1 // SPDX-License-Identifier: GPL-2.0
2 /* Copyright (c) 2017 - 2018 Covalent IO, Inc. http://covalent.io */
5 #include <linux/filter.h>
6 #include <linux/errno.h>
7 #include <linux/file.h>
9 #include <linux/workqueue.h>
10 #include <linux/skmsg.h>
11 #include <linux/list.h>
12 #include <linux/jhash.h>
17 struct sk_psock_progs progs;
21 #define SOCK_CREATE_FLAG_MASK \
22 (BPF_F_NUMA_NODE | BPF_F_RDONLY | BPF_F_WRONLY)
24 static struct bpf_map *sock_map_alloc(union bpf_attr *attr)
26 struct bpf_stab *stab;
30 if (!capable(CAP_NET_ADMIN))
31 return ERR_PTR(-EPERM);
32 if (attr->max_entries == 0 ||
33 attr->key_size != 4 ||
34 attr->value_size != 4 ||
35 attr->map_flags & ~SOCK_CREATE_FLAG_MASK)
36 return ERR_PTR(-EINVAL);
38 stab = kzalloc(sizeof(*stab), GFP_USER);
40 return ERR_PTR(-ENOMEM);
42 bpf_map_init_from_attr(&stab->map, attr);
43 raw_spin_lock_init(&stab->lock);
45 /* Make sure page count doesn't overflow. */
46 cost = (u64) stab->map.max_entries * sizeof(struct sock *);
47 err = bpf_map_charge_init(&stab->map.memory, cost);
51 stab->sks = bpf_map_area_alloc((u64) stab->map.max_entries *
52 sizeof(struct sock *),
57 bpf_map_charge_finish(&stab->map.memory);
63 int sock_map_get_from_fd(const union bpf_attr *attr, struct bpf_prog *prog)
65 u32 ufd = attr->target_fd;
71 map = __bpf_map_get(f);
74 ret = sock_map_prog_update(map, prog, NULL, attr->attach_type);
79 int sock_map_prog_detach(const union bpf_attr *attr, enum bpf_prog_type ptype)
81 u32 ufd = attr->target_fd;
82 struct bpf_prog *prog;
87 if (attr->attach_flags)
91 map = __bpf_map_get(f);
95 prog = bpf_prog_get(attr->attach_bpf_fd);
101 if (prog->type != ptype) {
106 ret = sock_map_prog_update(map, NULL, prog, attr->attach_type);
114 static void sock_map_sk_acquire(struct sock *sk)
115 __acquires(&sk->sk_lock.slock)
121 static void sock_map_sk_release(struct sock *sk)
122 __releases(&sk->sk_lock.slock)
128 static void sock_map_add_link(struct sk_psock *psock,
129 struct sk_psock_link *link,
130 struct bpf_map *map, void *link_raw)
132 link->link_raw = link_raw;
134 spin_lock_bh(&psock->link_lock);
135 list_add_tail(&link->list, &psock->link);
136 spin_unlock_bh(&psock->link_lock);
139 static void sock_map_del_link(struct sock *sk,
140 struct sk_psock *psock, void *link_raw)
142 struct sk_psock_link *link, *tmp;
143 bool strp_stop = false;
145 spin_lock_bh(&psock->link_lock);
146 list_for_each_entry_safe(link, tmp, &psock->link, list) {
147 if (link->link_raw == link_raw) {
148 struct bpf_map *map = link->map;
149 struct bpf_stab *stab = container_of(map, struct bpf_stab,
151 if (psock->parser.enabled && stab->progs.skb_parser)
153 list_del(&link->list);
154 sk_psock_free_link(link);
157 spin_unlock_bh(&psock->link_lock);
159 write_lock_bh(&sk->sk_callback_lock);
160 sk_psock_stop_strp(sk, psock);
161 write_unlock_bh(&sk->sk_callback_lock);
165 static void sock_map_unref(struct sock *sk, void *link_raw)
167 struct sk_psock *psock = sk_psock(sk);
170 sock_map_del_link(sk, psock, link_raw);
171 sk_psock_put(sk, psock);
