diff options
Diffstat (limited to 'tools/lib/bpf/libbpf.c')
| -rw-r--r-- | tools/lib/bpf/libbpf.c | 2286 |
1 files changed, 1475 insertions, 811 deletions
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index e4f83c304ec9..a1bea1953df6 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -195,6 +195,8 @@ enum kern_feature_id { FEAT_BTF_FLOAT, /* BPF perf link support */ FEAT_PERF_LINK, + /* BTF_KIND_DECL_TAG support */ + FEAT_BTF_DECL_TAG, __FEAT_CNT, }; @@ -218,18 +220,40 @@ struct reloc_desc { struct bpf_sec_def; -typedef struct bpf_link *(*attach_fn_t)(const struct bpf_sec_def *sec, - struct bpf_program *prog); +typedef int (*init_fn_t)(struct bpf_program *prog, long cookie); +typedef int (*preload_fn_t)(struct bpf_program *prog, struct bpf_prog_load_params *attr, long cookie); +typedef struct bpf_link *(*attach_fn_t)(const struct bpf_program *prog, long cookie); + +/* stored as sec_def->cookie for all libbpf-supported SEC()s */ +enum sec_def_flags { + SEC_NONE = 0, + /* expected_attach_type is optional, if kernel doesn't support that */ + SEC_EXP_ATTACH_OPT = 1, + /* legacy, only used by libbpf_get_type_names() and + * libbpf_attach_type_by_name(), not used by libbpf itself at all. + * This used to be associated with cgroup (and few other) BPF programs + * that were attachable through BPF_PROG_ATTACH command. Pretty + * meaningless nowadays, though. + */ + SEC_ATTACHABLE = 2, + SEC_ATTACHABLE_OPT = SEC_ATTACHABLE | SEC_EXP_ATTACH_OPT, + /* attachment target is specified through BTF ID in either kernel or + * other BPF program's BTF object */ + SEC_ATTACH_BTF = 4, + /* BPF program type allows sleeping/blocking in kernel */ + SEC_SLEEPABLE = 8, + /* allow non-strict prefix matching */ + SEC_SLOPPY_PFX = 16, +}; 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; + long cookie; + + init_fn_t init_fn; + preload_fn_t preload_fn; attach_fn_t attach_fn; }; @@ -261,7 +285,7 @@ struct bpf_program { size_t sub_insn_off; char *name; - /* sec_name with / replaced by _; makes recursive pinning + /* name with / replaced by _; makes recursive pinning * in bpf_object__pin_programs easier */ char *pin_name; @@ -346,15 +370,14 @@ enum libbpf_map_type { 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; + /* real_name is defined for special internal maps (.rodata*, + * .data*, .bss, .kconfig) and preserves their original ELF section + * name. This is important to be be able to find corresponding BTF + * DATASEC information. + */ + char *real_name; int fd; int sec_idx; size_t sec_offset; @@ -377,6 +400,7 @@ struct bpf_map { char *pin_path; bool pinned; bool reused; + __u64 map_extra; }; enum extern_type { @@ -419,6 +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; }; }; @@ -430,6 +459,41 @@ struct module_btf { char *name; __u32 id; int fd; + int fd_array_idx; +}; + +enum sec_type { + SEC_UNUSED = 0, + SEC_RELO, + SEC_BSS, + SEC_DATA, + SEC_RODATA, +}; + +struct elf_sec_desc { + enum sec_type sec_type; + Elf64_Shdr *shdr; + Elf_Data *data; +}; + +struct elf_state { + int fd; + const void *obj_buf; + size_t obj_buf_sz; + Elf *elf; + Elf64_Ehdr *ehdr; + Elf_Data *symbols; + Elf_Data *st_ops_data; + size_t shstrndx; /* section index for section name strings */ + size_t strtabidx; + struct elf_sec_desc *secs; + int sec_cnt; + int maps_shndx; + int btf_maps_shndx; + __u32 btf_maps_sec_btf_id; + int text_shndx; + int symbols_shndx; + int st_ops_shndx; }; struct bpf_object { @@ -447,47 +511,17 @@ struct bpf_object { struct extern_desc *externs; int nr_extern; int kconfig_map_idx; - int rodata_map_idx; bool loaded; bool has_subcalls; + bool has_rodata; struct bpf_gen *gen_loader; + /* Information when doing ELF related work. Only valid if efile.elf is not NULL */ + struct elf_state efile; /* - * 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 + * All loaded bpf_object are linked in a list, which is * hidden to caller. bpf_objects__<func> handlers deal with * all objects. */ @@ -515,17 +549,22 @@ 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) 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 Elf64_Shdr *elf_sec_hdr(const struct bpf_object *obj, Elf_Scn *scn); 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); +static Elf64_Sym *elf_sym_by_idx(const struct bpf_object *obj, size_t idx); +static Elf64_Rel *elf_rel_by_idx(Elf_Data *data, size_t idx); void bpf_program__unload(struct bpf_program *prog) { @@ -580,7 +619,16 @@ static char *__bpf_program__pin_name(struct bpf_program *prog) { char *name, *p; - name = p = strdup(prog->sec_name); + if (libbpf_mode & LIBBPF_STRICT_SEC_NAME) + name = strdup(prog->name); + else + name = strdup(prog->sec_name); + + if (!name) + return NULL; + + p = name; + while ((p = strchr(p, '/'))) *p = '_'; @@ -667,25 +715,25 @@ bpf_object__add_programs(struct bpf_object *obj, Elf_Data *sec_data, 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; + Elf64_Sym *sym; progs = obj->programs; nr_progs = obj->nr_programs; - nr_syms = symbols->d_size / sizeof(GElf_Sym); + nr_syms = symbols->d_size / sizeof(Elf64_Sym); sec_off = 0; for (i = 0; i < nr_syms; i++) { - if (!gelf_getsym(symbols, i, &sym)) - continue; - if (sym.st_shndx != sec_idx) + sym = elf_sym_by_idx(obj, i); + + if (sym->st_shndx != sec_idx) continue; - if (GELF_ST_TYPE(sym.st_info) != STT_FUNC) + if (ELF64_ST_TYPE(sym->st_info) != STT_FUNC) continue; - prog_sz = sym.st_size; - sec_off = sym.st_value; + prog_sz = sym->st_size; + sec_off = sym->st_value; - name = elf_sym_str(obj, sym.st_name); + 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); @@ -698,7 +746,7 @@ bpf_object__add_programs(struct bpf_object *obj, Elf_Data *sec_data, return -LIBBPF_ERRNO__FORMAT; } - if (sec_idx != obj->efile.text_shndx && GELF_ST_BIND(sym.st_info) == STB_LOCAL) { + if (sec_idx != obj->efile.text_shndx && ELF64_ST_BIND(sym->st_info) == STB_LOCAL) { pr_warn("sec '%s': program '%s' is static and not supported\n", sec_name, name); return -ENOTSUP; } @@ -731,9 +779,9 @@ bpf_object__add_programs(struct bpf_object *obj, Elf_Data *sec_data, * 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)) + if (ELF64_ST_BIND(sym->st_info) != STB_LOCAL + && (ELF64_ST_VISIBILITY(sym->st_other) == STV_HIDDEN + || ELF64_ST_VISIBILITY(sym->st_other) == STV_INTERNAL)) prog->mark_btf_static = true; nr_progs++; @@ -1101,6 +1149,7 @@ static struct bpf_object *bpf_object__new(const char *path, size_t obj_buf_sz, const char *obj_name) { + bool strict = (libbpf_mode & LIBBPF_STRICT_NO_OBJECT_LIST); struct bpf_object *obj; char *end; @@ -1134,24 +1183,21 @@ static struct bpf_object *bpf_object__new(const char *path, 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); + if (!strict) + list_add(&obj->list, &bpf_objects_list); return obj; } static void bpf_object__elf_finish(struct bpf_object *obj) { - if (!obj_elf_valid(obj)) + if (!obj->efile.elf) return; if (obj->efile.elf) { @@ -1159,13 +1205,10 @@ static void bpf_object__elf_finish(struct bpf_object *obj) 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; + zfree(&obj->efile.secs); + obj->efile.sec_cnt = 0; zclose(obj->efile.fd); obj->efile.obj_buf = NULL; obj->efile.obj_buf_sz = 0; @@ -1173,10 +1216,11 @@ static void bpf_object__elf_finish(struct bpf_object *obj) static int bpf_object__elf_init(struct bpf_object *obj) { + Elf64_Ehdr *ehdr; int err = 0; - GElf_Ehdr *ep; + Elf *elf; - if (obj_elf_valid(obj)) { + if (obj->efile.elf) { pr_warn("elf: init internal error\n"); return -LIBBPF_ERRNO__LIBELF; } @@ -1186,10 +1230,9 @@ static int bpf_object__elf_init(struct bpf_object *obj) * 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); + elf = elf_memory((char *)obj->efile.obj_buf, obj->efile.obj_buf_sz); } else { - obj->efile.fd = open(obj->path, O_RDONLY); + obj->efile.fd = open(obj->path, O_RDONLY | O_CLOEXEC); if (obj->efile.fd < 0) { char errmsg[STRERR_BUFSIZE], *cp; @@ -1199,23 +1242,37 @@ static int bpf_object__elf_init(struct bpf_object *obj) return err; } - obj->efile.elf = elf_begin(obj->efile.fd, ELF_C_READ_MMAP, NULL); + elf = elf_begin(obj->efile.fd, ELF_C_READ_MMAP, NULL); } - if (!obj->efile.elf) { + if (!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)) { + obj->efile.elf = elf; + + if (elf_kind(elf) != ELF_K_ELF) { + err = -LIBBPF_ERRNO__FORMAT; + pr_warn("elf: '%s' is not a proper ELF object\n", obj->path); + goto errout; + } + + if (gelf_getclass(elf) != ELFCLASS64) { + err = -LIBBPF_ERRNO__FORMAT; + pr_warn("elf: '%s' is not a 64-bit ELF object\n", obj->path); + goto errout; + } + + obj->efile.ehdr = ehdr = elf64_getehdr(elf); + if (!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)) { + if (elf_getshdrstrndx(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; @@ -1223,7 +1280,7 @@ static int bpf_object__elf_init(struct bpf_object *obj) } /* Elf is corrupted/truncated, avoid calling elf_strptr. */ - if (!elf_rawdata(elf_getscn(obj->efile.elf, obj->efile.shstrndx), NULL)) { + if (!elf_rawdata(elf_getscn(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; @@ -1231,8 +1288,7 @@ static int bpf_object__elf_init(struct bpf_object *obj) } /* Old LLVM set e_machine to EM_NONE */ - if (ep->e_type != ET_REL || - (ep->e_machine && ep->e_machine != EM_BPF)) { + if (ehdr->e_type != ET_REL || (ehdr->e_machine && ehdr->e_machine != EM_BPF)) { pr_warn("elf: %s is not a valid eBPF object file\n", obj->path); err = -LIBBPF_ERRNO__FORMAT; goto errout; @@ -1246,11 +1302,11 @@ errout: static int bpf_object__check_endianness(struct bpf_object *obj) { -#if __BYTE_ORDER == __LITTLE_ENDIAN - if (obj->efile.ehdr.e_ident[EI_DATA] == ELFDATA2LSB) +#if __BYTE_ORDER__ == __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) +#elif __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__ + if (obj->efile.ehdr->e_ident[EI_DATA] == ELFDATA2MSB) return 0; #else # error "Unrecognized __BYTE_ORDER__" @@ -1290,41 +1346,27 @@ static bool bpf_map_type__is_map_in_map(enum bpf_map_type type) return false; } -int bpf_object__section_size(const struct bpf_object *obj, const char *name, - __u32 *size) +static int find_elf_sec_sz(const struct bpf_object *obj, const char *name, __u32 *size) { int ret = -ENOENT; + Elf_Data *data; + Elf_Scn *scn; *size = 0; - if (!name) { + 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; - } + scn = elf_sec_by_name(obj, name); + 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) +static int find_elf_var_offset(const struct bpf_object *obj, const char *name, __u32 *off) { Elf_Data *symbols = obj->efile.symbols; const char *sname; @@ -1333,23 +1375,20 @@ int bpf_object__variable_offset(const struct bpf_object *obj, const char *name, if (!name || !off) return -EINVAL; - for (si = 0; si < symbols->d_size / sizeof(GElf_Sym); si++) { - GElf_Sym sym; + for (si = 0; si < symbols->d_size / sizeof(Elf64_Sym); si++) { + Elf64_Sym *sym = elf_sym_by_idx(obj, si); - if (!gelf_getsym(symbols, si, &sym)) - continue; - if (GELF_ST_BIND(sym.st_info) != STB_GLOBAL || - GELF_ST_TYPE(sym.st_info) != STT_OBJECT) + if (ELF64_ST_BIND(sym->st_info) != STB_GLOBAL || + ELF64_ST_TYPE(sym->st_info) != STT_OBJECT) continue; - sname = elf_sym_str(obj, sym.st_name); + sname = elf_sym_str(obj, sym->st_name); if (!sname) { - pr_warn("failed to get sym name string for var %s\n", - name); + pr_warn("failed to get sym name string for var %s\n", name); return -EIO; } if (strcmp(name, sname) == 0) { - *off = sym.st_value; + *off = sym->st_value; return 0; } } @@ -1401,17 +1440,55 @@ static size_t bpf_map_mmap_sz(const struct bpf_map *map) return map_sz; } -static char *internal_map_name(struct bpf_object *obj, - enum libbpf_map_type type) +static char *internal_map_name(struct bpf_object *obj, const char *real_name) { 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)); + int pfx_len, sfx_len = max((size_t)7, strlen(real_name)); + + /* This is one of the more confusing parts of libbpf for various + * reasons, some of which are historical. The original idea for naming + * internal names was to include as much of BPF object name prefix as + * possible, so that it can be distinguished from similar internal + * maps of a different BPF object. + * As an example, let's say we have bpf_object named 'my_object_name' + * and internal map corresponding to '.rodata' ELF section. The final + * map name advertised to user and to the kernel will be + * 'my_objec.rodata', taking first 8 characters of object name and + * entire 7 characters of '.rodata'. + * Somewhat confusingly, if internal map ELF section name is shorter + * than 7 characters, e.g., '.bss', we still reserve 7 characters + * for the suffix, even though we only have 4 actual characters, and + * resulting map will be called 'my_objec.bss', not even using all 15 + * characters allowed by the kernel. Oh well, at least the truncated + * object name is somewhat consistent in this case. But if the map + * name is '.kconfig', we'll still have entirety of '.kconfig' added + * (8 chars) and thus will be left with only first 7 characters of the + * object name ('my_obje'). Happy guessing, user, that the final map + * name will be "my_obje.kconfig". + * Now, with libbpf starting to support arbitrarily named .rodata.* + * and .data.