diff options
32 files changed, 1513 insertions, 140 deletions
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 28705ae67784..71d5ac83cf5d 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -1506,7 +1506,7 @@ union bpf_attr { __s32 map_token_fd; }; - struct { /* anonymous struct used by BPF_MAP_*_ELEM commands */ + struct { /* anonymous struct used by BPF_MAP_*_ELEM and BPF_MAP_FREEZE commands */ __u32 map_fd; __aligned_u64 key; union { @@ -1995,11 +1995,15 @@ union bpf_attr { * long bpf_skb_store_bytes(struct sk_buff *skb, u32 offset, const void *from, u32 len, u64 flags) * Description * Store *len* bytes from address *from* into the packet - * associated to *skb*, at *offset*. *flags* are a combination of - * **BPF_F_RECOMPUTE_CSUM** (automatically recompute the - * checksum for the packet after storing the bytes) and - * **BPF_F_INVALIDATE_HASH** (set *skb*\ **->hash**, *skb*\ - * **->swhash** and *skb*\ **->l4hash** to 0). + * associated to *skb*, at *offset*. The *flags* are a combination + * of the following values: + * + * **BPF_F_RECOMPUTE_CSUM** + * Automatically update *skb*\ **->csum** after storing the + * bytes. + * **BPF_F_INVALIDATE_HASH** + * Set *skb*\ **->hash**, *skb*\ **->swhash** and *skb*\ + * **->l4hash** to 0. * * A call to this helper is susceptible to change the underlying * packet buffer. Therefore, at load time, all checks on pointers @@ -2051,7 +2055,7 @@ union bpf_attr { * untouched (unless **BPF_F_MARK_ENFORCE** is added as well), and * for updates resulting in a null checksum the value is set to * **CSUM_MANGLED_0** instead. Flag **BPF_F_PSEUDO_HDR** indicates - * the checksum is to be computed against a pseudo-header. + * that the modified header field is part of the pseudo-header. * * This helper works in combination with **bpf_csum_diff**\ (), * which does not update the checksum in-place, but offers more diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c index 5a5adc66b8e2..2e18d7e50d9b 100644 --- a/kernel/bpf/hashtab.c +++ b/kernel/bpf/hashtab.c @@ -175,20 +175,30 @@ static bool htab_is_percpu(const struct bpf_htab *htab) htab->map.map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH; } +static inline bool is_fd_htab(const struct bpf_htab *htab) +{ + return htab->map.map_type == BPF_MAP_TYPE_HASH_OF_MAPS; +} + +static inline void *htab_elem_value(struct htab_elem *l, u32 key_size) +{ + return l->key + round_up(key_size, 8); +} + static inline void htab_elem_set_ptr(struct htab_elem *l, u32 key_size, void __percpu *pptr) { - *(void __percpu **)(l->key + roundup(key_size, 8)) = pptr; + *(void __percpu **)htab_elem_value(l, key_size) = pptr; } static inline void __percpu *htab_elem_get_ptr(struct htab_elem *l, u32 key_size) { - return *(void __percpu **)(l->key + roundup(key_size, 8)); + return *(void __percpu **)htab_elem_value(l, key_size); } static void *fd_htab_map_get_ptr(const struct bpf_map *map, struct htab_elem *l) { - return *(void **)(l->key + roundup(map->key_size, 8)); + return *(void **)htab_elem_value(l, map->key_size); } static struct htab_elem *get_htab_elem(struct bpf_htab *htab, int i) @@ -196,9 +206,13 @@ static struct htab_elem *get_htab_elem(struct bpf_htab *htab, int i) return (struct htab_elem *) (htab->elems + i * (u64)htab->elem_size); } +/* Both percpu and fd htab support in-place update, so no need for + * extra elem. LRU itself can remove the least used element, so + * there is no need for an extra elem during map_update. + */ static bool htab_has_extra_elems(struct bpf_htab *htab) { - return !htab_is_percpu(htab) && !htab_is_lru(htab); + return !htab_is_percpu(htab) && !htab_is_lru(htab) && !is_fd_htab(htab); } static void htab_free_prealloced_timers_and_wq(struct bpf_htab *htab) @@ -215,10 +229,10 @@ static void htab_free_prealloced_timers_and_wq(struct bpf_htab *htab) elem = get_htab_elem(htab, i); if (btf_record_has_field(htab->map.record, BPF_TIMER)) bpf_obj_free_timer(htab->map.record, - elem->key + round_up(htab->map.key_size, 8)); + htab_elem_value(elem, htab->map.key_size)); if (btf_record_has_field(htab->map.record, BPF_WORKQUEUE)) bpf_obj_free_workqueue(htab->map.record, - elem->key + round_up(htab->map.key_size, 8)); + htab_elem_value(elem, htab->map.key_size)); cond_resched(); } } @@ -245,7 +259,8 @@ static void htab_free_prealloced_fields(struct bpf_htab *htab) cond_resched(); } } else { - bpf_obj_free_fields(htab->map.record, elem->key + round_up(htab->map.key_size, 8)); + bpf_obj_free_fields(htab->map.record, + htab_elem_value(elem, htab->map.key_size)); cond_resched(); } cond_resched(); @@ -453,8 +468,6 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr) { bool percpu = (attr->map_type == BPF_MAP_TYPE_PERCPU_HASH || attr->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH); - bool lru = (attr->map_type == BPF_MAP_TYPE_LRU_HASH || - attr->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH); /* percpu_lru means each cpu has its own LRU list. * it is different from BPF_MAP_TYPE_PERCPU_HASH where * the map's value itself is percpu. percpu_lru has @@ -549,10 +562,7 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr) if (err) goto free_map_locked; - if (!percpu && !lru) { - /* lru itself can remove the least used element, so - * there is no need for an extra elem during map_update. - */ + if (htab_has_extra_elems(htab)) { err = alloc_extra_elems(htab); if (err) goto free_prealloc; @@ -670,7 +680,7 @@ static void *htab_map_lookup_elem(struct bpf_map *map, void *key) struct htab_elem *l = __htab_map_lookup_elem(map, key); if (l) - return l->key + round_up(map->key_size, 8); + return htab_elem_value(l, map->key_size); return NULL; } @@ -709,7 +719,7 @@ static __always_inline void *__htab_lru_map_lookup_elem(struct bpf_map *map, if (l) { if (mark) bpf_lru_node_set_ref(&l->lru_node); - return l->key + round_up(map->key_size, 8); + return htab_elem_value(l, map->key_size); } return NULL; @@ -763,7 +773,7 @@ static void check_and_free_fields(struct bpf_htab *htab, for_each_possible_cpu(cpu) bpf_obj_free_fields(htab->map.record, per_cpu_ptr(pptr, cpu)); } else { - void *map_value = elem->key + round_up(htab->map.key_size, 8); + void *map_value = htab_elem_value(elem, htab->map.key_size); bpf_obj_free_fields(htab->map.record, map_value); } @@ -968,8 +978,7 @@ static void pcpu_init_value(struct bpf_htab *htab, void __percpu *pptr, static bool fd_htab_map_needs_adjust(const struct bpf_htab *htab) { - return htab->map.map_type == BPF_MAP_TYPE_HASH_OF_MAPS && - BITS_PER_LONG == 64; + return is_fd_htab(htab) && BITS_PER_LONG == 64; } static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key, @@ -1039,11 +1048,9 @@ static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key, htab_elem_set_ptr(l_new, key_size, pptr); } else if (fd_htab_map_needs_adjust(htab)) { size = round_up(size, 8); - memcpy(l_new->key + round_up(key_size, 8), value, size); + memcpy(htab_elem_value(l_new, key_size), value, size); } else { - copy_map_value(&htab->map, - l_new->key + round_up(key_size, 8), - value); + copy_map_value(&htab->map, htab_elem_value(l_new, key_size), value); } l_new->hash = hash; @@ -1072,10 +1079,9 @@ static long htab_map_update_elem(struct bpf_map *map, void *key, void *value, u64 map_flags) { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); - struct htab_elem *l_new = NULL, *l_old; + struct htab_elem *l_new, *l_old; struct hlist_nulls_head *head; unsigned long flags; - void *old_map_ptr; struct bucket *b; u32 key_size, hash; int ret; @@ -1106,7 +1112,7 @@ static long htab_map_update_elem(struct bpf_map *map, void *key, void *value, if (l_old) { /* grab the element lock and update value in place */ copy_map_value_locked(map, - l_old->key + round_up(key_size, 8), + htab_elem_value(l_old, key_size), value, false); return 0; } @@ -1134,7 +1140,7 @@ static long htab_map_update_elem(struct bpf_map *map, void *key, void *value, * and update element in place */ copy_map_value_locked(map, - l_old->key + round_up(key_size, 8), + htab_elem_value(l_old, key_size), value, false); ret = 0; goto err; @@ -1156,24 +1162,14 @@ static long htab_map_update_elem(struct bpf_map *map, void *key, void *value, hlist_nulls_del_rcu(&l_old->hash_node); /* l_old has already been stashed in htab->extra_elems, free - * its special fields before it is available for reuse. Also - * save the old map pointer in htab of maps before unlock - * and release it after unlock. + * its special fields before it is available for reuse. */ - old_map_ptr = NULL; - if (htab_is_prealloc(htab)) { - if (map->ops->map_fd_put_ptr) - old_map_ptr = fd_htab_map_get_ptr(map, l_old); + if (htab_is_prealloc(htab)) check_and_free_fields(htab, l_old); - } } htab_unlock_bucket(b, flags); - if (l_old) { - if (old_map_ptr) - map->ops->map_fd_put_ptr(map, old_map_ptr, true); - if (!htab_is_prealloc(htab)) - free_htab_elem(htab, l_old); - } + if (l_old && !htab_is_prealloc(htab)) + free_htab_elem(htab, l_old); return 0; err: htab_unlock_bucket(b, flags); @@ -1220,8 +1216,7 @@ static long htab_lru_map_update_elem(struct bpf_map *map, void *key, void *value l_new = prealloc_lru_pop(htab, key, hash); if (!l_new) return -ENOMEM; - copy_map_value(&htab->map, - l_new->key + round_up(map->key_size, 