tools/lib/bpf/libbpf.c

// SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause)

/*
 * Common eBPF ELF object loading operations.
 *
 * Copyright (C) 2013-2015 Alexei Starovoitov <ast@kernel.org>
 * Copyright (C) 2015 Wang Nan <wangnan0@huawei.com>
 * Copyright (C) 2015 Huawei Inc.
 * Copyright (C) 2017 Nicira, Inc.
 * Copyright (C) 2019 Isovalent, Inc.
 */

#ifndef _GNU_SOURCE
#define _GNU_SOURCE
#endif
#include <stdlib.h>
#include <stdio.h>
#include <stdarg.h>
#include <libgen.h>
#include <inttypes.h>
#include <limits.h>
#include <string.h>
#include <unistd.h>
#include <endian.h>
#include <fcntl.h>
#include <errno.h>
#include <ctype.h>
#include <asm/unistd.h>
#include <linux/err.h>
#include <linux/kernel.h>
#include <linux/bpf.h>
#include <linux/btf.h>
#include <linux/filter.h>
#include <linux/list.h>
#include <linux/limits.h>
#include <linux/perf_event.h>
#include <linux/ring_buffer.h>
#include <linux/version.h>
#include <sys/epoll.h>
#include <sys/ioctl.h>
#include <sys/mman.h>
#include <sys/stat.h>
#include <sys/types.h>
#include <sys/vfs.h>
#include <sys/utsname.h>
#include <sys/resource.h>
#include <libelf.h>
#include <gelf.h>
#include <zlib.h>

#include "libbpf.h"
#include "bpf.h"
#include "btf.h"
#include "str_error.h"
#include "libbpf_internal.h"
#include "hashmap.h"
#include "bpf_gen_internal.h"

#ifndef BPF_FS_MAGIC
#define BPF_FS_MAGIC		0xcafe4a11
#endif

#define BPF_INSN_SZ (sizeof(struct bpf_insn))

/* vsprintf() in __base_pr() uses nonliteral format string. It may break
 * compilation if user enables corresponding warning. Disable it explicitly.
 */
#pragma GCC diagnostic ignored "-Wformat-nonliteral"

#define __printf(a, b)	__attribute__((format(printf, a, b)))

static struct bpf_map *bpf_object__add_map(struct bpf_object *obj);
static bool prog_is_subprog(const struct bpf_object *obj, const struct bpf_program *prog);

static int __base_pr(enum libbpf_print_level level, const char *format,
		     va_list args)
{
	if (level == LIBBPF_DEBUG)
		return 0;

	return vfprintf(stderr, format, args);
}

static libbpf_print_fn_t __libbpf_pr = __base_pr;

libbpf_print_fn_t libbpf_set_print(libbpf_print_fn_t fn)
{
	libbpf_print_fn_t old_print_fn = __libbpf_pr;

	__libbpf_pr = fn;
	return old_print_fn;
}

__printf(2, 3)
void libbpf_print(enum libbpf_print_level level, const char *format, ...)
{
	va_list args;

	if (!__libbpf_pr)
		return;

	va_start(args, format);
	__libbpf_pr(level, format, args);
	va_end(args);
}

static void pr_perm_msg(int err)
{
	struct rlimit limit;
	char buf[100];

	if (err != -EPERM || geteuid() != 0)
		return;

	err = getrlimit(RLIMIT_MEMLOCK, &limit);
	if (err)
		return;

	if (limit.rlim_cur == RLIM_INFINITY)
		return;

	if (limit.rlim_cur < 1024)
		snprintf(buf, sizeof(buf), "%zu bytes", (size_t)limit.rlim_cur);
	else if (limit.rlim_cur < 1024*1024)
		snprintf(buf, sizeof(buf), "%.1f KiB", (double)limit.rlim_cur / 1024);
	else
		snprintf(buf, sizeof(buf), "%.1f MiB", (double)limit.rlim_cur / (1024*1024));

	pr_warn("permission error while running as root; try raising 'ulimit -l'? current value: %s\n",
		buf);
}

#define STRERR_BUFSIZE  128

/* Copied from tools/perf/util/util.h */
#ifndef zfree
# define zfree(ptr) ({ free(*ptr); *ptr = NULL; })
#endif

#ifndef zclose
# define zclose(fd) ({			\
	int ___err = 0;			\
	if ((fd) >= 0)			\
		___err = close((fd));	\
	fd = -1;			\
	___err; })
#endif

static inline __u64 ptr_to_u64(const void *ptr)
{
	return (__u64) (unsigned long) ptr;
}

/* this goes away in libbpf 1.0 */
enum libbpf_strict_mode libbpf_mode = LIBBPF_STRICT_NONE;

int libbpf_set_strict_mode(enum libbpf_strict_mode mode)
{
	/* __LIBBPF_STRICT_LAST is the last power-of-2 value used + 1, so to
	 * get all possible values we compensate last +1, and then (2*x - 1)
	 * to get the bit mask
	 */
	if (mode != LIBBPF_STRICT_ALL
	    && (mode & ~((__LIBBPF_STRICT_LAST - 1) * 2 - 1)))
		return errno = EINVAL, -EINVAL;

	libbpf_mode = mode;
	return 0;
}

enum kern_feature_id {
	/* v4.14: kernel support for program & map names. */
	FEAT_PROG_NAME,
	/* v5.2: kernel support for global data sections. */
	FEAT_GLOBAL_DATA,
	/* BTF support */
	FEAT_BTF,
	/* BTF_KIND_FUNC and BTF_KIND_FUNC_PROTO support */
	FEAT_BTF_FUNC,
	/* BTF_KIND_VAR and BTF_KIND_DATASEC support */
	FEAT_BTF_DATASEC,
	/* BTF_FUNC_GLOBAL is supported */
	FEAT_BTF_GLOBAL_FUNC,
	/* BPF_F_MMAPABLE is supported for arrays */
	FEAT_ARRAY_MMAP,
	/* kernel support for expected_attach_type in BPF_PROG_LOAD */
	FEAT_EXP_ATTACH_TYPE,
	/* bpf_probe_read_{kernel,user}[_str] helpers */
	FEAT_PROBE_READ_KERN,
	/* BPF_PROG_BIND_MAP is supported */
	FEAT_PROG_BIND_MAP,
	/* Kernel support for module BTFs */
	FEAT_MODULE_BTF,
	/* BTF_KIND_FLOAT support */
	FEAT_BTF_FLOAT,
	__FEAT_CNT,
};

static bool kernel_supports(const struct bpf_object *obj, enum kern_feature_id feat_id);

enum reloc_type {
	RELO_LD64,
	RELO_CALL,
	RELO_DATA,
	RELO_EXTERN_VAR,
	RELO_EXTERN_FUNC,
	RELO_SUBPROG_ADDR,
};

struct reloc_desc {
	enum reloc_type type;
	int insn_idx;
	int map_idx;
	int sym_off;
};

struct bpf_sec_def;

typedef struct bpf_link *(*attach_fn_t)(const struct bpf_sec_def *sec,
					struct bpf_program *prog);

struct bpf_sec_def {
	const char *sec;
	size_t len;
	enum bpf_prog_type prog_type;
	enum bpf_attach_type expected_attach_type;
	bool is_exp_attach_type_optional;
	bool is_attachable;
	bool is_attach_btf;
	bool is_sleepable;
	attach_fn_t attach_fn;
};

/*
 * bpf_prog should be a better name but it has been used in
 * linux/filter.h.
 */
struct bpf_program {
	const struct bpf_sec_def *sec_def;
	char *sec_name;
	size_t sec_idx;
	/* this program's instruction offset (in number of instructions)
	 * within its containing ELF section
	 */
	size_t sec_insn_off;
	/* number of original instructions in ELF section belonging to this
	 * program, not taking into account subprogram instructions possible
	 * appended later during relocation
	 */
	size_t sec_insn_cnt;
	/* Offset (in number of instructions) of the start of instruction
	 * belonging to this BPF program  within its containing main BPF
	 * program. For the entry-point (main) BPF program, this is always
	 * zero. For a sub-program, this gets reset before each of main BPF
	 * programs are processed and relocated and is used to determined
	 * whether sub-program was already appended to the main program, and
	 * if yes, at which instruction offset.
	 */
	size_t sub_insn_off;

	char *name;
	/* sec_name with / replaced by _; makes recursive pinning
	 * in bpf_object__pin_programs easier
	 */
	char *pin_name;

	/* instructions that belong to BPF program; insns[0] is located at
	 * sec_insn_off instruction within its ELF section in ELF file, so
	 * when mapping ELF file instruction index to the local instruction,
	 * one needs to subtract sec_insn_off; and vice versa.
	 */
	struct bpf_insn *insns;
	/* actual number of instruction in this BPF program's image; for
	 * entry-point BPF programs this includes the size of main program
	 * itself plus all the used sub-programs, appended at the end
	 */
	size_t insns_cnt;

	struct reloc_desc *reloc_desc;
	int nr_reloc;
	int log_level;

	struct {
		int nr;
		int *fds;
	} instances;
	bpf_program_prep_t preprocessor;

	struct bpf_object *obj;
	void *priv;
	bpf_program_clear_priv_t clear_priv;

	bool load;
	bool mark_btf_static;
	enum bpf_prog_type type;
	enum bpf_attach_type expected_attach_type;
	int prog_ifindex;
	__u32 attach_btf_obj_fd;
	__u32 attach_btf_id;
	__u32 attach_prog_fd;
	void *func_info;
	__u32 func_info_rec_size;
	__u32 func_info_cnt;

	void *line_info;
	__u32 line_info_rec_size;
	__u32 line_info_cnt;
	__u32 prog_flags;
};

struct bpf_struct_ops {
	const char *tname;
	const struct btf_type *type;
	struct bpf_program **progs;
	__u32 *kern_func_off;
	/* e.g. struct tcp_congestion_ops in bpf_prog's btf format */
	void *data;
	/* e.g. struct bpf_struct_ops_tcp_congestion_ops in
	 *      btf_vmlinux's format.
	 * struct bpf_struct_ops_tcp_congestion_ops {
	 *	[... some other kernel fields ...]
	 *	struct tcp_congestion_ops data;
	 * }
	 * kern_vdata-size == sizeof(struct bpf_struct_ops_tcp_congestion_ops)
	 * bpf_map__init_kern_struct_ops() will populate the "kern_vdata"
	 * from "data".
	 */
	void *kern_vdata;
	__u32 type_id;
};

#define DATA_SEC ".data"
#define BSS_SEC ".bss"
#define RODATA_SEC ".rodata"
#define KCONFIG_SEC ".kconfig"
#define KSYMS_SEC ".ksyms"
#define STRUCT_OPS_SEC ".struct_ops"

enum libbpf_map_type {
	LIBBPF_MAP_UNSPEC,
	LIBBPF_MAP_DATA,
	LIBBPF_MAP_BSS,
	LIBBPF_MAP_RODATA,
	LIBBPF_MAP_KCONFIG,
};

static const char * const libbpf_type_to_btf_name[] = {
	[LIBBPF_MAP_DATA]	= DATA_SEC,
	[LIBBPF_MAP_BSS]	= BSS_SEC,
	[LIBBPF_MAP_RODATA]	= RODATA_SEC,
	[LIBBPF_MAP_KCONFIG]	= KCONFIG_SEC,
};

struct bpf_map {
	char *name;
	int fd;
	int sec_idx;
	size_t sec_offset;
	int map_ifindex;
	int inner_map_fd;
	struct bpf_map_def def;
	__u32 numa_node;
	__u32 btf_var_idx;
	__u32 btf_key_type_id;
	__u32 btf_value_type_id;
	__u32 btf_vmlinux_value_type_id;
	void *priv;
	bpf_map_clear_priv_t clear_priv;
	enum libbpf_map_type libbpf_type;
	void *mmaped;
	struct bpf_struct_ops *st_ops;
	struct bpf_map *inner_map;
	void **init_slots;
	int init_slots_sz;
	char *pin_path;
	bool pinned;
	bool reused;
};

enum extern_type {
	EXT_UNKNOWN,
	EXT_KCFG,
	EXT_KSYM,
};

enum kcfg_type {
	KCFG_UNKNOWN,
	KCFG_CHAR,
	KCFG_BOOL,
	KCFG_INT,
	KCFG_TRISTATE,
	KCFG_CHAR_ARR,
};

struct extern_desc {
	enum extern_type type;
	int sym_idx;
	int btf_id;
	int sec_btf_id;
	const char *name;
	bool is_set;
	bool is_weak;
	union {
		struct {
			enum kcfg_type type;
			int sz;
			int align;
			int data_off;
			bool is_signed;
		} kcfg;
		struct {
			unsigned long long addr;