* data sections, it's possible that ELF section name is + * longer than allowed 15 chars, so we now need to be careful to take + * only up to 15 first characters of ELF name, taking no BPF object + * name characters at all. So '.rodata.abracadabra' will result in + * '.rodata.abracad' kernel and user-visible name. + * We need to keep this convoluted logic intact for .data, .bss and + * .rodata maps, but for new custom .data.custom and .rodata.custom + * maps we use their ELF names as is, not prepending bpf_object name + * in front. We still need to truncate them to 15 characters for the + * kernel. Full name can be recovered for such maps by using DATASEC + * BTF type associated with such map's value type, though. + */ + if (sfx_len >= BPF_OBJ_NAME_LEN) + sfx_len = BPF_OBJ_NAME_LEN - 1; + + /* if there are two or more dots in map name, it's a custom dot map */ + if (strchr(real_name + 1, '.') != NULL) + pfx_len = 0; + else + 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]); + sfx_len, real_name); /* sanitise map name to characters allowed by kernel */ for (p = map_name; *p && p < map_name + sizeof(map_name); p++) @@ -1423,7 +1500,7 @@ static char *internal_map_name(struct bpf_object *obj, 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) + const char *real_name, int sec_idx, void *data, size_t data_sz) { struct bpf_map_def *def; struct bpf_map *map; @@ -1436,9 +1513,11 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type, 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"); + map->real_name = strdup(real_name); + map->name = internal_map_name(obj, real_name); + if (!map->real_name || !map->name) { + zfree(&map->real_name); + zfree(&map->name); return -ENOMEM; } @@ -1461,6 +1540,7 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type, map->mmaped = NULL; pr_warn("failed to alloc map '%s' content buffer: %d\n", map->name, err); + zfree(&map->real_name); zfree(&map->name); return err; } @@ -1474,34 +1554,43 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type, static int bpf_object__init_global_data_maps(struct bpf_object *obj) { - int err; + struct elf_sec_desc *sec_desc; + const char *sec_name; + int err = 0, sec_idx; /* * 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); + for (sec_idx = 1; sec_idx < obj->efile.sec_cnt; sec_idx++) { + sec_desc = &obj->efile.secs[sec_idx]; + + switch (sec_desc->sec_type) { + case SEC_DATA: + sec_name = elf_sec_name(obj, elf_sec_by_idx(obj, sec_idx)); + err = bpf_object__init_internal_map(obj, LIBBPF_MAP_DATA, + sec_name, sec_idx, + sec_desc->data->d_buf, + sec_desc->data->d_size); + break; + case SEC_RODATA: + obj->has_rodata = true; + sec_name = elf_sec_name(obj, elf_sec_by_idx(obj, sec_idx)); + err = bpf_object__init_internal_map(obj, LIBBPF_MAP_RODATA, + sec_name, sec_idx, + sec_desc->data->d_buf, + sec_desc->data->d_size); + break; + case SEC_BSS: + sec_name = elf_sec_name(obj, elf_sec_by_idx(obj, sec_idx)); + err = bpf_object__init_internal_map(obj, LIBBPF_MAP_BSS, + sec_name, sec_idx, + NULL, + sec_desc->data->d_size); + break; + default: + /* skip */ + break; + } if (err) return err; } @@ -1664,7 +1753,7 @@ static int bpf_object__process_kconfig_line(struct bpf_object *obj, void *ext_val; __u64 num; - if (strncmp(buf, "CONFIG_", 7)) + if (!str_has_pfx(buf, "CONFIG_")) return 0; sep = strchr(buf, '='); @@ -1798,7 +1887,7 @@ static int bpf_object__init_kconfig_map(struct bpf_object *obj) 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, + ".kconfig", obj->efile.symbols_shndx, NULL, map_sz); if (err) return err; @@ -1836,13 +1925,13 @@ static int bpf_object__init_user_maps(struct bpf_object *obj, bool strict) * * TODO: Detect array of map and report error. */ - nr_syms = symbols->d_size / sizeof(GElf_Sym); + nr_syms = symbols->d_size / sizeof(Elf64_Sym); for (i = 0; i < nr_syms; i++) { - GElf_Sym sym; + Elf64_Sym *sym = elf_sym_by_idx(obj, i); - if (!gelf_getsym(symbols, i, &sym)) + if (sym->st_shndx != obj->efile.maps_shndx) continue; - if (sym.st_shndx != obj->efile.maps_shndx) + if (ELF64_ST_TYPE(sym->st_info) == STT_SECTION) continue; nr_maps++; } @@ -1859,39 +1948,38 @@ static int bpf_object__init_user_maps(struct bpf_object *obj, bool strict) /* Fill obj->maps using data in "maps" section. */ for (i = 0; i < nr_syms; i++) { - GElf_Sym sym; + Elf64_Sym *sym = elf_sym_by_idx(obj, i); const char *map_name; struct bpf_map_def *def; struct bpf_map *map; - if (!gelf_getsym(symbols, i, &sym)) + if (sym->st_shndx != obj->efile.maps_shndx) continue; - if (sym.st_shndx != obj->efile.maps_shndx) + if (ELF64_ST_TYPE(sym->st_info) == STT_SECTION) continue; map = bpf_object__add_map(obj); if (IS_ERR(map)) return PTR_ERR(map); - map_name = elf_sym_str(obj, sym.st_name); + 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) { + if (ELF64_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; + 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) { + 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; @@ -1899,11 +1987,11 @@ static int bpf_object__init_user_maps(struct bpf_object *obj, bool strict) map->name = strdup(map_name); if (!map->name) { - pr_warn("failed to alloc map name\n"); + pr_warn("map '%s': failed to alloc map name\n", map_name); return -ENOMEM; } pr_debug("map %d is \"%s\"\n", i, map->name); - def = (struct bpf_map_def *)(data->d_buf + sym.st_value); + 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 @@ -1987,6 +2075,7 @@ static const char *__btf_kind_str(__u16 kind) case BTF_KIND_VAR: return "var"; case BTF_KIND_DATASEC: return "datasec"; case BTF_KIND_FLOAT: return "float"; + case BTF_KIND_DECL_TAG: return "decl_tag"; default: return "unknown"; } } @@ -2236,6 +2325,13 @@ int parse_btf_map_def(const char *map_name, struct btf *btf, } map_def->pinning = val; map_def->parts |= MAP_DEF_PINNING; + } else if (strcmp(name, "map_extra") == 0) { + __u32 map_extra; + + if (!get_map_field_int(map_name, btf, m, &map_extra)) + return -EINVAL; + map_def->map_extra = map_extra; + map_def->parts |= MAP_DEF_MAP_EXTRA; } else { if (strict) { pr_warn("map '%s': unknown field '%s'.\n", map_name, name); @@ -2260,6 +2356,7 @@ static void fill_map_from_def(struct bpf_map *map, const struct btf_map_def *def map->def.value_size = def->value_size; map->def.max_entries = def->max_entries; map->def.map_flags = def->map_flags; + map->map_extra = def->map_extra; map->numa_node = def->numa_node; map->btf_key_type_id = def->key_type_id; @@ -2283,7 +2380,10 @@ static void fill_map_from_def(struct bpf_map *map, const struct btf_map_def *def 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); + pr_debug("map '%s': found map_flags = 0x%x.\n", map->name, def->map_flags); + if (def->parts & MAP_DEF_MAP_EXTRA) + pr_debug("map '%s': found map_extra = 0x%llx.\n", map->name, + (unsigned long long)def->map_extra); 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) @@ -2420,8 +2520,8 @@ static int bpf_object__init_user_btf_maps(struct bpf_object *obj, bool strict, return -EINVAL; } - nr_types = btf__get_nr_types(obj->btf); - for (i = 1; i <= nr_types; i++) { + nr_types = btf__type_cnt(obj->btf); + for (i = 1; i < nr_types; i++) { t = btf__type_by_id(obj->btf, i); if (!btf_is_datasec(t)) continue; @@ -2472,12 +2572,13 @@ static int bpf_object__init_maps(struct bpf_object *obj, static bool section_have_execinstr(struct bpf_object *obj, int idx) { - GElf_Shdr sh; + Elf64_Shdr *sh; - if (elf_sec_hdr(obj, elf_sec_by_idx(obj, idx), &sh)) + sh = elf_sec_hdr(obj, elf_sec_by_idx(obj, idx)); + if (!sh) return false; - return sh.sh_flags & SHF_EXECINSTR; + return sh->sh_flags & SHF_EXECINSTR; } static bool btf_needs_sanitization(struct bpf_object *obj) @@ -2486,8 +2587,9 @@ static bool btf_needs_sanitization(struct bpf_object *obj) 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); + bool has_decl_tag = kernel_supports(obj, FEAT_BTF_DECL_TAG); - return !has_func || !has_datasec || !has_func_global || !has_float; + return !has_func || !has_datasec || !has_func_global || !has_float || !has_decl_tag; } static void bpf_object__sanitize_btf(struct bpf_object *obj, struct btf *btf) @@ -2496,14 +2598,15 @@ static void bpf_object__sanitize_btf(struct bpf_object *obj, struct btf *btf) 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); + bool has_decl_tag = kernel_supports(obj, FEAT_BTF_DECL_TAG); struct btf_type *t; int i, j, vlen; - for (i = 1; i <= btf__get_nr_types(btf); i++) { + for (i = 1; i < btf__type_cnt(btf); i++) { t = (struct btf_type *)btf__type_by_id(btf, i); - if (!has_datasec && btf_is_var(t)) { - /* replace VAR with INT */ + if ((!has_datasec && btf_is_var(t)) || (!has_decl_tag && btf_is_decl_tag(t))) { + /* replace VAR/DECL_TAG with INT */ t->info = BTF_INFO_ENC(BTF_KIND_INT, 0, 0); /* * using size = 1 is the safest choice, 4 will be too @@ -2610,6 +2713,104 @@ out: return 0; } +static int compare_vsi_off(const void *_a, const void *_b) +{ + const struct btf_var_secinfo *a = _a; + const struct btf_var_secinfo *b = _b; + + return a->offset - b->offset; +} + +static int btf_fixup_datasec(struct bpf_object *obj, struct btf *btf, + struct btf_type *t) +{ + __u32 size = 0, off = 0, i, vars = btf_vlen(t); + const char *name = btf__name_by_offset(btf, t->name_off); + const struct btf_type *t_var; + struct btf_var_secinfo *vsi; + const struct btf_var *var; + int ret; + + if (!name) { + pr_debug("No name found in string section for DATASEC kind.\n"); + return -ENOENT; + } + + /* .extern datasec size and var offsets were set correctly during + * extern collection step, so just skip straight to sorting variables + */ + if (t->size) + goto sort_vars; + + ret = find_elf_sec_sz(obj, name, &size); + if (ret || !size || (t->size && t->size != size)) { + pr_debug("Invalid size for section %s: %u bytes\n", name, size); + return -ENOENT; + } + + t->size = size; + + for (i = 0, vsi = btf_var_secinfos(t); i < vars; i++, vsi++) { + t_var = btf__type_by_id(btf, vsi->type); + var = btf_var(t_var); + + if (!btf_is_var(t_var)) { + pr_debug("Non-VAR type seen in section %s\n", name); + return -EINVAL; + } + + if (var->linkage == BTF_VAR_STATIC) + continue; + + name = btf__name_by_offset(btf, t_var->name_off); + if (!name) { + pr_debug("No name found in string section for VAR kind\n"); + return -ENOENT; + } + + ret = find_elf_var_offset(obj, name, &off); + if (ret) { + pr_debug("No offset found in symbol table for VAR %s\n", + name); + return -ENOENT; + } + + vsi->offset = off; + } + +sort_vars: + qsort(btf_var_secinfos(t), vars, sizeof(*vsi), compare_vsi_off); + return 0; +} + +static int btf_finalize_data(struct bpf_object *obj, struct btf *btf) +{ + int err = 0; + __u32 i, n = btf__type_cnt(btf); + + for (i = 1; i < n; i++) { + struct btf_type *t = btf_type_by_id(btf, i); + + /* Loader needs to fix up some of the things compiler + * couldn't get its hands on while emitting BTF. This + * is section size and global variable offset. We use + * the info from the ELF itself for this purpose. + */ + if (btf_is_datasec(t)) { + err = btf_fixup_datasec(obj, btf, t); + if (err) + break; + } + } + + return libbpf_err(err); +} + +int btf__finalize_data(struct bpf_object *obj, struct btf *btf) +{ + return btf_finalize_data(obj, btf); +} + static int bpf_object__finalize_btf(struct bpf_object *obj) { int err; @@ -2617,7 +2818,7 @@ static int bpf_object__finalize_btf(struct bpf_object *obj) if (!obj->btf) return 0; - err = btf__finalize_data(obj, obj->btf); + err = btf_finalize_data(obj, obj->btf); if (err) { pr_warn("Error finalizing %s: %d.\n", BTF_ELF_SEC, err); return err; @@ -2727,8 +2928,8 @@ static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj) if (!prog->mark_btf_static || !prog_is_subprog(obj, prog)) continue; - n = btf__get_nr_types(obj->btf); - for (j = 1; j <= n; j++) { + n = btf__type_cnt(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; @@ -2748,7 +2949,7 @@ static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj) __u32 sz; /* clone BTF to sanitize a copy and leave the original intact */ - raw_data = btf__get_raw_data(obj->btf, &sz); + raw_data = btf__raw_data(obj->btf, &sz); kern_btf = btf__new(raw_data, sz); err = libbpf_get_error(kern_btf); if (err) @@ -2761,7 +2962,7 @@ static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj) if (obj->gen_loader) { __u32 raw_size = 0; - const void *raw_data = btf__get_raw_data(kern_btf, &raw_size); + const void *raw_data = btf__raw_data(kern_btf, &raw_size); if (!raw_data) return -ENOMEM; @@ -2853,32 +3054,36 @@ static Elf_Scn *elf_sec_by_name(const struct bpf_object *obj, const char *name) return NULL; } -static int elf_sec_hdr(const struct bpf_object *obj, Elf_Scn *scn, GElf_Shdr *hdr) +static Elf64_Shdr *elf_sec_hdr(const struct bpf_object *obj, Elf_Scn *scn) { + Elf64_Shdr *shdr; + if (!scn) - return -EINVAL; + return NULL; - if (gelf_getshdr(scn, hdr) != hdr) { + shdr = elf64_getshdr(scn); + if (!shdr) { pr_warn("elf: failed to get section(%zu) header from %s: %s\n", elf_ndxscn(scn), obj->path, elf_errmsg(-1)); - return -EINVAL; + return NULL; } - return 0; + return shdr; } static const char *elf_sec_name(const struct bpf_object *obj, Elf_Scn *scn) { const char *name; - GElf_Shdr sh; + Elf64_Shdr *sh; if (!scn) return NULL; - if (elf_sec_hdr(obj, scn, &sh)) + sh = elf_sec_hdr(obj, scn); + if (!sh) return NULL; - name = elf_sec_str(obj, sh.sh_name); + 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)); @@ -2906,13 +3111,29 @@ static Elf_Data *elf_sec_data(const struct bpf_object *obj, Elf_Scn *scn) return data; } +static Elf64_Sym *elf_sym_by_idx(const struct bpf_object *obj, size_t idx) +{ + if (idx >= obj->efile.symbols->d_size / sizeof(Elf64_Sym)) + return NULL; + + return (Elf64_Sym *)obj->efile.symbols->d_buf + idx; +} + +static Elf64_Rel *elf_rel_by_idx(Elf_Data *data, size_t idx) +{ + if (idx >= data->d_size / sizeof(Elf64_Rel)) + return NULL; + + return (Elf64_Rel *)data->d_buf + idx; +} + 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; + return str_has_pfx(name, ".debug_"); } -static bool ignore_elf_section(GElf_Shdr *hdr, const char *name) +static bool ignore_elf_section(Elf64_Shdr *hdr, const char *name) { /* no special handling of .strtab */ if (hdr->sh_type == SHT_STRTAB) @@ -2931,7 +3152,7 @@ static bool ignore_elf_section(GElf_Shdr *hdr, const char *name) if (is_sec_name_dwarf(name)) return true; - if (strncmp(name, ".rel", sizeof(".rel") - 1) == 0) { + if (str_has_pfx(name, ".rel")) { name += sizeof(".rel") - 1; /* DWARF section relocations */ if (is_sec_name_dwarf(name)) @@ -2960,6 +3181,7 @@ static int cmp_progs(const void *_a, const void *_b) static int bpf_object__elf_collect(struct bpf_object *obj) { + struct elf_sec_desc *sec_desc; Elf *elf = obj->efile.elf; Elf_Data *btf_ext_data = NULL; Elf_Data *btf_data = NULL; @@ -2967,17 +3189,27 @@ static int bpf_object__elf_collect(struct bpf_object *obj) const char *name; Elf_Data *data; Elf_Scn *scn; - GElf_Shdr sh; + Elf64_Shdr *sh; + + /* ELF section indices are 1-based, so allocate +1 element to keep + * indexing simple. Also include 0th invalid section into sec_cnt for + * simpler and more traditional iteration logic. + */ + obj->efile.sec_cnt = 1 + obj->efile.ehdr->e_shnum; + obj->efile.secs = calloc(obj->efile.sec_cnt, sizeof(*obj->efile.secs)); + if (!obj->efile.secs) + return -ENOMEM; /* a bunch of ELF parsing functionality depends on processing symbols, * so do the first pass and find the symbol table */ scn = NULL; while ((scn = elf_nextscn(elf, scn)) != NULL) { - if (elf_sec_hdr(obj, scn, &sh)) + sh = elf_sec_hdr(obj, scn); + if (!sh) return -LIBBPF_ERRNO__FORMAT; - if (sh.sh_type == SHT_SYMTAB) { + if (sh->sh_type == SHT_SYMTAB) { if (obj->efile.symbols) { pr_warn("elf: multiple symbol tables in %s\n", obj->path); return -LIBBPF_ERRNO__FORMAT; @@ -2987,24 +3219,34 @@ static int bpf_object__elf_collect(struct bpf_object *obj) if (!data) return -LIBBPF_ERRNO__FORMAT; + idx = elf_ndxscn(scn); + obj->efile.symbols = data; - obj->efile.symbols_shndx = elf_ndxscn(scn); - obj->efile.strtabidx = sh.sh_link; + obj->efile.symbols_shndx = idx; + obj->efile.strtabidx = sh->sh_link; } } + if (!obj->efile.symbols) { + pr_warn("elf: couldn't find symbol table in %s, stripped object file?