diff options
Diffstat (limited to 'tools')
77 files changed, 1107 insertions, 236 deletions
diff --git a/tools/bpf/bpftool/btf_dumper.c b/tools/bpf/bpftool/btf_dumper.c index ede162f83eea..0e9310727281 100644 --- a/tools/bpf/bpftool/btf_dumper.c +++ b/tools/bpf/bpftool/btf_dumper.c @@ -67,7 +67,7 @@ static int dump_prog_id_as_func_ptr(const struct btf_dumper *d, if (!info->btf_id || !info->nr_func_info || btf__get_from_id(info->btf_id, &prog_btf)) goto print; - finfo = (struct bpf_func_info *)info->func_info; + finfo = u64_to_ptr(info->func_info); func_type = btf__type_by_id(prog_btf, finfo->type_id); if (!func_type || !btf_is_func(func_type)) goto print; diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c index 8a4c2b3b0cd6..f61184653633 100644 --- a/tools/bpf/bpftool/gen.c +++ b/tools/bpf/bpftool/gen.c @@ -143,6 +143,20 @@ static int codegen_datasec_def(struct bpf_object *obj, var_name, align); return -EINVAL; } + /* Assume 32-bit architectures when generating data section + * struct memory layout. Given bpftool can't know which target + * host architecture it's emitting skeleton for, we need to be + * conservative and assume 32-bit one to ensure enough padding + * bytes are generated for pointer and long types. This will + * still work correctly for 64-bit architectures, because in + * the worst case we'll generate unnecessary padding field, + * which on 64-bit architectures is not strictly necessary and + * would be handled by natural 8-byte alignment. But it still + * will be a correct memory layout, based on recorded offsets + * in BTF. + */ + if (align > 4) + align = 4; align_off = (off + align - 1) / align * align; if (align_off != need_off) { @@ -397,7 +411,7 @@ static int do_skeleton(int argc, char **argv) { \n\ struct %1$s *obj; \n\ \n\ - obj = (typeof(obj))calloc(1, sizeof(*obj)); \n\ + obj = (struct %1$s *)calloc(1, sizeof(*obj)); \n\ if (!obj) \n\ return NULL; \n\ if (%1$s__create_skeleton(obj)) \n\ @@ -461,7 +475,7 @@ static int do_skeleton(int argc, char **argv) { \n\ struct bpf_object_skeleton *s; \n\ \n\ - s = (typeof(s))calloc(1, sizeof(*s)); \n\ + s = (struct bpf_object_skeleton *)calloc(1, sizeof(*s));\n\ if (!s) \n\ return -1; \n\ obj->skeleton = s; \n\ @@ -479,7 +493,7 @@ static int do_skeleton(int argc, char **argv) /* maps */ \n\ s->map_cnt = %zu; \n\ s->map_skel_sz = sizeof(*s->maps); \n\ - s->maps = (typeof(s->maps))calloc(s->map_cnt, s->map_skel_sz);\n\ + s->maps = (struct bpf_map_skeleton *)calloc(s->map_cnt, s->map_skel_sz);\n\ if (!s->maps) \n\ goto err; \n\ ", @@ -515,7 +529,7 @@ static int do_skeleton(int argc, char **argv) /* programs */ \n\ s->prog_cnt = %zu; \n\ s->prog_skel_sz = sizeof(*s->progs); \n\ - s->progs = (typeof(s->progs))calloc(s->prog_cnt, s->prog_skel_sz);\n\ + s->progs = (struct bpf_prog_skeleton *)calloc(s->prog_cnt, s->prog_skel_sz);\n\ if (!s->progs) \n\ goto err; \n\ ", diff --git a/tools/bpf/bpftool/link.c b/tools/bpf/bpftool/link.c index 1b793759170e..a89f09e3c848 100644 --- a/tools/bpf/bpftool/link.c +++ b/tools/bpf/bpftool/link.c @@ -106,7 +106,7 @@ static int show_link_close_json(int fd, struct bpf_link_info *info) switch (info->type) { case BPF_LINK_TYPE_RAW_TRACEPOINT: jsonw_string_field(json_wtr, "tp_name", - (const char *)info->raw_tracepoint.tp_name); + u64_to_ptr(info->raw_tracepoint.tp_name)); break; case BPF_LINK_TYPE_TRACING: err = get_prog_info(info->prog_id, &prog_info); @@ -185,7 +185,7 @@ static int show_link_close_plain(int fd, struct bpf_link_info *info) switch (info->type) { case BPF_LINK_TYPE_RAW_TRACEPOINT: printf("\n\ttp '%s' ", - (const char *)info->raw_tracepoint.tp_name); + (const char *)u64_to_ptr(info->raw_tracepoint.tp_name)); break; case BPF_LINK_TYPE_TRACING: err = get_prog_info(info->prog_id, &prog_info); diff --git a/tools/bpf/bpftool/main.h b/tools/bpf/bpftool/main.h index e3a79b5a9960..c46e52137b87 100644 --- a/tools/bpf/bpftool/main.h +++ b/tools/bpf/bpftool/main.h @@ -21,7 +21,15 @@ /* Make sure we do not use kernel-only integer typedefs */ #pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64 -#define ptr_to_u64(ptr) ((__u64)(unsigned long)(ptr)) +static inline __u64 ptr_to_u64(const void *ptr) +{ + return (__u64)(unsigned long)ptr; +} + +static inline void *u64_to_ptr(__u64 ptr) +{ + return (void *)(unsigned long)ptr; +} #define NEXT_ARG() ({ argc--; argv++; if (argc < 0) usage(); }) #define NEXT_ARGP() ({ (*argc)--; (*argv)++; if (*argc < 0) usage(); }) diff --git a/tools/bpf/bpftool/pids.c b/tools/bpf/bpftool/pids.c index e3b116325403..df7d8ec76036 100644 --- a/tools/bpf/bpftool/pids.c +++ b/tools/bpf/bpftool/pids.c @@ -134,6 +134,8 @@ int build_obj_refs_table(struct obj_refs_table *table, enum bpf_obj_type type) while (true) { ret = read(fd, buf, sizeof(buf)); if (ret < 0) { + if (errno == EAGAIN) + continue; err = -errno; p_err("failed to read PID iterator output: %d", err); goto out; diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c index 158995d853b0..d393eb8263a6 100644 --- a/tools/bpf/bpftool/prog.c +++ b/tools/bpf/bpftool/prog.c @@ -428,14 +428,14 @@ prog_dump(struct bpf_prog_info *info, enum dump_mode mode, p_info("no instructions returned"); return -1; } - buf = (unsigned char *)(info->jited_prog_insns); + buf = u64_to_ptr(info->jited_prog_insns); member_len = info->jited_prog_len; } else { /* DUMP_XLATED */ if (info->xlated_prog_len == 0 || !info->xlated_prog_insns) { p_err("error retrieving insn dump: kernel.kptr_restrict set?"); return -1; } - buf = (unsigned char *)info->xlated_prog_insns; + buf = u64_to_ptr(info->xlated_prog_insns); member_len = info->xlated_prog_len; } @@ -444,7 +444,7 @@ prog_dump(struct bpf_prog_info *info, enum dump_mode mode, return -1; } - func_info = (void *)info->func_info; + func_info = u64_to_ptr(info->func_info); if (info->nr_line_info) { prog_linfo = bpf_prog_linfo__new(info); @@ -462,7 +462,7 @@ prog_dump(struct bpf_prog_info *info, enum dump_mode mode, n = write(fd, buf, member_len); close(fd); - if (n != member_len) { + if (n != (ssize_t)member_len) { p_err("error writing output file: %s", n < 0 ? strerror(errno) : "short write"); return -1; @@ -492,13 +492,13 @@ prog_dump(struct bpf_prog_info *info, enum dump_mode mode, __u32 i; if (info->nr_jited_ksyms) { kernel_syms_load(&dd); - ksyms = (__u64 *) info->jited_ksyms; + ksyms = u64_to_ptr(info->jited_ksyms); } if (json_output) jsonw_start_array(json_wtr); - lens = (__u32 *) info->jited_func_lens; + lens = u64_to_ptr(info->jited_func_lens); for (i = 0; i < info->nr_jited_func_lens; i++) { if (ksyms) { sym = kernel_syms_search(&dd, ksyms[i]); @@ -559,7 +559,7 @@ prog_dump(struct bpf_prog_info *info, enum dump_mode mode, } else { kernel_syms_load(&dd); dd.nr_jited_ksyms = info->nr_jited_ksyms; - dd.jited_ksyms = (__u64 *) info->jited_ksyms; + dd.jited_ksyms = u64_to_ptr(info->jited_ksyms); dd.btf = btf; dd.func_info = func_info; dd.finfo_rec_size = info->func_info_rec_size; @@ -1681,7 +1681,7 @@ static char *profile_target_name(int tgt_fd) goto out; } - func_info = (struct bpf_func_info *)(info_linear->info.func_info); + func_info = u64_to_ptr(info_linear->info.func_info); t = btf__type_by_id(btf, func_info[0].type_id); if (!t) { p_err("btf %d doesn't have type %d", diff --git a/tools/bpf/resolve_btfids/main.c b/tools/bpf/resolve_btfids/main.c index 4d9ecb975862..0def0bb1f783 100644 --- a/tools/bpf/resolve_btfids/main.c +++ b/tools/bpf/resolve_btfids/main.c @@ -233,6 +233,39 @@ static struct btf_id *add_symbol(struct rb_root *root, char *name, size_t size) return btf_id__add(root, id, false); } +/* + * The data of compressed section should be aligned to 4 + * (for 32bit) or 8 (for 64 bit) bytes. The binutils ld + * sets sh_addralign to 1, which makes libelf fail with + * misaligned section error during the update: + * FAILED elf_update(WRITE): invalid section alignment + * + * While waiting for ld fix, we fix the compressed sections + * sh_addralign value manualy. + */ +static int compressed_section_fix(Elf *elf, Elf_Scn *scn, GElf_Shdr *sh) +{ + int expected = gelf_getclass(elf) == ELFCLASS32 ? 4 : 8; + + if (!(sh->sh_flags & SHF_COMPRESSED)) + return 0; + + if (sh->sh_addralign == expected) + return 0; + + pr_debug2(" - fixing wrong alignment sh_addralign %u, expected %u\n", + sh->sh_addralign, expected); + + sh->sh_addralign = expected; + + if (gelf_update_shdr(scn, sh) == 0) { + printf("FAILED cannot update section header: %s\n", + elf_errmsg(-1)); + return -1; + } + return 0; +} + static int elf_collect(struct object *obj) { Elf_Scn *scn = NULL; @@ -309,6 +342,9 @@ static int elf_collect(struct object *obj) obj->efile.idlist_shndx = idx; obj->efile.idlist_addr = sh.sh_addr; } + + if (compressed_section_fix(elf, scn, &sh)) + return -1; } return 0; diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index 0480f893facd..b6238b2209b7 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -767,7 +767,7 @@ union bpf_attr { * * Also, note that **bpf_trace_printk**\ () is slow, and should * only be used for debugging purposes. For this reason, a notice - * bloc (spanning several lines) is printed to kernel logs and + * block (spanning several lines) is printed to kernel logs and * states that the helper should not be used "for production use" * the first time this helper is used (or more precisely, when * **trace_printk**\ () buffers are allocated). For passing values @@ -1033,14 +1033,14 @@ union bpf_attr { * * int ret; * struct bpf_tunnel_key key = {}; - * + * * ret = bpf_skb_get_tunnel_key(skb, &key, sizeof(key), 0); * if (ret < 0) * return TC_ACT_SHOT; // drop packet - * + * * if (key.remote_ipv4 != 0x0a000001) * return TC_ACT_SHOT; // drop packet - * + * * return TC_ACT_OK; // accept packet * * This interface can also be used with all encapsulation devices @@ -1147,7 +1147,7 @@ union bpf_attr { * Description * Retrieve the realm or the route, that is to say the * **tclassid** field of the destination for the *skb*. The - * indentifier retrieved is a user-provided tag, similar to the + * identifier retrieved is a user-provided tag, similar to the * one used with the net_cls cgroup (see description for * **bpf_get_cgroup_classid**\ () helper), but here this tag is * held by a route (a destination entry), not by a task. diff --git a/tools/include/uapi/linux/perf_event.h b/tools/include/uapi/linux/perf_event.h index 077e7ee69e3d..3e5dcdd48a49 100644 --- a/tools/include/uapi/linux/perf_event.h +++ b/tools/include/uapi/linux/perf_event.h @@ -1196,7 +1196,7 @@ union perf_mem_data_src { #define PERF_MEM_SNOOPX_FWD 0x01 /* forward */ /* 1 free */ -#define PERF_MEM_SNOOPX_SHIFT 37 +#define PERF_MEM_SNOOPX_SHIFT 38 /* locked instruction */ #define PERF_MEM_LOCK_NA 0x01 /* not available */ diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h index bc14db706b88..e9a4ecddb7a5 100644 --- a/tools/lib/bpf/bpf_helpers.h +++ b/tools/lib/bpf/bpf_helpers.h @@ -40,7 +40,7 @@ * Helper macro to manipulate data structures */ #ifndef offsetof -#define offsetof(TYPE, MEMBER) __builtin_offsetof(TYPE, MEMBER) +#define offsetof(TYPE, MEMBER) ((unsigned long)&((TYPE *)0)->MEMBER) #endif #ifndef container_of #define container_of(ptr, type, member) \ diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c index 4843e44916f7..7dfca7016aaa 100644 --- a/tools/lib/bpf/btf.c +++ b/tools/lib/bpf/btf.c @@ -41,6 +41,7 @@ struct btf { __u32 types_size; __u32 data_size; int fd; + int ptr_sz; }; static inline __u64 ptr_to_u64(const void *ptr) @@ -221,6 +222,70 @@ const struct btf_type *btf__type_by_id(const struct btf *btf, __u32 type_id) return btf->types[type_id]; } +static int determine_ptr_size(const struct btf *btf) +{ + const struct btf_type *t; + const char *name; + int i; + + for (i = 1; i <= btf->nr_types; i++) { + t = btf__type_by_id(btf, i); + if (!btf_is_int(t)) + continue; + + name = btf__name_by_offset(btf, t->name_off); + if (!name) + continue; + + if (strcmp(name, "long int") == 0 || + strcmp(name, "long unsigned int") == 0) { + if (t->size != 4 && t->size != 8) + continue; + return t->size; + } + } + + return -1; +} + +static size_t btf_ptr_sz(const struct btf *btf) +{ + if (!btf->ptr_sz) + ((struct btf *)btf)->ptr_sz = determine_ptr_size(btf); + return btf->ptr_sz < 0 ? sizeof(void *) : btf->ptr_sz; +} + +/* Return pointer size this BTF instance assumes. The size is heuristically + * determined by looking for 'long' or 'unsigned long' integer type and + * recording its size in bytes. If BTF type information doesn't have any such + * type, this function returns 0. In the latter case, native architecture's + * pointer size is assumed, so will be either 4 or 8, depending on + * architecture that libbpf was compiled for. It's possible to override + * guessed value by using btf__set_pointer_size() API. + */ +size_t btf__pointer_size(const struct btf *btf) +{ + if (!btf->ptr_sz) + ((struct btf *)btf)->ptr_sz = determine_ptr_size(btf); + + if (btf->ptr_sz < 0) + /* not enough BTF type info to guess */ + return 0; + + return btf->ptr_sz; +} + +/* Override or set pointer size in bytes. Only values of 4 and 8 are + * supported. + */ +int btf__set_pointer_size(struct btf *btf, size_t ptr_sz) +{ + if (ptr_sz != 4 && ptr_sz != 8) + return -EINVAL; + btf->ptr_sz = ptr_sz; + return 0; +} + static bool btf_type_is_void(const struct btf_type *t) { return t == &btf_void || btf_is_fwd(t); @@ -253,7 +318,7 @@ __s64 btf__resolve_size(const struct btf *btf, __u32 type_id) size = t->size; goto done; case BTF_KIND_PTR: - size = sizeof(void *); + size = btf_ptr_sz(btf); goto done; case BTF_KIND_TYPEDEF: case BTF_KIND_VOLATILE: @@ -293,9 +358,9 @@ int btf__align_of(const struct btf *btf, __u32 id) switch (kind) { case BTF_KIND_INT: case BTF_KIND_ENUM: - return min(sizeof(void *), (size_t)t->size); + return min(btf_ptr_sz(btf), (size_t)t->size); case BTF_KIND_PTR: - return sizeof(void *); + return btf_ptr_sz(btf); case BTF_KIND_TYPEDEF: case BTF_KIND_VOLATILE: case BTF_KIND_CONST: @@ -533,6 +598,18 @@ struct btf *btf__parse_elf(const char *path, struct btf_ext **btf_ext) if (IS_ERR(btf)) goto done; + switch (gelf_getclass(elf)) { + case ELFCLASS32: + btf__set_pointer_size(btf, 4); + break; + case ELFCLASS64: + btf__set_pointer_size(btf, 8); + break; + default: + pr_warn("failed to get ELF class (bitness) for %s\n", path); + break; + } + if (btf_ext && btf_ext_data) { *btf_ext = btf_ext__new(btf_ext_data->d_buf, btf_ext_data->d_size); diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h index f4a1a1d2b9a3..1ca14448df4c 100644 --- a/tools/lib/bpf/btf.h +++ b/tools/lib/bpf/btf.h @@ -76,6 +76,8 @@ LIBBPF_API __s32 btf__find_by_name_kind(const struct btf *btf, LIBBPF_API __u32 btf__get_nr_types(const struct btf *btf); LIBBPF_API const struct btf_type *btf__type_by_id(const struct btf *btf, __u32 id); +LIBBPF_API size_t btf__pointer_size(const struct btf *btf); +LIBBPF_API int btf__set_pointer_size(struct btf *btf, size_t ptr_sz); LIBBPF_API __s64 btf__resolve_size(const struct btf *btf, __u32 type_id); LIBBPF_API int btf__resolve_type(const struct btf *btf, __u32 type_id); LIBBPF_API int btf__align_of(const struct btf *btf, __u32 id); diff --git a/tools/lib/bpf/btf_dump.c b/tools/lib/bpf/btf_dump.c index cf711168d34a..57c00fa63932 100644 --- a/tools/lib/bpf/btf_dump.c +++ b/tools/lib/bpf/btf_dump.c @@ -13,6 +13,7 @@ #include <errno.h> #include <linux/err.h> #include <linux/btf.h> +#include <linux/kernel.h> #include "btf.h" #include "hashmap.h" #include "libbpf.h" @@ -60,6 +61,7 @@ struct btf_dump { const struct btf_ext *btf_ext; btf_dump_printf_fn_t printf_fn; struct btf_dump_opts opts; + int ptr_sz; bool strip_mods; /* per-type auxiliary state */ @@ -138,6 +140,7 @@ struct btf_dump *btf_dump__new(const struct btf *btf, d->btf_ext = btf_ext; d->printf_fn = printf_fn; d->opts.ctx = opts ? opts->ctx : NULL; + d->ptr_sz = btf__pointer_size(btf) ? : sizeof(void *); d->type_names = hashmap__new(str_hash_fn, str_equal_fn, NULL); if (IS_ERR(d->type_names)) { @@ -549,6 +552,9 @@ static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr) } } +static void btf_dump_emit_missing_aliases(struct btf_dump *d, __u32 id, + const struct btf_type *t); + static void btf_dump_emit_struct_fwd(struct btf_dump *d, __u32 id, const struct btf_type *t); static void btf_dump_emit_struct_def(struct btf_dump *d, __u32 id, @@ -671,6 +677,9 @@ static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id) switch (kind) { case BTF_KIND_INT: + /* Emit type alias definitions if necessary */ + btf_dump_emit_missing_aliases(d, id, t); + tstate->emit_state = EMITTED; break; case BTF_KIND_ENUM: @@ -797,7 +806,7 @@ static void btf_dump_emit_bit_padding(const struct btf_dump *d, int align, int lvl) { int off_diff = m_off - cur_off; - int ptr_bits = sizeof(void *) * 8; + int ptr_bits = d->ptr_sz * 8; if (off_diff <= 0) /* no gap */ @@ -870,7 +879,7 @@ static void btf_dump_emit_struct_def(struct btf_dump *d, btf_dump_printf(d, ": %d", m_sz); off = m_off + m_sz; } else { - m_sz = max(0, btf__resolve_size(d->btf, m->type)); + m_sz = max((__s64)0, btf__resolve_size(d->btf, m->type)); off = m_off + m_sz * 8; } btf_dump_printf(d, ";"); @@ -890,6 +899,32 @@ static void btf_dump_emit_struct_def(struct btf_dump *d, btf_dump_printf(d, " __attribute__((packed))"); } +static const char *missing_base_types[][2] = { + /* + * GCC emits typedefs to its internal __PolyX_t types when compiling Arm + * SIMD intrinsics. Alias them to standard base types. + */ + { "__Poly8_t", "unsigned char" }, + { "__Poly16_t", "unsigned short" }, + { "__Poly64_t", "unsigned long long" }, + { "__Poly128_t", "unsigned __int128" }, +}; + +static void btf_dump_emit_missing_aliases(struct btf_dump *d, __u32 id, + const struct btf_type *t) +{ + const char *name = btf_dump_type_name(d, id); + int i; + + for (i = 0; i < ARRAY_SIZE(missing_base_types); i++) { + if (strcmp(name, missing_base_types[i][0]) == 0) { + btf_dump_printf(d, "typedef %s %s;\n\n", + missing_base_types[i][1], name); + break; + } + } +} + static void btf_dump_emit_enum_fwd(struct btf_dump *d, __u32 id, const struct btf_type *t) { diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index 0a06124f7999..0ad0b0491e1f 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -2264,7 +2264,7 @@ static int bpf_object__init_user_btf_maps(struct bpf_object *obj, bool strict, data = elf_getdata(scn, NULL); if (!scn || !data) { pr_warn("failed to get Elf_Data from map section %d (%s)\n", - obj->efile.maps_shndx, MAPS_ELF_SEC); + obj->efile.btf_maps_shndx, MAPS_ELF_SEC); return -EINVAL; } @@ -2434,6 +2434,8 @@ static int bpf_object__init_btf(struct bpf_object *obj, BTF_ELF_SEC, err); goto out; } + /* enforce 8-byte pointers for BPF-targeted BTFs */ + btf__set_pointer_size(obj->btf, 8); err = 0; } if (btf_ext_data) { @@ -2542,6 +2544,8 @@ static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj) if (IS_ERR(kern_btf)) return PTR_ERR(kern_btf); + /* enforce 8-byte pointers for BPF-targeted BTFs */ + btf__set_pointer_size(obj->btf, 8); bpf_object__sanitize_btf(obj, kern_btf); } @@ -3478,10 +3482,11 @@ bpf_object__probe_global_data(struct bpf_object *obj) map = bpf_create_map_xattr(&map_attr); if (map < 0) { - cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg)); + ret = -errno; + cp = libbpf_strerror_r(ret, errmsg, sizeof(errmsg)); pr_warn("Error in %s():%s(%d). Couldn't create simple array map.\n", - __func__, cp, errno); - return -errno; + __func__, cp, -ret); + return ret; } insns[0].imm = map; @@ -5194,7 +5199,8 @@ static int bpf_object__collect_st_ops_relos(struct bpf_object *obj, static int bpf_object__collect_map_relos(struct bpf_object *obj, GElf_Shdr *shdr, Elf_Data *data) { - int i, j, nrels, new_sz, ptr_sz = sizeof(void *); + const int bpf_ptr_sz = 8, host_ptr_sz = sizeof(void *); + int i, j, nrels, new_sz; const struct btf_var_secinfo *vi = NULL; const struct btf_type *sec, *var, *def; const struct btf_member *member; @@ -5243,7 +5249,7 @@ static int bpf_object__collect_map_relos(struct bpf_object *obj, vi = btf_var_secinfos(sec) + map->btf_var_idx; if (vi->offset <= rel.r_offset && - rel.r_offset + sizeof(void *) <= vi->offset + vi->size) + rel.r_offset + bpf_ptr_sz <= vi->offset + vi->size) break; } if (j == obj->nr_maps) { @@ -5279,17 +5285,20 @@ static int bpf_object__collect_map_relos(struct bpf_object *obj, return -EINVAL; moff = rel.r_offset - vi->offset - moff; - if (moff % ptr_sz) + /* here we use BPF pointer size, which is always 64 bit, as we + * are parsing ELF that was built for BPF target + */ + if (moff % bpf_ptr_sz) return -EINVAL; - moff /= ptr_sz; + moff /= bpf_ptr_sz; if (moff >= map->init_slots_sz) { new_sz = moff + 1; - tmp = realloc(map->init_slots, new_sz * ptr_sz); + tmp = realloc(map->init_slots, new_sz * host_ptr_sz); if (!tmp) return -ENOMEM; map->init_slots = tmp; memset(map->init_slots + map->init_slots_sz, 0, - (new_sz - map->init_slots_sz) * ptr_sz); + (new_sz - map->init_slots_sz) * host_ptr_sz); map->init_slots_sz = new_sz; } map->init_slots[moff] = targ_map; @@ -6012,9 +6021,10 @@ int bpf_program__pin_instance(struct bpf_program *prog, const char *path, } if (bpf_obj_pin(prog->instances.fds[instance], path)) { - cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg)); + err = -errno; + cp = libbpf_strerror_r(err, errmsg, sizeof(errmsg)); pr_warn("failed to pin program: %s\n", cp); - return -errno; + return err; } pr_debug("pinned program '%s'\n", path); diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map index 0c4722bfdd0a..e35bd6cdbdbf 100644 --- a/tools/lib/bpf/libbpf.map +++ b/tools/lib/bpf/libbpf.map @@ -295,5 +295,7 @@ LIBBPF_0.1.0 { bpf_program__set_sk_lookup; btf__parse; btf__parse_raw; + btf__pointer_size; btf__set_fd; + btf__set_pointer_size; } LIBBPF_0.0.9; diff --git a/tools/lib/traceevent/event-parse.c b/tools/lib/traceevent/event-parse.c index 3ba566de821c..5acc18b32606 100644 --- a/tools/lib/traceevent/event-parse.c +++ b/tools/lib/traceevent/event-parse.c @@ -5259,7 +5259,7 @@ static int print_arg_pointer(struct trace_seq *s, const char *format, int plen, default: ret = 0; val = eval_num_arg(data, size, event, arg); - trace_seq_printf(s, "%p", (void *)val); + trace_seq_printf(s, "%p", (void *)(intptr_t)val); break; } diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt index 3f72d8e261f3..bd50cdff08a8 100644 --- a/tools/perf/Documentation/perf-record.txt +++ b/tools/perf/Documentation/perf-record.txt @@ -33,6 +33,10 @@ OPTIONS - a raw PMU event (eventsel+umask) in the form of rNNN where NNN is a hexadecimal event descriptor. + - a symbolic or raw PMU event followed by an optional colon + and a list of event modifiers, e.g., cpu-cycles:p. See the + linkperf:perf-list[1] man page for details on event modifiers. + - a symbolically formed PMU event like 'pmu/param1=0x3,param2/' where 'param1', 'param2', etc are defined as formats for the PMU in /sys/bus/event_source/devices/<pmu>/format/*. diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt index c9bfefc051fb..db420dd75e43 100644 --- a/tools/perf/Documentation/perf-stat.txt +++ b/tools/perf/Documentation/perf-stat.txt @@ -39,6 +39,10 @@ report:: - a raw PMU event (eventsel+umask) in the form of rNNN where NNN is a hexadecimal event descriptor. + - a symbolic or raw PMU event followed by an optional colon + and a list of event modifiers, e.g., cpu-cycles:p. See the + linkperf:perf-list[1] man page for details on event modifiers. + - a symbolically formed event like 'pmu/param1=0x3,param2/' where param1 and param2 are defined as formats for the