diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 6e947cd91152..6d7346c54d83 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -573,7 +573,6 @@ struct bpf_verifier_ops {
 				 const struct btf_type *t, int off, int size,
 				 enum bpf_access_type atype,
 				 u32 *next_btf_id);
-	bool (*check_kfunc_call)(u32 kfunc_btf_id, struct module *owner);
 };
 
 struct bpf_prog_offload_ops {
@@ -1719,7 +1718,6 @@ int bpf_prog_test_run_raw_tp(struct bpf_prog *prog,
 int bpf_prog_test_run_sk_lookup(struct bpf_prog *prog,
 				const union bpf_attr *kattr,
 				union bpf_attr __user *uattr);
-bool bpf_prog_test_check_kfunc_call(u32 kfunc_id, struct module *owner);
 bool btf_ctx_access(int off, int size, enum bpf_access_type type,
 		    const struct bpf_prog *prog,
 		    struct bpf_insn_access_aux *info);
@@ -1971,12 +1969,6 @@ static inline int bpf_prog_test_run_sk_lookup(struct bpf_prog *prog,
 	return -ENOTSUPP;
 }
 
-static inline bool bpf_prog_test_check_kfunc_call(u32 kfunc_id,
-						  struct module *owner)
-{
-	return false;
-}
-
 static inline void bpf_map_put(struct bpf_map *map)
 {
 }
diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index 143401d4c9d9..ac4797155412 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -521,6 +521,8 @@ bpf_prog_offload_remove_insns(struct bpf_verifier_env *env, u32 off, u32 cnt);
 
 int check_ctx_reg(struct bpf_verifier_env *env,
 		  const struct bpf_reg_state *reg, int regno);
+int check_kfunc_mem_size_reg(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
+			     u32 regno);
 int check_mem_reg(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
 		   u32 regno, u32 mem_size);
 
@@ -564,4 +566,9 @@ static inline u32 type_flag(u32 type)
 	return type & ~BPF_BASE_TYPE_MASK;
 }
 
+static inline enum bpf_prog_type resolve_prog_type(struct bpf_prog *prog)
+{
+	return prog->aux->dst_prog ? prog->aux->dst_prog->type : prog->type;
+}
+
 #endif /* _LINUX_BPF_VERIFIER_H */
diff --git a/include/linux/btf.h b/include/linux/btf.h
index 0c74348cbc9d..b12cfe3b12bb 100644
--- a/include/linux/btf.h
+++ b/include/linux/btf.h
@@ -12,11 +12,33 @@
 #define BTF_TYPE_EMIT(type) ((void)(type *)0)
 #define BTF_TYPE_EMIT_ENUM(enum_val) ((void)enum_val)
 
+enum btf_kfunc_type {
+	BTF_KFUNC_TYPE_CHECK,
+	BTF_KFUNC_TYPE_ACQUIRE,
+	BTF_KFUNC_TYPE_RELEASE,
+	BTF_KFUNC_TYPE_RET_NULL,
+	BTF_KFUNC_TYPE_MAX,
+};
+
 struct btf;
 struct btf_member;
 struct btf_type;
 union bpf_attr;
 struct btf_show;
+struct btf_id_set;
+
+struct btf_kfunc_id_set {
+	struct module *owner;
+	union {
+		struct {
+			struct btf_id_set *check_set;
+			struct btf_id_set *acquire_set;
+			struct btf_id_set *release_set;
+			struct btf_id_set *ret_null_set;
+		};
+		struct btf_id_set *sets[BTF_KFUNC_TYPE_MAX];
+	};
+};
 
 extern const struct file_operations btf_fops;
 
@@ -307,6 +329,11 @@ const struct btf_type *btf_type_by_id(const struct btf *btf, u32 type_id);
 const char *btf_name_by_offset(const struct btf *btf, u32 offset);
 struct btf *btf_parse_vmlinux(void);
 struct btf *bpf_prog_get_target_btf(const struct bpf_prog *prog);
+bool btf_kfunc_id_set_contains(const struct btf *btf,
+			       enum bpf_prog_type prog_type,
+			       enum btf_kfunc_type type, u32 kfunc_btf_id);
+int register_btf_kfunc_id_set(enum bpf_prog_type prog_type,
+			      const struct btf_kfunc_id_set *s);
 #else
 static inline const struct btf_type *btf_type_by_id(const struct btf *btf,
 						    u32 type_id)
@@ -318,50 +345,18 @@ static inline const char *btf_name_by_offset(const struct btf *btf,
 {
 	return NULL;
 }
-#endif
-
-struct kfunc_btf_id_set {
-	struct list_head list;
-	struct btf_id_set *set;
-	struct module *owner;
-};
-
-struct kfunc_btf_id_list {
-	struct list_head list;
-	struct mutex mutex;
-};
-
-#ifdef CONFIG_DEBUG_INFO_BTF_MODULES
-void register_kfunc_btf_id_set(struct kfunc_btf_id_list *l,
-			       struct kfunc_btf_id_set *s);
-void unregister_kfunc_btf_id_set(struct kfunc_btf_id_list *l,
-				 struct kfunc_btf_id_set *s);
-bool bpf_check_mod_kfunc_call(struct kfunc_btf_id_list *klist, u32 kfunc_id,
-			      struct module *owner);
-
-extern struct kfunc_btf_id_list bpf_tcp_ca_kfunc_list;
-extern struct kfunc_btf_id_list prog_test_kfunc_list;
-#else
-static inline void register_kfunc_btf_id_set(struct kfunc_btf_id_list *l,
-					     struct kfunc_btf_id_set *s)
-{
-}
-static inline void unregister_kfunc_btf_id_set(struct kfunc_btf_id_list *l,
-					       struct kfunc_btf_id_set *s)
-{
-}
-static inline bool bpf_check_mod_kfunc_call(struct kfunc_btf_id_list *klist,
-					    u32 kfunc_id, struct module *owner)
+static inline bool btf_kfunc_id_set_contains(const struct btf *btf,
+					     enum bpf_prog_type prog_type,
+					     enum btf_kfunc_type type,
+					     u32 kfunc_btf_id)
 {
 	return false;
 }
-
-static struct kfunc_btf_id_list bpf_tcp_ca_kfunc_list __maybe_unused;
-static struct kfunc_btf_id_list prog_test_kfunc_list __maybe_unused;
+static inline int register_btf_kfunc_id_set(enum bpf_prog_type prog_type,
+					    const struct btf_kfunc_id_set *s)
+{
+	return 0;
+}
 #endif
 
-#define DEFINE_KFUNC_BTF_ID_SET(set, name)                                     \
-	struct kfunc_btf_id_set name = { LIST_HEAD_INIT(name.list), (set),     \
-					 THIS_MODULE }
-
 #endif
diff --git a/include/linux/btf_ids.h b/include/linux/btf_ids.h
index 919c0fde1c51..bc5d9cc34e4c 100644
--- a/include/linux/btf_ids.h
+++ b/include/linux/btf_ids.h
@@ -11,6 +11,7 @@ struct btf_id_set {
 #ifdef CONFIG_DEBUG_INFO_BTF
 
 #include <linux/compiler.h> /* for __PASTE */
+#include <linux/compiler_attributes.h> /* for __maybe_unused */
 
 /*
  * Following macros help to define lists of BTF IDs placed
@@ -146,14 +147,14 @@ extern struct btf_id_set name;
 
 #else
 
-#define BTF_ID_LIST(name) static u32 name[5];
+#define BTF_ID_LIST(name) static u32 __maybe_unused name[5];
 #define BTF_ID(prefix, name)
 #define BTF_ID_UNUSED
-#define BTF_ID_LIST_GLOBAL(name, n) u32 name[n];
-#define BTF_ID_LIST_SINGLE(name, prefix, typename) static u32 name[1];
-#define BTF_ID_LIST_GLOBAL_SINGLE(name, prefix, typename) u32 name[1];
-#define BTF_SET_START(name) static struct btf_id_set name = { 0 };
-#define BTF_SET_START_GLOBAL(name) static struct btf_id_set name = { 0 };
+#define BTF_ID_LIST_GLOBAL(name, n) u32 __maybe_unused name[n];
+#define BTF_ID_LIST_SINGLE(name, prefix, typename) static u32 __maybe_unused name[1];
+#define BTF_ID_LIST_GLOBAL_SINGLE(name, prefix, typename) u32 __maybe_unused name[1];
+#define BTF_SET_START(name) static struct btf_id_set __maybe_unused name = { 0 };
+#define BTF_SET_START_GLOBAL(name) static struct btf_id_set __maybe_unused name = { 0 };
 #define BTF_SET_END(name)
 
 #endif /* CONFIG_DEBUG_INFO_BTF */
diff --git a/include/net/netfilter/nf_conntrack_bpf.h b/include/net/netfilter/nf_conntrack_bpf.h
new file mode 100644
index 000000000000..a473b56842c5
--- /dev/null
+++ b/include/net/netfilter/nf_conntrack_bpf.h
@@ -0,0 +1,23 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef _NF_CONNTRACK_BPF_H
+#define _NF_CONNTRACK_BPF_H
+
+#include <linux/btf.h>
+#include <linux/kconfig.h>
+
+#if (IS_BUILTIN(CONFIG_NF_CONNTRACK) && IS_ENABLED(CONFIG_DEBUG_INFO_BTF)) || \
+    (IS_MODULE(CONFIG_NF_CONNTRACK) && IS_ENABLED(CONFIG_DEBUG_INFO_BTF_MODULES))
+
+extern int register_nf_conntrack_bpf(void);
+
+#else
+
+static inline int register_nf_conntrack_bpf(void)
+{
+	return 0;
+}
+
+#endif
+
+#endif /* _NF_CONNTRACK_BPF_H */
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 33bb8ae4a804..57f5fd5af2f9 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -198,6 +198,21 @@
 DEFINE_IDR(btf_idr);
 DEFINE_SPINLOCK(btf_idr_lock);
 
+enum btf_kfunc_hook {
+	BTF_KFUNC_HOOK_XDP,
+	BTF_KFUNC_HOOK_TC,
+	BTF_KFUNC_HOOK_STRUCT_OPS,
+	BTF_KFUNC_HOOK_MAX,
+};
+
+enum {
+	BTF_KFUNC_SET_MAX_CNT = 32,
+};
+
+struct btf_kfunc_set_tab {
+	struct btf_id_set *sets[BTF_KFUNC_HOOK_MAX][BTF_KFUNC_TYPE_MAX];
+};
+
 struct btf {
 	void *data;
 	struct btf_type **types;
@@ -212,6 +227,7 @@ struct btf {
 	refcount_t refcnt;
 	u32 id;
 	struct rcu_head rcu;
+	struct btf_kfunc_set_tab *kfunc_set_tab;
 
 	/* split BTF support */
 	struct btf *base_btf;
@@ -1531,8 +1547,30 @@ static void btf_free_id(struct btf *btf)
 	spin_unlock_irqrestore(&btf_idr_lock, flags);
 }
 
+static void btf_free_kfunc_set_tab(struct btf *btf)
+{
+	struct btf_kfunc_set_tab *tab = btf->kfunc_set_tab;
+	int hook, type;
+
+	if (!tab)
+		return;
+	/* For module BTF, we directly assign the sets being registered, so
+	 * there is nothing to free except kfunc_set_tab.
+	 */
+	if (btf_is_module(btf))
+		goto free_tab;
+	for (hook = 0; hook < ARRAY_SIZE(tab->sets); hook++) {
+		for (type = 0; type < ARRAY_SIZE(tab->sets[0]); type++)
+			kfree(tab->sets[hook][type]);
+	}
+free_tab:
+	kfree(tab);
+	btf->kfunc_set_tab = NULL;
+}
+
 static void btf_free(struct btf *btf)
 {
+	btf_free_kfunc_set_tab(btf);
 	kvfree(btf->types);
 	kvfree(btf->resolved_sizes);
 	kvfree(btf->resolved_ids);
@@ -5616,17 +5654,45 @@ static bool __btf_type_is_scalar_struct(struct bpf_verifier_log *log,
 	return true;
 }
 
