From f3c93a93b564d25cab715cd71a6c3417e77b19f3 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Tue, 21 Jul 2020 21:38:04 -0700 Subject: tools/bpftool: Strip BPF .o files before skeleton generation Strip away DWARF info from .bpf.o files, before generating BPF skeletons. This reduces bpftool binary size from 3.43MB to 2.58MB. Signed-off-by: Andrii Nakryiko Signed-off-by: Alexei Starovoitov Acked-by: Quentin Monnet Link: https://lore.kernel.org/bpf/20200722043804.2373298-1-andriin@fb.com --- tools/bpf/bpftool/Makefile | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/bpf/bpftool/Makefile b/tools/bpf/bpftool/Makefile index 51bd520ed437..8462690a039b 100644 --- a/tools/bpf/bpftool/Makefile +++ b/tools/bpf/bpftool/Makefile @@ -59,6 +59,7 @@ endif INSTALL ?= install RM ?= rm -f CLANG ?= clang +LLVM_STRIP ?= llvm-strip FEATURE_USER = .bpftool FEATURE_TESTS = libbfd disassembler-four-args reallocarray zlib libcap \ @@ -147,7 +148,7 @@ $(OUTPUT)%.bpf.o: skeleton/%.bpf.c $(OUTPUT)vmlinux.h $(LIBBPF) -I$(srctree)/tools/include/uapi/ \ -I$(LIBBPF_PATH) \ -I$(srctree)/tools/lib \ - -g -O2 -target bpf -c $< -o $@ + -g -O2 -target bpf -c $< -o $@ && $(LLVM_STRIP) -g $@ $(OUTPUT)%.skel.h: $(OUTPUT)%.bpf.o $(BPFTOOL_BOOTSTRAP) $(QUIET_GEN)$(BPFTOOL_BOOTSTRAP) gen skeleton $< > $@ -- cgit v1.2.3 From 86176a1821a1237ad163d312ba0f2d7598757894 Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Wed, 22 Jul 2020 18:17:20 +0200 Subject: selftests/bpf: Test BPF socket lookup and reuseport with connections Cover the case when BPF socket lookup returns a socket that belongs to a reuseport group, and the reuseport group contains connected UDP sockets. Ensure that the presence of connected UDP sockets in reuseport group does not affect the socket lookup result. Socket selected by reuseport should always be used as result in such case. Signed-off-by: Jakub Sitnicki Signed-off-by: Alexei Starovoitov Acked-by: Kuniyuki Iwashima Link: https://lore.kernel.org/bpf/20200722161720.940831-3-jakub@cloudflare.com --- tools/testing/selftests/bpf/prog_tests/sk_lookup.c | 54 +++++++++++++++++++++- 1 file changed, 53 insertions(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c index f1784ae4565a..9bbd2b2b7630 100644 --- a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c +++ b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c @@ -74,6 +74,7 @@ struct test { struct inet_addr connect_to; struct inet_addr listen_at; enum server accept_on; + bool reuseport_has_conns; /* Add a connected socket to reuseport group */ }; static __u32 duration; /* for CHECK macro */ @@ -559,7 +560,8 @@ close: static void run_lookup_prog(const struct test *t) { - int client_fd, server_fds[MAX_SERVERS] = { -1 }; + int server_fds[MAX_SERVERS] = { -1 }; + int client_fd, reuse_conn_fd = -1; struct bpf_link *lookup_link; int i, err; @@ -583,6 +585,32 @@ static void run_lookup_prog(const struct test *t) break; } + /* Regular UDP socket lookup with reuseport behaves + * differently when reuseport group contains connected + * sockets. Check that adding a connected UDP socket to the + * reuseport group does not affect how reuseport works with + * BPF socket lookup. + */ + if (t->reuseport_has_conns) { + struct sockaddr_storage addr = {}; + socklen_t len = sizeof(addr); + + /* Add an extra socket to reuseport group */ + reuse_conn_fd = make_server(t->sotype, t->listen_at.ip, + t->listen_at.port, + t->reuseport_prog); + if (reuse_conn_fd < 0) + goto close; + + /* Connect the extra socket to itself */ + err = getsockname(reuse_conn_fd, (void *)&addr, &len); + if (CHECK(err, "getsockname", "errno %d\n", errno)) + goto close; + err = connect(reuse_conn_fd, (void *)&addr, len); + if (CHECK(err, "connect", "errno %d\n", errno)) + goto close; + } + client_fd = make_client(t->sotype, t->connect_to.ip, t->connect_to.port); if (client_fd < 0) goto close; @@ -594,6 +622,8 @@ static void run_lookup_prog(const struct test *t) close(client_fd); close: + if (reuse_conn_fd != -1) + close(reuse_conn_fd); for (i = 0; i < ARRAY_SIZE(server_fds); i++) { if (server_fds[i] != -1) close(server_fds[i]); @@ -710,6 +740,17 @@ static void test_redirect_lookup(struct test_sk_lookup *skel) .listen_at = { INT_IP4, INT_PORT }, .accept_on = SERVER_B, }, + { + .desc = "UDP IPv4 redir and reuseport with conns", + .lookup_prog = skel->progs.select_sock_a, + .reuseport_prog = skel->progs.select_sock_b, + .sock_map = skel->maps.redir_map, + .sotype = SOCK_DGRAM, + .connect_to = { EXT_IP4, EXT_PORT }, + .listen_at = { INT_IP4, INT_PORT }, + .accept_on = SERVER_B, + .reuseport_has_conns = true, + }, { .desc = "UDP IPv4 redir skip reuseport", .lookup_prog = skel->progs.select_sock_a_no_reuseport, @@ -754,6 +795,17 @@ static void test_redirect_lookup(struct test_sk_lookup *skel) .listen_at = { INT_IP6, INT_PORT }, .accept_on = SERVER_B, }, + { + .desc = "UDP IPv6 redir and reuseport with conns", + .lookup_prog = skel->progs.select_sock_a, + .reuseport_prog = skel->progs.select_sock_b, + .sock_map = skel->maps.redir_map, + .sotype = SOCK_DGRAM, + .connect_to = { EXT_IP6, EXT_PORT }, + .listen_at = { INT_IP6, INT_PORT }, + .accept_on = SERVER_B, + .reuseport_has_conns = true, + }, { .desc = "UDP IPv6 redir skip reuseport", .lookup_prog = skel->progs.select_sock_a_no_reuseport, -- cgit v1.2.3 From a5cbe05a6673b85bed2a63ffcfea6a96c6410cff Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Thu, 23 Jul 2020 11:41:12 -0700 Subject: bpf: Implement bpf iterator for map elements The bpf iterator for map elements are implemented. The bpf program will receive four parameters: bpf_iter_meta *meta: the meta data bpf_map *map: the bpf_map whose elements are traversed void *key: the key of one element void *value: the value of the same element Here, meta and map pointers are always valid, and key has register type PTR_TO_RDONLY_BUF_OR_NULL and value has register type PTR_TO_RDWR_BUF_OR_NULL. The kernel will track the access range of key and value during verification time. Later, these values will be compared against the values in the actual map to ensure all accesses are within range. A new field iter_seq_info is added to bpf_map_ops which is used to add map type specific information, i.e., seq_ops, init/fini seq_file func and seq_file private data size. Subsequent patches will have actual implementation for bpf_map_ops->iter_seq_info. In user space, BPF_ITER_LINK_MAP_FD needs to be specified in prog attr->link_create.flags, which indicates that attr->link_create.target_fd is a map_fd. The reason for such an explicit flag is for possible future cases where one bpf iterator may allow more than one possible customization, e.g., pid and cgroup id for task_file. Current kernel internal implementation only allows the target to register at most one required bpf_iter_link_info. To support the above case, optional bpf_iter_link_info's are needed, the target can be extended to register such link infos, and user provided link_info needs to match one of target supported ones. Signed-off-by: Yonghong Song Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/20200723184112.590360-1-yhs@fb.com --- include/linux/bpf.h | 16 ++++++++ include/uapi/linux/bpf.h | 7 ++++ kernel/bpf/bpf_iter.c | 85 ++++++++++++++++++++++++++++++++++-------- kernel/bpf/map_iter.c | 30 ++++++++++++++- tools/include/uapi/linux/bpf.h | 7 ++++ 5 files changed, 128 insertions(+), 17 deletions(-) (limited to 'tools') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index f9c4bb08f616..4175cf1f4665 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -107,6 +107,9 @@ struct bpf_map_ops { /* BTF name and id of struct allocated by map_alloc */ const char * const map_btf_name; int *map_btf_id; + + /* bpf_iter info used to open a seq_file */ + const struct bpf_iter_seq_info *iter_seq_info; }; struct bpf_map_memory { @@ -1207,12 +1210,18 @@ int bpf_obj_get_user(const char __user *pathname, int flags); int __init bpf_iter_ ## target(args) { return 0; } struct bpf_iter_aux_info { + struct bpf_map *map; }; +typedef int (*bpf_iter_check_target_t)(struct bpf_prog *prog, + struct bpf_iter_aux_info *aux); + #define BPF_ITER_CTX_ARG_MAX 2 struct bpf_iter_reg { const char *target; + bpf_iter_check_target_t check_target; u32 ctx_arg_info_size; + enum bpf_iter_link_info req_linfo; struct bpf_ctx_arg_aux ctx_arg_info[BPF_ITER_CTX_ARG_MAX]; const struct bpf_iter_seq_info *seq_info; }; @@ -1223,6 +1232,13 @@ struct bpf_iter_meta { u64 seq_num; }; +struct bpf_iter__bpf_map_elem { + __bpf_md_ptr(struct bpf_iter_meta *, meta); + __bpf_md_ptr(struct bpf_map *, map); + __bpf_md_ptr(void *, key); + __bpf_md_ptr(void *, value); +}; + int bpf_iter_reg_target(const struct bpf_iter_reg *reg_info); void bpf_iter_unreg_target(const struct bpf_iter_reg *reg_info); bool bpf_iter_prog_supported(struct bpf_prog *prog); diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 54d0c886e3ba..828c2f6438f2 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -246,6 +246,13 @@ enum bpf_link_type { MAX_BPF_LINK_TYPE, }; +enum bpf_iter_link_info { + BPF_ITER_LINK_UNSPEC = 0, + BPF_ITER_LINK_MAP_FD = 1, + + MAX_BPF_ITER_LINK_INFO, +}; + /* cgroup-bpf attach flags used in BPF_PROG_ATTACH command * * NONE(default): No further bpf programs allowed in the subtree. diff --git a/kernel/bpf/bpf_iter.c b/kernel/bpf/bpf_iter.c index 8fa94cb1b5a0..363b9cafc2d8 100644 --- a/kernel/bpf/bpf_iter.c +++ b/kernel/bpf/bpf_iter.c @@ -14,11 +14,13 @@ struct bpf_iter_target_info { struct bpf_iter_link { struct bpf_link link; + struct bpf_iter_aux_info aux; struct bpf_iter_target_info *tinfo; }; struct bpf_iter_priv_data { struct bpf_iter_target_info *tinfo; + const struct bpf_iter_seq_info *seq_info; struct bpf_prog *prog; u64 session_id; u64 seq_num; @@ -35,7 +37,8 @@ static DEFINE_MUTEX(link_mutex); /* incremented on every opened seq_file */ static atomic64_t session_id; -static int prepare_seq_file(struct file *file, struct bpf_iter_link *link); +static int prepare_seq_file(struct file *file, struct bpf_iter_link *link, + const struct bpf_iter_seq_info *seq_info); static void bpf_iter_inc_seq_num(struct seq_file *seq) { @@ -199,11 +202,25 @@ done: return copied; } +static const struct bpf_iter_seq_info * +__get_seq_info(struct bpf_iter_link *link) +{ + const struct bpf_iter_seq_info *seq_info; + + if (link->aux.map) { + seq_info = link->aux.map->ops->iter_seq_info; + if (seq_info) + return seq_info; + } + + return link->tinfo->reg_info->seq_info; +} + static int iter_open(struct inode *inode, struct file *file) { struct bpf_iter_link *link = inode->i_private; - return prepare_seq_file(file, link); + return prepare_seq_file(file, link, __get_seq_info(link)); } static int iter_release(struct inode *inode, struct file *file) @@ -218,8 +235,8 @@ static int iter_release(struct inode *inode, struct file *file) iter_priv = container_of(seq->private, struct bpf_iter_priv_data, target_private); - if (iter_priv->tinfo->reg_info->seq_info->fini_seq_private) - iter_priv->tinfo->reg_info->seq_info->fini_seq_private(seq->private); + if (iter_priv->seq_info->fini_seq_private) + iter_priv->seq_info->fini_seq_private(seq->private); bpf_prog_put(iter_priv->prog); seq->private = iter_priv; @@ -318,6 +335,11 @@ bool bpf_iter_prog_supported(struct bpf_prog *prog) static void bpf_iter_link_release(struct bpf_link *link) { + struct bpf_iter_link *iter_link = + container_of(link, struct bpf_iter_link, link); + + if (iter_link->aux.map) + bpf_map_put_with_uref(iter_link->aux.map); } static void bpf_iter_link_dealloc(struct bpf_link *link) @@ -370,14 +392,13 @@ int bpf_iter_link_attach(const union bpf_attr *attr, struct bpf_prog *prog) { struct bpf_link_primer link_primer; struct bpf_iter_target_info *tinfo; + struct bpf_iter_aux_info aux = {}; struct bpf_iter_link *link; + u32 prog_btf_id, target_fd; bool existed = false; - u32 prog_btf_id; + struct bpf_map *map; int err; - if (attr->link_create.target_fd || attr->link_create.flags) - return -EINVAL; - prog_btf_id = prog->aux->attach_btf_id; mutex_lock(&targets_mutex); list_for_each_entry(tinfo, &targets, list) { @@ -390,6 +411,13 @@ int bpf_iter_link_attach(const union bpf_attr *attr, struct bpf_prog *prog) if (!existed) return -ENOENT; + /* Make sure user supplied flags are target expected. */ + target_fd = attr->link_create.target_fd; + if (attr->link_create.flags != tinfo->reg_info->req_linfo) + return -EINVAL; + if (!attr->link_create.flags && target_fd) + return -EINVAL; + link = kzalloc(sizeof(*link), GFP_USER | __GFP_NOWARN); if (!link) return -ENOMEM; @@ -403,21 +431,45 @@ int bpf_iter_link_attach(const union bpf_attr *attr, struct bpf_prog *prog) return err; } + if (tinfo->reg_info->req_linfo == BPF_ITER_LINK_MAP_FD) { + map = bpf_map_get_with_uref(target_fd); + if (IS_ERR(map)) { + err = PTR_ERR(map); + goto cleanup_link; + } + + aux.map = map; + err = tinfo->reg_info->check_target(prog, &aux); + if (err) { + bpf_map_put_with_uref(map); + goto cleanup_link; + } + + link->aux.map = map; + } + return bpf_link_settle(&link_primer); + +cleanup_link: + bpf_link_cleanup(&link_primer); + return err; } static void init_seq_meta(struct bpf_iter_priv_data *priv_data, struct bpf_iter_target_info *tinfo, + const struct bpf_iter_seq_info *seq_info, struct bpf_prog *prog) { priv_data->tinfo = tinfo; + priv_data->seq_info = seq_info; priv_data->prog = prog; priv_data->session_id = atomic64_inc_return(&session_id); priv_data->seq_num = 0; priv_data->done_stop = false; } -static int prepare_seq_file(struct file *file, struct bpf_iter_link *link) +static int prepare_seq_file(struct file *file, struct bpf_iter_link *link, + const struct bpf_iter_seq_info *seq_info) { struct bpf_iter_priv_data *priv_data; struct bpf_iter_target_info *tinfo; @@ -433,21 +485,21 @@ static int prepare_seq_file(struct file *file, struct bpf_iter_link *link) tinfo = link->tinfo; total_priv_dsize = offsetof(struct bpf_iter_priv_data, target_private) + - tinfo->reg_info->seq_info->seq_priv_size; - priv_data = __seq_open_private(file, tinfo->reg_info->seq_info->seq_ops, + seq_info->seq_priv_size; + priv_data = __seq_open_private(file, seq_info->seq_ops, total_priv_dsize); if (!priv_data) { err = -ENOMEM; goto release_prog; } - if (tinfo->reg_info->seq_info->init_seq_private) { - err = tinfo->reg_info->seq_info->init_seq_private(priv_data->target_private, NULL); + if (seq_info->init_seq_private) { + err = seq_info->init_seq_private(priv_data->target_private, &link->aux); if (err) goto release_seq_file; } - init_seq_meta(priv_data, tinfo, prog); + init_seq_meta(priv_data, tinfo, seq_info, prog); seq = file->private_data; seq->private = priv_data->target_private; @@ -463,6 +515,7 @@ release_prog: int bpf_iter_new_fd(struct bpf_link *link) { + struct bpf_iter_link *iter_link; struct file *file; unsigned int flags; int err, fd; @@ -481,8 +534,8 @@ int bpf_iter_new_fd(struct bpf_link *link) goto free_fd; } - err = prepare_seq_file(file, - container_of(link, struct bpf_iter_link, link)); + iter_link = container_of(link, struct bpf_iter_link, link); + err = prepare_seq_file(file, iter_link, __get_seq_info(iter_link)); if (err) goto free_file; diff --git a/kernel/bpf/map_iter.c b/kernel/bpf/map_iter.c index 1a69241fb1e2..8a1f9b3355d0 100644 --- a/kernel/bpf/map_iter.c +++ b/kernel/bpf/map_iter.c @@ -98,10 +98,38 @@ static struct bpf_iter_reg bpf_map_reg_info = { .seq_info = &bpf_map_seq_info, }; +static int bpf_iter_check_map(struct bpf_prog *prog, + struct bpf_iter_aux_info *aux) +{ + return -EINVAL; +} + +DEFINE_BPF_ITER_FUNC(bpf_map_elem, struct bpf_iter_meta *meta, + struct bpf_map *map, void *key, void *value) + +static const struct bpf_iter_reg bpf_map_elem_reg_info = { + .target = "bpf_map_elem", + .check_target = bpf_iter_check_map, + .req_linfo = BPF_ITER_LINK_MAP_FD, + .ctx_arg_info_size = 2, + .ctx_arg_info = { + { offsetof(struct bpf_iter__bpf_map_elem, key), + PTR_TO_RDONLY_BUF_OR_NULL }, + { offsetof(struct bpf_iter__bpf_map_elem, value), + PTR_TO_RDWR_BUF_OR_NULL }, + }, +}; + static int __init bpf_map_iter_init(void) { + int ret; + bpf_map_reg_info.ctx_arg_info[0].btf_id = *btf_bpf_map_id; - return bpf_iter_reg_target(&bpf_map_reg_info); + ret = bpf_iter_reg_target(&bpf_map_reg_info); + if (ret) + return ret; + + return bpf_iter_reg_target(&bpf_map_elem_reg_info); } late_initcall(bpf_map_iter_init); diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index 54d0c886e3ba..828c2f6438f2 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -246,6 +246,13 @@ enum bpf_link_type { MAX_BPF_LINK_TYPE, }; +enum bpf_iter_link_info { + BPF_ITER_LINK_UNSPEC = 0, + BPF_ITER_LINK_MAP_FD = 1, + + MAX_BPF_ITER_LINK_INFO, +}; + /* cgroup-bpf attach flags used in BPF_PROG_ATTACH command * * NONE(default): No further bpf programs allowed in the subtree. -- cgit v1.2.3 From cd31039a7347610863aa8b77a9162048999723d0 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Thu, 23 Jul 2020 11:41:17 -0700 Subject: tools/libbpf: Add support for bpf map element iterator Add map_fd to bpf_iter_attach_opts and flags to bpf_link_create_opts. Later on, bpftool or selftest will be able to create a bpf map element iterator by passing