PMU in /sys/bus/event_source/devices/<pmu>/format/* @@ -416,6 +420,9 @@ counts for all hardware threads in a core but show the sum counts per hardware thread. This is essentially a replacement for the any bit and convenient for post processing. +--summary:: +Print summary for interval mode (-I). + EXAMPLES -------- diff --git a/tools/perf/bench/synthesize.c b/tools/perf/bench/synthesize.c index 8d624aea1c5e..b2924e3181dc 100644 --- a/tools/perf/bench/synthesize.c +++ b/tools/perf/bench/synthesize.c @@ -162,8 +162,8 @@ static int do_run_multi_threaded(struct target *target, init_stats(&event_stats); for (i = 0; i < multi_iterations; i++) { session = perf_session__new(NULL, false, NULL); - if (!session) - return -ENOMEM; + if (IS_ERR(session)) + return PTR_ERR(session); atomic_set(&event_count, 0); gettimeofday(&start, NULL); diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index f91352f847c0..772f1057647f 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -2452,7 +2452,7 @@ static struct option __record_options[] = { OPT_BOOLEAN(0, "tail-synthesize", &record.opts.tail_synthesize, "synthesize non-sample events at the end of output"), OPT_BOOLEAN(0, "overwrite", &record.opts.overwrite, "use overwrite mode"), - OPT_BOOLEAN(0, "no-bpf-event", &record.opts.no_bpf_event, "record bpf events"), + OPT_BOOLEAN(0, "no-bpf-event", &record.opts.no_bpf_event, "do not record bpf events"), OPT_BOOLEAN(0, "strict-freq", &record.opts.strict_freq, "Fail if the specified frequency can't be used"), OPT_CALLBACK('F', "freq", &record.opts, "freq or 'max'", diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index ece1cddfcd7c..3c74c9c0f3c3 100644 --- a/tools/perf/builtin-report.c +++ b/tools/perf/builtin-report.c @@ -1332,6 +1332,9 @@ int cmd_report(int argc, const char **argv) if (report.mmaps_mode) report.tasks_mode = true; + if (dump_trace) + report.tool.ordered_events = false; + if (quiet) perf_quiet_option(); diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c index 0c7d599fa555..e6fc297cee91 100644 --- a/tools/perf/builtin-sched.c +++ b/tools/perf/builtin-sched.c @@ -2584,7 +2584,8 @@ static int timehist_sched_change_event(struct perf_tool *tool, } if (!sched->idle_hist || thread->tid == 0) { - timehist_update_runtime_stats(tr, t, tprev); + if (!cpu_list || test_bit(sample->cpu, cpu_bitmap)) + timehist_update_runtime_stats(tr, t, tprev); if (sched->idle_hist) { struct idle_thread_runtime *itr = (void *)tr; @@ -2857,6 +2858,9 @@ static void timehist_print_summary(struct perf_sched *sched, printf("\nIdle stats:\n"); for (i = 0; i < idle_max_cpu; ++i) { + if (cpu_list && !test_bit(i, cpu_bitmap)) + continue; + t = idle_threads[i]; if (!t) continue; diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 483a28ef4ec4..fddc97cac984 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -404,7 +404,7 @@ static void read_counters(struct timespec *rs) { struct evsel *counter; - if (!stat_config.summary && (read_affinity_counters(rs) < 0)) + if (!stat_config.stop_read_counter && (read_affinity_counters(rs) < 0)) return; evlist__for_each_entry(evsel_list, counter) { @@ -897,9 +897,9 @@ try_again_reset: if (stat_config.walltime_run_table) stat_config.walltime_run[run_idx] = t1 - t0; - if (interval) { + if (interval && stat_config.summary) { stat_config.interval = 0; - stat_config.summary = true; + stat_config.stop_read_counter = true; init_stats(&walltime_nsecs_stats); update_stats(&walltime_nsecs_stats, t1 - t0); @@ -1164,6 +1164,8 @@ static struct option stat_options[] = { "Use with 'percore' event qualifier to show the event " "counts of one hardware thread by sum up total hardware " "threads of same physical core"), + OPT_BOOLEAN(0, "summary", &stat_config.summary, + "print summary for interval mode"), #ifdef HAVE_LIBPFM OPT_CALLBACK(0, "pfm-events", &evsel_list, "event", "libpfm4 event selector. use 'perf list' to list available events", diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c index 994c230027bb..7c64134472c7 100644 --- a/tools/perf/builtin-top.c +++ b/tools/perf/builtin-top.c @@ -1746,6 +1746,7 @@ int cmd_top(int argc, const char **argv) goto out_delete_evlist; } +#ifdef HAVE_LIBBPF_SUPPORT if (!top.record_opts.no_bpf_event) { top.sb_evlist = evlist__new(); @@ -1759,6 +1760,7 @@ int cmd_top(int argc, const char **argv) goto out_delete_evlist; } } +#endif if (perf_evlist__start_sb_thread(top.sb_evlist, target)) { pr_debug("Couldn't start the BPF side band thread:\nBPF programs starting from now on won't be annotatable\n"); diff --git a/tools/perf/pmu-events/jevents.c b/tools/perf/pmu-events/jevents.c index fa86c5f997cc..fc9c158bfa13 100644 --- a/tools/perf/pmu-events/jevents.c +++ b/tools/perf/pmu-events/jevents.c @@ -137,7 +137,7 @@ static char *fixregex(char *s) return s; /* allocate space for a new string */ - fixed = (char *) malloc(len + 1); + fixed = (char *) malloc(len + esc_count + 1); if (!fixed) return NULL; diff --git a/tools/perf/tests/bpf.c b/tools/perf/tests/bpf.c index 5d20bf8397f0..cd77e334e577 100644 --- a/tools/perf/tests/bpf.c +++ b/tools/perf/tests/bpf.c @@ -197,7 +197,7 @@ static int do_test(struct bpf_object *obj, int (*func)(void), perf_mmap__read_done(&md->core); } - if (count != expect) { + if (count != expect * evlist->core.nr_entries) { pr_debug("BPF filter result incorrect, expected %d, got %d samples\n", expect, count); goto out_delete_evlist; } diff --git a/tools/perf/tests/parse-events.c b/tools/perf/tests/parse-events.c index 7f9f87a470c3..aae0fd9045c1 100644 --- a/tools/perf/tests/parse-events.c +++ b/tools/perf/tests/parse-events.c @@ -719,7 +719,7 @@ static int test__group2(struct evlist *evlist) TEST_ASSERT_VAL("wrong exclude_user", !evsel->core.attr.exclude_user); TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->core.attr.exclude_kernel); TEST_ASSERT_VAL("wrong exclude_hv", evsel->core.attr.exclude_hv); - TEST_ASSERT_VAL("wrong exclude guest", !evsel->core.attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude guest", evsel->core.attr.exclude_guest); TEST_ASSERT_VAL("wrong exclude host", !evsel->core.attr.exclude_host); TEST_ASSERT_VAL("wrong precise_ip", !evsel->core.attr.precise_ip); TEST_ASSERT_VAL("wrong leader", evsel__is_group_leader(evsel)); @@ -842,7 +842,7 @@ static int test__group3(struct evlist *evlist __maybe_unused) TEST_ASSERT_VAL("wrong exclude_user", !evsel->core.attr.exclude_user); TEST_ASSERT_VAL("wrong exclude_kernel", evsel->core.attr.exclude_kernel); TEST_ASSERT_VAL("wrong exclude_hv", evsel->core.attr.exclude_hv); - TEST_ASSERT_VAL("wrong exclude guest", !evsel->core.attr.exclude_guest); + TEST_ASSERT_VAL("wrong exclude guest", evsel->core.attr.exclude_guest); TEST_ASSERT_VAL("wrong exclude host", !evsel->core.attr.exclude_host); TEST_ASSERT_VAL("wrong precise_ip", !evsel->core.attr.precise_ip); TEST_ASSERT_VAL("wrong leader", evsel__is_group_leader(evsel)); diff --git a/tools/perf/tests/parse-metric.c b/tools/perf/tests/parse-metric.c index fc0838a7abc2..23db8acc492d 100644 --- a/tools/perf/tests/parse-metric.c +++ b/tools/perf/tests/parse-metric.c @@ -70,6 +70,9 @@ static struct pmu_event pme_test[] = { { .metric_expr = "1/m3", .metric_name = "M3", +}, +{ + .name = NULL, } }; diff --git a/tools/perf/ui/browsers/hists.c b/tools/perf/ui/browsers/hists.c index be9c4c0549bc..a07626f07208 100644 --- a/tools/perf/ui/browsers/hists.c +++ b/tools/perf/ui/browsers/hists.c @@ -3629,8 +3629,8 @@ int perf_evlist__tui_browse_hists(struct evlist *evlist, const char *help, { int nr_entries = evlist->core.nr_entries; -single_entry: if (perf_evlist__single_entry(evlist)) { +single_entry: { struct evsel *first = evlist__first(evlist); return perf_evsel__hists_browse(first, nr_entries, help, @@ -3638,6 +3638,7 @@ single_entry: env, warn_lost_event, annotation_opts); } + } if (symbol_conf.event_group) { struct evsel *pos; diff --git a/tools/perf/util/arm-spe-decoder/arm-spe-decoder.c b/tools/perf/util/arm-spe-decoder/arm-spe-decoder.c index 302a14d0aca9..93e063f22be5 100644 --- a/tools/perf/util/arm-spe-decoder/arm-spe-decoder.c +++ b/tools/perf/util/arm-spe-decoder/arm-spe-decoder.c @@ -182,15 +182,15 @@ static int arm_spe_read_record(struct arm_spe_decoder *decoder) if (payload & BIT(EV_TLB_ACCESS)) decoder->record.type |= ARM_SPE_TLB_ACCESS; - if ((idx == 1 || idx == 2 || idx == 3) && + if ((idx == 2 || idx == 4 || idx == 8) && (payload & BIT(EV_LLC_MISS))) decoder->record.type |= ARM_SPE_LLC_MISS; - if ((idx == 1 || idx == 2 || idx == 3) && + if ((idx == 2 || idx == 4 || idx == 8) && (payload & BIT(EV_LLC_ACCESS))) decoder->record.type |= ARM_SPE_LLC_ACCESS; - if ((idx == 1 || idx == 2 || idx == 3) && + if ((idx == 2 || idx == 4 || idx == 8) && (payload & BIT(EV_REMOTE_ACCESS))) decoder->record.type |= ARM_SPE_REMOTE_ACCESS; diff --git a/tools/perf/util/cs-etm.c b/tools/perf/util/cs-etm.c index c283223fb31f..a2a369e2fbb6 100644 --- a/tools/perf/util/cs-etm.c +++ b/tools/perf/util/cs-etm.c @@ -1344,8 +1344,15 @@ static int cs_etm__synth_events(struct cs_etm_auxtrace *etm, attr.sample_type &= ~(u64)PERF_SAMPLE_ADDR; } - if (etm->synth_opts.last_branch) + if (etm->synth_opts.last_branch) { attr.sample_type |= PERF_SAMPLE_BRANCH_STACK; + /* + * We don't use the hardware index, but the sample generation + * code uses the new format branch_stack with this field, + * so the event attributes must indicate that it's present. + */ + attr.branch_sample_type |= PERF_SAMPLE_BRANCH_HW_INDEX; + } if (etm->synth_opts.instructions) { attr.config = PERF_COUNT_HW_INSTRUCTIONS; diff --git a/tools/perf/util/intel-pt.c b/tools/perf/util/intel-pt.c index 2a8d245351e7..0af4e81c46e2 100644 --- a/tools/perf/util/intel-pt.c +++ b/tools/perf/util/intel-pt.c @@ -3017,8 +3017,15 @@ static int intel_pt_synth_events(struct intel_pt *pt, if (pt->synth_opts.callchain) attr.sample_type |= PERF_SAMPLE_CALLCHAIN; - if (pt->synth_opts.last_branch) + if (pt->synth_opts.last_branch) { attr.sample_type |= PERF_SAMPLE_BRANCH_STACK; + /* + * We don't use the hardware index, but the sample generation + * code uses the new format branch_stack with this field, + * so the event attributes must indicate that it's present. + */ + attr.branch_sample_type |= PERF_SAMPLE_BRANCH_HW_INDEX; + } if (pt->synth_opts.instructions) { attr.config = PERF_COUNT_HW_INSTRUCTIONS; diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 208b813e00ea..85587de027a5 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -736,12 +736,6 @@ int machine__process_switch_event(struct machine *machine __maybe_unused, return 0; } -static int is_bpf_image(const char *name) -{ - return strncmp(name, "bpf_trampoline_", sizeof("bpf_trampoline_") - 1) == 0 || - strncmp(name, "bpf_dispatcher_", sizeof("bpf_dispatcher_") - 1) == 0; -} - static int machine__process_ksymbol_register(struct machine *machine, union perf_event *event, struct perf_sample *sample __maybe_unused) diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c index 1d7210804639..cc0faf8f1321 100644 --- a/tools/perf/util/map.c +++ b/tools/perf/util/map.c @@ -267,6 +267,22 @@ bool __map__is_bpf_prog(const struct map *map) return name && (strstr(name, "bpf_prog_") == name); } +bool __map__is_bpf_image(const struct map *map) +{ + const char *name; + + if (map->dso->binary_type == DSO_BINARY_TYPE__BPF_IMAGE) + return true; + + /* + * If PERF_RECORD_KSYMBOL is not included, the dso will not have + * type of DSO_BINARY_TYPE__BPF_IMAGE. In such cases, we can + * guess the type based on name. + */ + name = map->dso->short_name; + return name && is_bpf_image(name); +} + bool __map__is_ool(const struct map *map) { return map->dso && map->dso->binary_type == DSO_BINARY_TYPE__OOL; diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h index 9e312ae2d656..c2f5d28fe73a 100644 --- a/tools/perf/util/map.h +++ b/tools/perf/util/map.h @@ -147,12 +147,14 @@ int map__set_kallsyms_ref_reloc_sym(struct map *map, const char *symbol_name, bool __map__is_kernel(const struct map *map); bool __map__is_extra_kernel_map(const struct map *map); bool __map__is_bpf_prog(const struct map *map); +bool __map__is_bpf_image(const struct map *map); bool __map__is_ool(const struct map *map); static inline bool __map__is_kmodule(const struct map *map) { return !