			/* target btf_id of the corresponding kernel var. */
			int kernel_btf_obj_fd;
			int kernel_btf_id;

			/* local btf_id of the ksym extern's type. */
			__u32 type_id;
		} ksym;
	};
};

static LIST_HEAD(bpf_objects_list);

struct module_btf {
	struct btf *btf;
	char *name;
	__u32 id;
	int fd;
};

struct bpf_object {
	char name[BPF_OBJ_NAME_LEN];
	char license[64];
	__u32 kern_version;

	struct bpf_program *programs;
	size_t nr_programs;
	struct bpf_map *maps;
	size_t nr_maps;
	size_t maps_cap;

	char *kconfig;
	struct extern_desc *externs;
	int nr_extern;
	int kconfig_map_idx;
	int rodata_map_idx;

	bool loaded;
	bool has_subcalls;

	struct bpf_gen *gen_loader;

	/*
	 * Information when doing elf related work. Only valid if fd
	 * is valid.
	 */
	struct {
		int fd;
		const void *obj_buf;
		size_t obj_buf_sz;
		Elf *elf;
		GElf_Ehdr ehdr;
		Elf_Data *symbols;
		Elf_Data *data;
		Elf_Data *rodata;
		Elf_Data *bss;
		Elf_Data *st_ops_data;
		size_t shstrndx; /* section index for section name strings */
		size_t strtabidx;
		struct {
			GElf_Shdr shdr;
			Elf_Data *data;
		} *reloc_sects;
		int nr_reloc_sects;
		int maps_shndx;
		int btf_maps_shndx;
		__u32 btf_maps_sec_btf_id;
		int text_shndx;
		int symbols_shndx;
		int data_shndx;
		int rodata_shndx;
		int bss_shndx;
		int st_ops_shndx;
	} efile;
	/*
	 * All loaded bpf_object is linked in a list, which is
	 * hidden to caller. bpf_objects__<func> handlers deal with
	 * all objects.
	 */
	struct list_head list;

	struct btf *btf;
	struct btf_ext *btf_ext;

	/* Parse and load BTF vmlinux if any of the programs in the object need
	 * it at load time.
	 */
	struct btf *btf_vmlinux;
	/* vmlinux BTF override for CO-RE relocations */
	struct btf *btf_vmlinux_override;
	/* Lazily initialized kernel module BTFs */
	struct module_btf *btf_modules;
	bool btf_modules_loaded;
	size_t btf_module_cnt;
	size_t btf_module_cap;

	void *priv;
	bpf_object_clear_priv_t clear_priv;

	char path[];
};
#define obj_elf_valid(o)	((o)->efile.elf)

static const char *elf_sym_str(const struct bpf_object *obj, size_t off);
static const char *elf_sec_str(const struct bpf_object *obj, size_t off);
static Elf_Scn *elf_sec_by_idx(const struct bpf_object *obj, size_t idx);
static Elf_Scn *elf_sec_by_name(const struct bpf_object *obj, const char *name);
static int elf_sec_hdr(const struct bpf_object *obj, Elf_Scn *scn, GElf_Shdr *hdr);
static const char *elf_sec_name(const struct bpf_object *obj, Elf_Scn *scn);
static Elf_Data *elf_sec_data(const struct bpf_object *obj, Elf_Scn *scn);

void bpf_program__unload(struct bpf_program *prog)
{
	int i;

	if (!prog)
		return;

	/*
	 * If the object is opened but the program was never loaded,
	 * it is possible that prog->instances.nr == -1.
	 */
	if (prog->instances.nr > 0) {
		for (i = 0; i < prog->instances.nr; i++)
			zclose(prog->instances.fds[i]);
	} else if (prog->instances.nr != -1) {
		pr_warn("Internal error: instances.nr is %d\n",
			prog->instances.nr);
	}

	prog->instances.nr = -1;
	zfree(&prog->instances.fds);

	zfree(&prog->func_info);
	zfree(&prog->line_info);
}

static void bpf_program__exit(struct bpf_program *prog)
{
	if (!prog)
		return;

	if (prog->clear_priv)
		prog->clear_priv(prog, prog->priv);

	prog->priv = NULL;
	prog->clear_priv = NULL;

	bpf_program__unload(prog);
	zfree(&prog->name);
	zfree(&prog->sec_name);
	zfree(&prog->pin_name);
	zfree(&prog->insns);
	zfree(&prog->reloc_desc);

	prog->nr_reloc = 0;
	prog->insns_cnt = 0;
	prog->sec_idx = -1;
}

static char *__bpf_program__pin_name(struct bpf_program *prog)
{
	char *name, *p;

	name = p = strdup(prog->sec_name);
	while ((p = strchr(p, '/')))
		*p = '_';

	return name;
}

static bool insn_is_subprog_call(const struct bpf_insn *insn)
{
	return BPF_CLASS(insn->code) == BPF_JMP &&
	       BPF_OP(insn->code) == BPF_CALL &&
	       BPF_SRC(insn->code) == BPF_K &&
	       insn->src_reg == BPF_PSEUDO_CALL &&
	       insn->dst_reg == 0 &&
	       insn->off == 0;
}

static bool is_ldimm64_insn(struct bpf_insn *insn)
{
	return insn->code == (BPF_LD | BPF_IMM | BPF_DW);
}

static bool is_call_insn(const struct bpf_insn *insn)
{
	return insn->code == (BPF_JMP | BPF_CALL);
}

static bool insn_is_pseudo_func(struct bpf_insn *insn)
{
	return is_ldimm64_insn(insn) && insn->src_reg == BPF_PSEUDO_FUNC;
}

static int
bpf_object__init_prog(struct bpf_object *obj, struct bpf_program *prog,
		      const char *name, size_t sec_idx, const char *sec_name,
		      size_t sec_off, void *insn_data, size_t insn_data_sz)
{
	if (insn_data_sz == 0 || insn_data_sz % BPF_INSN_SZ || sec_off % BPF_INSN_SZ) {
		pr_warn("sec '%s': corrupted program '%s', offset %zu, size %zu\n",
			sec_name, name, sec_off, insn_data_sz);
		return -EINVAL;
	}

	memset(prog, 0, sizeof(*prog));
	prog->obj = obj;

	prog->sec_idx = sec_idx;
	prog->sec_insn_off = sec_off / BPF_INSN_SZ;
	prog->sec_insn_cnt = insn_data_sz / BPF_INSN_SZ;
	/* insns_cnt can later be increased by appending used subprograms */
	prog->insns_cnt = prog->sec_insn_cnt;

	prog->type = BPF_PROG_TYPE_UNSPEC;
	prog->load = true;

	prog->instances.fds = NULL;
	prog->instances.nr = -1;

	prog->sec_name = strdup(sec_name);
	if (!prog->sec_name)
		goto errout;

	prog->name = strdup(name);
	if (!prog->name)
		goto errout;

	prog->pin_name = __bpf_program__pin_name(prog);
	if (!prog->pin_name)
		goto errout;

	prog->insns = malloc(insn_data_sz);
	if (!prog->insns)
		goto errout;
	memcpy(prog->insns, insn_data, insn_data_sz);

	return 0;
errout:
	pr_warn("sec '%s': failed to allocate memory for prog '%s'\n", sec_name, name);
	bpf_program__exit(prog);
	return -ENOMEM;
}

static int
bpf_object__add_programs(struct bpf_object *obj, Elf_Data *sec_data,
			 const char *sec_name, int sec_idx)
{
	Elf_Data *symbols = obj->efile.symbols;
	struct bpf_program *prog, *progs;
	void *data = sec_data->d_buf;
	size_t sec_sz = sec_data->d_size, sec_off, prog_sz, nr_syms;
	int nr_progs, err, i;
	const char *name;
	GElf_Sym sym;

	progs = obj->programs;
	nr_progs = obj->nr_programs;
	nr_syms = symbols->d_size / sizeof(GElf_Sym);
	sec_off = 0;

	for (i = 0; i < nr_syms; i++) {
		if (!gelf_getsym(symbols, i, &sym))
			continue;
		if (sym.st_shndx != sec_idx)
			continue;
		if (GELF_ST_TYPE(sym.st_info) != STT_FUNC)
			continue;

		prog_sz = sym.st_size;
		sec_off = sym.st_value;

		name = elf_sym_str(obj, sym.st_name);
		if (!name) {
			pr_warn("sec '%s': failed to get symbol name for offset %zu\n",
				sec_name, sec_off);
			return -LIBBPF_ERRNO__FORMAT;
		}

		if (sec_off + prog_sz > sec_sz) {
			pr_warn("sec '%s': program at offset %zu crosses section boundary\n",
				sec_name, sec_off);
			return -LIBBPF_ERRNO__FORMAT;
		}

		if (sec_idx != obj->efile.text_shndx && GELF_ST_BIND(sym.st_info) == STB_LOCAL) {
			pr_warn("sec '%s': program '%s' is static and not supported\n", sec_name, name);
			return -ENOTSUP;
		}

		pr_debug("sec '%s': found program '%s' at insn offset %zu (%zu bytes), code size %zu insns (%zu bytes)\n",
			 sec_name, name, sec_off / BPF_INSN_SZ, sec_off, prog_sz / BPF_INSN_SZ, prog_sz);

		progs = libbpf_reallocarray(progs, nr_progs + 1, sizeof(*progs));
		if (!progs) {
			/*
			 * In this case the original obj->programs
			 * is still valid, so don't need special treat for
			 * bpf_close_object().
			 */
			pr_warn("sec '%s': failed to alloc memory for new program '%s'\n",
				sec_name, name);
			return -ENOMEM;
		}
		obj->programs = progs;

		prog = &progs[nr_progs];

		err = bpf_object__init_prog(obj, prog, name, sec_idx, sec_name,
					    sec_off, data + sec_off, prog_sz);
		if (err)
			return err;

		/* if function is a global/weak symbol, but has restricted
		 * (STV_HIDDEN or STV_INTERNAL) visibility, mark its BTF FUNC
		 * as static to enable more permissive BPF verification mode
		 * with more outside context available to BPF verifier
		 */
		if (GELF_ST_BIND(sym.st_info) != STB_LOCAL
		    && (GELF_ST_VISIBILITY(sym.st_other) == STV_HIDDEN
			|| GELF_ST_VISIBILITY(sym.st_other) == STV_INTERNAL))
			prog->mark_btf_static = true;

		nr_progs++;
		obj->nr_programs = nr_progs;
	}

	return 0;
}

static __u32 get_kernel_version(void)
{
	__u32 major, minor, patch;
	struct utsname info;

	uname(&info);
	if (sscanf(info.release, "%u.%u.%u", &major, &minor, &patch) != 3)
		return 0;
	return KERNEL_VERSION(major, minor, patch);
}

static const struct btf_member *
find_member_by_offset(const struct btf_type *t, __u32 bit_offset)
{
	struct btf_member *m;
	int i;

	for (i = 0, m = btf_members(t); i < btf_vlen(t); i++, m++) {
		if (btf_member_bit_offset(t, i) == bit_offset)
			return m;
	}

	return NULL;
}

static const struct btf_member *
find_member_by_name(const struct btf *btf, const struct btf_type *t,
		    const char *name)
{
	struct btf_member *m;
	int i;

	for (i = 0, m = btf_members(t); i < btf_vlen(t); i++, m++) {
		if (!strcmp(btf__name_by_offset(btf, m->name_off), name))
			return m;
	}

	return NULL;
}

#define STRUCT_OPS_VALUE_PREFIX "bpf_struct_ops_"
static int find_btf_by_prefix_kind(const struct btf *btf, const char *prefix,
				   const char *name, __u32 kind);

static int
find_struct_ops_kern_types(const struct btf *btf, const char *tname,
			   const struct btf_type **type, __u32 *type_id,
			   const struct btf_type **vtype, __u32 *vtype_id,
			   const struct btf_member **data_member)
{
	const struct btf_type *kern_type, *kern_vtype;
	const struct btf_member *kern_data_member;
	__s32 kern_vtype_id, kern_type_id;
	__u32 i;

	kern_type_id = btf__find_by_name_kind(btf, tname, BTF_KIND_STRUCT);
	if (kern_type_id < 0) {
		pr_warn("struct_ops init_kern: struct %s is not found in kernel BTF\n",
			tname);
		return kern_type_id;
	}
	kern_type = btf__type_by_id(btf, kern_type_id);

