at least */ @@ -592,20 +673,42 @@ static void unix_release_sock(struct sock *sk, int embrion) static void init_peercred(struct sock *sk) { - put_pid(sk->sk_peer_pid); - if (sk->sk_peer_cred) - put_cred(sk->sk_peer_cred); + const struct cred *old_cred; + struct pid *old_pid; + + spin_lock(&sk->sk_peer_lock); + old_pid = sk->sk_peer_pid; + old_cred = sk->sk_peer_cred; sk->sk_peer_pid = get_pid(task_tgid(current)); sk->sk_peer_cred = get_current_cred(); + spin_unlock(&sk->sk_peer_lock); + + put_pid(old_pid); + put_cred(old_cred); } static void copy_peercred(struct sock *sk, struct sock *peersk) { - put_pid(sk->sk_peer_pid); - if (sk->sk_peer_cred) - put_cred(sk->sk_peer_cred); + const struct cred *old_cred; + struct pid *old_pid; + + if (sk < peersk) { + spin_lock(&sk->sk_peer_lock); + spin_lock_nested(&peersk->sk_peer_lock, SINGLE_DEPTH_NESTING); + } else { + spin_lock(&peersk->sk_peer_lock); + spin_lock_nested(&sk->sk_peer_lock, SINGLE_DEPTH_NESTING); + } + old_pid = sk->sk_peer_pid; + old_cred = sk->sk_peer_cred; sk->sk_peer_pid = get_pid(peersk->sk_peer_pid); sk->sk_peer_cred = get_cred(peersk->sk_peer_cred); + + spin_unlock(&sk->sk_peer_lock); + spin_unlock(&peersk->sk_peer_lock); + + put_pid(old_pid); + put_cred(old_cred); } static int unix_listen(struct socket *sock, int backlog) @@ -613,7 +716,6 @@ static int unix_listen(struct socket *sock, int backlog) int err; struct sock *sk = sock->sk; struct unix_sock *u = unix_sk(sk); - struct pid *old_pid = NULL; err = -EOPNOTSUPP; if (sock->type != SOCK_STREAM && sock->type != SOCK_SEQPACKET) @@ -634,7 +736,6 @@ static int unix_listen(struct socket *sock, int backlog) out_unlock: unix_state_unlock(sk); - put_pid(old_pid); out: return err; } @@ -663,6 +764,8 @@ static ssize_t unix_stream_splice_read(struct socket *, loff_t *ppos, unsigned int flags); static int unix_dgram_sendmsg(struct socket *, struct msghdr *, size_t); static int unix_dgram_recvmsg(struct socket *, struct msghdr *, size_t, int); +static int unix_read_skb(struct sock *sk, skb_read_actor_t recv_actor); +static int unix_stream_read_skb(struct sock *sk, skb_read_actor_t recv_actor); static int unix_dgram_connect(struct socket *, struct sockaddr *, int, int); static int unix_seqpacket_sendmsg(struct socket *, struct msghdr *, size_t); @@ -683,14 +786,45 @@ static int unix_set_peek_off(struct sock *sk, int val) } #ifdef CONFIG_PROC_FS +static int unix_count_nr_fds(struct sock *sk) +{ + struct sk_buff *skb; + struct unix_sock *u; + int nr_fds = 0; + + spin_lock(&sk->sk_receive_queue.lock); + skb = skb_peek(&sk->sk_receive_queue); + while (skb) { + u = unix_sk(skb->sk); + nr_fds += atomic_read(&u->scm_stat.nr_fds); + skb = skb_peek_next(skb, &sk->sk_receive_queue); + } + spin_unlock(&sk->sk_receive_queue.lock); + + return nr_fds; +} + static void unix_show_fdinfo(struct seq_file *m, struct socket *sock) { struct sock *sk = sock->sk; struct unix_sock *u; + int nr_fds; if (sk) { - u = unix_sk(sock->sk); - seq_printf(m, "scm_fds: %u\n", READ_ONCE(u->scm_stat.nr_fds)); + u = unix_sk(sk); + if (sock->type == SOCK_DGRAM) { + nr_fds = atomic_read(&u->scm_stat.nr_fds); + goto out_print; + } + + unix_state_lock(sk); + if (sk->sk_state != TCP_LISTEN) + nr_fds = atomic_read(&u->scm_stat.nr_fds); + else + nr_fds = unix_count_nr_fds(sk); + unix_state_unlock(sk); +out_print: + seq_printf(m, "scm_fds: %u\n", nr_fds); } } #else @@ -713,10 +847,9 @@ static const struct proto_ops unix_stream_ops = { #endif .listen = unix_listen, .shutdown = unix_shutdown, - .setsockopt = sock_no_setsockopt, - .getsockopt = sock_no_getsockopt, .sendmsg = unix_stream_sendmsg, .recvmsg = unix_stream_recvmsg, + .read_skb = unix_stream_read_skb, .mmap = sock_no_mmap, .sendpage = unix_stream_sendpage, .splice_read = unix_stream_splice_read, @@ -740,9 +873,8 @@ static const struct proto_ops unix_dgram_ops = { #endif .listen = sock_no_listen, .shutdown = unix_shutdown, - .setsockopt = sock_no_setsockopt, - .getsockopt = sock_no_getsockopt, .sendmsg = unix_dgram_sendmsg, + .read_skb = unix_read_skb, .recvmsg = unix_dgram_recvmsg, .mmap = sock_no_mmap, .sendpage = sock_no_sendpage, @@ -766,8 +898,6 @@ static const struct proto_ops unix_seqpacket_ops = { #endif .listen = unix_listen, .shutdown = unix_shutdown, - .setsockopt = sock_no_setsockopt, - .getsockopt = sock_no_getsockopt, .sendmsg = unix_seqpacket_sendmsg, .recvmsg = unix_seqpacket_recvmsg, .mmap = sock_no_mmap, @@ -776,27 +906,66 @@ static const struct proto_ops unix_seqpacket_ops = { .show_fdinfo = unix_show_fdinfo, }; -static struct proto unix_proto = { +static void unix_close(struct sock *sk, long timeout) +{ + /* Nothing to do here, unix socket does not need a ->close(). + * This is merely for sockmap. + */ +} + +static void unix_unhash(struct sock *sk) +{ + /* Nothing to do here, unix socket does not need a ->unhash(). + * This is merely for sockmap. + */ +} + +struct proto unix_dgram_proto = { .name = "UNIX", .owner = THIS_MODULE, .obj_size = sizeof(struct unix_sock), + .close = unix_close, +#ifdef CONFIG_BPF_SYSCALL + .psock_update_sk_prot = unix_dgram_bpf_update_proto, +#endif }; -static struct sock *unix_create1(struct net *net, struct socket *sock, int kern) +struct proto unix_stream_proto = { + .name = "UNIX-STREAM", + .owner = THIS_MODULE, + .obj_size = sizeof(struct unix_sock), + .close = unix_close, + .unhash = unix_unhash, +#ifdef CONFIG_BPF_SYSCALL + .psock_update_sk_prot = unix_stream_bpf_update_proto, +#endif +}; + +static struct sock *unix_create1(struct net *net, struct socket *sock, int kern, int type) { - struct sock *sk = NULL; struct unix_sock *u; + struct sock *sk; + int err; atomic_long_inc(&unix_nr_socks); - if (atomic_long_read(&unix_nr_socks) > 2 * get_max_files()) - goto out; + if (atomic_long_read(&unix_nr_socks) > 2 * get_max_files()) { + err = -ENFILE; + goto err; + } - sk = sk_alloc(net, PF_UNIX, GFP_KERNEL, &unix_proto, kern); - if (!sk) - goto out; + if (type == SOCK_STREAM) + sk = sk_alloc(net, PF_UNIX, GFP_KERNEL, &unix_stream_proto, kern); + else /*dgram and seqpacket */ + sk = sk_alloc(net, PF_UNIX, GFP_KERNEL, &unix_dgram_proto, kern); + + if (!sk) { + err = -ENOMEM; + goto err; + } sock_init_data(sock, sk); + sk->sk_hash = unix_unbound_hash(sk); sk->sk_allocation = GFP_KERNEL_ACCOUNT; sk->sk_write_space = unix_write_space; sk->sk_max_ack_backlog = net->unx.sysctl_max_dgram_qlen; @@ -812,21 +981,22 @@ static struct sock *unix_create1(struct net *net, struct socket *sock, int kern) init_waitqueue_head(&u->peer_wait); init_waitqueue_func_entry(&u->peer_wake, unix_dgram_peer_wake_relay); memset(&u->scm_stat, 0, sizeof(struct scm_stat)); - unix_insert_socket(unix_sockets_unbound(sk), sk); -out: - if (sk == NULL) - atomic_long_dec(&unix_nr_socks); - else { - local_bh_disable(); - sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1); - local_bh_enable(); - } + unix_insert_unbound_socket(net, sk); + + sock_prot_inuse_add(net, sk->sk_prot, 1); + return sk; + +err: + atomic_long_dec(&unix_nr_socks); + return ERR_PTR(err); } static int unix_create(struct net *net, struct socket *sock, int protocol, int kern) { + struct sock *sk; + if (protocol && protocol != PF_UNIX) return -EPROTONOSUPPORT; @@ -842,7 +1012,7 @@ static int unix_create(struct net *net, struct socket *sock, int protocol, */ case SOCK_RAW: sock->type = SOCK_DGRAM; - /* fall through */ + fallthrough; case SOCK_DGRAM: sock->ops = &unix_dgram_ops; break; @@ -853,7 +1023,11 @@ static int unix_create(struct net *net, struct socket *sock, int protocol, return -ESOCKTNOSUPPORT; } - return unix_create1(net, sock, kern) ? 