map_fd to the kernel during link creation time. Signed-off-by: Yonghong Song Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/20200723184117.590673-1-yhs@fb.com --- tools/lib/bpf/bpf.c | 1 + tools/lib/bpf/bpf.h | 3 ++- tools/lib/bpf/libbpf.c | 10 +++++++++- tools/lib/bpf/libbpf.h | 3 ++- 4 files changed, 14 insertions(+), 3 deletions(-) (limited to 'tools') diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c index a7329b671c41..e1bdf214f75f 100644 --- a/tools/lib/bpf/bpf.c +++ b/tools/lib/bpf/bpf.c @@ -598,6 +598,7 @@ int bpf_link_create(int prog_fd, int target_fd, attr.link_create.prog_fd = prog_fd; attr.link_create.target_fd = target_fd; attr.link_create.attach_type = attach_type; + attr.link_create.flags = OPTS_GET(opts, flags, 0); return sys_bpf(BPF_LINK_CREATE, &attr, sizeof(attr)); } diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h index dbef24ebcfcb..6d367e01d05e 100644 --- a/tools/lib/bpf/bpf.h +++ b/tools/lib/bpf/bpf.h @@ -170,8 +170,9 @@ LIBBPF_API int bpf_prog_detach2(int prog_fd, int attachable_fd, struct bpf_link_create_opts { size_t sz; /* size of this struct for forward/backward compatibility */ + __u32 flags; }; -#define bpf_link_create_opts__last_field sz +#define bpf_link_create_opts__last_field flags LIBBPF_API int bpf_link_create(int prog_fd, int target_fd, enum bpf_attach_type attach_type, diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index 846164c79df1..a05aa7e2bab6 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -8282,13 +8282,20 @@ struct bpf_link * bpf_program__attach_iter(struct bpf_program *prog, const struct bpf_iter_attach_opts *opts) { + DECLARE_LIBBPF_OPTS(bpf_link_create_opts, link_create_opts); char errmsg[STRERR_BUFSIZE]; struct bpf_link *link; int prog_fd, link_fd; + __u32 target_fd = 0; if (!OPTS_VALID(opts, bpf_iter_attach_opts)) return ERR_PTR(-EINVAL); + if (OPTS_HAS(opts, map_fd)) { + target_fd = opts->map_fd; + link_create_opts.flags = BPF_ITER_LINK_MAP_FD; + } + prog_fd = bpf_program__fd(prog); if (prog_fd < 0) { pr_warn("program '%s': can't attach before loaded\n", @@ -8301,7 +8308,8 @@ bpf_program__attach_iter(struct bpf_program *prog, return ERR_PTR(-ENOMEM); link->detach = &bpf_link__detach_fd; - link_fd = bpf_link_create(prog_fd, 0, BPF_TRACE_ITER, NULL); + link_fd = bpf_link_create(prog_fd, target_fd, BPF_TRACE_ITER, + &link_create_opts); if (link_fd < 0) { link_fd = -errno; free(link); diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h index c2272132e929..c6813791fa7e 100644 --- a/tools/lib/bpf/libbpf.h +++ b/tools/lib/bpf/libbpf.h @@ -264,8 +264,9 @@ LIBBPF_API struct bpf_link *bpf_map__attach_struct_ops(struct bpf_map *map); struct bpf_iter_attach_opts { size_t sz; /* size of this struct for forward/backward compatibility */ + __u32 map_fd; }; -#define bpf_iter_attach_opts__last_field sz +#define bpf_iter_attach_opts__last_field map_fd LIBBPF_API struct bpf_link * bpf_program__attach_iter(struct bpf_program *prog, -- cgit v1.2.3 From d8793aca708602c676372b03d6493972457524af Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Thu, 23 Jul 2020 11:41:19 -0700 Subject: tools/bpftool: Add bpftool support for bpf map element iterator The optional parameter "map MAP" can be added to "bpftool iter" command to create a bpf iterator for map elements. For example, bpftool iter pin ./prog.o /sys/fs/bpf/p1 map id 333 For map element bpf iterator "map MAP" parameter is required. Otherwise, bpf link creation will return an error. Quentin Monnet kindly provided bash-completion implementation for new "map MAP" option. Signed-off-by: Yonghong Song Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/20200723184119.590799-1-yhs@fb.com --- tools/bpf/bpftool/Documentation/bpftool-iter.rst | 18 +++++++++++-- tools/bpf/bpftool/bash-completion/bpftool | 18 ++++++++++++- tools/bpf/bpftool/iter.c | 33 +++++++++++++++++++++--- 3 files changed, 62 insertions(+), 7 deletions(-) (limited to 'tools') diff --git a/tools/bpf/bpftool/Documentation/bpftool-iter.rst b/tools/bpf/bpftool/Documentation/bpftool-iter.rst index 8dce698eab79..070ffacb42b5 100644 --- a/tools/bpf/bpftool/Documentation/bpftool-iter.rst +++ b/tools/bpf/bpftool/Documentation/bpftool-iter.rst @@ -17,14 +17,15 @@ SYNOPSIS ITER COMMANDS =================== -| **bpftool** **iter pin** *OBJ* *PATH* +| **bpftool** **iter pin** *OBJ* *PATH* [**map** *MAP*] | **bpftool** **iter help** | | *OBJ* := /a/file/of/bpf_iter_target.o +| *MAP* := { **id** *MAP_ID* | **pinned** *FILE* } DESCRIPTION =========== - **bpftool iter pin** *OBJ* *PATH* + **bpftool iter pin** *OBJ* *PATH* [**map** *MAP*] A bpf iterator combines a kernel iterating of particular kernel data (e.g., tasks, bpf_maps, etc.) and a bpf program called for each kernel data object @@ -37,6 +38,12 @@ DESCRIPTION character ('.'), which is reserved for future extensions of *bpffs*. + Map element bpf iterator requires an additional parameter + *MAP* so bpf program can iterate over map elements for + that map. User can have a bpf program in kernel to run + with each map element, do checking, filtering, aggregation, + etc. without copying data to user space. + User can then *cat PATH* to see the bpf iterator output. **bpftool iter help** @@ -64,6 +71,13 @@ EXAMPLES Create a file-based bpf iterator from bpf_iter_netlink.o and pin it to /sys/fs/bpf/my_netlink +**# bpftool iter pin bpf_iter_hashmap.o /sys/fs/bpf/my_hashmap map id 20** + +:: + + Create a file-based bpf iterator from bpf_iter_hashmap.o and map with + id 20, and pin it to /sys/fs/bpf/my_hashmap + SEE ALSO ======== **bpf**\ (2), diff --git a/tools/bpf/bpftool/bash-completion/bpftool b/tools/bpf/bpftool/bash-completion/bpftool index 7b137264ea3a..257fa310ea2b 100644 --- a/tools/bpf/bpftool/bash-completion/bpftool +++ b/tools/bpf/bpftool/bash-completion/bpftool @@ -615,7 +615,23 @@ _bpftool() iter) case $command in pin) - _filedir + case $prev in + $command) + _filedir + ;; + id) + _bpftool_get_map_ids + ;; + name) + _bpftool_get_map_names + ;; + pinned) + _filedir + ;; + *) + _bpftool_one_of_list $MAP_TYPE + ;; + esac return 0 ;; *) diff --git a/tools/bpf/bpftool/iter.c b/tools/bpf/bpftool/iter.c index 33240fcc6319..c9dba7543dba 100644 --- a/tools/bpf/bpftool/iter.c +++ b/tools/bpf/bpftool/iter.c @@ -2,6 +2,7 @@ // Copyright (C) 2020 Facebook #define _GNU_SOURCE +#include #include #include @@ -9,11 +10,12 @@ static int do_pin(int argc, char **argv) { + DECLARE_LIBBPF_OPTS(bpf_iter_attach_opts, iter_opts); const char *objfile, *path; struct bpf_program *prog; struct bpf_object *obj; struct bpf_link *link; - int err; + int err = -1, map_fd = -1; if (!REQ_ARGS(2)) usage(); @@ -21,10 +23,26 @@ static int do_pin(int argc, char **argv) objfile = GET_ARG(); path = GET_ARG(); + /* optional arguments */ + if (argc) { + if (is_prefix(*argv, "map")) { + NEXT_ARG(); + + if (!REQ_ARGS(2)) { + p_err("incorrect map spec"); + return -1; + } + + map_fd = map_parse_fd(&argc, &argv); + if (map_fd < 0) + return -1; + } + } + obj = bpf_object__open(objfile); if (IS_ERR(obj)) { p_err("can't open objfile %s", objfile); - return -1; + goto close_map_fd; } err = bpf_object__load(obj); @@ -39,7 +57,10 @@ static int do_pin(int argc, char **argv) goto close_obj; } - link = bpf_program__attach_iter(prog, NULL); + if (map_fd >= 0) + iter_opts.map_fd = map_fd; + + link = bpf_program__attach_iter(prog, &iter_opts); if (IS_ERR(link)) { err = PTR_ERR(link); p_err("attach_iter failed for program %s", @@ -62,14 +83,18 @@ close_link: bpf_link__destroy(link); close_obj: bpf_object__close(obj); +close_map_fd: + if (map_fd >= 0) + close(map_fd); return err; } static int do_help(int argc, char **argv) { fprintf(stderr, - "Usage: %1$s %2$s pin OBJ PATH\n" + "Usage: %1$s %2$s pin OBJ PATH [map MAP]\n" " %1$s %2$s help\n" + " " HELP_SPEC_MAP "\n" "", bin_name, "iter"); -- cgit v1.2.3 From 2a7c2fff7dd6e87634e47ddb2d2c7f272708dbbf Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Thu, 23 Jul 2020 11:41:20 -0700 Subject: selftests/bpf: Add test for bpf hash map iterators Two subtests are added. $ ./test_progs -n 4 ... #4/18 bpf_hash_map:OK #4/19 bpf_percpu_hash_map:OK ... Signed-off-by: Yonghong Song Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/20200723184120.590916-1-yhs@fb.com --- tools/testing/selftests/bpf/prog_tests/bpf_iter.c | 187 +++++++++++++++++++++ .../selftests/bpf/progs/bpf_iter_bpf_hash_map.c | 100 +++++++++++ .../bpf/progs/bpf_iter_bpf_percpu_hash_map.c | 50 ++++++ 3 files changed, 337 insertions(+) create mode 100644 tools/testing/selftests/bpf/progs/bpf_iter_bpf_hash_map.c create mode 100644 tools/testing/selftests/bpf/progs/bpf_iter_bpf_percpu_hash_map.c (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c index fed42755416d..72790b600c62 100644 --- a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c +++ b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c @@ -15,6 +15,8 @@ #include "bpf_iter_test_kern2.skel.h" #include "bpf_iter_test_kern3.skel.h" #include "bpf_iter_test_kern4.skel.h" +#include "bpf_iter_bpf_hash_map.skel.h" +#include "bpf_iter_bpf_percpu_hash_map.skel.h" static int duration; @@ -455,6 +457,187 @@ out: bpf_iter_test_kern4__destroy(skel); } +static void test_bpf_hash_map(void) +{ + __u32 expected_key_a = 0, expected_key_b = 0, expected_key_c = 0; + DECLARE_LIBBPF_OPTS(bpf_iter_attach_opts, opts); + struct bpf_iter_bpf_hash_map *skel; + int err, i, len, map_fd, iter_fd; + __u64 val, expected_val = 0; + struct bpf_link *link; + struct key_t { + int a; + int b; + int c; + } key; + char buf[64]; + + skel = bpf_iter_bpf_hash_map__open(); + if (CHECK(!skel, "bpf_iter_bpf_hash_map__open", + "skeleton open failed\n")) + return; + + skel->bss->in_test_mode = true; + + err = bpf_iter_bpf_hash_map__load(skel); + if (CHECK(!skel, "bpf_iter_bpf_hash_map__load", + "skeleton load failed\n")) + goto out; + + /* iterator with hashmap2 and hashmap3 should fail */ + opts.map_fd = bpf_map__fd(skel->maps.hashmap2); + link = bpf_program__attach_iter(skel->progs.dump_bpf_hash_map, &opts); + if (CHECK(!IS_ERR(link), "attach_iter", + "attach_iter for hashmap2 unexpected succeeded\n")) + goto out; + + opts.map_fd = bpf_map__fd(skel->maps.hashmap3); + link = bpf_program__attach_iter(skel->progs.dump_bpf_hash_map, &opts); + if (CHECK(!IS_ERR(link), "attach_iter", + "attach_iter for hashmap3 unexpected succeeded\n")) + goto out; + + /* hashmap1 should be good, update map values here */ + map_fd = bpf_map__fd(skel->maps.hashmap1); + for (i = 0; i < bpf_map__max_entries(skel->maps.hashmap1); i++) { + key.a = i + 1; + key.b = i + 2; + key.c = i + 3; + val = i + 4; + expected_key_a += key.a; + expected_key_b += key.b; + expected_key_c += key.c; + expected_val += val; + + err = bpf_map_update_elem(map_fd, &key, &val, BPF_ANY); + if (CHECK(err, "map_update", "map_update failed\n")) + goto out; + } + + opts.map_fd = map_fd; + link = bpf_program__attach_iter(skel->progs.dump_bpf_hash_map, &opts); + if (CHECK(IS_ERR(link), "attach_iter", "attach_iter failed\n")) + goto out; + + iter_fd = bpf_iter_create(bpf_link__fd(link)); + if (CHECK(iter_fd < 0, "create_iter", "create_iter failed\n")) + goto free_link; + + /* do some tests */ + while ((len = read(iter_fd, buf, sizeof(buf))) > 0) + ; + if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + goto close_iter; + + /* test results */ + if (CHECK(skel->bss->key_sum_a != expected_key_a, + "key_sum_a", "got %u expected %u\n", + skel->bss->key_sum_a, expected_key_a)) + goto