\n", + obj->path); + return -ENOENT; + } + scn = NULL; while ((scn = elf_nextscn(elf, scn)) != NULL) { - idx++; + idx = elf_ndxscn(scn); + sec_desc = &obj->efile.secs[idx]; - if (elf_sec_hdr(obj, scn, &sh)) + sh = elf_sec_hdr(obj, scn); + if (!sh) return -LIBBPF_ERRNO__FORMAT; - name = elf_sec_str(obj, sh.sh_name); + name = elf_sec_str(obj, sh->sh_name); if (!name) return -LIBBPF_ERRNO__FORMAT; - if (ignore_elf_section(&sh, name)) + if (ignore_elf_section(sh, name)) continue; data = elf_sec_data(obj, scn); @@ -3013,8 +3255,8 @@ static int bpf_object__elf_collect(struct bpf_object *obj) pr_debug("elf: section(%d) %s, size %ld, link %d, flags %lx, type=%d\n", idx, name, (unsigned long)data->d_size, - (int)sh.sh_link, (unsigned long)sh.sh_flags, - (int)sh.sh_type); + (int)sh->sh_link, (unsigned long)sh->sh_flags, + (int)sh->sh_type); if (strcmp(name, "license") == 0) { err = bpf_object__init_license(obj, data->d_buf, data->d_size); @@ -3032,21 +3274,25 @@ static int bpf_object__elf_collect(struct bpf_object *obj) btf_data = data; } else if (strcmp(name, BTF_EXT_ELF_SEC) == 0) { btf_ext_data = data; - } else if (sh.sh_type == SHT_SYMTAB) { + } else if (sh->sh_type == SHT_SYMTAB) { /* already processed during the first pass above */ - } else if (sh.sh_type == SHT_PROGBITS && data->d_size > 0) { - if (sh.sh_flags & SHF_EXECINSTR) { + } else if (sh->sh_type == SHT_PROGBITS && data->d_size > 0) { + if (sh->sh_flags & SHF_EXECINSTR) { if (strcmp(name, ".text") == 0) obj->efile.text_shndx = idx; err = bpf_object__add_programs(obj, data, name, idx); if (err) return err; - } else if (strcmp(name, DATA_SEC) == 0) { - obj->efile.data = data; - obj->efile.data_shndx = idx; - } else if (strcmp(name, RODATA_SEC) == 0) { - obj->efile.rodata = data; - obj->efile.rodata_shndx = idx; + } else if (strcmp(name, DATA_SEC) == 0 || + str_has_pfx(name, DATA_SEC ".")) { + sec_desc->sec_type = SEC_DATA; + sec_desc->shdr = sh; + sec_desc->data = data; + } else if (strcmp(name, RODATA_SEC) == 0 || + str_has_pfx(name, RODATA_SEC ".")) { + sec_desc->sec_type = SEC_RODATA; + sec_desc->shdr = sh; + sec_desc->data = data; } else if (strcmp(name, STRUCT_OPS_SEC) == 0) { obj->efile.st_ops_data = data; obj->efile.st_ops_shndx = idx; @@ -3054,37 +3300,29 @@ static int bpf_object__elf_collect(struct bpf_object *obj) pr_info("elf: skipping unrecognized data section(%d) %s\n", idx, name); } - } else if (sh.sh_type == SHT_REL) { - int nr_sects = obj->efile.nr_reloc_sects; - void *sects = obj->efile.reloc_sects; - int sec = sh.sh_info; /* points to other section */ + } else if (sh->sh_type == SHT_REL) { + int targ_sec_idx = sh->sh_info; /* points to other section */ /* Only do relo for section with exec instructions */ - if (!section_have_execinstr(obj, sec) && + if (!section_have_execinstr(obj, targ_sec_idx) && strcmp(name, ".rel" STRUCT_OPS_SEC) && strcmp(name, ".rel" MAPS_ELF_SEC)) { pr_info("elf: skipping relo section(%d) %s for section(%d) %s\n", - idx, name, sec, - elf_sec_name(obj, elf_sec_by_idx(obj, sec)) ?: "<?>"); + idx, name, targ_sec_idx, + elf_sec_name(obj, elf_sec_by_idx(obj, targ_sec_idx)) ?: "<?>"); continue; } - sects = libbpf_reallocarray(sects, nr_sects + 1, - sizeof(*obj->efile.reloc_sects)); - if (!sects) - return -ENOMEM; - - obj->efile.reloc_sects = sects; - obj->efile.nr_reloc_sects++; - - obj->efile.reloc_sects[nr_sects].shdr = sh; - obj->efile.reloc_sects[nr_sects].data = data; - } else if (sh.sh_type == SHT_NOBITS && strcmp(name, BSS_SEC) == 0) { - obj->efile.bss = data; - obj->efile.bss_shndx = idx; + sec_desc->sec_type = SEC_RELO; + sec_desc->shdr = sh; + sec_desc->data = data; + } else if (sh->sh_type == SHT_NOBITS && strcmp(name, BSS_SEC) == 0) { + sec_desc->sec_type = SEC_BSS; + sec_desc->shdr = sh; + sec_desc->data = data; } else { pr_info("elf: skipping section(%d) %s (size %zu)\n", idx, name, - (size_t)sh.sh_size); + (size_t)sh->sh_size); } } @@ -3100,19 +3338,19 @@ static int bpf_object__elf_collect(struct bpf_object *obj) return bpf_object__init_btf(obj, btf_data, btf_ext_data); } -static bool sym_is_extern(const GElf_Sym *sym) +static bool sym_is_extern(const Elf64_Sym *sym) { - int bind = GELF_ST_BIND(sym->st_info); + int bind = ELF64_ST_BIND(sym->st_info); /* externs are symbols w/ type=NOTYPE, bind=GLOBAL|WEAK, section=UND */ return sym->st_shndx == SHN_UNDEF && (bind == STB_GLOBAL || bind == STB_WEAK) && - GELF_ST_TYPE(sym->st_info) == STT_NOTYPE; + ELF64_ST_TYPE(sym->st_info) == STT_NOTYPE; } -static bool sym_is_subprog(const GElf_Sym *sym, int text_shndx) +static bool sym_is_subprog(const Elf64_Sym *sym, int text_shndx) { - int bind = GELF_ST_BIND(sym->st_info); - int type = GELF_ST_TYPE(sym->st_info); + int bind = ELF64_ST_BIND(sym->st_info); + int type = ELF64_ST_TYPE(sym->st_info); /* in .text section */ if (sym->st_shndx != text_shndx) @@ -3135,8 +3373,8 @@ static int find_extern_btf_id(const struct btf *btf, const char *ext_name) if (!btf) return -ESRCH; - n = btf__get_nr_types(btf); - for (i = 1; i <= n; i++) { + n = btf__type_cnt(btf); + for (i = 1; i < n; i++) { t = btf__type_by_id(btf, i); if (!btf_is_var(t) && !btf_is_func(t)) @@ -3167,8 +3405,8 @@ static int find_extern_sec_btf_id(struct btf *btf, int ext_btf_id) { if (!btf) return -ESRCH; - n = btf__get_nr_types(btf); - for (i = 1; i <= n; i++) { + n = btf__type_cnt(btf); + for (i = 1; i < n; i++) { t = btf__type_by_id(btf, i); if (!btf_is_datasec(t)) @@ -3252,8 +3490,8 @@ static int find_int_btf_id(const struct btf *btf) const struct btf_type *t; int i, n; - n = btf__get_nr_types(btf); - for (i = 1; i <= n; i++) { + n = btf__type_cnt(btf); + for (i = 1; i < n; i++) { t = btf__type_by_id(btf, i); if (btf_is_int(t) && btf_int_bits(t) == 32) @@ -3310,30 +3548,31 @@ static int bpf_object__collect_externs(struct bpf_object *obj) int i, n, off, dummy_var_btf_id; const char *ext_name, *sec_name; Elf_Scn *scn; - GElf_Shdr sh; + Elf64_Shdr *sh; if (!obj->efile.symbols) return 0; scn = elf_sec_by_idx(obj, obj->efile.symbols_shndx); - if (elf_sec_hdr(obj, scn, &sh)) + sh = elf_sec_hdr(obj, scn); + if (!sh) return -LIBBPF_ERRNO__FORMAT; dummy_var_btf_id = add_dummy_ksym_var(obj->btf); if (dummy_var_btf_id < 0) return dummy_var_btf_id; - n = sh.sh_size / sh.sh_entsize; + n = sh->sh_size / sh->sh_entsize; pr_debug("looking for externs among %d symbols...\n", n); for (i = 0; i < n; i++) { - GElf_Sym sym; + Elf64_Sym *sym = elf_sym_by_idx(obj, i); - if (!gelf_getsym(obj->efile.symbols, i, &sym)) + if (!sym) return -LIBBPF_ERRNO__FORMAT; - if (!sym_is_extern(&sym)) + if (!sym_is_extern(sym)) continue; - ext_name = elf_sym_str(obj, sym.st_name); + ext_name = elf_sym_str(obj, sym->st_name); if (!ext_name || !ext_name[0]) continue; @@ -3355,7 +3594,7 @@ static int bpf_object__collect_externs(struct bpf_object *obj) t = btf__type_by_id(obj->btf, ext->btf_id); ext->name = btf__name_by_offset(obj->btf, t->name_off); ext->sym_idx = i; - ext->is_weak = GELF_ST_BIND(sym.st_info) == STB_WEAK; + ext->is_weak = ELF64_ST_BIND(sym->st_info) == STB_WEAK; ext->sec_btf_id = find_extern_sec_btf_id(obj->btf, ext->btf_id); if (ext->sec_btf_id <= 0) { @@ -3393,11 +3632,6 @@ static int bpf_object__collect_externs(struct bpf_object *obj) return -ENOTSUP; } } else if (strcmp(sec_name, KSYMS_SEC) == 0) { - if (btf_is_func(t) && ext->is_weak) { - pr_warn("extern weak function %s is unsupported\n", - ext->name); - return -ENOTSUP; - } ksym_sec = sec; ext->type = EXT_KSYM; skip_mods_and_typedefs(obj->btf, t->type, @@ -3565,9 +3799,14 @@ bpf_object__find_program_by_name(const struct bpf_object *obj, static bool bpf_object__shndx_is_data(const struct bpf_object *obj, int shndx) { - return shndx == obj->efile.data_shndx || - shndx == obj->efile.bss_shndx || - shndx == obj->efile.rodata_shndx; + switch (obj->efile.secs[shndx].sec_type) { + case SEC_BSS: + case SEC_DATA: + case SEC_RODATA: + return true; + default: + return false; + } } static bool bpf_object__shndx_is_maps(const struct bpf_object *obj, @@ -3580,22 +3819,25 @@ static bool bpf_object__shndx_is_maps(const struct bpf_object *obj, static enum libbpf_map_type bpf_object__section_to_libbpf_map_type(const struct bpf_object *obj, int shndx) { - if (shndx == obj->efile.data_shndx) - return LIBBPF_MAP_DATA; - else if (shndx == obj->efile.bss_shndx) + if (shndx == obj->efile.symbols_shndx) + return LIBBPF_MAP_KCONFIG; + + switch (obj->efile.secs[shndx].sec_type) { + case SEC_BSS: return LIBBPF_MAP_BSS; - else if (shndx == obj->efile.rodata_shndx) + case SEC_DATA: + return LIBBPF_MAP_DATA; + case SEC_RODATA: return LIBBPF_MAP_RODATA; - else if (shndx == obj->efile.symbols_shndx) - return LIBBPF_MAP_KCONFIG; - else + default: return LIBBPF_MAP_UNSPEC; + } } static int bpf_program__record_reloc(struct bpf_program *prog, struct reloc_desc *reloc_desc, __u32 insn_idx, const char *sym_name, - const GElf_Sym *sym, const GElf_Rel *rel) + const Elf64_Sym *sym, const Elf64_Rel *rel) { struct bpf_insn *insn = &prog->insns[insn_idx]; size_t map_idx, nr_maps = prog->obj->nr_maps; @@ -3612,7 +3854,7 @@ static int bpf_program__record_reloc(struct bpf_program *prog, } if (sym_is_extern(sym)) { - int sym_idx = GELF_R_SYM(rel->r_info); + int sym_idx = ELF64_R_SYM(rel->r_info); int i, n = obj->nr_extern; struct extern_desc *ext; @@ -3725,7 +3967,7 @@ static int bpf_program__record_reloc(struct bpf_program *prog, } for (map_idx = 0; map_idx < nr_maps; map_idx++) { map = &obj->maps[map_idx]; - if (map->libbpf_type != type) + if (map->libbpf_type != type || map->sec_idx != sym->st_shndx) continue; pr_debug("prog '%s': found data map %zd (%s, sec %d, off %zu) for insn %u\n", prog->name, map_idx, map->name, map->sec_idx, @@ -3777,9 +4019,8 @@ static struct bpf_program *find_prog_by_sec_insn(const struct bpf_object *obj, } static int -bpf_object__collect_prog_relos(struct bpf_object *obj, GElf_Shdr *shdr, Elf_Data *data) +bpf_object__collect_prog_relos(struct bpf_object *obj, Elf64_Shdr *shdr, Elf_Data *data) { - Elf_Data *symbols = obj->efile.symbols; const char *relo_sec_name, *sec_name; size_t sec_idx = shdr->sh_info; struct bpf_program *prog; @@ -3789,8 +4030,8 @@ bpf_object__collect_prog_relos(struct bpf_object *obj, GElf_Shdr *shdr, Elf_Data __u32 insn_idx; Elf_Scn *scn; Elf_Data *scn_data; - GElf_Sym sym; - GElf_Rel rel; + Elf64_Sym *sym; + Elf64_Rel *rel; scn = elf_sec_by_idx(obj, sec_idx); scn_data = elf_sec_data(obj, scn); @@ -3805,33 +4046,36 @@ bpf_object__collect_prog_relos(struct bpf_object *obj, GElf_Shdr *shdr, Elf_Data nrels = shdr->sh_size / shdr->sh_entsize; for (i = 0; i < nrels; i++) { - if (!gelf_getrel(data, i, &rel)) { + rel = elf_rel_by_idx(data, i); + if (!rel) { pr_warn("sec '%s': failed to get relo #%d\n", relo_sec_name, i); return -LIBBPF_ERRNO__FORMAT; } - if (!gelf_getsym(symbols, GELF_R_SYM(rel.r_info), &sym)) { + + sym = elf_sym_by_idx(obj, ELF64_R_SYM(rel->r_info)); + if (!sym) { pr_warn("sec '%s': symbol 0x%zx not found for relo #%d\n", - relo_sec_name, (size_t)GELF_R_SYM(rel.r_info), i); + relo_sec_name, (size_t)ELF64_R_SYM(rel->r_info), i); return -LIBBPF_ERRNO__FORMAT; } - if (rel.r_offset % BPF_INSN_SZ || rel.r_offset >= scn_data->d_size) { + if (rel->r_offset % BPF_INSN_SZ || rel->r_offset >= scn_data->d_size) { pr_warn("sec '%s': invalid offset 0x%zx for relo #%d\n", - relo_sec_name, (size_t)GELF_R_SYM(rel.r_info), i); + relo_sec_name, (size_t)ELF64_R_SYM(rel->r_info), i); return -LIBBPF_ERRNO__FORMAT; } - insn_idx = rel.r_offset / BPF_INSN_SZ; + insn_idx = rel->r_offset / BPF_INSN_SZ; /* relocations against static functions are recorded as * relocations against the section that contains a function; * in such case, symbol will be STT_SECTION and sym.st_name * will point to empty string (0), so fetch section name * instead */ - if (GELF_ST_TYPE(sym.st_info) == STT_SECTION && sym.st_name == 0) - sym_name = elf_sec_name(obj, elf_sec_by_idx(obj, sym.st_shndx)); + if (ELF64_ST_TYPE(sym->st_info) == STT_SECTION && sym->st_name == 0) + sym_name = elf_sec_name(obj, elf_sec_by_idx(obj, sym->st_shndx)); else - sym_name = elf_sym_str(obj, sym.st_name); + sym_name = elf_sym_str(obj, sym->st_name); sym_name = sym_name ?: "<?"; pr_debug("sec '%s': relo #%d: insn #%u against '%s'\n", @@ -3853,7 +4097,7 @@ bpf_object__collect_prog_relos(struct bpf_object *obj, GElf_Shdr *shdr, Elf_Data /* adjust insn_idx to local BPF program frame of reference */ insn_idx -= prog->sec_insn_off; err = bpf_program__record_reloc(prog, &relos[prog->nr_reloc], - insn_idx, sym_name, &sym, &rel); + insn_idx, sym_name, sym, rel); if (err) return err; @@ -3885,8 +4129,7 @@ static int bpf_map_find_btf_info(struct bpf_object *obj, struct bpf_map *map) * LLVM annotates global data differently in BTF, that is, * only as '.data', '.bss' or '.rodata'. */ - ret = btf__find_by_name(obj->btf, - libbpf_type_to_btf_name[map->libbpf_type]); + ret = btf__find_by_name(obj->btf, map->real_name); } if (ret < 0) return ret; @@ -3979,6 +4222,7 @@ int bpf_map__reuse_fd(struct bpf_map *map, int fd) map->btf_key_type_id = info.btf_key_type_id; map->btf_value_type_id = info.btf_value_type_id; map->reused = true; + map->map_extra = info.map_extra; return 0; @@ -4207,6 +4451,23 @@ static int probe_kern_btf_float(void) strs, sizeof(strs))); } +static int probe_kern_btf_decl_tag(void) +{ + static const char strs[] = "\0tag"; + __u32 types[] = { + /* int */ + BTF_TYPE_INT_ENC(0, BTF_INT_SIGNED, 0, 32, 4), /* [1] */ + /* VAR x */ /* [2] */ + BTF_TYPE_ENC(1, BTF_INFO_ENC(BTF_KIND_VAR, 0, 0), 1), + BTF_VAR_STATIC, + /* attr */ + BTF_TYPE_DECL_TAG_ENC(1, 2, -1), + }; + + return probe_fd(libbpf__load_raw_btf((char *)types, sizeof(types), + strs, sizeof(strs))); +} + static int probe_kern_array_mmap(void) { struct bpf_create_map_attr attr = { @@ -4423,6 +4684,9 @@ static struct kern_feature_desc { [FEAT_PERF_LINK] = { "BPF perf link support", probe_perf_link, }, + [FEAT_BTF_DECL_TAG] = { + "BTF_KIND_DECL_TAG support", probe_kern_btf_decl_tag, + }, }; static bool kernel_supports(const struct bpf_object *obj, enum kern_feature_id feat_id) @@ -4473,7 +4737,8 @@ static bool map_is_reuse_compat(const struct bpf_map *map, int map_fd) map_info.key_size == map->def.key_size && map_info.value_size == map->def.value_size && map_info.max_entries == map->def.max_entries && - map_info.map_flags == map->def.map_flags); + map_info.map_flags == map->def.map_flags && + map_info.map_extra == map->map_extra); } static int @@ -4556,7 +4821,7 @@ static void bpf_map__destroy(struct bpf_map *map); static int bpf_object__create_map(struct bpf_object *obj, struct bpf_map *map, bool is_inner) { - struct bpf_create_map_attr create_attr; + struct bpf_create_map_params create_attr; struct bpf_map_def *def = &map->def; int err = 0; @@ -4570,6 +4835,7 @@ static int bpf_object__create_map(struct bpf_object *obj, struct bpf_map *map, b create_attr.key_size = def->key_size; create_attr.value_size = def->value_size; create_attr.numa_node = map->numa_node; + create_attr.map_extra = map->map_extra; if (def->type == BPF_MAP_TYPE_PERF_EVENT_ARRAY && !def->max_entries) { int nr_cpus; @@ -4613,6 +4879,30 @@ static int bpf_object__create_map(struct bpf_object *obj, struct bpf_map *map, b create_attr.inner_map_fd = map->inner_map_fd; } + switch (def->type) { + case BPF_MAP_TYPE_PERF_EVENT_ARRAY: + case BPF_MAP_TYPE_CGROUP_ARRAY: + case BPF_MAP_TYPE_STACK_TRACE: + case BPF_MAP_TYPE_ARRAY_OF_MAPS: + case BPF_MAP_TYPE_HASH_OF_MAPS: + case BPF_MAP_TYPE_DEVMAP: + case BPF_MAP_TYPE_DEVMAP_HASH: + case BPF_MAP_TYPE_CPUMAP: + case BPF_MAP_TYPE_XSKMAP: + case BPF_MAP_TYPE_SOCKMAP: + case BPF_MAP_TYPE_SOCKHASH: + case BPF_MAP_TYPE_QUEUE: + case BPF_MAP_TYPE_STACK: + case BPF_MAP_TYPE_RINGBUF: + create_attr.btf_fd = 0; + create_attr.btf_key_type_id = 0; + create_attr.btf_value_type_id = 0; + map->btf_key_type_id = 0; + map->btf_value_type_id = 0; + default: + break; + } + if (obj->gen_loader) { bpf_gen__map_create(obj->gen_loader, &create_attr, is_inner ? -1 : map - obj->maps); /* Pretend to have valid FD to pass various fd >= 0 checks. @@ -4620,7 +4910,7 @@ static int bpf_object__create_map(struct bpf_object *obj, struct bpf_map *map, b */ map->fd = 0; } else { - map->fd = bpf_create_map_xattr(&create_attr); + map->fd = libbpf__bpf_create_map_xattr(&create_attr); } if (map->fd < 0 && (create_attr.btf_key_type_id || create_attr.btf_value_type_id)) { @@ -4635,7 +4925,7 @@ static int bpf_object__create_map(struct bpf_object *obj, struct bpf_map *map, b create_attr.btf_value_type_id = 0; map->btf_key_type_id = 0; map->btf_value_type_id = 0; - map->fd = bpf_create_map_xattr(&create_attr); + map->fd = libbpf__bpf_create_map_xattr(&create_attr); } err = map->fd < 0 ? -errno : 0; @@ -4812,8 +5102,8 @@ static int bpf_core_add_cands(struct bpf_core_cand *local_cand, size_t targ_essent_len; int n, i; - n = btf__get_nr_types(targ_btf); - for (i = targ_start_id; i <= n; i++) { + n = btf__type_cnt(targ_btf); + for (i = targ_start_id; i < n; i++) { t = btf__type_by_id(targ_btf, i); if (btf_kind(t) != btf_kind(local_cand->t)) continue; @@ -4988,7 +5278,7 @@ bpf_core_find_cands(struct bpf_object *obj, const struct btf *local_btf, __u32 l err = bpf_core_add_cands(&local_cand, local_essent_len, obj->btf_modules[i].btf, obj->btf_modules[i].name, - btf__get_nr_types(obj->btf_vmlinux) + 1, + btf__type_cnt(obj->btf_vmlinux), cands); if (err) goto err_out; @@ -5132,7 +5422,7 @@ static int bpf_core_apply_relo(struct bpf_program *prog, * relocated, so it's enough to just subtract in-section offset */ insn_idx = insn_idx - prog->sec_insn_off; - if (insn_idx > prog->insns_cnt) + if (insn_idx >= prog->insns_cnt) return -EINVAL; insn = &prog->insns[insn_idx]; @@ -5326,7 +5616,13 @@ bpf_object__relocate_data(struct bpf_object *obj, struct bpf_program *prog) case RELO_EXTERN_FUNC: ext = &obj->externs[relo->sym_off]; insn[0].src_reg = BPF_PSEUDO_KFUNC_CALL; - insn[0].imm = ext->ksym.kernel_btf_id; + if (ext->is_set) { + insn[0].imm = ext->ksym.kernel_btf_id; + insn[0].off = ext->ksym.btf_fd_idx; + } else { /* unresolved weak kfunc */ + insn[0].imm = 0; + insn[0].off = 0; + } break; case RELO_SUBPROG_ADDR: if (insn[0].src_reg != BPF_PSEUDO_FUNC) { @@ -5851,10 +6147,10 @@ bpf_object__relocate(struct bpf_object *obj, const char *targ_btf_path) } static int bpf_object__collect_st_ops_relos(struct bpf_object *obj, - GElf_Shdr *shdr, Elf_Data *data); + Elf64_Shdr *shdr, Elf_Data *data); static int bpf_object__collect_map_relos(struct bpf_object *obj, - GElf_Shdr *shdr, Elf_Data *data) + Elf64_Shdr *shdr, Elf_Data *data) { const int bpf_ptr_sz = 8, host_ptr_sz = sizeof(void *); int i, j, nrels, new_sz; @@ -5863,10 +6159,9 @@ static int bpf_object__collect_map_relos(struct bpf_object *obj, struct bpf_map *map = NULL, *targ_map; const struct btf_member *member; const char *name, *mname; - Elf_Data *symbols; unsigned int moff; - GElf_Sym sym; - GElf_Rel rel; + Elf64_Sym *sym; + Elf64_Rel *rel; void *tmp; if (!obj->efile.btf_maps_sec_btf_id || !obj->btf) @@ -5875,28 +6170,30 @@ static int bpf_object__collect_map_relos(struct bpf_object *obj, if (!sec) return -EINVAL; - symbols = obj->efile.symbols; nrels = shdr->sh_size / shdr->sh_entsize; for (i = 0; i < nrels; i++) { - if (!gelf_getrel(data, i, &rel)) { + rel = elf_rel_by_idx(data, i); + if (!rel) { pr_warn(".maps relo #%d: failed to get ELF relo\n", i); return -LIBBPF_ERRNO__FORMAT; } - if (!gelf_getsym(symbols, GELF_R_SYM(rel.r_info), &sym)) { + + sym = elf_sym_by_idx(obj, ELF64_R_SYM(rel->r_info)); + if (!sym) { pr_warn(".maps relo #%d: symbol %zx not found\n", - i, (size_t)GELF_R_SYM(rel.r_info)); + i, (size_t)ELF64_R_SYM(rel->r_info)); return -LIBBPF_ERRNO__FORMAT; } - name = elf_sym_str(obj, sym.st_name) ?: "<?>"; - if (sym.st_shndx != obj->efile.btf_maps_shndx) { + name = elf_sym_str(obj, sym->st_name) ?: "<?>"; + if (sym->st_shndx != obj->efile.btf_maps_shndx) { pr_warn(".maps relo #%d: '%s' isn't a BTF-defined map\n", i, name); return -LIBBPF_ERRNO__RELOC; } - pr_debug(".maps relo #%d: for %zd value %zd rel.r_offset %zu name %d ('%s')\n", - i, (ssize_t)(rel.r_info >> 32), (size_t)sym.st_value, - (size_t)rel.r_offset, sym.st_name, name); + pr_debug(".maps relo #%d: for %zd value %zd rel->r_offset %zu name %d ('%s')\n", + i, (ssize_t)(rel->r_info >> 32), (size_t)sym->st_value, + (size_t)rel->r_offset, sym->st_name, name); for (j = 0; j < obj->nr_maps; j++) { map = &obj->maps[j]; @@ -5904,13 +6201,13 @@ static int bpf_object__collect_map_relos(struct bpf_object *obj, continue; vi = btf_var_secinfos(sec) + map->btf_var_idx; - if (vi->offset <= rel.r_offset && - rel.r_offset + bpf_ptr_sz <= vi->offset + vi->size) + if (vi->offset <= rel->r_offset && + rel->r_offset + bpf_ptr_sz <= vi->offset + vi->size) break; } if (j == obj->nr_maps) { - pr_warn(".maps relo #%d: cannot find map '%s' at rel.r_offset %zu\n", - i, name, (size_t)rel.r_offset); + pr_warn(".maps relo #%d: cannot find map '%s' at rel->r_offset %zu\n", + i, name, (size_t)rel->r_offset); return -EINVAL; } @@ -5937,10 +6234,10 @@ static int bpf_object__collect_map_relos(struct bpf_object *obj, return -EINVAL; moff = btf_member_bit_offset(def, btf_vlen(def) - 1) / 8; - if (rel.r_offset - vi->offset < moff) + if (rel->r_offset - vi->offset < moff) return -EINVAL; - moff = rel.r_offset - vi->offset - moff; + moff = rel->r_offset - vi->offset - moff; /* here we use BPF pointer size, which is always 64 bit, as we * are parsing ELF that was built for BPF target */ @@ -5985,10 +6282,18 @@ static int bpf_object__collect_relos(struct bpf_object *obj) { int i, err; - for (i = 0; i < obj->efile.nr_reloc_sects; i++) { - GElf_Shdr *shdr = &obj->efile.reloc_sects[i].shdr; - Elf_Data *data = obj->efile.reloc_sects[i].data; - int idx = shdr->sh_info; + for (i = 0; i < obj->efile.sec_cnt; i++) { + struct elf_sec_desc *sec_desc = &obj->efile.secs[i]; + Elf64_Shdr *shdr; + Elf_Data *data; + int idx; + + if (sec_desc->sec_type != SEC_RELO) + continue; + + shdr = sec_desc->shdr; + data = sec_desc->data; + idx = shdr->sh_info; if (shdr->sh_type != SHT_REL) { pr_warn("internal error at %d\n", __LINE__); @@ -6064,15 +6369,58 @@ static int bpf_object__sanitize_prog(struct bpf_object *obj, struct bpf_program return 0; } +static int libbpf_find_attach_btf_id(struct bpf_program *prog, const char *attach_name, + int *btf_obj_fd, int *btf_type_id); + +/* this is called as prog->sec_def->preload_fn for libbpf-supported sec_defs */ +static int libbpf_preload_prog(struct bpf_program *prog, + struct bpf_prog_load_params *attr, long cookie) +{ + enum sec_def_flags def = cookie; + + /* old kernels might not support specifying expected_attach_type */ + if ((def & SEC_EXP_ATTACH_OPT) && !kernel_supports(prog->obj, FEAT_EXP_ATTACH_TYPE)) + attr->expected_attach_type = 0; + + if (def & SEC_SLEEPABLE) + attr->prog_flags |= BPF_F_SLEEPABLE; + + if ((prog->type == BPF_PROG_TYPE_TRACING || + prog->type == BPF_PROG_TYPE_LSM || + prog->type == BPF_PROG_TYPE_EXT) && !prog->attach_btf_id) { + int btf_obj_fd = 0, btf_type_id = 0, err; + const char *attach_name; + + attach_name = strchr(prog->sec_name, '/') + 1; + err = libbpf_find_attach_btf_id(prog, attach_name, &btf_obj_fd, &btf_type_id); + if (err) + return err; + + /* cache resolved BTF FD and BTF type ID in the prog */ + prog->attach_btf_obj_fd = btf_obj_fd; + prog->attach_btf_id = btf_type_id; + + /* but by now libbpf common logic is not utilizing + * prog->atach_btf_obj_fd/prog->attach_btf_id anymore because + * this callback is called after attrs were populated by + * libbpf, so this callback has to update attr explicitly here + */ + attr->attach_btf_obj_fd = btf_obj_fd; + attr->attach_btf_id = btf_type_id; + } + return 0; +} + static int load_program(struct bpf_program *prog, struct bpf_insn *insns, int insns_cnt, char *license, __u32 kern_version, int *pfd) { struct bpf_prog_load_params load_attr = {}; + struct bpf_object *obj = prog->obj; char *cp, errmsg[STRERR_BUFSIZE]; size_t log_buf_size = 0; char *log_buf = NULL; - int btf_fd, ret; + int btf_fd, ret, err; if (prog->type == BPF_PROG_TYPE_UNSPEC) { /* @@ -6088,29 +6436,22 @@ load_program(struct bpf_program *prog, struct bpf_insn *insns, int insns_cnt, return -EINVAL; load_attr.prog_type = prog->type; - /* old kernels might not support specifying expected_attach_type */ - if (!kernel_supports(prog->obj, FEAT_EXP_ATTACH_TYPE) && prog->sec_def && - prog->sec_def->is_exp_attach_type_optional) - load_attr.expected_attach_type = 0; - else - load_attr.expected_attach_type = prog->expected_attach_type; - if (kernel_supports(prog->obj, FEAT_PROG_NAME)) + load_attr.expected_attach_type = prog->expected_attach_type; + if (kernel_supports(obj, FEAT_PROG_NAME)) load_attr.name = prog->name; load_attr.insns = insns; load_attr.insn_cnt = insns_cnt; load_attr.license = license; load_attr.attach_btf_id = prog->attach_btf_id; - if (prog->attach_prog_fd) - load_attr.attach_prog_fd = prog->attach_prog_fd; - else - load_attr.attach_btf_obj_fd = prog->attach_btf_obj_fd; + load_attr.attach_prog_fd = prog->attach_prog_fd; + load_attr.attach_btf_obj_fd = prog->attach_btf_obj_fd; load_attr.attach_btf_id = prog->attach_btf_id; load_attr.kern_version = kern_version; load_attr.prog_ifindex = prog->prog_ifindex; /* specify func_info/line_info only if kernel supports them */ - btf_fd = bpf_object__btf_fd(prog->obj); - if (btf_fd >= 0 && kernel_supports(prog->obj, FEAT_BTF_FUNC)) { + btf_fd = bpf_object__btf_fd(obj); + if (btf_fd >= 0 && kernel_supports(obj, FEAT_BTF_FUNC)) { load_attr.prog_btf_fd = btf_fd; load_attr.func_info = prog->func_info; load_attr.func_info_rec_size = prog->func_info_rec_size; @@ -6121,10 +6462,21 @@ 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 = obj->fd_array; - if (prog->obj->gen_loader) { - bpf_gen__prog_load(prog->obj->gen_loader, &load_attr, - prog - prog->obj->programs); + /* adjust load_attr if sec_def provides custom preload callback */ + if (prog->sec_def && prog->sec_def->preload_fn) { + err = prog->sec_def->preload_fn(prog, &load_attr, prog->sec_def->cookie); + if (err < 0) { + pr_warn("prog '%s': failed to prepare load attributes: %d\n", + prog->name, err); + return err; + } + } + + if (obj->gen_loader) { + bpf_gen__prog_load(obj->gen_loader, &load_attr, + prog - obj->programs); *pfd = -1; return 0; } @@ -6145,16 +6497,21 @@ retry_load: if (log_buf && load_attr.log_level) pr_debug("verifier log:\n%s", log_buf); - if (prog->obj->rodata_map_idx >= 0 && - kernel_supports(prog->obj, FEAT_PROG_BIND_MAP)) { - struct bpf_map *rodata_map = - &prog->obj->maps[prog->obj->rodata_map_idx]; + if (obj->has_rodata && kernel_supports(obj, FEAT_PROG_BIND_MAP)) { + struct bpf_map *map; + int i; - if (bpf_prog_bind_map(ret, bpf_map__fd(rodata_map), NULL)) { - cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg)); - pr_warn("prog '%s': failed to bind .rodata map: %s\n", - prog->name, cp); - /* Don't fail hard if can't bind rodata. */ + for (i = 0; i < obj->nr_maps; i++) { + map = &prog->obj->maps[i]; + if (map->libbpf_type != LIBBPF_MAP_RODATA) + continue; + + if (bpf_prog_bind_map(ret, bpf_map__fd(map), NULL)) { + cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg)); + pr_warn("prog '%s': failed to bind .rodata map: %s\n", + prog->name, cp); + /* Don't fail hard if can't bind rodata. */ + } } } @@ -6218,16 +6575,13 @@ static int bpf_program__record_externs(struct bpf_program *prog) case RELO_EXTERN_VAR: if (ext->type != EXT_KSYM) continue; - if (!ext->ksym.type_id) { - pr_warn("typeless ksym %s is not supported yet\n", - ext->name); - return -ENOTSUP; - } - bpf_gen__record_extern(obj->gen_loader, ext->name, BTF_KIND_VAR, - relo->insn_idx); + bpf_gen__record_extern(obj->gen_loader, ext->name, + ext->is_weak, !ext->ksym.type_id, + BTF_KIND_VAR, relo->insn_idx); break; case RELO_EXTERN_FUNC: - bpf_gen__record_extern(obj->gen_loader, ext->name, BTF_KIND_FUNC, + bpf_gen__record_extern(obj->gen_loader, ext->name, + ext->is_weak, false, BTF_KIND_FUNC, relo->insn_idx); break; default: @@ -6237,8 +6591,6 @@ static int bpf_program__record_externs(struct bpf_program *prog) return 0; } -static int libbpf_find_attach_btf_id(struct bpf_program *prog, int *btf_obj_fd, int *btf_type_id); - int bpf_program__load(struct bpf_program *prog, char *license, __u32 kern_ver) { int err = 0, fd, i; @@ -6248,19 +6600,6 @@ int bpf_program__load(struct bpf_program *prog, char *license, __u32 kern_ver) return libbpf_err(-EINVAL); } - if ((prog->type == BPF_PROG_TYPE_TRACING || - prog->type == BPF_PROG_TYPE_LSM || - prog->type == BPF_PROG_TYPE_EXT) && !prog->attach_btf_id) { - int btf_obj_fd = 0, btf_type_id = 0; - - err = libbpf_find_attach_btf_id(prog, &btf_obj_fd, &btf_type_id); - if (err) - return libbpf_err(err); - - prog->attach_btf_obj_fd = btf_obj_fd; - prog->attach_btf_id = btf_type_id; - } - if (prog->instances.nr < 0 || !prog->instances.fds) { if (prog->preprocessor) { pr_warn("Internal error: can't load program '%s'\n", @@ -6328,8 +6667,6 @@ int bpf_program__load(struct bpf_program *prog, char *license, __u32 kern_ver) out: if (err) pr_warn("failed to load program '%s'\n", prog->name); - zfree(&prog->insns); - prog->insns_cnt = 0; return libbpf_err(err); } @@ -6367,12 +6704,51 @@ bpf_object__load_progs(struct bpf_object *obj, int log_level) static const struct bpf_sec_def *find_sec_def(const char *sec_name); +static int bpf_object_init_progs(struct bpf_object *obj, const struct bpf_object_open_opts *opts) +{ + struct bpf_program *prog; + int err; + + bpf_object__for_each_program(prog, obj) { + prog->sec_def = find_sec_def(prog->sec_name); + if (!prog->sec_def) { + /* couldn't guess, but user might manually specify */ + pr_debug("prog '%s': unrecognized ELF section name '%s'\n", + prog->name, prog->sec_name); + continue; + } + + bpf_program__set_type(prog, prog->sec_def->prog_type); + bpf_program__set_expected_attach_type(prog, prog->sec_def->expected_attach_type); + +#pragma GCC diagnostic push +#pragma GCC diagnostic ignored "-Wdeprecated-declarations" + if (prog->sec_def->prog_type == BPF_PROG_TYPE_TRACING || + prog->sec_def->prog_type == BPF_PROG_TYPE_EXT) + prog->attach_prog_fd = OPTS_GET(opts, attach_prog_fd, 0); +#pragma GCC diagnostic pop + + /* sec_def can have custom callback which should be called + * after bpf_program is initialized to adjust its properties + */ + if (prog->sec_def->init_fn) { + err = prog->sec_def->init_fn(prog, prog->sec_def->cookie); + if (err < 0) { + pr_warn("prog '%s': failed to initialize: %d\n", + prog->name, err); + return err; + } + } + } + + return 0; +} + static struct bpf_object * __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz, const struct bpf_object_open_opts *opts) { const char *obj_name, *kconfig, *btf_tmp_path; - struct bpf_program *prog; struct bpf_object *obj; char tmp_name[64]; int err; @@ -6430,30 +6806,12 @@ __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz, err = err ? : bpf_object__collect_externs(obj); err = err ? : bpf_object__finalize_btf(obj); err = err ? : bpf_object__init_maps(obj, opts); + err = err ? : bpf_object_init_progs(obj, opts); err = err ? : bpf_object__collect_relos(obj); if (err) goto out; - bpf_object__elf_finish(obj); - bpf_object__for_each_program(prog, obj) { - prog->sec_def = find_sec_def(prog->sec_name); - if (!prog->sec_def) { - /* couldn't guess, but user might manually specify */ - pr_debug("prog '%s': unrecognized ELF section name '%s'\n", - prog->name, prog->sec_name); - continue; - } - - if (prog->sec_def->is_sleepable) - prog->prog_flags |= BPF_F_SLEEPABLE; - bpf_program__set_type(prog, prog->sec_def->prog_type); - bpf_program__set_expected_attach_type(prog, - prog->sec_def->expected_attach_type); - - if (prog->sec_def->prog_type == BPF_PROG_TYPE_TRACING || - prog->sec_def->prog_type == BPF_PROG_TYPE_EXT) - prog->attach_prog_fd = OPTS_GET(opts, attach_prog_fd, 0); - } + bpf_object__elf_finish(obj); return obj; out: @@ -6529,7 +6887,7 @@ bpf_object__open_buffer(const void *obj_buf, size_t obj_buf_sz, return libbpf_ptr(__bpf_object__open(NULL, obj_buf, obj_buf_sz, &opts)); } -int bpf_object__unload(struct bpf_object *obj) +static int bpf_object_unload(struct bpf_object *obj) { size_t i; @@ -6548,6 +6906,8 @@ int bpf_object__unload(struct bpf_object *obj) return 0; } +int bpf_object__unload(struct bpf_object *obj) __attribute__((alias("bpf_object_unload"))); + static int bpf_object__sanitize_maps(struct bpf_object *obj) { struct bpf_map *m; @@ -6621,13 +6981,14 @@ out: 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) { @@ -6636,10 +6997,10 @@ 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; } @@ -6648,7 +7009,7 @@ 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; } @@ -6657,14 +7018,15 @@ 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); - if (id == -ESRCH && ext->is_weak) { - return 0; - } else if (id < 0) { + id = find_ksym_btf_id(obj, ext->name, BTF_KIND_VAR, &btf, &mod_btf); + if (id < 0) { + if (id == -ESRCH && ext->is_weak) + return 0; pr_warn("extern (var ksym) '%s': not found in kernel BTF\n", ext->name); return id; @@ -6696,7 +7058,7 @@ 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); @@ -6708,26 +7070,22 @@ 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", + if (kfunc_id == -ESRCH && ext->is_weak) + return 0; + pr_warn("extern (func ksym) '%s': not found in kernel or module BTFs\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; @@ -6739,9 +7097,30 @@ 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); @@ -6807,8 +7186,7 @@ static int bpf_object__resolve_externs(struct bpf_object *obj, if (err) return err; pr_debug("extern (kcfg) %s=0x%x\n", ext->name, kver); - } else if (ext->type == EXT_KCFG && - strncmp(ext->name, "CONFIG_", 7) == 0) { + } else if (ext->type == EXT_KCFG && str_has_pfx(ext->name, "CONFIG_")) { need_config = true; } else if (ext->type == EXT_KSYM) { if (ext->ksym.type_id) @@ -6902,6 +7280,9 @@ 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); @@ -6926,7 +7307,7 @@ out: if (obj->maps[i].pinned && !obj->maps[i].reused) bpf_map__unpin(&obj->maps[i], NULL); - bpf_object__unload(obj); + bpf_object_unload(obj); pr_warn("failed to load object '%s'\n", obj->path); return libbpf_err(err); } @@ -6992,8 +7373,7 @@ static int check_path(const char *path) return err; } -int bpf_program__pin_instance(struct bpf_program *prog, const char *path, - int instance) +static int bpf_program_pin_instance(struct bpf_program *prog, const char *path, int instance) { char *cp, errmsg[STRERR_BUFSIZE]; int err; @@ -7028,8 +7408,7 @@ int bpf_program__pin_instance(struct bpf_program *prog, const char *path, return 0; } -int bpf_program__unpin_instance(struct bpf_program *prog, const char *path, - int instance) +static int bpf_program_unpin_instance(struct bpf_program *prog, const char *path, int instance) { int err; @@ -7057,6 +7436,12 @@ int bpf_program__unpin_instance(struct bpf_program *prog, const char *path, return 0; } +__attribute__((alias("bpf_program_pin_instance"))) +int bpf_object__pin_instance(struct bpf_program *prog, const char *path, int instance); + +__attribute__((alias("bpf_program_unpin_instance"))) +int bpf_program__unpin_instance(struct bpf_program *prog, const char *path, int instance); + int bpf_program__pin(struct bpf_program *prog, const char *path) { int i, err; @@ -7081,7 +7466,7 @@ int bpf_program__pin(struct bpf_program *prog, const char *path) if (prog->instances.nr == 1) { /* don't create subdirs when pinning single instance */ - return bpf_program__pin_instance(prog, path, 0); + return bpf_program_pin_instance(prog, path, 0); } for (i = 0; i < prog->instances.nr; i++) { @@ -7097,7 +7482,7 @@ int bpf_program__pin(struct bpf_program *prog, const char *path) goto err_unpin; } - err = bpf_program__pin_instance(prog, buf, i); + err = bpf_program_pin_instance(prog, buf, i); if (err) goto err_unpin; } @@ -7115,7 +7500,7 @@ err_unpin: else if (len >= PATH_MAX) continue; - bpf_program__unpin_instance(prog, buf, i); + bpf_program_unpin_instance(prog, buf, i); } rmdir(path); @@ -7143,7 +7528,7 @@ int bpf_program__unpin(struct bpf_program *prog, const char *path) if (prog->instances.nr == 1) { /* don't create subdirs when pinning single instance */ - return bpf_program__unpin_instance(prog, path, 0); + return bpf_program_unpin_instance(prog, path, 0); } for (i = 0; i < prog->instances.nr; i++) { @@ -7156,7 +7541,7 @@ int bpf_program__unpin(struct bpf_program *prog, const char *path) else if (len >= PATH_MAX) return libbpf_err(-ENAMETOOLONG); - err = bpf_program__unpin_instance(prog, buf, i); + err = bpf_program_unpin_instance(prog, buf, i); if (err) return err; } @@ -7517,6 +7902,7 @@ static void bpf_map__destroy(struct bpf_map *map) } zfree(&map->name); + zfree(&map->real_name); zfree(&map->pin_path); if (map->fd >= 0) @@ -7535,7 +7921,7 @@ void bpf_object__close(struct bpf_object *obj) bpf_gen__free(obj->gen_loader); bpf_object__elf_finish(obj); - bpf_object__unload(obj); + bpf_object_unload(obj); btf__free(obj->btf); btf_ext__free(obj->btf_ext); @@ -7564,6 +7950,10 @@ struct bpf_object * bpf_object__next(struct bpf_object *prev) { struct bpf_object *next; + bool strict = (libbpf_mode & LIBBPF_STRICT_NO_OBJECT_LIST); + + if (strict) + return NULL; if (!prev) next = list_first_entry(&bpf_objects_list, @@ -7670,6 +8060,12 @@ __bpf_program__iter(const struct bpf_program *p, const struct bpf_object *obj, struct bpf_program * bpf_program__next(struct bpf_program *prev, const struct bpf_object *obj) { + return bpf_object__next_program(obj, prev); +} + +struct bpf_program * +bpf_object__next_program(const struct bpf_object *obj, struct bpf_program *prev) +{ struct bpf_program *prog = prev; do { @@ -7682,6 +8078,12 @@ bpf_program__next(struct bpf_program *prev, const struct bpf_object *obj) struct bpf_program * bpf_program__prev(struct bpf_program *next, const struct bpf_object *obj) { + return bpf_object__prev_program(obj, next); +} + +struct bpf_program * +bpf_object__prev_program(const struct bpf_object *obj, struct bpf_program *next) +{ struct bpf_program *prog = next; do { @@ -7762,6 +8164,16 @@ size_t bpf_program__size(const struct bpf_program *prog) return prog->insns_cnt * BPF_INSN_SZ; } +const struct bpf_insn *bpf_program__insns(const struct bpf_program *prog) +{ + return prog->insns; +} + +size_t bpf_program__insn_cnt(const struct bpf_program *prog) +{ + return prog->insns_cnt; +} + int bpf_program__set_prep(struct bpf_program *prog, int nr_instances, bpf_program_prep_t prep) { @@ -7869,223 +8281,145 @@ void bpf_program__set_expected_attach_type(struct bpf_program *prog, prog->expected_attach_type = type; } -#define BPF_PROG_SEC_IMPL(string, ptype, eatype, eatype_optional, \ - attachable, attach_btf) \ - { \ - .sec = string, \ - .len = sizeof(string) - 1, \ - .prog_type = ptype, \ - .expected_attach_type = eatype, \ - .is_exp_attach_type_optional = eatype_optional, \ - .is_attachable = attachable, \ - .is_attach_btf = attach_btf, \ - } - -/* Programs that can NOT be attached. */ -#define BPF_PROG_SEC(string, ptype) BPF_PROG_SEC_IMPL(string, ptype, 0, 0, 0, 0) - -/* Programs that can be attached. */ -#define BPF_APROG_SEC(string, ptype, atype) \ - BPF_PROG_SEC_IMPL(string, ptype, atype, true, 1, 0) - -/* Programs that must specify expected attach type at load time. */ -#define BPF_EAPROG_SEC(string, ptype, eatype) \ - BPF_PROG_SEC_IMPL(string, ptype, eatype, false, 1, 0) - -/* Programs that use BTF to identify attach point */ -#define BPF_PROG_BTF(string, ptype, eatype) \ - BPF_PROG_SEC_IMPL(string, ptype, eatype, false, 0, 1) - -/* Programs that can be attached but attach type can't be identified by section - * name. Kept for backward compatibility. - */ -#define BPF_APROG_COMPAT(string, ptype) BPF_PROG_SEC(string, ptype) - -#define SEC_DEF(sec_pfx, ptype, ...) { \ +#define SEC_DEF(sec_pfx, ptype, atype, flags, ...) { \ .sec = sec_pfx, \ - .len = sizeof(sec_pfx) - 1, \ .prog_type = BPF_PROG_TYPE_##ptype, \ + .expected_attach_type = atype, \ + .cookie = (long)(flags), \ + .preload_fn = libbpf_preload_prog, \ __VA_ARGS__ \ } -static struct bpf_link *attach_kprobe(const struct bpf_sec_def *sec, - struct bpf_program *prog); -static struct bpf_link *attach_tp(const struct bpf_sec_def *sec, - struct bpf_program *prog); -static struct bpf_link *attach_raw_tp(const struct bpf_sec_def *sec, - struct bpf_program *prog); -static struct bpf_link *attach_trace(const struct bpf_sec_def *sec, - struct bpf_program *prog); -static struct bpf_link *attach_lsm(const struct bpf_sec_def *sec, - struct bpf_program *prog); -static struct bpf_link *attach_iter(const struct bpf_sec_def *sec, - struct bpf_program *prog); +static struct bpf_link *attach_kprobe(const struct bpf_program *prog, long cookie); +static struct bpf_link *attach_tp(const struct bpf_program *prog, long cookie); +static struct bpf_link *attach_raw_tp(const struct bpf_program *prog, long cookie); +static struct bpf_link *attach_trace(const struct bpf_program *prog, long cookie); +static struct bpf_link *attach_lsm(const struct bpf_program *prog, long cookie); +static struct bpf_link *attach_iter(const struct bpf_program *prog, long cookie); static const struct bpf_sec_def section_defs[] = { - BPF_PROG_SEC("socket", BPF_PROG_TYPE_SOCKET_FILTER), - BPF_EAPROG_SEC("sk_reuseport/migrate", BPF_PROG_TYPE_SK_REUSEPORT, - BPF_SK_REUSEPORT_SELECT_OR_MIGRATE), - BPF_EAPROG_SEC("sk_reuseport", BPF_PROG_TYPE_SK_REUSEPORT, - BPF_SK_REUSEPORT_SELECT), - SEC_DEF("kprobe/", KPROBE, - .attach_fn = attach_kprobe), - BPF_PROG_SEC("uprobe/", BPF_PROG_TYPE_KPROBE), - SEC_DEF("kretprobe/", KPROBE, - .attach_fn = attach_kprobe), - BPF_PROG_SEC("uretprobe/", BPF_PROG_TYPE_KPROBE), - BPF_PROG_SEC("classifier", BPF_PROG_TYPE_SCHED_CLS), - BPF_PROG_SEC("action", BPF_PROG_TYPE_SCHED_ACT), - SEC_DEF("tracepoint/", TRACEPOINT, - .attach_fn = attach_tp), - SEC_DEF("tp/", TRACEPOINT, - .attach_fn = attach_tp), - SEC_DEF("raw_tracepoint/", RAW_TRACEPOINT, - .attach_fn = attach_raw_tp), - SEC_DEF("raw_tp/", RAW_TRACEPOINT, - .attach_fn = attach_raw_tp), - SEC_DEF("tp_btf/", TRACING, - .expected_attach_type = BPF_TRACE_RAW_TP, - .is_attach_btf = true, - .attach_fn = attach_trace), - SEC_DEF("fentry/", TRACING, - .expected_attach_type = BPF_TRACE_FENTRY, - .is_attach_btf = true, - .attach_fn = attach_trace), - SEC_DEF("fmod_ret/", TRACING, - .expected_attach_type = BPF_MODIFY_RETURN, - .is_attach_btf = true, - .attach_fn = attach_trace), - SEC_DEF("fexit/", TRACING, - .expected_attach_type = BPF_TRACE_FEXIT, - .is_attach_btf = true, - .attach_fn = attach_trace), - SEC_DEF("fentry.s/", TRACING, - .expected_attach_type = BPF_TRACE_FENTRY, - .is_attach_btf = true, - .is_sleepable = true, - .attach_fn = attach_trace), - SEC_DEF("fmod_ret.s/", TRACING, - .expected_attach_type = BPF_MODIFY_RETURN, - .is_attach_btf = true, - .is_sleepable = true, - .attach_fn = attach_trace), - SEC_DEF("fexit.s/", TRACING, - .expected_attach_type = BPF_TRACE_FEXIT, - .is_attach_btf = true, - .is_sleepable = true, - .attach_fn = attach_trace), - SEC_DEF("freplace/", EXT, - .is_attach_btf = true, - .attach_fn = attach_trace), - SEC_DEF("lsm/", LSM, - .is_attach_btf = true, - .expected_attach_type = BPF_LSM_MAC, - .attach_fn = attach_lsm), - SEC_DEF("lsm.s/", LSM, - .is_attach_btf = true, - .is_sleepable = true, - .expected_attach_type = BPF_LSM_MAC, - .attach_fn = attach_lsm), - SEC_DEF("iter/", TRACING, - .expected_attach_type = BPF_TRACE_ITER, - .is_attach_btf = true, - .attach_fn = attach_iter), - SEC_DEF("syscall", SYSCALL, - .is_sleepable = true), - BPF_EAPROG_SEC("xdp_devmap/", BPF_PROG_TYPE_XDP, - BPF_XDP_DEVMAP), - BPF_EAPROG_SEC("xdp_cpumap/", BPF_PROG_TYPE_XDP, - BPF_XDP_CPUMAP), - BPF_APROG_SEC("xdp", BPF_PROG_TYPE_XDP, - BPF_XDP), - BPF_PROG_SEC("perf_event", BPF_PROG_TYPE_PERF_EVENT), - BPF_PROG_SEC("lwt_in", BPF_PROG_TYPE_LWT_IN), - BPF_PROG_SEC("lwt_out", BPF_PROG_TYPE_LWT_OUT), - BPF_PROG_SEC("lwt_xmit", BPF_PROG_TYPE_LWT_XMIT), - BPF_PROG_SEC("lwt_seg6local", BPF_PROG_TYPE_LWT_SEG6LOCAL), - BPF_APROG_SEC("cgroup_skb/ingress", BPF_PROG_TYPE_CGROUP_SKB, - BPF_CGROUP_INET_INGRESS), - BPF_APROG_SEC("cgroup_skb/egress", BPF_PROG_TYPE_CGROUP_SKB, - BPF_CGROUP_INET_EGRESS), - BPF_APROG_COMPAT("cgroup/skb", BPF_PROG_TYPE_CGROUP_SKB), - BPF_EAPROG_SEC("cgroup/sock_create", BPF_PROG_TYPE_CGROUP_SOCK, - BPF_CGROUP_INET_SOCK_CREATE), - BPF_EAPROG_SEC("cgroup/sock_release", BPF_PROG_TYPE_CGROUP_SOCK, - BPF_CGROUP_INET_SOCK_RELEASE), - BPF_APROG_SEC("cgroup/sock", BPF_PROG_TYPE_CGROUP_SOCK, - BPF_CGROUP_INET_SOCK_CREATE), - BPF_EAPROG_SEC("cgroup/post_bind4", BPF_PROG_TYPE_CGROUP_SOCK, - BPF_CGROUP_INET4_POST_BIND), - BPF_EAPROG_SEC("cgroup/post_bind6", BPF_PROG_TYPE_CGROUP_SOCK, - BPF_CGROUP_INET6_POST_BIND), - BPF_APROG_SEC("cgroup/dev", BPF_PROG_TYPE_CGROUP_DEVICE, - BPF_CGROUP_DEVICE), - BPF_APROG_SEC("sockops", BPF_PROG_TYPE_SOCK_OPS, - BPF_CGROUP_SOCK_OPS), - BPF_APROG_SEC("sk_skb/stream_parser", BPF_PROG_TYPE_SK_SKB, - BPF_SK_SKB_STREAM_PARSER), - BPF_APROG_SEC("sk_skb/stream_verdict", BPF_PROG_TYPE_SK_SKB, - BPF_SK_SKB_STREAM_VERDICT), - BPF_APROG_COMPAT("sk_skb", BPF_PROG_TYPE_SK_SKB), - BPF_APROG_SEC("sk_msg", BPF_PROG_TYPE_SK_MSG, - BPF_SK_MSG_VERDICT), - BPF_APROG_SEC("lirc_mode2", BPF_PROG_TYPE_LIRC_MODE2, - BPF_LIRC_MODE2), - BPF_APROG_SEC("flow_dissector", BPF_PROG_TYPE_FLOW_DISSECTOR, - BPF_FLOW_DISSECTOR), - BPF_EAPROG_SEC("cgroup/bind4", BPF_PROG_TYPE_CGROUP_SOCK_ADDR, - BPF_CGROUP_INET4_BIND), - BPF_EAPROG_SEC("cgroup/bind6", BPF_PROG_TYPE_CGROUP_SOCK_ADDR, - BPF_CGROUP_INET6_BIND), - BPF_EAPROG_SEC("cgroup/connect4", BPF_PROG_TYPE_CGROUP_SOCK_ADDR, - BPF_CGROUP_INET4_CONNECT), - BPF_EAPROG_SEC("cgroup/connect6", BPF_PROG_TYPE_CGROUP_SOCK_ADDR, - BPF_CGROUP_INET6_CONNECT), - BPF_EAPROG_SEC("cgroup/sendmsg4", BPF_PROG_TYPE_CGROUP_SOCK_ADDR, - BPF_CGROUP_UDP4_SENDMSG), - BPF_EAPROG_SEC("cgroup/sendmsg6", BPF_PROG_TYPE_CGROUP_SOCK_ADDR, - BPF_CGROUP_UDP6_SENDMSG), - BPF_EAPROG_SEC("cgroup/recvmsg4", BPF_PROG_TYPE_CGROUP_SOCK_ADDR, - BPF_CGROUP_UDP4_RECVMSG), - BPF_EAPROG_SEC("cgroup/recvmsg6", BPF_PROG_TYPE_CGROUP_SOCK_ADDR, - BPF_CGROUP_UDP6_RECVMSG), - BPF_EAPROG_SEC("cgroup/getpeername4", BPF_PROG_TYPE_CGROUP_SOCK_ADDR, - BPF_CGROUP_INET4_GETPEERNAME), - BPF_EAPROG_SEC("cgroup/getpeername6", BPF_PROG_TYPE_CGROUP_SOCK_ADDR, - BPF_CGROUP_INET6_GETPEERNAME), - BPF_EAPROG_SEC("cgroup/getsockname4", BPF_PROG_TYPE_CGROUP_SOCK_ADDR, - BPF_CGROUP_INET4_GETSOCKNAME), - BPF_EAPROG_SEC("cgroup/getsockname6", BPF_PROG_TYPE_CGROUP_SOCK_ADDR, - BPF_CGROUP_INET6_GETSOCKNAME), - BPF_EAPROG_SEC("cgroup/sysctl", BPF_PROG_TYPE_CGROUP_SYSCTL, - BPF_CGROUP_SYSCTL), - BPF_EAPROG_SEC("cgroup/getsockopt", BPF_PROG_TYPE_CGROUP_SOCKOPT, - BPF_CGROUP_GETSOCKOPT), - BPF_EAPROG_SEC("cgroup/setsockopt", BPF_PROG_TYPE_CGROUP_SOCKOPT, - BPF_CGROUP_SETSOCKOPT), - BPF_PROG_SEC("struct_ops", BPF_PROG_TYPE_STRUCT_OPS), - BPF_EAPROG_SEC("sk_lookup/", BPF_PROG_TYPE_SK_LOOKUP, - BPF_SK_LOOKUP), + SEC_DEF("socket", SOCKET_FILTER, 0, SEC_NONE | SEC_SLOPPY_PFX), + SEC_DEF("sk_reuseport/migrate", SK_REUSEPORT, BPF_SK_REUSEPORT_SELECT_OR_MIGRATE, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("sk_reuseport", SK_REUSEPORT, BPF_SK_REUSEPORT_SELECT, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("kprobe/", KPROBE, 0, SEC_NONE, attach_kprobe), + SEC_DEF("uprobe/", KPROBE, 0, SEC_NONE), + SEC_DEF("kretprobe/", KPROBE, 0, SEC_NONE, attach_kprobe), + SEC_DEF("uretprobe/", KPROBE, 0, SEC_NONE), + SEC_DEF("tc", SCHED_CLS, 0, SEC_NONE), + SEC_DEF("classifier", SCHED_CLS, 0, SEC_NONE | SEC_SLOPPY_PFX), + SEC_DEF("action", SCHED_ACT, 0, SEC_NONE | SEC_SLOPPY_PFX), + SEC_DEF("tracepoint/", TRACEPOINT, 0, SEC_NONE, attach_tp), + SEC_DEF("tp/", TRACEPOINT, 0, SEC_NONE, attach_tp), + SEC_DEF("raw_tracepoint/", RAW_TRACEPOINT, 0, SEC_NONE, attach_raw_tp), + SEC_DEF("raw_tp/", RAW_TRACEPOINT, 0, SEC_NONE, attach_raw_tp), + SEC_DEF("raw_tracepoint.w/", RAW_TRACEPOINT_WRITABLE, 0, SEC_NONE, attach_raw_tp), + SEC_DEF("raw_tp.w/", RAW_TRACEPOINT_WRITABLE, 0, SEC_NONE, attach_raw_tp), + SEC_DEF("tp_btf/", TRACING, BPF_TRACE_RAW_TP, SEC_ATTACH_BTF, attach_trace), + SEC_DEF("fentry/", TRACING, BPF_TRACE_FENTRY, SEC_ATTACH_BTF, attach_trace), + SEC_DEF("fmod_ret/", TRACING, BPF_MODIFY_RETURN, SEC_ATTACH_BTF, attach_trace), + SEC_DEF("fexit/", TRACING, BPF_TRACE_FEXIT, SEC_ATTACH_BTF, attach_trace), + SEC_DEF("fentry.s/", TRACING, BPF_TRACE_FENTRY, SEC_ATTACH_BTF | SEC_SLEEPABLE, attach_trace), + SEC_DEF("fmod_ret.s/", TRACING, BPF_MODIFY_RETURN, SEC_ATTACH_BTF | SEC_SLEEPABLE, attach_trace), + SEC_DEF("fexit.s/", TRACING, BPF_TRACE_FEXIT, SEC_ATTACH_BTF | SEC_SLEEPABLE, attach_trace), + SEC_DEF("freplace/", EXT, 0, SEC_ATTACH_BTF, attach_trace), + SEC_DEF("lsm/", LSM, BPF_LSM_MAC, SEC_ATTACH_BTF, attach_lsm), + SEC_DEF("lsm.s/", LSM, BPF_LSM_MAC, SEC_ATTACH_BTF | SEC_SLEEPABLE, attach_lsm), + SEC_DEF("iter/", TRACING, BPF_TRACE_ITER, SEC_ATTACH_BTF, attach_iter), + SEC_DEF("syscall", SYSCALL, 0, SEC_SLEEPABLE), + SEC_DEF("xdp_devmap/", XDP, BPF_XDP_DEVMAP, SEC_ATTACHABLE), + SEC_DEF("xdp_cpumap/", XDP, BPF_XDP_CPUMAP, SEC_ATTACHABLE), + SEC_DEF("xdp", XDP, BPF_XDP, SEC_ATTACHABLE_OPT | SEC_SLOPPY_PFX), + SEC_DEF("perf_event", PERF_EVENT, 0, SEC_NONE | SEC_SLOPPY_PFX), + SEC_DEF("lwt_in", LWT_IN, 0, SEC_NONE | SEC_SLOPPY_PFX), + SEC_DEF("lwt_out", LWT_OUT, 0, SEC_NONE | SEC_SLOPPY_PFX), + SEC_DEF("lwt_xmit", LWT_XMIT, 0, SEC_NONE | SEC_SLOPPY_PFX), + SEC_DEF("lwt_seg6local", LWT_SEG6LOCAL, 0, SEC_NONE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup_skb/ingress", CGROUP_SKB, BPF_CGROUP_INET_INGRESS, SEC_ATTACHABLE_OPT | SEC_SLOPPY_PFX), + SEC_DEF("cgroup_skb/egress", CGROUP_SKB, BPF_CGROUP_INET_EGRESS, SEC_ATTACHABLE_OPT | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/skb", CGROUP_SKB, 0, SEC_NONE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/sock_create", CGROUP_SOCK, BPF_CGROUP_INET_SOCK_CREATE, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/sock_release", CGROUP_SOCK, BPF_CGROUP_INET_SOCK_RELEASE, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/sock", CGROUP_SOCK, BPF_CGROUP_INET_SOCK_CREATE, SEC_ATTACHABLE_OPT | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/post_bind4", CGROUP_SOCK, BPF_CGROUP_INET4_POST_BIND, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/post_bind6", CGROUP_SOCK, BPF_CGROUP_INET6_POST_BIND, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/dev", CGROUP_DEVICE, BPF_CGROUP_DEVICE, SEC_ATTACHABLE_OPT | SEC_SLOPPY_PFX), + SEC_DEF("sockops", SOCK_OPS, BPF_CGROUP_SOCK_OPS, SEC_ATTACHABLE_OPT | SEC_SLOPPY_PFX), + SEC_DEF("sk_skb/stream_parser", SK_SKB, BPF_SK_SKB_STREAM_PARSER, SEC_ATTACHABLE_OPT | SEC_SLOPPY_PFX), + SEC_DEF("sk_skb/stream_verdict",SK_SKB, BPF_SK_SKB_STREAM_VERDICT, SEC_ATTACHABLE_OPT | SEC_SLOPPY_PFX), + SEC_DEF("sk_skb", SK_SKB, 0, SEC_NONE | SEC_SLOPPY_PFX), + SEC_DEF("sk_msg", SK_MSG, BPF_SK_MSG_VERDICT, SEC_ATTACHABLE_OPT | SEC_SLOPPY_PFX), + SEC_DEF("lirc_mode2", LIRC_MODE2, BPF_LIRC_MODE2, SEC_ATTACHABLE_OPT | SEC_SLOPPY_PFX), + SEC_DEF("flow_dissector", FLOW_DISSECTOR, BPF_FLOW_DISSECTOR, SEC_ATTACHABLE_OPT | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/bind4", CGROUP_SOCK_ADDR, BPF_CGROUP_INET4_BIND, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/bind6", CGROUP_SOCK_ADDR, BPF_CGROUP_INET6_BIND, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/connect4", CGROUP_SOCK_ADDR, BPF_CGROUP_INET4_CONNECT, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/connect6", CGROUP_SOCK_ADDR, BPF_CGROUP_INET6_CONNECT, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/sendmsg4", CGROUP_SOCK_ADDR, BPF_CGROUP_UDP4_SENDMSG, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/sendmsg6", CGROUP_SOCK_ADDR, BPF_CGROUP_UDP6_SENDMSG, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/recvmsg4", CGROUP_SOCK_ADDR, BPF_CGROUP_UDP4_RECVMSG, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/recvmsg6", CGROUP_SOCK_ADDR, BPF_CGROUP_UDP6_RECVMSG, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/getpeername4", CGROUP_SOCK_ADDR, BPF_CGROUP_INET4_GETPEERNAME, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/getpeername6", CGROUP_SOCK_ADDR, BPF_CGROUP_INET6_GETPEERNAME, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/getsockname4", CGROUP_SOCK_ADDR, BPF_CGROUP_INET4_GETSOCKNAME, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/getsockname6", CGROUP_SOCK_ADDR, BPF_CGROUP_INET6_GETSOCKNAME, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/sysctl", CGROUP_SYSCTL, BPF_CGROUP_SYSCTL, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/getsockopt", CGROUP_SOCKOPT, BPF_CGROUP_GETSOCKOPT, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("cgroup/setsockopt", CGROUP_SOCKOPT, BPF_CGROUP_SETSOCKOPT, SEC_ATTACHABLE | SEC_SLOPPY_PFX), + SEC_DEF("struct_ops+", STRUCT_OPS, 0, SEC_NONE), + SEC_DEF("sk_lookup", SK_LOOKUP, BPF_SK_LOOKUP, SEC_ATTACHABLE | SEC_SLOPPY_PFX), }; -#undef BPF_PROG_SEC_IMPL -#undef BPF_PROG_SEC -#undef BPF_APROG_SEC -#undef BPF_EAPROG_SEC -#undef BPF_APROG_COMPAT -#undef SEC_DEF - #define MAX_TYPE_NAME_SIZE 32 static const struct bpf_sec_def *find_sec_def(const char *sec_name) { - int i, n = ARRAY_SIZE(section_defs); + const struct bpf_sec_def *sec_def; + enum sec_def_flags sec_flags; + int i, n = ARRAY_SIZE(section_defs), len; + bool strict = libbpf_mode & LIBBPF_STRICT_SEC_NAME; for (i = 0; i < n; i++) { - if (strncmp(sec_name, - section_defs[i].sec, section_defs[i].len)) + sec_def = §ion_defs[i]; + sec_flags = sec_def->cookie; + len = strlen(sec_def->sec); + + /* "type/" always has to have proper SEC("type/extras") form */ + if (sec_def->sec[len - 1] == '/') { + if (str_has_pfx(sec_name, sec_def->sec)) + return sec_def; continue; - return §ion_defs[i]; + } + + /* "type+" means it can be either exact SEC("type") or + * well-formed SEC("type/extras") with proper '/' separator + */ + if (sec_def->sec[len - 1] == '+') { + len--; + /* not even a prefix */ + if (strncmp(sec_name, sec_def->sec, len) != 0) + continue; + /* exact match or has '/' separator */ + if (sec_name[len] == '\0' || sec_name[len] == '/') + return sec_def; + continue; + } + + /* SEC_SLOPPY_PFX definitions are allowed to be just prefix + * matches, unless strict section name mode + * (LIBBPF_STRICT_SEC_NAME) is enabled, in which case the + * match has to be exact. + */ + if ((sec_flags & SEC_SLOPPY_PFX) && !strict) { + if (str_has_pfx(sec_name, sec_def->sec)) + return sec_def; + continue; + } + + /* Definitions not marked SEC_SLOPPY_PFX (e.g., + * SEC("syscall")) are exact matches in both modes. + */ + if (strcmp(sec_name, sec_def->sec) == 0) + return sec_def; } return NULL; } @@ -8102,8 +8436,15 @@ static char *libbpf_get_type_names(bool attach_type) buf[0] = '\0'; /* Forge string buf with all available names */ for (i = 0; i < ARRAY_SIZE(section_defs); i++) { - if (attach_type && !section_defs[i].is_attachable) - continue; + const struct bpf_sec_def *sec_def = §ion_defs[i]; + + if (attach_type) { + if (sec_def->preload_fn != libbpf_preload_prog) + continue; + + if (!(sec_def->cookie & SEC_ATTACHABLE)) + continue; + } if (strlen(buf) + strlen(section_defs[i].sec) + 2 > len) { free(buf); @@ -8162,7 +8503,7 @@ static struct bpf_map *find_struct_ops_map_by_offset(struct bpf_object *obj, /* Collect the reloc from ELF and populate the st_ops->progs[] */ static int bpf_object__collect_st_ops_relos(struct bpf_object *obj, - GElf_Shdr *shdr, Elf_Data *data) + Elf64_Shdr *shdr, Elf_Data *data) { const struct btf_member *member; struct bpf_struct_ops *st_ops; @@ -8170,58 +8511,58 @@ static int bpf_object__collect_st_ops_relos(struct bpf_object *obj, unsigned int shdr_idx; const struct btf *btf; struct bpf_map *map; - Elf_Data *symbols; unsigned int moff, insn_idx; const char *name; __u32 member_idx; - GElf_Sym sym; - GElf_Rel rel; + Elf64_Sym *sym; + Elf64_Rel *rel; int i, nrels; - symbols = obj->efile.symbols; btf = obj->btf; nrels = shdr->sh_size / shdr->sh_entsize; for (i = 0; i < nrels; i++) { - if (!gelf_getrel(data, i, &rel)) { + rel = elf_rel_by_idx(data, i); + if (!rel) { pr_warn("struct_ops reloc: failed to get %d reloc\n", i); return -LIBBPF_ERRNO__FORMAT; } - if (!gelf_getsym(symbols, GELF_R_SYM(rel.r_info), &sym)) { + sym = elf_sym_by_idx(obj, ELF64_R_SYM(rel->r_info)); + if (!sym) { pr_warn("struct_ops reloc: symbol %zx not found\n", - (size_t)GELF_R_SYM(rel.r_info)); + (size_t)ELF64_R_SYM(rel->r_info)); return -LIBBPF_ERRNO__FORMAT; } - name = elf_sym_str(obj, sym.st_name) ?: "<?>"; - map = find_struct_ops_map_by_offset(obj, rel.r_offset); + name = elf_sym_str(obj, sym->st_name) ?: "<?>"; + map = find_struct_ops_map_by_offset(obj, rel->r_offset); if (!map) { - pr_warn("struct_ops reloc: cannot find map at rel.r_offset %zu\n", - (size_t)rel.r_offset); + pr_warn("struct_ops reloc: cannot find map at rel->r_offset %zu\n", + (size_t)rel->r_offset); return -EINVAL; } - moff = rel.r_offset - map->sec_offset; - shdr_idx = sym.st_shndx; + moff = rel->r_offset - map->sec_offset; + shdr_idx = sym->st_shndx; st_ops = map->st_ops; - pr_debug("struct_ops reloc %s: for %lld value %lld shdr_idx %u rel.r_offset %zu map->sec_offset %zu name %d (\'%s\')\n", + pr_debug("struct_ops reloc %s: for %lld value %lld shdr_idx %u rel->r_offset %zu map->sec_offset %zu name %d (\'%s\')\n", map->name, - (long long)(rel.r_info >> 32), - (long long)sym.st_value, - shdr_idx, (size_t)rel.r_offset, - map->sec_offset, sym.st_name, name); + (long long)(rel->r_info >> 32), + (long long)sym->st_value, + shdr_idx, (size_t)rel->r_offset, + map->sec_offset, sym->st_name, name); if (shdr_idx >= SHN_LORESERVE) { - pr_warn("struct_ops reloc %s: rel.r_offset %zu shdr_idx %u unsupported non-static function\n", - map->name, (size_t)rel.r_offset, shdr_idx); + pr_warn("struct_ops reloc %s: rel->r_offset %zu shdr_idx %u unsupported non-static function\n", + map->name, (size_t)rel->r_offset, shdr_idx); return -LIBBPF_ERRNO__RELOC; } - if (sym.st_value % BPF_INSN_SZ) { + if (sym->st_value % BPF_INSN_SZ) { pr_warn("struct_ops reloc %s: invalid target program offset %llu\n", - map->name, (unsigned long long)sym.st_value); + map->name, (unsigned long long)sym->st_value); return -LIBBPF_ERRNO__FORMAT; } - insn_idx = sym.st_value / BPF_INSN_SZ; + insn_idx = sym->st_value / BPF_INSN_SZ; member = find_member_by_offset(st_ops->type, moff * 8); if (!member) { @@ -8245,35 +8586,37 @@ static int bpf_object__collect_st_ops_relos(struct bpf_object *obj, return -EINVAL; } - if (prog->type == BPF_PROG_TYPE_UNSPEC) { - const struct bpf_sec_def *sec_def; - - sec_def = find_sec_def(prog->sec_name); - if (sec_def && - sec_def->prog_type != BPF_PROG_TYPE_STRUCT_OPS) { - /* for pr_warn */ - prog->type = sec_def->prog_type; - goto invalid_prog; - } + /* prevent the use of BPF prog with invalid type */ + if (prog->type != BPF_PROG_TYPE_STRUCT_OPS) { + pr_warn("struct_ops reloc %s: prog %s is not struct_ops BPF program\n", + map->name, prog->name); + return -EINVAL; + } - prog->type = BPF_PROG_TYPE_STRUCT_OPS; + /* if we haven't