+static bool is_kfunc_arg_mem_size(const struct btf *btf,
+				  const struct btf_param *arg,
+				  const struct bpf_reg_state *reg)
+{
+	int len, sfx_len = sizeof("__sz") - 1;
+	const struct btf_type *t;
+	const char *param_name;
+
+	t = btf_type_skip_modifiers(btf, arg->type, NULL);
+	if (!btf_type_is_scalar(t) || reg->type != SCALAR_VALUE)
+		return false;
+
+	/* In the future, this can be ported to use BTF tagging */
+	param_name = btf_name_by_offset(btf, arg->name_off);
+	if (str_is_empty(param_name))
+		return false;
+	len = strlen(param_name);
+	if (len < sfx_len)
+		return false;
+	param_name += len - sfx_len;
+	if (strncmp(param_name, "__sz", sfx_len))
+		return false;
+
+	return true;
+}
+
 static int btf_check_func_arg_match(struct bpf_verifier_env *env,
 				    const struct btf *btf, u32 func_id,
 				    struct bpf_reg_state *regs,
 				    bool ptr_to_mem_ok)
 {
 	struct bpf_verifier_log *log = &env->log;
+	u32 i, nargs, ref_id, ref_obj_id = 0;
 	bool is_kfunc = btf_is_kernel(btf);
 	const char *func_name, *ref_tname;
 	const struct btf_type *t, *ref_t;
 	const struct btf_param *args;
-	u32 i, nargs, ref_id;
+	int ref_regno = 0;
+	bool rel = false;
 
 	t = btf_type_by_id(btf, func_id);
 	if (!t || !btf_type_is_func(t)) {
@@ -5704,6 +5770,16 @@ static int btf_check_func_arg_match(struct bpf_verifier_env *env,
 			if (reg->type == PTR_TO_BTF_ID) {
 				reg_btf = reg->btf;
 				reg_ref_id = reg->btf_id;
+				/* Ensure only one argument is referenced PTR_TO_BTF_ID */
+				if (reg->ref_obj_id) {
+					if (ref_obj_id) {
+						bpf_log(log, "verifier internal error: more than one arg with ref_obj_id R%d %u %u\n",
+							regno, reg->ref_obj_id, ref_obj_id);
+						return -EFAULT;
+					}
+					ref_regno = regno;
+					ref_obj_id = reg->ref_obj_id;
+				}
 			} else {
 				reg_btf = btf_vmlinux;
 				reg_ref_id = *reg2btf_ids[reg->type];
@@ -5727,17 +5803,33 @@ static int btf_check_func_arg_match(struct bpf_verifier_env *env,
 			u32 type_size;
 
 			if (is_kfunc) {
+				bool arg_mem_size = i + 1 < nargs && is_kfunc_arg_mem_size(btf, &args[i + 1], &regs[regno + 1]);
+
 				/* Permit pointer to mem, but only when argument
 				 * type is pointer to scalar, or struct composed
 				 * (recursively) of scalars.
+				 * When arg_mem_size is true, the pointer can be
+				 * void *.
 				 */
 				if (!btf_type_is_scalar(ref_t) &&
-				    !__btf_type_is_scalar_struct(log, btf, ref_t, 0)) {
+				    !__btf_type_is_scalar_struct(log, btf, ref_t, 0) &&
+				    (arg_mem_size ? !btf_type_is_void(ref_t) : 1)) {
 					bpf_log(log,
-						"arg#%d pointer type %s %s must point to scalar or struct with scalar\n",
-						i, btf_type_str(ref_t), ref_tname);
+						"arg#%d pointer type %s %s must point to %sscalar, or struct with scalar\n",
+						i, btf_type_str(ref_t), ref_tname, arg_mem_size ? "void, " : "");
 					return -EINVAL;
 				}
+
+				/* Check for mem, len pair */
+				if (arg_mem_size) {
+					if (check_kfunc_mem_size_reg(env, &regs[regno + 1], regno + 1)) {
+						bpf_log(log, "arg#%d arg#%d memory, len pair leads to invalid memory access\n",
+							i, i + 1);
+						return -EINVAL;
+					}
+					i++;
+					continue;
+				}
 			}
 
 			resolve_ret = btf_resolve_size(btf, ref_t, &type_size);
@@ -5758,7 +5850,23 @@ static int btf_check_func_arg_match(struct bpf_verifier_env *env,
 		}
 	}
 
-	return 0;
+	/* Either both are set, or neither */
+	WARN_ON_ONCE((ref_obj_id && !ref_regno) || (!ref_obj_id && ref_regno));
+	if (is_kfunc) {
+		rel = btf_kfunc_id_set_contains(btf, resolve_prog_type(env->prog),
+						BTF_KFUNC_TYPE_RELEASE, func_id);
+		/* We already made sure ref_obj_id is set only for one argument */
+		if (rel && !ref_obj_id) {
+			bpf_log(log, "release kernel function %s expects refcounted PTR_TO_BTF_ID\n",
+				func_name);
+			return -EINVAL;
+		}
+		/* Allow (!rel && ref_obj_id), so that passing such referenced PTR_TO_BTF_ID to
+		 * other kfuncs works
+		 */
+	}
+	/* returns argument register number > 0 in case of reference release kfunc */
+	return rel ? ref_regno : 0;
 }
 
 /* Compare BTF of a function with given bpf_reg_state.
@@ -6200,12 +6308,17 @@ bool btf_id_set_contains(const struct btf_id_set *set, u32 id)
 	return bsearch(&id, set->ids, set->cnt, sizeof(u32), btf_id_cmp_func) != NULL;
 }
 
+enum {
+	BTF_MODULE_F_LIVE = (1 << 0),
+};
+
 #ifdef CONFIG_DEBUG_INFO_BTF_MODULES
 struct btf_module {
 	struct list_head list;
 	struct module *module;
 	struct btf *btf;
 	struct bin_attribute *sysfs_attr;
+	int flags;
 };
 
 static LIST_HEAD(btf_modules);
@@ -6233,7 +6346,8 @@ static int btf_module_notify(struct notifier_block *nb, unsigned long op,
 	int err = 0;
 
 	if (mod->btf_data_size == 0 ||
-	    (op != MODULE_STATE_COMING && op != MODULE_STATE_GOING))
+	    (op != MODULE_STATE_COMING && op != MODULE_STATE_LIVE &&
+	     op != MODULE_STATE_GOING))
 		goto out;
 
 	switch (op) {
@@ -6291,6 +6405,17 @@ static int btf_module_notify(struct notifier_block *nb, unsigned long op,
 			btf_mod->sysfs_attr = attr;
 		}
 
+		break;
+	case MODULE_STATE_LIVE:
+		mutex_lock(&btf_module_mutex);
+		list_for_each_entry_safe(btf_mod, tmp, &btf_modules, list) {
+			if (btf_mod->module != module)
+				continue;
+
+			btf_mod->flags |= BTF_MODULE_F_LIVE;
+			break;
+		}
+		mutex_unlock(&btf_module_mutex);
 		break;
 	case MODULE_STATE_GOING:
 		mutex_lock(&btf_module_mutex);
@@ -6338,7 +6463,12 @@ struct module *btf_try_get_module(const struct btf *btf)
 		if (btf_mod->btf != btf)
 			continue;
 
-		if (try_module_get(btf_mod->module))
+		/* We must only consider module whose __init routine has
+		 * finished, hence we must check for BTF_MODULE_F_LIVE flag,
+		 * which is set from the notifier callback for
+		 * MODULE_STATE_LIVE.
+		 */
+		if ((btf_mod->flags & BTF_MODULE_F_LIVE) && try_module_get(btf_mod->module))
 			res = btf_mod->module;
 
 		break;
@@ -6349,6 +6479,36 @@ struct module *btf_try_get_module(const struct btf *btf)
 	return res;
 }
 
+/* Returns struct btf corresponding to the struct module
+ *
+ * This function can return NULL or ERR_PTR. Note that caller must
+ * release reference for struct btf iff btf_is_module is true.
+ */
+static struct btf *btf_get_module_btf(const struct module *module)
+{
+	struct btf *btf = NULL;
+#ifdef CONFIG_DEBUG_INFO_BTF_MODULES
+	struct btf_module *btf_mod, *tmp;
+#endif
+
+	if (!module)
+		return bpf_get_btf_vmlinux();
+#ifdef CONFIG_DEBUG_INFO_BTF_MODULES
+	mutex_lock(&btf_module_mutex);
+	list_for_each_entry_safe(btf_mod, tmp, &btf_modules, list) {
+		if (btf_mod->module != module)
+			continue;
+
+		btf_get(btf_mod->btf);
+		btf = btf_mod->btf;
+		break;
+	}
+	mutex_unlock(&btf_module_mutex);
+#endif
+
+	return btf;
+}
+
 BPF_CALL_4(bpf_btf_find_by_name_kind, char *, name, int, name_sz, u32, kind, int, flags)
 {
 	struct btf *btf;
@@ -6416,53 +6576,181 @@ BTF_ID_LIST_GLOBAL(btf_tracing_ids, MAX_BTF_TRACING_TYPE)
 BTF_TRACING_TYPE_xxx
 #undef BTF_TRACING_TYPE
 
-/* BTF ID set registration API for modules */
+/* Kernel Function (kfunc) BTF ID set registration API */
 
-#ifdef CONFIG_DEBUG_INFO_BTF_MODULES
-
-void register_kfunc_btf_id_set(struct kfunc_btf_id_list *l,
-			       struct kfunc_btf_id_set *s)
+static int __btf_populate_kfunc_set(struct btf *btf, enum btf_kfunc_hook hook,
+				    enum btf_kfunc_type type,
+				    struct btf_id_set *add_set, bool vmlinux_set)
 {
-	mutex_lock(&l->mutex);
-	list_add(&s->list, &l->list);
-	mutex_unlock(&l->mutex);
-}
-EXPORT_SYMBOL_GPL(register_kfunc_btf_id_set);
+	struct btf_kfunc_set_tab *tab;
+	struct btf_id_set *set;
+	u32 set_cnt;
+	int ret;
 
-void unregister_kfunc_btf_id_set(struct kfunc_btf_id_list *l,
-				 struct kfunc_btf_id_set *s)
-{
-	mutex_lock(&l->mutex);
-	list_del_init(&s->list);
-	mutex_unlock(&l->mutex);
-}
-EXPORT_SYMBOL_GPL(unregister_kfunc_btf_id_set);
-
-bool bpf_check_mod_kfunc_call(struct kfunc_btf_id_list *klist, u32 kfunc_id,
-			      struct module *owner)
-{
-	struct kfunc_btf_id_set *s;
-
-	mutex_lock(&klist->mutex);
-	list_for_each_entry(s, &klist->list, list) {
-		if (s->owner == owner && btf_id_set_contains(s->set, kfunc_id)) {
-			mutex_unlock(&klist->mutex);
-			return true;
-		}
+	if (hook >= BTF_KFUNC_HOOK_MAX || type >= BTF_KFUNC_TYPE_MAX) {
+		ret = -EINVAL;
+		goto end;
 	}
-	mutex_unlock(&klist->mutex);
-	return false;
+
+	if (!add_set->cnt)
+		return 0;
+
+	tab = btf->kfunc_set_tab;
+	if (!tab) {
+		tab = kzalloc(sizeof(*tab), GFP_KERNEL | __GFP_NOWARN);
+		if (!tab)
+			return -ENOMEM;
+		btf->kfunc_set_tab = tab;
+	}
+
+	set = tab->sets[hook][type];
+	/* Warn when register_btf_kfunc_id_set is called twice for the same hook
+	 * for module sets.
+	 */
+	if (WARN_ON_ONCE(set && !vmlinux_set)) {
+		ret = -EINVAL;
+		goto end;
+	}
+
+	/* We don't need to allocate, concatenate, and sort module sets, because
+	 * only one is allowed per hook. Hence, we can directly assign the
+	 * pointer and return.
+	 */
+	if (!vmlinux_set) {
+		tab->sets[hook][type] = add_set;
+		return 0;
+	}
+
+	/* In case of vmlinux sets, there may be more than one set being
+	 * registered per hook. To create a unified set, we allocate a new set
+	 * and concatenate all individual sets being registered. While each set
+	 * is individually sorted, they may become unsorted when concatenated,
+	 * hence re-sorting the final set again is required to make binary
+	 * searching the set using btf_id_set_contains function work.
+	 */
+	set_cnt = set ? set->cnt : 0;
+
+	if (set_cnt > U32_MAX - add_set->cnt) {
+		ret = -EOVERFLOW;
+		goto end;
+	}
+
+	if (set_cnt + add_set->cnt > BTF_KFUNC_SET_MAX_CNT) {
+		ret = -E2BIG;
+		goto end;
+	}
+
+	/* Grow set */
+	set = krealloc(tab->sets[hook][type],
+		       offsetof(struct btf_id_set, ids[set_cnt + add_set->cnt]),
+		       GFP_KERNEL | __GFP_NOWARN);
+	if (!set) {
+		ret = -ENOMEM;
+		goto end;
+	}
+
+	/* For newly allocated set, initialize set->cnt to 0 */
+	if (!tab->sets[hook][type])
+		set->cnt = 0;
+	tab->sets[hook][type] = set;
+
+	/* Concatenate the two sets */
+	memcpy(set->ids + set->cnt, add_set->ids, add_set->cnt * sizeof(set->ids[0]));
+	set->cnt += add_set->cnt;
+
+	sort(set->ids, set->cnt, sizeof(set->ids[0]), btf_id_cmp_func, NULL);
+
+	return 0;
+end:
+	btf_free_kfunc_set_tab(btf);
+	return ret;
 }
 
-#define DEFINE_KFUNC_BTF_ID_LIST(name)                                         \
-	struct kfunc_btf_id_list name = { LIST_HEAD_INIT(name.list),           \
-					  __MUTEX_INITIALIZER(name.mutex) };   \
-	EXPORT_SYMBOL_GPL(name)
+static int btf_populate_kfunc_set(struct btf *btf, enum btf_kfunc_hook hook,
+				  const struct btf_kfunc_id_set *kset)
+{
+	bool vmlinux_set = !btf_is_module(btf);
+	int type, ret;
 
-DEFINE_KFUNC_BTF_ID_LIST(bpf_tcp_ca_kfunc_list);
-DEFINE_KFUNC_BTF_ID_LIST(prog_test_kfunc_list);
+	for (type = 0; type < ARRAY_SIZE(kset->sets); type++) {
+		if (!kset->sets[type])
+			continue;
 