175 static int sock_map_link(struct bpf_map *map, struct sk_psock_progs *progs,
178 struct bpf_prog *msg_parser, *skb_parser, *skb_verdict;
179 bool skb_progs, sk_psock_is_new = false;
180 struct sk_psock *psock;
183 skb_verdict = READ_ONCE(progs->skb_verdict);
184 skb_parser = READ_ONCE(progs->skb_parser);
185 skb_progs = skb_parser && skb_verdict;
187 skb_verdict = bpf_prog_inc_not_zero(skb_verdict);
188 if (IS_ERR(skb_verdict))
189 return PTR_ERR(skb_verdict);
190 skb_parser = bpf_prog_inc_not_zero(skb_parser);
191 if (IS_ERR(skb_parser)) {
192 bpf_prog_put(skb_verdict);
193 return PTR_ERR(skb_parser);
197 msg_parser = READ_ONCE(progs->msg_parser);
199 msg_parser = bpf_prog_inc_not_zero(msg_parser);
200 if (IS_ERR(msg_parser)) {
201 ret = PTR_ERR(msg_parser);
206 psock = sk_psock_get_checked(sk);
208 ret = PTR_ERR(psock);
213 if ((msg_parser && READ_ONCE(psock->progs.msg_parser)) ||
214 (skb_progs && READ_ONCE(psock->progs.skb_parser))) {
215 sk_psock_put(sk, psock);
220 psock = sk_psock_init(sk, map->numa_node);
225 sk_psock_is_new = true;
229 psock_set_prog(&psock->progs.msg_parser, msg_parser);
230 if (sk_psock_is_new) {
231 ret = tcp_bpf_init(sk);
238 write_lock_bh(&sk->sk_callback_lock);
239 if (skb_progs && !psock->parser.enabled) {
240 ret = sk_psock_init_strp(sk, psock);
242 write_unlock_bh(&sk->sk_callback_lock);
245 psock_set_prog(&psock->progs.skb_verdict, skb_verdict);
246 psock_set_prog(&psock->progs.skb_parser, skb_parser);
247 sk_psock_start_strp(sk, psock);
249 write_unlock_bh(&sk->sk_callback_lock);
252 sk_psock_put(sk, psock);
255 bpf_prog_put(msg_parser);
258 bpf_prog_put(skb_verdict);
259 bpf_prog_put(skb_parser);
264 static void sock_map_free(struct bpf_map *map)
266 struct bpf_stab *stab = container_of(map, struct bpf_stab, map);
269 /* After the sync no updates or deletes will be in-flight so it
270 * is safe to walk map and remove entries without risking a race
271 * in EEXIST update case.
274 for (i = 0; i < stab->map.max_entries; i++) {
275 struct sock **psk = &stab->sks[i];
278 sk = xchg(psk, NULL);
283 sock_map_unref(sk, psk);
290 /* wait for psock readers accessing its map link */
293 bpf_map_area_free(stab->sks);
297 static void sock_map_release_progs(struct bpf_map *map)
299 psock_progs_drop(&container_of(map, struct bpf_stab, map)->progs);
302 static struct sock *__sock_map_lookup_elem(struct bpf_map *map, u32 key)
304 struct bpf_stab *stab = container_of(map, struct bpf_stab, map);
306 WARN_ON_ONCE(!rcu_read_lock_held());
308 if (unlikely(key >= map->max_entries))
310 return READ_ONCE(stab->sks[key]);
313 static void *sock_map_lookup(struct bpf_map *map, void *key)
315 return ERR_PTR(-EOPNOTSUPP);
318 static int __sock_map_delete(struct bpf_stab *stab, struct sock *sk_test,
324 raw_spin_lock_bh(&stab->lock);
326 if (!sk_test || sk_test == sk)
327 sk = xchg(psk, NULL);
330 sock_map_unref(sk, psk);
334 raw_spin_unlock_bh(&stab->lock);
338 static void sock_map_delete_from_link(struct bpf_map *map, struct sock *sk,
341 struct bpf_stab *stab = container_of(map, struct bpf_stab, map);