yet processed this BPF program, record proper + * attach_btf_id and member_idx + */ + if (!prog->attach_btf_id) { prog->attach_btf_id = st_ops->type_id; prog->expected_attach_type = member_idx; - } else if (prog->type != BPF_PROG_TYPE_STRUCT_OPS || - prog->attach_btf_id != st_ops->type_id || - prog->expected_attach_type != member_idx) { - goto invalid_prog; } + + /* struct_ops BPF prog can be re-used between multiple + * .struct_ops as long as it's the same struct_ops struct + * definition and the same function pointer field + */ + if (prog->attach_btf_id != st_ops->type_id || + prog->expected_attach_type != member_idx) { + pr_warn("struct_ops reloc %s: cannot use prog %s in sec %s with type %u attach_btf_id %u expected_attach_type %u for func ptr %s\n", + map->name, prog->name, prog->sec_name, prog->type, + prog->attach_btf_id, prog->expected_attach_type, name); + return -EINVAL; + } + st_ops->progs[member_idx] = prog; } return 0; - -invalid_prog: - pr_warn("struct_ops reloc %s: cannot use prog %s in sec %s with type %u attach_btf_id %u expected_attach_type %u for func ptr %s\n", - map->name, prog->name, prog->sec_name, prog->type, - prog->attach_btf_id, prog->expected_attach_type, name); - return -EINVAL; } #define BTF_TRACE_PREFIX "btf_trace_" @@ -8353,28 +8696,27 @@ int libbpf_find_vmlinux_btf_id(const char *name, static int libbpf_find_prog_btf_id(const char *name, __u32 attach_prog_fd) { - struct bpf_prog_info_linear *info_linear; - struct bpf_prog_info *info; + struct bpf_prog_info info = {}; + __u32 info_len = sizeof(info); struct btf *btf; int err; - info_linear = bpf_program__get_prog_info_linear(attach_prog_fd, 0); - err = libbpf_get_error(info_linear); + err = bpf_obj_get_info_by_fd(attach_prog_fd, &info, &info_len); if (err) { - pr_warn("failed get_prog_info_linear for FD %d\n", - attach_prog_fd); + pr_warn("failed bpf_obj_get_info_by_fd for FD %d: %d\n", + attach_prog_fd, err); return err; } err = -EINVAL; - info = &info_linear->info; - if (!info->btf_id) { + if (!info.btf_id) { pr_warn("The target program doesn't have BTF\n"); goto out; } - btf = btf__load_from_kernel_by_id(info->btf_id); - if (libbpf_get_error(btf)) { - pr_warn("Failed to get BTF of the program\n"); + btf = btf__load_from_kernel_by_id(info.btf_id); + err = libbpf_get_error(btf); + if (err) { + pr_warn("Failed to get BTF %d of the program: %d\n", info.btf_id, err); goto out; } err = btf__find_by_name_kind(btf, name, BTF_KIND_FUNC); @@ -8384,7 +8726,6 @@ static int libbpf_find_prog_btf_id(const char *name, __u32 attach_prog_fd) goto out; } out: - free(info_linear); return err; } @@ -8425,32 +8766,12 @@ static int find_kernel_btf_id(struct bpf_object *obj, const char *attach_name, return -ESRCH; } -static int libbpf_find_attach_btf_id(struct bpf_program *prog, int *btf_obj_fd, int *btf_type_id) +static int libbpf_find_attach_btf_id(struct bpf_program *prog, const char *attach_name, + int *btf_obj_fd, int *btf_type_id) { enum bpf_attach_type attach_type = prog->expected_attach_type; __u32 attach_prog_fd = prog->attach_prog_fd; - const char *name = prog->sec_name, *attach_name; - const struct bpf_sec_def *sec = NULL; - int i, err = 0; - - if (!name) - return -EINVAL; - - for (i = 0; i < ARRAY_SIZE(section_defs); i++) { - if (!section_defs[i].is_attach_btf) - continue; - if (strncmp(name, section_defs[i].sec, section_defs[i].len)) - continue; - - sec = §ion_defs[i]; - break; - } - - if (!sec) { - pr_warn("failed to identify BTF ID based on ELF section name '%s'\n", name); - return -ESRCH; - } - attach_name = name + sec->len; + int err = 0; /* BPF program's BTF ID */ if (attach_prog_fd) { @@ -8484,27 +8805,30 @@ int libbpf_attach_type_by_name(const char *name, enum bpf_attach_type *attach_type) { char *type_names; - int i; + const struct bpf_sec_def *sec_def; if (!name) return libbpf_err(-EINVAL); - for (i = 0; i < ARRAY_SIZE(section_defs); i++) { - if (strncmp(name, section_defs[i].sec, section_defs[i].len)) - continue; - if (!section_defs[i].is_attachable) - return libbpf_err(-EINVAL); - *attach_type = section_defs[i].expected_attach_type; - return 0; - } - pr_debug("failed to guess attach type based on ELF section name '%s'\n", name); - type_names = libbpf_get_type_names(true); - if (type_names != NULL) { - pr_debug("attachable section(type) names are:%s\n", type_names); - free(type_names); + sec_def = find_sec_def(name); + if (!sec_def) { + pr_debug("failed to guess attach type based on ELF section name '%s'\n", name); + type_names = libbpf_get_type_names(true); + if (type_names != NULL) { + pr_debug("attachable section(type) names are:%s\n", type_names); + free(type_names); + } + + return libbpf_err(-EINVAL); } - return libbpf_err(-EINVAL); + if (sec_def->preload_fn != libbpf_preload_prog) + return libbpf_err(-EINVAL); + if (!(sec_def->cookie & SEC_ATTACHABLE)) + return libbpf_err(-EINVAL); + + *attach_type = sec_def->expected_attach_type; + return 0; } int bpf_map__fd(const struct bpf_map *map) @@ -8517,9 +8841,30 @@ const struct bpf_map_def *bpf_map__def(const struct bpf_map *map) return map ? &map->def : libbpf_err_ptr(-EINVAL); } +static bool map_uses_real_name(const struct bpf_map *map) +{ + /* Since libbpf started to support custom .data.* and .rodata.* maps, + * their user-visible name differs from kernel-visible name. Users see + * such map's corresponding ELF section name as a map name. + * This check distinguishes .data/.rodata from .data.* and .rodata.* + * maps to know which name has to be returned to the user. + */ + if (map->libbpf_type == LIBBPF_MAP_DATA && strcmp(map->real_name, DATA_SEC) != 0) + return true; + if (map->libbpf_type == LIBBPF_MAP_RODATA && strcmp(map->real_name, RODATA_SEC) != 0) + return true; + return false; +} + const char *bpf_map__name(const struct bpf_map *map) { - return map ? map->name : NULL; + if (!map) + return NULL; + + if (map_uses_real_name(map)) + return map->real_name; + + return map->name; } enum bpf_map_type bpf_map__type(const struct bpf_map *map) @@ -8548,6 +8893,19 @@ int bpf_map__set_map_flags(struct bpf_map *map, __u32 flags) return 0; } +__u64 bpf_map__map_extra(const struct bpf_map *map) +{ + return map->map_extra; +} + +int bpf_map__set_map_extra(struct bpf_map *map, __u64 map_extra) +{ + if (map->fd >= 0) + return libbpf_err(-EBUSY); + map->map_extra = map_extra; + return 0; +} + __u32 bpf_map__numa_node(const struct bpf_map *map) { return map->numa_node; @@ -8702,6 +9060,12 @@ __bpf_map__iter(const struct bpf_map *m, const struct bpf_object *obj, int i) struct bpf_map * bpf_map__next(const struct bpf_map *prev, const struct bpf_object *obj) { + return bpf_object__next_map(obj, prev); +} + +struct bpf_map * +bpf_object__next_map(const struct bpf_object *obj, const struct bpf_map *prev) +{ if (prev == NULL) return obj->maps; @@ -8711,6 +9075,12 @@ bpf_map__next(const struct bpf_map *prev, const struct bpf_object *obj) struct bpf_map * bpf_map__prev(const struct bpf_map *next, const struct bpf_object *obj) { + return bpf_object__prev_map(obj, next); +} + +struct bpf_map * +bpf_object__prev_map(const struct bpf_object *obj, const struct bpf_map *next) +{ if (next == NULL) { if (!obj->nr_maps) return NULL; @@ -8726,7 +9096,22 @@ bpf_object__find_map_by_name(const struct bpf_object *obj, const char *name) struct bpf_map *pos; bpf_object__for_each_map(pos, obj) { - if (pos->name && !strcmp(pos->name, name)) + /* if it's a special internal map name (which always starts + * with dot) then check if that special name matches the + * real map name (ELF section name) + */ + if (name[0] == '.') { + if (pos->real_name && strcmp(pos->real_name, name) == 0) + return pos; + continue; + } + /* otherwise map name has to be an exact match */ + if (map_uses_real_name(pos)) { + if (strcmp(pos->real_name, name) == 0) + return pos; + continue; + } + if (strcmp(pos->name, name) == 0) return pos; } return errno = ENOENT, NULL; @@ -8991,8 +9376,15 @@ int bpf_link__unpin(struct bpf_link *link) struct bpf_link_perf { struct bpf_link link; int perf_event_fd; + /* legacy kprobe support: keep track of probe identifier and type */ + char *legacy_probe_name; + bool legacy_is_kprobe; + bool legacy_is_retprobe; }; +static int remove_kprobe_event_legacy(const char *probe_name, bool retprobe); +static int remove_uprobe_event_legacy(const char *probe_name, bool retprobe); + static int bpf_link_perf_detach(struct bpf_link *link) { struct bpf_link_perf *perf_link = container_of(link, struct bpf_link_perf, link); @@ -9005,17 +9397,29 @@ static int bpf_link_perf_detach(struct bpf_link *link) close(perf_link->perf_event_fd); close(link->fd); - return libbpf_err(err); + /* legacy uprobe/kprobe needs to be removed after perf event fd closure */ + if (perf_link->legacy_probe_name) { + if (perf_link->legacy_is_kprobe) { + err = remove_kprobe_event_legacy(perf_link->legacy_probe_name, + perf_link->legacy_is_retprobe); + } else { + err = remove_uprobe_event_legacy(perf_link->legacy_probe_name, + perf_link->legacy_is_retprobe); + } + } + + return err; } static void bpf_link_perf_dealloc(struct bpf_link *link) { struct bpf_link_perf *perf_link = container_of(link, struct bpf_link_perf, link); + free(perf_link->legacy_probe_name); free(perf_link); } -struct bpf_link *bpf_program__attach_perf_event_opts(struct bpf_program *prog, int pfd, +struct bpf_link *bpf_program__attach_perf_event_opts(const struct bpf_program *prog, int pfd, const struct bpf_perf_event_opts *opts) { char errmsg[STRERR_BUFSIZE]; @@ -9090,7 +9494,7 @@ err_out: return libbpf_err_ptr(err); } -struct bpf_link *bpf_program__attach_perf_event(struct bpf_program *prog, int pfd) +struct bpf_link *bpf_program__attach_perf_event(const struct bpf_program *prog, int pfd) { return bpf_program__attach_perf_event_opts(prog, pfd, NULL); } @@ -9207,16 +9611,110 @@ static int perf_event_open_probe(bool uprobe, bool retprobe, const char *name, return pfd; } +static int append_to_file(const char *file, const char *fmt, ...) +{ + int fd, n, err = 0; + va_list ap; + + fd = open(file, O_WRONLY | O_APPEND | O_CLOEXEC, 0); + if (fd < 0) + return -errno; + + va_start(ap, fmt); + n = vdprintf(fd, fmt, ap); + va_end(ap); + + if (n < 0) + err = -errno; + + close(fd); + return err; +} + +static void gen_kprobe_legacy_event_name(char *buf, size_t buf_sz, + const char *kfunc_name, size_t offset) +{ + snprintf(buf, buf_sz, "libbpf_%u_%s_0x%zx", getpid(), kfunc_name, offset); +} + +static int add_kprobe_event_legacy(const char *probe_name, bool retprobe, + const char *kfunc_name, size_t offset) +{ + const char *file = "/sys/kernel/debug/tracing/kprobe_events"; + + return append_to_file(file, "%c:%s/%s %s+0x%zx", + retprobe ? 'r' : 'p', + retprobe ? "kretprobes" : "kprobes", + probe_name, kfunc_name, offset); +} + +static int remove_kprobe_event_legacy(const char *probe_name, bool retprobe) +{ + const char *file = "/sys/kernel/debug/tracing/kprobe_events"; + + return append_to_file(file, "-:%s/%s", retprobe ? "kretprobes" : "kprobes", probe_name); +} + +static int determine_kprobe_perf_type_legacy(const char *probe_name, bool retprobe) +{ + char file[256]; + + snprintf(file, sizeof(file), + "/sys/kernel/debug/tracing/events/%s/%s/id", + retprobe ? "kretprobes" : "kprobes", probe_name); + + return parse_uint_from_file(file, "%d\n"); +} + +static int perf_event_kprobe_open_legacy(const char *probe_name, bool retprobe, + const char *kfunc_name, size_t offset, int pid) +{ + struct perf_event_attr attr = {}; + char errmsg[STRERR_BUFSIZE]; + int type, pfd, err; + + err = add_kprobe_event_legacy(probe_name, retprobe, kfunc_name, offset); + if (err < 0) { + pr_warn("failed to add legacy kprobe event for '%s+0x%zx': %s\n", + kfunc_name, offset, + libbpf_strerror_r(err, errmsg, sizeof(errmsg))); + return err; + } + type = determine_kprobe_perf_type_legacy(probe_name, retprobe); + if (type < 0) { + pr_warn("failed to determine legacy kprobe event id for '%s+0x%zx': %s\n", + kfunc_name, offset, + libbpf_strerror_r(type, errmsg, sizeof(errmsg))); + return type; + } + attr.size = sizeof(attr); + attr.config = type; + attr.type = PERF_TYPE_TRACEPOINT; + + pfd = syscall(__NR_perf_event_open, &attr, + pid < 0 ? -1 : pid, /* pid */ + pid == -1 ? 0 : -1, /* cpu */ + -1 /* group_fd */, PERF_FLAG_FD_CLOEXEC); + if (pfd < 0) { + err = -errno; + pr_warn("legacy kprobe perf_event_open() failed: %s\n", + libbpf_strerror_r(err, errmsg, sizeof(errmsg))); + return err; + } + return pfd; +} + struct bpf_link * -bpf_program__attach_kprobe_opts(struct bpf_program *prog, +bpf_program__attach_kprobe_opts(const struct bpf_program *prog, const char *func_name, const struct bpf_kprobe_opts *opts) { DECLARE_LIBBPF_OPTS(bpf_perf_event_opts, pe_opts); char errmsg[STRERR_BUFSIZE]; + char *legacy_probe = NULL; struct bpf_link *link; - unsigned long offset; - bool retprobe; + size_t offset; + bool retprobe, legacy; int pfd, err; if (!OPTS_VALID(opts, bpf_kprobe_opts)) @@ -9226,27 +9724,57 @@ bpf_program__attach_kprobe_opts(struct bpf_program *prog, offset = OPTS_GET(opts, offset, 0); pe_opts.bpf_cookie = OPTS_GET(opts, bpf_cookie, 0); - pfd = perf_event_open_probe(false /* uprobe */, retprobe, func_name, - offset, -1 /* pid */, 0 /* ref_ctr_off */); + legacy = determine_kprobe_perf_type() < 0; + if (!legacy) { + pfd = perf_event_open_probe(false /* uprobe */, retprobe, + func_name, offset, + -1 /* pid */, 0 /* ref_ctr_off */); + } else { + char probe_name[256]; + + gen_kprobe_legacy_event_name(probe_name, sizeof(probe_name), + func_name, offset); + + legacy_probe = strdup(func_name); + if (!legacy_probe) + return libbpf_err_ptr(-ENOMEM); + + pfd = perf_event_kprobe_open_legacy(legacy_probe, retprobe, func_name, + offset, -1 /* pid */); + } if (pfd < 0) { - pr_warn("prog '%s': failed to create %s '%s' perf event: %s\n", - prog->name, retprobe ? "kretprobe" : "kprobe", func_name, - libbpf_strerror_r(pfd, errmsg, sizeof(errmsg))); - return libbpf_err_ptr(pfd); + err = -errno; + pr_warn("prog '%s': failed to create %s '%s+0x%zx' perf event: %s\n", + prog->name, retprobe ? "kretprobe" : "kprobe", + func_name, offset, + libbpf_strerror_r(err, errmsg, sizeof(errmsg))); + goto err_out; } link = bpf_program__attach_perf_event_opts(prog, pfd, &pe_opts); err = libbpf_get_error(link); if (err) { close(pfd); - pr_warn("prog '%s': failed to attach to %s '%s': %s\n", - prog->name, retprobe ? "kretprobe" : "kprobe", func_name, + pr_warn("prog '%s': failed to attach to %s '%s+0x%zx': %s\n", + prog->name, retprobe ? "kretprobe" : "kprobe", + func_name, offset, libbpf_strerror_r(err, errmsg, sizeof(errmsg))); - return libbpf_err_ptr(err); + goto err_out; } + if (legacy) { + struct bpf_link_perf *perf_link = container_of(link, struct bpf_link_perf, link); + + perf_link->legacy_probe_name = legacy_probe; + perf_link->legacy_is_kprobe = true; + perf_link->legacy_is_retprobe = retprobe; + } + return link; +err_out: + free(legacy_probe); + return libbpf_err_ptr(err); } -struct bpf_link *bpf_program__attach_kprobe(struct bpf_program *prog, +struct bpf_link *bpf_program__attach_kprobe(const struct bpf_program *prog, bool retprobe, const char *func_name) { @@ -9257,8 +9785,7 @@ struct bpf_link *bpf_program__attach_kprobe(struct bpf_program *prog, return bpf_program__attach_kprobe_opts(prog, func_name, &opts); } -static struct bpf_link *attach_kprobe(const struct bpf_sec_def *sec, - struct bpf_program *prog) +static struct bpf_link *attach_kprobe(const struct bpf_program *prog, long cookie) { DECLARE_LIBBPF_OPTS(bpf_kprobe_opts, opts); unsigned long offset = 0; @@ -9267,8 +9794,11 @@ static struct bpf_link *attach_kprobe(const struct bpf_sec_def *sec, char *func; int n, err; - func_name = prog->sec_name + sec->len; - opts.retprobe = strcmp(sec->sec, "kretprobe/") == 0; + opts.retprobe = str_has_pfx(prog->sec_name, "kretprobe/"); + if (opts.retprobe) + func_name = prog->sec_name + sizeof("kretprobe/") - 1; + else + func_name = prog->sec_name + sizeof("kprobe/") - 1; n = sscanf(func_name, "%m[a-zA-Z0-9_.]