0 : -ENOMEM; + sk = unix_create1(net, sock, kern, sock->type); + if (IS_ERR(sk)) + return PTR_ERR(sk); + + return 0; } static int unix_release(struct socket *sock) @@ -863,245 +1037,284 @@ static int unix_release(struct socket *sock) if (!sk) return 0; + sk->sk_prot->close(sk, 0); unix_release_sock(sk, 0); sock->sk = NULL; return 0; } -static int unix_autobind(struct socket *sock) +static struct sock *unix_find_bsd(struct sockaddr_un *sunaddr, int addr_len, + int type) { - struct sock *sk = sock->sk; - struct net *net = sock_net(sk); + struct inode *inode; + struct path path; + struct sock *sk; + int err; + + unix_mkname_bsd(sunaddr, addr_len); + err = kern_path(sunaddr->sun_path, LOOKUP_FOLLOW, &path); + if (err) + goto fail; + + err = path_permission(&path, MAY_WRITE); + if (err) + goto path_put; + + err = -ECONNREFUSED; + inode = d_backing_inode(path.dentry); + if (!S_ISSOCK(inode->i_mode)) + goto path_put; + + sk = unix_find_socket_byinode(inode); + if (!sk) + goto path_put; + + err = -EPROTOTYPE; + if (sk->sk_type == type) + touch_atime(&path); + else + goto sock_put; + + path_put(&path); + + return sk; + +sock_put: + sock_put(sk); +path_put: + path_put(&path); +fail: + return ERR_PTR(err); +} + +static struct sock *unix_find_abstract(struct net *net, + struct sockaddr_un *sunaddr, + int addr_len, int type) +{ + unsigned int hash = unix_abstract_hash(sunaddr, addr_len, type); + struct dentry *dentry; + struct sock *sk; + + sk = unix_find_socket_byname(net, sunaddr, addr_len, hash); + if (!sk) + return ERR_PTR(-ECONNREFUSED); + + dentry = unix_sk(sk)->path.dentry; + if (dentry) + touch_atime(&unix_sk(sk)->path); + + return sk; +} + +static struct sock *unix_find_other(struct net *net, + struct sockaddr_un *sunaddr, + int addr_len, int type) +{ + struct sock *sk; + + if (sunaddr->sun_path[0]) + sk = unix_find_bsd(sunaddr, addr_len, type); + else + sk = unix_find_abstract(net, sunaddr, addr_len, type); + + return sk; +} + +static int unix_autobind(struct sock *sk) +{ + unsigned int new_hash, old_hash = sk->sk_hash; struct unix_sock *u = unix_sk(sk); - static u32 ordernum = 1; + struct net *net = sock_net(sk); struct unix_address *addr; + u32 lastnum, ordernum; int err; - unsigned int retries = 0; err = mutex_lock_interruptible(&u->bindlock); if (err) return err; - err = 0; if (u->addr) goto out; err = -ENOMEM; - addr = kzalloc(sizeof(*addr) + sizeof(short) + 16, GFP_KERNEL); + addr = kzalloc(sizeof(*addr) + + offsetof(struct sockaddr_un, sun_path) + 16, GFP_KERNEL); if (!addr) goto out; + addr->len = offsetof(struct sockaddr_un, sun_path) + 6; addr->name->sun_family = AF_UNIX; refcount_set(&addr->refcnt, 1); + ordernum = get_random_u32(); + lastnum = ordernum & 0xFFFFF; retry: - addr->len = sprintf(addr->name->sun_path+1, "%05x", ordernum) + 1 + sizeof(short); - addr->hash = unix_hash_fold(csum_partial(addr->name, addr->len, 0)); + ordernum = (ordernum + 1) & 0xFFFFF; + sprintf(addr->name->sun_path + 1, "%05x", ordernum); - spin_lock(&unix_table_lock); - ordernum = (ordernum+1)&0xFFFFF; + new_hash = unix_abstract_hash(addr->name, addr->len, sk->sk_type); + unix_table_double_lock(net, old_hash, new_hash); - if (__unix_find_socket_byname(net, addr->name, addr->len, sock->type, - addr->hash)) { - spin_unlock(&unix_table_lock); - /* - * __unix_find_socket_byname() may take long time if many names + if (__unix_find_socket_byname(net, addr->name, addr->len, new_hash)) { + unix_table_double_unlock(net, old_hash, new_hash); + + /* __unix_find_socket_byname() may take long time if many names * are already in use. */ cond_resched(); - /* Give up if all names seems to be in use. */ - if (retries++ == 0xFFFFF) { + + if (ordernum == lastnum) { + /* Give up if all names seems to be in use. */ err = -ENOSPC; - kfree(addr); + unix_release_addr(addr); goto out; } + goto retry; } - addr->hash ^= sk->sk_type; - __unix_remove_socket(sk); - smp_store_release(&u->addr, addr); - __unix_insert_socket(&unix_socket_table[addr->hash], sk); - spin_unlock(&unix_table_lock); + __unix_set_addr_hash(net, sk, addr, new_hash); + unix_table_double_unlock(net, old_hash, new_hash); err = 0; out: mutex_unlock(&u->bindlock); return err; } -static struct sock *unix_find_other(struct net *net, - struct sockaddr_un *sunname, int len, - int type, unsigned int hash, int *error) +static int unix_bind_bsd(struct sock *sk, struct sockaddr_un *sunaddr, + int addr_len) { - struct sock *u; - struct path path; - int err = 0; - - if (sunname->sun_path[0]) { - struct inode *inode; - err = kern_path(sunname->sun_path, LOOKUP_FOLLOW, &path); - if (err) - goto fail; - inode = d_backing_inode(path.dentry); - err = inode_permission(inode, MAY_WRITE); - if (err) - goto put_fail; - - err = -ECONNREFUSED; - if (!S_ISSOCK(inode->i_mode)) - goto put_fail; - u = unix_find_socket_byinode(inode); - if (!u) - goto put_fail; - - if (u->sk_type == type) - touch_atime(&path); + umode_t mode = S_IFSOCK | + (SOCK_INODE(sk->sk_socket)->i_mode & ~current_umask()); + unsigned int new_hash, old_hash = sk->sk_hash; + struct unix_sock *u = unix_sk(sk); + struct net *net = sock_net(sk); + struct user_namespace *ns; // barf... + struct unix_address *addr; + struct dentry *dentry; + struct path parent; + int err; - path_put(&path); + unix_mkname_bsd(sunaddr, addr_len); + addr_len = strlen(sunaddr->sun_path) + + offsetof(struct sockaddr_un, sun_path) + 1; - err = -EPROTOTYPE; - if (u->sk_type != type) { - sock_put(u); - goto fail; - } - } else { - err = -ECONNREFUSED; - u = unix_find_socket_byname(net, sunname, len, type, hash); - if (u) { - struct dentry *dentry; - dentry = unix_sk(u)->path.dentry; - if (dentry) - touch_atime(&unix_sk(u)->path); - } else - goto fail; - } - return u; - -put_fail: - path_put(&path); -fail: - *error = err; - return NULL; -} + addr = unix_create_addr(sunaddr, addr_len); + if (!addr) + return -ENOMEM; -static int unix_mknod(const char *sun_path, umode_t mode, struct path *res) -{ - struct dentry *dentry; - struct path path; - int err = 0; /* * Get the parent directory, calculate the hash for last * component. */ - dentry = kern_path_create(AT_FDCWD, sun_path, &path, 0); - err = PTR_ERR(dentry); - if (IS_ERR(dentry)) - return err; + dentry = kern_path_create(AT_FDCWD, addr->name->sun_path, &parent, 0); + if (IS_ERR(dentry)) { + err = PTR_ERR(dentry); + goto out; + } /* * All right, let's create it. */ - err = security_path_mknod(&path, dentry, mode, 0); - if (!err) { - err = vfs_mknod(d_inode(path.dentry), dentry, mode, 0); - if (!err) { - res->mnt = mntget(path.mnt); - res->dentry = dget(dentry); - } - } - done_path_create(&path, dentry); - return err; + ns = mnt_user_ns(parent.mnt); + err = security_path_mknod(&parent, dentry, mode, 0); + if (!err) + err = vfs_mknod(ns, d_inode(parent.dentry), dentry, mode, 0); + if (err) + goto out_path; + err = mutex_lock_interruptible(&u->bindlock); + if (err) + goto out_unlink; + if (u->addr) + goto out_unlock; + + new_hash = unix_bsd_hash(d_backing_inode(dentry)); + unix_table_double_lock(net, old_hash, new_hash); + u->path.mnt = mntget(parent.mnt); + u->path.dentry = dget(dentry); + __unix_set_addr_hash(net, sk, addr, new_hash); + unix_table_double_unlock(net, old_hash, new_hash); + unix_insert_bsd_socket(sk); + mutex_unlock(&u->bindlock); + done_path_create(&parent, dentry); + return 0; + +out_unlock: + mutex_unlock(&u->bindlock); + err = -EINVAL; +out_unlink: + /* failed after successful mknod? unlink what we'd created... */ + vfs_unlink(ns, d_inode(parent.dentry), dentry, NULL); +out_path: + done_path_create(&parent, dentry); +out: + unix_release_addr(addr); + return err == -EEXIST ? -EADDRINUSE : err; } -static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) +static int unix_bind_abstract(struct sock *sk, struct sockaddr_un *sunaddr, + int addr_len) { - struct sock *sk = sock->sk; - struct net *net = sock_net(sk); + unsigned int new_hash, old_hash = sk->sk_hash; struct unix_sock *u = unix_sk(sk); - struct sockaddr_un *sunaddr = (struct sockaddr_un *)uaddr; - char *sun_path = sunaddr->sun_path; - int err; - unsigned int hash; + struct net *net = sock_net(sk); struct unix_address *addr; - struct hlist_head *list; - struct path path = { }; - - err = -EINVAL; - if (addr_len < offsetofend(struct sockaddr_un, sun_family) || - sunaddr->sun_family != AF_UNIX) - goto out; + int err; - if (addr_len == sizeof(short)) { - err = unix_autobind(sock); - goto out; - } + addr = unix_create_addr(sunaddr, addr_len); + if (!addr) + return -ENOMEM; - err = unix_mkname(sunaddr, addr_len, &hash); - if (err < 0) + err = mutex_lock_interruptible(&u->bindlock); + if (err) goto out; - addr_len = err; - if (sun_path[0]) { - umode_t mode = S_IFSOCK | - (SOCK_INODE(sock)->i_mode & ~current_umask()); - err = unix_mknod(sun_path, mode, &path); - if (err) { - if (err == -EEXIST) - err = -EADDRINUSE; - goto out; - } + if (u->addr) { + err = -EINVAL; + goto out_mutex; } - err = mutex_lock_interruptible(&u->bindlock); - if (err) - goto out_put; + new_hash = unix_abstract_hash(addr->name, addr->len, sk->sk_type); + unix_table_double_lock(net, old_hash, new_hash); - err = -EINVAL; - if (u->addr) - goto out_up; + if (__unix_find_socket_byname(net, addr->name, addr->len, new_hash)) + goto out_spin; - err = -ENOMEM; - addr = kmalloc(sizeof(*addr)+addr_len, GFP_KERNEL); - if (!addr) - goto out_up; - - memcpy(addr->name, sunaddr, addr_len); - addr->len = addr_len; - addr->hash = hash ^ sk->sk_type; - refcount_set(&addr->refcnt, 1); + __unix_set_addr_hash(net, sk, addr, new_hash); + unix_table_double_unlock(net, old_hash, new_hash); + mutex_unlock(&u->bindlock); + return 0; - if (sun_path[0]) { - addr->hash = UNIX_HASH_SIZE; - hash = d_backing_inode(path.dentry)->i_ino & (UNIX_HASH_SIZE - 1); - spin_lock(&unix_table_lock); - u->path = path; - list = &unix_socket_table[hash]; - } else { - spin_lock(&unix_table_lock); - err = -EADDRINUSE; - if (__unix_find_socket_byname(net, sunaddr, addr_len, - sk->sk_type, hash)) { - unix_release_addr(addr); - goto out_unlock; - } +out_spin: + unix_table_double_unlock(net, old_hash, new_hash); + err = -EADDRINUSE; +out_mutex: + mutex_unlock(&u->bindlock); +out: + unix_release_addr(addr); + return err; +} - list = &unix_socket_table[addr->hash]; - } +static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) +{ + struct sockaddr_un *sunaddr = (struct sockaddr_un *)uaddr; + struct sock *sk = sock->sk; + int err; - err = 0; - __unix_remove_socket(sk); - smp_store_release(&u->addr, addr); - __unix_insert_socket(list, sk); + if (addr_len == offsetof(struct sockaddr_un, sun_path) && + sunaddr->sun_family == AF_UNIX) + return unix_autobind(sk); -out_unlock: - spin_unlock(&unix_table_lock); -out_up: - mutex_unlock(&u->bindlock); -out_put: + err = unix_validate_addr(sunaddr, addr_len); if (err) - path_put(&path); -out: + return err; + + if (sunaddr->sun_path[0]) + err = unix_bind_bsd(sk, sunaddr, addr_len); + else + err = unix_bind_abstract(sk, sunaddr, addr_len); + return err; } @@ -1133,11 +1346,9 @@ static void unix_state_double_unlock(struct sock *sk1, struct sock *sk2) static int unix_dgram_connect(struct socket *sock, struct sockaddr *addr, int alen, int flags) { - struct sock *sk = sock->sk; - struct net *net = sock_net(sk); struct sockaddr_un *sunaddr = (struct sockaddr_un *)addr; + struct sock *sk = sock->sk; struct sock *other; - unsigned int hash; int err; err = -EINVAL; @@ -1145,19 +1356,23 @@ static int unix_dgram_connect(struct socket *sock, struct sockaddr *addr, goto out; if (addr->sa_family != AF_UNSPEC) { - err = unix_mkname(sunaddr, alen, &hash); - if (err < 0) + err = unix_validate_addr(sunaddr, alen); + if (err) goto out; - alen = err; if (test_bit(SOCK_PASSCRED, &sock->flags) && - !unix_sk(sk)->addr && (err = unix_autobind(sock)) != 0) - goto out; + !unix_sk(sk)->addr) { + err = unix_autobind(sk); + if (err) + goto out; + } restart: - other = unix_find_other(net, sunaddr, alen, sock->type, hash, &err); - if (!other) + other = unix_find_other(sock_net(sk), sunaddr, alen, sock->type); + if (IS_ERR(other)) { + err = PTR_ERR(other); goto out; + } unix_state_double_lock(sk, other); @@ -1176,6 +1391,7 @@ restart: if (err) goto out_unlock; + sk->sk_state = other->sk_state = TCP_ESTABLISHED; } else { /* * 1003.1g breaking connected state with AF_UNSPEC @@ -1189,7 +1405,10 @@ restart: */ if (unix_peer(sk)) { struct sock *old_peer = unix_peer(sk); + unix_peer(sk) = other; + if (!other) + sk->sk_state = TCP_CLOSE; unix_dgram_peer_wake_disconnect_wakeup(sk, old_peer); unix_state_double_unlock(sk, other); @@ -1201,6 +1420,7 @@ restart: unix_peer(sk) = other; unix_state_double_unlock(sk, other); } + return 0; out_unlock: @@ -1211,6 +1431,7 @@ out: } static long unix_wait_for_peer(struct sock *other, long timeo) + __releases(&unix_sk(other)->lock) { struct unix_sock *u = unix_sk(other); int sched; @@ -1235,25 +1456,23 @@ static int unix_stream_connect(struct socket *sock, struct sockaddr *uaddr, int addr_len, int flags) { struct sockaddr_un *sunaddr = (struct sockaddr_un *)uaddr; - struct sock *sk = sock->sk; - struct net *net = sock_net(sk); + struct sock *sk = sock->sk, *newsk = NULL, *other = NULL; struct unix_sock *u = unix_sk(sk), *newu, *otheru; - struct sock *newsk = NULL; - struct sock *other = NULL; + struct net *net = sock_net(sk); struct sk_buff *skb = NULL; - unsigned int hash; - int st; - int err; long timeo; + int err; + int st; - err = unix_mkname(sunaddr, addr_len, &hash); - if (err < 0) + err = unix_validate_addr(sunaddr, addr_len); + if (err) goto out; - addr_len = err; - if (test_bit(SOCK_PASSCRED, &sock->flags) && !u->addr && - (err = unix_autobind(sock)) != 0) - goto out; + if (test_bit(SOCK_PASSCRED, &sock->flags) && !u->addr) { + err = unix_autobind(sk); + if (err) + goto out; + } timeo = sock_sndtimeo(sk, flags & O_NONBLOCK); @@ -1262,12 +1481,15 @@ static int unix_stream_connect(struct socket *sock, struct sockaddr *uaddr, we will have to recheck all again in any case. */ - err = -ENOMEM; - /* create new sock for complete connection */ - newsk = unix_create1(sock_net(sk), NULL, 0); - if (newsk == NULL) + newsk = unix_create1(net, NULL, 0, sock->type); + if (IS_ERR(newsk)) { + err = PTR_ERR(newsk); + newsk = NULL; goto out; + } + + err = -ENOMEM; /* Allocate skb for sending to listening sock */ skb = sock_wmalloc(newsk, 1, 0, GFP_KERNEL); @@ -1276,9 +1498,12 @@ static int unix_stream_connect(struct socket *sock, struct sockaddr *uaddr, restart: /* Find listening sock. */ - other = unix_find_other(net, sunaddr, addr_len, sk->sk_type, hash, &err); - if (!other) + other = unix_find_other(net, sunaddr, addr_len, sk->sk_type); + if (IS_ERR(other)) { + err = PTR_ERR(other); + other = NULL; goto out; + } /* Latch state of peer */ unix_state_lock(other); @@ -1366,9 +1591,9 @@ restart: * * The contents of *(otheru->addr) and otheru->path * are seen fully set up here, since we have found - * otheru in hash under unix_table_lock. Insertion - * into the hash chain we'd found it in had been done - * in an earlier critical area protected by unix_table_lock, + * otheru in hash under its lock. Insertion into the + * hash chain we'd found it in had been done in an + * earlier critical area protected by the chain's lock, * the same one where we'd set *(otheru->addr) contents, * as well as otheru->path and otheru->addr itself. * @@ -1398,7 +1623,7 @@ restart: unix_state_unlock(sk); - /* take ten and and send info to listening sock */ + /* take ten and send info to listening sock */ spin_lock(&other->sk_receive_queue.lock); __skb_queue_tail(&other->sk_receive_queue, skb); spin_unlock(&other->sk_receive_queue.lock); @@ -1432,12 +1657,10 @@ static int unix_socketpair(struct socket *socka, struct socket *sockb) init_peercred(ska); init_peercred(skb); - if (ska->sk_type != SOCK_DGRAM) { - ska->sk_state = TCP_ESTABLISHED; - skb->sk_state = TCP_ESTABLISHED; - socka->state = SS_CONNECTED; - sockb->state = SS_CONNECTED; - } + ska->sk_state = TCP_ESTABLISHED; + skb->sk_state = TCP_ESTABLISHED; + socka->state = SS_CONNECTED; + sockb->state = SS_CONNECTED; return 0; } @@ -1470,7 +1693,8 @@ static int unix_accept(struct socket *sock, struct socket *newsock, int flags, * so that no locks are necessary. */ - skb = skb_recv_datagram(sk, 0, flags&O_NONBLOCK, &err); + skb = skb_recv_datagram(sk, (flags & O_NONBLOCK) ? MSG_DONTWAIT : 0, + &err); if (!skb) { /* This means receive shutdown. */ if (err == 0) @@ -1517,7 +1741,7 @@ static int unix_getname(struct socket *sock, struct sockaddr *uaddr, int peer) if (!addr) { sunaddr->sun_family = AF_UNIX; sunaddr->sun_path[0] = 0; - err = sizeof(short); + err = offsetof(struct sockaddr_un, sun_path); } else { err = addr->len; memcpy(sunaddr, addr->name, addr->len); @@ -1527,6 +1751,53 @@ out: return err; } +static void unix_peek_fds(struct scm_cookie *scm, struct sk_buff *skb) +{ + scm->fp = scm_fp_dup(UNIXCB(skb).fp); + + /* + * Garbage collection of unix sockets starts by selecting a set of + * candidate sockets which have reference only from being in flight + * (total_refs == inflight_refs). This condition is checked once during + * the candidate collection phase, and candidates are marked as such, so + * that non-candidates can later be ignored. While inflight_refs is + * protected by unix_gc_lock, total_refs (file count) is not, hence this + * is an instantaneous decision. + * + * Once a candidate, however, the