343 __sock_map_delete(stab, sk, link_raw);
346 static int sock_map_delete_elem(struct bpf_map *map, void *key)
348 struct bpf_stab *stab = container_of(map, struct bpf_stab, map);
352 if (unlikely(i >= map->max_entries))
356 return __sock_map_delete(stab, NULL, psk);
359 static int sock_map_get_next_key(struct bpf_map *map, void *key, void *next)
361 struct bpf_stab *stab = container_of(map, struct bpf_stab, map);
362 u32 i = key ? *(u32 *)key : U32_MAX;
363 u32 *key_next = next;
365 if (i == stab->map.max_entries - 1)
367 if (i >= stab->map.max_entries)
374 static int sock_map_update_common(struct bpf_map *map, u32 idx,
375 struct sock *sk, u64 flags)
377 struct bpf_stab *stab = container_of(map, struct bpf_stab, map);
378 struct inet_connection_sock *icsk = inet_csk(sk);
379 struct sk_psock_link *link;
380 struct sk_psock *psock;
384 WARN_ON_ONCE(!rcu_read_lock_held());
385 if (unlikely(flags > BPF_EXIST))
387 if (unlikely(idx >= map->max_entries))
389 if (unlikely(rcu_access_pointer(icsk->icsk_ulp_data)))
392 link = sk_psock_init_link();
396 ret = sock_map_link(map, &stab->progs, sk);
400 psock = sk_psock(sk);
401 WARN_ON_ONCE(!psock);
403 raw_spin_lock_bh(&stab->lock);
404 osk = stab->sks[idx];
405 if (osk && flags == BPF_NOEXIST) {
408 } else if (!osk && flags == BPF_EXIST) {
413 sock_map_add_link(psock, link, map, &stab->sks[idx]);
416 sock_map_unref(osk, &stab->sks[idx]);
417 raw_spin_unlock_bh(&stab->lock);
420 raw_spin_unlock_bh(&stab->lock);
422 sk_psock_put(sk, psock);
424 sk_psock_free_link(link);
428 static bool sock_map_op_okay(const struct bpf_sock_ops_kern *ops)
430 return ops->op == BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB ||
431 ops->op == BPF_SOCK_OPS_ACTIVE_ESTABLISHED_CB;
434 static bool sock_map_sk_is_suitable(const struct sock *sk)
436 return sk->sk_type == SOCK_STREAM &&
437 sk->sk_protocol == IPPROTO_TCP;
440 static int sock_map_update_elem(struct bpf_map *map, void *key,
441 void *value, u64 flags)
443 u32 ufd = *(u32 *)value;
444 u32 idx = *(u32 *)key;
449 sock = sockfd_lookup(ufd, &ret);
457 if (!sock_map_sk_is_suitable(sk)) {
462 sock_map_sk_acquire(sk);
463 if (sk->sk_state != TCP_ESTABLISHED)
466 ret = sock_map_update_common(map, idx, sk, flags);
467 sock_map_sk_release(sk);
473 BPF_CALL_4(bpf_sock_map_update, struct bpf_sock_ops_kern *, sops,
474 struct bpf_map *, map, void *, key, u64, flags)
476 WARN_ON_ONCE(!rcu_read_lock_held());
478 if (likely(sock_map_sk_is_suitable(sops->sk) &&
479 sock_map_op_okay(sops)))
480 return sock_map_update_common(map, *(u32 *)key, sops->sk,
485 const struct bpf_func_proto bpf_sock_map_update_proto = {
486 .func = bpf_sock_map_update,
489 .ret_type = RET_INTEGER,
490 .arg1_type = ARG_PTR_TO_CTX,
491 .arg2_type = ARG_CONST_MAP_PTR,
492 .arg3_type = ARG_PTR_TO_MAP_KEY,
493 .arg4_type = ARG_ANYTHING,
496 BPF_CALL_4(bpf_sk_redirect_map, struct sk_buff *, skb,
497 struct bpf_map *, map, u32, key, u64, flags)
499 struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);
501 if (unlikely(flags & ~(BPF_F_INGRESS)))