-#endif
+		ret = __btf_populate_kfunc_set(btf, hook, type, kset->sets[type], vmlinux_set);
+		if (ret)
+			break;
+	}
+	return ret;
+}
+
+static bool __btf_kfunc_id_set_contains(const struct btf *btf,
+					enum btf_kfunc_hook hook,
+					enum btf_kfunc_type type,
+					u32 kfunc_btf_id)
+{
+	struct btf_id_set *set;
+
+	if (hook >= BTF_KFUNC_HOOK_MAX || type >= BTF_KFUNC_TYPE_MAX)
+		return false;
+	if (!btf->kfunc_set_tab)
+		return false;
+	set = btf->kfunc_set_tab->sets[hook][type];
+	if (!set)
+		return false;
+	return btf_id_set_contains(set, kfunc_btf_id);
+}
+
+static int bpf_prog_type_to_kfunc_hook(enum bpf_prog_type prog_type)
+{
+	switch (prog_type) {
+	case BPF_PROG_TYPE_XDP:
+		return BTF_KFUNC_HOOK_XDP;
+	case BPF_PROG_TYPE_SCHED_CLS:
+		return BTF_KFUNC_HOOK_TC;
+	case BPF_PROG_TYPE_STRUCT_OPS:
+		return BTF_KFUNC_HOOK_STRUCT_OPS;
+	default:
+		return BTF_KFUNC_HOOK_MAX;
+	}
+}
+
+/* Caution:
+ * Reference to the module (obtained using btf_try_get_module) corresponding to
+ * the struct btf *MUST* be held when calling this function from verifier
+ * context. This is usually true as we stash references in prog's kfunc_btf_tab;
+ * keeping the reference for the duration of the call provides the necessary
+ * protection for looking up a well-formed btf->kfunc_set_tab.
+ */
+bool btf_kfunc_id_set_contains(const struct btf *btf,
+			       enum bpf_prog_type prog_type,
+			       enum btf_kfunc_type type, u32 kfunc_btf_id)
+{
+	enum btf_kfunc_hook hook;
+
+	hook = bpf_prog_type_to_kfunc_hook(prog_type);
+	return __btf_kfunc_id_set_contains(btf, hook, type, kfunc_btf_id);
+}
+
+/* This function must be invoked only from initcalls/module init functions */
+int register_btf_kfunc_id_set(enum bpf_prog_type prog_type,
+			      const struct btf_kfunc_id_set *kset)
+{
+	enum btf_kfunc_hook hook;
+	struct btf *btf;
+	int ret;
+
+	btf = btf_get_module_btf(kset->owner);
+	if (IS_ERR_OR_NULL(btf))
+		return btf ? PTR_ERR(btf) : -ENOENT;
+
+	hook = bpf_prog_type_to_kfunc_hook(prog_type);
+	ret = btf_populate_kfunc_set(btf, hook, kset);
+	/* reference is only taken for module BTF */
+	if (btf_is_module(btf))
+		btf_put(btf);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(register_btf_kfunc_id_set);
 
 int bpf_core_types_are_compat(const struct btf *local_btf, __u32 local_id,
 			      const struct btf *targ_btf, __u32 targ_id)
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index bfb45381fb3f..8c5a46d41f28 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -452,7 +452,8 @@ static bool reg_type_may_be_refcounted_or_null(enum bpf_reg_type type)
 {
 	return base_type(type) == PTR_TO_SOCKET ||
 		base_type(type) == PTR_TO_TCP_SOCK ||
-		base_type(type) == PTR_TO_MEM;
+		base_type(type) == PTR_TO_MEM ||
+		base_type(type) == PTR_TO_BTF_ID;
 }
 
 static bool type_is_rdonly_mem(u32 type)
@@ -1741,7 +1742,7 @@ find_kfunc_desc(const struct bpf_prog *prog, u32 func_id, u16 offset)
 }
 
 static struct btf *__find_kfunc_desc_btf(struct bpf_verifier_env *env,
-					 s16 offset, struct module **btf_modp)
+					 s16 offset)
 {
 	struct bpf_kfunc_btf kf_btf = { .offset = offset };
 	struct bpf_kfunc_btf_tab *tab;
@@ -1795,8 +1796,6 @@ static struct btf *__find_kfunc_desc_btf(struct bpf_verifier_env *env,
 		sort(tab->descs, tab->nr_descs, sizeof(tab->descs[0]),
 		     kfunc_btf_cmp_by_off, NULL);
 	}
-	if (btf_modp)
-		*btf_modp = b->module;
 	return b->btf;
 }
 
@@ -1813,8 +1812,7 @@ void bpf_free_kfunc_btf_tab(struct bpf_kfunc_btf_tab *tab)
 }
 
 static struct btf *find_kfunc_desc_btf(struct bpf_verifier_env *env,
-				       u32 func_id, s16 offset,
-				       struct module **btf_modp)
+				       u32 func_id, s16 offset)
 {
 	if (offset) {
 		if (offset < 0) {
@@ -1825,7 +1823,7 @@ static struct btf *find_kfunc_desc_btf(struct bpf_verifier_env *env,
 			return ERR_PTR(-EINVAL);
 		}
 
-		return __find_kfunc_desc_btf(env, offset, btf_modp);
+		return __find_kfunc_desc_btf(env, offset);
 	}
 	return btf_vmlinux ?: ERR_PTR(-ENOENT);
 }
@@ -1888,7 +1886,7 @@ static int add_kfunc_call(struct bpf_verifier_env *env, u32 func_id, s16 offset)
 		prog_aux->kfunc_btf_tab = btf_tab;
 	}
 
-	desc_btf = find_kfunc_desc_btf(env, func_id, offset, NULL);
+	desc_btf = find_kfunc_desc_btf(env, func_id, offset);
 	if (IS_ERR(desc_btf)) {
 		verbose(env, "failed to find BTF for kernel function\n");
 		return PTR_ERR(desc_btf);
@@ -2349,7 +2347,7 @@ static const char *disasm_kfunc_name(void *data, const struct bpf_insn *insn)
 	if (insn->src_reg != BPF_PSEUDO_KFUNC_CALL)
 		return NULL;
 
-	desc_btf = find_kfunc_desc_btf(data, insn->imm, insn->off, NULL);
+	desc_btf = find_kfunc_desc_btf(data, insn->imm, insn->off);
 	if (IS_ERR(desc_btf))
 		return "<error>";
 
@@ -3496,11 +3494,6 @@ static int check_map_access(struct bpf_verifier_env *env, u32 regno,
 
 #define MAX_PACKET_OFF 0xffff
 
-static enum bpf_prog_type resolve_prog_type(struct bpf_prog *prog)
-{
-	return prog->aux->dst_prog ? prog->aux->dst_prog->type : prog->type;
-}
-
 static bool may_access_direct_pkt_data(struct bpf_verifier_env *env,
 				       const struct bpf_call_arg_meta *meta,
 				       enum bpf_access_type t)
@@ -4867,6 +4860,62 @@ static int check_helper_mem_access(struct bpf_verifier_env *env, int regno,
 	}
 }
 
+static int check_mem_size_reg(struct bpf_verifier_env *env,
+			      struct bpf_reg_state *reg, u32 regno,
+			      bool zero_size_allowed,
+			      struct bpf_call_arg_meta *meta)
+{
+	int err;
+
+	/* This is used to refine r0 return value bounds for helpers
+	 * that enforce this value as an upper bound on return values.
+	 * See do_refine_retval_range() for helpers that can refine
+	 * the return value. C type of helper is u32 so we pull register
+	 * bound from umax_value however, if negative verifier errors
+	 * out. Only upper bounds can be learned because retval is an
+	 * int type and negative retvals are allowed.
+	 */
+	if (meta)
+		meta->msize_max_value = reg->umax_value;
+
+	/* The register is SCALAR_VALUE; the access check
+	 * happens using its boundaries.
+	 */
+	if (!tnum_is_const(reg->var_off))
+		/* For unprivileged variable accesses, disable raw
+		 * mode so that the program is required to
+		 * initialize all the memory that the helper could
+		 * just partially fill up.
+		 */
+		meta = NULL;
+
+	if (reg->smin_value < 0) {
+		verbose(env, "R%d min value is negative, either use unsigned or 'var &= const'\n",
+			regno);
+		return -EACCES;
+	}
+
+	if (reg->umin_value == 0) {
+		err = check_helper_mem_access(env, regno - 1, 0,
+					      zero_size_allowed,
+					      meta);
+		if (err)
+			return err;
+	}
+
+	if (reg->umax_value >= BPF_MAX_VAR_SIZ) {
+		verbose(env, "R%d unbounded memory access, use 'var &= const' or 'if (var < const)'\n",
+			regno);
+		return -EACCES;
+	}
+	err = check_helper_mem_access(env, regno - 1,
+				      reg->umax_value,
+				      zero_size_allowed, meta);
+	if (!err)
+		err = mark_chain_precision(env, regno);
+	return err;
+}
+
 int check_mem_reg(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
 		   u32 regno, u32 mem_size)
 {
@@ -4890,6 +4939,28 @@ int check_mem_reg(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
 	return check_helper_mem_access(env, regno, mem_size, true, NULL);
 }
 
+int check_kfunc_mem_size_reg(struct bpf_verifier_env *env, struct bpf_reg_state *reg,
+			     u32 regno)
+{
+	struct bpf_reg_state *mem_reg = &cur_regs(env)[regno - 1];
+	bool may_be_null = type_may_be_null(mem_reg->type);
+	struct bpf_reg_state saved_reg;
+	int err;
+
+	WARN_ON_ONCE(regno < BPF_REG_2 || regno > BPF_REG_5);
+
+	if (may_be_null) {
+		saved_reg = *mem_reg;
+		mark_ptr_not_null_reg(mem_reg);
+	}
+
+	err = check_mem_size_reg(env, reg, regno, true, NULL);
+
+	if (may_be_null)
+		*mem_reg = saved_reg;
+	return err;
+}
+
 /* Implementation details:
  * bpf_map_lookup returns PTR_TO_MAP_VALUE_OR_NULL
  * Two bpf_map_lookups (even with the same key) will have different reg->id.
@@ -5411,51 +5482,7 @@ skip_type_check:
 	} else if (arg_type_is_mem_size(arg_type)) {
 		bool zero_size_allowed = (arg_type == ARG_CONST_SIZE_OR_ZERO);
 
-		/* This is used to refine r0 return value bounds for helpers
-		 * that enforce this value as an upper bound on return values.
-		 * See do_refine_retval_range() for helpers that can refine
-		 * the return value. C type of helper is u32 so we pull register
-		 * bound from umax_value however, if negative verifier errors
-		 * out. Only upper bounds can be learned because retval is an
-		 * int type and negative retvals are allowed.
-		 */
-		meta->msize_max_value = reg->umax_value;
-
-		/* The register is SCALAR_VALUE; the access check
-		 * happens using its boundaries.
-		 */
-		if (!tnum_is_const(reg->var_off))
-			/* For unprivileged variable accesses, disable raw
-			 * mode so that the program is required to
-			 * initialize all the memory that the helper could
-			 * just partially fill up.
-			 */
-			meta = NULL;
-
-		if (reg->smin_value < 0) {
-			verbose(env, "R%d min value is negative, either use unsigned or 'var &= const'\n",
-				regno);
-			return -EACCES;
-		}
-
-		if (reg->umin_value == 0) {
-			err = check_helper_mem_access(env, regno - 1, 0,
-						      zero_size_allowed,
-						      meta);
-			if (err)
-				return err;
-		}
-
-		if (reg->umax_value >= BPF_MAX_VAR_SIZ) {
-			verbose(env, "R%d unbounded memory access, use 'var &= const' or 'if (var < const)'\n",
-				regno);
-			return -EACCES;
-		}
-		err = check_helper_mem_access(env, regno - 1,
-					      reg->umax_value,
-					      zero_size_allowed, meta);
-		if (!err)
-			err = mark_chain_precision(env, regno);
+		err = check_mem_size_reg(env, reg, regno, zero_size_allowed, meta);
 	} else if (arg_type_is_alloc_size(arg_type)) {
 		if (!tnum_is_const(reg->var_off)) {
 			verbose(env, "R%d is not a known constant'\n",
@@ -6814,22 +6841,23 @@ static void mark_btf_func_reg_size(struct bpf_verifier_env *env, u32 regno,
 	}
 }
 
-static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn)
+static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
+			    int *insn_idx_p)
 {
 	const struct btf_type *t, *func, *func_proto, *ptr_type;
 	struct bpf_reg_state *regs = cur_regs(env);
 	const char *func_name, *ptr_type_name;
 	u32 i, nargs, func_id, ptr_type_id;
-	struct module *btf_mod = NULL;
+	int err, insn_idx = *insn_idx_p;
 	const struct btf_param *args;
 	struct btf *desc_btf;
-	int err;
+	bool acq;
 
 	/* skip for now, but return error when we find this in fixup_kfunc_call */
 	if (!insn->imm)
 		return 0;
 
-	desc_btf = find_kfunc_desc_btf(env, insn->imm, insn->off, &btf_mod);
+	desc_btf = find_kfunc_desc_btf(env, insn->imm, insn->off);
 	if (IS_ERR(desc_btf))
 		return PTR_ERR(desc_btf);
 
@@ -6838,23 +6866,43 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn)
 	func_name = btf_name_by_offset(desc_btf, func->name_off);
 	func_proto = btf_type_by_id(desc_btf, func->type);
 