8), value); + copy_map_value(&htab->map, htab_elem_value(l_new, map->key_size), value); ret = htab_lock_bucket(b, &flags); if (ret) @@ -1255,13 +1250,14 @@ err_lock_bucket: return ret; } -static long __htab_percpu_map_update_elem(struct bpf_map *map, void *key, +static long htab_map_update_elem_in_place(struct bpf_map *map, void *key, void *value, u64 map_flags, - bool onallcpus) + bool percpu, bool onallcpus) { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); - struct htab_elem *l_new = NULL, *l_old; + struct htab_elem *l_new, *l_old; struct hlist_nulls_head *head; + void *old_map_ptr = NULL; unsigned long flags; struct bucket *b; u32 key_size, hash; @@ -1292,21 +1288,29 @@ static long __htab_percpu_map_update_elem(struct bpf_map *map, void *key, goto err; if (l_old) { - /* per-cpu hash map can update value in-place */ - pcpu_copy_value(htab, htab_elem_get_ptr(l_old, key_size), - value, onallcpus); + /* Update value in-place */ + if (percpu) { + pcpu_copy_value(htab, htab_elem_get_ptr(l_old, key_size), + value, onallcpus); + } else { + void **inner_map_pptr = htab_elem_value(l_old, key_size); + + old_map_ptr = *inner_map_pptr; + WRITE_ONCE(*inner_map_pptr, *(void **)value); + } } else { l_new = alloc_htab_elem(htab, key, value, key_size, - hash, true, onallcpus, NULL); + hash, percpu, onallcpus, NULL); if (IS_ERR(l_new)) { ret = PTR_ERR(l_new); goto err; } hlist_nulls_add_head_rcu(&l_new->hash_node, head); } - ret = 0; err: htab_unlock_bucket(b, flags); + if (old_map_ptr) + map->ops->map_fd_put_ptr(map, old_map_ptr, true); return ret; } @@ -1383,7 +1387,7 @@ err_lock_bucket: static long htab_percpu_map_update_elem(struct bpf_map *map, void *key, void *value, u64 map_flags) { - return __htab_percpu_map_update_elem(map, key, value, map_flags, false); + return htab_map_update_elem_in_place(map, key, value, map_flags, true, false); } static long htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key, @@ -1500,10 +1504,10 @@ static void htab_free_malloced_timers_and_wq(struct bpf_htab *htab) /* We only free timer on uref dropping to zero */ if (btf_record_has_field(htab->map.record, BPF_TIMER)) bpf_obj_free_timer(htab->map.record, - l->key + round_up(htab->map.key_size, 8)); + htab_elem_value(l, htab->map.key_size)); if (btf_record_has_field(htab->map.record, BPF_WORKQUEUE)) bpf_obj_free_workqueue(htab->map.record, - l->key + round_up(htab->map.key_size, 8)); + htab_elem_value(l, htab->map.key_size)); } cond_resched_rcu(); } @@ -1615,15 +1619,12 @@ static int __htab_map_lookup_and_delete_elem(struct bpf_map *map, void *key, off += roundup_value_size; } } else { - u32 roundup_key_size = round_up(map->key_size, 8); + void *src = htab_elem_value(l, map->key_size); if (flags & BPF_F_LOCK) - copy_map_value_locked(map, value, l->key + - roundup_key_size, - true); + copy_map_value_locked(map, value, src, true); else - copy_map_value(map, value, l->key + - roundup_key_size); + copy_map_value(map, value, src); /* Zeroing special fields in the temp buffer */ check_and_init_map_value(map, value); } @@ -1680,12 +1681,12 @@ __htab_map_lookup_and_delete_batch(struct bpf_map *map, bool is_percpu) { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); - u32 bucket_cnt, total, key_size, value_size, roundup_key_size; void *keys = NULL, *values = NULL, *value, *dst_key, *dst_val; void __user *uvalues = u64_to_user_ptr(attr->batch.values); void __user *ukeys = u64_to_user_ptr(attr->batch.keys); void __user *ubatch = u64_to_user_ptr(attr->batch.in_batch); u32 batch, max_count, size, bucket_size, map_id; + u32 bucket_cnt, total, key_size, value_size; struct htab_elem *node_to_free = NULL; u64 elem_map_flags, map_flags; struct hlist_nulls_head *head; @@ -1720,7 +1721,6 @@ __htab_map_lookup_and_delete_batch(struct bpf_map *map, return -ENOENT; key_size = htab->map.key_size; - roundup_key_size = round_up(htab->map.key_size, 8); value_size = htab->map.value_size; size = round_up(value_size, 8); if (is_percpu) @@ -1812,8 +1812,8 @@ again_nocopy: off += size; } } else { - value = l->key + roundup_key_size; - if (map->map_type == BPF_MAP_TYPE_HASH_OF_MAPS) { + value = htab_elem_value(l, key_size); + if (is_fd_htab(htab)) { struct bpf_map **inner_map = value; /* Actual value is the id of the inner map */ @@ -2063,11 +2063,11 @@ static void *bpf_hash_map_seq_next(struct seq_file *seq, void *v, loff_t *pos) static int __bpf_hash_map_seq_show(struct seq_file *seq, struct htab_elem *elem) { struct bpf_iter_seq_hash_map_info *info = seq->private; - u32 roundup_key_size, roundup_value_size; struct bpf_iter__bpf_map_elem ctx = {}; struct bpf_map *map = info->map; struct bpf_iter_meta meta; int ret = 0, off = 0, cpu; + u32 roundup_value_size; struct bpf_prog *prog; void __percpu *pptr; @@ -2077,10 +2077,9 @@ static int __bpf_hash_map_seq_show(struct seq_file *seq, struct htab_elem *elem) ctx.meta = &meta; ctx.map = info->map; if (elem) { - roundup_key_size = round_up(map->key_size, 8); ctx.key = elem->key; if (!info->percpu_value_buf) { - ctx.value = elem->key + roundup_key_size; + ctx.value = htab_elem_value(elem, map->key_size); } else { roundup_value_size = round_up(map->value_size, 8); pptr = htab_elem_get_ptr(elem, map->key_size); @@ -2165,7 +2164,6 @@ static long bpf_for_each_hash_elem(struct bpf_map *map, bpf_callback_t callback_ struct hlist_nulls_head *head; struct hlist_nulls_node *n; struct htab_elem *elem; - u32 roundup_key_size; int i, num_elems = 0; void __percpu *pptr; struct bucket *b; @@ -2180,7 +2178,6 @@ static long bpf_for_each_hash_elem(struct bpf_map *map, bpf_callback_t callback_ is_percpu = htab_is_percpu(htab); - roundup_key_size = round_up(map->key_size, 8); /* migration has been disabled, so percpu value prepared here will be * the same as the one seen by the bpf program with * bpf_map_lookup_elem(). @@ -2196,7 +2193,7 @@ static long bpf_for_each_hash_elem(struct bpf_map *map, bpf_callback_t callback_ pptr = htab_elem_get_ptr(elem, map->key_size); val = this_cpu_ptr(pptr); } else { - val = elem->key + roundup_key_size; + val = htab_elem_value(elem, map->key_size); } num_elems++; ret = callback_fn((u64)(long)map, (u64)(long)key, @@ -2411,8 +2408,8 @@ int bpf_percpu_hash_update(struct bpf_map *map, void *key, void *value, ret = __htab_lru_percpu_map_update_elem(map, key, value, map_flags, true); else - ret = __htab_percpu_map_update_elem(map, key, value, map_flags, - true); + ret = htab_map_update_elem_in_place(map, key, value, map_flags, + true, true); rcu_read_unlock(); return ret; @@ -2536,24 +2533,23 @@ int bpf_fd_htab_map_lookup_elem(struct bpf_map *map, void *key, u32 *value) return ret; } -/* only called from syscall */ +/* Only called from syscall */ int bpf_fd_htab_map_update_elem(struct bpf_map *map, struct file *map_file, void *key, void *value, u64 map_flags) { void *ptr; int ret; - u32 ufd = *(u32 *)value; - ptr = map->ops->map_fd_get_ptr(map, map_file, ufd); + ptr = map->ops->map_fd_get_ptr(map, map_file, *(int *)value); if (IS_ERR(ptr)) return PTR_ERR(ptr); /* The htab bucket lock is always held during update operations in fd * htab map, and the following rcu_read_lock() is only used to avoid - * the WARN_ON_ONCE in htab_map_update_elem(). + * the WARN_ON_ONCE in htab_map_update_elem_in_place(). */ rcu_read_lock(); - ret = htab_map_update_elem(map, key, &ptr, map_flags); + ret = htab_map_update_elem_in_place(map, key, &ptr, map_flags, false, false); rcu_read_unlock(); if (ret) map->ops->map_fd_put_ptr(map, ptr, false); diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 187dc37d61d4..0f5906f43d7c 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -2987,6 +2987,9 @@ int bpf_kprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *pr if (sizeof(u64) != sizeof(void *)) return -EOPNOTSUPP; + if (attr->link_create.flags) + return -EINVAL; + if (!is_kprobe_multi(prog)) return -EINVAL; @@ -3376,6 +3379,9 @@ int bpf_uprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *pr if (sizeof(u64) != sizeof(void *)) return -EOPNOTSUPP; + if (attr->link_create.flags) + return -EINVAL; + if (!is_uprobe_multi(prog)) return -EINVAL; diff --git a/net/core/skmsg.c b/net/core/skmsg.c index 0ddc4c718833..276934673066 100644 --- a/net/core/skmsg.c +++ b/net/core/skmsg.c @@ -530,16 +530,22 @@ static int sk_psock_skb_ingress_enqueue(struct sk_buff *skb, u32 off, u32 len, struct sk_psock *psock, struct sock *sk, - struct sk_msg *msg) + struct sk_msg *msg, + bool take_ref) { int num_sge, copied; + /* skb_to_sgvec will fail when the total number of fragments in + * frag_list and frags exceeds MAX_MSG_FRAGS. For example, the + * caller may aggregate multiple skbs. + */ num_sge = skb_to_sgvec(skb, msg->sg.data, off, len); if (num_sge < 0) { /* skb linearize may fail with ENOMEM, but lets simply try again * later if this happens. Under memory pressure we don't want to * drop the skb. We need to linearize the skb so that the mapping * in skb_to_sgvec can not error. + * Note that skb_linearize requires the skb not to be shared. */ if (skb_linearize(skb)) return -EAGAIN; @@ -556,7 +562,7 @@ static int sk_psock_skb_ingress_enqueue(struct sk_buff *skb, msg->sg.start = 0; msg->sg.size = copied; msg->sg.end = num_sge; - msg->skb = skb; + msg->skb = take_ref ? skb_get(skb) : skb; sk_psock_queue_msg(psock, msg); sk_psock_data_ready(sk, psock); @@ -564,7 +570,7 @@ static int sk_psock_skb_ingress_enqueue(struct sk_buff *skb, } static int sk_psock_skb_ingress_self(struct sk_psock *psock, struct sk_buff *skb, - u32 off, u32 len); + u32 off, u32 len, bool take_ref); static int sk_psock_skb_ingress(struct sk_psock *psock, struct sk_buff *skb, u32 off, u32 len) @@ -578,7 +584,7 @@ static int sk_psock_skb_ingress(struct sk_psock *psock, struct sk_buff *skb, * correctly. */ if (unlikely(skb->sk == sk)) - return sk_psock_skb_ingress_self(psock, skb, off, len); + return sk_psock_skb_ingress_self(psock, skb, off, len, true); msg = sk_psock_create_ingress_msg(sk, skb); if (!msg) return -EAGAIN; @@ -590,7 +596,7 @@ static int sk_psock_skb_ingress(struct sk_psock *psock, struct sk_buff *skb, * into user buffers. */ skb_set_owner_r(skb, sk); - err = sk_psock_skb_ingress_enqueue(skb, off, len, psock, sk, msg); + err = sk_psock_skb_ingress_enqueue(skb, off, len, psock, sk, msg, true); if (err < 0) kfree(msg); return err; @@ -601,7 +607,7 @@ static int sk_psock_skb_ingress(struct sk_psock *psock, struct sk_buff *skb, * because the skb is already accounted for here. */ static int sk_psock_skb_ingress_self(struct sk_psock *psock, struct sk_buff *skb, - u32 off, u32 len) + u32 off, u32 len, bool take_ref) { struct sk_msg *msg = alloc_sk_msg(GFP_ATOMIC); struct sock *sk = psock->sk; @@ -610,7 +616,7 @@ static int sk_psock_skb_ingress_self(struct sk_psock *psock, struct sk_buff *skb if (unlikely(!msg)) return -EAGAIN; skb_set_owner_r(skb, sk); - err = sk_psock_skb_ingress_enqueue(skb, off, len, psock, sk, msg); + err = sk_psock_skb_ingress_enqueue(skb, off, len, psock, sk, msg, take_ref); if (err < 0) kfree(msg); return err; @@ -619,18 +625,13 @@ static int sk_psock_skb_ingress_self(struct sk_psock *psock, struct sk_buff *skb static int sk_psock_handle_skb(struct sk_psock *psock, struct sk_buff *skb, u32 off, u32 len, bool ingress) { - int err = 0; - if (!ingress) { if (!sock_writeable(psock->sk)) return -EAGAIN; return skb_send_sock(psock->sk, skb, off, len); } - skb_get(skb); - err = sk_psock_skb_ingress(psock, skb, off, len); - if (err < 0) - kfree_skb(skb); - return err; + + return sk_psock_skb_ingress(psock, skb, off, len); } static void sk_psock_skb_state(struct sk_psock *psock, @@ -656,11 +657,6 @@ static void sk_psock_backlog(struct work_struct *work) int ret; mutex_lock(&psock->work_mutex); - if (unlikely(state->len)) { - len = state->len; - off = state->off; - } - while ((skb = skb_peek(&psock->ingress_skb))) { len = skb->len; off = 0; @@ -670,6 +666,13 @@ static void sk_psock_backlog(struct work_struct *work) off = stm->offset; len = stm->full_len; } + + /* Resume processing from previous partial state */ + if (unlikely(state->len)) { + len = state->len; + off = state->off; + } + ingress = skb_bpf_ingress(skb); skb_bpf_redirect_clear(skb); do { @@ -680,7 +683,8 @@ static void sk_psock_backlog(struct work_struct *work) if (ret <= 0) { if (ret == -EAGAIN) { sk_psock_skb_state(psock, state, len, off); - + /* Restore redir info we cleared before */ + skb_bpf_set_redir(skb, psock->sk, ingress); /* Delay slightly to prioritize any * other work that might be here. */ @@ -697,6 +701,8 @@ static void sk_psock_backlog(struct work_struct *work) len -= ret; } while (len); + /* The entire skb sent, clear state */ + sk_psock_skb_state(psock, state, 0, 0); skb = skb_dequeue(&psock->ingress_skb); kfree_skb(skb); } @@ -1014,7 +1020,7 @@ static int sk_psock_verdict_apply(struct sk_psock *psock, struct sk_buff *skb, off = stm->offset; len = stm->full_len; } - err = sk_psock_skb_ingress_self(psock, skb, off, len); + err = sk_psock_skb_ingress_self(psock, skb, off, len, false); } if (err < 0) { spin_lock_bh(&psock->ingress_lock); diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 914d4e1516a3..f3d7d19482da 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -1120,9 +1120,13 @@ alloc_encrypted: num_async++; else if (ret == -ENOMEM) goto wait_for_memory; - else if (ctx->open_rec && ret == -ENOSPC) + else if (ctx->open_rec && ret == -ENOSPC) { + if (msg_pl->cork_bytes) { + ret = 0; + goto send_end; + } goto rollback_iter; - else if (ret != -EAGAIN) + } else if (ret != -EAGAIN) goto send_end; } continue; diff --git a/scripts/Makefile.btf b/scripts/Makefile.btf index fbaaec2187e5..db76335dd917 100644 --- a/scripts/Makefile.btf +++ b/scripts/Makefile.btf @@ -23,6 +23,8 @@ else # Switch to using --btf_features for v1.26 and later. pahole-flags-$(call test-ge, $(pahole-ver), 126) = -j$(JOBS) --btf_features=encode_force,var,float,enum64,decl_tag,type_tag,optimized_func,consistent_func,decl_tag_kfuncs +pahole-flags-$(call test-ge, $(pahole-ver), 130) += --btf_features=attributes + ifneq ($(KBUILD_EXTMOD),) module-pahole-flags-$(call test-ge, $(pahole-ver), 128) += --btf_features=distilled_base endif diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index 28705ae67784..71d5ac83cf5d 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -1506,7 +1506,7 @@ union bpf_attr { __s32 map_token_fd; }; - struct { /* anonymous struct used by BPF_MAP_*_ELEM commands */ + struct { /* anonymous struct used by BPF_MAP_*_ELEM and BPF_MAP_FREEZE commands */ __u32 map_fd; __aligned_u64 key; union { @@ -1995,11 +1995,15 @@ union bpf_attr { * long bpf_skb_store_bytes(struct sk_buff *skb, u32 offset, const void *from, u32 len, u64 flags) * Description * Store *len* bytes from address *from* into the packet - * associated to *skb*, at *offset*. *flags* are a combination of - * **BPF_F_RECOMPUTE_CSUM** (automatically recompute the - * checksum for the packet after storing the bytes) and - * **BPF_F_INVALIDATE_HASH** (set *skb*\ **->hash**, *skb*\ - * **->swhash** and *skb*\ **->l4hash** to 0). + * associated to *skb*, at *offset*. The *flags* are a combination + * of the following values: + * + * **BPF_F_RECOMPUTE_CSUM** + * Automatically update *skb*\ **->csum** after storing the + * bytes. + * **BPF_F_INVALIDATE_HASH** + * Set *skb*\ **->hash**, *skb*\ **->swhash** and *skb*\ + * **->l4hash** to 0. * * A call to this helper is susceptible to change the underlying * packet buffer. Therefore, at load time, all checks on pointers @@ -2051,7 +2055,7 @@ union bpf_attr { * untouched (unless **BPF_F_MARK_ENFORCE** is added as well), and * for updates resulting in a null checksum the value is set to * **CSUM_MANGLED_0** instead. Flag **BPF_F_PSEUDO_HDR** indicates - * the checksum is to be computed against a pseudo-header. + * that the modified header field is part of the pseudo-header. * * This helper works in combination with **bpf_csum_diff**\ (), * which does not update the checksum in-place, but offers more diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h index 686824b8b413..a50773d4616e 100644 --- a/tools/lib/bpf/bpf_helpers.h +++ b/tools/lib/bpf/bpf_helpers.h @@ -15,6 +15,14 @@ #define __array(name, val) typeof(val) *name[] #define __ulong(name, val) enum { ___bpf_concat(__unique_value, __COUNTER__) = val } name +#ifndef likely +#define likely(x) (__builtin_expect(!!(x), 1)) +#endif + +#ifndef unlikely +#define unlikely(x) (__builtin_expect(!!(x), 0)) +#endif + /* * Helper macro to place programs, maps, license in * different sections in elf_bpf file. Section names diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c index 38bc6b14b066..24fc71ce5631 100644 --- a/tools/lib/bpf/btf.c +++ b/tools/lib/bpf/btf.c @@ -1148,6 +1148,12 @@ static int btf_find_elf_sections(Elf *elf, const char *path, struct btf_elf_secs else continue; + if (sh.sh_type != SHT_PROGBITS) { + pr_warn("unexpected section type (%d) of section(%d, %s) from %s\n", + sh.sh_type, idx, name, path); + goto err; + } + data = elf_getdata(scn, 0); if (!data) { pr_warn("failed to get section(%d, %s) data from %s\n", diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index 6b85060f07b3..56250b5ac5b0 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -896,7 +896,7 @@ bpf_object__add_programs(struct bpf_object *obj, Elf_Data *sec_data, return -LIBBPF_ERRNO__FORMAT; } - if (sec_off + prog_sz > sec_sz) { + if (sec_off + prog_sz > sec_sz || sec_off + prog_sz < sec_off) { pr_warn("sec '%s': program at offset %zu crosses section boundary\n", sec_name, sec_off); return -LIBBPF_ERRNO__FORMAT; @@ -1725,15 +1725,6 @@ static Elf64_Sym *find_elf_var_sym(const struct bpf_object *obj, const char *nam return ERR_PTR(-ENOENT); } -/* Some versions of Android don't provide memfd_create() in their libc - * implementation, so avoid complications and just go straight to Linux - * syscall. - */ -static int sys_memfd_create(const char *name, unsigned flags) -{ - return syscall(__NR_memfd_create, name, flags); -} - #ifndef MFD_CLOEXEC #define MFD_CLOEXEC 0x0001U #endif @@ -9455,6 +9446,30 @@ int bpf_program__set_log_buf(struct bpf_program *prog, char *log_buf, size_t log return 0; } +struct bpf_func_info *bpf_program__func_info(const struct bpf_program *prog) +{ + if (prog->func_info_rec_size != sizeof(struct bpf_func_info)) + return libbpf_err_ptr(-EOPNOTSUPP); + return prog->func_info; +} + +__u32 bpf_program__func_info_cnt(const struct bpf_program *prog) +{ + return prog->func_info_cnt; +} + +struct bpf_line_info *bpf_program__line_info(const struct bpf_program *prog) +{ + if (prog->line_info_rec_size != sizeof(struct