503 tcb->bpf.flags = flags;
504 tcb->bpf.sk_redir = __sock_map_lookup_elem(map, key);
505 if (!tcb->bpf.sk_redir)
510 const struct bpf_func_proto bpf_sk_redirect_map_proto = {
511 .func = bpf_sk_redirect_map,
513 .ret_type = RET_INTEGER,
514 .arg1_type = ARG_PTR_TO_CTX,
515 .arg2_type = ARG_CONST_MAP_PTR,
516 .arg3_type = ARG_ANYTHING,
517 .arg4_type = ARG_ANYTHING,
520 BPF_CALL_4(bpf_msg_redirect_map, struct sk_msg *, msg,
521 struct bpf_map *, map, u32, key, u64, flags)
523 if (unlikely(flags & ~(BPF_F_INGRESS)))
526 msg->sk_redir = __sock_map_lookup_elem(map, key);
532 const struct bpf_func_proto bpf_msg_redirect_map_proto = {
533 .func = bpf_msg_redirect_map,
535 .ret_type = RET_INTEGER,
536 .arg1_type = ARG_PTR_TO_CTX,
537 .arg2_type = ARG_CONST_MAP_PTR,
538 .arg3_type = ARG_ANYTHING,
539 .arg4_type = ARG_ANYTHING,
542 const struct bpf_map_ops sock_map_ops = {
543 .map_alloc = sock_map_alloc,
544 .map_free = sock_map_free,
545 .map_get_next_key = sock_map_get_next_key,
546 .map_update_elem = sock_map_update_elem,
547 .map_delete_elem = sock_map_delete_elem,
548 .map_lookup_elem = sock_map_lookup,
549 .map_release_uref = sock_map_release_progs,
550 .map_check_btf = map_check_no_btf,
553 struct bpf_htab_elem {
557 struct hlist_node node;
561 struct bpf_htab_bucket {
562 struct hlist_head head;
568 struct bpf_htab_bucket *buckets;
571 struct sk_psock_progs progs;
575 static inline u32 sock_hash_bucket_hash(const void *key, u32 len)
577 return jhash(key, len, 0);
580 static struct bpf_htab_bucket *sock_hash_select_bucket(struct bpf_htab *htab,
583 return &htab->buckets[hash & (htab->buckets_num - 1)];
586 static struct bpf_htab_elem *
587 sock_hash_lookup_elem_raw(struct hlist_head *head, u32 hash, void *key,
590 struct bpf_htab_elem *elem;
592 hlist_for_each_entry_rcu(elem, head, node) {
593 if (elem->hash == hash &&
594 !memcmp(&elem->key, key, key_size))
601 static struct sock *__sock_hash_lookup_elem(struct bpf_map *map, void *key)
603 struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
604 u32 key_size = map->key_size, hash;
605 struct bpf_htab_bucket *bucket;
606 struct bpf_htab_elem *elem;
608 WARN_ON_ONCE(!rcu_read_lock_held());
610 hash = sock_hash_bucket_hash(key, key_size);
611 bucket = sock_hash_select_bucket(htab, hash);
612 elem = sock_hash_lookup_elem_raw(&bucket->head, hash, key, key_size);
614 return elem ? elem->sk : NULL;
617 static void sock_hash_free_elem(struct bpf_htab *htab,
618 struct bpf_htab_elem *elem)
620 atomic_dec(&htab->count);
621 kfree_rcu(elem, rcu);
624 static void sock_hash_delete_from_link(struct bpf_map *map, struct sock *sk,
627 struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
628 struct bpf_htab_elem *elem_probe, *elem = link_raw;
629 struct bpf_htab_bucket *bucket;
631 WARN_ON_ONCE(!rcu_read_lock_held());
632 bucket = sock_hash_select_bucket(htab, elem->hash);
634 /* elem may be deleted in parallel from the map, but access here
635 * is okay since it's going away only after RCU grace period.
636 * However, we need to check whether it's still present.