socket must not be reinstalled into a + * file descriptor while the garbage collection is in progress. + * + * If the above conditions are met, then the directed graph of + * candidates (*) does not change while unix_gc_lock is held. + * + * Any operations that changes the file count through file descriptors + * (dup, close, sendmsg) does not change the graph since candidates are + * not installed in fds. + * + * Dequeing a candidate via recvmsg would install it into an fd, but + * that takes unix_gc_lock to decrement the inflight count, so it's + * serialized with garbage collection. + * + * MSG_PEEK is special in that it does not change the inflight count, + * yet does install the socket into an fd. The following lock/unlock + * pair is to ensure serialization with garbage collection. It must be + * done between incrementing the file count and installing the file into + * an fd. + * + * If garbage collection starts after the barrier provided by the + * lock/unlock, then it will see the elevated refcount and not mark this + * as a candidate. If a garbage collection is already in progress + * before the file count was incremented, then the lock/unlock pair will + * ensure that garbage collection is finished before progressing to + * installing the fd. + * + * (*) A -> B where B is on the queue of A or B is on the queue of C + * which is on the queue of listening socket A. + */ + spin_lock(&unix_gc_lock); + spin_unlock(&unix_gc_lock); +} + static int unix_scm_to_skb(struct scm_cookie *scm, struct sk_buff *skb, bool send_fds) { int err = 0; @@ -1588,11 +1859,9 @@ static int maybe_init_creds(struct scm_cookie *scm, static bool unix_skb_scm_eq(struct sk_buff *skb, struct scm_cookie *scm) { - const struct unix_skb_parms *u = &UNIXCB(skb); - - return u->pid == scm->pid && - uid_eq(u->uid, scm->creds.uid) && - gid_eq(u->gid, scm->creds.gid) && + return UNIXCB(skb).pid == scm->pid && + uid_eq(UNIXCB(skb).uid, scm->creds.uid) && + gid_eq(UNIXCB(skb).gid, scm->creds.gid) && unix_secdata_eq(scm, skb); } @@ -1601,10 +1870,8 @@ static void scm_stat_add(struct sock *sk, struct sk_buff *skb) struct scm_fp_list *fp = UNIXCB(skb).fp; struct unix_sock *u = unix_sk(sk); - lockdep_assert_held(&sk->sk_receive_queue.lock); - if (unlikely(fp && fp->count)) - u->scm_stat.nr_fds += fp->count; + atomic_add(fp->count, &u->scm_stat.nr_fds); } static void scm_stat_del(struct sock *sk, struct sk_buff *skb) @@ -1612,10 +1879,8 @@ static void scm_stat_del(struct sock *sk, struct sk_buff *skb) struct scm_fp_list *fp = UNIXCB(skb).fp; struct unix_sock *u = unix_sk(sk); - lockdep_assert_held(&sk->sk_receive_queue.lock); - if (unlikely(fp && fp->count)) - u->scm_stat.nr_fds -= fp->count; + atomic_sub(fp->count, &u->scm_stat.nr_fds); } /* @@ -1625,19 +1890,15 @@ static void scm_stat_del(struct sock *sk, struct sk_buff *skb) static int unix_dgram_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) { - struct sock *sk = sock->sk; - struct net *net = sock_net(sk); - struct unix_sock *u = unix_sk(sk); DECLARE_SOCKADDR(struct sockaddr_un *, sunaddr, msg->msg_name); - struct sock *other = NULL; - int namelen = 0; /* fake GCC */ - int err; - unsigned int hash; - struct sk_buff *skb; - long timeo; + struct sock *sk = sock->sk, *other = NULL; + struct unix_sock *u = unix_sk(sk); struct scm_cookie scm; + struct sk_buff *skb; int data_len = 0; int sk_locked; + long timeo; + int err; wait_for_unix_gc(); err = scm_send(sock, msg, &scm, false); @@ -1649,10 +1910,9 @@ static int unix_dgram_sendmsg(struct socket *sock, struct msghdr *msg, goto out; if (msg->msg_namelen) { - err = unix_mkname(sunaddr, msg->msg_namelen, &hash); - if (err < 0) + err = unix_validate_addr(sunaddr, msg->msg_namelen); + if (err) goto out; - namelen = err; } else { sunaddr = NULL; err = -ENOTCONN; @@ -1661,9 +1921,11 @@ static int unix_dgram_sendmsg(struct socket *sock, struct msghdr *msg, goto out; } - if (test_bit(SOCK_PASSCRED, &sock->flags) && !u->addr - && (err = unix_autobind(sock)) != 0) - goto out; + if (test_bit(SOCK_PASSCRED, &sock->flags) && !u->addr) { + err = unix_autobind(sk); + if (err) + goto out; + } err = -EMSGSIZE; if (len > sk->sk_sndbuf - 32) @@ -1703,10 +1965,13 @@ restart: if (sunaddr == NULL) goto out_free; - other = unix_find_other(net, sunaddr, namelen, sk->sk_type, - hash, &err); - if (other == NULL) + other = unix_find_other(sock_net(sk), sunaddr, msg->msg_namelen, + sk->sk_type); + if (IS_ERR(other)) { + err = PTR_ERR(other); + other = NULL; goto out_free; + } } if (sk_filter(other, skb) < 0) { @@ -1740,6 +2005,7 @@ restart_locked: unix_state_unlock(sk); + sk->sk_state = TCP_CLOSE; unix_dgram_disconnected(sk, other); sock_put(other); err = -ECONNREFUSED; @@ -1804,10 +2070,8 @@ restart_locked: if (sock_flag(other, SOCK_RCVTSTAMP)) __net_timestamp(skb); maybe_add_creds(skb, sock, other); - spin_lock(&other->sk_receive_queue.lock); scm_stat_add(other, skb); - __skb_queue_tail(&other->sk_receive_queue, skb); - spin_unlock(&other->sk_receive_queue.lock); + skb_queue_tail(&other->sk_receive_queue, skb); unix_state_unlock(other); other->sk_data_ready(other); sock_put(other); @@ -1832,6 +2096,53 @@ out: */ #define UNIX_SKB_FRAGS_SZ (PAGE_SIZE << get_order(32768)) +#if IS_ENABLED(CONFIG_AF_UNIX_OOB) +static int queue_oob(struct socket *sock, struct msghdr *msg, struct sock *other) +{ + struct unix_sock *ousk = unix_sk(other); + struct sk_buff *skb; + int err = 0; + + skb = sock_alloc_send_skb(sock->sk, 1, msg->msg_flags & MSG_DONTWAIT, &err); + + if (!skb) + return err; + + skb_put(skb, 1); + err = skb_copy_datagram_from_iter(skb, 0, &msg->msg_iter, 1); + + if (err) { + kfree_skb(skb); + return err; + } + + unix_state_lock(other); + + if (sock_flag(other, SOCK_DEAD) || + (other->sk_shutdown & RCV_SHUTDOWN)) { + unix_state_unlock(other); + kfree_skb(skb); + return -EPIPE; + } + + maybe_add_creds(skb, sock, other); + skb_get(skb); + + if (ousk->oob_skb) + consume_skb(ousk->oob_skb); + + WRITE_ONCE(ousk->oob_skb, skb); + + scm_stat_add(other, skb); + skb_queue_tail(&other->sk_receive_queue, skb); + sk_send_sigurg(other); + unix_state_unlock(other); + other->sk_data_ready(other); + + return err; +} +#endif + static int unix_stream_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) { @@ -1850,8 +2161,14 @@ static int unix_stream_sendmsg(struct socket *sock, struct msghdr *msg, return err; err = -EOPNOTSUPP; - if (msg->msg_flags&MSG_OOB) - goto out_err; + if (msg->msg_flags & MSG_OOB) { +#if IS_ENABLED(CONFIG_AF_UNIX_OOB) + if (len) + len--; + else +#endif + goto out_err; + } if (msg->msg_namelen) { err = sk->sk_state == TCP_ESTABLISHED ? -EISCONN : -EOPNOTSUPP; @@ -1909,15 +2226,22 @@ static int unix_stream_sendmsg(struct socket *sock, struct msghdr *msg, goto pipe_err_free; maybe_add_creds(skb, sock, other); - spin_lock(&other->sk_receive_queue.lock); scm_stat_add(other, skb); - __skb_queue_tail(&other->sk_receive_queue, skb); - spin_unlock(&other->sk_receive_queue.lock); + skb_queue_tail(&other->sk_receive_queue, skb); unix_state_unlock(other); other->sk_data_ready(other); sent += size; } +#if IS_ENABLED(CONFIG_AF_UNIX_OOB) + if (msg->msg_flags & MSG_OOB) { + err = queue_oob(sock, msg, other); + if (err) + goto out_err; + sent++; + } +#endif + scm_destroy(&scm); return sent; @@ -2090,11 +2414,11 @@ static void unix_copy_addr(struct msghdr *msg, struct sock *sk) } } -static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg, - size_t size, int flags) +int __unix_dgram_recvmsg(struct sock *sk, struct msghdr *msg, size_t size, + int flags) { struct scm_cookie scm; - struct sock *sk = sock->sk; + struct socket *sock = sk->sk_socket; struct unix_sock *u = unix_sk(sk); struct sk_buff *skb, *last; long timeo; @@ -2112,9 +2436,12 @@ static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg, skip = sk_peek_offset(sk, flags); skb = __skb_try_recv_datagram(sk, &sk->sk_receive_queue, flags, - scm_stat_del, &skip, &err, &last); - if (skb) + &skip, &err, &last); + if (skb) { + if (!