	/* Find the corresponding "map_value" type that will be used
	 * in map_update(BPF_MAP_TYPE_STRUCT_OPS).  For example,
	 * find "struct bpf_struct_ops_tcp_congestion_ops" from the
	 * btf_vmlinux.
	 */
	kern_vtype_id = find_btf_by_prefix_kind(btf, STRUCT_OPS_VALUE_PREFIX,
						tname, BTF_KIND_STRUCT);
	if (kern_vtype_id < 0) {
		pr_warn("struct_ops init_kern: struct %s%s is not found in kernel BTF\n",
			STRUCT_OPS_VALUE_PREFIX, tname);
		return kern_vtype_id;
	}
	kern_vtype = btf__type_by_id(btf, kern_vtype_id);

	/* Find "struct tcp_congestion_ops" from
	 * struct bpf_struct_ops_tcp_congestion_ops {
	 *	[ ... ]
	 *	struct tcp_congestion_ops data;
	 * }
	 */
	kern_data_member = btf_members(kern_vtype);
	for (i = 0; i < btf_vlen(kern_vtype); i++, kern_data_member++) {
		if (kern_data_member->type == kern_type_id)
			break;
	}
	if (i == btf_vlen(kern_vtype)) {
		pr_warn("struct_ops init_kern: struct %s data is not found in struct %s%s\n",
			tname, STRUCT_OPS_VALUE_PREFIX, tname);
		return -EINVAL;
	}

	*type = kern_type;
	*type_id = kern_type_id;
	*vtype = kern_vtype;
	*vtype_id = kern_vtype_id;
	*data_member = kern_data_member;

	return 0;
}

static bool bpf_map__is_struct_ops(const struct bpf_map *map)
{
	return map->def.type == BPF_MAP_TYPE_STRUCT_OPS;
}

/* Init the map's fields that depend on kern_btf */
static int bpf_map__init_kern_struct_ops(struct bpf_map *map,
					 const struct btf *btf,
					 const struct btf *kern_btf)
{
	const struct btf_member *member, *kern_member, *kern_data_member;
	const struct btf_type *type, *kern_type, *kern_vtype;
	__u32 i, kern_type_id, kern_vtype_id, kern_data_off;
	struct bpf_struct_ops *st_ops;
	void *data, *kern_data;
	const char *tname;
	int err;

	st_ops = map->st_ops;
	type = st_ops->type;
	tname = st_ops->tname;
	err = find_struct_ops_kern_types(kern_btf, tname,
					 &kern_type, &kern_type_id,
					 &kern_vtype, &kern_vtype_id,
					 &kern_data_member);
	if (err)
		return err;

	pr_debug("struct_ops init_kern %s: type_id:%u kern_type_id:%u kern_vtype_id:%u\n",
		 map->name, st_ops->type_id, kern_type_id, kern_vtype_id);

	map->def.value_size = kern_vtype->size;
	map->btf_vmlinux_value_type_id = kern_vtype_id;

	st_ops->kern_vdata = calloc(1, kern_vtype->size);
	if (!st_ops->kern_vdata)
		return -ENOMEM;

	data = st_ops->data;
	kern_data_off = kern_data_member->offset / 8;
	kern_data = st_ops->kern_vdata + kern_data_off;

	member = btf_members(type);
	for (i = 0; i < btf_vlen(type); i++, member++) {
		const struct btf_type *mtype, *kern_mtype;
		__u32 mtype_id, kern_mtype_id;
		void *mdata, *kern_mdata;
		__s64 msize, kern_msize;
		__u32 moff, kern_moff;
		__u32 kern_member_idx;
		const char *mname;

		mname = btf__name_by_offset(btf, member->name_off);
		kern_member = find_member_by_name(kern_btf, kern_type, mname);
		if (!kern_member) {
			pr_warn("struct_ops init_kern %s: Cannot find member %s in kernel BTF\n",
				map->name, mname);
			return -ENOTSUP;
		}

		kern_member_idx = kern_member - btf_members(kern_type);
		if (btf_member_bitfield_size(type, i) ||
		    btf_member_bitfield_size(kern_type, kern_member_idx)) {
			pr_warn("struct_ops init_kern %s: bitfield %s is not supported\n",
				map->name, mname);
			return -ENOTSUP;
		}

		moff = member->offset / 8;
		kern_moff = kern_member->offset / 8;

		mdata = data + moff;
		kern_mdata = kern_data + kern_moff;

		mtype = skip_mods_and_typedefs(btf, member->type, &mtype_id);
		kern_mtype = skip_mods_and_typedefs(kern_btf, kern_member->type,
						    &kern_mtype_id);
		if (BTF_INFO_KIND(mtype->info) !=
		    BTF_INFO_KIND(kern_mtype->info)) {
			pr_warn("struct_ops init_kern %s: Unmatched member type %s %u != %u(kernel)\n",
				map->name, mname, BTF_INFO_KIND(mtype->info),
				BTF_INFO_KIND(kern_mtype->info));
			return -ENOTSUP;
		}

		if (btf_is_ptr(mtype)) {
			struct bpf_program *prog;

			prog = st_ops->progs[i];
			if (!prog)
				continue;

			kern_mtype = skip_mods_and_typedefs(kern_btf,
							    kern_mtype->type,
							    &kern_mtype_id);

			/* mtype->type must be a func_proto which was
			 * guaranteed in bpf_object__collect_st_ops_relos(),
			 * so only check kern_mtype for func_proto here.
			 */
			if (!btf_is_func_proto(kern_mtype)) {
				pr_warn("struct_ops init_kern %s: kernel member %s is not a func ptr\n",
					map->name, mname);
				return -ENOTSUP;
			}

			prog->attach_btf_id = kern_type_id;
			prog->expected_attach_type = kern_member_idx;

			st_ops->kern_func_off[i] = kern_data_off + kern_moff;

			pr_debug("struct_ops init_kern %s: func ptr %s is set to prog %s from data(+%u) to kern_data(+%u)\n",
				 map->name, mname, prog->name, moff,
				 kern_moff);

			continue;
		}

		msize = btf__resolve_size(btf, mtype_id);
		kern_msize = btf__resolve_size(kern_btf, kern_mtype_id);
		if (msize < 0 || kern_msize < 0 || msize != kern_msize) {
			pr_warn("struct_ops init_kern %s: Error in size of member %s: %zd != %zd(kernel)\n",
				map->name, mname, (ssize_t)msize,
				(ssize_t)kern_msize);
			return -ENOTSUP;
		}

		pr_debug("struct_ops init_kern %s: copy %s %u bytes from data(+%u) to kern_data(+%u)\n",
			 map->name, mname, (unsigned int)msize,
			 moff, kern_moff);
		memcpy(kern_mdata, mdata, msize);
	}

	return 0;
}

static int bpf_object__init_kern_struct_ops_maps(struct bpf_object *obj)
{
	struct bpf_map *map;
	size_t i;
	int err;

	for (i = 0; i < obj->nr_maps; i++) {
		map = &obj->maps[i];

		if (!bpf_map__is_struct_ops(map))
			continue;

		err = bpf_map__init_kern_struct_ops(map, obj->btf,
						    obj->btf_vmlinux);
		if (err)
			return err;
	}

	return 0;
}

static int bpf_object__init_struct_ops_maps(struct bpf_object *obj)
{
	const struct btf_type *type, *datasec;
	const struct btf_var_secinfo *vsi;
	struct bpf_struct_ops *st_ops;
	const char *tname, *var_name;
	__s32 type_id, datasec_id;
	const struct btf *btf;
	struct bpf_map *map;
	__u32 i;

	if (obj->efile.st_ops_shndx == -1)
		return 0;

	btf = obj->btf;
	datasec_id = btf__find_by_name_kind(btf, STRUCT_OPS_SEC,
					    BTF_KIND_DATASEC);
	if (datasec_id < 0) {
		pr_warn("struct_ops init: DATASEC %s not found\n",
			STRUCT_OPS_SEC);
		return -EINVAL;
	}

	datasec = btf__type_by_id(btf, datasec_id);
	vsi = btf_var_secinfos(datasec);
	for (i = 0; i < btf_vlen(datasec); i++, vsi++) {
		type = btf__type_by_id(obj->btf, vsi->type);
		var_name = btf__name_by_offset(obj->btf, type->name_off);

		type_id = btf__resolve_type(obj->btf, vsi->type);
		if (type_id < 0) {
			pr_warn("struct_ops init: Cannot resolve var type_id %u in DATASEC %s\n",
				vsi->type, STRUCT_OPS_SEC);
			return -EINVAL;
		}

		type = btf__type_by_id(obj->btf, type_id);
		tname = btf__name_by_offset(obj->btf, type->name_off);
		if (!tname[0]) {
			pr_warn("struct_ops init: anonymous type is not supported\n");
			return -ENOTSUP;
		}
		if (!btf_is_struct(type)) {
			pr_warn("struct_ops init: %s is not a struct\n", tname);
			return -EINVAL;
		}

		map = bpf_object__add_map(obj);
		if (IS_ERR(map))
			return PTR_ERR(map);

		map->sec_idx = obj->efile.st_ops_shndx;
		map->sec_offset = vsi->offset;
		map->name = strdup(var_name);
		if (!map->name)
			return -ENOMEM;

		map->def.type = BPF_MAP_TYPE_STRUCT_OPS;
		map->def.key_size = sizeof(int);
		map->def.value_size = type->size;
		map->def.max_entries = 1;

		map->st_ops = calloc(1, sizeof(*map->st_ops));
		if (!map->st_ops)
			return -ENOMEM;
		st_ops = map->st_ops;
		st_ops->data = malloc(type->size);
		st_ops->progs = calloc(btf_vlen(type), sizeof(*st_ops->progs));
		st_ops->kern_func_off = malloc(btf_vlen(type) *
					       sizeof(*st_ops->kern_func_off));
		if (!st_ops->data || !st_ops->progs || !st_ops->kern_func_off)
			return -ENOMEM;

		if (vsi->offset + type->size > obj->efile.st_ops_data->d_size) {
			pr_warn("struct_ops init: var %s is beyond the end of DATASEC %s\n",
				var_name, STRUCT_OPS_SEC);
			return -EINVAL;
		}

		memcpy(st_ops->data,
		       obj->efile.st_ops_data->d_buf + vsi->offset,
		       type->size);
		st_ops->tname = tname;
		st_ops->type = type;
		st_ops->type_id = type_id;

		pr_debug("struct_ops init: struct %s(type_id=%u) %s found at offset %u\n",
			 tname, type_id, var_name, vsi->offset);
	}

	return 0;
}

static struct bpf_object *bpf_object__new(const char *path,
					  const void *obj_buf,
					  size_t obj_buf_sz,
					  const char *obj_name)
{
	struct bpf_object *obj;
	char *end;

	obj = calloc(1, sizeof(struct bpf_object) + strlen(path) + 1);
	if (!obj) {
		pr_warn("alloc memory failed for %s\n", path);
		return ERR_PTR(-ENOMEM);
	}

	strcpy(obj->path, path);
	if (obj_name) {
		strncpy(obj->name, obj_name, sizeof(obj->name) - 1);
		obj->name[sizeof(obj->name) - 1] = 0;
	} else {
		/* Using basename() GNU version which doesn't modify arg. */
		strncpy(obj->name, basename((void *)path),
			sizeof(obj->name) - 1);
		end = strchr(obj->name, '.');
		if (end)
			*end = 0;
	}

	obj->efile.fd = -1;
	/*
	 * Caller of this function should also call
	 * bpf_object__elf_finish() after data collection to return
	 * obj_buf to user. If not, we should duplicate the buffer to
	 * avoid user freeing them before elf finish.
	 */
	obj->efile.obj_buf = obj_buf;
	obj->efile.obj_buf_sz = obj_buf_sz;
	obj->efile.maps_shndx = -1;
	obj->efile.btf_maps_shndx = -1;
	obj->efile.data_shndx = -1;
	obj->efile.rodata_shndx = -1;
	obj->efile.bss_shndx = -1;
	obj->efile.st_ops_shndx = -1;
	obj->kconfig_map_idx = -1;
	obj->rodata_map_idx = -1;

	obj->kern_version = get_kernel_version();
	obj->loaded = false;