638 raw_spin_lock_bh(&bucket->lock);
639 elem_probe = sock_hash_lookup_elem_raw(&bucket->head, elem->hash,
640 elem->key, map->key_size);
641 if (elem_probe && elem_probe == elem) {
642 hlist_del_rcu(&elem->node);
643 sock_map_unref(elem->sk, elem);
644 sock_hash_free_elem(htab, elem);
646 raw_spin_unlock_bh(&bucket->lock);
649 static int sock_hash_delete_elem(struct bpf_map *map, void *key)
651 struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
652 u32 hash, key_size = map->key_size;
653 struct bpf_htab_bucket *bucket;
654 struct bpf_htab_elem *elem;
657 hash = sock_hash_bucket_hash(key, key_size);
658 bucket = sock_hash_select_bucket(htab, hash);
660 raw_spin_lock_bh(&bucket->lock);
661 elem = sock_hash_lookup_elem_raw(&bucket->head, hash, key, key_size);
663 hlist_del_rcu(&elem->node);
664 sock_map_unref(elem->sk, elem);
665 sock_hash_free_elem(htab, elem);
668 raw_spin_unlock_bh(&bucket->lock);
672 static struct bpf_htab_elem *sock_hash_alloc_elem(struct bpf_htab *htab,
673 void *key, u32 key_size,
674 u32 hash, struct sock *sk,
675 struct bpf_htab_elem *old)
677 struct bpf_htab_elem *new;
679 if (atomic_inc_return(&htab->count) > htab->map.max_entries) {
681 atomic_dec(&htab->count);
682 return ERR_PTR(-E2BIG);
686 new = kmalloc_node(htab->elem_size, GFP_ATOMIC | __GFP_NOWARN,
687 htab->map.numa_node);
689 atomic_dec(&htab->count);
690 return ERR_PTR(-ENOMEM);
692 memcpy(new->key, key, key_size);
698 static int sock_hash_update_common(struct bpf_map *map, void *key,
699 struct sock *sk, u64 flags)
701 struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
702 struct inet_connection_sock *icsk = inet_csk(sk);
703 u32 key_size = map->key_size, hash;
704 struct bpf_htab_elem *elem, *elem_new;
705 struct bpf_htab_bucket *bucket;
706 struct sk_psock_link *link;
707 struct sk_psock *psock;
710 WARN_ON_ONCE(!rcu_read_lock_held());
711 if (unlikely(flags > BPF_EXIST))
713 if (unlikely(icsk->icsk_ulp_data))
716 link = sk_psock_init_link();
720 ret = sock_map_link(map, &htab->progs, sk);
724 psock = sk_psock(sk);
725 WARN_ON_ONCE(!psock);
727 hash = sock_hash_bucket_hash(key, key_size);
728 bucket = sock_hash_select_bucket(htab, hash);
730 raw_spin_lock_bh(&bucket->lock);
731 elem = sock_hash_lookup_elem_raw(&bucket->head, hash, key, key_size);
732 if (elem && flags == BPF_NOEXIST) {
735 } else if (!elem && flags == BPF_EXIST) {
740 elem_new = sock_hash_alloc_elem(htab, key, key_size, hash, sk, elem);
741 if (IS_ERR(elem_new)) {
742 ret = PTR_ERR(elem_new);
746 sock_map_add_link(psock, link, map, elem_new);
747 /* Add new element to the head of the list, so that
748 * concurrent search will find it before old elem.
750 hlist_add_head_rcu(&elem_new->node, &bucket->head);
752 hlist_del_rcu(&elem->node);
753 sock_map_unref(elem->sk, elem);
754 sock_hash_free_elem(htab, elem);
756 raw_spin_unlock_bh(&bucket->lock);
759 raw_spin_unlock_bh(&bucket->lock);
760 sk_psock_put(sk, psock);
762 sk_psock_free_link(link);
766 static int sock_hash_update_elem(struct bpf_map *map, void *key,
767 void *value, u64 flags)
769 u32 ufd = *(u32 *)value;
774 sock = sockfd_lookup(ufd, &ret);
782 if (!sock_map_sk_is_suitable(sk)) {
