libbpf: Support kernel module function calls

This patch adds libbpf support for kernel module function call support.
The fd_array parameter is used during BPF program load to pass module
BTFs referenced by the program. insn->off is set to index into this
array, but starts from 1, because insn->off as 0 is reserved for
btf_vmlinux.

We try to use existing insn->off for a module, since the kernel limits
the maximum distinct module BTFs for kfuncs to 256, and also because
index must never exceed the maximum allowed value that can fit in
insn->off (INT16_MAX). In the future, if kernel interprets signed offset
as unsigned for kfunc calls, this limit can be increased to UINT16_MAX.

Also introduce a btf__find_by_name_kind_own helper to start searching
from module BTF's start id when we know that the BTF ID is not present
in vmlinux BTF (in find_ksym_btf_id).
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211002011757.311265-7-memxor@gmail.com

tools/lib/bpf/bpf.c

@@ -264,6 +264,7 @@ int libbpf__bpf_prog_load(const struct bpf_prog_load_params *load_attr)
 	attr.line_info_rec_size = load_attr->line_info_rec_size;
 	attr.line_info_cnt = load_attr->line_info_cnt;
 	attr.line_info = ptr_to_u64(load_attr->line_info);
+	attr.fd_array = ptr_to_u64(load_attr->fd_array);
 
 	if (load_attr->name)
 		memcpy(attr.prog_name, load_attr->name,

tools/lib/bpf/btf.c

@@ -695,15 +695,15 @@ __s32 btf__find_by_name(const struct btf *btf, const char *type_name)
 	return libbpf_err(-ENOENT);
 }
 
-__s32 btf__find_by_name_kind(const struct btf *btf, const char *type_name,
-			     __u32 kind)
+static __s32 btf_find_by_name_kind(const struct btf *btf, int start_id,
+				   const char *type_name, __u32 kind)
 {
 	__u32 i, nr_types = btf__get_nr_types(btf);
 
 	if (kind == BTF_KIND_UNKN || !strcmp(type_name, "void"))
 		return 0;
 
-	for (i = 1; i <= nr_types; i++) {
+	for (i = start_id; i <= nr_types; i++) {
 		const struct btf_type *t = btf__type_by_id(btf, i);
 		const char *name;
 
@@ -717,6 +717,18 @@ __s32 btf__find_by_name_kind(const struct btf *btf, const char *type_name,
 	return libbpf_err(-ENOENT);
 }
 
+__s32 btf__find_by_name_kind_own(const struct btf *btf, const char *type_name,
+				 __u32 kind)
+{
+	return btf_find_by_name_kind(btf, btf->start_id, type_name, kind);
+}
+
+__s32 btf__find_by_name_kind(const struct btf *btf, const char *type_name,
+			     __u32 kind)
+{
+	return btf_find_by_name_kind(btf, 1, type_name, kind);
+}
+
 static bool btf_is_modifiable(const struct btf *btf)
 {
 	return (void *)btf->hdr != btf->raw_data;

tools/lib/bpf/libbpf.c

@@ -443,6 +443,11 @@ struct extern_desc {
 
 			/* local btf_id of the ksym extern's type. */
 			__u32 type_id;
+			/* BTF fd index to be patched in for insn->off, this is
+			 * 0 for vmlinux BTF, index in obj->fd_array for module
+			 * BTF
+			 */
+			__s16 btf_fd_idx;
 		} ksym;
 	};
 };
@@ -454,6 +459,7 @@ struct module_btf {
 	char *name;
 	__u32 id;
 	int fd;
+	int fd_array_idx;
 };
 
 struct bpf_object {
@@ -539,6 +545,10 @@ struct bpf_object {
 	void *priv;
 	bpf_object_clear_priv_t clear_priv;
 