(flags & MSG_PEEK)) + scm_stat_del(sk, skb); break; + } mutex_unlock(&u->iolock); @@ -2181,7 +2508,7 @@ static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg, sk_peek_offset_fwd(sk, size); if (UNIXCB(skb).fp) - scm.fp = scm_fp_dup(UNIXCB(skb).fp); + unix_peek_fds(&scm, skb); } err = (flags & MSG_TRUNC) ? skb->len - skip : size; @@ -2194,6 +2521,38 @@ out: return err; } +static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, + int flags) +{ + struct sock *sk = sock->sk; + +#ifdef CONFIG_BPF_SYSCALL + const struct proto *prot = READ_ONCE(sk->sk_prot); + + if (prot != &unix_dgram_proto) + return prot->recvmsg(sk, msg, size, flags, NULL); +#endif + return __unix_dgram_recvmsg(sk, msg, size, flags); +} + +static int unix_read_skb(struct sock *sk, skb_read_actor_t recv_actor) +{ + struct unix_sock *u = unix_sk(sk); + struct sk_buff *skb; + int err, copied; + + mutex_lock(&u->iolock); + skb = skb_recv_datagram(sk, MSG_DONTWAIT, &err); + mutex_unlock(&u->iolock); + if (!skb) + return err; + + copied = recv_actor(sk, skb); + kfree_skb(skb); + + return copied; +} + /* * Sleep until more data has arrived. But check for races.. */ @@ -2201,13 +2560,14 @@ static long unix_stream_data_wait(struct sock *sk, long timeo, struct sk_buff *last, unsigned int last_len, bool freezable) { + unsigned int state = TASK_INTERRUPTIBLE | freezable * TASK_FREEZABLE; struct sk_buff *tail; DEFINE_WAIT(wait); unix_state_lock(sk); for (;;) { - prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); + prepare_to_wait(sk_sleep(sk), &wait, state); tail = skb_peek_tail(&sk->sk_receive_queue); if (tail != last || @@ -2220,10 +2580,7 @@ static long unix_stream_data_wait(struct sock *sk, long timeo, sk_set_bit(SOCKWQ_ASYNC_WAITDATA, sk); unix_state_unlock(sk); - if (freezable) - timeo = freezable_schedule_timeout(timeo); - else - timeo = schedule_timeout(timeo); + timeo = schedule_timeout(timeo); unix_state_lock(sk); if (sock_flag(sk, SOCK_DEAD)) @@ -2253,6 +2610,84 @@ struct unix_stream_read_state { unsigned int splice_flags; }; +#if IS_ENABLED(CONFIG_AF_UNIX_OOB) +static int unix_stream_recv_urg(struct unix_stream_read_state *state) +{ + struct socket *sock = state->socket; + struct sock *sk = sock->sk; + struct unix_sock *u = unix_sk(sk); + int chunk = 1; + struct sk_buff *oob_skb; + + mutex_lock(&u->iolock); + unix_state_lock(sk); + + if (sock_flag(sk, SOCK_URGINLINE) || !u->oob_skb) { + unix_state_unlock(sk); + mutex_unlock(&u->iolock); + return -EINVAL; + } + + oob_skb = u->oob_skb; + + if (!(state->flags & MSG_PEEK)) + WRITE_ONCE(u->oob_skb, NULL); + + unix_state_unlock(sk); + + chunk = state->recv_actor(oob_skb, 0, chunk, state); + + if (!(state->flags & MSG_PEEK)) { + UNIXCB(oob_skb).consumed += 1; + kfree_skb(oob_skb); + } + + mutex_unlock(&u->iolock); + + if (chunk < 0) + return -EFAULT; + + state->msg->msg_flags |= MSG_OOB; + return 1; +} + +static struct sk_buff *manage_oob(struct sk_buff *skb, struct sock *sk, + int flags, int copied) +{ + struct unix_sock *u = unix_sk(sk); + + if (!unix_skb_len(skb) && !(flags & MSG_PEEK)) { + skb_unlink(skb, &sk->sk_receive_queue); + consume_skb(skb); + skb = NULL; + } else { + if (skb == u->oob_skb) { + if (copied) { + skb = NULL; + } else if (sock_flag(sk, SOCK_URGINLINE)) { + if (!(flags & MSG_PEEK)) { + WRITE_ONCE(u->oob_skb, NULL); + consume_skb(skb); + } + } else if (!(flags & MSG_PEEK)) { + skb_unlink(skb, &sk->sk_receive_queue); + consume_skb(skb); + skb = skb_peek(&sk->sk_receive_queue); + } + } + } + return skb; +} +#endif + +static int unix_stream_read_skb(struct sock *sk, skb_read_actor_t recv_actor) +{ + if (unlikely(sk->sk_state != TCP_ESTABLISHED)) + return -ENOTCONN; + + return unix_read_skb(sk, recv_actor); +} + static int unix_stream_read_generic(struct unix_stream_read_state *state, bool freezable) { @@ -2278,6 +2713,9 @@ static int unix_stream_read_generic(struct unix_stream_read_state *state, if (unlikely(flags & MSG_OOB)) { err = -EOPNOTSUPP; +#if IS_ENABLED(CONFIG_AF_UNIX_OOB) + err = unix_stream_recv_urg(state); +#endif goto out; } @@ -2306,6 +2744,18 @@ redo: } last = skb = skb_peek(&sk->sk_receive_queue); last_len = last ? last->len : 0; + +#if IS_ENABLED(CONFIG_AF_UNIX_OOB) + if (skb) { + skb = manage_oob(skb, sk, flags, copied); + if (!skb) { + unix_state_unlock(sk); + if (copied) + break; + goto redo; + } + } +#endif again: if (skb == NULL) { if (copied >= target) @@ -2408,9 +2858,7 @@ unlock: sk_peek_offset_bwd(sk, chunk); if (UNIXCB(skb).fp) { - spin_lock(&sk->sk_receive_queue.lock); scm_stat_del(sk, skb); - spin_unlock(&sk->sk_receive_queue.lock); unix_detach_fds(&scm, skb); } @@ -2426,7 +2874,7 @@ unlock: /* It is questionable, see note in unix_dgram_recvmsg. */ if (UNIXCB(skb).fp) - scm.fp = scm_fp_dup(UNIXCB(skb).fp); + unix_peek_fds(&scm, skb); sk_peek_offset_fwd(sk, chunk); @@ -2465,6 +2913,20 @@ static int unix_stream_read_actor(struct sk_buff *skb, return ret ?: chunk; } +int __unix_stream_recvmsg(struct sock *sk, struct msghdr *msg, + size_t size, int flags) +{ + struct unix_stream_read_state state = { + .recv_actor = unix_stream_read_actor, + .socket = sk->sk_socket, + .msg = msg, + .size = size, + .flags = flags + }; + + return unix_stream_read_generic(&state, true); +} + static int unix_stream_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, int flags) { @@ -2476,6 +2938,13 @@ static int unix_stream_recvmsg(struct socket *sock, struct msghdr *msg, .flags = flags }; +#ifdef CONFIG_BPF_SYSCALL + struct sock *sk = sock->sk; + const struct proto *prot = READ_ONCE(sk->sk_prot); + + if (prot != &unix_stream_proto) + return prot->recvmsg(sk, msg, size, flags, NULL); +#endif return unix_stream_read_generic(&state, true); } @@ -2536,7 +3005,10 @@ static int unix_shutdown(struct socket *sock, int mode) (sk->sk_type == SOCK_STREAM || sk->sk_type == SOCK_SEQPACKET)) { int peer_mode = 0; + const struct proto *prot = READ_ONCE(other->sk_prot); + if (prot->unhash) + prot->unhash(other); if (mode&RCV_SHUTDOWN) peer_mode |= SEND_SHUTDOWN; if (mode&SEND_SHUTDOWN) @@ -2643,6 +3115,19 @@ static int unix_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) case SIOCUNIXFILE: err = unix_open_file(sk); break; +#if IS_ENABLED(CONFIG_AF_UNIX_OOB) + case SIOCATMARK: + { + struct sk_buff *skb; + int answ = 0; + + skb = skb_peek(&sk->sk_receive_queue); + if (skb && skb == READ_ONCE(unix_sk(sk)->oob_skb)) + answ = 1; + err = put_user(answ, (int __user *)arg); + } + break; +#endif default: err = -ENOIOCTLCMD; break; @@ -2676,6 +3161,12 @@ static __poll_t unix_poll(struct file *file, struct socket *sock, poll_table *wa /* readable? */ if (!skb_queue_empty_lockless(&sk->sk_receive_queue)) mask |= EPOLLIN | EPOLLRDNORM; + if (sk_is_readable(sk)) + mask |= EPOLLIN | EPOLLRDNORM; +#if IS_ENABLED(CONFIG_AF_UNIX_OOB) + if (READ_ONCE(unix_sk(sk)->oob_skb)) + mask |= EPOLLPRI; +#endif /* Connection-based need to check for termination and startup */ if ((sk->sk_type == SOCK_STREAM || sk->sk_type == SOCK_SEQPACKET) && @@ -2715,6 +3206,8 @@ static __poll_t unix_dgram_poll(struct file *file, struct socket *sock, /* readable? */ if (!skb_queue_empty_lockless(&sk->sk_receive_queue)) mask |= EPOLLIN | EPOLLRDNORM; + if (sk_is_readable(sk)) + mask |= EPOLLIN | EPOLLRDNORM; /* Connection-based need to check for termination and startup */ if (sk->sk_type == SOCK_SEQPACKET) { @@ -2735,7 +3228,7 @@ static __poll_t unix_dgram_poll(struct file *file, struct socket *sock, other = unix_peer(sk); if (other && unix_peer(other) != sk && - unix_recvq_full(other) && + unix_recvq_full_lockless(other) && unix_dgram_peer_wake_me(sk, other)) writable = 0; @@ -2755,19 +3248,18 @@ static __poll_t unix_dgram_poll(struct file *file, struct socket *sock, #define BUCKET_SPACE (BITS_PER_LONG - (UNIX_HASH_BITS + 1) - 1) #define get_bucket(x) ((x) >> BUCKET_SPACE) -#define get_offset(x) ((x) & ((1L << BUCKET_SPACE) - 1)) +#define get_offset(x) ((x) & ((1UL << BUCKET_SPACE) - 1)) #define set_bucket_offset(b, o) ((b) << BUCKET_SPACE | (o)) static struct sock *unix_from_bucket(struct seq_file *seq, loff_t *pos) { unsigned long offset = get_offset(*pos); unsigned long bucket = get_bucket(*pos); - struct sock *sk; unsigned long count = 0; + struct sock *sk; - for (sk = sk_head(&unix_socket_table[bucket]); sk; sk = sk_next(sk)) { - if (sock_net(sk) != seq_file_net(seq)) - continue; + for (sk = sk_head(&seq_file_net(seq)->unx.table.buckets[bucket]); + sk; sk = sk_next(sk)) { if (++count == offset) break; } @@ -2775,57 +3267,68 @@ static struct sock *unix_from_bucket(struct seq_file *seq, loff_t *pos) return sk; } -static struct sock *unix_next_socket(struct seq_file *seq, - struct sock *sk, - loff_t *pos) +static struct sock *unix_get_first(struct seq_file *seq, loff_t *pos) { - unsigned long bucket; + unsigned long bucket = get_bucket(*pos); + struct net *net = seq_file_net(seq); + struct sock *sk; - while (sk > (struct sock *)SEQ_START_TOKEN) { - sk = sk_next(sk); - if (!sk) - goto next_bucket; - if (sock_net(sk) == seq_file_net(seq)) - return sk; - } + while (bucket < UNIX_HASH_SIZE) { + spin_lock(&net->unx.table.locks[bucket]); - do { sk = unix_from_bucket(seq, pos); if (sk) return sk; -next_bucket: - bucket = get_bucket(*pos) + 1; - *pos = set_bucket_offset(bucket, 1); - } while (bucket < ARRAY_SIZE(unix_socket_table)); + spin_unlock(&net->unx.table.locks[bucket]); + + *pos = set_bucket_offset(++bucket, 1); + } return NULL; } -static void *unix_seq_start(struct seq_file *seq, loff_t *pos) - __acquires(unix_table_lock) +static struct sock *unix_get_next(struct seq_file *seq, struct sock *sk, + loff_t *pos) { - spin_lock(&unix_table_lock); + unsigned long bucket = get_bucket(*pos); + + sk = sk_next(sk); + if (sk) + return sk; + + spin_unlock(&seq_file_net(seq)->unx.table.locks[bucket]); + + *pos = set_bucket_offset(++bucket, 1); + + return unix_get_first(seq, pos); +} + +static void *unix_seq_start(struct seq_file *seq, loff_t *pos) +{ if (!*pos) return SEQ_START_TOKEN; - if (get_bucket(*pos) >= ARRAY_SIZE(unix_socket_table)) - return NULL; - - return unix_next_socket(seq, NULL, pos); + return unix_get_first(seq, pos); } static void *unix_seq_next(struct seq_file *seq, void *v, loff_t *pos) { ++*pos; - return unix_next_socket(seq, v, pos); + + if (v == SEQ_START_TOKEN) + return unix_get_first(seq, pos); + + return unix_get_next(seq, v, pos); } static void unix_seq_stop(struct seq_file *seq, void *v) - __releases(unix_table_lock) { - spin_unlock(&unix_table_lock); + struct sock *sk = v; + + if (sk) + spin_unlock(&seq_file_net(seq)->unx.table.locks[sk->sk_hash]); } static int unix_seq_show(struct seq_file *seq, void *v) @@ -2850,15 +3353,16 @@ static int unix_seq_show(struct seq_file *seq, void *v) (s->sk_state == TCP_ESTABLISHED ? SS_CONNECTING : SS_DISCONNECTING), sock_i_ino(s)); - if (u->addr) { // under unix_table_lock here + if (u->addr) { // under a hash table lock here int i, len; seq_putc(seq, ' '); i = 0; - len = u->addr->len - sizeof(short); - if (!UNIX_ABSTRACT(s)) + len = u->addr->len - + offsetof(struct sockaddr_un, sun_path); + if (u->addr->name->sun_path[0]) { len--; - else { + } else { seq_putc(seq, '@'); i++; } @@ -2879,6 +3383,203 @@ static const struct seq_operations unix_seq_ops = { .stop = unix_seq_stop, .show = unix_seq_show, }; + +#if IS_BUILTIN(CONFIG_UNIX) && defined(CONFIG_BPF_SYSCALL) +struct bpf_unix_iter_state { + struct seq_net_private p; + unsigned int cur_sk; + unsigned int end_sk; + unsigned int max_sk; + struct sock **batch; + bool st_bucket_done; +}; + +struct bpf_iter__unix { + __bpf_md_ptr(struct bpf_iter_meta *, meta); + __bpf_md_ptr(struct unix_sock *, unix_sk); + uid_t uid __aligned(8); +}; + +static int unix_prog_seq_show(struct bpf_prog *prog, struct bpf_iter_meta *meta, + struct unix_sock *unix_sk, uid_t uid) +{ + struct bpf_iter__unix ctx; + + meta->seq_num--; /* skip SEQ_START_TOKEN */ + ctx.meta = meta; + ctx.unix_sk = unix_sk; + ctx.uid = uid; + return bpf_iter_run_prog(prog, &ctx); +} + +static int bpf_iter_unix_hold_batch(struct seq_file *seq, struct sock *start_sk) + +{ + struct bpf_unix_iter_state *iter = seq->private; + unsigned int expected = 1; + struct sock *sk; + + sock_hold(start_sk); + iter->batch[iter->end_sk++] = start_sk; + + for (sk = sk_next(start_sk); sk; sk = sk_next(sk)) { + if (iter->end_sk < iter->max_sk) { + sock_hold(sk); + iter->batch[iter->end_sk++] = sk; + } + + expected++; + } + + spin_unlock(&seq_file_net(seq)->unx.table.locks[start_sk->sk_hash]); + + return expected; +} + +static void bpf_iter_unix_put_batch(struct bpf_unix_iter_state *iter) +{ + while (iter->cur_sk < iter->end_sk) + sock_put(iter->batch[iter->cur_sk++]); +} + +static int bpf_iter_unix_realloc_batch(struct bpf_unix_iter_state *iter, + unsigned int new_batch_sz) +{ + struct sock **new_batch; + + new_batch = kvmalloc(sizeof(*new_batch) * new_batch_sz, + GFP_USER | __GFP_NOWARN); + if (!new_batch) + return -ENOMEM; + + bpf_iter_unix_put_batch(iter); + kvfree(iter->batch); + iter->batch = new_batch; + iter->max_sk = new_batch_sz; + + return 0; +} + +static struct sock *bpf_iter_unix_batch(struct seq_file *seq, + loff_t *pos) +{ + struct bpf_unix_iter_state *iter = seq->private; + unsigned int expected; + bool resized = false; + struct sock *sk; + + if (iter->st_bucket_done) + *pos = set_bucket_offset(get_bucket(*pos) + 1, 1); + +again: + /* Get a new batch */ + iter->cur_sk = 0; + iter->end_sk = 0; + + sk = unix_get_first(seq, pos); + if (!sk) + return NULL; /* Done */ + + expected = bpf_iter_unix_hold_batch(seq, sk); + + if (iter->end_sk == expected) { + iter->st_bucket_done = true; + return sk; + } + + if (!resized && !bpf_iter_unix_realloc_batch(iter, expected * 3 / 2)) { + resized = true; + goto again; + } + + return sk; +} + +static void *bpf_iter_unix_seq_start(struct seq_file *seq, loff_t *pos) +{ + if (!*pos) + return SEQ_START_TOKEN; + + /* bpf iter does not support lseek, so it always + * continue from where it was stop()-ped. + */ + return bpf_iter_unix_batch(seq, pos); +} + +static void *bpf_iter_unix_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct bpf_unix_iter_state *iter = seq->private; + struct sock *sk; + + /* Whenever seq_next() is called, the iter->cur_sk is + * done with seq_show(), so advance to the next sk in + * the batch. + */ + if (iter->cur_sk < iter->end_sk) + sock_put(iter->batch[iter->cur_sk++]); + + ++*pos; + + if (iter->cur_sk < iter->end_sk) + sk = iter->batch[iter->cur_sk]; + else + sk = bpf_iter_unix_batch(seq, pos); + + return sk; +} + +static int bpf_iter_unix_seq_show(struct seq_file *seq, void *v) +{ + struct bpf_iter_meta meta; + struct bpf_prog *prog; + struct sock *sk = v; + uid_t uid; + bool slow; + int ret; + + if (v == SEQ_START_TOKEN) + return 0; + + slow = lock_sock_fast(sk); + + if (unlikely(sk_unhashed(sk))) { + ret = SEQ_SKIP; + goto unlock; + } + + uid = from_kuid_munged(seq_user_ns(seq), sock_i_uid(sk)); + meta.seq = seq; + prog = bpf_iter_get_info(&meta, false); + ret = unix_prog_seq_show(prog, &meta, v, uid); +unlock: + unlock_sock_fast(sk, slow); + return ret; +} + +static void bpf_iter_unix_seq_stop(struct seq_file *seq, void *v) +{ + struct bpf_unix_iter_state *iter = seq->private; + struct bpf_iter_meta meta; + struct bpf_prog *prog; + + if (!v) { + meta.seq = seq; + prog = bpf_iter_get_info(&meta, true); + if (prog) + (void)unix_prog_seq_show(prog, &meta, v, 0); + } + + if (iter->cur_sk < iter->end_sk) + bpf_iter_unix_put_batch(iter); +} + +static const struct seq_operations bpf_iter_unix_seq_ops = { + .start = bpf_iter_unix_seq_start, + .next = bpf_iter_unix_seq_next, + .stop = bpf_iter_unix_seq_stop, + .show = bpf_iter_unix_seq_show, +}; +#endif #endif static const struct net_proto_family unix_family_ops = { @@ -2890,7 +3591,7 @@ static const struct net_proto_family unix_family_ops = { static int __net_init unix_net_init(struct net *net) { - int error = -ENOMEM; + int i; net->unx.sysctl_max_dgram_qlen = 10; if (unix_sysctl_register(net)) @@ -2898,18 +3599,44 @@ static int __net_init unix_net_init(struct net *net) #ifdef CONFIG_PROC_FS if (!proc_create_net("unix", 0, net->proc_net, &unix_seq_ops, - sizeof(struct seq_net_private))) { - unix_sysctl_unregister(net); - goto out; + sizeof(struct seq_net_private))) + goto err_sysctl; +#endif + + net->unx.table.locks = kvmalloc_array(UNIX_HASH_SIZE, + sizeof(spinlock_t), GFP_KERNEL); + if (!net->unx.table.locks) + goto err_proc; + + net->unx.table.buckets = kvmalloc_array(UNIX_HASH_SIZE, + sizeof(struct hlist_head), + GFP_KERNEL); + if (!net->unx.table.buckets) + goto free_locks; + + for (i = 0; i < UNIX_HASH_SIZE; i++) { + spin_lock_init(&net->unx.table.locks[i]); + INIT_HLIST_HEAD(&net->unx.table.buckets[i]); } + + return 0; + +free_locks: + kvfree(net->unx.table.locks); +err_proc: +#ifdef CONFIG_PROC_FS + remove_proc_entry("unix", net->proc_net); +err_sysctl: #endif - error = 0; + unix_sysctl_unregister(net); out: - return error; + return -ENOMEM; } static void __net_exit unix_net_exit(struct