-	if (!env->ops->check_kfunc_call ||
-	    !env->ops->check_kfunc_call(func_id, btf_mod)) {
+	if (!btf_kfunc_id_set_contains(desc_btf, resolve_prog_type(env->prog),
+				      BTF_KFUNC_TYPE_CHECK, func_id)) {
 		verbose(env, "calling kernel function %s is not allowed\n",
 			func_name);
 		return -EACCES;
 	}
 
+	acq = btf_kfunc_id_set_contains(desc_btf, resolve_prog_type(env->prog),
+					BTF_KFUNC_TYPE_ACQUIRE, func_id);
+
 	/* Check the arguments */
 	err = btf_check_kfunc_arg_match(env, desc_btf, func_id, regs);
-	if (err)
+	if (err < 0)
 		return err;
+	/* In case of release function, we get register number of refcounted
+	 * PTR_TO_BTF_ID back from btf_check_kfunc_arg_match, do the release now
+	 */
+	if (err) {
+		err = release_reference(env, regs[err].ref_obj_id);
+		if (err) {
+			verbose(env, "kfunc %s#%d reference has not been acquired before\n",
+				func_name, func_id);
+			return err;
+		}
+	}
 
 	for (i = 0; i < CALLER_SAVED_REGS; i++)
 		mark_reg_not_init(env, regs, caller_saved[i]);
 
 	/* Check return type */
 	t = btf_type_skip_modifiers(desc_btf, func_proto->type, NULL);
+
+	if (acq && !btf_type_is_ptr(t)) {
+		verbose(env, "acquire kernel function does not return PTR_TO_BTF_ID\n");
+		return -EINVAL;
+	}
+
 	if (btf_type_is_scalar(t)) {
 		mark_reg_unknown(env, regs, BPF_REG_0);
 		mark_btf_func_reg_size(env, BPF_REG_0, t->size);
@@ -6873,7 +6921,21 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn)
 		regs[BPF_REG_0].btf = desc_btf;
 		regs[BPF_REG_0].type = PTR_TO_BTF_ID;
 		regs[BPF_REG_0].btf_id = ptr_type_id;
+		if (btf_kfunc_id_set_contains(desc_btf, resolve_prog_type(env->prog),
+					      BTF_KFUNC_TYPE_RET_NULL, func_id)) {
+			regs[BPF_REG_0].type |= PTR_MAYBE_NULL;
+			/* For mark_ptr_or_null_reg, see 93c230e3f5bd6 */
+			regs[BPF_REG_0].id = ++env->id_gen;
+		}
 		mark_btf_func_reg_size(env, BPF_REG_0, sizeof(void *));
+		if (acq) {
+			int id = acquire_reference_state(env, insn_idx);
+
+			if (id < 0)
+				return id;
+			regs[BPF_REG_0].id = id;
+			regs[BPF_REG_0].ref_obj_id = id;
+		}
 	} /* else { add_kfunc_call() ensures it is btf_type_is_void(t) } */
 
 	nargs = btf_type_vlen(func_proto);
@@ -11518,7 +11580,7 @@ static int do_check(struct bpf_verifier_env *env)
 				if (insn->src_reg == BPF_PSEUDO_CALL)
 					err = check_func_call(env, insn, &env->insn_idx);
 				else if (insn->src_reg == BPF_PSEUDO_KFUNC_CALL)
-					err = check_kfunc_call(env, insn);
+					err = check_kfunc_call(env, insn, &env->insn_idx);
 				else
 					err = check_helper_call(env, insn, &env->insn_idx);
 				if (err)
diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index 46dd95755967..3a5bf8ad834d 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -5,6 +5,7 @@
 #include <linux/btf.h>
 #include <linux/btf_ids.h>
 #include <linux/slab.h>
+#include <linux/init.h>
 #include <linux/vmalloc.h>
 #include <linux/etherdevice.h>
 #include <linux/filter.h>
@@ -171,6 +172,8 @@ int noinline bpf_fentry_test1(int a)
 {
 	return a + 1;
 }
+EXPORT_SYMBOL_GPL(bpf_fentry_test1);
+ALLOW_ERROR_INJECTION(bpf_fentry_test1, ERRNO);
 
 int noinline bpf_fentry_test2(int a, u64 b)
 {
@@ -232,22 +235,137 @@ struct sock * noinline bpf_kfunc_call_test3(struct sock *sk)
 	return sk;
 }
 
+struct prog_test_ref_kfunc {
+	int a;
+	int b;
+	struct prog_test_ref_kfunc *next;
+};
+
+static struct prog_test_ref_kfunc prog_test_struct = {
+	.a = 42,
+	.b = 108,
+	.next = &prog_test_struct,
+};
+
+noinline struct prog_test_ref_kfunc *
+bpf_kfunc_call_test_acquire(unsigned long *scalar_ptr)
+{
+	/* randomly return NULL */
+	if (get_jiffies_64() % 2)
+		return NULL;
+	return &prog_test_struct;
+}
+
+noinline void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p)
+{
+}
+
+struct prog_test_pass1 {
+	int x0;
+	struct {
+		int x1;
+		struct {
+			int x2;
+			struct {
+				int x3;
+			};
+		};
+	};
+};
+
+struct prog_test_pass2 {
+	int len;
+	short arr1[4];
+	struct {
+		char arr2[4];
+		unsigned long arr3[8];
+	} x;
+};
+
+struct prog_test_fail1 {
+	void *p;
+	int x;
+};
+
+struct prog_test_fail2 {
+	int x8;
+	struct prog_test_pass1 x;
+};
+
+struct prog_test_fail3 {
+	int len;
+	char arr1[2];
+	char arr2[0];
+};
+
+noinline void bpf_kfunc_call_test_pass_ctx(struct __sk_buff *skb)
+{
+}
+
+noinline void bpf_kfunc_call_test_pass1(struct prog_test_pass1 *p)
+{
+}
+
+noinline void bpf_kfunc_call_test_pass2(struct prog_test_pass2 *p)
+{
+}
+
+noinline void bpf_kfunc_call_test_fail1(struct prog_test_fail1 *p)
+{
+}
+
+noinline void bpf_kfunc_call_test_fail2(struct prog_test_fail2 *p)
+{
+}
+
+noinline void bpf_kfunc_call_test_fail3(struct prog_test_fail3 *p)
+{
+}
+
+noinline void bpf_kfunc_call_test_mem_len_pass1(void *mem, int mem__sz)
+{
+}
+
+noinline void bpf_kfunc_call_test_mem_len_fail1(void *mem, int len)
+{
+}
+
+noinline void bpf_kfunc_call_test_mem_len_fail2(u64 *mem, int len)
+{
+}
+
 __diag_pop();
 
 ALLOW_ERROR_INJECTION(bpf_modify_return_test, ERRNO);
 
-BTF_SET_START(test_sk_kfunc_ids)
+BTF_SET_START(test_sk_check_kfunc_ids)
 BTF_ID(func, bpf_kfunc_call_test1)
 BTF_ID(func, bpf_kfunc_call_test2)
 BTF_ID(func, bpf_kfunc_call_test3)
-BTF_SET_END(test_sk_kfunc_ids)
+BTF_ID(func, bpf_kfunc_call_test_acquire)
+BTF_ID(func, bpf_kfunc_call_test_release)
+BTF_ID(func, bpf_kfunc_call_test_pass_ctx)
+BTF_ID(func, bpf_kfunc_call_test_pass1)
+BTF_ID(func, bpf_kfunc_call_test_pass2)
+BTF_ID(func, bpf_kfunc_call_test_fail1)
+BTF_ID(func, bpf_kfunc_call_test_fail2)
+BTF_ID(func, bpf_kfunc_call_test_fail3)
+BTF_ID(func, bpf_kfunc_call_test_mem_len_pass1)
+BTF_ID(func, bpf_kfunc_call_test_mem_len_fail1)
+BTF_ID(func, bpf_kfunc_call_test_mem_len_fail2)
+BTF_SET_END(test_sk_check_kfunc_ids)
 
-bool bpf_prog_test_check_kfunc_call(u32 kfunc_id, struct module *owner)
-{
-	if (btf_id_set_contains(&test_sk_kfunc_ids, kfunc_id))
-		return true;
-	return bpf_check_mod_kfunc_call(&prog_test_kfunc_list, kfunc_id, owner);
-}
+BTF_SET_START(test_sk_acquire_kfunc_ids)
+BTF_ID(func, bpf_kfunc_call_test_acquire)
+BTF_SET_END(test_sk_acquire_kfunc_ids)
+
+BTF_SET_START(test_sk_release_kfunc_ids)
+BTF_ID(func, bpf_kfunc_call_test_release)
+BTF_SET_END(test_sk_release_kfunc_ids)
+
+BTF_SET_START(test_sk_ret_null_kfunc_ids)
+BTF_ID(func, bpf_kfunc_call_test_acquire)
+BTF_SET_END(test_sk_ret_null_kfunc_ids)
 
 static void *bpf_test_init(const union bpf_attr *kattr, u32 size,
 			   u32 headroom, u32 tailroom)
@@ -1067,3 +1185,17 @@ out:
 	kfree(ctx);
 	return err;
 }
+
+static const struct btf_kfunc_id_set bpf_prog_test_kfunc_set = {
+	.owner        = THIS_MODULE,
+	.check_set    = &test_sk_check_kfunc_ids,
+	.acquire_set  = &test_sk_acquire_kfunc_ids,
+	.release_set  = &test_sk_release_kfunc_ids,
+	.ret_null_set = &test_sk_ret_null_kfunc_ids,
+};
+
+static int __init bpf_prog_test_run_init(void)
+{
+	return register_btf_kfunc_id_set(BPF_PROG_TYPE_SCHED_CLS, &bpf_prog_test_kfunc_set);
+}
+late_initcall(bpf_prog_test_run_init);
diff --git a/net/core/filter.c b/net/core/filter.c
index 4603b7cd3cd1..f73a84c75970 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -10062,7 +10062,6 @@ const struct bpf_verifier_ops tc_cls_act_verifier_ops = {
 	.convert_ctx_access	= tc_cls_act_convert_ctx_access,
 	.gen_prologue		= tc_cls_act_prologue,
 	.gen_ld_abs		= bpf_gen_ld_abs,
-	.check_kfunc_call	= bpf_prog_test_check_kfunc_call,
 };
 
 const struct bpf_prog_ops tc_cls_act_prog_ops = {
diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
index 9b7171c40434..3b471781327f 100644
--- a/net/core/net_namespace.c
+++ b/net/core/net_namespace.c
@@ -299,6 +299,7 @@ struct net *get_net_ns_by_id(const struct net *net, int id)
 
 	return peer;
 }
+EXPORT_SYMBOL_GPL(get_net_ns_by_id);
 
 /*
  * setup_net runs the initializers for the network namespace object.
diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c
index de610cb83694..b60c9fd7147e 100644
--- a/net/ipv4/bpf_tcp_ca.c
+++ b/net/ipv4/bpf_tcp_ca.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2019 Facebook  */
 
+#include <linux/init.h>
 #include <linux/types.h>
 #include <linux/bpf_verifier.h>
 #include <linux/bpf.h>
@@ -212,26 +213,23 @@ bpf_tcp_ca_get_func_proto(enum bpf_func_id func_id,
 	}
 }
 
-BTF_SET_START(bpf_tcp_ca_kfunc_ids)
+BTF_SET_START(bpf_tcp_ca_check_kfunc_ids)
 BTF_ID(func, tcp_reno_ssthresh)
 BTF_ID(func, tcp_reno_cong_avoid)
 BTF_ID(func, tcp_reno_undo_cwnd)
 BTF_ID(func, tcp_slow_start)
 BTF_ID(func, tcp_cong_avoid_ai)
-BTF_SET_END(bpf_tcp_ca_kfunc_ids)
+BTF_SET_END(bpf_tcp_ca_check_kfunc_ids)
 
-static bool bpf_tcp_ca_check_kfunc_call(u32 kfunc_btf_id, struct module *owner)
-{
-	if (btf_id_set_contains(&bpf_tcp_ca_kfunc_ids, kfunc_btf_id))
-		return true;
-	return bpf_check_mod_kfunc_call(&bpf_tcp_ca_kfunc_list, kfunc_btf_id, owner);
-}
+static const struct btf_kfunc_id_set bpf_tcp_ca_kfunc_set = {
+	.owner     = THIS_MODULE,
+	.check_set = &bpf_tcp_ca_check_kfunc_ids,
+};
 
 static const struct bpf_verifier_ops bpf_tcp_ca_verifier_ops = {
 	.get_func_proto		= bpf_tcp_ca_get_func_proto,
 	.is_valid_access	= bpf_tcp_ca_is_valid_access,
 	.btf_struct_access	= bpf_tcp_ca_btf_struct_access,
-	.check_kfunc_call	= bpf_tcp_ca_check_kfunc_call,
 };
 
 static int bpf_tcp_ca_init_member(const struct btf_type *t,
@@ -300,3 +298,9 @@ struct bpf_struct_ops bpf_tcp_congestion_ops = {
 	.init = bpf_tcp_ca_init,
 	.name = "tcp_congestion_ops",
 };
+
+static int __init bpf_tcp_ca_kfunc_init(void)
+{
+	return register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &bpf_tcp_ca_kfunc_set);
+}
+late_initcall(bpf_tcp_ca_kfunc_init);
diff --git a/net/ipv4/tcp_bbr.c b/net/ipv4/tcp_bbr.c
index ec5550089b4d..02e8626ccb27 100644
--- a/net/ipv4/tcp_bbr.c
+++ b/net/ipv4/tcp_bbr.c
@@ -1154,7 +1154,7 @@ static struct tcp_congestion_ops tcp_bbr_cong_ops __read_mostly = {
 	.set_state	= bbr_set_state,
 };
 