+	int *fd_array;
+	size_t fd_array_cap;
+	size_t fd_array_cnt;
+
 	char path[];
 };
 #define obj_elf_valid(o)	((o)->efile.elf)
@@ -5407,6 +5417,7 @@ bpf_object__relocate_data(struct bpf_object *obj, struct bpf_program *prog)
 			ext = &obj->externs[relo->sym_off];
 			insn[0].src_reg = BPF_PSEUDO_KFUNC_CALL;
 			insn[0].imm = ext->ksym.kernel_btf_id;
+			insn[0].off = ext->ksym.btf_fd_idx;
 			break;
 		case RELO_SUBPROG_ADDR:
 			if (insn[0].src_reg != BPF_PSEUDO_FUNC) {
@@ -6236,6 +6247,7 @@ load_program(struct bpf_program *prog, struct bpf_insn *insns, int insns_cnt,
 	}
 	load_attr.log_level = prog->log_level;
 	load_attr.prog_flags = prog->prog_flags;
+	load_attr.fd_array = prog->obj->fd_array;
 
 	/* adjust load_attr if sec_def provides custom preload callback */
 	if (prog->sec_def && prog->sec_def->preload_fn) {
@@ -6752,13 +6764,14 @@ static int bpf_object__read_kallsyms_file(struct bpf_object *obj)
 
 static int find_ksym_btf_id(struct bpf_object *obj, const char *ksym_name,
 			    __u16 kind, struct btf **res_btf,
-			    int *res_btf_fd)
+			    struct module_btf **res_mod_btf)
 {
-	int i, id, btf_fd, err;
+	struct module_btf *mod_btf;
 	struct btf *btf;
+	int i, id, err;
 
 	btf = obj->btf_vmlinux;
-	btf_fd = 0;
+	mod_btf = NULL;
 	id = btf__find_by_name_kind(btf, ksym_name, kind);
 
 	if (id == -ENOENT) {
@@ -6767,10 +6780,10 @@ static int find_ksym_btf_id(struct bpf_object *obj, const char *ksym_name,
 			return err;
 
 		for (i = 0; i < obj->btf_module_cnt; i++) {
-			btf = obj->btf_modules[i].btf;
-			/* we assume module BTF FD is always >0 */
-			btf_fd = obj->btf_modules[i].fd;
-			id = btf__find_by_name_kind(btf, ksym_name, kind);
+			/* we assume module_btf's BTF FD is always >0 */
+			mod_btf = &obj->btf_modules[i];
+			btf = mod_btf->btf;
+			id = btf__find_by_name_kind_own(btf, ksym_name, kind);
 			if (id != -ENOENT)
 				break;
 		}
@@ -6779,7 +6792,7 @@ static int find_ksym_btf_id(struct bpf_object *obj, const char *ksym_name,
 		return -ESRCH;
 
 	*res_btf = btf;
-	*res_btf_fd = btf_fd;
+	*res_mod_btf = mod_btf;
 	return id;
 }
 
@@ -6788,11 +6801,12 @@ static int bpf_object__resolve_ksym_var_btf_id(struct bpf_object *obj,
 {
 	const struct btf_type *targ_var, *targ_type;
 	__u32 targ_type_id, local_type_id;
+	struct module_btf *mod_btf = NULL;
 	const char *targ_var_name;
-	int id, btf_fd = 0, err;
 	struct btf *btf = NULL;
+	int id, err;
 
-	id = find_ksym_btf_id(obj, ext->name, BTF_KIND_VAR, &btf, &btf_fd);
+	id = find_ksym_btf_id(obj, ext->name, BTF_KIND_VAR, &btf, &mod_btf);
 	if (id == -ESRCH && ext->is_weak) {
 		return 0;
 	} else if (id < 0) {
@@ -6827,7 +6841,7 @@ static int bpf_object__resolve_ksym_var_btf_id(struct bpf_object *obj,
 	}
 
 	ext->is_set = true;
-	ext->ksym.kernel_btf_obj_fd = btf_fd;
+	ext->ksym.kernel_btf_obj_fd = mod_btf ? mod_btf->fd : 0;
 	ext->ksym.kernel_btf_id = id;
 	pr_debug("extern (var ksym) '%s': resolved to [%d] %s %s\n",
 		 ext->name, id, btf_kind_str(targ_var), targ_var_name);
@@ -6839,26 +6853,20 @@ static int bpf_object__resolve_ksym_func_btf_id(struct bpf_object *obj,
 						struct extern_desc *ext)
 {
 	int local_func_proto_id, kfunc_proto_id, kfunc_id;
+	struct module_btf *mod_btf = NULL;
 	const struct btf_type *kern_func;
 	struct btf *kern_btf = NULL;
-	int ret, kern_btf_fd = 0;
+	int ret;
 
 	local_func_proto_id = ext->ksym.type_id;
 