	INIT_LIST_HEAD(&obj->list);
	list_add(&obj->list, &bpf_objects_list);
	return obj;
}

static void bpf_object__elf_finish(struct bpf_object *obj)
{
	if (!obj_elf_valid(obj))
		return;

	if (obj->efile.elf) {
		elf_end(obj->efile.elf);
		obj->efile.elf = NULL;
	}
	obj->efile.symbols = NULL;
	obj->efile.data = NULL;
	obj->efile.rodata = NULL;
	obj->efile.bss = NULL;
	obj->efile.st_ops_data = NULL;

	zfree(&obj->efile.reloc_sects);
	obj->efile.nr_reloc_sects = 0;
	zclose(obj->efile.fd);
	obj->efile.obj_buf = NULL;
	obj->efile.obj_buf_sz = 0;
}

static int bpf_object__elf_init(struct bpf_object *obj)
{
	int err = 0;
	GElf_Ehdr *ep;

	if (obj_elf_valid(obj)) {
		pr_warn("elf: init internal error\n");
		return -LIBBPF_ERRNO__LIBELF;
	}

	if (obj->efile.obj_buf_sz > 0) {
		/*
		 * obj_buf should have been validated by
		 * bpf_object__open_buffer().
		 */
		obj->efile.elf = elf_memory((char *)obj->efile.obj_buf,
					    obj->efile.obj_buf_sz);
	} else {
		obj->efile.fd = open(obj->path, O_RDONLY);
		if (obj->efile.fd < 0) {
			char errmsg[STRERR_BUFSIZE], *cp;

			err = -errno;
			cp = libbpf_strerror_r(err, errmsg, sizeof(errmsg));
			pr_warn("elf: failed to open %s: %s\n", obj->path, cp);
			return err;
		}

		obj->efile.elf = elf_begin(obj->efile.fd, ELF_C_READ_MMAP, NULL);
	}

	if (!obj->efile.elf) {
		pr_warn("elf: failed to open %s as ELF file: %s\n", obj->path, elf_errmsg(-1));
		err = -LIBBPF_ERRNO__LIBELF;
		goto errout;
	}

	if (!gelf_getehdr(obj->efile.elf, &obj->efile.ehdr)) {
		pr_warn("elf: failed to get ELF header from %s: %s\n", obj->path, elf_errmsg(-1));
		err = -LIBBPF_ERRNO__FORMAT;
		goto errout;
	}
	ep = &obj->efile.ehdr;

	if (elf_getshdrstrndx(obj->efile.elf, &obj->efile.shstrndx)) {
		pr_warn("elf: failed to get section names section index for %s: %s\n",
			obj->path, elf_errmsg(-1));
		err = -LIBBPF_ERRNO__FORMAT;
		goto errout;
	}

	/* Elf is corrupted/truncated, avoid calling elf_strptr. */
	if (!elf_rawdata(elf_getscn(obj->efile.elf, obj->efile.shstrndx), NULL)) {
		pr_warn("elf: failed to get section names strings from %s: %s\n",
			obj->path, elf_errmsg(-1));
		err = -LIBBPF_ERRNO__FORMAT;
		goto errout;
	}

	/* Old LLVM set e_machine to EM_NONE */
	if (ep->e_type != ET_REL ||
	    (ep->e_machine && ep->e_machine != EM_BPF)) {
		pr_warn("elf: %s is not a valid eBPF object file\n", obj->path);
		err = -LIBBPF_ERRNO__FORMAT;
		goto errout;
	}

	return 0;
errout:
	bpf_object__elf_finish(obj);
	return err;
}

static int bpf_object__check_endianness(struct bpf_object *obj)
{
#if __BYTE_ORDER == __LITTLE_ENDIAN
	if (obj->efile.ehdr.e_ident[EI_DATA] == ELFDATA2LSB)
		return 0;
#elif __BYTE_ORDER == __BIG_ENDIAN
	if (obj->efile.ehdr.e_ident[EI_DATA] == ELFDATA2MSB)
		return 0;
#else
# error "Unrecognized __BYTE_ORDER__"
#endif
	pr_warn("elf: endianness mismatch in %s.\n", obj->path);
	return -LIBBPF_ERRNO__ENDIAN;
}

static int
bpf_object__init_license(struct bpf_object *obj, void *data, size_t size)
{
	memcpy(obj->license, data, min(size, sizeof(obj->license) - 1));
	pr_debug("license of %s is %s\n", obj->path, obj->license);
	return 0;
}

static int
bpf_object__init_kversion(struct bpf_object *obj, void *data, size_t size)
{
	__u32 kver;

	if (size != sizeof(kver)) {
		pr_warn("invalid kver section in %s\n", obj->path);
		return -LIBBPF_ERRNO__FORMAT;
	}
	memcpy(&kver, data, sizeof(kver));
	obj->kern_version = kver;
	pr_debug("kernel version of %s is %x\n", obj->path, obj->kern_version);
	return 0;
}

static bool bpf_map_type__is_map_in_map(enum bpf_map_type type)
{
	if (type == BPF_MAP_TYPE_ARRAY_OF_MAPS ||
	    type == BPF_MAP_TYPE_HASH_OF_MAPS)
		return true;
	return false;
}

int bpf_object__section_size(const struct bpf_object *obj, const char *name,
			     __u32 *size)
{
	int ret = -ENOENT;

	*size = 0;
	if (!name) {
		return -EINVAL;
	} else if (!strcmp(name, DATA_SEC)) {
		if (obj->efile.data)
			*size = obj->efile.data->d_size;
	} else if (!strcmp(name, BSS_SEC)) {
		if (obj->efile.bss)
			*size = obj->efile.bss->d_size;
	} else if (!strcmp(name, RODATA_SEC)) {
		if (obj->efile.rodata)
			*size = obj->efile.rodata->d_size;
	} else if (!strcmp(name, STRUCT_OPS_SEC)) {
		if (obj->efile.st_ops_data)
			*size = obj->efile.st_ops_data->d_size;
	} else {
		Elf_Scn *scn = elf_sec_by_name(obj, name);
		Elf_Data *data = elf_sec_data(obj, scn);

		if (data) {
			ret = 0; /* found it */
			*size = data->d_size;
		}
	}

	return *size ? 0 : ret;
}

int bpf_object__variable_offset(const struct bpf_object *obj, const char *name,
				__u32 *off)
{
	Elf_Data *symbols = obj->efile.symbols;
	const char *sname;
	size_t si;

	if (!name || !off)
		return -EINVAL;

	for (si = 0; si < symbols->d_size / sizeof(GElf_Sym); si++) {
		GElf_Sym sym;

		if (!gelf_getsym(symbols, si, &sym))
			continue;
		if (GELF_ST_BIND(sym.st_info) != STB_GLOBAL ||
		    GELF_ST_TYPE(sym.st_info) != STT_OBJECT)
			continue;

		sname = elf_sym_str(obj, sym.st_name);
		if (!sname) {
			pr_warn("failed to get sym name string for var %s\n",
				name);
			return -EIO;
		}
		if (strcmp(name, sname) == 0) {
			*off = sym.st_value;
			return 0;
		}
	}

	return -ENOENT;
}

static struct bpf_map *bpf_object__add_map(struct bpf_object *obj)
{
	struct bpf_map *new_maps;
	size_t new_cap;
	int i;

	if (obj->nr_maps < obj->maps_cap)
		return &obj->maps[obj->nr_maps++];

	new_cap = max((size_t)4, obj->maps_cap * 3 / 2);
	new_maps = libbpf_reallocarray(obj->maps, new_cap, sizeof(*obj->maps));
	if (!new_maps) {
		pr_warn("alloc maps for object failed\n");
		return ERR_PTR(-ENOMEM);
	}

	obj->maps_cap = new_cap;
	obj->maps = new_maps;

	/* zero out new maps */
	memset(obj->maps + obj->nr_maps, 0,
	       (obj->maps_cap - obj->nr_maps) * sizeof(*obj->maps));
	/*
	 * fill all fd with -1 so won't close incorrect fd (fd=0 is stdin)
	 * when failure (zclose won't close negative fd)).
	 */
	for (i = obj->nr_maps; i < obj->maps_cap; i++) {
		obj->maps[i].fd = -1;
		obj->maps[i].inner_map_fd = -1;
	}

	return &obj->maps[obj->nr_maps++];
}

static size_t bpf_map_mmap_sz(const struct bpf_map *map)
{
	long page_sz = sysconf(_SC_PAGE_SIZE);
	size_t map_sz;

	map_sz = (size_t)roundup(map->def.value_size, 8) * map->def.max_entries;
	map_sz = roundup(map_sz, page_sz);
	return map_sz;
}

static char *internal_map_name(struct bpf_object *obj,
			       enum libbpf_map_type type)
{
	char map_name[BPF_OBJ_NAME_LEN], *p;
	const char *sfx = libbpf_type_to_btf_name[type];
	int sfx_len = max((size_t)7, strlen(sfx));
	int pfx_len = min((size_t)BPF_OBJ_NAME_LEN - sfx_len - 1,
			  strlen(obj->name));

	snprintf(map_name, sizeof(map_name), "%.*s%.*s", pfx_len, obj->name,
		 sfx_len, libbpf_type_to_btf_name[type]);

	/* sanitise map name to characters allowed by kernel */
	for (p = map_name; *p && p < map_name + sizeof(map_name); p++)
		if (!isalnum(*p) && *p != '_' && *p != '.')
			*p = '_';

	return strdup(map_name);
}

static int
bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type,
			      int sec_idx, void *data, size_t data_sz)
{
	struct bpf_map_def *def;
	struct bpf_map *map;
	int err;

	map = bpf_object__add_map(obj);
	if (IS_ERR(map))
		return PTR_ERR(map);

	map->libbpf_type = type;
	map->sec_idx = sec_idx;
	map->sec_offset = 0;
	map->name = internal_map_name(obj, type);
	if (!map->name) {
		pr_warn("failed to alloc map name\n");
		return -ENOMEM;
	}

	def = &map->def;
	def->type = BPF_MAP_TYPE_ARRAY;
	def->key_size = sizeof(int);
	def->value_size = data_sz;
	def->max_entries = 1;
	def->map_flags = type == LIBBPF_MAP_RODATA || type == LIBBPF_MAP_KCONFIG
			 ? BPF_F_RDONLY_PROG : 0;
	def->map_flags |= BPF_F_MMAPABLE;

	pr_debug("map '%s' (global data): at sec_idx %d, offset %zu, flags %x.\n",
		 map->name, map->sec_idx, map->sec_offset, def->map_flags);

	map->mmaped = mmap(NULL, bpf_map_mmap_sz(map), PROT_READ | PROT_WRITE,
			   MAP_SHARED | MAP_ANONYMOUS, -1, 0);
	if (map->mmaped == MAP_FAILED) {
		err = -errno;
		map->mmaped = NULL;
		pr_warn("failed to alloc map '%s' content buffer: %d\n",
			map->name, err);
		zfree(&map->name);
		return err;
	}

	if (data)
		memcpy(map->mmaped, data, data_sz);

	pr_debug("map %td is \"%s\"\n", map - obj->maps, map->name);
	return 0;
}

static int bpf_object__init_global_data_maps(struct bpf_object *obj)
{
	int err;