-BTF_SET_START(tcp_bbr_kfunc_ids)
+BTF_SET_START(tcp_bbr_check_kfunc_ids)
 #ifdef CONFIG_X86
 #ifdef CONFIG_DYNAMIC_FTRACE
 BTF_ID(func, bbr_init)
@@ -1167,25 +1167,27 @@ BTF_ID(func, bbr_min_tso_segs)
 BTF_ID(func, bbr_set_state)
 #endif
 #endif
-BTF_SET_END(tcp_bbr_kfunc_ids)
+BTF_SET_END(tcp_bbr_check_kfunc_ids)
 
-static DEFINE_KFUNC_BTF_ID_SET(&tcp_bbr_kfunc_ids, tcp_bbr_kfunc_btf_set);
+static const struct btf_kfunc_id_set tcp_bbr_kfunc_set = {
+	.owner     = THIS_MODULE,
+	.check_set = &tcp_bbr_check_kfunc_ids,
+};
 
 static int __init bbr_register(void)
 {
 	int ret;
 
 	BUILD_BUG_ON(sizeof(struct bbr) > ICSK_CA_PRIV_SIZE);
-	ret = tcp_register_congestion_control(&tcp_bbr_cong_ops);
-	if (ret)
+
+	ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &tcp_bbr_kfunc_set);
+	if (ret < 0)
 		return ret;
-	register_kfunc_btf_id_set(&bpf_tcp_ca_kfunc_list, &tcp_bbr_kfunc_btf_set);
-	return 0;
+	return tcp_register_congestion_control(&tcp_bbr_cong_ops);
 }
 
 static void __exit bbr_unregister(void)
 {
-	unregister_kfunc_btf_id_set(&bpf_tcp_ca_kfunc_list, &tcp_bbr_kfunc_btf_set);
 	tcp_unregister_congestion_control(&tcp_bbr_cong_ops);
 }
 
diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c
index e07837e23b3f..24d562dd6225 100644
--- a/net/ipv4/tcp_cubic.c
+++ b/net/ipv4/tcp_cubic.c
@@ -485,7 +485,7 @@ static struct tcp_congestion_ops cubictcp __read_mostly = {
 	.name		= "cubic",
 };
 
-BTF_SET_START(tcp_cubic_kfunc_ids)
+BTF_SET_START(tcp_cubic_check_kfunc_ids)
 #ifdef CONFIG_X86
 #ifdef CONFIG_DYNAMIC_FTRACE
 BTF_ID(func, cubictcp_init)
@@ -496,9 +496,12 @@ BTF_ID(func, cubictcp_cwnd_event)
 BTF_ID(func, cubictcp_acked)
 #endif
 #endif
-BTF_SET_END(tcp_cubic_kfunc_ids)
+BTF_SET_END(tcp_cubic_check_kfunc_ids)
 
-static DEFINE_KFUNC_BTF_ID_SET(&tcp_cubic_kfunc_ids, tcp_cubic_kfunc_btf_set);
+static const struct btf_kfunc_id_set tcp_cubic_kfunc_set = {
+	.owner     = THIS_MODULE,
+	.check_set = &tcp_cubic_check_kfunc_ids,
+};
 
 static int __init cubictcp_register(void)
 {
@@ -534,16 +537,14 @@ static int __init cubictcp_register(void)
 	/* divide by bic_scale and by constant Srtt (100ms) */
 	do_div(cube_factor, bic_scale * 10);
 
-	ret = tcp_register_congestion_control(&cubictcp);
-	if (ret)
+	ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &tcp_cubic_kfunc_set);
+	if (ret < 0)
 		return ret;
-	register_kfunc_btf_id_set(&bpf_tcp_ca_kfunc_list, &tcp_cubic_kfunc_btf_set);
-	return 0;
+	return tcp_register_congestion_control(&cubictcp);
 }
 
 static void __exit cubictcp_unregister(void)
 {
-	unregister_kfunc_btf_id_set(&bpf_tcp_ca_kfunc_list, &tcp_cubic_kfunc_btf_set);
 	tcp_unregister_congestion_control(&cubictcp);
 }
 
diff --git a/net/ipv4/tcp_dctcp.c b/net/ipv4/tcp_dctcp.c
index 0d7ab3cc7b61..1943a6630341 100644
--- a/net/ipv4/tcp_dctcp.c
+++ b/net/ipv4/tcp_dctcp.c
@@ -238,7 +238,7 @@ static struct tcp_congestion_ops dctcp_reno __read_mostly = {
 	.name		= "dctcp-reno",
 };
 
-BTF_SET_START(tcp_dctcp_kfunc_ids)
+BTF_SET_START(tcp_dctcp_check_kfunc_ids)
 #ifdef CONFIG_X86
 #ifdef CONFIG_DYNAMIC_FTRACE
 BTF_ID(func, dctcp_init)
@@ -249,25 +249,27 @@ BTF_ID(func, dctcp_cwnd_undo)
 BTF_ID(func, dctcp_state)
 #endif
 #endif
-BTF_SET_END(tcp_dctcp_kfunc_ids)
+BTF_SET_END(tcp_dctcp_check_kfunc_ids)
 
-static DEFINE_KFUNC_BTF_ID_SET(&tcp_dctcp_kfunc_ids, tcp_dctcp_kfunc_btf_set);
+static const struct btf_kfunc_id_set tcp_dctcp_kfunc_set = {
+	.owner     = THIS_MODULE,
+	.check_set = &tcp_dctcp_check_kfunc_ids,
+};
 
 static int __init dctcp_register(void)
 {
 	int ret;
 
 	BUILD_BUG_ON(sizeof(struct dctcp) > ICSK_CA_PRIV_SIZE);
-	ret = tcp_register_congestion_control(&dctcp);
-	if (ret)
+
+	ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &tcp_dctcp_kfunc_set);
+	if (ret < 0)
 		return ret;
-	register_kfunc_btf_id_set(&bpf_tcp_ca_kfunc_list, &tcp_dctcp_kfunc_btf_set);
-	return 0;
+	return tcp_register_congestion_control(&dctcp);
 }
 
 static void __exit dctcp_unregister(void)
 {
-	unregister_kfunc_btf_id_set(&bpf_tcp_ca_kfunc_list, &tcp_dctcp_kfunc_btf_set);
 	tcp_unregister_congestion_control(&dctcp);
 }
 
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index a135b1a46014..238b6a620e88 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -14,6 +14,11 @@ nf_conntrack-$(CONFIG_NF_CONNTRACK_LABELS) += nf_conntrack_labels.o
 nf_conntrack-$(CONFIG_NF_CT_PROTO_DCCP) += nf_conntrack_proto_dccp.o
 nf_conntrack-$(CONFIG_NF_CT_PROTO_SCTP) += nf_conntrack_proto_sctp.o
 nf_conntrack-$(CONFIG_NF_CT_PROTO_GRE) += nf_conntrack_proto_gre.o
+ifeq ($(CONFIG_NF_CONNTRACK),m)
+nf_conntrack-$(CONFIG_DEBUG_INFO_BTF_MODULES) += nf_conntrack_bpf.o
+else ifeq ($(CONFIG_NF_CONNTRACK),y)
+nf_conntrack-$(CONFIG_DEBUG_INFO_BTF) += nf_conntrack_bpf.o
+endif
 