bpf_line_info)) + return libbpf_err_ptr(-EOPNOTSUPP); + return prog->line_info; +} + +__u32 bpf_program__line_info_cnt(const struct bpf_program *prog) +{ + return prog->line_info_cnt; +} + #define SEC_DEF(sec_pfx, ptype, atype, flags, ...) { \ .sec = (char *)sec_pfx, \ .prog_type = BPF_PROG_TYPE_##ptype, \ diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h index e0605403f977..d39f19c8396d 100644 --- a/tools/lib/bpf/libbpf.h +++ b/tools/lib/bpf/libbpf.h @@ -940,6 +940,12 @@ LIBBPF_API int bpf_program__set_log_level(struct bpf_program *prog, __u32 log_le LIBBPF_API const char *bpf_program__log_buf(const struct bpf_program *prog, size_t *log_size); LIBBPF_API int bpf_program__set_log_buf(struct bpf_program *prog, char *log_buf, size_t log_size); +LIBBPF_API struct bpf_func_info *bpf_program__func_info(const struct bpf_program *prog); +LIBBPF_API __u32 bpf_program__func_info_cnt(const struct bpf_program *prog); + +LIBBPF_API struct bpf_line_info *bpf_program__line_info(const struct bpf_program *prog); +LIBBPF_API __u32 bpf_program__line_info_cnt(const struct bpf_program *prog); + /** * @brief **bpf_program__set_attach_target()** sets BTF-based attach target * for supported BPF program types: diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map index d8b71f22f197..1205f9a4fe04 100644 --- a/tools/lib/bpf/libbpf.map +++ b/tools/lib/bpf/libbpf.map @@ -437,6 +437,10 @@ LIBBPF_1.6.0 { bpf_linker__add_fd; bpf_linker__new_fd; bpf_object__prepare; + bpf_program__func_info; + bpf_program__func_info_cnt; + bpf_program__line_info; + bpf_program__line_info_cnt; btf__add_decl_attr; btf__add_type_attr; } LIBBPF_1.5.0; diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h index 76669c73dcd1..477a3b3389a0 100644 --- a/tools/lib/bpf/libbpf_internal.h +++ b/tools/lib/bpf/libbpf_internal.h @@ -667,6 +667,15 @@ static inline int sys_dup3(int oldfd, int newfd, int flags) return syscall(__NR_dup3, oldfd, newfd, flags); } +/* Some versions of Android don't provide memfd_create() in their libc + * implementation, so avoid complications and just go straight to Linux + * syscall. + */ +static inline int sys_memfd_create(const char *name, unsigned flags) +{ + return syscall(__NR_memfd_create, name, flags); +} + /* Point *fixed_fd* to the same file that *tmp_fd* points to. * Regardless of success, *tmp_fd* is closed. * Whatever *fixed_fd* pointed to is closed silently. diff --git a/tools/lib/bpf/linker.c b/tools/lib/bpf/linker.c index 800e0ef09c37..56f5068e2eba 100644 --- a/tools/lib/bpf/linker.c +++ b/tools/lib/bpf/linker.c @@ -573,7 +573,7 @@ int bpf_linker__add_buf(struct bpf_linker *linker, void *buf, size_t buf_sz, snprintf(filename, sizeof(filename), "mem:%p+%zu", buf, buf_sz); - fd = memfd_create(filename, 0); + fd = sys_memfd_create(filename, 0); if (fd < 0) { ret = -errno; pr_warn("failed to create memfd '%s': %s\n", filename, errstr(ret)); diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 66bb50356be0..03663222a0a5 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -811,6 +811,7 @@ $(OUTPUT)/bench_local_storage_create.o: $(OUTPUT)/bench_local_storage_create.ske $(OUTPUT)/bench_bpf_hashmap_lookup.o: $(OUTPUT)/bpf_hashmap_lookup.skel.h $(OUTPUT)/bench_htab_mem.o: $(OUTPUT)/htab_mem_bench.skel.h $(OUTPUT)/bench_bpf_crypto.o: $(OUTPUT)/crypto_bench.skel.h +$(OUTPUT)/bench_sockmap.o: $(OUTPUT)/bench_sockmap_prog.skel.h $(OUTPUT)/bench.o: bench.h testing_helpers.h $(BPFOBJ) $(OUTPUT)/bench: LDLIBS += -lm $(OUTPUT)/bench: $(OUTPUT)/bench.o \ @@ -831,6 +832,7 @@ $(OUTPUT)/bench: $(OUTPUT)/bench.o \ $(OUTPUT)/bench_local_storage_create.o \ $(OUTPUT)/bench_htab_mem.o \ $(OUTPUT)/bench_bpf_crypto.o \ + $(OUTPUT)/bench_sockmap.o \ # $(call msg,BINARY,,$@) $(Q)$(CC) $(CFLAGS) $(LDFLAGS) $(filter %.a %.o,$^) $(LDLIBS) -o $@ diff --git a/tools/testing/selftests/bpf/bench.c b/tools/testing/selftests/bpf/bench.c index 1bd403a5ef7b..c80df9b75094 100644 --- a/tools/testing/selftests/bpf/bench.c +++ b/tools/testing/selftests/bpf/bench.c @@ -283,6 +283,7 @@ extern struct argp bench_local_storage_create_argp; extern struct argp bench_htab_mem_argp; extern struct argp bench_trigger_batch_argp; extern struct argp bench_crypto_argp; +extern struct argp bench_sockmap_argp; static const struct argp_child bench_parsers[] = { { &bench_ringbufs_argp, 0, "Ring buffers benchmark", 0 }, @@ -297,6 +298,7 @@ static const struct argp_child bench_parsers[] = { { &bench_htab_mem_argp, 0, "hash map memory benchmark", 0 }, { &bench_trigger_batch_argp, 0, "BPF triggering benchmark", 0 }, { &bench_crypto_argp, 0, "bpf crypto benchmark", 0 }, + { &bench_sockmap_argp, 0, "bpf sockmap benchmark", 0 }, {}, }; @@ -549,6 +551,7 @@ extern const struct bench bench_local_storage_create; extern const struct bench bench_htab_mem; extern const struct bench bench_crypto_encrypt; extern const struct bench bench_crypto_decrypt; +extern const struct bench bench_sockmap; static const struct bench *benchs[] = { &bench_count_global, @@ -609,6 +612,7 @@ static const struct bench *benchs[] = { &bench_htab_mem, &bench_crypto_encrypt, &bench_crypto_decrypt, + &bench_sockmap, }; static void find_benchmark(void) diff --git a/tools/testing/selftests/bpf/benchs/bench_sockmap.c b/tools/testing/selftests/bpf/benchs/bench_sockmap.c new file mode 100644 index 000000000000..8ebf563a67a2 --- /dev/null +++ b/tools/testing/selftests/bpf/benchs/bench_sockmap.c @@ -0,0 +1,598 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include <error.h> +#include <sys/types.h> +#include <sys/socket.h> +#include <netinet/in.h> +#include <sys/sendfile.h> +#include <arpa/inet.h> +#include <fcntl.h> +#include <argp.h> +#include "bench.h" +#include "bench_sockmap_prog.skel.h" + +#define FILE_SIZE (128 * 1024) +#define DATA_REPEAT_SIZE 10 + +static const char snd_data[DATA_REPEAT_SIZE] = {0, 1, 2, 3, 4, 5, 6, 7, 8, 9}; + +/* c1 <-> [p1, p2] <-> c2 + * RX bench(BPF_SK_SKB_STREAM_VERDICT): + * ARG_FW_RX_PASS: + * send(p2) -> recv(c2) -> bpf skb passthrough -> recv(c2) + * ARG_FW_RX_VERDICT_EGRESS: + * send(c1) -> verdict skb to tx queuec of p2 -> recv(c2) + * ARG_FW_RX_VERDICT_INGRESS: + * send(c1) -> verdict skb to rx queuec of c2 -> recv(c2) + * + * TX bench(BPF_SK_MSG_VERDIC): + * ARG_FW_TX_PASS: + * send(p2) -> bpf msg passthrough -> send(p2) -> recv(c2) + * ARG_FW_TX_VERDICT_INGRESS: + * send(p2) -> verdict msg to rx queue of c2 -> recv(c2) + * ARG_FW_TX_VERDICT_EGRESS: + * send(p1) -> verdict msg to tx queue of p2 -> recv(c2) + */ +enum SOCKMAP_ARG_FLAG { + ARG_FW_RX_NORMAL = 11000, + ARG_FW_RX_PASS, + ARG_FW_RX_VERDICT_EGRESS, + ARG_FW_RX_VERDICT_INGRESS, + ARG_FW_TX_NORMAL, + ARG_FW_TX_PASS, + ARG_FW_TX_VERDICT_INGRESS, + ARG_FW_TX_VERDICT_EGRESS, + ARG_CTL_RX_STRP, + ARG_CONSUMER_DELAY_TIME, + ARG_PRODUCER_DURATION, +}; + +#define TXMODE_NORMAL() \ + ((ctx.mode) == ARG_FW_TX_NORMAL) + +#define TXMODE_BPF_INGRESS() \ + ((ctx.mode) == ARG_FW_TX_VERDICT_INGRESS) + +#define TXMODE_BPF_EGRESS() \ + ((ctx.mode) == ARG_FW_TX_VERDICT_EGRESS) + +#define TXMODE_BPF_PASS() \ + ((ctx.mode) == ARG_FW_TX_PASS) + +#define TXMODE_BPF() ( \ + TXMODE_BPF_PASS() || \ + TXMODE_BPF_INGRESS() || \ + TXMODE_BPF_EGRESS()) + +#define TXMODE() ( \ + TXMODE_NORMAL() || \ + TXMODE_BPF()) + +#define RXMODE_NORMAL() \ + ((ctx.mode) == ARG_FW_RX_NORMAL) + +#define RXMODE_BPF_PASS() \ + ((ctx.mode) == ARG_FW_RX_PASS) + +#define RXMODE_BPF_VERDICT_EGRESS() \ + ((ctx.mode) == ARG_FW_RX_VERDICT_EGRESS) + +#define RXMODE_BPF_VERDICT_INGRESS() \ + ((ctx.mode) == ARG_FW_RX_VERDICT_INGRESS) + +#define RXMODE_BPF_VERDICT() ( \ + RXMODE_BPF_VERDICT_INGRESS() || \ + RXMODE_BPF_VERDICT_EGRESS()) + +#define RXMODE_BPF() ( \ + RXMODE_BPF_PASS() || \ + RXMODE_BPF_VERDICT()) + +#define RXMODE() ( \ + RXMODE_NORMAL() || \ + RXMODE_BPF()) + +static struct socmap_ctx { + struct bench_sockmap_prog *skel; + enum SOCKMAP_ARG_FLAG mode; + #define c1 fds[0] + #define p1 fds[1] + #define c2 fds[2] + #define p2 fds[3] + #define sfd fds[4] + int fds[5]; + long send_calls; + long read_calls; + long prod_send; + long user_read; + int file_size; + int delay_consumer; + int prod_run_time; + int strp_size; +} ctx = { + .prod_send = 0, + .user_read = 0, + .file_size = FILE_SIZE, + .mode = ARG_FW_RX_VERDICT_EGRESS, + .fds = {0}, + .delay_consumer = 0, + .prod_run_time = 0, + .strp_size = 0, +}; + +static void bench_sockmap_prog_destroy(void) +{ + int i; + + for (i = 0; i < sizeof(ctx.fds); i++) { + if (ctx.fds[0] > 0) + close(ctx.fds[i]); + } + + bench_sockmap_prog__destroy(ctx.skel); +} + +static void init_addr(struct sockaddr_storage *ss, + socklen_t *len) +{ + struct sockaddr_in *addr4 = memset(ss, 0, sizeof(*ss)); + + addr4->sin_family = AF_INET; + addr4->sin_port = 0; + addr4->sin_addr.s_addr = htonl(INADDR_LOOPBACK); + *len = sizeof(*addr4); +} + +static bool set_non_block(int fd, bool blocking) +{ + int flags = fcntl(fd, F_GETFL, 0); + + if (flags == -1) + return false; + flags = blocking ? (flags | O_NONBLOCK) : (flags & ~O_NONBLOCK); + return (fcntl(fd, F_SETFL, flags) == 0); +} + +static int create_pair(int *c, int *p, int type) +{ + struct sockaddr_storage addr; + int err, cfd, pfd; + socklen_t addr_len = sizeof(struct sockaddr_storage); + + err = getsockname(ctx.sfd, (struct sockaddr *)&addr, &addr_len); + if (err) { + fprintf(stderr, "getsockname error %d\n", errno); + return err; + } + cfd = socket(AF_INET, type, 0); + if (cfd < 0) { + fprintf(stderr, "socket error %d\n", errno); + return err; + } + + err = connect(cfd, (struct sockaddr *)&addr, addr_len); + if (err && errno != EINPROGRESS) { + fprintf(stderr, "connect error %d\n", errno); + return err; + } + + pfd = accept(ctx.sfd, NULL, NULL); + if (pfd < 0) { + fprintf(stderr, "accept error %d\n", errno); + return err; + } + *c = cfd; + *p = pfd; + return 0; +} + +static int create_sockets(void) +{ + struct sockaddr_storage addr; + int err, one = 1; + socklen_t addr_len; + + init_addr(&addr, &addr_len); + ctx.sfd = socket(AF_INET, SOCK_STREAM, 0); + if (ctx.sfd < 0) { + fprintf(stderr, "socket error:%d\n", errno); + return ctx.sfd; + } + err = setsockopt(ctx.sfd, SOL_SOCKET, SO_REUSEPORT, &one, sizeof(one)); + if (err) { + fprintf(stderr, "setsockopt error:%d\n", errno); + return err; + } + + err = bind(ctx.sfd, (struct sockaddr *)&addr, addr_len); + if (err) { + fprintf(stderr, "bind error:%d\n", errno); + return err; + } + + err = listen(ctx.sfd, SOMAXCONN); + if (err) { + fprintf(stderr, "listen error:%d\n", errno); + return err; + } + + err = create_pair(&ctx.c1, &ctx.p1, SOCK_STREAM); + if (err) { + fprintf(stderr, "create_pair 1 error\n"); + return err; + } + + err = create_pair(&ctx.c2, &ctx.p2, SOCK_STREAM); + if (err) { + fprintf(stderr, "create_pair 2 error\n"); + return err; + } + printf("create socket fd c1:%d p1:%d c2:%d p2:%d\n", + ctx.c1, ctx.p1, ctx.c2, ctx.p2); + return 0; +} + +static void validate(void) +{ + if (env.consumer_cnt != 2 || env.producer_cnt != 1 || + !env.affinity) + goto err; + return; +err: + fprintf(stderr, "argument '-c 2 -p 1 -a' is necessary"); + exit(1); +} + +static int setup_rx_sockmap(void) +{ + int verdict, pass, parser, map; + int zero = 0, one = 1; + int err; + + parser = bpf_program__fd(ctx.skel->progs.prog_skb_parser); + verdict = bpf_program__fd(ctx.skel->progs.prog_skb_verdict); + pass = bpf_program__fd(ctx.skel->progs.prog_skb_pass); + map = bpf_map__fd(ctx.skel->maps.sock_map_rx); + + if (ctx.strp_size != 0) { + ctx.skel->bss->pkt_size = ctx.strp_size; + err = bpf_prog_attach(parser, map, BPF_SK_SKB_STREAM_PARSER, 0); + if (err) + return err; + } + + if (RXMODE_BPF_VERDICT()) + err = bpf_prog_attach(verdict, map, BPF_SK_SKB_STREAM_VERDICT, 0); + else if (RXMODE_BPF_PASS()) + err = bpf_prog_attach(pass, map, BPF_SK_SKB_STREAM_VERDICT, 0); + if (err) + return err; + + if (RXMODE_BPF_PASS()) + return bpf_map_update_elem(map, &zero, &ctx.c2, BPF_NOEXIST); + + err = bpf_map_update_elem(map, &zero, &ctx.p1, BPF_NOEXIST); + if (err < 0) + return err; + + if (RXMODE_BPF_VERDICT_INGRESS()) { + ctx.skel->bss->verdict_dir = BPF_F_INGRESS; + err = bpf_map_update_elem(map, &one, &ctx.c2, BPF_NOEXIST); + } else { + err = bpf_map_update_elem(map, &one, &ctx.p2, BPF_NOEXIST); + } + if (err < 0) + return err; + + return 0; +} + +static int setup_tx_sockmap(void) +{ + int zero = 0, one = 1; + int prog, map; + int err; + + map = bpf_map__fd(ctx.skel->maps.sock_map_tx); + prog = TXMODE_BPF_PASS() ? + bpf_program__fd(ctx.skel->progs.prog_skmsg_pass) : + bpf_program__fd(ctx.skel->progs.prog_skmsg_verdict); + + err = bpf_prog_attach(prog, map, BPF_SK_MSG_VERDICT, 0); + if (err) + return err; + + if (TXMODE_BPF_EGRESS()) { + err = bpf_map_update_elem(map, &zero, &ctx.p1, BPF_NOEXIST); + err |= bpf_map_update_elem(map, &one, &ctx.p2, BPF_NOEXIST); + } else { + ctx.skel->bss->verdict_dir = BPF_F_INGRESS; + err = bpf_map_update_elem(map, &zero, &ctx.p2, BPF_NOEXIST); + err |= bpf_map_update_elem(map, &one, &ctx.c2, BPF_NOEXIST); + } + + if (err < 0) + return err; + + return 0; +} + +static void setup(void) +{ + int err; + + ctx.skel = bench_sockmap_prog__open_and_load(); + if (!ctx.skel) { + fprintf(stderr, "error loading skel\n"); + exit(1); + } + + if (create_sockets()) { + fprintf(stderr, "create_net_mode error\n"); + goto err; + } + + if (RXMODE_BPF()) { + err = setup_rx_sockmap(); + if (err) { + fprintf(stderr, "setup_rx_sockmap error:%d\n", err); + goto err; + } + } else if (TXMODE_BPF()) { + err = setup_tx_sockmap(); + if (err) { + fprintf(stderr, "setup_tx_sockmap error:%d\n", err); + goto err; + } + } else { + fprintf(stderr, "unknown sockmap bench mode: %d\n", ctx.mode); + goto err; + } + + return; + +err: + bench_sockmap_prog_destroy(); + exit(1); +} + +static void measure(struct bench_res *res) +{ + res->drops = atomic_swap(&ctx.prod_send, 0); + res->hits = atomic_swap(&ctx.skel->bss->process_byte, 0); + res->false_hits = atomic_swap(&ctx.user_read, 0); + res->important_hits = atomic_swap(&ctx.send_calls, 0); + res->important_hits |= atomic_swap(&ctx.read_calls, 0) << 32; +} + +static void verify_data(int *check_pos, char *buf, int rcv) +{ + for (int i = 0 ; i < rcv; i++) { + if (buf[i] != snd_data[(*check_pos) % DATA_REPEAT_SIZE]) { + fprintf(stderr, "verify data fail"); + exit(1); + } + (*check_pos)++; + if (*check_pos >= FILE_SIZE) + *check_pos = 0; + } +} + +static void *consumer(void *input) +{ + int rcv, sent; + int check_pos = 0; + int tid = (long)input; + int recv_buf_size = FILE_SIZE; + char *buf = malloc(recv_buf_size); + int delay_read = ctx.delay_consumer; + + if (!buf) { + fprintf(stderr, "fail to init read buffer"); + return NULL; + } + + while (true) { + if (tid == 1) { + /* consumer 1 is unused for tx test and stream verdict test */ + if (RXMODE_BPF() || TXMODE()) + return NULL; + /* it's only for RX_NORMAL which service as reserve-proxy mode */ + rcv = read(ctx.p1, buf, recv_buf_size); + if (rcv < 0) { + fprintf(stderr, "fail to read p1"); + return NULL; + } + + sent = send(ctx.p2, buf, recv_buf_size, 0); + if (sent < 0) { + fprintf(stderr, "fail to send p2"); + return NULL; + } + } else { + if (delay_read != 0) { + if (delay_read < 0) + return NULL; + sleep(delay_read); + delay_read = 0; + } + /* read real endpoint by consumer 0 */ + atomic_inc(&ctx.read_calls); + rcv = read(ctx.c2, buf, recv_buf_size); + if (rcv < 0 && errno != EAGAIN) { + fprintf(stderr, "%s fail to read c2 %d\n", __func__, errno); + return NULL; + } + verify_data(&check_pos, buf, rcv); + atomic_add(&ctx.user_read, rcv); + } + } + + return NULL; +} + +static void *producer(void *input) +{ + int off = 0, fp, need_sent, sent; + int file_size = ctx.file_size; + struct timespec ts1, ts2; + int target; + FILE *file; + + file = tmpfile(); + if (!file) { + fprintf(stderr, "create file for sendfile"); + return NULL; + } + + /* we need simple verify */ + for (int i = 0; i < file_size; i++) { + if (fwrite(&snd_data[off], sizeof(char), 1, file) != 1) { + fprintf(stderr, "init tmpfile error"); + return NULL; + } + if (++off >= sizeof(snd_data)) + off = 0; + } + fflush(file); + fseek(file, 0, SEEK_SET); + + fp = fileno(file); + need_sent = file_size; + clock_gettime(CLOCK_MONOTONIC, &ts1); + + if (RXMODE_BPF_VERDICT()) + target = ctx.c1; + else if (TXMODE_BPF_EGRESS()) + target = ctx.p1; + else + target = ctx.p2; + set_non_block(target, true); + while (true) { + if (ctx.prod_run_time) { + clock_gettime(CLOCK_MONOTONIC, &ts2); + if (ts2.tv_sec - ts1.tv_sec > ctx.prod_run_time) + return NULL; + } + + errno = 0; + atomic_inc(&ctx.send_calls); + sent = sendfile(target, fp, NULL, need_sent); + if (sent < 0) { + if (errno != EAGAIN && errno != ENOMEM && errno != ENOBUFS) { + fprintf(stderr, "sendfile return %d, errorno %d:%s\n", + sent, errno, strerror(errno)); + return NULL; + } + continue; + } else if (sent < need_sent) { + need_sent -= sent; + atomic_add(&ctx.prod_send, sent); + continue; + } + atomic_add(&ctx.prod_send, need_sent); + need_sent = file_size; + lseek(fp, 0, SEEK_SET); + } + + return NULL; +} + +static void report_progress(int iter, struct bench_res *res, long delta_ns) +{ + double speed_mbs, prod_mbs, bpf_mbs, send_hz, read_hz; + + prod_mbs = res->drops / 1000000.0 / (delta_ns / 1000000000.0); + speed_mbs = res->false_hits / 1000000.0 / (delta_ns / 1000000000.0); + bpf_mbs = res->hits / 1000000.0 / (delta_ns / 1000000000.0); + send_hz = (res->important_hits & 0xFFFFFFFF) / (delta_ns / 1000000000.0); + read_hz = (res->important_hits >> 32) / (delta_ns / 1000000000.0); + + printf("Iter %3d (%7.3lfus): ", + iter, (delta_ns - 1000000000) / 1000.0); + printf("Send Speed %8.3lf MB/s (%8.3lf calls/s), BPF Speed %8.3lf MB/s, " + "Rcv Speed %8.3lf MB/s (%8.3lf calls/s)\n", + prod_mbs, send_hz, bpf_mbs, speed_mbs, read_hz); +} + +static void report_final(struct bench_res res[], int res_cnt) +{ + double verdict_mbs_mean = 0.0; + long verdict_total = 0; + int i; + + for (i = 0; i < res_cnt; i++) { + verdict_mbs_mean += res[i].hits / 1000000.0 / (0.0 + res_cnt); + verdict_total += res[i].hits / 1000000.0; + } + + printf("Summary: total trans %8.3lu MB \u00B1 %5.3lf MB/s\n", + verdict_total, verdict_mbs_mean); +} + +static const struct argp_option opts[] = { + { "rx-normal", ARG_FW_RX_NORMAL, NULL, 0, + "simple reserve-proxy mode, no bfp enabled"}, + { "rx-pass", ARG_FW_RX_PASS, NULL, 0, + "run bpf prog but no redir applied"}, + { "rx-strp", ARG_CTL_RX_STRP, "Byte", 0, + "enable strparser and set the encapsulation size"}, + { "rx-verdict-egress", ARG_FW_RX_VERDICT_EGRESS, NULL, 0, + "forward data with bpf(stream