-	kfunc_id = find_ksym_btf_id(obj, ext->name, BTF_KIND_FUNC,
-				    &kern_btf, &kern_btf_fd);
+	kfunc_id = find_ksym_btf_id(obj, ext->name, BTF_KIND_FUNC, &kern_btf, &mod_btf);
 	if (kfunc_id < 0) {
 		pr_warn("extern (func ksym) '%s': not found in kernel BTF\n",
 			ext->name);
 		return kfunc_id;
 	}
 
-	if (kern_btf != obj->btf_vmlinux) {
-		pr_warn("extern (func ksym) '%s': function in kernel module is not supported\n",
-			ext->name);
-		return -ENOTSUP;
-	}
-
 	kern_func = btf__type_by_id(kern_btf, kfunc_id);
 	kfunc_proto_id = kern_func->type;
 
@@ -6870,9 +6878,30 @@ static int bpf_object__resolve_ksym_func_btf_id(struct bpf_object *obj,
 		return -EINVAL;
 	}
 
+	/* set index for module BTF fd in fd_array, if unset */
+	if (mod_btf && !mod_btf->fd_array_idx) {
+		/* insn->off is s16 */
+		if (obj->fd_array_cnt == INT16_MAX) {
+			pr_warn("extern (func ksym) '%s': module BTF fd index %d too big to fit in bpf_insn offset\n",
+				ext->name, mod_btf->fd_array_idx);
+			return -E2BIG;
+		}
+		/* Cannot use index 0 for module BTF fd */
+		if (!obj->fd_array_cnt)
+			obj->fd_array_cnt = 1;
+
+		ret = libbpf_ensure_mem((void **)&obj->fd_array, &obj->fd_array_cap, sizeof(int),
+					obj->fd_array_cnt + 1);
+		if (ret)
+			return ret;
+		mod_btf->fd_array_idx = obj->fd_array_cnt;
+		/* we assume module BTF FD is always >0 */
+		obj->fd_array[obj->fd_array_cnt++] = mod_btf->fd;
+	}
+
 	ext->is_set = true;
-	ext->ksym.kernel_btf_obj_fd = kern_btf_fd;
 	ext->ksym.kernel_btf_id = kfunc_id;
+	ext->ksym.btf_fd_idx = mod_btf ? mod_btf->fd_array_idx : 0;
 	pr_debug("extern (func ksym) '%s': resolved to kernel [%d]\n",
 		 ext->name, kfunc_id);
 
@@ -7031,6 +7060,9 @@ int bpf_object__load_xattr(struct bpf_object_load_attr *attr)
 			err = bpf_gen__finish(obj->gen_loader);
 	}
 
+	/* clean up fd_array */
+	zfree(&obj->fd_array);
+
 	/* clean up module BTFs */
 	for (i = 0; i < obj->btf_module_cnt; i++) {
 		close(obj->btf_modules[i].fd);

tools/lib/bpf/libbpf_internal.h

@@ -298,6 +298,7 @@ struct bpf_prog_load_params {
 	__u32 log_level;
 	char *log_buf;
 	size_t log_buf_sz;
+	int *fd_array;
 };
 
 int libbpf__bpf_prog_load(const struct bpf_prog_load_params *load_attr);
@@ -408,6 +409,8 @@ int btf_type_visit_type_ids(struct btf_type *t, type_id_visit_fn visit, void *ct
 int btf_type_visit_str_offs(struct btf_type *t, str_off_visit_fn visit, void *ctx);
 int btf_ext_visit_type_ids(struct btf_ext *btf_ext, type_id_visit_fn visit, void *ctx);
 int btf_ext_visit_str_offs(struct btf_ext *btf_ext, str_off_visit_fn visit, void *ctx);
+__s32 btf__find_by_name_kind_own(const struct btf *btf, const char *type_name,
+				 __u32 kind);
 
 extern enum libbpf_strict_mode libbpf_mode;