 obj-$(CONFIG_NETFILTER) = netfilter.o
 
diff --git a/net/netfilter/nf_conntrack_bpf.c b/net/netfilter/nf_conntrack_bpf.c
new file mode 100644
index 000000000000..8ad3f52579f3
--- /dev/null
+++ b/net/netfilter/nf_conntrack_bpf.c
@@ -0,0 +1,257 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Unstable Conntrack Helpers for XDP and TC-BPF hook
+ *
+ * These are called from the XDP and SCHED_CLS BPF programs. Note that it is
+ * allowed to break compatibility for these functions since the interface they
+ * are exposed through to BPF programs is explicitly unstable.
+ */
+
+#include <linux/bpf.h>
+#include <linux/btf.h>
+#include <linux/types.h>
+#include <linux/btf_ids.h>
+#include <linux/net_namespace.h>
+#include <net/netfilter/nf_conntrack.h>
+#include <net/netfilter/nf_conntrack_core.h>
+
+/* bpf_ct_opts - Options for CT lookup helpers
+ *
+ * Members:
+ * @netns_id   - Specify the network namespace for lookup
+ *		 Values:
+ *		   BPF_F_CURRENT_NETNS (-1)
+ *		     Use namespace associated with ctx (xdp_md, __sk_buff)
+ *		   [0, S32_MAX]
+ *		     Network Namespace ID
+ * @error      - Out parameter, set for any errors encountered
+ *		 Values:
+ *		   -EINVAL - Passed NULL for bpf_tuple pointer
+ *		   -EINVAL - opts->reserved is not 0
+ *		   -EINVAL - netns_id is less than -1
+ *		   -EINVAL - opts__sz isn't NF_BPF_CT_OPTS_SZ (12)
+ *		   -EPROTO - l4proto isn't one of IPPROTO_TCP or IPPROTO_UDP
+ *		   -ENONET - No network namespace found for netns_id
+ *		   -ENOENT - Conntrack lookup could not find entry for tuple
+ *		   -EAFNOSUPPORT - tuple__sz isn't one of sizeof(tuple->ipv4)
+ *				   or sizeof(tuple->ipv6)
+ * @l4proto    - Layer 4 protocol
+ *		 Values:
+ *		   IPPROTO_TCP, IPPROTO_UDP
+ * @reserved   - Reserved member, will be reused for more options in future
+ *		 Values:
+ *		   0
+ */
+struct bpf_ct_opts {
+	s32 netns_id;
+	s32 error;
+	u8 l4proto;
+	u8 reserved[3];
+};
+
+enum {
+	NF_BPF_CT_OPTS_SZ = 12,
+};
+
+static struct nf_conn *__bpf_nf_ct_lookup(struct net *net,
+					  struct bpf_sock_tuple *bpf_tuple,
+					  u32 tuple_len, u8 protonum,
+					  s32 netns_id)
+{
+	struct nf_conntrack_tuple_hash *hash;
+	struct nf_conntrack_tuple tuple;
+
+	if (unlikely(protonum != IPPROTO_TCP && protonum != IPPROTO_UDP))
+		return ERR_PTR(-EPROTO);
+	if (unlikely(netns_id < BPF_F_CURRENT_NETNS))
+		return ERR_PTR(-EINVAL);
+
+	memset(&tuple, 0, sizeof(tuple));
+	switch (tuple_len) {
+	case sizeof(bpf_tuple->ipv4):
+		tuple.src.l3num = AF_INET;
+		tuple.src.u3.ip = bpf_tuple->ipv4.saddr;
+		tuple.src.u.tcp.port = bpf_tuple->ipv4.sport;
+		tuple.dst.u3.ip = bpf_tuple->ipv4.daddr;
+		tuple.dst.u.tcp.port = bpf_tuple->ipv4.dport;
+		break;
+	case sizeof(bpf_tuple->ipv6):
+		tuple.src.l3num = AF_INET6;
+		memcpy(tuple.src.u3.ip6, bpf_tuple->ipv6.saddr, sizeof(bpf_tuple->ipv6.saddr));
+		tuple.src.u.tcp.port = bpf_tuple->ipv6.sport;
+		memcpy(tuple.dst.u3.ip6, bpf_tuple->ipv6.daddr, sizeof(bpf_tuple->ipv6.daddr));
+		tuple.dst.u.tcp.port = bpf_tuple->ipv6.dport;
+		break;
+	default:
+		return ERR_PTR(-EAFNOSUPPORT);
+	}
+
+	tuple.dst.protonum = protonum;
+
+	if (netns_id >= 0) {
+		net = get_net_ns_by_id(net, netns_id);
+		if (unlikely(!net))
+			return ERR_PTR(-ENONET);
+	}
+
+	hash = nf_conntrack_find_get(net, &nf_ct_zone_dflt, &tuple);
+	if (netns_id >= 0)
+		put_net(net);
+	if (!hash)
+		return ERR_PTR(-ENOENT);
+	return nf_ct_tuplehash_to_ctrack(hash);
+}
+
+__diag_push();
+__diag_ignore(GCC, 8, "-Wmissing-prototypes",
+	      "Global functions as their definitions will be in nf_conntrack BTF");
+
+/* bpf_xdp_ct_lookup - Lookup CT entry for the given tuple, and acquire a
+ *		       reference to it
+ *
+ * Parameters:
+ * @xdp_ctx	- Pointer to ctx (xdp_md) in XDP program
+ *		    Cannot be NULL
+ * @bpf_tuple	- Pointer to memory representing the tuple to look up
+ *		    Cannot be NULL
+ * @tuple__sz	- Length of the tuple structure
+ *		    Must be one of sizeof(bpf_tuple->ipv4) or
+ *		    sizeof(bpf_tuple->ipv6)
+ * @opts	- Additional options for lookup (documented above)
+ *		    Cannot be NULL
+ * @opts__sz	- Length of the bpf_ct_opts structure
+ *		    Must be NF_BPF_CT_OPTS_SZ (12)
+ */
+struct nf_conn *
+bpf_xdp_ct_lookup(struct xdp_md *xdp_ctx, struct bpf_sock_tuple *bpf_tuple,
+		  u32 tuple__sz, struct bpf_ct_opts *opts, u32 opts__sz)
+{
+	struct xdp_buff *ctx = (struct xdp_buff *)xdp_ctx;
+	struct net *caller_net;
+	struct nf_conn *nfct;
+
+	BUILD_BUG_ON(sizeof(struct bpf_ct_opts) != NF_BPF_CT_OPTS_SZ);
+
+	if (!opts)
+		return NULL;
+	if (!bpf_tuple || opts->reserved[0] || opts->reserved[1] ||
+	    opts->reserved[2] || opts__sz != NF_BPF_CT_OPTS_SZ) {
+		opts->error = -EINVAL;
+		return NULL;
+	}
+	caller_net = dev_net(ctx->rxq->dev);
+	nfct = __bpf_nf_ct_lookup(caller_net, bpf_tuple, tuple__sz, opts->l4proto,
+				  opts->netns_id);
+	if (IS_ERR(nfct)) {
+		opts->error = PTR_ERR(nfct);
+		return NULL;
+	}
+	return nfct;
+}
+
+/* bpf_skb_ct_lookup - Lookup CT entry for the given tuple, and acquire a
+ *		       reference to it
+ *
+ * Parameters:
+ * @skb_ctx	- Pointer to ctx (__sk_buff) in TC program
+ *		    Cannot be NULL
+ * @bpf_tuple	- Pointer to memory representing the tuple to look up
+ *		    Cannot be NULL
+ * @tuple__sz	- Length of the tuple structure
+ *		    Must be one of sizeof(bpf_tuple->ipv4) or
+ *		    sizeof(bpf_tuple->ipv6)
+ * @opts	- Additional options for lookup (documented above)
+ *		    Cannot be NULL
+ * @opts__sz	- Length of the bpf_ct_opts structure
+ *		    Must be NF_BPF_CT_OPTS_SZ (12)
+ */
+struct nf_conn *
+bpf_skb_ct_lookup(struct __sk_buff *skb_ctx, struct bpf_sock_tuple *bpf_tuple,
+		  u32 tuple__sz, struct bpf_ct_opts *opts, u32 opts__sz)
+{
+	struct sk_buff *skb = (struct sk_buff *)skb_ctx;
+	struct net *caller_net;
+	struct nf_conn *nfct;
+
+	BUILD_BUG_ON(sizeof(struct bpf_ct_opts) != NF_BPF_CT_OPTS_SZ);
+
+	if (!opts)
+		return NULL;
+	if (!bpf_tuple || opts->reserved[0] || opts->reserved[1] ||
+	    opts->reserved[2] || opts__sz != NF_BPF_CT_OPTS_SZ) {
+		opts->error = -EINVAL;
+		return NULL;
+	}
+	caller_net = skb->dev ? dev_net(skb->dev) : sock_net(skb->sk);
+	nfct = __bpf_nf_ct_lookup(caller_net, bpf_tuple, tuple__sz, opts->l4proto,
+				  opts->netns_id);
+	if (IS_ERR(nfct)) {
+		opts->error = PTR_ERR(nfct);
+		return NULL;
+	}
+	return nfct;
+}
+
+/* bpf_ct_release - Release acquired nf_conn object
+ *
+ * This must be invoked for referenced PTR_TO_BTF_ID, and the verifier rejects
+ * the program if any references remain in the program in all of the explored
+ * states.
+ *
+ * Parameters:
+ * @nf_conn	 - Pointer to referenced nf_conn object, obtained using
+ *		   bpf_xdp_ct_lookup or bpf_skb_ct_lookup.
+ */
+void bpf_ct_release(struct nf_conn *nfct)
+{
+	if (!nfct)
+		return;
+	nf_ct_put(nfct);
+}
+
+__diag_pop()
+
+BTF_SET_START(nf_ct_xdp_check_kfunc_ids)
+BTF_ID(func, bpf_xdp_ct_lookup)
+BTF_ID(func, bpf_ct_release)
+BTF_SET_END(nf_ct_xdp_check_kfunc_ids)
+
+BTF_SET_START(nf_ct_tc_check_kfunc_ids)
+BTF_ID(func, bpf_skb_ct_lookup)
+BTF_ID(func, bpf_ct_release)
+BTF_SET_END(nf_ct_tc_check_kfunc_ids)
+
+BTF_SET_START(nf_ct_acquire_kfunc_ids)
+BTF_ID(func, bpf_xdp_ct_lookup)
+BTF_ID(func, bpf_skb_ct_lookup)
+BTF_SET_END(nf_ct_acquire_kfunc_ids)
+
+BTF_SET_START(nf_ct_release_kfunc_ids)
+BTF_ID(func, bpf_ct_release)
+BTF_SET_END(nf_ct_release_kfunc_ids)
+
+/* Both sets are identical */
+#define nf_ct_ret_null_kfunc_ids nf_ct_acquire_kfunc_ids
+
+static const struct btf_kfunc_id_set nf_conntrack_xdp_kfunc_set = {
+	.owner        = THIS_MODULE,
+	.check_set    = &nf_ct_xdp_check_kfunc_ids,
+	.acquire_set  = &nf_ct_acquire_kfunc_ids,
+	.release_set  = &nf_ct_release_kfunc_ids,
+	.ret_null_set = &nf_ct_ret_null_kfunc_ids,
+};
+
+static const struct btf_kfunc_id_set nf_conntrack_tc_kfunc_set = {
+	.owner        = THIS_MODULE,
+	.check_set    = &nf_ct_tc_check_kfunc_ids,
+	.acquire_set  = &nf_ct_acquire_kfunc_ids,
+	.release_set  = &nf_ct_release_kfunc_ids,
+	.ret_null_set = &nf_ct_ret_null_kfunc_ids,
+};
+
+int register_nf_conntrack_bpf(void)
+{
+	int ret;
+
+	ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_XDP, &nf_conntrack_xdp_kfunc_set);
+	return ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_SCHED_CLS, &nf_conntrack_tc_kfunc_set);
+}
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 894a325d39f2..3b60fca61920 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -34,6 +34,7 @@
 #include <linux/rculist_nulls.h>
 
 #include <net/netfilter/nf_conntrack.h>
+#include <net/netfilter/nf_conntrack_bpf.h>
 #include <net/netfilter/nf_conntrack_l4proto.h>
 #include <net/netfilter/nf_conntrack_expect.h>
 #include <net/netfilter/nf_conntrack_helper.h>
@@ -2748,8 +2749,15 @@ int nf_conntrack_init_start(void)
 	conntrack_gc_work_init(&conntrack_gc_work);
 	queue_delayed_work(system_power_efficient_wq, &conntrack_gc_work.dwork, HZ);
 
+	ret = register_nf_conntrack_bpf();
+	if (ret < 0)
+		goto err_kfunc;
+
 	return 0;
 
+err_kfunc:
+	cancel_delayed_work_sync(&conntrack_gc_work.dwork);
+	nf_conntrack_proto_fini();
 err_proto:
 	nf_conntrack_seqadj_fini();
 err_seqadj:
diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
index df3b292a8ffe..bdbacf5adcd2 100644
--- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
+++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
@@ -109,26 +109,31 @@ static struct bin_attribute bin_attr_bpf_testmod_file __ro_after_init = {
 	.write = bpf_testmod_test_write,
 };
 
-BTF_SET_START(bpf_testmod_kfunc_ids)
+BTF_SET_START(bpf_testmod_check_kfunc_ids)
 BTF_ID(func, bpf_testmod_test_mod_kfunc)
-BTF_SET_END(bpf_testmod_kfunc_ids)
+BTF_SET_END(bpf_testmod_check_kfunc_ids)
 
-static DEFINE_KFUNC_BTF_ID_SET(&bpf_testmod_kfunc_ids, bpf_testmod_kfunc_btf_set);
+static const struct btf_kfunc_id_set bpf_testmod_kfunc_set = {
+	.owner     = THIS_MODULE,
+	.check_set = &bpf_testmod_check_kfunc_ids,
+};
+
+extern int bpf_fentry_test1(int a);
 
 static int bpf_testmod_init(void)
 {
 	int ret;
 
-	ret = sysfs_create_bin_file(kernel_kobj, &bin_attr_bpf_testmod_file);
-	if (ret)
+	ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_SCHED_CLS, &bpf_testmod_kfunc_set);
+	if (ret < 0)
 		return ret;
-	register_kfunc_btf_id_set(&prog_test_kfunc_list, &bpf_testmod_kfunc_btf_set);
-	return 0;
+	if (bpf_fentry_test1(0) < 0)
+		return -EINVAL;
+	return sysfs_create_bin_file(kernel_kobj, &bin_attr_bpf_testmod_file);
 }
 
 static void bpf_testmod_exit(void)
 {
-	unregister_kfunc_btf_id_set(&prog_test_kfunc_list, &bpf_testmod_kfunc_btf_set);
 	return sysfs_remove_bin_file(kernel_kobj, &bin_attr_bpf_testmod_file);
 }
 