verdict)"}, + { "rx-verdict-ingress", ARG_FW_RX_VERDICT_INGRESS, NULL, 0, + "forward data with bpf(stream verdict)"}, + { "tx-normal", ARG_FW_TX_NORMAL, NULL, 0, + "simple c-s mode, no bfp enabled"}, + { "tx-pass", ARG_FW_TX_PASS, NULL, 0, + "run bpf prog but no redir applied"}, + { "tx-verdict-ingress", ARG_FW_TX_VERDICT_INGRESS, NULL, 0, + "forward msg to ingress queue of another socket"}, + { "tx-verdict-egress", ARG_FW_TX_VERDICT_EGRESS, NULL, 0, + "forward msg to egress queue of another socket"}, + { "delay-consumer", ARG_CONSUMER_DELAY_TIME, "SEC", 0, + "delay consumer start"}, + { "producer-duration", ARG_PRODUCER_DURATION, "SEC", 0, + "producer duration"}, + {}, +}; + +static error_t parse_arg(int key, char *arg, struct argp_state *state) +{ + switch (key) { + case ARG_FW_RX_NORMAL...ARG_FW_TX_VERDICT_EGRESS: + ctx.mode = key; + break; + case ARG_CONSUMER_DELAY_TIME: + ctx.delay_consumer = strtol(arg, NULL, 10); + break; + case ARG_PRODUCER_DURATION: + ctx.prod_run_time = strtol(arg, NULL, 10); + break; + case ARG_CTL_RX_STRP: + ctx.strp_size = strtol(arg, NULL, 10); + break; + default: + return ARGP_ERR_UNKNOWN; + } + + return 0; +} + +/* exported into benchmark runner */ +const struct argp bench_sockmap_argp = { + .options = opts, + .parser = parse_arg, +}; + +/* Benchmark performance of creating bpf local storage */ +const struct bench bench_sockmap = { + .name = "sockmap", + .argp = &bench_sockmap_argp, + .validate = validate, + .setup = setup, + .producer_thread = producer, + .consumer_thread = consumer, + .measure = measure, + .report_progress = report_progress, + .report_final = report_final, +}; diff --git a/tools/testing/selftests/bpf/bpf_arena_spin_lock.h b/tools/testing/selftests/bpf/bpf_arena_spin_lock.h index fb8dc0768999..4e29c31c4ef8 100644 --- a/tools/testing/selftests/bpf/bpf_arena_spin_lock.h +++ b/tools/testing/selftests/bpf/bpf_arena_spin_lock.h @@ -95,9 +95,6 @@ struct arena_qnode { #define _Q_LOCKED_VAL (1U << _Q_LOCKED_OFFSET) #define _Q_PENDING_VAL (1U << _Q_PENDING_OFFSET) -#define likely(x) __builtin_expect(!!(x), 1) -#define unlikely(x) __builtin_expect(!!(x), 0) - struct arena_qnode __arena qnodes[_Q_MAX_CPUS][_Q_MAX_NODES]; static inline u32 encode_tail(int cpu, int idx) diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_nf.c b/tools/testing/selftests/bpf/prog_tests/bpf_nf.c index dbd13f8e42a7..dd6512fa652b 100644 --- a/tools/testing/selftests/bpf/prog_tests/bpf_nf.c +++ b/tools/testing/selftests/bpf/prog_tests/bpf_nf.c @@ -63,6 +63,12 @@ static void test_bpf_nf_ct(int mode) .repeat = 1, ); + if (SYS_NOFAIL("iptables-legacy --version")) { + fprintf(stdout, "Missing required iptables-legacy tool\n"); + test__skip(); + return; + } + skel = test_bpf_nf__open_and_load(); if (!ASSERT_OK_PTR(skel, "test_bpf_nf__open_and_load")) return; diff --git a/tools/testing/selftests/bpf/prog_tests/fd_htab_lookup.c b/tools/testing/selftests/bpf/prog_tests/fd_htab_lookup.c new file mode 100644 index 000000000000..ca46fdd6e1ae --- /dev/null +++ b/tools/testing/selftests/bpf/prog_tests/fd_htab_lookup.c @@ -0,0 +1,192 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (C) 2025. Huawei Technologies Co., Ltd */ +#define _GNU_SOURCE +#include <stdbool.h> +#include <test_progs.h> +#include "fd_htab_lookup.skel.h" + +struct htab_op_ctx { + int fd; + int loop; + unsigned int entries; + bool stop; +}; + +#define ERR_TO_RETVAL(where, err) ((void *)(long)(((where) << 12) | (-err))) + +static void *htab_lookup_fn(void *arg) +{ + struct htab_op_ctx *ctx = arg; + int i = 0; + + while (i++ < ctx->loop && !ctx->stop) { + unsigned int j; + + for (j = 0; j < ctx->entries; j++) { + unsigned int key = j, zero = 0, value; + int inner_fd, err; + + err = bpf_map_lookup_elem(ctx->fd, &key, &value); + if (err) { + ctx->stop = true; + return ERR_TO_RETVAL(1, err); + } + + inner_fd = bpf_map_get_fd_by_id(value); + if (inner_fd < 0) { + /* The old map has been freed */ + if (inner_fd == -ENOENT) + continue; + ctx->stop = true; + return ERR_TO_RETVAL(2, inner_fd); + } + + err = bpf_map_lookup_elem(inner_fd, &zero, &value); + if (err) { + close(inner_fd); + ctx->stop = true; + return ERR_TO_RETVAL(3, err); + } + close(inner_fd); + + if (value != key) { + ctx->stop = true; + return ERR_TO_RETVAL(4, -EINVAL); + } + } + } + + return NULL; +} + +static void *htab_update_fn(void *arg) +{ + struct htab_op_ctx *ctx = arg; + int i = 0; + + while (i++ < ctx->loop && !ctx->stop) { + unsigned int j; + + for (j = 0; j < ctx->entries; j++) { + unsigned int key = j, zero = 0; + int inner_fd, err; + + inner_fd = bpf_map_create(BPF_MAP_TYPE_ARRAY, NULL, 4, 4, 1, NULL); + if (inner_fd < 0) { + ctx->stop = true; + return ERR_TO_RETVAL(1, inner_fd); + } + + err = bpf_map_update_elem(inner_fd, &zero, &key, 0); + if (err) { + close(inner_fd); + ctx->stop = true; + return ERR_TO_RETVAL(2, err); + } + + err = bpf_map_update_elem(ctx->fd, &key, &inner_fd, BPF_EXIST); + if (err) { + close(inner_fd); + ctx->stop = true; + return ERR_TO_RETVAL(3, err); + } + close(inner_fd); + } + } + + return NULL; +} + +static int setup_htab(int fd, unsigned int entries) +{ + unsigned int i; + + for (i = 0; i < entries; i++) { + unsigned int key = i, zero = 0; + int inner_fd, err; + + inner_fd = bpf_map_create(BPF_MAP_TYPE_ARRAY, NULL, 4, 4, 1, NULL); + if (!ASSERT_OK_FD(inner_fd, "new array")) + return -1; + + err = bpf_map_update_elem(inner_fd, &zero, &key, 0); + if (!ASSERT_OK(err, "init array")) { + close(inner_fd); + return -1; + } + + err = bpf_map_update_elem(fd, &key, &inner_fd, 0); + if (!ASSERT_OK(err, "init outer")) { + close(inner_fd); + return -1; + } + close(inner_fd); + } + + return 0; +} + +static int get_int_from_env(const char *name, int dft) +{ + const char *value; + + value = getenv(name); + if (!value) + return dft; + + return atoi(value); +} + +void test_fd_htab_lookup(void) +{ + unsigned int i, wr_nr = 8, rd_nr = 16; + pthread_t tids[wr_nr + rd_nr]; + struct fd_htab_lookup *skel; + struct htab_op_ctx ctx; + int err; + + skel = fd_htab_lookup__open_and_load(); + if (!ASSERT_OK_PTR(skel, "fd_htab_lookup__open_and_load")) + return; + + ctx.fd = bpf_map__fd(skel->maps.outer_map); + ctx.loop = get_int_from_env("FD_HTAB_LOOP_NR", 5); + ctx.stop = false; + ctx.entries = 8; + + err = setup_htab(ctx.fd, ctx.entries); + if (err) + goto destroy; + + memset(tids, 0, sizeof(tids)); + for (i = 0; i < wr_nr; i++) { + err = pthread_create(&tids[i], NULL, htab_update_fn, &ctx); + if (!ASSERT_OK(err, "pthread_create")) { + ctx.stop = true; + goto reap; + } + } + for (i = 0; i < rd_nr; i++) { + err = pthread_create(&tids[i + wr_nr], NULL, htab_lookup_fn, &ctx); + if (!ASSERT_OK(err, "pthread_create")) { + ctx.stop = true; + goto reap; + } + } + +reap: + for (i = 0; i < wr_nr + rd_nr; i++) { + void *ret = NULL; + char desc[32]; + + if (!tids[i]) + continue; + + snprintf(desc, sizeof(desc), "thread %u", i + 1); + err = pthread_join(tids[i], &ret); + ASSERT_OK(err, desc); + ASSERT_EQ(ret, NULL, desc); + } +destroy: + fd_htab_lookup__destroy(skel); +} diff --git a/tools/testing/selftests/bpf/prog_tests/sockmap_ktls.c b/tools/testing/selftests/bpf/prog_tests/sockmap_ktls.c index 2d0796314862..49b85c1c7552 100644 --- a/tools/testing/selftests/bpf/prog_tests/sockmap_ktls.c +++ b/tools/testing/selftests/bpf/prog_tests/sockmap_ktls.c @@ -3,13 +3,64 @@ /* * Tests for sockmap/sockhash holding kTLS sockets. */ - +#include <error.h> #include <netinet/tcp.h> +#include <linux/tls.h> #include "test_progs.h" +#include "sockmap_helpers.h" +#include "test_skmsg_load_helpers.skel.h" +#include "test_sockmap_ktls.skel.h" #define MAX_TEST_NAME 80 #define TCP_ULP 31 +static int init_ktls_pairs(int c, int p) +{ + int err; + struct tls12_crypto_info_aes_gcm_128 crypto_rx; + struct tls12_crypto_info_aes_gcm_128 crypto_tx; + + err = setsockopt(c, IPPROTO_TCP, TCP_ULP, "tls", strlen("tls")); + if (!ASSERT_OK(err, "setsockopt(TCP_ULP)")) + goto out; + + err = setsockopt(p, IPPROTO_TCP, TCP_ULP, "tls", strlen("tls")); + if (!ASSERT_OK(err, "setsockopt(TCP_ULP)")) + goto out; + + memset(&crypto_rx, 0, sizeof(crypto_rx)); + memset(&crypto_tx, 0, sizeof(crypto_tx)); + crypto_rx.info.version = TLS_1_2_VERSION; + crypto_tx.info.version = TLS_1_2_VERSION; + crypto_rx.info.cipher_type = TLS_CIPHER_AES_GCM_128; + crypto_tx.info.cipher_type = TLS_CIPHER_AES_GCM_128; + + err = setsockopt(c, SOL_TLS, TLS_TX, &crypto_tx, sizeof(crypto_tx)); + if (!ASSERT_OK(err, "setsockopt(TLS_TX)")) + goto out; + + err = setsockopt(p, SOL_TLS, TLS_RX, &crypto_rx, sizeof(crypto_rx)); + if (!ASSERT_OK(err, "setsockopt(TLS_RX)")) + goto out; + return 0; +out: + return -1; +} + +static int create_ktls_pairs(int family, int sotype, int *c, int *p) +{ + int err; + + err = create_pair(family, sotype, c, p); + if (!ASSERT_OK(err, "create_pair()")) + return -1; + + err = init_ktls_pairs(*c, *p); + if (!ASSERT_OK(err, "init_ktls_pairs(c, p)")) + return -1; + return 0; +} + static int tcp_server(int family) { int err, s; @@ -146,6 +197,115 @@ static const char *fmt_test_name(const char *subtest_name, int family, return test_name; } +static void test_sockmap_ktls_offload(int family, int sotype) +{ + int err; + int c = 0, p = 0, sent, recvd; + char msg[12] = "hello world\0"; + char rcv[13]; + + err = create_ktls_pairs(family, sotype, &c, &p); + if (!ASSERT_OK(err, "create_ktls_pairs()")) + goto out; + + sent = send(c, msg, sizeof(msg), 0); + if (!ASSERT_OK(err, "send(msg)")) + goto out; + + recvd = recv(p, rcv, sizeof(rcv), 0); + if (!ASSERT_OK(err, "recv(msg)") || + !ASSERT_EQ(recvd, sent, "length mismatch")) + goto out; + + ASSERT_OK(memcmp(msg, rcv, sizeof(msg)), "data mismatch"); + +out: + if (c) + close(c); + if (p) + close(p); +} + +static void test_sockmap_ktls_tx_cork(int family, int sotype, bool push) +{ + int err, off; + int i, j; + int start_push = 0, push_len = 0; + int c = 0, p = 0, one = 1, sent, recvd; + int prog_fd, map_fd; + char msg[12] = "hello world\0"; + char rcv[20] = {0}; + struct test_sockmap_ktls *skel; + + skel = test_sockmap_ktls__open_and_load(); + if (!ASSERT_TRUE(skel, "open ktls skel")) + return; + + err = create_pair(family, sotype, &c, &p); + if (!ASSERT_OK(err, "create_pair()")) + goto out; + + prog_fd = bpf_program__fd(skel->progs.prog_sk_policy); + map_fd = bpf_map__fd(skel->maps.sock_map); + + err = bpf_prog_attach(prog_fd, map_fd, BPF_SK_MSG_VERDICT, 0); + if (!ASSERT_OK(err, "bpf_prog_attach sk msg")) + goto out; + + err = bpf_map_update_elem(map_fd, &one, &c, BPF_NOEXIST); + if (!ASSERT_OK(err, "bpf_map_update_elem(c)")) + goto out; + + err = init_ktls_pairs(c, p); + if (!ASSERT_OK(err, "init_ktls_pairs(c, p)")) + goto out; + + skel->bss->cork_byte = sizeof(msg); + if (push) { + start_push = 1; + push_len = 2; + } + skel->bss->push_start = start_push; + skel->bss->push_end = push_len; + + off = sizeof(msg) / 2; + sent = send(c, msg, off, 0); + if (!ASSERT_EQ(sent, off, "send(msg)")) + goto out; + + recvd = recv_timeout(p, rcv, sizeof(rcv), MSG_DONTWAIT, 1); + if (!ASSERT_EQ(-1, recvd, "expected no data")) + goto out; + + /* send remaining msg */ + sent = send(c, msg + off, sizeof(msg) - off, 0); + if (!ASSERT_EQ(sent, sizeof(msg) - off, "send remaining data")) + goto out; + + recvd = recv_timeout(p, rcv, sizeof(rcv), MSG_DONTWAIT, 1); + if (!ASSERT_OK(err, "recv(msg)") || + !ASSERT_EQ(recvd, sizeof(msg) + push_len, "check length mismatch")) + goto out; + + for (i = 0, j = 0; i < recvd;) { + /* skip checking the data that has been pushed in */ + if (i >= start_push && i <= start_push + push_len - 1) { + i++; + continue; + } + if (!ASSERT_EQ(rcv[i], msg[j], "data mismatch")) + goto out; + i++; + j++; + } +out: + if (c) + close(c); + if (p) + close(p); + test_sockmap_ktls__destroy(skel); +} + static void run_tests(int family, enum bpf_map_type map_type) { int map; @@ -162,10 +322,22 @@ static void run_tests(int family, enum bpf_map_type map_type) close(map); } +static void run_ktls_test(int family, int sotype) +{ + if (test__start_subtest("tls simple offload")) + test_sockmap_ktls_offload(family, sotype); + if (test__start_subtest("tls tx cork")) + test_sockmap_ktls_tx_cork(family, sotype, false); + if (test__start_subtest("tls tx cork with push")) + test_sockmap_ktls_tx_cork(family, sotype, true); +} + void test_sockmap_ktls(void) { run_tests(AF_INET, BPF_MAP_TYPE_SOCKMAP); run_tests(AF_INET, BPF_MAP_TYPE_SOCKHASH); run_tests(AF_INET6, BPF_MAP_TYPE_SOCKMAP); run_tests(AF_INET6, BPF_MAP_TYPE_SOCKHASH); + run_ktls_test(AF_INET, SOCK_STREAM); + run_ktls_test(AF_INET6, SOCK_STREAM); } diff --git a/tools/testing/selftests/bpf/prog_tests/test_btf_ext.c b/tools/testing/selftests/bpf/prog_tests/test_btf_ext.c new file mode 100644 index 000000000000..7d1b478c99a0 --- /dev/null +++ b/tools/testing/selftests/bpf/prog_tests/test_btf_ext.c @@ -0,0 +1,64 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2025 Meta Platforms Inc. */ +#include <test_progs.h> +#include "test_btf_ext.skel.h" +#include "btf_helpers.h" + +static void subtest_line_func_info(void) +{ + struct test_btf_ext *skel; + struct bpf_prog_info info; + struct bpf_line_info line_info[128], *libbpf_line_info; + struct bpf_func_info func_info[128], *libbpf_func_info; + __u32 info_len = sizeof(info), libbbpf_line_info_cnt, libbbpf_func_info_cnt; + int err, fd; + + skel = test_btf_ext__open_and_load(); + if (!ASSERT_OK_PTR(skel, "skel_open_and_load")) + return; + + fd = bpf_program__fd(skel->progs.global_func); + + memset(&info, 0, sizeof(info)); + info.line_info = ptr_to_u64(&line_info); + info.nr_line_info = sizeof(line_info); + info.line_info_rec_size = sizeof(*line_info); + err = bpf_prog_get_info_by_fd(fd, &info, &info_len); + if (!ASSERT_OK(err, "prog_line_info")) + goto out; + + libbpf_line_info = bpf_program__line_info(skel->progs.global_func); + libbbpf_line_info_cnt = bpf_program__line_info_cnt(skel->progs.global_func); + + memset(&info, 0, sizeof(info)); + info.func_info = ptr_to_u64(&func_info); + info.nr_func_info = sizeof(func_info); + info.func_info_rec_size = sizeof(*func_info); + err = bpf_prog_get_info_by_fd(fd, &info, &info_len); + if (!ASSERT_OK(err, "prog_func_info")) + goto out; + + libbpf_func_info = bpf_program__func_info(skel->progs.global_func); + libbbpf_func_info_cnt = bpf_program__func_info_cnt(skel->progs.global_func); + + if (!ASSERT_OK_PTR(libbpf_line_info, "bpf_program__line_info")) + goto out; + if (!ASSERT_EQ(libbbpf_line_info_cnt, info.nr_line_info, "line_info_cnt")) + goto out; + if (!ASSERT_OK_PTR(libbpf_func_info, "bpf_program__func_info")) + goto out; + if (!ASSERT_EQ(libbbpf_func_info_cnt, info.nr_func_info, "func_info_cnt")) + goto out; + ASSERT_MEMEQ(libbpf_line_info, line_info, libbbpf_line_info_cnt * sizeof(*line_info), + "line_info"); + ASSERT_MEMEQ(libbpf_func_info, func_info, libbbpf_func_info_cnt * sizeof(*func_info), + "func_info"); +out: + test_btf_ext__destroy(skel); +} + +void test_btf_ext(void) +{ + if (test__start_subtest("line_func_info")) + subtest_line_func_info(); +} diff --git a/tools/testing/selftests/bpf/prog_tests/test_veristat.c b/tools/testing/selftests/bpf/prog_tests/test_veristat.c index a95b42bf744a..47b56c258f3f 100644 --- a/tools/testing/selftests/bpf/prog_tests/test_veristat.c +++ b/tools/testing/selftests/bpf/prog_tests/test_veristat.c @@ -63,6 +63,9 @@ static void test_set_global_vars_succeeds(void) " -G \"var_eb = EB2\" "\ " -G \"var_ec = EC2\" "\ " -G \"var_b = 1\" "\ + " -G \"struct1.struct2.u.var_u8 = 170\" "\ + " -G \"union1.struct3.var_u8_l = 0xaa\" "\ + " -G \"union1.struct3.var_u8_h = 0xaa\" "\ "-vl2 > %s", fix->veristat, fix->tmpfile); read(fix->fd, fix->output, fix->sz); @@ -78,6 +81,8 @@ static void test_set_global_vars_succeeds(void) __CHECK_STR("_w=12 ", "var_eb = EB2"); __CHECK_STR("_w=13 ", "var_ec = EC2"); __CHECK_STR("_w=1 ", "var_b = 1"); + __CHECK_STR("_w=170 ", "struct1.struct2.u.var_u8 = 170"); + __CHECK_STR("_w=0xaaaa ", "union1.var_u16 = 0xaaaa"); out: teardown_fixture(fix); diff --git a/tools/testing/selftests/bpf/progs/bench_sockmap_prog.c b/tools/testing/selftests/bpf/progs/bench_sockmap_prog.c new file mode 100644 index 000000000000..079bf3794b3a --- /dev/null +++ b/tools/testing/selftests/bpf/progs/bench_sockmap_prog.c @@ -0,0 +1,65 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/bpf.h> +#include <bpf/bpf_helpers.h> +#include <bpf/bpf_endian.h> + +long process_byte = 0; +int verdict_dir = 0; +int dropped = 0; +int pkt_size = 0; +struct { + __uint(type, BPF_MAP_TYPE_SOCKMAP); + __uint(max_entries, 20); + __type(key, int); + __type(value, int); +} sock_map_rx SEC(".maps"); + +struct { + __uint(type, BPF_MAP_TYPE_SOCKMAP); + __uint(max_entries, 20); + __type(key, int); + __type(value, int); +} sock_map_tx SEC(".maps"); + +SEC("sk_skb/stream_parser") +int prog_skb_parser(struct __sk_buff *skb) +{ + return pkt_size; +} + +SEC("sk_skb/stream_verdict") +int prog_skb_verdict(struct __sk_buff *skb) +{ + int one = 1; + int ret = bpf_sk_redirect_map(skb, &sock_map_rx, one, verdict_dir); + + if (ret == SK_DROP) + dropped++; + __sync_fetch_and_add(&process_byte, skb->len); + return ret; +} + +SEC("sk_skb/stream_verdict") +int prog_skb_pass(struct __sk_buff *skb) +{ + __sync_fetch_and_add(&process_byte, skb->len); + return SK_PASS; +} + +SEC("sk_msg") +int prog_skmsg_verdict(struct sk_msg_md *msg) +{ + int one = 1; + + __sync_fetch_and_add(&process_byte, msg->size); + return bpf_msg_redirect_map(msg, &sock_map_tx, one, verdict_dir); +} + +SEC("sk_msg") +int prog_skmsg_pass(struct sk_msg_md *msg) +{ + __sync_fetch_and_add(&process_byte, msg->size); + return SK_PASS; +} + +char _license[] SEC("license") = "GPL"; diff --git a/tools/testing/selftests/bpf/progs/fd_htab_lookup.c b/tools/testing/selftests/bpf/progs/fd_htab_lookup.c new file mode 100644 index 000000000000..a4a9e1db626f --- /dev/null +++ b/tools/testing/selftests/bpf/progs/fd_htab_lookup.c @@ -0,0 +1,25 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (C) 2025. Huawei Technologies Co., Ltd */ +#include <linux/bpf.h> +#include <bpf/bpf_helpers.h> + +char _license[] SEC("license") = "GPL"; + +struct inner_map_type { + __uint(type, BPF_MAP_TYPE_ARRAY); + __uint(key_size, 4); + __uint(value_size, 4); + __uint(max_entries, 1); +} inner_map SEC(".maps"); + +struct { + __uint(type, BPF_MAP_TYPE_HASH_OF_MAPS); + __uint(max_entries, 64); + __type(key, int); + __type(value, int); + __array(values, struct inner_map_type); +} outer_map SEC(".maps") = { + .values = { + [0] = &inner_map, + }, +}; diff --git a/tools/testing/selftests/bpf/progs/iters.c b/tools/testing/selftests/bpf/progs/iters.c index 427b72954b87..76adf4a8f2da 100644 --- a/tools/testing/selftests/bpf/progs/iters.c +++ b/tools/testing/selftests/bpf/progs/iters.c @@ -7,8 +7,6 @@ #include "bpf_misc.h" #include "bpf_compiler.h" -#define unlikely(x) __builtin_expect(!!