__map__is_kernel(map) && !__map__is_extra_kernel_map(map) && - !__map__is_bpf_prog(map) && !__map__is_ool(map); + !__map__is_bpf_prog(map) && !__map__is_ool(map) && + !__map__is_bpf_image(map); } bool map__has_symbols(const struct map *map); @@ -164,4 +166,9 @@ static inline bool is_entry_trampoline(const char *name) return !strcmp(name, ENTRY_TRAMPOLINE_NAME); } +static inline bool is_bpf_image(const char *name) +{ + return strncmp(name, "bpf_trampoline_", sizeof("bpf_trampoline_") - 1) == 0 || + strncmp(name, "bpf_dispatcher_", sizeof("bpf_dispatcher_") - 1) == 0; +} #endif /* __PERF_MAP_H */ diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c index 9f7260e69113..c4d2394e2b2d 100644 --- a/tools/perf/util/parse-events.c +++ b/tools/perf/util/parse-events.c @@ -37,6 +37,7 @@ #include "util/evsel_config.h" #include "util/event.h" #include "util/pfm.h" +#include "perf.h" #define MAX_NAME_LEN 100 @@ -1533,19 +1534,23 @@ int parse_events_add_pmu(struct parse_events_state *parse_state, evsel = __add_event(list, &parse_state->idx, &attr, true, get_config_name(head_config), pmu, &config_terms, auto_merge_stats, NULL); - if (evsel) { - evsel->unit = info.unit; - evsel->scale = info.scale; - evsel->per_pkg = info.per_pkg; - evsel->snapshot = info.snapshot; - evsel->metric_expr = info.metric_expr; - evsel->metric_name = info.metric_name; - evsel->pmu_name = name ? strdup(name) : NULL; - evsel->use_uncore_alias = use_uncore_alias; - evsel->percore = config_term_percore(&evsel->config_terms); - } + if (!evsel) + return -ENOMEM; + + evsel->pmu_name = name ? strdup(name) : NULL; + evsel->use_uncore_alias = use_uncore_alias; + evsel->percore = config_term_percore(&evsel->config_terms); - return evsel ? 0 : -ENOMEM; + if (parse_state->fake_pmu) + return 0; + + evsel->unit = info.unit; + evsel->scale = info.scale; + evsel->per_pkg = info.per_pkg; + evsel->snapshot = info.snapshot; + evsel->metric_expr = info.metric_expr; + evsel->metric_name = info.metric_name; + return 0; } int parse_events_multi_pmu_add(struct parse_events_state *parse_state, @@ -1794,6 +1799,8 @@ static int get_event_modifier(struct event_modifier *mod, char *str, if (*str == 'u') { if (!exclude) exclude = eu = ek = eh = 1; + if (!exclude_GH && !perf_guest) + eG = 1; eu = 0; } else if (*str == 'k') { if (!exclude) diff --git a/tools/perf/util/parse-events.y b/tools/perf/util/parse-events.y index b9fb91fdc5de..645bf4f1859f 100644 --- a/tools/perf/util/parse-events.y +++ b/tools/perf/util/parse-events.y @@ -511,7 +511,7 @@ PE_PREFIX_MEM PE_VALUE '/' PE_VALUE ':' PE_MODIFIER_BP sep_dc list = alloc_list(); ABORT_ON(!list); err = parse_events_add_breakpoint(list, &parse_state->idx, - (void *) $2, $6, $4); + (void *)(uintptr_t) $2, $6, $4); free($6); if (err) { free(list); @@ -528,7 +528,7 @@ PE_PREFIX_MEM PE_VALUE '/' PE_VALUE sep_dc list = alloc_list(); ABORT_ON(!list); if (parse_events_add_breakpoint(list, &parse_state->idx, - (void *) $2, NULL, $4)) { + (void *)(uintptr_t) $2, NULL, $4)) { free(list); YYABORT; } @@ -544,7 +544,7 @@ PE_PREFIX_MEM PE_VALUE ':' PE_MODIFIER_BP sep_dc list = alloc_list(); ABORT_ON(!list); err = parse_events_add_breakpoint(list, &parse_state->idx, - (void *) $2, $4, 0); + (void *)(uintptr_t) $2, $4, 0); free($4); if (err) { free(list); @@ -561,7 +561,7 @@ PE_PREFIX_MEM PE_VALUE sep_dc list = alloc_list(); ABORT_ON(!list); if (parse_events_add_breakpoint(list, &parse_state->idx, - (void *) $2, NULL, 0)) { + (void *)(uintptr_t) $2, NULL, 0)) { free(list); YYABORT; } diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c index ffbc9d35a383..7a5f03764702 100644 --- a/tools/perf/util/session.c +++ b/tools/perf/util/session.c @@ -87,7 +87,7 @@ static int perf_session__process_compressed_event(struct perf_session *session, session->decomp_last = decomp; } - pr_debug("decomp (B): %ld to %ld\n", src_size, decomp_size); + pr_debug("decomp (B): %zd to %zd\n", src_size, decomp_size); return 0; } diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c index 57d0706e1330..493ec372fdec 100644 --- a/tools/perf/util/stat-display.c +++ b/tools/perf/util/stat-display.c @@ -117,7 +117,7 @@ static void aggr_printout(struct perf_stat_config *config, cpu_map__id_to_die(id), config->csv_output ? 0 : -3, cpu_map__id_to_cpu(id), config->csv_sep); - } else { + } else if (id > -1) { fprintf(config->output, "CPU%*d%s", config->csv_output ? 0 : -7, evsel__cpus(evsel)->map[id], diff --git a/tools/perf/util/stat.h b/tools/perf/util/stat.h index f8778cffd941..aa3bed48511b 100644 --- a/tools/perf/util/stat.h +++ b/tools/perf/util/stat.h @@ -113,6 +113,7 @@ struct perf_stat_config { bool summary; bool metric_no_group; bool metric_no_merge; + bool stop_read_counter; FILE *output; unsigned int interval; unsigned int timeout; diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c index 1f5fcb828a21..5151a8c0b791 100644 --- a/tools/perf/util/symbol.c +++ b/tools/perf/util/symbol.c @@ -663,6 +663,7 @@ static bool symbol__is_idle(const char *name) "exit_idle", "mwait_idle", "mwait_idle_with_hints", + "mwait_idle_with_hints.constprop.0", "poll_idle", "ppc64_runlatch_off", "pseries_dedicated_idle_sleep", diff --git a/tools/perf/util/zstd.c b/tools/perf/util/zstd.c index d2202392ffdb..48dd2b018c47 100644 --- a/tools/perf/util/zstd.c +++ b/tools/perf/util/zstd.c @@ -99,7 +99,7 @@ size_t zstd_decompress_stream(struct zstd_data *data, void *src, size_t src_size while (input.pos < input.size) { ret = ZSTD_decompressStream(data->dstream, &output, &input); if (ZSTD_isError(ret)) { - pr_err("failed to decompress (B): %ld -> %ld, dst_size %ld : %s\n", + pr_err("failed to decompress (B): %zd -> %zd, dst_size %zd : %s\n", src_size, output.size, dst_size, ZSTD_getErrorName(ret)); break; } diff --git a/tools/testing/selftests/bpf/.gitignore b/tools/testing/selftests/bpf/.gitignore index 1bb204cee853..9a0946ddb705 100644 --- a/tools/testing/selftests/bpf/.gitignore +++ b/tools/testing/selftests/bpf/.gitignore @@ -6,7 +6,6 @@ test_lpm_map test_tag FEATURE-DUMP.libbpf fixdep -test_align test_dev_cgroup /test_progs* test_tcpbpf_user diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index a83b5827532f..fc946b7ac288 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -32,7 +32,7 @@ LDLIBS += -lcap -lelf -lz -lrt -lpthread # Order correspond to 'make run_tests' order TEST_GEN_PROGS = test_verifier test_tag test_maps test_lru_map test_lpm_map test_progs \ - test_align test_verifier_log test_dev_cgroup test_tcpbpf_user \ + test_verifier_log test_dev_cgroup test_tcpbpf_user \ test_sock test_btf test_sockmap get_cgroup_id_user test_socket_cookie \ test_cgroup_storage \ test_netcnt test_tcpnotify_user test_sock_fields test_sysctl \ diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_obj_id.c b/tools/testing/selftests/bpf/prog_tests/bpf_obj_id.c index 7afa4160416f..284d5921c345 100644 --- a/tools/testing/selftests/bpf/prog_tests/bpf_obj_id.c +++ b/tools/testing/selftests/bpf/prog_tests/bpf_obj_id.c @@ -159,15 +159,15 @@ void test_bpf_obj_id(void) /* Check getting link info */ info_len = sizeof(struct bpf_link_info) * 2; bzero(&link_infos[i], info_len); - link_infos[i].raw_tracepoint.tp_name = (__u64)&tp_name; + link_infos[i].raw_tracepoint.tp_name = ptr_to_u64(&tp_name); link_infos[i].raw_tracepoint.tp_name_len = sizeof(tp_name); err = bpf_obj_get_info_by_fd(bpf_link__fd(links[i]), &link_infos[i], &info_len); if (CHECK(err || link_infos[i].type != BPF_LINK_TYPE_RAW_TRACEPOINT || link_infos[i].prog_id != prog_infos[i].id || - link_infos[i].raw_tracepoint.tp_name != (__u64)&tp_name || - strcmp((char *)link_infos[i].raw_tracepoint.tp_name, + link_infos[i].raw_tracepoint.tp_name != ptr_to_u64(&tp_name) || + strcmp(u64_to_ptr(link_infos[i].raw_tracepoint.tp_name), "sys_enter") || info_len != sizeof(struct bpf_link_info), "get-link-info(fd)", @@ -178,7 +178,7 @@ void test_bpf_obj_id(void) link_infos[i].type, BPF_LINK_TYPE_RAW_TRACEPOINT, link_infos[i].id, link_infos[i].prog_id, prog_infos[i].id, - (char *)link_infos[i].raw_tracepoint.tp_name, + (const char *)u64_to_ptr(link_infos[i].raw_tracepoint.tp_name), "sys_enter")) goto done; diff --git a/tools/testing/selftests/bpf/prog_tests/btf_dump.c b/tools/testing/selftests/bpf/prog_tests/btf_dump.c index cb33a7ee4e04..39fb81d9daeb 100644 --- a/tools/testing/selftests/bpf/prog_tests/btf_dump.c +++ b/tools/testing/selftests/bpf/prog_tests/btf_dump.c @@ -12,15 +12,16 @@ void btf_dump_printf(void *ctx, const char *fmt, va_list args) static struct btf_dump_test_case { const char *name; const char *file; + bool known_ptr_sz; struct btf_dump_opts opts; } btf_dump_test_cases[] = { - {"btf_dump: syntax", "btf_dump_test_case_syntax", {}}, - {"btf_dump: ordering", "btf_dump_test_case_ordering", {}}, - {"btf_dump: padding", "btf_dump_test_case_padding", {}}, - {"btf_dump: packing", "btf_dump_test_case_packing", {}}, - {"btf_dump: bitfields", "btf_dump_test_case_bitfields", {}}, - {"btf_dump: multidim", "btf_dump_test_case_multidim", {}}, - {"btf_dump: namespacing", "btf_dump_test_case_namespacing", {}}, + {"btf_dump: syntax", "btf_dump_test_case_syntax", true, {}}, + {"btf_dump: ordering", "btf_dump_test_case_ordering", false, {}}, + {"btf_dump: padding", "btf_dump_test_case_padding", true, {}}, + {"btf_dump: packing", "btf_dump_test_case_packing", true, {}}, + {"btf_dump: bitfields", "btf_dump_test_case_bitfields", true, {}}, + {"btf_dump: multidim", "btf_dump_test_case_multidim", false, {}}, + {"btf_dump: namespacing", "btf_dump_test_case_namespacing", false, {}}, }; static int btf_dump_all_types(const struct btf *btf, @@ -62,6 +63,18 @@ static int test_btf_dump_case(int n, struct btf_dump_test_case *t) goto done; } + /* tests with t->known_ptr_sz have no "long" or "unsigned long" type, + * so it's impossible to determine correct pointer size; but if they + * do, it should be 8 regardless of host architecture, becaues BPF + * target is always 64-bit + */ + if (!t->known_ptr_sz) { + btf__set_pointer_size(btf, 8); + } else { + CHECK(btf__pointer_size(btf) != 8, "ptr_sz", "exp %d, got %zu\n", + 8, btf__pointer_size(btf)); + } + snprintf(out_file, sizeof(out_file), "/tmp/%s.output.XXXXXX", t->file); fd = mkstemp(out_file); if (CHECK(fd < 0, "create_tmp", "failed to create file: %d\n", fd)) { diff --git a/tools/testing/selftests/bpf/prog_tests/core_extern.c b/tools/testing/selftests/bpf/prog_tests/core_extern.c index b093787e9448..1931a158510e 100644 --- a/tools/testing/selftests/bpf/prog_tests/core_extern.c +++ b/tools/testing/selftests/bpf/prog_tests/core_extern.c @@ -159,8 +159,8 @@ void test_core_extern(void) exp = (uint64_t *)&t->data; for (j = 0; j < n; j++) { CHECK(got[j] != exp[j], "check_res", - "result #%d: expected %lx, but got %lx\n", - j, exp[j], got[j]); + "result #%d: expected %llx, but got %llx\n", + j, (__u64)exp[j], (__u64)got[j]); } cleanup: test_core_extern__destroy(skel); diff --git a/tools/testing/selftests/bpf/prog_tests/core_reloc.c b/tools/testing/selftests/bpf/prog_tests/core_reloc.c index 084ed26a7d78..a54eafc5e4b3 100644 --- a/tools/testing/selftests/bpf/prog_tests/core_reloc.c +++ b/tools/testing/selftests/bpf/prog_tests/core_reloc.c @@ -237,8 +237,8 @@ .union_sz = sizeof(((type *)0)->union_field), \ .arr_sz = sizeof(((type *)0)->arr_field), \ .arr_elem_sz = sizeof(((type *)0)->arr_field[0]), \ - .ptr_sz = sizeof(((type *)0)->ptr_field), \ - .enum_sz = sizeof(((type *)0)->enum_field), \ + .ptr_sz = 8, /* always 8-byte pointer for BPF */ \ + .enum_sz = sizeof(((type *)0)->enum_field), \ } #define SIZE_CASE(name) { \ @@ -432,20 +432,20 @@ static struct core_reloc_test_case test_cases[] = { .sb4 = -1, .sb20 = -0x17654321, .u32 = 0xBEEF, - .s32 = -0x3FEDCBA987654321, + .s32 = -0x3FEDCBA987654321LL, }), BITFIELDS_CASE(bitfields___bitfield_vs_int, { - .ub1 = 0xFEDCBA9876543210, + .ub1 = 0xFEDCBA9876543210LL, .ub2 = 0xA6, - .ub7 = -0x7EDCBA987654321, - .sb4 = -0x6123456789ABCDE, - .sb20 = 0xD00D, + .ub7 = -0x7EDCBA987654321LL, + .sb4 = -0x6123456789ABCDELL, + .sb20 = 0xD00DLL, .u32 = -0x76543, - .s32 = 0x0ADEADBEEFBADB0B, + .s32 = 0x0ADEADBEEFBADB0BLL, }), BITFIELDS_CASE(bitfields___just_big_enough, { - .ub1 = 0xF, - .ub2 = 0x0812345678FEDCBA, + .ub1 = 0xFLL, + .ub2 = 0x0812345678FEDCBALL, }), BITFIELDS_ERR_CASE(bitfields___err_too_big_bitfield), diff --git a/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c b/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c index a895bfed55db..197d0d217b56 100644 --- a/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c +++ b/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c @@ -16,7 +16,7 @@ static void test_fexit_bpf2bpf_common(const char *obj_file, __u32 duration = 0, retval; struct bpf_map *data_map; const int zero = 0; - u64 *result = NULL; + __u64 *result = NULL; err = bpf_prog_load(target_obj_file, BPF_PROG_TYPE_UNSPEC, &pkt_obj, &pkt_fd); @@ -29,7 +29,7 @@ static void test_fexit_bpf2bpf_common(const char *obj_file, link = calloc(sizeof(struct bpf_link *), prog_cnt); prog = calloc(sizeof(struct bpf_program *), prog_cnt); - result = malloc((prog_cnt + 32 /* spare */) * sizeof(u64)); + result = malloc((prog_cnt + 32 /* spare */) * sizeof(__u64)); if (CHECK(!link || !prog || !result, "alloc_memory", "failed to alloc memory")) goto close_prog; @@ -72,7 +72,7 @@ static void test_fexit_bpf2bpf_common(const char *obj_file, goto close_prog; for (i = 0; i < prog_cnt; i++) - if (CHECK(result[i] != 1, "result", "fexit_bpf2bpf failed err %ld\n", + if (CHECK(result[i] != 1, "result", "fexit_bpf2bpf failed err %llu\n", result[i])) goto close_prog; diff --git a/tools/testing/selftests/bpf/prog_tests/flow_dissector.c b/tools/testing/selftests/bpf/prog_tests/flow_dissector.c index f11f187990e9..cd6dc80edf18 100644 --- a/tools/testing/selftests/bpf/prog_tests/flow_dissector.c +++ b/tools/testing/selftests/bpf/prog_tests/flow_dissector.c @@ -591,7 +591,7 @@ void test_flow_dissector(void) CHECK_ATTR(tattr.data_size_out != sizeof(flow_keys) || err || tattr.retval != 1, tests[i].name, - "err %d errno %d retval %d duration %d size %u/%lu\n", + "err %d errno %d retval %d duration %d size %u/%zu\n", err, errno, tattr.retval, tattr.duration, tattr.data_size_out, sizeof(flow_keys)); CHECK_FLOW_KEYS(tests[i].name, flow_keys, tests[i].keys); diff --git a/tools/testing/selftests/bpf/prog_tests/global_data.c b/tools/testing/selftests/bpf/prog_tests/global_data.c index e3cb62b0a110..9efa7e50eab2 100644 --- a/tools/testing/selftests/bpf/prog_tests/global_data.c +++ b/tools/testing/selftests/bpf/prog_tests/global_data.c @@ -5,7 +5,7 @@ static void test_global_data_number(struct bpf_object *obj, __u32 duration) { int i, err, map_fd; - uint64_t num; + __u64 num; map_fd = bpf_find_map(__func__, obj, "result_number"); if (CHECK_FAIL(map_fd < 0)) @@ -14,7 +14,7 @@ static void test_global_data_number(struct bpf_object *obj, __u32 duration) struct { char *name; uint32_t key; - uint64_t num; + __u64 num; } tests[] = { { "relocate .bss reference", 0, 0 }, { "relocate .data reference", 1, 42 }, @@ -32,7 +32,7 @@ static void test_global_data_number(struct bpf_object *obj, __u32 duration) for (i = 0; i < sizeof(tests) / sizeof(tests[0]); i++) { err = bpf_map_lookup_elem(map_fd, &tests[i].key, &num); CHECK(err || num != tests[i].num, tests[i].name, - "err %d result %lx expected %lx\n", + "err %d result %llx expected %llx\n", err, num, tests[i].num); } } diff --git a/tools/testing/selftests/bpf/prog_tests/mmap.c b/tools/testing/selftests/bpf/prog_tests/mmap.c index 43d0b5578f46..9c3c5c0f068f 100644 --- a/tools/testing/selftests/bpf/prog_tests/mmap.c +++ b/tools/testing/selftests/bpf/prog_tests/mmap.c @@ -21,7 +21,7 @@ void test_mmap(void) const long page_size = sysconf(_SC_PAGE_SIZE); int err, duration = 0, i, data_map_fd, data_map_id, tmp_fd, rdmap_fd; struct bpf_map *data_map, *bss_map; - void *bss_mmaped = NULL, *map_mmaped = NULL, *tmp1, *tmp2; + void *bss_mmaped = NULL, *map_mmaped = NULL, *tmp0, *tmp1, *tmp2; struct test_mmap__bss *bss_data; struct bpf_map_info map_info; __u32 map_info_sz = sizeof(map_info); @@ -183,16 +183,23 @@ void test_mmap(void) /* check some more advanced mmap() manipulations */ + tmp0 = mmap(NULL, 4 * page_size, PROT_READ, MAP_SHARED | MAP_ANONYMOUS, + -1, 0); + if (CHECK(tmp0 == MAP_FAILED, "adv_mmap0", "errno %d\n", errno)) + goto cleanup; + /* map all but last page: pages 1-3 mapped */ - tmp1 = mmap(NULL, 3 * page_size, PROT_READ, MAP_SHARED, + tmp1 = mmap(tmp0, 3 * page_size, PROT_READ, MAP_SHARED | MAP_FIXED, data_map_fd, 0); - if (CHECK(tmp1 == MAP_FAILED, "adv_mmap1", "errno %d\n", errno)) + if (CHECK(tmp0 != tmp1, "adv_mmap1", "tmp0: %p, tmp1: %p\n", tmp0, tmp1)) { + munmap(tmp0, 4 * page_size); goto cleanup; + } /* unmap second page: pages 1, 3 mapped */ err = munmap(tmp1 + page_size, page_size); if (CHECK(err, "adv_mmap2", "errno %d\n", errno)) { - munmap(tmp1, map_sz); + munmap(tmp1, 4 * page_size); goto cleanup; } @@ -201,7 +208,7 @@ void test_mmap(void) MAP_SHARED | MAP_FIXED, data_map_fd, 0); if (CHECK(tmp2 == MAP_FAILED, "adv_mmap3", "errno %d\n", errno)) { munmap(tmp1, page_size); - munmap(tmp1 + 2*page_size, page_size); + munmap(tmp1 + 2*page_size, 2 * page_size); goto cleanup; } CHECK(tmp1 + page_size != tmp2, "adv_mmap4", @@ -211,7 +218,7 @@ void test_mmap(void) tmp2 = mmap(tmp1, 4 * page_size, PROT_READ, MAP_SHARED | MAP_FIXED, data_map_fd, 0); if (CHECK(tmp2 == MAP_FAILED, "adv_mmap5", "errno %d\n", errno)) { - munmap(tmp1, 3 * page_size); /* unmap page 1 */ + munmap(tmp1, 4 * page_size); /* unmap page 1 */ goto cleanup; } CHECK(tmp1 != tmp2, "adv_mmap6", "tmp1: %p, tmp2: %p\n", tmp1, tmp2); diff --git a/tools/testing/selftests/bpf/prog_tests/prog_run_xattr.c b/tools/testing/selftests/bpf/prog_tests/prog_run_xattr.c index dde2b7ae7bc9..935a294f049a 100644 --- a/tools/testing/selftests/bpf/prog_tests/prog_run_xattr.c +++ b/tools/testing/selftests/bpf/prog_tests/prog_run_xattr.c @@ -28,7 +28,7 @@ void test_prog_run_xattr(void) "err %d errno %d retval %d\n", err, errno, tattr.retval); CHECK_ATTR(tattr.data_size_out != sizeof(pkt_v4), "data_size_out", - "incorrect output size, want %lu have %u\n", + "incorrect output size, want %zu have %u\n", sizeof(pkt_v4), tattr.data_size_out); CHECK_ATTR(buf[5] != 0, "overflow", diff --git a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c index c571584c00f5..9ff0412e1fd3 100644 --- a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c +++ b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c @@ -309,6 +309,7 @@ static void v4_to_v6(struct sockaddr_storage *ss) v6->sin6_addr.s6_addr[10] = 0xff; v6->sin6_addr.s6_addr[11] = 0xff; memcpy(&v6->sin6_addr.s6_addr[12], &v4.sin_addr.s_addr, 4); + memset(&v6->sin6_addr.s6_addr[0], 0, 10); } static int udp_recv_send(int server_fd) diff --git a/tools/testing/selftests/bpf/prog_tests/skb_ctx.c b/tools/testing/selftests/bpf/prog_tests/skb_ctx.c index 25de86af2d03..fafeddaad6a9 100644 --- a/tools/testing/selftests/bpf/prog_tests/skb_ctx.c +++ b/tools/testing/selftests/bpf/prog_tests/skb_ctx.c @@ -81,7 +81,7 @@ void test_skb_ctx(void) CHECK_ATTR(tattr.ctx_size_out != sizeof(skb), "ctx_size_out", - "incorrect output size, want %lu have %u\n", + "incorrect output size, want %zu have %u\n", sizeof(skb), tattr.ctx_size_out); for (i = 0; i < 5; i++) diff --git a/tools/testing/selftests/bpf/prog_tests/test_global_funcs.c b/tools/testing/selftests/bpf/prog_tests/test_global_funcs.c index 25b068591e9a..193002b14d7f 100644 --- a/tools/testing/selftests/bpf/prog_tests/test_global_funcs.c +++ b/tools/testing/selftests/bpf/prog_tests/test_global_funcs.c @@ -19,7 +19,7 @@ static int libbpf_debug_print(enum libbpf_print_level level, log_buf = va_arg(args, char *); if (!log_buf) goto out; - if (strstr(log_buf, err_str) == 0) + if (err_str && strstr(log_buf, err_str) == 0) found = true; out: printf(format, log_buf); diff --git a/tools/testing/selftests/bpf/prog_tests/varlen.c b/tools/testing/selftests/bpf/prog_tests/varlen.c index c75525eab02c..dd324b4933db 100644 --- a/tools/testing/selftests/bpf/prog_tests/varlen.c +++ b/tools/testing/selftests/bpf/prog_tests/varlen.c @@ -44,25 +44,25 @@ void test_varlen(void) CHECK_VAL(bss->payload1_len2, size2); CHECK_VAL(bss->total1, size1 + size2); CHECK(memcmp(bss->payload1, exp_str, size1 + size2), "content_check", - "doesn't match!"); + "doesn't match!\n"); CHECK_VAL(data->payload2_len1, size1); CHECK_VAL(data->payload2_len2, size2); CHECK_VAL(data->total2, size1 + size2); CHECK(memcmp(data->payload2, exp_str, size1 + size2), "content_check", - "doesn't match!"); + "doesn't match!\n"); CHECK_VAL(data->payload3_len1, size1); CHECK_VAL(data->payload3_len2, size2); CHECK_VAL(data->total3, size1 + size2); CHECK(memcmp(data->payload3, exp_str, size1 + size2), "content_check", - "doesn't match!"); + "doesn't match!\n"); CHECK_VAL(data->payload4_len1, size1); CHECK_VAL(data->payload4_len2, size2); CHECK_VAL(data->total4, size1 + size2); CHECK(memcmp(data->payload4, exp_str, size1 + size2), "content_check", - "doesn't match!"); + "doesn't match!\n"); cleanup: test_varlen__destroy(skel); } diff --git a/tools/testing/selftests/bpf/progs/core_reloc_types.h b/tools/testing/selftests/bpf/progs/core_reloc_types.h index 34d84717c946..69139ed66216 100644 --- a/tools/testing/selftests/bpf/progs/core_reloc_types.h +++ b/tools/testing/selftests/bpf/progs/core_reloc_types.h @@ -1,5 +1,10 @@ #include <stdint.h> #include <stdbool.h> + +void preserce_ptr_sz_fn(long x) {} + +#define __bpf_aligned __attribute__((aligned(8))) + /* * KERNEL */ @@ -444,51 +449,51 @@ struct core_reloc_primitives { char a; int b; enum core_reloc_primitives_enum c; - void *d; - int (*f)(const char *); + void *d __bpf_aligned; + int (*f)(const char *) __bpf_aligned; }; struct core_reloc_primitives___diff_enum_def { char a; int b; - void *d; - int (*f)(const char *); + void *d __bpf_aligned; + int (*f)(const char *) __bpf_aligned; enum { X = 100, Y = 200, - } c; /* inline enum def with differing set of values */ + } c __bpf_aligned; /* inline enum def with differing set of values */ }; struct core_reloc_primitives___diff_func_proto { - void (*f)(int); /* incompatible function prototype */ - void *d; - enum core_reloc_primitives_enum c; + void (*f)(int) __bpf_aligned; /* incompatible function prototype */ + void *d __bpf_aligned; + enum core_reloc_primitives_enum c __bpf_aligned; int b; char a; }; struct core_reloc_primitives___diff_ptr_type { - const char * const d; /* different pointee type + modifiers */ - char a; + const char * const d __bpf_aligned; /* different pointee type + modifiers */ + char a __bpf_aligned; int b; enum core_reloc_primitives_enum c; - int (*f)(const char *); + int (*f)(const char *) __bpf_aligned; }; struct core_reloc_primitives___err_non_enum { char a[1]; int b; int c; /* int instead of enum */ - void *d; - int (*f)(const char *); + void *d __bpf_aligned; + int (*f)(const char *) __bpf_aligned; }; struct core_reloc_primitives___err_non_int { char a[1]; - int *b; /* ptr instead of int */ - enum core_reloc_primitives_enum