diff --git a/tools/testing/selftests/bpf/config b/tools/testing/selftests/bpf/config
index f6287132fa89..763db63a3890 100644
--- a/tools/testing/selftests/bpf/config
+++ b/tools/testing/selftests/bpf/config
@@ -48,3 +48,8 @@ CONFIG_IMA_READ_POLICY=y
 CONFIG_BLK_DEV_LOOP=y
 CONFIG_FUNCTION_TRACER=y
 CONFIG_DYNAMIC_FTRACE=y
+CONFIG_NETFILTER=y
+CONFIG_NF_DEFRAG_IPV4=y
+CONFIG_NF_DEFRAG_IPV6=y
+CONFIG_NF_CONNTRACK=y
+CONFIG_USERFAULTFD=y
diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_mod_race.c b/tools/testing/selftests/bpf/prog_tests/bpf_mod_race.c
new file mode 100644
index 000000000000..d43f548c572c
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_mod_race.c
@@ -0,0 +1,230 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <unistd.h>
+#include <pthread.h>
+#include <sys/mman.h>
+#include <stdatomic.h>
+#include <test_progs.h>
+#include <sys/syscall.h>
+#include <linux/module.h>
+#include <linux/userfaultfd.h>
+
+#include "ksym_race.skel.h"
+#include "bpf_mod_race.skel.h"
+#include "kfunc_call_race.skel.h"
+
+/* This test crafts a race between btf_try_get_module and do_init_module, and
+ * checks whether btf_try_get_module handles the invocation for a well-formed
+ * but uninitialized module correctly. Unless the module has completed its
+ * initcalls, the verifier should fail the program load and return ENXIO.
+ *
+ * userfaultfd is used to trigger a fault in an fmod_ret program, and make it
+ * sleep, then the BPF program is loaded and the return value from verifier is
+ * inspected. After this, the userfaultfd is closed so that the module loading
+ * thread makes forward progress, and fmod_ret injects an error so that the
+ * module load fails and it is freed.
+ *
+ * If the verifier succeeded in loading the supplied program, it will end up
+ * taking reference to freed module, and trigger a crash when the program fd
+ * is closed later. This is true for both kfuncs and ksyms. In both cases,
+ * the crash is triggered inside bpf_prog_free_deferred, when module reference
+ * is finally released.
+ */
+
+struct test_config {
+	const char *str_open;
+	void *(*bpf_open_and_load)();
+	void (*bpf_destroy)(void *);
+};
+
+enum test_state {
+	_TS_INVALID,
+	TS_MODULE_LOAD,
+	TS_MODULE_LOAD_FAIL,
+};
+
+static _Atomic enum test_state state = _TS_INVALID;
+
+static int sys_finit_module(int fd, const char *param_values, int flags)
+{
+	return syscall(__NR_finit_module, fd, param_values, flags);
+}
+
+static int sys_delete_module(const char *name, unsigned int flags)
+{
+	return syscall(__NR_delete_module, name, flags);
+}
+
+static int load_module(const char *mod)
+{
+	int ret, fd;
+
+	fd = open("bpf_testmod.ko", O_RDONLY);
+	if (fd < 0)
+		return fd;
+
+	ret = sys_finit_module(fd, "", 0);
+	close(fd);
+	if (ret < 0)
+		return ret;
+	return 0;
+}
+
+static void *load_module_thread(void *p)
+{
+
+	if (!ASSERT_NEQ(load_module("bpf_testmod.ko"), 0, "load_module_thread must fail"))
+		atomic_store(&state, TS_MODULE_LOAD);
+	else
+		atomic_store(&state, TS_MODULE_LOAD_FAIL);
+	return p;
+}
+
+static int sys_userfaultfd(int flags)
+{
+	return syscall(__NR_userfaultfd, flags);
+}
+
+static int test_setup_uffd(void *fault_addr)
+{
+	struct uffdio_register uffd_register = {};
+	struct uffdio_api uffd_api = {};
+	int uffd;
+
+	uffd = sys_userfaultfd(O_CLOEXEC);
+	if (uffd < 0)
+		return -errno;
+
+	uffd_api.api = UFFD_API;
+	uffd_api.features = 0;
+	if (ioctl(uffd, UFFDIO_API, &uffd_api)) {
+		close(uffd);
+		return -1;
+	}
+
+	uffd_register.range.start = (unsigned long)fault_addr;
+	uffd_register.range.len = 4096;
+	uffd_register.mode = UFFDIO_REGISTER_MODE_MISSING;
+	if (ioctl(uffd, UFFDIO_REGISTER, &uffd_register)) {
+		close(uffd);
+		return -1;
+	}
+	return uffd;
+}
+
+static void test_bpf_mod_race_config(const struct test_config *config)
+{
+	void *fault_addr, *skel_fail;
+	struct bpf_mod_race *skel;
+	struct uffd_msg uffd_msg;
+	pthread_t load_mod_thrd;
+	_Atomic int *blockingp;
+	int uffd, ret;
+
+	fault_addr = mmap(0, 4096, PROT_READ, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
+	if (!ASSERT_NEQ(fault_addr, MAP_FAILED, "mmap for uffd registration"))
+		return;
+
+	if (!ASSERT_OK(sys_delete_module("bpf_testmod", 0), "unload bpf_testmod"))
+		goto end_mmap;
+
+	skel = bpf_mod_race__open();
+	if (!ASSERT_OK_PTR(skel, "bpf_mod_kfunc_race__open"))
+		goto end_module;
+
+	skel->rodata->bpf_mod_race_config.tgid = getpid();
+	skel->rodata->bpf_mod_race_config.inject_error = -4242;
+	skel->rodata->bpf_mod_race_config.fault_addr = fault_addr;
+	if (!ASSERT_OK(bpf_mod_race__load(skel), "bpf_mod___load"))
+		goto end_destroy;
+	blockingp = (_Atomic int *)&skel->bss->bpf_blocking;
+
+	if (!ASSERT_OK(bpf_mod_race__attach(skel), "bpf_mod_kfunc_race__attach"))
+		goto end_destroy;
+
+	uffd = test_setup_uffd(fault_addr);
+	if (!ASSERT_GE(uffd, 0, "userfaultfd open + register address"))
+		goto end_destroy;
+
+	if (!ASSERT_OK(pthread_create(&load_mod_thrd, NULL, load_module_thread, NULL),
+		       "load module thread"))
+		goto end_uffd;
+
+	/* Now, we either fail loading module, or block in bpf prog, spin to find out */
+	while (!atomic_load(&state) && !atomic_load(blockingp))
+		;
+	if (!ASSERT_EQ(state, _TS_INVALID, "module load should block"))
+		goto end_join;
+	if (!ASSERT_EQ(*blockingp, 1, "module load blocked")) {
+		pthread_kill(load_mod_thrd, SIGKILL);
+		goto end_uffd;
+	}
+
+	/* We might have set bpf_blocking to 1, but may have not blocked in
+	 * bpf_copy_from_user. Read userfaultfd descriptor to verify that.
+	 */
+	if (!ASSERT_EQ(read(uffd, &uffd_msg, sizeof(uffd_msg)), sizeof(uffd_msg),
+		       "read uffd block event"))
+		goto end_join;
+	if (!ASSERT_EQ(uffd_msg.event, UFFD_EVENT_PAGEFAULT, "read uffd event is pagefault"))
+		goto end_join;
+
+	/* We know that load_mod_thrd is blocked in the fmod_ret program, the
+	 * module state is still MODULE_STATE_COMING because mod->init hasn't
+	 * returned. This is the time we try to load a program calling kfunc and
+	 * check if we get ENXIO from verifier.
+	 */
+	skel_fail = config->bpf_open_and_load();
+	ret = errno;
+	if (!ASSERT_EQ(skel_fail, NULL, config->str_open)) {
+		/* Close uffd to unblock load_mod_thrd */
+		close(uffd);
+		uffd = -1;
+		while (atomic_load(blockingp) != 2)
+			;
+		ASSERT_OK(kern_sync_rcu(), "kern_sync_rcu");
+		config->bpf_destroy(skel_fail);
+		goto end_join;
+
+	}
+	ASSERT_EQ(ret, ENXIO, "verifier returns ENXIO");
+	ASSERT_EQ(skel->data->res_try_get_module, false, "btf_try_get_module == false");
+
+	close(uffd);
+	uffd = -1;
+end_join:
+	pthread_join(load_mod_thrd, NULL);
+	if (uffd < 0)
+		ASSERT_EQ(atomic_load(&state), TS_MODULE_LOAD_FAIL, "load_mod_thrd success");
+end_uffd:
+	if (uffd >= 0)
+		close(uffd);
+end_destroy:
+	bpf_mod_race__destroy(skel);
+	ASSERT_OK(kern_sync_rcu(), "kern_sync_rcu");
+end_module:
+	sys_delete_module("bpf_testmod", 0);
+	ASSERT_OK(load_module("bpf_testmod.ko"), "restore bpf_testmod");
+end_mmap:
+	munmap(fault_addr, 4096);
+	atomic_store(&state, _TS_INVALID);
+}
+
+static const struct test_config ksym_config = {
+	.str_open = "ksym_race__open_and_load",
+	.bpf_open_and_load = (void *)ksym_race__open_and_load,
+	.bpf_destroy = (void *)ksym_race__destroy,
+};
+
+static const struct test_config kfunc_config = {
+	.str_open = "kfunc_call_race__open_and_load",
+	.bpf_open_and_load = (void *)kfunc_call_race__open_and_load,
+	.bpf_destroy = (void *)kfunc_call_race__destroy,
+};
+
+void serial_test_bpf_mod_race(void)
+{
+	if (test__start_subtest("ksym (used_btfs UAF)"))
+		test_bpf_mod_race_config(&ksym_config);
+	if (test__start_subtest("kfunc (kfunc_btf_tab UAF)"))
+		test_bpf_mod_race_config(&kfunc_config);
+}
diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_nf.c b/tools/testing/selftests/bpf/prog_tests/bpf_nf.c
new file mode 100644
index 000000000000..e3166a81e989
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_nf.c
@@ -0,0 +1,48 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <test_progs.h>
+#include <network_helpers.h>
+#include "test_bpf_nf.skel.h"
+
+enum {
+	TEST_XDP,
+	TEST_TC_BPF,
+};
+
+void test_bpf_nf_ct(int mode)
+{
+	struct test_bpf_nf *skel;
+	int prog_fd, err, retval;
+
+	skel = test_bpf_nf__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "test_bpf_nf__open_and_load"))
+		return;
+
+	if (mode == TEST_XDP)
+		prog_fd = bpf_program__fd(skel->progs.nf_xdp_ct_test);
+	else
+		prog_fd = bpf_program__fd(skel->progs.nf_skb_ct_test);
+
+	err = bpf_prog_test_run(prog_fd, 1, &pkt_v4, sizeof(pkt_v4), NULL, NULL,
+				(__u32 *)&retval, NULL);
+	if (!ASSERT_OK(err, "bpf_prog_test_run"))
+		goto end;
+
+	ASSERT_EQ(skel->bss->test_einval_bpf_tuple, -EINVAL, "Test EINVAL for NULL bpf_tuple");
+	ASSERT_EQ(skel->bss->test_einval_reserved, -EINVAL, "Test EINVAL for reserved not set to 0");
+	ASSERT_EQ(skel->bss->test_einval_netns_id, -EINVAL, "Test EINVAL for netns_id < -1");
+	ASSERT_EQ(skel->bss->test_einval_len_opts, -EINVAL, "Test EINVAL for len__opts != NF_BPF_CT_OPTS_SZ");
+	ASSERT_EQ(skel->bss->test_eproto_l4proto, -EPROTO, "Test EPROTO for l4proto != TCP or UDP");
+	ASSERT_EQ(skel->bss->test_enonet_netns_id, -ENONET, "Test ENONET for bad but valid netns_id");
+	ASSERT_EQ(skel->bss->test_enoent_lookup, -ENOENT, "Test ENOENT for failed lookup");
+	ASSERT_EQ(skel->bss->test_eafnosupport, -EAFNOSUPPORT, "Test EAFNOSUPPORT for invalid len__tuple");
+end:
+	test_bpf_nf__destroy(skel);
+}
+
+void test_bpf_nf(void)
+{
+	if (test__start_subtest("xdp-ct"))
+		test_bpf_nf_ct(TEST_XDP);
+	if (test__start_subtest("tc-bpf-ct"))
+		test_bpf_nf_ct(TEST_TC_BPF);
+}
diff --git a/tools/testing/selftests/bpf/prog_tests/kfunc_call.c b/tools/testing/selftests/bpf/prog_tests/kfunc_call.c
index 7d7445ccc141..b39a4f09aefd 100644
--- a/tools/testing/selftests/bpf/prog_tests/kfunc_call.c
+++ b/tools/testing/selftests/bpf/prog_tests/kfunc_call.c
@@ -27,6 +27,12 @@ static void test_main(void)
 	ASSERT_OK(err, "bpf_prog_test_run(test2)");
 	ASSERT_EQ(retval, 3, "test2-retval");
 
+	prog_fd = skel->progs.kfunc_call_test_ref_btf_id.prog_fd;
+	err = bpf_prog_test_run(prog_fd, 1, &pkt_v4, sizeof(pkt_v4),
+				NULL, NULL, (__u32 *)&retval, NULL);
+	ASSERT_OK(err, "bpf_prog_test_run(test_ref_btf_id)");
+	ASSERT_EQ(retval, 0, "test_ref_btf_id-retval");
+
 	kfunc_call_test_lskel__destroy(skel);
 }
 
diff --git a/tools/testing/selftests/bpf/progs/bpf_mod_race.c b/tools/testing/selftests/bpf/progs/bpf_mod_race.c
new file mode 100644
index 000000000000..82a5c6c6ba83
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/bpf_mod_race.c
@@ -0,0 +1,100 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
+
+const volatile struct {
+	/* thread to activate trace programs for */
+	pid_t tgid;
+	/* return error from __init function */
+	int inject_error;
+	/* uffd monitored range start address */
+	void *fault_addr;
+} bpf_mod_race_config = { -1 };
+
+int bpf_blocking = 0;
+int res_try_get_module = -1;
+
+static __always_inline bool check_thread_id(void)
+{
+	struct task_struct *task = bpf_get_current_task_btf();
+
+	return task->tgid == bpf_mod_race_config.tgid;
+}
+
+/* The trace of execution is something like this:
+ *
+ * finit_module()
+ *   load_module()
+ *     prepare_coming_module()
+ *       notifier_call(MODULE_STATE_COMING)
+ *         btf_parse_module()
+ *         btf_alloc_id()		// Visible to userspace at this point
+ *         list_add(btf_mod->list, &btf_modules)
+ *     do_init_module()
+ *       freeinit = kmalloc()
+ *       ret = mod->init()
+ *         bpf_prog_widen_race()
+ *           bpf_copy_from_user()
+ *             ...<sleep>...
+ *       if (ret < 0)
+ *         ...
+ *         free_module()
+ * return ret
+ *
+ * At this point, module loading thread is blocked, we now load the program:
+ *
+ * bpf_check
+ *   add_kfunc_call/check_pseudo_btf_id
+ *     btf_try_get_module
+ *       try_get_module_live == false
+ *     return -ENXIO
+ *
+ * Without the fix (try_get_module_live in btf_try_get_module):
+ *
+ * bpf_check
+ *   add_kfunc_call/check_pseudo_btf_id
+ *     btf_try_get_module
+ *       try_get_module == true
+ *     <store module reference in btf_kfunc_tab or used_btf array>
+ *   ...
+ * return fd
+ *
+ * Now, if we inject an error in the blocked program, our module will be freed
+ * (going straight from MODULE_STATE_COMING to MODULE_STATE_GOING).
+ * Later, when bpf program is freed, it will try to module_put already freed
+ * module. This is why try_get_module_live returns false if mod->state is not
+ * MODULE_STATE_LIVE.
+ */
+
+SEC("fmod_ret.s/bpf_fentry_test1")
+int BPF_PROG(widen_race, int a, int ret)
+{
+	char dst;
+
+	if (!check_thread_id())
+		return 0;
+	/* Indicate that we will attempt to block */
+	bpf_blocking = 1;
+	bpf_copy_from_user(&dst, 1, bpf_mod_race_config.fault_addr);
+	return bpf_mod_race_config.inject_error;
+}
+
+SEC("fexit/do_init_module")
+int BPF_PROG(fexit_init_module, struct module *mod, int ret)
+{
+	if (!check_thread_id())
+		return 0;
+	/* Indicate that we finished blocking */
+	bpf_blocking = 2;
+	return 0;
+}
+
+SEC("fexit/btf_try_get_module")
+int BPF_PROG(fexit_module_get, const struct btf *btf, struct module *mod)
+{
+	res_try_get_module = !!mod;
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/kfunc_call_race.c b/tools/testing/selftests/bpf/progs/kfunc_call_race.c
new file mode 100644
index 000000000000..4e8fed75a4e0
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/kfunc_call_race.c
@@ -0,0 +1,14 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+
+extern void bpf_testmod_test_mod_kfunc(int i) __ksym;
+
+SEC("tc")
+int kfunc_call_fail(struct __sk_buff *ctx)
+{
+	bpf_testmod_test_mod_kfunc(0);
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/kfunc_call_test.c b/tools/testing/selftests/bpf/progs/kfunc_call_test.c
index 8a8cf59017aa..5aecbb9fdc68 100644
--- a/tools/testing/selftests/bpf/progs/kfunc_call_test.c
+++ b/tools/testing/selftests/bpf/progs/kfunc_call_test.c
@@ -1,13 +1,20 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2021 Facebook */
-#include <linux/bpf.h>
+#include <vmlinux.h>
 #include <bpf/bpf_helpers.h>
-#include "bpf_tcp_helpers.h"
 
 extern int bpf_kfunc_call_test2(struct sock *sk, __u32 a, __u32 b) __ksym;
 extern __u64 bpf_kfunc_call_test1(struct sock *sk, __u32 a, __u64 b,
 				  __u32 c, __u64 d) __ksym;
 