787 sock_map_sk_acquire(sk);
788 if (sk->sk_state != TCP_ESTABLISHED)
791 ret = sock_hash_update_common(map, key, sk, flags);
792 sock_map_sk_release(sk);
798 static int sock_hash_get_next_key(struct bpf_map *map, void *key,
801 struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
802 struct bpf_htab_elem *elem, *elem_next;
803 u32 hash, key_size = map->key_size;
804 struct hlist_head *head;
808 goto find_first_elem;
809 hash = sock_hash_bucket_hash(key, key_size);
810 head = &sock_hash_select_bucket(htab, hash)->head;
811 elem = sock_hash_lookup_elem_raw(head, hash, key, key_size);
813 goto find_first_elem;
815 elem_next = hlist_entry_safe(rcu_dereference_raw(hlist_next_rcu(&elem->node)),
816 struct bpf_htab_elem, node);
818 memcpy(key_next, elem_next->key, key_size);
822 i = hash & (htab->buckets_num - 1);
825 for (; i < htab->buckets_num; i++) {
826 head = &sock_hash_select_bucket(htab, i)->head;
827 elem_next = hlist_entry_safe(rcu_dereference_raw(hlist_first_rcu(head)),
828 struct bpf_htab_elem, node);
830 memcpy(key_next, elem_next->key, key_size);
838 static struct bpf_map *sock_hash_alloc(union bpf_attr *attr)
840 struct bpf_htab *htab;
844 if (!capable(CAP_NET_ADMIN))
845 return ERR_PTR(-EPERM);
846 if (attr->max_entries == 0 ||
847 attr->key_size == 0 ||
848 attr->value_size != 4 ||
849 attr->map_flags & ~SOCK_CREATE_FLAG_MASK)
850 return ERR_PTR(-EINVAL);
851 if (attr->key_size > MAX_BPF_STACK)
852 return ERR_PTR(-E2BIG);
854 htab = kzalloc(sizeof(*htab), GFP_USER);
856 return ERR_PTR(-ENOMEM);
858 bpf_map_init_from_attr(&htab->map, attr);
860 htab->buckets_num = roundup_pow_of_two(htab->map.max_entries);
861 htab->elem_size = sizeof(struct bpf_htab_elem) +
862 round_up(htab->map.key_size, 8);
863 if (htab->buckets_num == 0 ||
864 htab->buckets_num > U32_MAX / sizeof(struct bpf_htab_bucket)) {
869 cost = (u64) htab->buckets_num * sizeof(struct bpf_htab_bucket) +
870 (u64) htab->elem_size * htab->map.max_entries;
871 if (cost >= U32_MAX - PAGE_SIZE) {
875 err = bpf_map_charge_init(&htab->map.memory, cost);
879 htab->buckets = bpf_map_area_alloc(htab->buckets_num *
880 sizeof(struct bpf_htab_bucket),
881 htab->map.numa_node);
882 if (!htab->buckets) {
883 bpf_map_charge_finish(&htab->map.memory);
888 for (i = 0; i < htab->buckets_num; i++) {
889 INIT_HLIST_HEAD(&htab->buckets[i].head);
890 raw_spin_lock_init(&htab->buckets[i].lock);
899 static void sock_hash_free(struct bpf_map *map)
901 struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
902 struct bpf_htab_bucket *bucket;
903 struct hlist_head unlink_list;
904 struct bpf_htab_elem *elem;
905 struct hlist_node *node;
908 /* After the sync no updates or deletes will be in-flight so it
909 * is safe to walk map and remove entries without risking a race
910 * in EEXIST update case.
913 for (i = 0; i < htab->buckets_num; i++) {
914 bucket = sock_hash_select_bucket(htab, i);
916 /* We are racing with sock_hash_delete_from_link to
917 * enter the spin-lock critical section. Every socket on
918 * the list is still linked to sockhash. Since link
919 * exists, psock exists and holds a ref to socket. That
920 * lets us to grab a socket ref too.