(x), 0) - static volatile int zero = 0; int my_pid; diff --git a/tools/testing/selftests/bpf/progs/prepare.c b/tools/testing/selftests/bpf/progs/prepare.c index 1f1dd547e4ee..cfc1f48e0d28 100644 --- a/tools/testing/selftests/bpf/progs/prepare.c +++ b/tools/testing/selftests/bpf/progs/prepare.c @@ -2,7 +2,6 @@ /* Copyright (c) 2025 Meta */ #include <vmlinux.h> #include <bpf/bpf_helpers.h> -//#include <bpf/bpf_tracing.h> char _license[] SEC("license") = "GPL"; diff --git a/tools/testing/selftests/bpf/progs/set_global_vars.c b/tools/testing/selftests/bpf/progs/set_global_vars.c index 9adb5ba4cd4d..90f5656c3991 100644 --- a/tools/testing/selftests/bpf/progs/set_global_vars.c +++ b/tools/testing/selftests/bpf/progs/set_global_vars.c @@ -24,6 +24,44 @@ const volatile enum Enumu64 var_eb = EB1; const volatile enum Enums64 var_ec = EC1; const volatile bool var_b = false; +struct Struct { + int:16; + __u16 filler; + struct { + const __u16 filler2; + }; + struct Struct2 { + __u16 filler; + volatile struct { + const int:1; + union { + const volatile __u8 var_u8; + const volatile __s16 filler3; + const int:1; + } u; + }; + } struct2; +}; + +const volatile __u32 stru = 0; /* same prefix as below */ +const volatile struct Struct struct1 = {.struct2 = {.u = {.var_u8 = 1}}}; + +union Union { + __u16 var_u16; + struct Struct3 { + struct { + __u8 var_u8_l; + }; + struct { + struct { + __u8 var_u8_h; + }; + }; + } struct3; +}; + +const volatile union Union union1 = {.var_u16 = -1}; + char arr[4] = {0}; SEC("socket") @@ -43,5 +81,8 @@ int test_set_globals(void *ctx) a = var_eb; a = var_ec; a = var_b; + a = struct1.struct2.u.var_u8; + a = union1.var_u16; + return a; } diff --git a/tools/testing/selftests/bpf/progs/test_btf_ext.c b/tools/testing/selftests/bpf/progs/test_btf_ext.c new file mode 100644 index 000000000000..cdf20331db04 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/test_btf_ext.c @@ -0,0 +1,22 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2025 Meta Platforms Inc. */ + +#include <linux/bpf.h> +#include <bpf/bpf_helpers.h> +#include "bpf_misc.h" + +char _license[] SEC("license") = "GPL"; + +__noinline static void f0(void) +{ + __u64 a = 1; + + __sink(a); +} + +SEC("xdp") +__u64 global_func(struct xdp_md *xdp) +{ + f0(); + return XDP_DROP; +} diff --git a/tools/testing/selftests/bpf/progs/test_sockmap_ktls.c b/tools/testing/selftests/bpf/progs/test_sockmap_ktls.c new file mode 100644 index 000000000000..e0f757929ef4 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/test_sockmap_ktls.c @@ -0,0 +1,26 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/bpf.h> +#include <bpf/bpf_helpers.h> +#include <bpf/bpf_endian.h> + +int cork_byte; +int push_start; +int push_end; + +struct { + __uint(type, BPF_MAP_TYPE_SOCKMAP); + __uint(max_entries, 20); + __type(key, int); + __type(value, int); +} sock_map SEC(".maps"); + +SEC("sk_msg") +int prog_sk_policy(struct sk_msg_md *msg) +{ + if (cork_byte > 0) + bpf_msg_cork_bytes(msg, cork_byte); + if (push_start > 0 && push_end > 0) + bpf_msg_push_data(msg, push_start, push_end, 0); + + return SK_PASS; +} diff --git a/tools/testing/selftests/bpf/test_kmods/bpf_testmod.c b/tools/testing/selftests/bpf/test_kmods/bpf_testmod.c index 3220f1d28697..f38eaf0d35ef 100644 --- a/tools/testing/selftests/bpf/test_kmods/bpf_testmod.c +++ b/tools/testing/selftests/bpf/test_kmods/bpf_testmod.c @@ -1340,7 +1340,7 @@ static int st_ops_gen_prologue_with_kfunc(struct bpf_insn *insn_buf, bool direct *insn++ = BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_7, offsetof(struct st_ops_args, a)); *insn++ = BPF_JMP_IMM(BPF_JA, 0, 0, 2); *insn++ = BPF_MOV64_REG(BPF_REG_1, BPF_REG_0); - *insn++ = BPF_CALL_KFUNC(0, bpf_cgroup_release_id), + *insn++ = BPF_CALL_KFUNC(0, bpf_cgroup_release_id); *insn++ = BPF_MOV64_REG(BPF_REG_1, BPF_REG_8); *insn++ = prog->insnsi[0]; @@ -1379,7 +1379,7 @@ static int st_ops_gen_epilogue_with_kfunc(struct bpf_insn *insn_buf, const struc *insn++ = BPF_STX_MEM(BPF_DW, BPF_REG_1, BPF_REG_6, offsetof(struct st_ops_args, a)); *insn++ = BPF_JMP_IMM(BPF_JA, 0, 0, 2); *insn++ = BPF_MOV64_REG(BPF_REG_1, BPF_REG_0); - *insn++ = BPF_CALL_KFUNC(0, bpf_cgroup_release_id), + *insn++ = BPF_CALL_KFUNC(0, bpf_cgroup_release_id); *insn++ = BPF_MOV64_REG(BPF_REG_0, BPF_REG_6); *insn++ = BPF_ALU64_IMM(BPF_MUL, BPF_REG_0, 2); *insn++ = BPF_EXIT_INSN(); diff --git a/tools/testing/selftests/bpf/veristat.c b/tools/testing/selftests/bpf/veristat.c index a18972ffdeb6..b2bb20b00952 100644 --- a/tools/testing/selftests/bpf/veristat.c +++ b/tools/testing/selftests/bpf/veristat.c @@ -1486,7 +1486,84 @@ static bool is_preset_supported(const struct btf_type *t) return btf_is_int(t) || btf_is_enum(t) || btf_is_enum64(t); } -static int set_global_var(struct bpf_object *obj, struct btf *btf, const struct btf_type *t, +const int btf_find_member(const struct btf *btf, + const struct btf_type *parent_type, + __u32 parent_offset, + const char *member_name, + int *member_tid, + __u32 *member_offset) +{ + int i; + + if (!btf_is_composite(parent_type)) + return -EINVAL; + + for (i = 0; i < btf_vlen(parent_type); ++i) { + const struct btf_member *member; + const struct btf_type *member_type; + int tid; + + member = btf_members(parent_type) + i; + tid = btf__resolve_type(btf, member->type); + if (tid < 0) + return -EINVAL; + + member_type = btf__type_by_id(btf, tid); + if (member->name_off) { + const char *name = btf__name_by_offset(btf, member->name_off); + + if (strcmp(member_name, name) == 0) { + if (btf_member_bitfield_size(parent_type, i) != 0) { + fprintf(stderr, "Bitfield presets are not supported %s\n", + name); + return -EINVAL; + } + *member_offset = parent_offset + member->offset; + *member_tid = tid; + return 0; + } + } else if (btf_is_composite(member_type)) { + int err; + + err = btf_find_member(btf, member_type, parent_offset + member->offset, + member_name, member_tid, member_offset); + if (!err) + return 0; + } + } + + return -EINVAL; +} + +static int adjust_var_secinfo(struct btf *btf, const struct btf_type *t, + struct btf_var_secinfo *sinfo, const char *var) +{ + char expr[256], *saveptr; + const struct btf_type *base_type, *member_type; + int err, member_tid; + char *name; + __u32 member_offset = 0; + + base_type = btf__type_by_id(btf, btf__resolve_type(btf, t->type)); + snprintf(expr, sizeof(expr), "%s", var); + strtok_r(expr, ".", &saveptr); + + while ((name = strtok_r(NULL, ".", &saveptr))) { + err = btf_find_member(btf, base_type, 0, name, &member_tid, &member_offset); + if (err) { + fprintf(stderr, "Could not find member %s for variable %s\n", name, var); + return err; + } + member_type = btf__type_by_id(btf, member_tid); + sinfo->offset += member_offset / 8; + sinfo->size = member_type->size; + sinfo->type = member_tid; + base_type = member_type; + } + return 0; +} + +static int set_global_var(struct bpf_object *obj, struct btf *btf, struct bpf_map *map, struct btf_var_secinfo *sinfo, struct var_preset *preset) { @@ -1495,9 +1572,9 @@ static int set_global_var(struct bpf_object *obj, struct btf *btf, const struct long long value = preset->ivalue; size_t size; - base_type = btf__type_by_id(btf, btf__resolve_type(btf, t->type)); + base_type = btf__type_by_id(btf, btf__resolve_type(btf, sinfo->type)); if (!base_type) { - fprintf(stderr, "Failed to resolve type %d\n", t->type); + fprintf(stderr, "Failed to resolve type %d\n", sinfo->type); return -EINVAL; } if (!is_preset_supported(base_type)) { @@ -1530,7 +1607,7 @@ static int set_global_var(struct bpf_object *obj, struct btf *btf, const struct if (value >= max_val || value < -max_val) { fprintf(stderr, "Variable %s value %lld is out of range [%lld; %lld]\n", - btf__name_by_offset(btf, t->name_off), value, + btf__name_by_offset(btf, base_type->name_off), value, is_signed ? -max_val : 0, max_val - 1); return -EINVAL; } @@ -1583,14 +1660,20 @@ static int set_global_vars(struct bpf_object *obj, struct var_preset *presets, i for (j = 0; j < n; ++j, ++sinfo) { const struct btf_type *var_type = btf__type_by_id(btf, sinfo->type); const char *var_name; + int var_len; if (!btf_is_var(var_type)) continue; var_name = btf__name_by_offset(btf, var_type->name_off); + var_len = strlen(var_name); for (k = 0; k < npresets; ++k) { - if (strcmp(var_name, presets[k].name) != 0) + struct btf_var_secinfo tmp_sinfo; + + if (strncmp(var_name, presets[k].name, var_len) != 0 || + (presets[k].name[var_len] != '\0' && + presets[k].name[var_len] != '.')) continue; if (presets[k].applied) { @@ -1598,13 +1681,17 @@ static int set_global_vars(struct bpf_object *obj, struct var_preset *presets, i var_name); return -EINVAL; } + tmp_sinfo = *sinfo; + err = adjust_var_secinfo(btf, var_type, + &tmp_sinfo, presets[k].name); + if (err) + return err; - err = set_global_var(obj, btf, var_type, map, sinfo, presets + k); + err = set_global_var(obj, btf, map, &tmp_sinfo, presets + k); if (err) return err; presets[k].applied = true; - break; } } } |