	/*
	 * Populate obj->maps with libbpf internal maps.
	 */
	if (obj->efile.data_shndx >= 0) {
		err = bpf_object__init_internal_map(obj, LIBBPF_MAP_DATA,
						    obj->efile.data_shndx,
						    obj->efile.data->d_buf,
						    obj->efile.data->d_size);
		if (err)
			return err;
	}
	if (obj->efile.rodata_shndx >= 0) {
		err = bpf_object__init_internal_map(obj, LIBBPF_MAP_RODATA,
						    obj->efile.rodata_shndx,
						    obj->efile.rodata->d_buf,
						    obj->efile.rodata->d_size);
		if (err)
			return err;

		obj->rodata_map_idx = obj->nr_maps - 1;
	}
	if (obj->efile.bss_shndx >= 0) {
		err = bpf_object__init_internal_map(obj, LIBBPF_MAP_BSS,
						    obj->efile.bss_shndx,
						    NULL,
						    obj->efile.bss->d_size);
		if (err)
			return err;
	}
	return 0;
}


static struct extern_desc *find_extern_by_name(const struct bpf_object *obj,
					       const void *name)
{
	int i;

	for (i = 0; i < obj->nr_extern; i++) {
		if (strcmp(obj->externs[i].name, name) == 0)
			return &obj->externs[i];
	}
	return NULL;
}

static int set_kcfg_value_tri(struct extern_desc *ext, void *ext_val,
			      char value)
{
	switch (ext->kcfg.type) {
	case KCFG_BOOL:
		if (value == 'm') {
			pr_warn("extern (kcfg) %s=%c should be tristate or char\n",
				ext->name, value);
			return -EINVAL;
		}
		*(bool *)ext_val = value == 'y' ? true : false;
		break;
	case KCFG_TRISTATE:
		if (value == 'y')
			*(enum libbpf_tristate *)ext_val = TRI_YES;
		else if (value == 'm')
			*(enum libbpf_tristate *)ext_val = TRI_MODULE;
		else /* value == 'n' */
			*(enum libbpf_tristate *)ext_val = TRI_NO;
		break;
	case KCFG_CHAR:
		*(char *)ext_val = value;
		break;
	case KCFG_UNKNOWN:
	case KCFG_INT:
	case KCFG_CHAR_ARR:
	default:
		pr_warn("extern (kcfg) %s=%c should be bool, tristate, or char\n",
			ext->name, value);
		return -EINVAL;
	}
	ext->is_set = true;
	return 0;
}

static int set_kcfg_value_str(struct extern_desc *ext, char *ext_val,
			      const char *value)
{
	size_t len;

	if (ext->kcfg.type != KCFG_CHAR_ARR) {
		pr_warn("extern (kcfg) %s=%s should be char array\n", ext->name, value);
		return -EINVAL;
	}

	len = strlen(value);
	if (value[len - 1] != '"') {
		pr_warn("extern (kcfg) '%s': invalid string config '%s'\n",
			ext->name, value);
		return -EINVAL;
	}

	/* strip quotes */
	len -= 2;
	if (len >= ext->kcfg.sz) {
		pr_warn("extern (kcfg) '%s': long string config %s of (%zu bytes) truncated to %d bytes\n",
			ext->name, value, len, ext->kcfg.sz - 1);
		len = ext->kcfg.sz - 1;
	}
	memcpy(ext_val, value + 1, len);
	ext_val[len] = '\0';
	ext->is_set = true;
	return 0;
}

static int parse_u64(const char *value, __u64 *res)
{
	char *value_end;
	int err;

	errno = 0;
	*res = strtoull(value, &value_end, 0);
	if (errno) {
		err = -errno;
		pr_warn("failed to parse '%s' as integer: %d\n", value, err);
		return err;
	}
	if (*value_end) {
		pr_warn("failed to parse '%s' as integer completely\n", value);
		return -EINVAL;
	}
	return 0;
}

static bool is_kcfg_value_in_range(const struct extern_desc *ext, __u64 v)
{
	int bit_sz = ext->kcfg.sz * 8;

	if (ext->kcfg.sz == 8)
		return true;

	/* Validate that value stored in u64 fits in integer of `ext->sz`
	 * bytes size without any loss of information. If the target integer
	 * is signed, we rely on the following limits of integer type of
	 * Y bits and subsequent transformation:
	 *
	 *     -2^(Y-1) <= X           <= 2^(Y-1) - 1
	 *            0 <= X + 2^(Y-1) <= 2^Y - 1
	 *            0 <= X + 2^(Y-1) <  2^Y
	 *
	 *  For unsigned target integer, check that all the (64 - Y) bits are
	 *  zero.
	 */
	if (ext->kcfg.is_signed)
		return v + (1ULL << (bit_sz - 1)) < (1ULL << bit_sz);
	else
		return (v >> bit_sz) == 0;
}

static int set_kcfg_value_num(struct extern_desc *ext, void *ext_val,
			      __u64 value)
{
	if (ext->kcfg.type != KCFG_INT && ext->kcfg.type != KCFG_CHAR) {
		pr_warn("extern (kcfg) %s=%llu should be integer\n",
			ext->name, (unsigned long long)value);
		return -EINVAL;
	}
	if (!is_kcfg_value_in_range(ext, value)) {
		pr_warn("extern (kcfg) %s=%llu value doesn't fit in %d bytes\n",
			ext->name, (unsigned long long)value, ext->kcfg.sz);
		return -ERANGE;
	}
	switch (ext->kcfg.sz) {
		case 1: *(__u8 *)ext_val = value; break;
		case 2: *(__u16 *)ext_val = value; break;
		case 4: *(__u32 *)ext_val = value; break;
		case 8: *(__u64 *)ext_val = value; break;
		default:
			return -EINVAL;
	}
	ext->is_set = true;
	return 0;
}

static int bpf_object__process_kconfig_line(struct bpf_object *obj,
					    char *buf, void *data)
{
	struct extern_desc *ext;
	char *sep, *value;
	int len, err = 0;
	void *ext_val;
	__u64 num;

	if (strncmp(buf, "CONFIG_", 7))
		return 0;

	sep = strchr(buf, '=');
	if (!sep) {
		pr_warn("failed to parse '%s': no separator\n", buf);
		return -EINVAL;
	}

	/* Trim ending '\n' */
	len = strlen(buf);
	if (buf[len - 1] == '\n')
		buf[len - 1] = '\0';
	/* Split on '=' and ensure that a value is present. */
	*sep = '\0';
	if (!sep[1]) {
		*sep = '=';
		pr_warn("failed to parse '%s': no value\n", buf);
		return -EINVAL;
	}

	ext = find_extern_by_name(obj, buf);
	if (!ext || ext->is_set)
		return 0;

	ext_val = data + ext->kcfg.data_off;
	value = sep + 1;

	switch (*value) {
	case 'y': case 'n': case 'm':
		err = set_kcfg_value_tri(ext, ext_val, *value);
		break;
	case '"':
		err = set_kcfg_value_str(ext, ext_val, value);
		break;
	default:
		/* assume integer */
		err = parse_u64(value, &num);
		if (err) {
			pr_warn("extern (kcfg) %s=%s should be integer\n",
				ext->name, value);
			return err;
		}
		err = set_kcfg_value_num(ext, ext_val, num);
		break;
	}
	if (err)
		return err;
	pr_debug("extern (kcfg) %s=%s\n", ext->name, value);
	return 0;
}

static int bpf_object__read_kconfig_file(struct bpf_object *obj, void *data)
{
	char buf[PATH_MAX];
	struct utsname uts;
	int len, err = 0;
	gzFile file;

	uname(&uts);
	len = snprintf(buf, PATH_MAX, "/boot/config-%s", uts.release);
	if (len < 0)
		return -EINVAL;
	else if (len >= PATH_MAX)
		return -ENAMETOOLONG;

	/* gzopen also accepts uncompressed files. */
	file = gzopen(buf, "r");
	if (!file)
		file = gzopen("/proc/config.gz", "r");

	if (!file) {
		pr_warn("failed to open system Kconfig\n");
		return -ENOENT;
	}

	while (gzgets(file, buf, sizeof(buf))) {
		err = bpf_object__process_kconfig_line(obj, buf, data);
		if (err) {
			pr_warn("error parsing system Kconfig line '%s': %d\n",
				buf, err);
			goto out;
		}
	}

out:
	gzclose(file);
	return err;
}

static int bpf_object__read_kconfig_mem(struct bpf_object *obj,
					const char *config, void *data)
{
	char buf[PATH_MAX];
	int err = 0;
	FILE *file;

	file = fmemopen((void *)config, strlen(config), "r");
	if (!file) {
		err = -errno;
		pr_warn("failed to open in-memory Kconfig: %d\n", err);
		return err;
	}

	while (fgets(buf, sizeof(buf), file)) {
		err = bpf_object__process_kconfig_line(obj, buf, data);
		if (err) {
			pr_warn("error parsing in-memory Kconfig line '%s': %d\n",
				buf, err);
			break;
		}
	}

	fclose(file);
	return err;
}

static int bpf_object__init_kconfig_map(struct bpf_object *obj)
{
	struct extern_desc *last_ext = NULL, *ext;
	size_t map_sz;
	int i, err;

	for (i = 0; i < obj->nr_extern; i++) {
		ext = &obj->externs[i];
		if (ext->type == EXT_KCFG)
			last_ext = ext;
	}

	if (!last_ext)
		return 0;

	map_sz = last_ext->kcfg.data_off + last_ext->kcfg.sz;
	err = bpf_object__init_internal_map(obj, LIBBPF_MAP_KCONFIG,
					    obj->efile.symbols_shndx,
					    NULL, map_sz);
	if (err)
		return err;

	obj->kconfig_map_idx = obj->nr_maps - 1;

	return 0;
}

static int bpf_object__init_user_maps(struct bpf_object *obj, bool strict)
{
	Elf_Data *symbols = obj->efile.symbols;
	int i, map_def_sz = 0, nr_maps = 0, nr_syms;
	Elf_Data *data = NULL;
	Elf_Scn *scn;

	if (obj->efile.maps_shndx < 0)
		return 0;

	if (!symbols)
		return -EINVAL;

	scn = elf_sec_by_idx(obj, obj->efile.maps_shndx);
	data = elf_sec_data(obj, scn);
	if (!scn || !data) {
		pr_warn("elf: failed to get legacy map definitions for %s\n",
			obj->path);
		return -EINVAL;
	}

	/*
	 * Count number of maps. Each map has a name.
	 * Array of maps is not supported: only the first element is
	 * considered.
	 *
	 * TODO: Detect array of map and report error.
	 */
	nr_syms = symbols->d_size / sizeof(GElf_Sym);
	for (i = 0; i < nr_syms; i++) {
		GElf_Sym sym;

		if (!gelf_getsym(symbols, i, &sym))
			continue;
		if (sym.st_shndx != obj->efile.maps_shndx)
			continue;
		nr_maps++;
	}
	/* Assume equally sized map definitions */
	pr_debug("elf: found %d legacy map definitions (%zd bytes) in %s\n",
		 nr_maps, data->d_size, obj->path);

	if (!data->d_size || nr_maps == 0 || (data->d_size % nr_maps) != 0) {
		pr_warn("elf: unable to determine legacy map definition size in %s\n",
			obj->path);
		return -EINVAL;
	}
	map_def_sz = data->d_size / nr_maps;