net *net) { + kvfree(net->unx.table.buckets); + kvfree(net->unx.table.locks); unix_sysctl_unregister(net); remove_proc_entry("unix", net->proc_net); } @@ -2919,13 +3646,96 @@ static struct pernet_operations unix_net_ops = { .exit = unix_net_exit, }; +#if IS_BUILTIN(CONFIG_UNIX) && defined(CONFIG_BPF_SYSCALL) && defined(CONFIG_PROC_FS) +DEFINE_BPF_ITER_FUNC(unix, struct bpf_iter_meta *meta, + struct unix_sock *unix_sk, uid_t uid) + +#define INIT_BATCH_SZ 16 + +static int bpf_iter_init_unix(void *priv_data, struct bpf_iter_aux_info *aux) +{ + struct bpf_unix_iter_state *iter = priv_data; + int err; + + err = bpf_iter_init_seq_net(priv_data, aux); + if (err) + return err; + + err = bpf_iter_unix_realloc_batch(iter, INIT_BATCH_SZ); + if (err) { + bpf_iter_fini_seq_net(priv_data); + return err; + } + + return 0; +} + +static void bpf_iter_fini_unix(void *priv_data) +{ + struct bpf_unix_iter_state *iter = priv_data; + + bpf_iter_fini_seq_net(priv_data); + kvfree(iter->batch); +} + +static const struct bpf_iter_seq_info unix_seq_info = { + .seq_ops = &bpf_iter_unix_seq_ops, + .init_seq_private = bpf_iter_init_unix, + .fini_seq_private = bpf_iter_fini_unix, + .seq_priv_size = sizeof(struct bpf_unix_iter_state), +}; + +static const struct bpf_func_proto * +bpf_iter_unix_get_func_proto(enum bpf_func_id func_id, + const struct bpf_prog *prog) +{ + switch (func_id) { + case BPF_FUNC_setsockopt: + return &bpf_sk_setsockopt_proto; + case BPF_FUNC_getsockopt: + return &bpf_sk_getsockopt_proto; + default: + return NULL; + } +} + +static struct bpf_iter_reg unix_reg_info = { + .target = "unix", + .ctx_arg_info_size = 1, + .ctx_arg_info = { + { offsetof(struct bpf_iter__unix, unix_sk), + PTR_TO_BTF_ID_OR_NULL }, + }, + .get_func_proto = bpf_iter_unix_get_func_proto, + .seq_info = &unix_seq_info, +}; + +static void __init bpf_iter_register(void) +{ + unix_reg_info.ctx_arg_info[0].btf_id = btf_sock_ids[BTF_SOCK_TYPE_UNIX]; + if (bpf_iter_reg_target(&unix_reg_info)) + pr_warn("Warning: could not register bpf iterator unix\n"); +} +#endif + static int __init af_unix_init(void) { - int rc = -1; + int i, rc = -1; BUILD_BUG_ON(sizeof(struct unix_skb_parms) > sizeof_field(struct sk_buff, cb)); - rc = proto_register(&unix_proto, 1); + for (i = 0; i < UNIX_HASH_SIZE / 2; i++) { + spin_lock_init(&bsd_socket_locks[i]); + INIT_HLIST_HEAD(&bsd_socket_buckets[i]); + } + + rc = proto_register(&unix_dgram_proto, 1); + if (rc != 0) { + pr_crit("%s: Cannot create unix_sock SLAB cache!\n", __func__); + goto out; + } + + rc = proto_register(&unix_stream_proto, 1); if (rc != 0) { pr_crit("%s: Cannot create unix_sock SLAB cache!\n", __func__); goto out; @@ -2933,6 +3743,12 @@ static int __init af_unix_init(void) sock_register(&unix_family_ops); register_pernet_subsys(&unix_net_ops); + unix_bpf_build_proto(); + +#if IS_BUILTIN(CONFIG_UNIX) && defined(CONFIG_BPF_SYSCALL) && defined(CONFIG_PROC_FS) + bpf_iter_register(); +#endif + out: return rc; } @@ -2940,7 +3756,8 @@ out: static void __exit af_unix_exit(void) { sock_unregister(PF_UNIX); - proto_unregister(&unix_proto); + proto_unregister(&unix_dgram_proto); + proto_unregister(&unix_stream_proto); unregister_pernet_subsys(&unix_net_ops); } diff --git a/net/unix/diag.c b/net/unix/diag.c index 9ff64f9df1f3..105f522a89fe 100644 --- a/net/unix/diag.c +++ b/net/unix/diag.c @@ -13,13 +13,14 @@ static int sk_diag_dump_name(struct sock *sk, struct sk_buff *nlskb) { - /* might or might not have unix_table_lock */ + /* might or might not have a hash table lock */ struct unix_address *addr = smp_load_acquire(&unix_sk(sk)->addr); if (!addr) return 0; - return nla_put(nlskb, UNIX_DIAG_NAME, addr->len - sizeof(short), + return nla_put(nlskb, UNIX_DIAG_NAME, + addr->len - offsetof(struct sockaddr_un, sun_path), addr->name->sun_path); } @@ -194,25 +195,21 @@ static int sk_diag_dump(struct sock *sk, struct sk_buff *skb, struct unix_diag_r static int unix_diag_dump(struct sk_buff *skb, struct netlink_callback *cb) { - struct unix_diag_req *req; - int num, s_num, slot, s_slot; struct net *net = sock_net(skb->sk); + int num, s_num, slot, s_slot; + struct unix_diag_req *req; req = nlmsg_data(cb->nlh); s_slot = cb->args[0]; num = s_num = cb->args[1]; - spin_lock(&unix_table_lock); - for (slot = s_slot; - slot < ARRAY_SIZE(unix_socket_table); - s_num = 0, slot++) { + for (slot = s_slot; slot < UNIX_HASH_SIZE; s_num = 0, slot++) { struct sock *sk; num = 0; - sk_for_each(sk, &unix_socket_table[slot]) { - if (!net_eq(sock_net(sk), net)) - continue; + spin_lock(&net->unx.table.locks[slot]); + sk_for_each(sk, &net->unx.table.buckets[slot]) { if (num < s_num) goto next; if (!(req->udiag_states & (1 << sk->sk_state))) @@ -220,37 +217,38 @@ static int unix_diag_dump(struct sk_buff *skb, struct netlink_callback *cb) if (sk_diag_dump(sk, skb, req, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, - NLM_F_MULTI) < 0) + NLM_F_MULTI) < 0) { + spin_unlock(&net->unx.table.locks[slot]); goto done; + } next: num++; } + spin_unlock(&net->unx.table.locks[slot]); } done: - spin_unlock(&unix_table_lock); cb->args[0] = slot; cb->args[1] = num; return skb->len; } -static struct sock *unix_lookup_by_ino(unsigned int ino) +static struct sock *unix_lookup_by_ino(struct net *net, unsigned int ino) { - int i; struct sock *sk; + int i; - spin_lock(&unix_table_lock); - for (i = 0; i < ARRAY_SIZE(unix_socket_table); i++) { - sk_for_each(sk, &unix_socket_table[i]) + for (i = 0; i < UNIX_HASH_SIZE; i++) { + spin_lock(&net->unx.table.locks[i]); + sk_for_each(sk, &net->unx.table.buckets[i]) { if (ino == sock_i_ino(sk)) { sock_hold(sk); - spin_unlock(&unix_table_lock); - + spin_unlock(&net->unx.table.locks[i]); return sk; } + } + spin_unlock(&net->unx.table.locks[i]); } - - spin_unlock(&unix_table_lock); return NULL; } @@ -258,21 +256,20 @@ static int unix_diag_get_exact(struct sk_buff *in_skb, const struct nlmsghdr *nlh, struct unix_diag_req *req) { - int err = -EINVAL; - struct sock *sk; - struct sk_buff *rep; - unsigned int extra_len; struct net *net = sock_net(in_skb->sk); + unsigned int extra_len; + struct sk_buff *rep; + struct sock *sk; + int err; + err = -EINVAL; if (req->udiag_ino == 0) goto out_nosk; - sk = unix_lookup_by_ino(req->udiag_ino); + sk = unix_lookup_by_ino(net, req->udiag_ino); err = -ENOENT; if (sk == NULL) goto out_nosk; - if (!net_eq(sock_net(sk), net)) - goto out; err = sock_diag_check_cookie(sk, req->udiag_cookie); if (err) @@ -295,10 +292,8 @@ again: goto again; } - err = netlink_unicast(net->diag_nlsk, rep, NETLINK_CB(in_skb).portid, - MSG_DONTWAIT); - if (err > 0) - err = 0; + err = nlmsg_unicast(net->diag_nlsk, rep, NETLINK_CB(in_skb).portid); + out: if (sk) sock_put(sk); @@ -309,7 +304,6 @@ out_nosk: static int unix_diag_handler_dump(struct sk_buff *skb, struct nlmsghdr *h) { int hdrlen = sizeof(struct unix_diag_req); - struct net *net = sock_net(skb->sk); if (nlmsg_len(h) < hdrlen) return -EINVAL; @@ -318,7 +312,7 @@ static int unix_diag_handler_dump(struct sk_buff *skb, struct nlmsghdr *h) struct netlink_dump_control c = { .dump = unix_diag_dump, }; - return netlink_dump_start(net->diag_nlsk, skb, h, &c); + return netlink_dump_start(sock_net(skb->sk)->diag_nlsk, skb, h, &c); } else return unix_diag_get_exact(skb, h, nlmsg_data(h)); } diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 12e2ddaf887f..dc2763540393 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -192,8 +192,11 @@ void wait_for_unix_gc(void) { /* If number of inflight sockets is insane, * force a garbage collect right now. + * Paired with the WRITE_ONCE() in unix_inflight(), + * unix_notinflight() and gc_in_progress(). */ - if (unix_tot_inflight > UNIX_INFLIGHT_TRIGGER_GC && !gc_in_progress) + if (READ_ONCE(unix_tot_inflight) > UNIX_INFLIGHT_TRIGGER_GC && + !READ_ONCE(gc_in_progress)) unix_gc(); wait_event(unix_gc_wait, gc_in_progress == false); } @@ -201,6 +204,7 @@ void wait_for_unix_gc(void) /* The external entry point: unix_gc() */ void unix_gc(void) { + struct sk_buff *next_skb, *skb; struct unix_sock *u; struct unix_sock *next; struct sk_buff_head hitlist; @@ -213,7 +217,9 @@ void unix_gc(void) if (gc_in_progress) goto out; - gc_in_progress = true; + /* Paired with READ_ONCE() in wait_for_unix_gc(). */ + WRITE_ONCE(gc_in_progress, true); + /* First, select candidates for garbage collection. Only * in-flight sockets are considered, and from those only ones * which don't have any external reference. @@ -292,14 +298,36 @@ void unix_gc(void) spin_unlock(&unix_gc_lock); + /* We need io_uring to clean its registered files, ignore all io_uring + * originated skbs. It's fine as io_uring doesn't keep references to + * other io_uring instances and so killing all other files in the cycle + * will put all io_uring references forcing it to go through normal + * release.path eventually putting registered files. + */ + skb_queue_walk_safe(&hitlist, skb, next_skb) { + if (skb->scm_io_uring) { + __skb_unlink(skb, &hitlist); + skb_queue_tail(&skb->sk->sk_receive_queue, skb); + } + } + /* Here we are. Hitlist is filled. Die. */ __skb_queue_purge(&hitlist); spin_lock(&unix_gc_lock); + /* There could be io_uring registered files, just push them back to + * the inflight list + */ + list_for_each_entry_safe(u, next, &gc_candidates, link) + list_move_tail(&u->link, &gc_inflight_list); + /* All candidates should have been detached by now. */ BUG_ON(!list_empty(&gc_candidates)); - gc_in_progress = false; + + /* Paired with READ_ONCE() in wait_for_unix_gc(). */ + WRITE_ONCE(gc_in_progress, false); + wake_up(&unix_gc_wait); out: diff --git a/net/unix/scm.c b/net/unix/scm.c index 8c40f2b32392..aa27a02478dc 100644 --- a/net/unix/scm.c +++ b/net/unix/scm.c @@ -8,6 +8,7 @@ #include <net/af_unix.h> #include <net/scm.h> #include <linux/init.h> +#include <linux/io_uring.h> #include "scm.h" @@ -59,7 +60,8 @@ void unix_inflight(struct user_struct *user, struct file *fp) } else { BUG_ON(list_empty(&u->link)); } - unix_tot_inflight++; + /* Paired with READ_ONCE() in wait_for_unix_gc() */ + WRITE_ONCE(unix_tot_inflight, unix_tot_inflight + 1); } user->unix_inflight++; spin_unlock(&unix_gc_lock); @@ -79,7 +81,8 @@ void unix_notinflight(struct user_struct *user, struct file *fp) if (atomic_long_dec_and_test(&u->inflight)) list_del_init(&u->link); - unix_tot_inflight--; + /* Paired with READ_ONCE() in wait_for_unix_gc() */ + WRITE_ONCE(unix_tot_inflight, unix_tot_inflight - 1); } user->unix_inflight--; spin_unlock(&unix_gc_lock); diff --git a/net/unix/sysctl_net_unix.c b/net/unix/sysctl_net_unix.c index c09bea89151b..500129aa710c 100644 --- a/net/unix/sysctl_net_unix.c +++ b/net/unix/sysctl_net_unix.c @@ -26,15 +26,16 @@ int __net_init unix_sysctl_register(struct net *net) { struct ctl_table *table; - table = kmemdup(unix_table, sizeof(unix_table), GFP_KERNEL); - if (table == NULL) - goto err_alloc; + if (net_eq(net, &init_net)) { + table = unix_table; + } else { + table = kmemdup(unix_table, sizeof(unix_table), GFP_KERNEL); + if (!table) + goto err_alloc; - /* Don't export sysctls to unprivileged users */ - if (net->user_ns != &init_user_ns) - table[0].procname = NULL; + table[0].data = &net->unx.sysctl_max_dgram_qlen; + } - table[0].data = &net->unx.sysctl_max_dgram_qlen; net->unx.ctl = register_net_sysctl(net, "net/unix", table); if (net->unx.ctl == NULL) goto err_reg; @@ -42,7 +43,8 @@ int __net_init unix_sysctl_register(struct net *net) return 0; err_reg: - kfree(table); + if (!net_eq(net, &init_net)) + kfree(table); err_alloc: return -ENOMEM; } @@ -53,5 +55,6 @@ void unix_sysctl_unregister(struct net *net) table = net->unx.ctl->ctl_table_arg; unregister_net_sysctl_table(net->unx.ctl); - kfree(table); + if (!net_eq(net, &init_net)) + kfree(table); } diff --git a/net/unix/unix_bpf.c b/net/unix/unix_bpf.c new file mode 100644 index 000000000000..e9bf15513961 --- /dev/null +++ b/net/unix/unix_bpf.c @@ -0,0 +1,175 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2021 Cong Wang <cong.wang@bytedance.com> */ + +#include <linux/skmsg.h> +#include <linux/bpf.h> +#include <net/sock.h> +#include <net/af_unix.h> + +#define unix_sk_has_data(__sk, __psock) \ + ({ !skb_queue_empty(&__sk->sk_receive_queue) || \ + !skb_queue_empty(&__psock->ingress_skb) || \ + !list_empty(&__psock->ingress_msg); \ + }) + +static int unix_msg_wait_data(struct sock *sk, struct sk_psock *psock, + long timeo) +{ + DEFINE_WAIT_FUNC(wait, woken_wake_function); + struct unix_sock *u = unix_sk(sk); + int ret = 0; + + if (sk->sk_shutdown & RCV_SHUTDOWN) + return 1; + + if (!timeo) + return ret; + + add_wait_queue(sk_sleep(sk), &wait); + sk_set_bit(SOCKWQ_ASYNC_WAITDATA, sk); + if (!unix_sk_has_data(sk, psock)) { + mutex_unlock(&u->iolock); + wait_woken(&wait, TASK_INTERRUPTIBLE, timeo); + mutex_lock(&u->iolock); + ret = unix_sk_has_data(sk, psock); + } + sk_clear_bit(SOCKWQ_ASYNC_WAITDATA, sk); + remove_wait_queue(sk_sleep(sk), &wait); + return ret; +} + +static int __unix_recvmsg(struct sock *sk, struct msghdr *msg, + size_t len, int flags) +{ + if (sk->sk_type == SOCK_DGRAM) + return __unix_dgram_recvmsg(sk, msg, len, flags); + else + return __unix_stream_recvmsg(sk, msg, len, flags); +} + +static int unix_bpf_recvmsg(struct sock *sk, struct msghdr *msg, + size_t len, int flags, int *addr_len) +{ + struct unix_sock *u = unix_sk(sk); + struct sk_psock *psock; + int copied; + + psock = sk_psock_get(sk); + if (unlikely(!psock)) + return __unix_recvmsg(sk, msg, len, flags); + + mutex_lock(&u->iolock); + if (!skb_queue_empty(&sk->sk_receive_queue) && + sk_psock_queue_empty(psock)) { + mutex_unlock(&u->iolock); + sk_psock_put(sk, psock); + return __unix_recvmsg(sk, msg, len, flags); + } + +msg_bytes_ready: + copied = sk_msg_recvmsg(sk, psock, msg, len, flags); + if (!copied) { + long timeo; + int data; + + timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); + data = unix_msg_wait_data(sk, psock, timeo); + if (data) { + if (!sk_psock_queue_empty(psock)) + goto msg_bytes_ready; + mutex_unlock(&u->iolock); + sk_psock_put(sk, psock); + return __unix_recvmsg(sk, msg, len, flags); + } + copied = -EAGAIN; + } + mutex_unlock(&u->iolock); + sk_psock_put(sk, psock); + return copied; +} + +static struct proto *unix_dgram_prot_saved __read_mostly; +static DEFINE_SPINLOCK(unix_dgram_prot_lock); +static struct proto unix_dgram_bpf_prot; + +static struct proto *unix_stream_prot_saved __read_mostly; +static DEFINE_SPINLOCK(unix_stream_prot_lock); +static struct proto unix_stream_bpf_prot; + +static void unix_dgram_bpf_rebuild_protos(struct proto *prot, const struct proto *base) +{ + *prot = *base; + prot->close = sock_map_close; + prot->recvmsg = unix_bpf_recvmsg; + prot->sock_is_readable = sk_msg_is_readable; +} + +static void unix_stream_bpf_rebuild_protos(struct proto *prot, + const struct proto *base) +{ + *prot = *base; + prot->close = sock_map_close; + prot->recvmsg = unix_bpf_recvmsg; + prot->sock_is_readable = sk_msg_is_readable; + prot->unhash = sock_map_unhash; +} + +static void unix_dgram_bpf_check_needs_rebuild(struct proto *ops) +{ + if (unlikely(ops != smp_load_acquire(&unix_dgram_prot_saved))) { + spin_lock_bh(&unix_dgram_prot_lock); + if (likely(ops != unix_dgram_prot_saved)) { + unix_dgram_bpf_rebuild_protos(&unix_dgram_bpf_prot, ops); + smp_store_release(&unix_dgram_prot_saved, ops); + } + spin_unlock_bh(&unix_dgram_prot_lock); + } +} + +static void unix_stream_bpf_check_needs_rebuild(struct proto *ops) +{ + if (unlikely(ops != smp_load_acquire(&unix_stream_prot_saved))) { + spin_lock_bh(&unix_stream_prot_lock); + if (likely(ops != unix_stream_prot_saved)) { + unix_stream_bpf_rebuild_protos(&unix_stream_bpf_prot, ops); + smp_store_release(&unix_stream_prot_saved, ops); + } + spin_unlock_bh(&unix_stream_prot_lock); + } +} + +int unix_dgram_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore) +{ + if (sk->sk_type != SOCK_DGRAM) + return -EOPNOTSUPP; + + if (restore) { + sk->sk_write_space = psock->saved_write_space; + sock_replace_proto(sk, psock->sk_proto); + return 0; + } + + unix_dgram_bpf_check_needs_rebuild(psock->sk_proto); + sock_replace_proto(sk, &unix_dgram_bpf_prot); + return 0; +} + +int unix_stream_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore) +{ + if (restore) { + sk->sk_write_space = psock->saved_write_space; + sock_replace_proto(sk, psock->sk_proto); + return 0; + } + + unix_stream_bpf_check_needs_rebuild(psock->sk_proto); + sock_replace_proto(sk, &unix_stream_bpf_prot); + return 0; +} + +void __init unix_bpf_build_proto(void) +{ + unix_dgram_bpf_rebuild_protos(&unix_dgram_bpf_prot, &unix_dgram_proto); + unix_stream_bpf_rebuild_protos(&unix_stream_bpf_prot, &unix_stream_proto); + +} |