922 raw_spin_lock_bh(&bucket->lock);
923 hlist_for_each_entry(elem, &bucket->head, node)
925 hlist_move_list(&bucket->head, &unlink_list);
926 raw_spin_unlock_bh(&bucket->lock);
928 /* Process removed entries out of atomic context to
929 * block for socket lock before deleting the psock's
932 hlist_for_each_entry_safe(elem, node, &unlink_list, node) {
933 hlist_del(&elem->node);
936 sock_map_unref(elem->sk, elem);
938 release_sock(elem->sk);
940 sock_hash_free_elem(htab, elem);
944 /* wait for psock readers accessing its map link */
947 /* wait for psock readers accessing its map link */
950 bpf_map_area_free(htab->buckets);
954 static void sock_hash_release_progs(struct bpf_map *map)
956 psock_progs_drop(&container_of(map, struct bpf_htab, map)->progs);
959 BPF_CALL_4(bpf_sock_hash_update, struct bpf_sock_ops_kern *, sops,
960 struct bpf_map *, map, void *, key, u64, flags)
962 WARN_ON_ONCE(!rcu_read_lock_held());
964 if (likely(sock_map_sk_is_suitable(sops->sk) &&
965 sock_map_op_okay(sops)))
966 return sock_hash_update_common(map, key, sops->sk, flags);
970 const struct bpf_func_proto bpf_sock_hash_update_proto = {
971 .func = bpf_sock_hash_update,
974 .ret_type = RET_INTEGER,
975 .arg1_type = ARG_PTR_TO_CTX,
976 .arg2_type = ARG_CONST_MAP_PTR,
977 .arg3_type = ARG_PTR_TO_MAP_KEY,
978 .arg4_type = ARG_ANYTHING,
981 BPF_CALL_4(bpf_sk_redirect_hash, struct sk_buff *, skb,
982 struct bpf_map *, map, void *, key, u64, flags)
984 struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);
986 if (unlikely(flags & ~(BPF_F_INGRESS)))
988 tcb->bpf.flags = flags;
989 tcb->bpf.sk_redir = __sock_hash_lookup_elem(map, key);
990 if (!tcb->bpf.sk_redir)
995 const struct bpf_func_proto bpf_sk_redirect_hash_proto = {
996 .func = bpf_sk_redirect_hash,
998 .ret_type = RET_INTEGER,
999 .arg1_type = ARG_PTR_TO_CTX,
1000 .arg2_type = ARG_CONST_MAP_PTR,
1001 .arg3_type = ARG_PTR_TO_MAP_KEY,
1002 .arg4_type = ARG_ANYTHING,
1005 BPF_CALL_4(bpf_msg_redirect_hash, struct sk_msg *, msg,
1006 struct bpf_map *, map, void *, key, u64, flags)
1008 if (unlikely(flags & ~(BPF_F_INGRESS)))
1011 msg->sk_redir = __sock_hash_lookup_elem(map, key);
1017 const struct bpf_func_proto bpf_msg_redirect_hash_proto = {
1018 .func = bpf_msg_redirect_hash,
1020 .ret_type = RET_INTEGER,
1021 .arg1_type = ARG_PTR_TO_CTX,
1022 .arg2_type = ARG_CONST_MAP_PTR,
1023 .arg3_type = ARG_PTR_TO_MAP_KEY,
1024 .arg4_type = ARG_ANYTHING,
1027 const struct bpf_map_ops sock_hash_ops = {
1028 .map_alloc = sock_hash_alloc,
1029 .map_free = sock_hash_free,
1030 .map_get_next_key = sock_hash_get_next_key,
1031 .map_update_elem = sock_hash_update_elem,
1032 .map_delete_elem = sock_hash_delete_elem,
1033 .map_lookup_elem = sock_map_lookup,
1034 .map_release_uref = sock_hash_release_progs,
1035 .map_check_btf = map_check_no_btf,
1038 static struct sk_psock_progs *sock_map_progs(struct bpf_map *map)
1040 switch (map->map_type) {
1041 case BPF_MAP_TYPE_SOCKMAP:
1042 return &container_of(map, struct bpf_stab, map)->progs;
1043 case BPF_MAP_TYPE_SOCKHASH:
1044 return &container_of(map, struct bpf_htab, map)->progs;
1052 int sock_map_prog_update(struct bpf_map *map, struct bpf_prog *prog,
1053 struct bpf_prog *old, u32 which)
1055 struct sk_psock_progs *progs = sock_map_progs(map);
1056 struct bpf_prog **pprog;
1062 case BPF_SK_MSG_VERDICT:
1063 pprog = &progs->msg_parser;
1065 case BPF_SK_SKB_STREAM_PARSER:
1066 pprog = &progs->skb_parser;
1068 case BPF_SK_SKB_STREAM_VERDICT:
1069 pprog = &progs->skb_verdict;
1076 return psock_replace_prog(pprog, prog, old);
1078 psock_set_prog(pprog, prog);
1082 void sk_psock_unlink(struct sock *sk, struct sk_psock_link *link)
1084 switch (link->map->map_type) {
1085 case BPF_MAP_TYPE_SOCKMAP:
1086 return sock_map_delete_from_link(link->map, sk,
1088 case BPF_MAP_TYPE_SOCKHASH:
1089 return sock_hash_delete_from_link(link->map, sk,