	/* Fill obj->maps using data in "maps" section.  */
	for (i = 0; i < nr_syms; i++) {
		GElf_Sym sym;
		const char *map_name;
		struct bpf_map_def *def;
		struct bpf_map *map;

		if (!gelf_getsym(symbols, i, &sym))
			continue;
		if (sym.st_shndx != obj->efile.maps_shndx)
			continue;

		map = bpf_object__add_map(obj);
		if (IS_ERR(map))
			return PTR_ERR(map);

		map_name = elf_sym_str(obj, sym.st_name);
		if (!map_name) {
			pr_warn("failed to get map #%d name sym string for obj %s\n",
				i, obj->path);
			return -LIBBPF_ERRNO__FORMAT;
		}

		if (GELF_ST_TYPE(sym.st_info) == STT_SECTION
		    || GELF_ST_BIND(sym.st_info) == STB_LOCAL) {
			pr_warn("map '%s' (legacy): static maps are not supported\n", map_name);
			return -ENOTSUP;
		}

		map->libbpf_type = LIBBPF_MAP_UNSPEC;
		map->sec_idx = sym.st_shndx;
		map->sec_offset = sym.st_value;
		pr_debug("map '%s' (legacy): at sec_idx %d, offset %zu.\n",
			 map_name, map->sec_idx, map->sec_offset);
		if (sym.st_value + map_def_sz > data->d_size) {
			pr_warn("corrupted maps section in %s: last map \"%s\" too small\n",
				obj->path, map_name);
			return -EINVAL;
		}

		map->name = strdup(map_name);
		if (!map->name) {
			pr_warn("failed to alloc map name\n");
			return -ENOMEM;
		}
		pr_debug("map %d is \"%s\"\n", i, map->name);
		def = (struct bpf_map_def *)(data->d_buf + sym.st_value);
		/*
		 * If the definition of the map in the object file fits in
		 * bpf_map_def, copy it.  Any extra fields in our version
		 * of bpf_map_def will default to zero as a result of the
		 * calloc above.
		 */
		if (map_def_sz <= sizeof(struct bpf_map_def)) {
			memcpy(&map->def, def, map_def_sz);
		} else {
			/*
			 * Here the map structure being read is bigger than what
			 * we expect, truncate if the excess bits are all zero.
			 * If they are not zero, reject this map as
			 * incompatible.
			 */
			char *b;

			for (b = ((char *)def) + sizeof(struct bpf_map_def);
			     b < ((char *)def) + map_def_sz; b++) {
				if (*b != 0) {
					pr_warn("maps section in %s: \"%s\" has unrecognized, non-zero options\n",
						obj->path, map_name);
					if (strict)
						return -EINVAL;
				}
			}
			memcpy(&map->def, def, sizeof(struct bpf_map_def));
		}
	}
	return 0;
}

const struct btf_type *
skip_mods_and_typedefs(const struct btf *btf, __u32 id, __u32 *res_id)
{
	const struct btf_type *t = btf__type_by_id(btf, id);

	if (res_id)
		*res_id = id;

	while (btf_is_mod(t) || btf_is_typedef(t)) {
		if (res_id)
			*res_id = t->type;
		t = btf__type_by_id(btf, t->type);
	}

	return t;
}

static const struct btf_type *
resolve_func_ptr(const struct btf *btf, __u32 id, __u32 *res_id)
{
	const struct btf_type *t;

	t = skip_mods_and_typedefs(btf, id, NULL);
	if (!btf_is_ptr(t))
		return NULL;

	t = skip_mods_and_typedefs(btf, t->type, res_id);

	return btf_is_func_proto(t) ? t : NULL;
}

static const char *__btf_kind_str(__u16 kind)
{
	switch (kind) {
	case BTF_KIND_UNKN: return "void";
	case BTF_KIND_INT: return "int";
	case BTF_KIND_PTR: return "ptr";
	case BTF_KIND_ARRAY: return "array";
	case BTF_KIND_STRUCT: return "struct";
	case BTF_KIND_UNION: return "union";
	case BTF_KIND_ENUM: return "enum";
	case BTF_KIND_FWD: return "fwd";
	case BTF_KIND_TYPEDEF: return "typedef";
	case BTF_KIND_VOLATILE: return "volatile";
	case BTF_KIND_CONST: return "const";
	case BTF_KIND_RESTRICT: return "restrict";
	case BTF_KIND_FUNC: return "func";
	case BTF_KIND_FUNC_PROTO: return "func_proto";
	case BTF_KIND_VAR: return "var";
	case BTF_KIND_DATASEC: return "datasec";
	case BTF_KIND_FLOAT: return "float";
	default: return "unknown";
	}
}

const char *btf_kind_str(const struct btf_type *t)
{
	return __btf_kind_str(btf_kind(t));
}

/*
 * Fetch integer attribute of BTF map definition. Such attributes are
 * represented using a pointer to an array, in which dimensionality of array
 * encodes specified integer value. E.g., int (*type)[BPF_MAP_TYPE_ARRAY];
 * encodes `type => BPF_MAP_TYPE_ARRAY` key/value pair completely using BTF
 * type definition, while using only sizeof(void *) space in ELF data section.
 */
static bool get_map_field_int(const char *map_name, const struct btf *btf,
			      const struct btf_member *m, __u32 *res)
{
	const struct btf_type *t = skip_mods_and_typedefs(btf, m->type, NULL);
	const char *name = btf__name_by_offset(btf, m->name_off);
	const struct btf_array *arr_info;
	const struct btf_type *arr_t;

	if (!btf_is_ptr(t)) {
		pr_warn("map '%s': attr '%s': expected PTR, got %s.\n",
			map_name, name, btf_kind_str(t));
		return false;
	}

	arr_t = btf__type_by_id(btf, t->type);
	if (!arr_t) {
		pr_warn("map '%s': attr '%s': type [%u] not found.\n",
			map_name, name, t->type);
		return false;
	}
	if (!btf_is_array(arr_t)) {
		pr_warn("map '%s': attr '%s': expected ARRAY, got %s.\n",
			map_name, name, btf_kind_str(arr_t));
		return false;
	}
	arr_info = btf_array(arr_t);
	*res = arr_info->nelems;
	return true;
}

static int build_map_pin_path(struct bpf_map *map, const char *path)
{
	char buf[PATH_MAX];
	int len;

	if (!path)
		path = "/sys/fs/bpf";

	len = snprintf(buf, PATH_MAX, "%s/%s", path, bpf_map__name(map));
	if (len < 0)
		return -EINVAL;
	else if (len >= PATH_MAX)
		return -ENAMETOOLONG;

	return bpf_map__set_pin_path(map, buf);
}

int parse_btf_map_def(const char *map_name, struct btf *btf,
		      const struct btf_type *def_t, bool strict,
		      struct btf_map_def *map_def, struct btf_map_def *inner_def)
{
	const struct btf_type *t;
	const struct btf_member *m;
	bool is_inner = inner_def == NULL;
	int vlen, i;

	vlen = btf_vlen(def_t);
	m = btf_members(def_t);
	for (i = 0; i < vlen; i++, m++) {
		const char *name = btf__name_by_offset(btf, m->name_off);

		if (!name) {
			pr_warn("map '%s': invalid field #%d.\n", map_name, i);
			return -EINVAL;
		}
		if (strcmp(name, "type") == 0) {
			if (!get_map_field_int(map_name, btf, m, &map_def->map_type))
				return -EINVAL;
			map_def->parts |= MAP_DEF_MAP_TYPE;
		} else if (strcmp(name, "max_entries") == 0) {
			if (!get_map_field_int(map_name, btf, m, &map_def->max_entries))
				return -EINVAL;
			map_def->parts |= MAP_DEF_MAX_ENTRIES;
		} else if (strcmp(name, "map_flags") == 0) {
			if (!get_map_field_int(map_name, btf, m, &map_def->map_flags))
				return -EINVAL;
			map_def->parts |= MAP_DEF_MAP_FLAGS;
		} else if (strcmp(name, "numa_node") == 0) {
			if (!get_map_field_int(map_name, btf, m, &map_def->numa_node))
				return -EINVAL;
			map_def->parts |= MAP_DEF_NUMA_NODE;
		} else if (strcmp(name, "key_size") == 0) {
			__u32 sz;

			if (!get_map_field_int(map_name, btf, m, &sz))
				return -EINVAL;
			if (map_def->key_size && map_def->key_size != sz) {
				pr_warn("map '%s': conflicting key size %u != %u.\n",
					map_name, map_def->key_size, sz);
				return -EINVAL;
			}
			map_def->key_size = sz;
			map_def->parts |= MAP_DEF_KEY_SIZE;
		} else if (strcmp(name, "key") == 0) {
			__s64 sz;

			t = btf__type_by_id(btf, m->type);
			if (!t) {
				pr_warn("map '%s': key type [%d] not found.\n",
					map_name, m->type);
				return -EINVAL;
			}
			if (!btf_is_ptr(t)) {
				pr_warn("map '%s': key spec is not PTR: %s.\n",
					map_name, btf_kind_str(t));
				return -EINVAL;
			}
			sz = btf__resolve_size(btf, t->type);
			if (sz < 0) {
				pr_warn("map '%s': can't determine key size for type [%u]: %zd.\n",
					map_name, t->type, (ssize_t)sz);
				return sz;
			}
			if (map_def->key_size && map_def->key_size != sz) {
				pr_warn("map '%s': conflicting key size %u != %zd.\n",
					map_name, map_def->key_size, (ssize_t)sz);
				return -EINVAL;
			}
			map_def->key_size = sz;
			map_def->key_type_id = t->type;
			map_def->parts |= MAP_DEF_KEY_SIZE | MAP_DEF_KEY_TYPE;
		} else if (strcmp(name, "value_size") == 0) {
			__u32 sz;

			if (!get_map_field_int(map_name, btf, m, &sz))
				return -EINVAL;
			if (map_def->value_size && map_def->value_size != sz) {
				pr_warn("map '%s': conflicting value size %u != %u.\n",
					map_name, map_def->value_size, sz);
				return -EINVAL;
			}
			map_def->value_size = sz;
			map_def->parts |= MAP_DEF_VALUE_SIZE;
		} else if (strcmp(name, "value") == 0) {
			__s64 sz;

			t = btf__type_by_id(btf, m->type);
			if (!t) {
				pr_warn("map '%s': value type [%d] not found.\n",
					map_name, m->type);
				return -EINVAL;
			}
			if (!btf_is_ptr(t)) {
				pr_warn("map '%s': value spec is not PTR: %s.\n",
					map_name, btf_kind_str(t));
				return -EINVAL;
			}
			sz = btf__resolve_size(btf, t->type);
			if (sz < 0) {
				pr_warn("map '%s': can't determine value size for type [%u]: %zd.\n",
					map_name, t->type, (ssize_t)sz);
				return sz;
			}
			if (map_def->value_size && map_def->value_size != sz) {
				pr_warn("map '%s': conflicting value size %u != %zd.\n",
					map_name, map_def->value_size, (ssize_t)sz);
				return -EINVAL;
			}
			map_def->value_size = sz;
			map_def->value_type_id = t->type;
			map_def->parts |= MAP_DEF_VALUE_SIZE | MAP_DEF_VALUE_TYPE;
		}
		else if (strcmp(name, "values") == 0) {
			char inner_map_name[128];
			int err;

			if (is_inner) {
				pr_warn("map '%s': multi-level inner maps not supported.\n",
					map_name);
				return -ENOTSUP;
			}
			if (i != vlen - 1) {
				pr_warn("map '%s': '%s' member should be last.\n",
					map_name, name);
				return -EINVAL;
			}
			if (!bpf_map_type__is_map_in_map(map_def->map_type)) {
				pr_warn("map '%s': should be map-in-map.\n",
					map_name);
				return -ENOTSUP;
			}
			if (map_def->value_size && map_def->value_size != 4) {
				pr_warn("map '%s': conflicting value size %u != 4.\n",
					map_name, map_def->value_size);
				return -EINVAL;
			}
			map_def->value_size = 4;
			t = btf__type_by_id(btf, m->type);
			if (!t) {
				pr_warn("map '%s': map-in-map inner type [%d] not found.\n",
					map_name, m->type);
				return -EINVAL;
			}
			if (!btf_is_array(t) || btf_array(t)->nelems) {
				pr_warn("map '%s': map-in-map inner spec is not a zero-sized array.\n",
					map_name);
				return -EINVAL;
			}
			t = skip_mods_and_typedefs(btf, btf_array(t)->type, NULL);
			if (!btf_is_ptr(t)) {
				pr_warn("map '%s': map-in-map inner def is of unexpected kind %s.\n",
					map_name, btf_kind_str(t));
				return -EINVAL;
			}
			t = skip_mods_and_typedefs(btf, t->type, NULL);
			if (!btf_is_struct(t)) {
				pr_warn("map '%s': map-in-map inner def is of unexpected kind %s.\n",
					map_name, btf_kind_str(t));
				return -EINVAL;
			}

			snprintf(inner_map_name, sizeof(inner_map_name), "%s.inner", map_name);
			err = parse_btf_map_def(inner_map_name, btf, t, strict, inner_def, NULL);
			if (err)
				return err;

			map_def->parts |= MAP_DEF_INNER_MAP;
		} else if (strcmp(name, "pinning") == 0) {
			__u32 val;