+%li", &func, &offset); if (n < 1) { @@ -9289,17 +9819,96 @@ static struct bpf_link *attach_kprobe(const struct bpf_sec_def *sec, return link; } +static void gen_uprobe_legacy_event_name(char *buf, size_t buf_sz, + const char *binary_path, uint64_t offset) +{ + int i; + + snprintf(buf, buf_sz, "libbpf_%u_%s_0x%zx", getpid(), binary_path, (size_t)offset); + + /* sanitize binary_path in the probe name */ + for (i = 0; buf[i]; i++) { + if (!isalnum(buf[i])) + buf[i] = '_'; + } +} + +static inline int add_uprobe_event_legacy(const char *probe_name, bool retprobe, + const char *binary_path, size_t offset) +{ + const char *file = "/sys/kernel/debug/tracing/uprobe_events"; + + return append_to_file(file, "%c:%s/%s %s:0x%zx", + retprobe ? 'r' : 'p', + retprobe ? "uretprobes" : "uprobes", + probe_name, binary_path, offset); +} + +static inline int remove_uprobe_event_legacy(const char *probe_name, bool retprobe) +{ + const char *file = "/sys/kernel/debug/tracing/uprobe_events"; + + return append_to_file(file, "-:%s/%s", retprobe ? "uretprobes" : "uprobes", probe_name); +} + +static int determine_uprobe_perf_type_legacy(const char *probe_name, bool retprobe) +{ + char file[512]; + + snprintf(file, sizeof(file), + "/sys/kernel/debug/tracing/events/%s/%s/id", + retprobe ? "uretprobes" : "uprobes", probe_name); + + return parse_uint_from_file(file, "%d\n"); +} + +static int perf_event_uprobe_open_legacy(const char *probe_name, bool retprobe, + const char *binary_path, size_t offset, int pid) +{ + struct perf_event_attr attr; + int type, pfd, err; + + err = add_uprobe_event_legacy(probe_name, retprobe, binary_path, offset); + if (err < 0) { + pr_warn("failed to add legacy uprobe event for %s:0x%zx: %d\n", + binary_path, (size_t)offset, err); + return err; + } + type = determine_uprobe_perf_type_legacy(probe_name, retprobe); + if (type < 0) { + pr_warn("failed to determine legacy uprobe event id for %s:0x%zx: %d\n", + binary_path, offset, err); + return type; + } + + memset(&attr, 0, sizeof(attr)); + attr.size = sizeof(attr); + attr.config = type; + attr.type = PERF_TYPE_TRACEPOINT; + + pfd = syscall(__NR_perf_event_open, &attr, + pid < 0 ? -1 : pid, /* pid */ + pid == -1 ? 0 : -1, /* cpu */ + -1 /* group_fd */, PERF_FLAG_FD_CLOEXEC); + if (pfd < 0) { + err = -errno; + pr_warn("legacy uprobe perf_event_open() failed: %d\n", err); + return err; + } + return pfd; +} + LIBBPF_API struct bpf_link * -bpf_program__attach_uprobe_opts(struct bpf_program *prog, pid_t pid, +bpf_program__attach_uprobe_opts(const struct bpf_program *prog, pid_t pid, const char *binary_path, size_t func_offset, const struct bpf_uprobe_opts *opts) { DECLARE_LIBBPF_OPTS(bpf_perf_event_opts, pe_opts); - char errmsg[STRERR_BUFSIZE]; + char errmsg[STRERR_BUFSIZE], *legacy_probe = NULL; struct bpf_link *link; size_t ref_ctr_off; int pfd, err; - bool retprobe; + bool retprobe, legacy; if (!OPTS_VALID(opts, bpf_uprobe_opts)) return libbpf_err_ptr(-EINVAL); @@ -9308,15 +9917,35 @@ bpf_program__attach_uprobe_opts(struct bpf_program *prog, pid_t pid, ref_ctr_off = OPTS_GET(opts, ref_ctr_offset, 0); pe_opts.bpf_cookie = OPTS_GET(opts, bpf_cookie, 0); - pfd = perf_event_open_probe(true /* uprobe */, retprobe, binary_path, - func_offset, pid, ref_ctr_off); + legacy = determine_uprobe_perf_type() < 0; + if (!legacy) { + pfd = perf_event_open_probe(true /* uprobe */, retprobe, binary_path, + func_offset, pid, ref_ctr_off); + } else { + char probe_name[512]; + + if (ref_ctr_off) + return libbpf_err_ptr(-EINVAL); + + gen_uprobe_legacy_event_name(probe_name, sizeof(probe_name), + binary_path, func_offset); + + legacy_probe = strdup(probe_name); + if (!legacy_probe) + return libbpf_err_ptr(-ENOMEM); + + pfd = perf_event_uprobe_open_legacy(legacy_probe, retprobe, + binary_path, func_offset, pid); + } if (pfd < 0) { + err = -errno; pr_warn("prog '%s': failed to create %s '%s:0x%zx' perf event: %s\n", prog->name, retprobe ? "uretprobe" : "uprobe", binary_path, func_offset, - libbpf_strerror_r(pfd, errmsg, sizeof(errmsg))); - return libbpf_err_ptr(pfd); + libbpf_strerror_r(err, errmsg, sizeof(errmsg))); + goto err_out; } + link = bpf_program__attach_perf_event_opts(prog, pfd, &pe_opts); err = libbpf_get_error(link); if (err) { @@ -9325,12 +9954,23 @@ bpf_program__attach_uprobe_opts(struct bpf_program *prog, pid_t pid, prog->name, retprobe ? "uretprobe" : "uprobe", binary_path, func_offset, libbpf_strerror_r(err, errmsg, sizeof(errmsg))); - return libbpf_err_ptr(err); + goto err_out; + } + if (legacy) { + struct bpf_link_perf *perf_link = container_of(link, struct bpf_link_perf, link); + + perf_link->legacy_probe_name = legacy_probe; + perf_link->legacy_is_kprobe = false; + perf_link->legacy_is_retprobe = retprobe; } return link; +err_out: + free(legacy_probe); + return libbpf_err_ptr(err); + } -struct bpf_link *bpf_program__attach_uprobe(struct bpf_program *prog, +struct bpf_link *bpf_program__attach_uprobe(const struct bpf_program *prog, bool retprobe, pid_t pid, const char *binary_path, size_t func_offset) @@ -9390,7 +10030,7 @@ static int perf_event_open_tracepoint(const char *tp_category, return pfd; } -struct bpf_link *bpf_program__attach_tracepoint_opts(struct bpf_program *prog, +struct bpf_link *bpf_program__attach_tracepoint_opts(const struct bpf_program *prog, const char *tp_category, const char *tp_name, const struct bpf_tracepoint_opts *opts) @@ -9424,15 +10064,14 @@ struct bpf_link *bpf_program__attach_tracepoint_opts(struct bpf_program *prog, return link; } -struct bpf_link *bpf_program__attach_tracepoint(struct bpf_program *prog, +struct bpf_link *bpf_program__attach_tracepoint(const struct bpf_program *prog, const char *tp_category, const char *tp_name) { return bpf_program__attach_tracepoint_opts(prog, tp_category, tp_name, NULL); } -static struct bpf_link *attach_tp(const struct bpf_sec_def *sec, - struct bpf_program *prog) +static struct bpf_link *attach_tp(const struct bpf_program *prog, long cookie) { char *sec_name, *tp_cat, *tp_name; struct bpf_link *link; @@ -9441,8 +10080,11 @@ static struct bpf_link *attach_tp(const struct bpf_sec_def *sec, if (!sec_name) return libbpf_err_ptr(-ENOMEM); - /* extract "tp/<category>/<name>" */ - tp_cat = sec_name + sec->len; + /* extract "tp/<category>/<name>" or "tracepoint/<category>/<name>" */ + if (str_has_pfx(prog->sec_name, "tp/")) + tp_cat = sec_name + sizeof("tp/") - 1; + else + tp_cat = sec_name + sizeof("tracepoint/") - 1; tp_name = strchr(tp_cat, '/'); if (!tp_name) { free(sec_name); @@ -9456,7 +10098,7 @@ static struct bpf_link *attach_tp(const struct bpf_sec_def *sec, return link; } -struct bpf_link *bpf_program__attach_raw_tracepoint(struct bpf_program *prog, +struct bpf_link *bpf_program__attach_raw_tracepoint(const struct bpf_program *prog, const char *tp_name) { char errmsg[STRERR_BUFSIZE]; @@ -9486,16 +10128,34 @@ struct bpf_link *bpf_program__attach_raw_tracepoint(struct bpf_program *prog, return link; } -static struct bpf_link *attach_raw_tp(const struct bpf_sec_def *sec, - struct bpf_program *prog) +static struct bpf_link *attach_raw_tp(const struct bpf_program *prog, long cookie) { - const char *tp_name = prog->sec_name + sec->len; + static const char *const prefixes[] = { + "raw_tp/", + "raw_tracepoint/", + "raw_tp.w/", + "raw_tracepoint.w/", + }; + size_t i; + const char *tp_name = NULL; + + for (i = 0; i < ARRAY_SIZE(prefixes); i++) { + if (str_has_pfx(prog->sec_name, prefixes[i])) { + tp_name = prog->sec_name + strlen(prefixes[i]); + break; + } + } + if (!tp_name) { + pr_warn("prog '%s': invalid section name '%s'\n", + prog->name, prog->sec_name); + return libbpf_err_ptr(-EINVAL); + } return bpf_program__attach_raw_tracepoint(prog, tp_name); } /* Common logic for all BPF program types that attach to a btf_id */ -static struct bpf_link *bpf_program__attach_btf_id(struct bpf_program *prog) +static struct bpf_link *bpf_program__attach_btf_id(const struct bpf_program *prog) { char errmsg[STRERR_BUFSIZE]; struct bpf_link *link; @@ -9524,30 +10184,28 @@ static struct bpf_link *bpf_program__attach_btf_id(struct bpf_program *prog) return (struct bpf_link *)link; } -struct bpf_link *bpf_program__attach_trace(struct bpf_program *prog) +struct bpf_link *bpf_program__attach_trace(const struct bpf_program *prog) { return bpf_program__attach_btf_id(prog); } -struct bpf_link *bpf_program__attach_lsm(struct bpf_program *prog) +struct bpf_link *bpf_program__attach_lsm(const struct bpf_program *prog) { return bpf_program__attach_btf_id(prog); } -static struct bpf_link *attach_trace(const struct bpf_sec_def *sec, - struct bpf_program *prog) +static struct bpf_link *attach_trace(const struct bpf_program *prog, long cookie) { return bpf_program__attach_trace(prog); } -static struct bpf_link *attach_lsm(const struct bpf_sec_def *sec, - struct bpf_program *prog) +static struct bpf_link *attach_lsm(const struct bpf_program *prog, long cookie) { return bpf_program__attach_lsm(prog); } static struct bpf_link * -bpf_program__attach_fd(struct bpf_program *prog, int target_fd, int btf_id, +bpf_program__attach_fd(const struct bpf_program *prog, int target_fd, int btf_id, const char *target_name) { DECLARE_LIBBPF_OPTS(bpf_link_create_opts, opts, @@ -9583,24 +10241,24 @@ bpf_program__attach_fd(struct bpf_program *prog, int target_fd, int btf_id, } struct bpf_link * -bpf_program__attach_cgroup(struct bpf_program *prog, int cgroup_fd) +bpf_program__attach_cgroup(const struct bpf_program *prog, int cgroup_fd) { return bpf_program__attach_fd(prog, cgroup_fd, 0, "cgroup"); } struct bpf_link * -bpf_program__attach_netns(struct bpf_program *prog, int netns_fd) +bpf_program__attach_netns(const struct bpf_program *prog, int netns_fd) { return bpf_program__attach_fd(prog, netns_fd, 0, "netns"); } -struct bpf_link *bpf_program__attach_xdp(struct bpf_program *prog, int ifindex) +struct bpf_link *bpf_program__attach_xdp(const struct bpf_program *prog, int ifindex) { /* target_fd/target_ifindex use the same field in LINK_CREATE */ return bpf_program__attach_fd(prog, ifindex, 0, "xdp"); } -struct bpf_link *bpf_program__attach_freplace(struct bpf_program *prog, +struct bpf_link *bpf_program__attach_freplace(const struct bpf_program *prog, int target_fd, const char *attach_func_name) { @@ -9633,7 +10291,7 @@ struct bpf_link *bpf_program__attach_freplace(struct bpf_program *prog, } struct bpf_link * -bpf_program__attach_iter(struct bpf_program *prog, +bpf_program__attach_iter(const struct bpf_program *prog, const struct bpf_iter_attach_opts *opts) { DECLARE_LIBBPF_OPTS(bpf_link_create_opts, link_create_opts); @@ -9672,21 +10330,17 @@ bpf_program__attach_iter(struct bpf_program *prog, return link; } -static struct bpf_link *attach_iter(const struct bpf_sec_def *sec, - struct bpf_program *prog) +static struct bpf_link *attach_iter(const struct bpf_program *prog, long cookie) { return bpf_program__attach_iter(prog, NULL); } -struct bpf_link *bpf_program__attach(struct bpf_program *prog) +struct bpf_link *bpf_program__attach(const struct bpf_program *prog) { - const struct bpf_sec_def *sec_def; - - sec_def = find_sec_def(prog->sec_name); - if (!sec_def || !sec_def->attach_fn) + if (!prog->sec_def || !prog->sec_def->attach_fn) return libbpf_err_ptr(-ESRCH); - return sec_def->attach_fn(sec_def, prog); + return prog->sec_def->attach_fn(prog, prog->sec_def->cookie); } static int bpf_link__detach_struct_ops(struct bpf_link *link) @@ -9699,7 +10353,7 @@ static int bpf_link__detach_struct_ops(struct bpf_link *link) return 0; } -struct bpf_link *bpf_map__attach_struct_ops(struct bpf_map *map) +struct bpf_link *bpf_map__attach_struct_ops(const struct bpf_map *map) { struct bpf_struct_ops *st_ops; struct bpf_link *link; @@ -10512,18 +11166,29 @@ int bpf_program__set_attach_target(struct bpf_program *prog, { int btf_obj_fd = 0, btf_id = 0, err; - if (!prog || attach_prog_fd < 0 || !attach_func_name) + if (!prog || attach_prog_fd < 0) return libbpf_err(-EINVAL); if (prog->obj->loaded) return libbpf_err(-EINVAL); + if (attach_prog_fd && !attach_func_name) { + /* remember attach_prog_fd and let bpf_program__load() find + * BTF ID during the program load + */ + prog->attach_prog_fd = attach_prog_fd; + return 0; + } + if (attach_prog_fd) { btf_id = libbpf_find_prog_btf_id(attach_func_name, attach_prog_fd); if (btf_id < 0) return libbpf_err(btf_id); } else { + if (!attach_func_name) + return libbpf_err(-EINVAL); + /* load btf_vmlinux, if not yet */ err = bpf_object__load_vmlinux_btf(prog->obj, true); if (err) @@ -10596,7 +11261,7 @@ int parse_cpu_mask_file(const char *fcpu, bool **mask, int *mask_sz) int fd, err = 0, len; char buf[128]; - fd = open(fcpu, O_RDONLY); + fd = open(fcpu, O_RDONLY | O_CLOEXEC); if (fd < 0) { err = -errno; pr_warn("Failed to open cpu mask file %s: %d\n", fcpu, err); @@ -10765,16 +11430,15 @@ int bpf_object__attach_skeleton(struct bpf_object_skeleton *s) for (i = 0; i < s->prog_cnt; i++) { struct bpf_program *prog = *s->progs[i].prog; struct bpf_link **link = s->progs[i].link; - const struct bpf_sec_def *sec_def; if (!prog->load) continue; - sec_def = find_sec_def(prog->sec_name); - if (!sec_def || !sec_def->attach_fn) + /* auto-attaching not supported for this program */ + if (!prog->sec_def || !prog->sec_def->attach_fn) continue; - *link = sec_def->attach_fn(sec_def, prog); + *link = bpf_program__attach(prog); err = libbpf_get_error(*link); if (err) { pr_warn("failed to auto-attach program '%s': %d\n", |