+extern struct prog_test_ref_kfunc *bpf_kfunc_call_test_acquire(unsigned long *sp) __ksym;
+extern void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p) __ksym;
+extern void bpf_kfunc_call_test_pass_ctx(struct __sk_buff *skb) __ksym;
+extern void bpf_kfunc_call_test_pass1(struct prog_test_pass1 *p) __ksym;
+extern void bpf_kfunc_call_test_pass2(struct prog_test_pass2 *p) __ksym;
+extern void bpf_kfunc_call_test_mem_len_pass1(void *mem, int len) __ksym;
+extern void bpf_kfunc_call_test_mem_len_fail2(__u64 *mem, int len) __ksym;
+
 SEC("tc")
 int kfunc_call_test2(struct __sk_buff *skb)
 {
@@ -44,4 +51,45 @@ int kfunc_call_test1(struct __sk_buff *skb)
 	return ret;
 }
 
+SEC("tc")
+int kfunc_call_test_ref_btf_id(struct __sk_buff *skb)
+{
+	struct prog_test_ref_kfunc *pt;
+	unsigned long s = 0;
+	int ret = 0;
+
+	pt = bpf_kfunc_call_test_acquire(&s);
+	if (pt) {
+		if (pt->a != 42 || pt->b != 108)
+			ret = -1;
+		bpf_kfunc_call_test_release(pt);
+	}
+	return ret;
+}
+
+SEC("tc")
+int kfunc_call_test_pass(struct __sk_buff *skb)
+{
+	struct prog_test_pass1 p1 = {};
+	struct prog_test_pass2 p2 = {};
+	short a = 0;
+	__u64 b = 0;
+	long c = 0;
+	char d = 0;
+	int e = 0;
+
+	bpf_kfunc_call_test_pass_ctx(skb);
+	bpf_kfunc_call_test_pass1(&p1);
+	bpf_kfunc_call_test_pass2(&p2);
+
+	bpf_kfunc_call_test_mem_len_pass1(&a, sizeof(a));
+	bpf_kfunc_call_test_mem_len_pass1(&b, sizeof(b));
+	bpf_kfunc_call_test_mem_len_pass1(&c, sizeof(c));
+	bpf_kfunc_call_test_mem_len_pass1(&d, sizeof(d));
+	bpf_kfunc_call_test_mem_len_pass1(&e, sizeof(e));
+	bpf_kfunc_call_test_mem_len_fail2(&b, -1);
+
+	return 0;
+}
+
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/ksym_race.c b/tools/testing/selftests/bpf/progs/ksym_race.c
new file mode 100644
index 000000000000..def97f2fed90
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/ksym_race.c
@@ -0,0 +1,13 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+
+extern int bpf_testmod_ksym_percpu __ksym;
+
+SEC("tc")
+int ksym_fail(struct __sk_buff *ctx)
+{
+	return *(int *)bpf_this_cpu_ptr(&bpf_testmod_ksym_percpu);
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/test_bpf_nf.c b/tools/testing/selftests/bpf/progs/test_bpf_nf.c
new file mode 100644
index 000000000000..6f131c993c0b
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_bpf_nf.c
@@ -0,0 +1,109 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+
+#define EAFNOSUPPORT 97
+#define EPROTO 71
+#define ENONET 64
+#define EINVAL 22
+#define ENOENT 2
+
+int test_einval_bpf_tuple = 0;
+int test_einval_reserved = 0;
+int test_einval_netns_id = 0;
+int test_einval_len_opts = 0;
+int test_eproto_l4proto = 0;
+int test_enonet_netns_id = 0;
+int test_enoent_lookup = 0;
+int test_eafnosupport = 0;
+
+struct nf_conn *bpf_xdp_ct_lookup(struct xdp_md *, struct bpf_sock_tuple *, u32,
+				  struct bpf_ct_opts *, u32) __ksym;
+struct nf_conn *bpf_skb_ct_lookup(struct __sk_buff *, struct bpf_sock_tuple *, u32,
+				  struct bpf_ct_opts *, u32) __ksym;
+void bpf_ct_release(struct nf_conn *) __ksym;
+
+static __always_inline void
+nf_ct_test(struct nf_conn *(*func)(void *, struct bpf_sock_tuple *, u32,
+				   struct bpf_ct_opts *, u32),
+	   void *ctx)
+{
+	struct bpf_ct_opts opts_def = { .l4proto = IPPROTO_TCP, .netns_id = -1 };
+	struct bpf_sock_tuple bpf_tuple;
+	struct nf_conn *ct;
+
+	__builtin_memset(&bpf_tuple, 0, sizeof(bpf_tuple.ipv4));
+
+	ct = func(ctx, NULL, 0, &opts_def, sizeof(opts_def));
+	if (ct)
+		bpf_ct_release(ct);
+	else
+		test_einval_bpf_tuple = opts_def.error;
+
+	opts_def.reserved[0] = 1;
+	ct = func(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4), &opts_def, sizeof(opts_def));
+	opts_def.reserved[0] = 0;
+	opts_def.l4proto = IPPROTO_TCP;
+	if (ct)
+		bpf_ct_release(ct);
+	else
+		test_einval_reserved = opts_def.error;
+
+	opts_def.netns_id = -2;
+	ct = func(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4), &opts_def, sizeof(opts_def));
+	opts_def.netns_id = -1;
+	if (ct)
+		bpf_ct_release(ct);
+	else
+		test_einval_netns_id = opts_def.error;
+
+	ct = func(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4), &opts_def, sizeof(opts_def) - 1);
+	if (ct)
+		bpf_ct_release(ct);
+	else
+		test_einval_len_opts = opts_def.error;
+
+	opts_def.l4proto = IPPROTO_ICMP;
+	ct = func(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4), &opts_def, sizeof(opts_def));
+	opts_def.l4proto = IPPROTO_TCP;
+	if (ct)
+		bpf_ct_release(ct);
+	else
+		test_eproto_l4proto = opts_def.error;
+
+	opts_def.netns_id = 0xf00f;
+	ct = func(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4), &opts_def, sizeof(opts_def));
+	opts_def.netns_id = -1;
+	if (ct)
+		bpf_ct_release(ct);
+	else
+		test_enonet_netns_id = opts_def.error;
+
+	ct = func(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4), &opts_def, sizeof(opts_def));
+	if (ct)
+		bpf_ct_release(ct);
+	else
+		test_enoent_lookup = opts_def.error;
+
+	ct = func(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4) - 1, &opts_def, sizeof(opts_def));
+	if (ct)
+		bpf_ct_release(ct);
+	else
+		test_eafnosupport = opts_def.error;
+}
+
+SEC("xdp")
+int nf_xdp_ct_test(struct xdp_md *ctx)
+{
+	nf_ct_test((void *)bpf_xdp_ct_lookup, ctx);
+	return 0;
+}
+
+SEC("tc")
+int nf_skb_ct_test(struct __sk_buff *ctx)
+{
+	nf_ct_test((void *)bpf_skb_ct_lookup, ctx);
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/test_verifier.c b/tools/testing/selftests/bpf/test_verifier.c
index 76cd903117af..29bbaa58233c 100644
--- a/tools/testing/selftests/bpf/test_verifier.c
+++ b/tools/testing/selftests/bpf/test_verifier.c
@@ -31,6 +31,7 @@
 #include <linux/if_ether.h>
 #include <linux/btf.h>
 
+#include <bpf/btf.h>
 #include <bpf/bpf.h>
 #include <bpf/libbpf.h>
 
@@ -66,6 +67,11 @@ static bool unpriv_disabled = false;
 static int skips;
 static bool verbose = false;
 
+struct kfunc_btf_id_pair {
+	const char *kfunc;
+	int insn_idx;
+};
+
 struct bpf_test {
 	const char *descr;
 	struct bpf_insn	insns[MAX_INSNS];
@@ -92,6 +98,7 @@ struct bpf_test {
 	int fixup_map_reuseport_array[MAX_FIXUPS];
 	int fixup_map_ringbuf[MAX_FIXUPS];
 	int fixup_map_timer[MAX_FIXUPS];
+	struct kfunc_btf_id_pair fixup_kfunc_btf_id[MAX_FIXUPS];
 	/* Expected verifier log output for result REJECT or VERBOSE_ACCEPT.
 	 * Can be a tab-separated sequence of expected strings. An empty string
 	 * means no log verification.
@@ -744,6 +751,7 @@ static void do_test_fixup(struct bpf_test *test, enum bpf_prog_type prog_type,
 	int *fixup_map_reuseport_array = test->fixup_map_reuseport_array;
 	int *fixup_map_ringbuf = test->fixup_map_ringbuf;
 	int *fixup_map_timer = test->fixup_map_timer;
+	struct kfunc_btf_id_pair *fixup_kfunc_btf_id = test->fixup_kfunc_btf_id;
 
 	if (test->fill_helper) {
 		test->fill_insns = calloc(MAX_TEST_INSNS, sizeof(struct bpf_insn));
@@ -936,6 +944,26 @@ static void do_test_fixup(struct bpf_test *test, enum bpf_prog_type prog_type,
 			fixup_map_timer++;
 		} while (*fixup_map_timer);
 	}
+
+	/* Patch in kfunc BTF IDs */
+	if (fixup_kfunc_btf_id->kfunc) {
+		struct btf *btf;
+		int btf_id;
+
+		do {
+			btf_id = 0;
+			btf = btf__load_vmlinux_btf();
+			if (btf) {
+				btf_id = btf__find_by_name_kind(btf,
+								fixup_kfunc_btf_id->kfunc,
+								BTF_KIND_FUNC);
+				btf_id = btf_id < 0 ? 0 : btf_id;
+			}
+			btf__free(btf);
+			prog[fixup_kfunc_btf_id->insn_idx].imm = btf_id;
+			fixup_kfunc_btf_id++;
+		} while (fixup_kfunc_btf_id->kfunc);
+	}
 }
 
 struct libcap {
diff --git a/tools/testing/selftests/bpf/verifier/calls.c b/tools/testing/selftests/bpf/verifier/calls.c
index d7b74eb28333..829be2b9e08e 100644
--- a/tools/testing/selftests/bpf/verifier/calls.c
+++ b/tools/testing/selftests/bpf/verifier/calls.c
@@ -21,6 +21,81 @@
 	.prog_type = BPF_PROG_TYPE_TRACEPOINT,
 	.result  = ACCEPT,
 },
+{
+	"calls: invalid kfunc call: ptr_to_mem to struct with non-scalar",
+	.insns = {
+	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
+	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8),
+	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
+	BPF_EXIT_INSN(),
+	},
+	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
+	.result = REJECT,
+	.errstr = "arg#0 pointer type STRUCT prog_test_fail1 must point to scalar",
+	.fixup_kfunc_btf_id = {
+		{ "bpf_kfunc_call_test_fail1", 2 },
+	},
+},
+{
+	"calls: invalid kfunc call: ptr_to_mem to struct with nesting depth > 4",
+	.insns = {
+	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
+	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8),
+	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
+	BPF_EXIT_INSN(),
+	},
+	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
+	.result = REJECT,
+	.errstr = "max struct nesting depth exceeded\narg#0 pointer type STRUCT prog_test_fail2",
+	.fixup_kfunc_btf_id = {
+		{ "bpf_kfunc_call_test_fail2", 2 },
+	},
+},
+{
+	"calls: invalid kfunc call: ptr_to_mem to struct with FAM",
+	.insns = {
+	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
+	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8),
+	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
+	BPF_EXIT_INSN(),
+	},
+	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
+	.result = REJECT,
+	.errstr = "arg#0 pointer type STRUCT prog_test_fail3 must point to scalar",
+	.fixup_kfunc_btf_id = {
+		{ "bpf_kfunc_call_test_fail3", 2 },
+	},
+},
+{
+	"calls: invalid kfunc call: reg->type != PTR_TO_CTX",
+	.insns = {
+	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
+	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8),
+	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
+	BPF_EXIT_INSN(),
+	},
+	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
+	.result = REJECT,
+	.errstr = "arg#0 expected pointer to ctx, but got PTR",
+	.fixup_kfunc_btf_id = {
+		{ "bpf_kfunc_call_test_pass_ctx", 2 },
+	},
+},
+{
+	"calls: invalid kfunc call: void * not allowed in func proto without mem size arg",
+	.insns = {
+	BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),
+	BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8),
+	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
+	BPF_EXIT_INSN(),
+	},
+	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
+	.result = REJECT,
+	.errstr = "arg#0 pointer type UNKNOWN  must point to scalar",
+	.fixup_kfunc_btf_id = {
+		{ "bpf_kfunc_call_test_mem_len_fail1", 2 },
+	},
+},
 {
 	"calls: basic sanity",
 	.insns = {