			if (is_inner) {
				pr_warn("map '%s': inner def can't be pinned.\n", map_name);
				return -EINVAL;
			}
			if (!get_map_field_int(map_name, btf, m, &val))
				return -EINVAL;
			if (val != LIBBPF_PIN_NONE && val != LIBBPF_PIN_BY_NAME) {
				pr_warn("map '%s': invalid pinning value %u.\n",
					map_name, val);
				return -EINVAL;
			}
			map_def->pinning = val;
			map_def->parts |= MAP_DEF_PINNING;
		} else {
			if (strict) {
				pr_warn("map '%s': unknown field '%s'.\n", map_name, name);
				return -ENOTSUP;
			}
			pr_debug("map '%s': ignoring unknown field '%s'.\n", map_name, name);
		}
	}

	if (map_def->map_type == BPF_MAP_TYPE_UNSPEC) {
		pr_warn("map '%s': map type isn't specified.\n", map_name);
		return -EINVAL;
	}

	return 0;
}

static void fill_map_from_def(struct bpf_map *map, const struct btf_map_def *def)
{
	map->def.type = def->map_type;
	map->def.key_size = def->key_size;
	map->def.value_size = def->value_size;
	map->def.max_entries = def->max_entries;
	map->def.map_flags = def->map_flags;

	map->numa_node = def->numa_node;
	map->btf_key_type_id = def->key_type_id;
	map->btf_value_type_id = def->value_type_id;

	if (def->parts & MAP_DEF_MAP_TYPE)
		pr_debug("map '%s': found type = %u.\n", map->name, def->map_type);

	if (def->parts & MAP_DEF_KEY_TYPE)
		pr_debug("map '%s': found key [%u], sz = %u.\n",
			 map->name, def->key_type_id, def->key_size);
	else if (def->parts & MAP_DEF_KEY_SIZE)
		pr_debug("map '%s': found key_size = %u.\n", map->name, def->key_size);

	if (def->parts & MAP_DEF_VALUE_TYPE)
		pr_debug("map '%s': found value [%u], sz = %u.\n",
			 map->name, def->value_type_id, def->value_size);
	else if (def->parts & MAP_DEF_VALUE_SIZE)
		pr_debug("map '%s': found value_size = %u.\n", map->name, def->value_size);

	if (def->parts & MAP_DEF_MAX_ENTRIES)
		pr_debug("map '%s': found max_entries = %u.\n", map->name, def->max_entries);
	if (def->parts & MAP_DEF_MAP_FLAGS)
		pr_debug("map '%s': found map_flags = %u.\n", map->name, def->map_flags);
	if (def->parts & MAP_DEF_PINNING)
		pr_debug("map '%s': found pinning = %u.\n", map->name, def->pinning);
	if (def->parts & MAP_DEF_NUMA_NODE)
		pr_debug("map '%s': found numa_node = %u.\n", map->name, def->numa_node);

	if (def->parts & MAP_DEF_INNER_MAP)
		pr_debug("map '%s': found inner map definition.\n", map->name);
}

static const char *btf_var_linkage_str(__u32 linkage)
{
	switch (linkage) {
	case BTF_VAR_STATIC: return "static";
	case BTF_VAR_GLOBAL_ALLOCATED: return "global";
	case BTF_VAR_GLOBAL_EXTERN: return "extern";
	default: return "unknown";
	}
}

static int bpf_object__init_user_btf_map(struct bpf_object *obj,
					 const struct btf_type *sec,
					 int var_idx, int sec_idx,
					 const Elf_Data *data, bool strict,
					 const char *pin_root_path)
{
	struct btf_map_def map_def = {}, inner_def = {};
	const struct btf_type *var, *def;
	const struct btf_var_secinfo *vi;
	const struct btf_var *var_extra;
	const char *map_name;
	struct bpf_map *map;
	int err;

	vi = btf_var_secinfos(sec) + var_idx;
	var = btf__type_by_id(obj->btf, vi->type);
	var_extra = btf_var(var);
	map_name = btf__name_by_offset(obj->btf, var->name_off);

	if (map_name == NULL || map_name[0] == '\0') {
		pr_warn("map #%d: empty name.\n", var_idx);
		return -EINVAL;
	}
	if ((__u64)vi->offset + vi->size > data->d_size) {
		pr_warn("map '%s' BTF data is corrupted.\n", map_name);
		return -EINVAL;
	}
	if (!btf_is_var(var)) {
		pr_warn("map '%s': unexpected var kind %s.\n",
			map_name, btf_kind_str(var));
		return -EINVAL;
	}
	if (var_extra->linkage != BTF_VAR_GLOBAL_ALLOCATED) {
		pr_warn("map '%s': unsupported map linkage %s.\n",
			map_name, btf_var_linkage_str(var_extra->linkage));
		return -EOPNOTSUPP;
	}

	def = skip_mods_and_typedefs(obj->btf, var->type, NULL);
	if (!btf_is_struct(def)) {
		pr_warn("map '%s': unexpected def kind %s.\n",
			map_name, btf_kind_str(var));
		return -EINVAL;
	}
	if (def->size > vi->size) {
		pr_warn("map '%s': invalid def size.\n", map_name);
		return -EINVAL;
	}

	map = bpf_object__add_map(obj);
	if (IS_ERR(map))
		return PTR_ERR(map);
	map->name = strdup(map_name);
	if (!map->name) {
		pr_warn("map '%s': failed to alloc map name.\n", map_name);
		return -ENOMEM;
	}
	map->libbpf_type = LIBBPF_MAP_UNSPEC;
	map->def.type = BPF_MAP_TYPE_UNSPEC;
	map->sec_idx = sec_idx;
	map->sec_offset = vi->offset;
	map->btf_var_idx = var_idx;
	pr_debug("map '%s': at sec_idx %d, offset %zu.\n",
		 map_name, map->sec_idx, map->sec_offset);

	err = parse_btf_map_def(map->name, obj->btf, def, strict, &map_def, &inner_def);
	if (err)
		return err;

	fill_map_from_def(map, &map_def);

	if (map_def.pinning == LIBBPF_PIN_BY_NAME) {
		err = build_map_pin_path(map, pin_root_path);
		if (err) {
			pr_warn("map '%s': couldn't build pin path.\n", map->name);
			return err;
		}
	}

	if (map_def.parts & MAP_DEF_INNER_MAP) {
		map->inner_map = calloc(1, sizeof(*map->inner_map));
		if (!map->inner_map)
			return -ENOMEM;
		map->inner_map->fd = -1;
		map->inner_map->sec_idx = sec_idx;
		map->inner_map->name = malloc(strlen(map_name) + sizeof(".inner") + 1);
		if (!map->inner_map->name)
			return -ENOMEM;
		sprintf(map->inner_map->name, "%s.inner", map_name);

		fill_map_from_def(map->inner_map, &inner_def);
	}

	return 0;
}

static int bpf_object__init_user_btf_maps(struct bpf_object *obj, bool strict,
					  const char *pin_root_path)
{
	const struct btf_type *sec = NULL;
	int nr_types, i, vlen, err;
	const struct btf_type *t;
	const char *name;
	Elf_Data *data;
	Elf_Scn *scn;

	if (obj->efile.btf_maps_shndx < 0)
		return 0;

	scn = elf_sec_by_idx(obj, obj->efile.btf_maps_shndx);
	data = elf_sec_data(obj, scn);
	if (!scn || !data) {
		pr_warn("elf: failed to get %s map definitions for %s\n",
			MAPS_ELF_SEC, obj->path);
		return -EINVAL;
	}

	nr_types = btf__get_nr_types(obj->btf);
	for (i = 1; i <= nr_types; i++) {
		t = btf__type_by_id(obj->btf, i);
		if (!btf_is_datasec(t))
			continue;
		name = btf__name_by_offset(obj->btf, t->name_off);
		if (strcmp(name, MAPS_ELF_SEC) == 0) {
			sec = t;
			obj->efile.btf_maps_sec_btf_id = i;
			break;
		}
	}

	if (!sec) {
		pr_warn("DATASEC '%s' not found.\n", MAPS_ELF_SEC);
		return -ENOENT;
	}

	vlen = btf_vlen(sec);
	for (i = 0; i < vlen; i++) {
		err = bpf_object__init_user_btf_map(obj, sec, i,
						    obj->efile.btf_maps_shndx,
						    data, strict,
						    pin_root_path);
		if (err)
			return err;
	}

	return 0;
}

static int bpf_object__init_maps(struct bpf_object *obj,
				 const struct bpf_object_open_opts *opts)
{
	const char *pin_root_path;
	bool strict;
	int err;

	strict = !OPTS_GET(opts, relaxed_maps, false);
	pin_root_path = OPTS_GET(opts, pin_root_path, NULL);

	err = bpf_object__init_user_maps(obj, strict);
	err = err ?: bpf_object__init_user_btf_maps(obj, strict, pin_root_path);
	err = err ?: bpf_object__init_global_data_maps(obj);
	err = err ?: bpf_object__init_kconfig_map(obj);
	err = err ?: bpf_object__init_struct_ops_maps(obj);

	return err;
}

static bool section_have_execinstr(struct bpf_object *obj, int idx)
{
	GElf_Shdr sh;

	if (elf_sec_hdr(obj, elf_sec_by_idx(obj, idx), &sh))
		return false;

	return sh.sh_flags & SHF_EXECINSTR;
}

static bool btf_needs_sanitization(struct bpf_object *obj)
{
	bool has_func_global = kernel_supports(obj, FEAT_BTF_GLOBAL_FUNC);
	bool has_datasec = kernel_supports(obj, FEAT_BTF_DATASEC);
	bool has_float = kernel_supports(obj, FEAT_BTF_FLOAT);
	bool has_func = kernel_supports(obj, FEAT_BTF_FUNC);

	return !has_func || !has_datasec || !has_func_global || !has_float;
}

static void bpf_object__sanitize_btf(struct bpf_object *obj, struct btf *btf)
{
	bool has_func_global = kernel_supports(obj, FEAT_BTF_GLOBAL_FUNC);
	bool has_datasec = kernel_supports(obj, FEAT_BTF_DATASEC);
	bool has_float = kernel_supports(obj, FEAT_BTF_FLOAT);
	bool has_func = kernel_supports(obj, FEAT_BTF_FUNC);
	struct btf_type *t;
	int i, j, vlen;

	for (i = 1; i <= btf__get_nr_types(btf); i++) {
		t = (struct btf_type *)btf__type_by_id(btf, i);

		if (!has_datasec && btf_is_var(t)) {
			/* replace VAR with INT */
			t->info = BTF_INFO_ENC(BTF_KIND_INT, 0, 0);
			/*
			 * using size = 1 is the safest choice, 4 will be too
			 * big and cause kernel BTF validation failure if
			 * original variable took less than 4 bytes
			 */
			t->size = 1;
			*(int *)(t + 1) = BTF_INT_ENC(0, 0, 8);
		} else if (!has_datasec && btf_is_datasec(t)) {
			/* replace DATASEC with STRUCT */
			const struct btf_var_secinfo *v = btf_var_secinfos(t);
			struct btf_member *m = btf_members(t);
			struct btf_type *vt;
			char *name;

			name = (char *)btf__name_by_offset(btf, t->name_off);
			while (*name) {
				if (*name == '.')
					*name = '_';
				name++;
			}

			vlen = btf_vlen(t);
			t->info = BTF_INFO_ENC(BTF_KIND_STRUCT, 0, vlen);
			for (j = 0; j < vlen; j++, v++, m++) {
				/* order of field assignments is important */
				m->offset = v->offset * 8;
				m->type = v->type;
				/* preserve variable name as member name */
				vt = (void *)btf__type_by_id(btf, v->type);
				m->name_off = vt->name_off;
			}
		} else if (!has_func && btf_is_func_proto(t)) {
			/* replace FUNC_PROTO with ENUM */
			vlen = btf_vlen(t);
			t->info = BTF_INFO_ENC(BTF_KIND_ENUM, 0, vlen);
			t->size = sizeof(__u32); /* kernel enforced */
		} else if (!has_func && btf_is_func(t)) {
			/* replace FUNC with TYPEDEF */
			t->info = BTF_INFO_ENC(BTF_KIND_TYPEDEF, 0, 0);
		} else if (!has_func_global && btf_is_func(t)) {
			/* replace BTF_FUNC_GLOBAL with BTF_FUNC_STATIC */
			t->info = BTF_INFO_ENC(BTF_KIND_FUNC, 0, 0);
		} else if (!has_float && btf_is_float(t)) {
			/* replace FLOAT with an equally-sized empty STRUCT;
			 * since C compilers do not accept e.g. "float" as a
			 * valid struct name, make it anonymous
			 */
			t->name_off = 0;
			t->info = BTF_INFO_ENC(BTF_KIND_STRUCT, 0, 0);
		}
	}
}