c; - void *d; - int (*f)(const char *); + int *b __bpf_aligned; /* ptr instead of int */ + enum core_reloc_primitives_enum c __bpf_aligned; + void *d __bpf_aligned; + int (*f)(const char *) __bpf_aligned; }; struct core_reloc_primitives___err_non_ptr { @@ -496,7 +501,7 @@ struct core_reloc_primitives___err_non_ptr { int b; enum core_reloc_primitives_enum c; int d; /* int instead of ptr */ - int (*f)(const char *); + int (*f)(const char *) __bpf_aligned; }; /* @@ -507,7 +512,7 @@ struct core_reloc_mods_output { }; typedef const int int_t; -typedef const char *char_ptr_t; +typedef const char *char_ptr_t __bpf_aligned; typedef const int arr_t[7]; struct core_reloc_mods_substruct { @@ -523,9 +528,9 @@ typedef struct { struct core_reloc_mods { int a; int_t b; - char *c; + char *c __bpf_aligned; char_ptr_t d; - int e[3]; + int e[3] __bpf_aligned; arr_t f; struct core_reloc_mods_substruct g; core_reloc_mods_substruct_t h; @@ -535,9 +540,9 @@ struct core_reloc_mods { struct core_reloc_mods___mod_swap { int b; int_t a; - char *d; + char *d __bpf_aligned; char_ptr_t c; - int f[3]; + int f[3] __bpf_aligned; arr_t e; struct { int y; @@ -555,7 +560,7 @@ typedef arr1_t arr2_t; typedef arr2_t arr3_t; typedef arr3_t arr4_t; -typedef const char * const volatile fancy_char_ptr_t; +typedef const char * const volatile fancy_char_ptr_t __bpf_aligned; typedef core_reloc_mods_substruct_t core_reloc_mods_substruct_tt; @@ -567,7 +572,7 @@ struct core_reloc_mods___typedefs { arr4_t e; fancy_char_ptr_t d; fancy_char_ptr_t c; - int3_t b; + int3_t b __bpf_aligned; int3_t a; }; @@ -739,19 +744,19 @@ struct core_reloc_bitfields___bit_sz_change { int8_t sb4: 1; /* 4 -> 1 */ int32_t sb20: 30; /* 20 -> 30 */ /* non-bitfields */ - uint16_t u32; /* 32 -> 16 */ - int64_t s32; /* 32 -> 64 */ + uint16_t u32; /* 32 -> 16 */ + int64_t s32 __bpf_aligned; /* 32 -> 64 */ }; /* turn bitfield into non-bitfield and vice versa */ struct core_reloc_bitfields___bitfield_vs_int { uint64_t ub1; /* 3 -> 64 non-bitfield */ uint8_t ub2; /* 20 -> 8 non-bitfield */ - int64_t ub7; /* 7 -> 64 non-bitfield signed */ - int64_t sb4; /* 4 -> 64 non-bitfield signed */ - uint64_t sb20; /* 20 -> 16 non-bitfield unsigned */ - int32_t u32: 20; /* 32 non-bitfield -> 20 bitfield */ - uint64_t s32: 60; /* 32 non-bitfield -> 60 bitfield */ + int64_t ub7 __bpf_aligned; /* 7 -> 64 non-bitfield signed */ + int64_t sb4 __bpf_aligned; /* 4 -> 64 non-bitfield signed */ + uint64_t sb20 __bpf_aligned; /* 20 -> 16 non-bitfield unsigned */ + int32_t u32: 20; /* 32 non-bitfield -> 20 bitfield */ + uint64_t s32: 60 __bpf_aligned; /* 32 non-bitfield -> 60 bitfield */ }; struct core_reloc_bitfields___just_big_enough { diff --git a/tools/testing/selftests/bpf/progs/test_tcpbpf_kern.c b/tools/testing/selftests/bpf/progs/test_tcpbpf_kern.c index 1f1966e86e9f..3e6912e4df3d 100644 --- a/tools/testing/selftests/bpf/progs/test_tcpbpf_kern.c +++ b/tools/testing/selftests/bpf/progs/test_tcpbpf_kern.c @@ -54,6 +54,7 @@ SEC("sockops") int bpf_testcb(struct bpf_sock_ops *skops) { char header[sizeof(struct ipv6hdr) + sizeof(struct tcphdr)]; + struct bpf_sock_ops *reuse = skops; struct tcphdr *thdr; int good_call_rv = 0; int bad_call_rv = 0; @@ -62,6 +63,46 @@ int bpf_testcb(struct bpf_sock_ops *skops) int v = 0; int op; + /* Test reading fields in bpf_sock_ops using single register */ + asm volatile ( + "%[reuse] = *(u32 *)(%[reuse] +96)" + : [reuse] "+r"(reuse) + :); + + asm volatile ( + "%[op] = *(u32 *)(%[skops] +96)" + : [op] "+r"(op) + : [skops] "r"(skops) + :); + + asm volatile ( + "r9 = %[skops];\n" + "r8 = *(u32 *)(r9 +164);\n" + "*(u32 *)(r9 +164) = r8;\n" + :: [skops] "r"(skops) + : "r9", "r8"); + + asm volatile ( + "r1 = %[skops];\n" + "r1 = *(u64 *)(r1 +184);\n" + "if r1 == 0 goto +1;\n" + "r1 = *(u32 *)(r1 +4);\n" + :: [skops] "r"(skops):"r1"); + + asm volatile ( + "r9 = %[skops];\n" + "r9 = *(u64 *)(r9 +184);\n" + "if r9 == 0 goto +1;\n" + "r9 = *(u32 *)(r9 +4);\n" + :: [skops] "r"(skops):"r9"); + + asm volatile ( + "r1 = %[skops];\n" + "r2 = *(u64 *)(r1 +184);\n" + "if r2 == 0 goto +1;\n" + "r2 = *(u32 *)(r2 +4);\n" + :: [skops] "r"(skops):"r1", "r2"); + op = (int) skops->op; update_event_map(op); diff --git a/tools/testing/selftests/bpf/progs/test_varlen.c b/tools/testing/selftests/bpf/progs/test_varlen.c index cd4b72c55dfe..913acdffd90f 100644 --- a/tools/testing/selftests/bpf/progs/test_varlen.c +++ b/tools/testing/selftests/bpf/progs/test_varlen.c @@ -15,9 +15,9 @@ int test_pid = 0; bool capture = false; /* .bss */ -long payload1_len1 = 0; -long payload1_len2 = 0; -long total1 = 0; +__u64 payload1_len1 = 0; +__u64 payload1_len2 = 0; +__u64 total1 = 0; char payload1[MAX_LEN + MAX_LEN] = {}; /* .data */ diff --git a/tools/testing/selftests/bpf/test_btf.c b/tools/testing/selftests/bpf/test_btf.c index 305fae8f80a9..c75fc6447186 100644 --- a/tools/testing/selftests/bpf/test_btf.c +++ b/tools/testing/selftests/bpf/test_btf.c @@ -3883,7 +3883,7 @@ static int test_big_btf_info(unsigned int test_num) info_garbage.garbage = 0; err = bpf_obj_get_info_by_fd(btf_fd, info, &info_len); if (CHECK(err || info_len != sizeof(*info), - "err:%d errno:%d info_len:%u sizeof(*info):%lu", + "err:%d errno:%d info_len:%u sizeof(*info):%zu", err, errno, info_len, sizeof(*info))) { err = -1; goto done; @@ -4094,7 +4094,7 @@ static int do_test_get_info(unsigned int test_num) if (CHECK(err || !info.id || info_len != sizeof(info) || info.btf_size != raw_btf_size || (ret = memcmp(raw_btf, user_btf, expected_nbytes)), - "err:%d errno:%d info.id:%u info_len:%u sizeof(info):%lu raw_btf_size:%u info.btf_size:%u expected_nbytes:%u memcmp:%d", + "err:%d errno:%d info.id:%u info_len:%u sizeof(info):%zu raw_btf_size:%u info.btf_size:%u expected_nbytes:%u memcmp:%d", err, errno, info.id, info_len, sizeof(info), raw_btf_size, info.btf_size, expected_nbytes, ret)) { err = -1; @@ -4730,7 +4730,7 @@ ssize_t get_pprint_expected_line(enum pprint_mapv_kind_t mapv_kind, nexpected_line = snprintf(expected_line, line_size, "%s%u: {%u,0,%d,0x%x,0x%x,0x%x," - "{%lu|[%u,%u,%u,%u,%u,%u,%u,%u]},%s," + "{%llu|[%u,%u,%u,%u,%u,%u,%u,%u]},%s," "%u,0x%x,[[%d,%d],[%d,%d]]}\n", percpu_map ? "\tcpu" : "", percpu_map ? cpu : next_key, @@ -4738,7 +4738,7 @@ ssize_t get_pprint_expected_line(enum pprint_mapv_kind_t mapv_kind, v->unused_bits2a, v->bits28, v->unused_bits2b, - v->ui64, + (__u64)v->ui64, v->ui8a[0], v->ui8a[1], v->ui8a[2], v->ui8a[3], v->ui8a[4], v->ui8a[5], diff --git a/tools/testing/selftests/bpf/test_maps.c b/tools/testing/selftests/bpf/test_maps.c index 754cf611723e..0d92ebcb335d 100644 --- a/tools/testing/selftests/bpf/test_maps.c +++ b/tools/testing/selftests/bpf/test_maps.c @@ -1274,6 +1274,8 @@ static void __run_parallel(unsigned int tasks, pid_t pid[tasks]; int i; + fflush(stdout); + for (i = 0; i < tasks; i++) { pid[i] = fork(); if (pid[i] == 0) { diff --git a/tools/testing/selftests/bpf/test_progs.c b/tools/testing/selftests/bpf/test_progs.c index b1e4dadacd9b..22943b58d752 100644 --- a/tools/testing/selftests/bpf/test_progs.c +++ b/tools/testing/selftests/bpf/test_progs.c @@ -618,7 +618,9 @@ int cd_flavor_subdir(const char *exec_name) if (!flavor) return 0; flavor++; - fprintf(stdout, "Switching to flavor '%s' subdirectory...\n", flavor); + if (env.verbosity > VERBOSE_NONE) + fprintf(stdout, "Switching to flavor '%s' subdirectory...\n", flavor); + return chdir(flavor); } diff --git a/tools/testing/selftests/bpf/test_progs.h b/tools/testing/selftests/bpf/test_progs.h index 6e09bf738473..dbb820dde138 100644 --- a/tools/testing/selftests/bpf/test_progs.h +++ b/tools/testing/selftests/bpf/test_progs.h @@ -135,6 +135,11 @@ static inline __u64 ptr_to_u64(const void *ptr) return (__u64) (unsigned long) ptr; } +static inline void *u64_to_ptr(__u64 ptr) +{ + return (void *) (unsigned long) ptr; +} + int bpf_find_map(const char *test, struct bpf_object *obj, const char *name); int compare_map_keys(int map1_fd, int map2_fd); int compare_stack_ips(int smap_fd, int amap_fd, int stack_trace_len); diff --git a/tools/testing/selftests/kvm/.gitignore b/tools/testing/selftests/kvm/.gitignore index 452787152748..307ceaadbbb9 100644 --- a/tools/testing/selftests/kvm/.gitignore +++ b/tools/testing/selftests/kvm/.gitignore @@ -11,6 +11,7 @@ /x86_64/set_sregs_test /x86_64/smm_test /x86_64/state_test +/x86_64/user_msr_test /x86_64/vmx_preemption_timer_test /x86_64/svm_vmcall_test /x86_64/sync_regs_test diff --git a/tools/testing/selftests/kvm/Makefile b/tools/testing/selftests/kvm/Makefile index 4a166588d99f..7ebe71fbca53 100644 --- a/tools/testing/selftests/kvm/Makefile +++ b/tools/testing/selftests/kvm/Makefile @@ -55,6 +55,8 @@ TEST_GEN_PROGS_x86_64 += x86_64/vmx_set_nested_state_test TEST_GEN_PROGS_x86_64 += x86_64/vmx_tsc_adjust_test TEST_GEN_PROGS_x86_64 += x86_64/xss_msr_test TEST_GEN_PROGS_x86_64 += x86_64/debug_regs +TEST_GEN_PROGS_x86_64 += x86_64/tsc_msrs_test +TEST_GEN_PROGS_x86_64 += x86_64/user_msr_test TEST_GEN_PROGS_x86_64 += clear_dirty_log_test TEST_GEN_PROGS_x86_64 += demand_paging_test TEST_GEN_PROGS_x86_64 += dirty_log_test diff --git a/tools/testing/selftests/kvm/include/x86_64/vmx.h b/tools/testing/selftests/kvm/include/x86_64/vmx.h index 16fa21ebb99c..54d624dd6c10 100644 --- a/tools/testing/selftests/kvm/include/x86_64/vmx.h +++ b/tools/testing/selftests/kvm/include/x86_64/vmx.h @@ -48,7 +48,7 @@ #define SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES 0x00000001 #define SECONDARY_EXEC_ENABLE_EPT 0x00000002 #define SECONDARY_EXEC_DESC 0x00000004 -#define SECONDARY_EXEC_RDTSCP 0x00000008 +#define SECONDARY_EXEC_ENABLE_RDTSCP 0x00000008 #define SECONDARY_EXEC_VIRTUALIZE_X2APIC_MODE 0x00000010 #define SECONDARY_EXEC_ENABLE_VPID 0x00000020 #define SECONDARY_EXEC_WBINVD_EXITING 0x00000040 diff --git a/tools/testing/selftests/kvm/x86_64/tsc_msrs_test.c b/tools/testing/selftests/kvm/x86_64/tsc_msrs_test.c new file mode 100644 index 000000000000..f8e761149daa --- /dev/null +++ b/tools/testing/selftests/kvm/x86_64/tsc_msrs_test.c @@ -0,0 +1,168 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Tests for MSR_IA32_TSC and MSR_IA32_TSC_ADJUST. + * + * Copyright (C) 2020, Red Hat, Inc. + */ +#include <stdio.h> +#include <string.h> +#include "kvm_util.h" +#include "processor.h" + +#define VCPU_ID 0 + +#define UNITY (1ull << 30) +#define HOST_ADJUST (UNITY * 64) +#define GUEST_STEP (UNITY * 4) +#define ROUND(x) ((x + UNITY / 2) & -UNITY) +#define rounded_rdmsr(x) ROUND(rdmsr(x)) +#define rounded_host_rdmsr(x) ROUND(vcpu_get_msr(vm, 0, x)) + +#define GUEST_ASSERT_EQ(a, b) do { \ + __typeof(a) _a = (a); \ + __typeof(b) _b = (b); \ + if (_a != _b) \ + ucall(UCALL_ABORT, 4, \ + "Failed guest assert: " \ + #a " == " #b, __LINE__, _a, _b); \ + } while(0) + +static void guest_code(void) +{ + u64 val = 0; + + GUEST_ASSERT_EQ(rounded_rdmsr(MSR_IA32_TSC), val); + GUEST_ASSERT_EQ(rounded_rdmsr(MSR_IA32_TSC_ADJUST), val); + + /* Guest: writes to MSR_IA32_TSC affect both MSRs. */ + val = 1ull * GUEST_STEP; + wrmsr(MSR_IA32_TSC, val); + GUEST_ASSERT_EQ(rounded_rdmsr(MSR_IA32_TSC), val); + GUEST_ASSERT_EQ(rounded_rdmsr(MSR_IA32_TSC_ADJUST), val); + + /* Guest: writes to MSR_IA32_TSC_ADJUST affect both MSRs. */ + GUEST_SYNC(2); + val = 2ull * GUEST_STEP; + wrmsr(MSR_IA32_TSC_ADJUST, val); + GUEST_ASSERT_EQ(rounded_rdmsr(MSR_IA32_TSC), val); + GUEST_ASSERT_EQ(rounded_rdmsr(MSR_IA32_TSC_ADJUST), val); + + /* Host: setting the TSC offset. */ + GUEST_SYNC(3); + GUEST_ASSERT_EQ(rounded_rdmsr(MSR_IA32_TSC), HOST_ADJUST + val); + GUEST_ASSERT_EQ(rounded_rdmsr(MSR_IA32_TSC_ADJUST), val); + + /* + * Guest: writes to MSR_IA32_TSC_ADJUST do not destroy the + * host-side offset and affect both MSRs. + */ + GUEST_SYNC(4); + val = 3ull * GUEST_STEP; + wrmsr(MSR_IA32_TSC_ADJUST, val); + GUEST_ASSERT_EQ(rounded_rdmsr(MSR_IA32_TSC), HOST_ADJUST + val); + GUEST_ASSERT_EQ(rounded_rdmsr(MSR_IA32_TSC_ADJUST), val); + + /* + * Guest: writes to MSR_IA32_TSC affect both MSRs, so the host-side + * offset is now visible in MSR_IA32_TSC_ADJUST. + */ + GUEST_SYNC(5); + val = 4ull * GUEST_STEP; + wrmsr(MSR_IA32_TSC, val); + GUEST_ASSERT_EQ(rounded_rdmsr(MSR_IA32_TSC), val); + GUEST_ASSERT_EQ(rounded_rdmsr(MSR_IA32_TSC_ADJUST), val - HOST_ADJUST); + + GUEST_DONE(); +} + +static void run_vcpu(struct kvm_vm *vm, uint32_t vcpuid, int stage) +{ + struct ucall uc; + + vcpu_args_set(vm, vcpuid, 1, vcpuid); + + vcpu_ioctl(vm, vcpuid, KVM_RUN, NULL); + + switch (get_ucall(vm, vcpuid, &uc)) { + case UCALL_SYNC: + TEST_ASSERT(!strcmp((const char *)uc.args[0], "hello") && + uc.args[1] == stage + 1, "Stage %d: Unexpected register values vmexit, got %lx", + stage + 1, (ulong)uc.args[1]); + return; + case UCALL_DONE: + return; + case UCALL_ABORT: + TEST_ASSERT(false, "%s at %s:%ld\n" \ + "\tvalues: %#lx, %#lx", (const char *)uc.args[0], + __FILE__, uc.args[1], uc.args[2], uc.args[3]); + default: + TEST_ASSERT(false, "Unexpected exit: %s", + exit_reason_str(vcpu_state(vm, vcpuid)->exit_reason)); + } +} + +int main(void) +{ + struct kvm_vm *vm; + uint64_t val; + + vm = vm_create_default(VCPU_ID, 0, guest_code); + vcpu_set_cpuid(vm, VCPU_ID, kvm_get_supported_cpuid()); + + val = 0; + ASSERT_EQ(rounded_host_rdmsr(MSR_IA32_TSC), val); + ASSERT_EQ(rounded_host_rdmsr(MSR_IA32_TSC_ADJUST), val); + + /* Guest: writes to MSR_IA32_TSC affect both MSRs. */ + run_vcpu(vm, VCPU_ID, 1); + val = 1ull * GUEST_STEP; + ASSERT_EQ(rounded_host_rdmsr(MSR_IA32_TSC), val); + ASSERT_EQ(rounded_host_rdmsr(MSR_IA32_TSC_ADJUST), val); + + /* Guest: writes to MSR_IA32_TSC_ADJUST affect both MSRs. */ + run_vcpu(vm, VCPU_ID, 2); + val = 2ull * GUEST_STEP; + ASSERT_EQ(rounded_host_rdmsr(MSR_IA32_TSC), val); + ASSERT_EQ(rounded_host_rdmsr(MSR_IA32_TSC_ADJUST), val); + + /* + * Host: writes to MSR_IA32_TSC set the host-side offset + * and therefore do not change MSR_IA32_TSC_ADJUST. + */ + vcpu_set_msr(vm, 0, MSR_IA32_TSC, HOST_ADJUST + val); + ASSERT_EQ(rounded_host_rdmsr(MSR_IA32_TSC), HOST_ADJUST + val); + ASSERT_EQ(rounded_host_rdmsr(MSR_IA32_TSC_ADJUST), val); + run_vcpu(vm, VCPU_ID, 3); + + /* Host: writes to MSR_IA32_TSC_ADJUST do not modify the TSC. */ + vcpu_set_msr(vm, 0, MSR_IA32_TSC_ADJUST, UNITY * 123456); + ASSERT_EQ(rounded_host_rdmsr(MSR_IA32_TSC), HOST_ADJUST + val); + ASSERT_EQ(vcpu_get_msr(vm, 0, MSR_IA32_TSC_ADJUST), UNITY * 123456); + + /* Restore previous value. */ + vcpu_set_msr(vm, 0, MSR_IA32_TSC_ADJUST, val); + ASSERT_EQ(rounded_host_rdmsr(MSR_IA32_TSC), HOST_ADJUST + val); + ASSERT_EQ(rounded_host_rdmsr(MSR_IA32_TSC_ADJUST), val); + + /* + * Guest: writes to MSR_IA32_TSC_ADJUST do not destroy the + * host-side offset and affect both MSRs. + */ + run_vcpu(vm, VCPU_ID, 4); + val = 3ull * GUEST_STEP; + ASSERT_EQ(rounded_host_rdmsr(MSR_IA32_TSC), HOST_ADJUST + val); + ASSERT_EQ(rounded_host_rdmsr(MSR_IA32_TSC_ADJUST), val); + + /* + * Guest: writes to MSR_IA32_TSC affect both MSRs, so the host-side + * offset is now visible in MSR_IA32_TSC_ADJUST. + */ + run_vcpu(vm, VCPU_ID, 5); + val = 4ull * GUEST_STEP; + ASSERT_EQ(rounded_host_rdmsr(MSR_IA32_TSC), val); + ASSERT_EQ(rounded_host_rdmsr(MSR_IA32_TSC_ADJUST), val - HOST_ADJUST); + + kvm_vm_free(vm); + + return 0; +} diff --git a/tools/testing/selftests/kvm/x86_64/user_msr_test.c b/tools/testing/selftests/kvm/x86_64/user_msr_test.c new file mode 100644 index 000000000000..cbe1b08890ff --- /dev/null +++ b/tools/testing/selftests/kvm/x86_64/user_msr_test.c @@ -0,0 +1,248 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * tests for KVM_CAP_X86_USER_SPACE_MSR and KVM_X86_SET_MSR_FILTER + * + * Copyright (C) 2020, Amazon Inc. + * + * This is a functional test to verify that we can deflect MSR events + * into user space. + */ +#define _GNU_SOURCE /* for program_invocation_short_name */ +#include <fcntl.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <sys/ioctl.h> + +#include "test_util.h" + +#include "kvm_util.h" +#include "processor.h" + +#define VCPU_ID 5 + +static u32 msr_reads, msr_writes; + +static u8 bitmap_00000000[KVM_MSR_FILTER_MAX_BITMAP_SIZE]; +static u8 bitmap_00000000_write[KVM_MSR_FILTER_MAX_BITMAP_SIZE]; +static u8 bitmap_40000000[KVM_MSR_FILTER_MAX_BITMAP_SIZE]; +static u8 bitmap_c0000000[KVM_MSR_FILTER_MAX_BITMAP_SIZE]; +static u8 bitmap_c0000000_read[KVM_MSR_FILTER_MAX_BITMAP_SIZE]; +static u8 bitmap_deadbeef[1] = { 0x1 }; + +static void deny_msr(uint8_t *bitmap, u32 msr) +{ + u32 idx = msr & (KVM_MSR_FILTER_MAX_BITMAP_SIZE - 1); + + bitmap[idx / 8] &= ~(1 << (idx % 8)); +} + +static void prepare_bitmaps(void) +{ + memset(bitmap_00000000, 0xff, sizeof(bitmap_00000000)); + memset(bitmap_00000000_write, 0xff, sizeof(bitmap_00000000_write)); + memset(bitmap_40000000, 0xff, sizeof(bitmap_40000000)); + memset(bitmap_c0000000, 0xff, sizeof(bitmap_c0000000)); + memset(bitmap_c0000000_read, 0xff, sizeof(bitmap_c0000000_read)); + + deny_msr(bitmap_00000000_write, MSR_IA32_POWER_CTL); + deny_msr(bitmap_c0000000_read, MSR_SYSCALL_MASK); + deny_msr(bitmap_c0000000_read, MSR_GS_BASE); +} + +struct kvm_msr_filter filter = { + .flags = KVM_MSR_FILTER_DEFAULT_DENY, + .ranges = { + { + .flags = KVM_MSR_FILTER_READ, + .base = 0x00000000, + .nmsrs = KVM_MSR_FILTER_MAX_BITMAP_SIZE * BITS_PER_BYTE, + .bitmap = bitmap_00000000, + }, { + .flags = KVM_MSR_FILTER_WRITE, + .base = 0x00000000, + .nmsrs = KVM_MSR_FILTER_MAX_BITMAP_SIZE * BITS_PER_BYTE, + .bitmap = bitmap_00000000_write, + }, { + .flags = KVM_MSR_FILTER_READ | KVM_MSR_FILTER_WRITE, + .base = 0x40000000, + .nmsrs = KVM_MSR_FILTER_MAX_BITMAP_SIZE * BITS_PER_BYTE, + .bitmap = bitmap_40000000, + }, { + .flags = KVM_MSR_FILTER_READ, + .base = 0xc0000000, + .nmsrs = KVM_MSR_FILTER_MAX_BITMAP_SIZE * BITS_PER_BYTE, + .bitmap = bitmap_c0000000_read, + }, { + .flags = KVM_MSR_FILTER_WRITE, + .base = 0xc0000000, + .nmsrs = KVM_MSR_FILTER_MAX_BITMAP_SIZE * BITS_PER_BYTE, + .bitmap = bitmap_c0000000, + }, { + .flags = KVM_MSR_FILTER_WRITE | KVM_MSR_FILTER_READ, + .base = 0xdeadbeef, + .nmsrs = 1, + .bitmap = bitmap_deadbeef, + }, + }, +}; + +struct kvm_msr_filter no_filter = { + .flags = KVM_MSR_FILTER_DEFAULT_ALLOW, +}; + +static void guest_msr_calls(bool trapped) +{ + /* This goes into the in-kernel emulation */ + wrmsr(MSR_SYSCALL_MASK, 0); + + if (trapped) { + /* This goes into user space emulation */ + GUEST_ASSERT(rdmsr(MSR_SYSCALL_MASK) == MSR_SYSCALL_MASK); + GUEST_ASSERT(rdmsr(MSR_GS_BASE) == MSR_GS_BASE); + } else { + GUEST_ASSERT(rdmsr(MSR_SYSCALL_MASK) != MSR_SYSCALL_MASK); + GUEST_ASSERT(rdmsr(MSR_GS_BASE) != MSR_GS_BASE); + } + + /* If trapped == true, this goes into user space emulation */ + wrmsr(MSR_IA32_POWER_CTL, 0x1234); + + /* This goes into the in-kernel emulation */ + rdmsr(MSR_IA32_POWER_CTL); + + /* Invalid MSR, should always be handled by user space exit */ + GUEST_ASSERT(rdmsr(0xdeadbeef) == 0xdeadbeef); + wrmsr(0xdeadbeef, 0x1234); +} + +static void guest_code(void) +{ + guest_msr_calls(true); + + /* + * Disable msr filtering, so that the kernel + * handles everything in the next round + */ + GUEST_SYNC(0); + + guest_msr_calls(false); + + GUEST_DONE(); +} + +static int handle_ucall(struct kvm_vm *vm) +{ + struct ucall uc; + + switch (get_ucall(vm, VCPU_ID, &uc)) { + case UCALL_ABORT: + TEST_FAIL("Guest assertion not met"); + break; + case UCALL_SYNC: + vm_ioctl(vm, KVM_X86_SET_MSR_FILTER, &no_filter); + break; + case UCALL_DONE: + return 1; + default: + TEST_FAIL("Unknown ucall %lu", uc.cmd); + } + + return 0; +} + +static void handle_rdmsr(struct kvm_run *run) +{ + run->msr.data = run->msr.index; + msr_reads++; + + if (run->msr.index == MSR_SYSCALL_MASK || + run->msr.index == MSR_GS_BASE) { + TEST_ASSERT(run->msr.reason == KVM_MSR_EXIT_REASON_FILTER, + "MSR read trap w/o access fault"); + } + + if (run->msr.index == 0xdeadbeef) { + TEST_ASSERT(run->msr.reason == KVM_MSR_EXIT_REASON_UNKNOWN, + "MSR deadbeef read trap w/o inval fault"); + } +} + +static void handle_wrmsr(struct kvm_run *run) +{ + /* ignore */ + msr_writes++; + + if (run->msr.index == MSR_IA32_POWER_CTL) { + TEST_ASSERT(run->msr.data == 0x1234, + "MSR data for MSR_IA32_POWER_CTL incorrect"); + TEST_ASSERT(run->msr.reason == KVM_MSR_EXIT_REASON_FILTER, + "MSR_IA32_POWER_CTL trap w/o access fault"); + } + + if (run->msr.index == 0xdeadbeef) { + TEST_ASSERT(run->msr.data == 0x1234, + "MSR data for deadbeef incorrect"); + TEST_ASSERT(run->msr.reason == KVM_MSR_EXIT_REASON_UNKNOWN, + "deadbeef trap w/o inval fault"); + } +} + +int main(int argc, char *argv[]) +{ + struct kvm_enable_cap cap = { + .cap = KVM_CAP_X86_USER_SPACE_MSR, + .args[0] = KVM_MSR_EXIT_REASON_INVAL | + KVM_MSR_EXIT_REASON_UNKNOWN | + KVM_MSR_EXIT_REASON_FILTER, + }; + struct kvm_vm *vm; + struct kvm_run *run; + int rc; + + /* Tell stdout not to buffer its content */ + setbuf(stdout, NULL); + + /* Create VM */ + vm = vm_create_default(VCPU_ID, 0, guest_code); + vcpu_set_cpuid(vm, VCPU_ID, kvm_get_supported_cpuid()); + run = vcpu_state(vm, VCPU_ID); + + rc = kvm_check_cap(KVM_CAP_X86_USER_SPACE_MSR); + TEST_ASSERT(rc, "KVM_CAP_X86_USER_SPACE_MSR is available"); + vm_enable_cap(vm, &cap); + + rc = kvm_check_cap(KVM_CAP_X86_MSR_FILTER); + TEST_ASSERT(rc, "KVM_CAP_X86_MSR_FILTER is available"); + + prepare_bitmaps(); + vm_ioctl(vm, KVM_X86_SET_MSR_FILTER, &filter); + + while (1) { + rc = _vcpu_run(vm, VCPU_ID); + + TEST_ASSERT(rc == 0, "vcpu_run failed: %d\n", rc); + + switch (run->exit_reason) { + case KVM_EXIT_X86_RDMSR: + handle_rdmsr(run); + break; + case KVM_EXIT_X86_WRMSR: + handle_wrmsr(run); + break; + case KVM_EXIT_IO: + if (handle_ucall(vm)) + goto done; + break; + } + + } + +done: + TEST_ASSERT(msr_reads == 4, "Handled 4 rdmsr in user space"); + TEST_ASSERT(msr_writes == 3, "Handled 3 wrmsr in user space"); + + kvm_vm_free(vm); + + return 0; +} diff --git a/tools/testing/selftests/net/icmp_redirect.sh b/tools/testing/selftests/net/icmp_redirect.sh index 18c5de53558a..bf361f30d6ef 100755 --- a/tools/testing/selftests/net/icmp_redirect.sh +++ b/tools/testing/selftests/net/icmp_redirect.sh @@ -180,6 +180,8 @@ setup() ;; r[12]) ip netns exec $ns sysctl -q -w net.ipv4.ip_forward=1 ip netns exec $ns sysctl -q -w net.ipv4.conf.all.send_redirects=1 + ip netns exec $ns sysctl -q -w net.ipv4.conf.default.rp_filter=0 + ip netns exec $ns sysctl -q -w net.ipv4.conf.all.rp_filter=0 ip netns exec $ns sysctl -q -w net.ipv6.conf.all.forwarding=1 ip netns exec $ns sysctl -q -w net.ipv6.route.mtu_expires=10 diff --git a/tools/testing/selftests/netfilter/nft_flowtable.sh b/tools/testing/selftests/netfilter/nft_flowtable.sh index d3e0809ab368..431296c0f91c 100755 --- a/tools/testing/selftests/netfilter/nft_flowtable.sh +++ b/tools/testing/selftests/netfilter/nft_flowtable.sh @@ -2,13 +2,18 @@ # SPDX-License-Identifier: GPL-2.0 # # This tests basic flowtable functionality. -# Creates following topology: +# Creates following default topology: # # Originator (MTU 9000) <-Router1-> MTU 1500 <-Router2-> Responder (MTU 2000) # Router1 is the one doing flow offloading, Router2 has no special # purpose other than having a link that is smaller than either Originator # and responder, i.e. TCPMSS announced values are too large and will still # result in fragmentation and/or PMTU discovery. +# +# You can check with different Orgininator/Link/Responder MTU eg: +# nft_flowtable.sh -o8000 -l1500 -r2000 +# + # Kselftest framework requirement - SKIP code is 4. ksft_skip=4 @@ -21,29 +26,17 @@ ns2out="" log_netns=$(sysctl -n net.netfilter.nf_log_all_netns) -nft --version > /dev/null 