static bool libbpf_needs_btf(const struct bpf_object *obj)
{
	return obj->efile.btf_maps_shndx >= 0 ||
	       obj->efile.st_ops_shndx >= 0 ||
	       obj->nr_extern > 0;
}

static bool kernel_needs_btf(const struct bpf_object *obj)
{
	return obj->efile.st_ops_shndx >= 0;
}

static int bpf_object__init_btf(struct bpf_object *obj,
				Elf_Data *btf_data,
				Elf_Data *btf_ext_data)
{
	int err = -ENOENT;

	if (btf_data) {
		obj->btf = btf__new(btf_data->d_buf, btf_data->d_size);
		err = libbpf_get_error(obj->btf);
		if (err) {
			obj->btf = NULL;
			pr_warn("Error loading ELF section %s: %d.\n", BTF_ELF_SEC, err);
			goto out;
		}
		/* enforce 8-byte pointers for BPF-targeted BTFs */
		btf__set_pointer_size(obj->btf, 8);
	}
	if (btf_ext_data) {
		if (!obj->btf) {
			pr_debug("Ignore ELF section %s because its depending ELF section %s is not found.\n",
				 BTF_EXT_ELF_SEC, BTF_ELF_SEC);
			goto out;
		}
		obj->btf_ext = btf_ext__new(btf_ext_data->d_buf, btf_ext_data->d_size);
		err = libbpf_get_error(obj->btf_ext);
		if (err) {
			pr_warn("Error loading ELF section %s: %d. Ignored and continue.\n",
				BTF_EXT_ELF_SEC, err);
			obj->btf_ext = NULL;
			goto out;
		}
	}
out:
	if (err && libbpf_needs_btf(obj)) {
		pr_warn("BTF is required, but is missing or corrupted.\n");
		return err;
	}
	return 0;
}

static int bpf_object__finalize_btf(struct bpf_object *obj)
{
	int err;

	if (!obj->btf)
		return 0;

	err = btf__finalize_data(obj, obj->btf);
	if (err) {
		pr_warn("Error finalizing %s: %d.\n", BTF_ELF_SEC, err);
		return err;
	}

	return 0;
}

static bool prog_needs_vmlinux_btf(struct bpf_program *prog)
{
	if (prog->type == BPF_PROG_TYPE_STRUCT_OPS ||
	    prog->type == BPF_PROG_TYPE_LSM)
		return true;

	/* BPF_PROG_TYPE_TRACING programs which do not attach to other programs
	 * also need vmlinux BTF
	 */
	if (prog->type == BPF_PROG_TYPE_TRACING && !prog->attach_prog_fd)
		return true;

	return false;
}

static bool obj_needs_vmlinux_btf(const struct bpf_object *obj)
{
	struct bpf_program *prog;
	int i;

	/* CO-RE relocations need kernel BTF */
	if (obj->btf_ext && obj->btf_ext->core_relo_info.len)
		return true;

	/* Support for typed ksyms needs kernel BTF */
	for (i = 0; i < obj->nr_extern; i++) {
		const struct extern_desc *ext;

		ext = &obj->externs[i];
		if (ext->type == EXT_KSYM && ext->ksym.type_id)
			return true;
	}

	bpf_object__for_each_program(prog, obj) {
		if (!prog->load)
			continue;
		if (prog_needs_vmlinux_btf(prog))
			return true;
	}

	return false;
}

static int bpf_object__load_vmlinux_btf(struct bpf_object *obj, bool force)
{
	int err;

	/* btf_vmlinux could be loaded earlier */
	if (obj->btf_vmlinux || obj->gen_loader)
		return 0;

	if (!force && !obj_needs_vmlinux_btf(obj))
		return 0;

	obj->btf_vmlinux = libbpf_find_kernel_btf();
	err = libbpf_get_error(obj->btf_vmlinux);
	if (err) {
		pr_warn("Error loading vmlinux BTF: %d\n", err);
		obj->btf_vmlinux = NULL;
		return err;
	}
	return 0;
}

static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj)
{
	struct btf *kern_btf = obj->btf;
	bool btf_mandatory, sanitize;
	int i, err = 0;

	if (!obj->btf)
		return 0;

	if (!kernel_supports(obj, FEAT_BTF)) {
		if (kernel_needs_btf(obj)) {
			err = -EOPNOTSUPP;
			goto report;
		}
		pr_debug("Kernel doesn't support BTF, skipping uploading it.\n");
		return 0;
	}

	/* Even though some subprogs are global/weak, user might prefer more
	 * permissive BPF verification process that BPF verifier performs for
	 * static functions, taking into account more context from the caller
	 * functions. In such case, they need to mark such subprogs with
	 * __attribute__((visibility("hidden"))) and libbpf will adjust
	 * corresponding FUNC BTF type to be marked as static and trigger more
	 * involved BPF verification process.
	 */
	for (i = 0; i < obj->nr_programs; i++) {
		struct bpf_program *prog = &obj->programs[i];
		struct btf_type *t;
		const char *name;
		int j, n;

		if (!prog->mark_btf_static || !prog_is_subprog(obj, prog))
			continue;

		n = btf__get_nr_types(obj->btf);
		for (j = 1; j <= n; j++) {
			t = btf_type_by_id(obj->btf, j);
			if (!btf_is_func(t) || btf_func_linkage(t) != BTF_FUNC_GLOBAL)
				continue;

			name = btf__str_by_offset(obj->btf, t->name_off);
			if (strcmp(name, prog->name) != 0)
				continue;

			t->info = btf_type_info(BTF_KIND_FUNC, BTF_FUNC_STATIC, 0);
			break;
		}
	}

	sanitize = btf_needs_sanitization(obj);
	if (sanitize) {
		const void *raw_data;
		__u32 sz;

		/* clone BTF to sanitize a copy and leave the original intact */
		raw_data = btf__get_raw_data(obj->btf, &sz);
		kern_btf = btf__new(raw_data, sz);
		err = libbpf_get_error(kern_btf);
		if (err)
			return err;

		/* enforce 8-byte pointers for BPF-targeted BTFs */
		btf__set_pointer_size(obj->btf, 8);
		bpf_object__sanitize_btf(obj, kern_btf);
	}

	if (obj->gen_loader) {
		__u32 raw_size = 0;
		const void *raw_data = btf__get_raw_data(kern_btf, &raw_size);

		if (!raw_data)
			return -ENOMEM;
		bpf_gen__load_btf(obj->gen_loader, raw_data, raw_size);
		/* Pretend to have valid FD to pass various fd >= 0 checks.
		 * This fd == 0 will not be used with any syscall and will be reset to -1 eventually.
		 */
		btf__set_fd(kern_btf, 0);
	} else {
		err = btf__load(kern_btf);
	}
	if (sanitize) {
		if (!err) {
			/* move fd to libbpf's BTF */
			btf__set_fd(obj->btf, btf__fd(kern_btf));
			btf__set_fd(kern_btf, -1);
		}
		btf__free(kern_btf);
	}
report:
	if (err) {
		btf_mandatory = kernel_needs_btf(obj);
		pr_warn("Error loading .BTF into kernel: %d. %s\n", err,
			btf_mandatory ? "BTF is mandatory, can't proceed."
				      : "BTF is optional, ignoring.");
		if (!btf_mandatory)
			err = 0;
	}
	return err;
}

static const char *elf_sym_str(const struct bpf_object *obj, size_t off)
{
	const char *name;

	name = elf_strptr(obj->efile.elf, obj->efile.strtabidx, off);
	if (!name) {
		pr_warn("elf: failed to get section name string at offset %zu from %s: %s\n",
			off, obj->path, elf_errmsg(-1));
		return NULL;
	}

	return name;
}

static const char *elf_sec_str(const struct bpf_object *obj, size_t off)
{
	const char *name;

	name = elf_strptr(obj->efile.elf, obj->efile.shstrndx, off);
	if (!name) {
		pr_warn("elf: failed to get section name string at offset %zu from %s: %s\n",
			off, obj->path, elf_errmsg(-1));
		return NULL;
	}

	return name;
}

static Elf_Scn *elf_sec_by_idx(const struct bpf_object *obj, size_t idx)
{
	Elf_Scn *scn;

	scn = elf_getscn(obj->efile.elf, idx);
	if (!scn) {
		pr_warn("elf: failed to get section(%zu) from %s: %s\n",
			idx, obj->path, elf_errmsg(-1));
		return NULL;
	}
	return scn;
}

static Elf_Scn *elf_sec_by_name(const struct bpf_object *obj, const char *name)
{
	Elf_Scn *scn = NULL;
	Elf *elf = obj->efile.elf;
	const char *sec_name;

	while ((scn = elf_nextscn(elf, scn)) != NULL) {
		sec_name = elf_sec_name(obj, scn);
		if (!sec_name)
			return NULL;

		if (strcmp(sec_name, name) != 0)
			continue;

		return scn;
	}
	return NULL;
}

static int elf_sec_hdr(const struct bpf_object *obj, Elf_Scn *scn, GElf_Shdr *hdr)
{
	if (!scn)
		return -EINVAL;

	if (gelf_getshdr(scn, hdr) != hdr) {
		pr_warn("elf: failed to get section(%zu) header from %s: %s\n",
			elf_ndxscn(scn), obj->path, elf_errmsg(-1));
		return -EINVAL;
	}

	return 0;
}

static const char *elf_sec_name(const struct bpf_object *obj, Elf_Scn *scn)
{
	const char *name;
	GElf_Shdr sh;

	if (!scn)
		return NULL;

	if (elf_sec_hdr(obj, scn, &sh))
		return NULL;

	name = elf_sec_str(obj, sh.sh_name);
	if (!name) {
		pr_warn("elf: failed to get section(%zu) name from %s: %s\n",
			elf_ndxscn(scn), obj->path, elf_errmsg(-1));
		return NULL;
	}

	return name;
}

static Elf_Data *elf_sec_data(const struct bpf_object *obj, Elf_Scn *scn)
{
	Elf_Data *data;

	if (!scn)
		return NULL;

	data = elf_getdata(scn, 0);
	if (!data) {
		pr_warn("elf: failed to get section(%zu) %s data from %s: %s\n",
			elf_ndxscn(scn), elf_sec_name(obj, scn) ?: "<?>",
			obj->path, elf_errmsg(-1));
		return NULL;
	}

	return data;
}

static bool is_sec_name_dwarf(const char *name)
{
	/* approximation, but the actual list is too long */
	return strncmp(name, ".debug_", sizeof(".debug_") - 1) == 0;
}

static bool ignore_elf_section(GElf_Shdr *hdr, const char *name)
{
	/* no special handling of .strtab */
	if (hdr->sh_type == SHT_STRTAB)
		return true;

	/* ignore .llvm_addrsig section as well */
	if (hdr->sh_type == SHT_LLVM_ADDRSIG)
		return true;

	/* no subprograms will lead to an empty .text section, ignore it */
	if (hdr->sh_type == SHT_PROGBITS && hdr->sh_size == 0 &&
	    strcmp(name, ".text") == 0)
		return true;

	/* DWARF sections */
	if (is_sec_name_dwarf(name))
		return true;

	if (strncmp(name, ".rel", sizeof(".rel") - 1) == 0) {
		name += sizeof(".rel") - 1;
		/* DWARF section relocations */
		if (is_sec_name_dwarf(name))
			return true;

		/* .BTF and .BTF.ext don't need relocations */
		if (strcmp(name, BTF_ELF_SEC) == 0 ||
		    strcmp(name, BTF_EXT_ELF_SEC) == 0)