2>&1 -if [ $? -ne 0 ];then - echo "SKIP: Could not run test without nft tool" - exit $ksft_skip -fi - -ip -Version > /dev/null 2>&1 -if [ $? -ne 0 ];then - echo "SKIP: Could not run test without ip tool" - exit $ksft_skip -fi - -which nc > /dev/null 2>&1 -if [ $? -ne 0 ];then - echo "SKIP: Could not run test without nc (netcat)" - exit $ksft_skip -fi +checktool (){ + if ! $1 > /dev/null 2>&1; then + echo "SKIP: Could not $2" + exit $ksft_skip + fi +} -ip netns add nsr1 -if [ $? -ne 0 ];then - echo "SKIP: Could not create net namespace" - exit $ksft_skip -fi +checktool "nft --version" "run test without nft tool" +checktool "ip -Version" "run test without ip tool" +checktool "which nc" "run test without nc (netcat)" +checktool "ip netns add nsr1" "create net namespace" ip netns add ns1 ip netns add ns2 @@ -89,11 +82,41 @@ ip -net nsr2 addr add dead:2::1/64 dev veth1 # ns2 is going via nsr2 with a smaller mtu, so that TCPMSS announced by both peers # is NOT the lowest link mtu. -ip -net nsr1 link set veth0 mtu 9000 -ip -net ns1 link set eth0 mtu 9000 +omtu=9000 +lmtu=1500 +rmtu=2000 + +usage(){ + echo "nft_flowtable.sh [OPTIONS]" + echo + echo "MTU options" + echo " -o originator" + echo " -l link" + echo " -r responder" + exit 1 +} + +while getopts "o:l:r:" o +do + case $o in + o) omtu=$OPTARG;; + l) lmtu=$OPTARG;; + r) rmtu=$OPTARG;; + *) usage;; + esac +done + +if ! ip -net nsr1 link set veth0 mtu $omtu; then + exit 1 +fi + +ip -net ns1 link set eth0 mtu $omtu + +if ! ip -net nsr2 link set veth1 mtu $rmtu; then + exit 1 +fi -ip -net nsr2 link set veth1 mtu 2000 -ip -net ns2 link set eth0 mtu 2000 +ip -net ns2 link set eth0 mtu $rmtu # transfer-net between nsr1 and nsr2. # these addresses are not used for connections. @@ -113,7 +136,10 @@ for i in 1 2; do ip -net ns$i route add default via 10.0.$i.1 ip -net ns$i addr add dead:$i::99/64 dev eth0 ip -net ns$i route add default via dead:$i::1 - ip netns exec ns$i sysctl net.ipv4.tcp_no_metrics_save=1 > /dev/null + if ! ip netns exec ns$i sysctl net.ipv4.tcp_no_metrics_save=1 > /dev/null; then + echo "ERROR: Check Originator/Responder values (problem during address addition)" + exit 1 + fi # don't set ip DF bit for first two tests ip netns exec ns$i sysctl net.ipv4.ip_no_pmtu_disc=1 > /dev/null @@ -147,7 +173,7 @@ table inet filter { # as PMTUd is off. # This rule is deleted for the last test, when we expect PMTUd # to kick in and ensure all packets meet mtu requirements. - meta length gt 1500 accept comment something-to-grep-for + meta length gt $lmtu accept comment something-to-grep-for # next line blocks connection w.o. working offload. # we only do this for reverse dir, because we expect packets to @@ -171,15 +197,13 @@ if [ $? -ne 0 ]; then fi # test basic connectivity -ip netns exec ns1 ping -c 1 -q 10.0.2.99 > /dev/null -if [ $? -ne 0 ];then +if ! ip netns exec ns1 ping -c 1 -q 10.0.2.99 > /dev/null; then echo "ERROR: ns1 cannot reach ns2" 1>&2 bash exit 1 fi -ip netns exec ns2 ping -c 1 -q 10.0.1.99 > /dev/null -if [ $? -ne 0 ];then +if ! ip netns exec ns2 ping -c 1 -q 10.0.1.99 > /dev/null; then echo "ERROR: ns2 cannot reach ns1" 1>&2 exit 1 fi @@ -196,7 +220,6 @@ ns2out=$(mktemp) make_file() { name=$1 - who=$2 SIZE=$((RANDOM % (1024 * 8))) TSIZE=$((SIZE * 1024)) @@ -215,8 +238,7 @@ check_transfer() out=$2 what=$3 - cmp "$in" "$out" > /dev/null 2>&1 - if [ $? -ne 0 ] ;then + if ! cmp "$in" "$out" > /dev/null 2>&1; then echo "FAIL: file mismatch for $what" 1>&2 ls -l "$in" ls -l "$out" @@ -243,17 +265,21 @@ test_tcp_forwarding_ip() sleep 3 - kill $lpid - kill $cpid + if ps -p $lpid > /dev/null;then + kill $lpid + fi + + if ps -p $cpid > /dev/null;then + kill $cpid + fi + wait - check_transfer "$ns1in" "$ns2out" "ns1 -> ns2" - if [ $? -ne 0 ];then + if ! check_transfer "$ns1in" "$ns2out" "ns1 -> ns2"; then lret=1 fi - check_transfer "$ns2in" "$ns1out" "ns1 <- ns2" - if [ $? -ne 0 ];then + if ! check_transfer "$ns2in" "$ns1out" "ns1 <- ns2"; then lret=1 fi @@ -282,13 +308,12 @@ test_tcp_forwarding_nat() return $lret } -make_file "$ns1in" "ns1" -make_file "$ns2in" "ns2" +make_file "$ns1in" +make_file "$ns2in" # First test: # No PMTU discovery, nsr1 is expected to fragment packets from ns1 to ns2 as needed. -test_tcp_forwarding ns1 ns2 -if [ $? -eq 0 ] ;then +if test_tcp_forwarding ns1 ns2; then echo "PASS: flow offloaded for ns1/ns2" else echo "FAIL: flow offload for ns1/ns2:" 1>&2 @@ -319,9 +344,7 @@ table ip nat { } EOF -test_tcp_forwarding_nat ns1 ns2 - -if [ $? -eq 0 ] ;then +if test_tcp_forwarding_nat ns1 ns2; then echo "PASS: flow offloaded for ns1/ns2 with NAT" else echo "FAIL: flow offload for ns1/ns2 with NAT" 1>&2 @@ -333,8 +356,7 @@ fi # Same as second test, but with PMTU discovery enabled. handle=$(ip netns exec nsr1 nft -a list table inet filter | grep something-to-grep-for | cut -d \# -f 2) -ip netns exec nsr1 nft delete rule inet filter forward $handle -if [ $? -ne 0 ] ;then +if ! ip netns exec nsr1 nft delete rule inet filter forward $handle; then echo "FAIL: Could not delete large-packet accept rule" exit 1 fi @@ -342,8 +364,7 @@ fi ip netns exec ns1 sysctl net.ipv4.ip_no_pmtu_disc=0 > /dev/null ip netns exec ns2 sysctl net.ipv4.ip_no_pmtu_disc=0 > /dev/null -test_tcp_forwarding_nat ns1 ns2 -if [ $? -eq 0 ] ;then +if test_tcp_forwarding_nat ns1 ns2; then echo "PASS: flow offloaded for ns1/ns2 with NAT and pmtu discovery" else echo "FAIL: flow offload for ns1/ns2 with NAT and pmtu discovery" 1>&2 @@ -389,8 +410,7 @@ ip -net ns2 route del 192.168.10.1 via 10.0.2.1 ip -net ns2 route add default via 10.0.2.1 ip -net ns2 route add default via dead:2::1 -test_tcp_forwarding ns1 ns2 -if [ $? -eq 0 ] ;then +if test_tcp_forwarding ns1 ns2; then echo "PASS: ipsec tunnel mode for ns1/ns2" else echo "FAIL: ipsec tunnel mode for ns1/ns2" diff --git a/tools/testing/selftests/powerpc/mm/.gitignore b/tools/testing/selftests/powerpc/mm/.gitignore index 91c775c23c66..aac4a59f9e28 100644 --- a/tools/testing/selftests/powerpc/mm/.gitignore +++ b/tools/testing/selftests/powerpc/mm/.gitignore @@ -2,6 +2,7 @@ hugetlb_vs_thp_test subpage_prot tempfile +prot_sao segv_errors wild_bctr large_vm_fork_separation diff --git a/tools/testing/selftests/powerpc/mm/Makefile b/tools/testing/selftests/powerpc/mm/Makefile index 250ce172e0da..defe488d6bf1 100644 --- a/tools/testing/selftests/powerpc/mm/Makefile +++ b/tools/testing/selftests/powerpc/mm/Makefile @@ -2,7 +2,7 @@ noarg: $(MAKE) -C ../ -TEST_GEN_PROGS := hugetlb_vs_thp_test subpage_prot segv_errors wild_bctr \ +TEST_GEN_PROGS := hugetlb_vs_thp_test subpage_prot prot_sao segv_errors wild_bctr \ large_vm_fork_separation bad_accesses pkey_exec_prot \ pkey_siginfo stack_expansion_signal stack_expansion_ldst @@ -14,6 +14,8 @@ include ../../lib.mk $(TEST_GEN_PROGS): ../harness.c ../utils.c +$(OUTPUT)/prot_sao: ../utils.c + $(OUTPUT)/wild_bctr: CFLAGS += -m64 $(OUTPUT)/large_vm_fork_separation: CFLAGS += -m64 $(OUTPUT)/bad_accesses: CFLAGS += -m64 diff --git a/tools/testing/selftests/powerpc/mm/prot_sao.c b/tools/testing/selftests/powerpc/mm/prot_sao.c new file mode 100644 index 000000000000..e0cf8ebbf8cd --- /dev/null +++ b/tools/testing/selftests/powerpc/mm/prot_sao.c @@ -0,0 +1,43 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright 2016, Michael Ellerman, IBM Corp. + */ + +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <sys/mman.h> + +#include <asm/cputable.h> + +#include "utils.h" + +#define SIZE (64 * 1024) + +int test_prot_sao(void) +{ + char *p; + + /* SAO was introduced in 2.06 and removed in 3.1 */ + SKIP_IF(!have_hwcap(PPC_FEATURE_ARCH_2_06) || + have_hwcap2(PPC_FEATURE2_ARCH_3_1)); + + /* + * Ensure we can ask for PROT_SAO. + * We can't really verify that it does the right thing, but at least we + * confirm the kernel will accept it. + */ + p = mmap(NULL, SIZE, PROT_READ | PROT_WRITE | PROT_SAO, + MAP_ANONYMOUS | MAP_PRIVATE, -1, 0); + FAIL_IF(p == MAP_FAILED); + + /* Write to the mapping, to at least cause a fault */ + memset(p, 0xaa, SIZE); + + return 0; +} + +int main(void) +{ + return test_harness(test_prot_sao, "prot-sao"); +} diff --git a/tools/testing/selftests/x86/test_vsyscall.c b/tools/testing/selftests/x86/test_vsyscall.c index c41f24b517f4..65c141ebfbbd 100644 --- a/tools/testing/selftests/x86/test_vsyscall.c +++ b/tools/testing/selftests/x86/test_vsyscall.c @@ -462,6 +462,17 @@ static int test_vsys_x(void) return 0; } +/* + * Debuggers expect ptrace() to be able to peek at the vsyscall page. + * Use process_vm_readv() as a proxy for ptrace() to test this. We + * want it to work in the vsyscall=emulate case and to fail in the + * vsyscall=xonly case. + * + * It's worth noting that this ABI is a bit nutty. write(2) can't + * read from the vsyscall page on any kernel version or mode. The + * fact that ptrace() ever worked was a nice courtesy of old kernels, + * but the code to support it is fairly gross. + */ static int test_process_vm_readv(void) { #ifdef __x86_64__ @@ -477,8 +488,12 @@ static int test_process_vm_readv(void) remote.iov_len = 4096; ret = process_vm_readv(getpid(), &local, 1, &remote, 1, 0); if (ret != 4096) { - printf("[OK]\tprocess_vm_readv() failed (ret = %d, errno = %d)\n", ret, errno); - return 0; + /* + * We expect process_vm_readv() to work if and only if the + * vsyscall page is readable. + */ + printf("[%s]\tprocess_vm_readv() failed (ret = %d, errno = %d)\n", vsyscall_map_r ? "FAIL" : "OK", ret, errno); + return vsyscall_map_r ? 1 : 0; } if (vsyscall_map_r) { @@ -488,6 +503,9 @@ static int test_process_vm_readv(void) printf("[FAIL]\tIt worked but returned incorrect data\n"); return 1; } + } else { + printf("[FAIL]\tprocess_rm_readv() succeeded, but it should have failed in this configuration\n"); + return 1; } #endif diff --git a/tools/usb/Build b/tools/usb/Build new file mode 100644 index 000000000000..2ad6f9745816 --- /dev/null +++ b/tools/usb/Build @@ -0,0 +1,2 @@ +testusb-y += testusb.o +ffs-test-y += ffs-test.o diff --git a/tools/usb/Makefile b/tools/usb/Makefile index 01d758d73b6d..1b128e551b2e 100644 --- a/tools/usb/Makefile +++ b/tools/usb/Makefile @@ -1,14 +1,51 @@ # SPDX-License-Identifier: GPL-2.0 # Makefile for USB tools +include ../scripts/Makefile.include -PTHREAD_LIBS = -lpthread -WARNINGS = -Wall -Wextra -CFLAGS = $(WARNINGS) -g -I../include -LDFLAGS = $(PTHREAD_LIBS) +bindir ?= /usr/bin -all: testusb ffs-test -%: %.c - $(CC) $(CFLAGS) -o $@ $^ $(LDFLAGS) +ifeq ($(srctree),) +srctree := $(patsubst %/,%,$(dir $(CURDIR))) +srctree := $(patsubst %/,%,$(dir $(srctree))) +endif + +# Do not use make's built-in rules +# (this improves performance and avoids hard-to-debug behaviour); +MAKEFLAGS += -r + +override CFLAGS += -O2 -Wall -Wextra -g -D_GNU_SOURCE -I$(OUTPUT)include -I$(srctree)/tools/include +override LDFLAGS += -lpthread + +ALL_TARGETS := testusb ffs-test +ALL_PROGRAMS := $(patsubst %,$(OUTPUT)%,$(ALL_TARGETS)) + +all: $(ALL_PROGRAMS) + +export srctree OUTPUT CC LD CFLAGS +include $(srctree)/tools/build/Makefile.include + +TESTUSB_IN := $(OUTPUT)testusb-in.o +$(TESTUSB_IN): FORCE + $(Q)$(MAKE) $(build)=testusb +$(OUTPUT)testusb: $(TESTUSB_IN) + $(QUIET_LINK)$(CC) $(CFLAGS) $< -o $@ $(LDFLAGS) + +FFS_TEST_IN := $(OUTPUT)ffs-test-in.o +$(FFS_TEST_IN): FORCE + $(Q)$(MAKE) $(build)=ffs-test +$(OUTPUT)ffs-test: $(FFS_TEST_IN) + $(QUIET_LINK)$(CC) $(CFLAGS) $< -o $@ $(LDFLAGS) clean: - $(RM) testusb ffs-test + rm -f $(ALL_PROGRAMS) + find $(if $(OUTPUT),$(OUTPUT),.) -name '*.o' -delete -o -name '\.*.d' -delete -o -name '\.*.o.cmd' -delete + +install: $(ALL_PROGRAMS) + install -d -m 755 $(DESTDIR)$(bindir); \ + for program in $(ALL_PROGRAMS); do \ + install $$program $(DESTDIR)$(bindir); \ + done + +FORCE: + +.PHONY: all install clean FORCE prepare |