close_iter; + if (CHECK(skel->bss->key_sum_b != expected_key_b, + "key_sum_b", "got %u expected %u\n", + skel->bss->key_sum_b, expected_key_b)) + goto close_iter; + if (CHECK(skel->bss->val_sum != expected_val, + "val_sum", "got %llu expected %llu\n", + skel->bss->val_sum, expected_val)) + goto close_iter; + +close_iter: + close(iter_fd); +free_link: + bpf_link__destroy(link); +out: + bpf_iter_bpf_hash_map__destroy(skel); +} + +static void test_bpf_percpu_hash_map(void) +{ + __u32 expected_key_a = 0, expected_key_b = 0, expected_key_c = 0; + DECLARE_LIBBPF_OPTS(bpf_iter_attach_opts, opts); + struct bpf_iter_bpf_percpu_hash_map *skel; + int err, i, j, len, map_fd, iter_fd; + __u32 expected_val = 0; + struct bpf_link *link; + struct key_t { + int a; + int b; + int c; + } key; + char buf[64]; + void *val; + + val = malloc(8 * bpf_num_possible_cpus()); + + skel = bpf_iter_bpf_percpu_hash_map__open(); + if (CHECK(!skel, "bpf_iter_bpf_percpu_hash_map__open", + "skeleton open failed\n")) + return; + + skel->rodata->num_cpus = bpf_num_possible_cpus(); + + err = bpf_iter_bpf_percpu_hash_map__load(skel); + if (CHECK(!skel, "bpf_iter_bpf_percpu_hash_map__load", + "skeleton load failed\n")) + goto out; + + /* update map values here */ + map_fd = bpf_map__fd(skel->maps.hashmap1); + for (i = 0; i < bpf_map__max_entries(skel->maps.hashmap1); i++) { + key.a = i + 1; + key.b = i + 2; + key.c = i + 3; + expected_key_a += key.a; + expected_key_b += key.b; + expected_key_c += key.c; + + for (j = 0; j < bpf_num_possible_cpus(); j++) { + *(__u32 *)(val + j * 8) = i + j; + expected_val += i + j; + } + + err = bpf_map_update_elem(map_fd, &key, val, BPF_ANY); + if (CHECK(err, "map_update", "map_update failed\n")) + goto out; + } + + opts.map_fd = map_fd; + link = bpf_program__attach_iter(skel->progs.dump_bpf_percpu_hash_map, &opts); + if (CHECK(IS_ERR(link), "attach_iter", "attach_iter failed\n")) + goto out; + + iter_fd = bpf_iter_create(bpf_link__fd(link)); + if (CHECK(iter_fd < 0, "create_iter", "create_iter failed\n")) + goto free_link; + + /* do some tests */ + while ((len = read(iter_fd, buf, sizeof(buf))) > 0) + ; + if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + goto close_iter; + + /* test results */ + if (CHECK(skel->bss->key_sum_a != expected_key_a, + "key_sum_a", "got %u expected %u\n", + skel->bss->key_sum_a, expected_key_a)) + goto close_iter; + if (CHECK(skel->bss->key_sum_b != expected_key_b, + "key_sum_b", "got %u expected %u\n", + skel->bss->key_sum_b, expected_key_b)) + goto close_iter; + if (CHECK(skel->bss->val_sum != expected_val, + "val_sum", "got %u expected %u\n", + skel->bss->val_sum, expected_val)) + goto close_iter; + +close_iter: + close(iter_fd); +free_link: + bpf_link__destroy(link); +out: + bpf_iter_bpf_percpu_hash_map__destroy(skel); +} + void test_bpf_iter(void) { if (test__start_subtest("btf_id_or_null")) @@ -491,4 +674,8 @@ void test_bpf_iter(void) test_overflow(true, false); if (test__start_subtest("prog-ret-1")) test_overflow(false, true); + if (test__start_subtest("bpf_hash_map")) + test_bpf_hash_map(); + if (test__start_subtest("bpf_percpu_hash_map")) + test_bpf_percpu_hash_map(); } diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_bpf_hash_map.c b/tools/testing/selftests/bpf/progs/bpf_iter_bpf_hash_map.c new file mode 100644 index 000000000000..07ddbfdbcab7 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/bpf_iter_bpf_hash_map.c @@ -0,0 +1,100 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2020 Facebook */ +#include "bpf_iter.h" +#include +#include + +char _license[] SEC("license") = "GPL"; + +struct key_t { + int a; + int b; + int c; +}; + +struct { + __uint(type, BPF_MAP_TYPE_HASH); + __uint(max_entries, 3); + __type(key, struct key_t); + __type(value, __u64); +} hashmap1 SEC(".maps"); + +struct { + __uint(type, BPF_MAP_TYPE_HASH); + __uint(max_entries, 3); + __type(key, __u64); + __type(value, __u64); +} hashmap2 SEC(".maps"); + +struct { + __uint(type, BPF_MAP_TYPE_HASH); + __uint(max_entries, 3); + __type(key, struct key_t); + __type(value, __u32); +} hashmap3 SEC(".maps"); + +/* will set before prog run */ +bool in_test_mode = 0; + +/* will collect results during prog run */ +__u32 key_sum_a = 0, key_sum_b = 0, key_sum_c = 0; +__u64 val_sum = 0; + +SEC("iter/bpf_map_elem") +int dump_bpf_hash_map(struct bpf_iter__bpf_map_elem *ctx) +{ + struct seq_file *seq = ctx->meta->seq; + __u32 seq_num = ctx->meta->seq_num; + struct bpf_map *map = ctx->map; + struct key_t *key = ctx->key; + __u64 *val = ctx->value; + + if (in_test_mode) { + /* test mode is used by selftests to + * test functionality of bpf_hash_map iter. + * + * the above hashmap1 will have correct size + * and will be accepted, hashmap2 and hashmap3 + * should be rejected due to smaller key/value + * size. + */ + if (key == (void *)0 || val == (void *)0) + return 0; + + key_sum_a += key->a; + key_sum_b += key->b; + key_sum_c += key->c; + val_sum += *val; + return 0; + } + + /* non-test mode, the map is prepared with the + * below bpftool command sequence: + * bpftool map create /sys/fs/bpf/m1 type hash \ + * key 12 value 8 entries 3 name map1 + * bpftool map update id 77 key 0 0 0 1 0 0 0 0 0 0 0 1 \ + * value 0 0 0 1 0 0 0 1 + * bpftool map update id 77 key 0 0 0 1 0 0 0 0 0 0 0 2 \ + * value 0 0 0 1 0 0 0 2 + * The bpftool iter command line: + * bpftool iter pin ./bpf_iter_bpf_hash_map.o /sys/fs/bpf/p1 \ + * map id 77 + * The below output will be: + * map dump starts + * 77: (1000000 0 2000000) (200000001000000) + * 77: (1000000 0 1000000) (100000001000000) + * map dump ends + */ + if (seq_num == 0) + BPF_SEQ_PRINTF(seq, "map dump starts\n"); + + if (key == (void *)0 || val == (void *)0) { + BPF_SEQ_PRINTF(seq, "map dump ends\n"); + return 0; + } + + BPF_SEQ_PRINTF(seq, "%d: (%x %d %x) (%llx)\n", map->id, + key->a, key->b, key->c, *val); + + return 0; +} diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_bpf_percpu_hash_map.c b/tools/testing/selftests/bpf/progs/bpf_iter_bpf_percpu_hash_map.c new file mode 100644 index 000000000000..feaaa2b89c57 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/bpf_iter_bpf_percpu_hash_map.c @@ -0,0 +1,50 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2020 Facebook */ +#include "bpf_iter.h" +#include +#include + +char _license[] SEC("license") = "GPL"; + +struct key_t { + int a; + int b; + int c; +}; + +struct { + __uint(type, BPF_MAP_TYPE_PERCPU_HASH); + __uint(max_entries, 3); + __type(key, struct key_t); + __type(value, __u32); +} hashmap1 SEC(".maps"); + +/* will set before prog run */ +volatile const __u32 num_cpus = 0; + +/* will collect results during prog run */ +__u32 key_sum_a = 0, key_sum_b = 0, key_sum_c = 0; +__u32 val_sum = 0; + +SEC("iter/bpf_map_elem") +int dump_bpf_percpu_hash_map(struct bpf_iter__bpf_map_elem *ctx) +{ + struct key_t *key = ctx->key; + void *pptr = ctx->value; + __u32 step; + int i; + + if (key == (void *)0 || pptr == (void *)0) + return 0; + + key_sum_a += key->a; + key_sum_b += key->b; + key_sum_c += key->c; + + step = 8; + for (i = 0; i < num_cpus; i++) { + val_sum += *(__u32 *)pptr; + pptr += step; + } + return 0; +} -- cgit v1.2.3 From 60dd49ea65390986a665c462da704927e861e67e Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Thu, 23 Jul 2020 11:41:21 -0700 Subject: selftests/bpf: Add test for bpf array map iterators Two subtests are added. $ ./test_progs -n 4 ... #4/20 bpf_array_map:OK #4/21 bpf_percpu_array_map:OK ... The bpf_array_map subtest also tested bpf program changing array element values and send key/value to user space through bpf_seq_write() interface. Signed-off-by: Yonghong Song Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/20200723184121.591367-1-yhs@fb.com --- tools/testing/selftests/bpf/prog_tests/bpf_iter.c | 161 +++++++++++++++++++++ .../selftests/bpf/progs/bpf_iter_bpf_array_map.c | 40 +++++ .../bpf/progs/bpf_iter_bpf_percpu_array_map.c | 46 ++++++ 3 files changed, 247 insertions(+) create mode 100644 tools/testing/selftests/bpf/progs/bpf_iter_bpf_array_map.c create mode 100644 tools/testing/selftests/bpf/progs/bpf_iter_bpf_percpu_array_map.c (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c index 72790b600c62..4a02b2222a6d 100644 --- a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c +++ b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c @@ -17,6 +17,8 @@ #include "bpf_iter_test_kern4.skel.h" #include "bpf_iter_bpf_hash_map.skel.h" #include "bpf_iter_bpf_percpu_hash_map.skel.h" +#include "bpf_iter_bpf_array_map.skel.h" +#include "bpf_iter_bpf_percpu_array_map.skel.h" static int duration; @@ -638,6 +640,161 @@ out: bpf_iter_bpf_percpu_hash_map__destroy(skel); } +static void test_bpf_array_map(void) +{ + __u64 val, expected_val = 0, res_first_val, first_val = 0; + DECLARE_LIBBPF_OPTS(bpf_iter_attach_opts, opts); + __u32 expected_key = 0, res_first_key; + struct bpf_iter_bpf_array_map *skel; + int err, i, map_fd, iter_fd; + struct bpf_link *link; + char buf[64] = {}; + int len, start; + + skel = bpf_iter_bpf_array_map__open_and_load(); + if (CHECK(!skel, "bpf_iter_bpf_array_map__open_and_load", + "skeleton open_and_load failed\n")) + return; + + map_fd = bpf_map__fd(skel->maps.arraymap1); + for (i = 0; i < bpf_map__max_entries(skel->maps.arraymap1); i++) { + val = i + 4; + expected_key += i; + expected_val += val; + + if (i == 0) + first_val = val; + + err = bpf_map_update_elem(map_fd, &i, &val, BPF_ANY); + if (CHECK(err, "map_update", "map_update failed\n")) + goto out; + } + + opts.map_fd = map_fd; + link = bpf_program__attach_iter(skel->progs.dump_bpf_array_map, &opts); + if (CHECK(IS_ERR(link), "attach_iter", "attach_iter failed\n")) + goto out; + + iter_fd = bpf_iter_create(bpf_link__fd(link)); + if (CHECK(iter_fd < 0, "create_iter", "create_iter failed\n")) + goto free_link; + + /* do some tests */ + start = 0; + while ((len = read(iter_fd, buf + start, sizeof(buf) - start)) > 0) + start += len; + if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + goto close_iter; + + /* test results */ + res_first_key = *(__u32 *)buf; + res_first_val = *(__u64 *)(buf + sizeof(__u32)); + if (CHECK(res_first_key != 0 || res_first_val != first_val, + "bpf_seq_write", + "seq_write failure: first key %u vs expected 0, " + " first value %llu vs expected %llu\n", + res_first_key, res_first_val, first_val)) + goto close_iter; + + if (CHECK(skel->bss->key_sum != expected_key, + "key_sum", "got %u expected %u\n", + skel->bss->key_sum, expected_key)) + goto close_iter; + if (CHECK(skel->bss->val_sum != expected_val, + "val_sum", "got %llu expected %llu\n", + skel->bss->val_sum, expected_val)) + goto close_iter; + + for (i = 0; i < bpf_map__max_entries(skel->maps.arraymap1); i++) { + err = bpf_map_lookup_elem(map_fd, &i, &val); + if (CHECK(err, "map_lookup", "map_lookup failed\n")) + goto out; + if (CHECK(i != val, "invalid_val", + "got value %llu expected %u\n", val, i)) + goto out; + } + +close_iter: + close(iter_fd); +free_link: + bpf_link__destroy(link); +out: + bpf_iter_bpf_array_map__destroy(skel); +} + +static void test_bpf_percpu_array_map(void) +{ + DECLARE_LIBBPF_OPTS(bpf_iter_attach_opts, opts); + struct bpf_iter_bpf_percpu_array_map *skel; + __u32 expected_key = 0, expected_val = 0; + int err, i, j, map_fd, iter_fd; + struct bpf_link *link; + char buf[64]; + void *val; + int len; + + val = malloc(8 * bpf_num_possible_cpus()); + + skel = bpf_iter_bpf_percpu_array_map__open(); + if (CHECK(!skel, "bpf_iter_bpf_percpu_array_map__open", + "skeleton open failed\n")) + return; + + skel->rodata->num_cpus = bpf_num_possible_cpus(); + + err = bpf_iter_bpf_percpu_array_map__load(skel); + if (CHECK(!skel, "bpf_iter_bpf_percpu_array_map__load", + "skeleton load failed\n")) + goto out; + + /* update map values here */ + map_fd = bpf_map__fd(skel->maps.arraymap1); + for (i = 0; i < bpf_map__max_entries(skel->maps.arraymap1); i++) { + expected_key += i; + + for (j = 0; j < bpf_num_possible_cpus(); j++) { + *(__u32 *)(val + j * 8) = i + j; + expected_val += i + j; + } + + err = bpf_map_update_elem(map_fd, &i, val, BPF_ANY); + if (CHECK(err, "map_update", "map_update failed\n")) + goto out; + } + + opts.map_fd = map_fd; + link = bpf_program__attach_iter(skel->progs.dump_bpf_percpu_array_map, &opts); + if (CHECK(IS_ERR(link), "attach_iter", "attach_iter failed\n")) + goto out; + + iter_fd = bpf_iter_create(bpf_link__fd(link)); + if (CHECK(iter_fd < 0, "create_iter", "create_iter failed\n")) + goto free_link; + + /* do some tests */ + while ((len = read(iter_fd, buf, sizeof(buf))) > 0) + ; + if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + goto close_iter; + + /* test results */ + if (CHECK(skel->bss->key_sum != expected_key, + "key_sum", "got %u expected %u\n", + skel->bss->key_sum, expected_key)) + goto close_iter; + if (CHECK(skel->bss->val_sum != expected_val, + "val_sum", "got %u expected %u\n", + skel->bss->val_sum, expected_val)) + goto close_iter; + +close_iter: + close(iter_fd); +free_link: + bpf_link__destroy(link); +out: + bpf_iter_bpf_percpu_array_map__destroy(skel); +} + void test_bpf_iter(void) { if (test__start_subtest("btf_id_or_null")) @@ -678,4 +835,8 @@ void test_bpf_iter(void) test_bpf_hash_map(); if (test__start_subtest("bpf_percpu_hash_map")) test_bpf_percpu_hash_map(); + if (test__start_subtest("bpf_array_map")) + test_bpf_array_map(); + if (test__start_subtest("bpf_percpu_array_map")) + test_bpf_percpu_array_map(); } diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_bpf_array_map.c b/tools/testing/selftests/bpf/progs/bpf_iter_bpf_array_map.c new file mode 100644 index 000000000000..6286023fd62b --- /dev/null +++ b/tools/testing/selftests/bpf/progs/bpf_iter_bpf_array_map.c @@ -0,0 +1,40 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2020 Facebook */ +#include "bpf_iter.h" +#include +#include + +char _license[] SEC("license") = "GPL"; + +struct key_t { + int a; + int b; + int c; +}; + +struct { + __uint(type, BPF_MAP_TYPE_ARRAY); + __uint(max_entries, 3); + __type(key, __u32); + __type(value, __u64); +} arraymap1 SEC(".maps"); + +__u32 key_sum = 0; +__u64 val_sum = 0; + +SEC("iter/bpf_map_elem") +int dump_bpf_array_map(struct bpf_iter__bpf_map_elem *ctx) +{ + __u32 *key = ctx->key; + __u64 *val = ctx->value; + + if (key == (void *)0 || val == (void *)0) + return 0; + + bpf_seq_write(ctx->meta->seq, key, sizeof(__u32)); + bpf_seq_write(ctx->meta->seq, val, sizeof(__u64)); + key_sum += *key; + val_sum += *val; + *val = *key; + return 0; +} diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_bpf_percpu_array_map.c b/tools/testing/selftests/bpf/progs/bpf_iter_bpf_percpu_array_map.c new file mode 100644 index 000000000000..85fa710fad90 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/bpf_iter_bpf_percpu_array_map.c @@ -0,0 +1,46 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2020 Facebook */ +#include "bpf_iter.h" +#include +#include + +char _license[] SEC("license") = "GPL"; + +struct key_t { + int a; + int b; + int c; +}; + +struct { + __uint(type, BPF_MAP_TYPE_PERCPU_ARRAY); + __uint(max_entries, 3); + __type(key, __u32); + __type(value, __u32); +} arraymap1 SEC(".maps"); + +/* will set before prog run */ +volatile const __u32 num_cpus = 0; + +__u32 key_sum = 0, val_sum = 0; + +SEC("iter/bpf_map_elem") +int dump_bpf_percpu_array_map(struct bpf_iter__bpf_map_elem *ctx) +{ + __u32 *key = ctx->key; + void *pptr = ctx->value; + __u32 step; + int i; + + if (key == (void *)0 || pptr == (void *)0) + return 0; + + key_sum += *key; + + step = 8; + for (i = 0; i < num_cpus; i++) { + val_sum += *(__u32 *)pptr; + pptr += step; + } + return 0; +} -- cgit v1.2.3 From 3b1c420bd882115eb7a3d2335cc00d7b9974eb0b Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Thu, 23 Jul 2020 11:41:22 -0700 Subject: selftests/bpf: Add a test for bpf sk_storage_map iterator Added one test for bpf sk_storage_map_iterator. $ ./test_progs -n 4 ... #4/22 bpf_sk_storage_map:OK ... Signed-off-by: Yonghong Song Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/20200723184122.591591-1-yhs@fb.com --- tools/testing/selftests/bpf/prog_tests/bpf_iter.c | 72 ++++++++++++++++++++++ .../bpf/progs/bpf_iter_bpf_sk_storage_map.c | 34 ++++++++++ 2 files changed, 106 insertions(+) create mode 100644 tools/testing/selftests/bpf/progs/bpf_iter_bpf_sk_storage_map.c (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c index 4a02b2222a6d..ffbbeb9fa268 100644 --- a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c +++ b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c @@ -19,6 +19,7 @@ #include "bpf_iter_bpf_percpu_hash_map.skel.h" #include "bpf_iter_bpf_array_map.skel.h" #include "bpf_iter_bpf_percpu_array_map.skel.h" +#include "bpf_iter_bpf_sk_storage_map.skel.h" static int duration; @@ -795,6 +796,75 @@ out: bpf_iter_bpf_percpu_array_map__destroy(skel); } +static void test_bpf_sk_storage_map(void) +{ + DECLARE_LIBBPF_OPTS(bpf_iter_attach_opts, opts); + int err, i, len, map_fd, iter_fd, num_sockets; + struct bpf_iter_bpf_sk_storage_map *skel; + int sock_fd[3] = {-1, -1, -1}; + __u32 val, expected_val = 0; + struct bpf_link *link; + char buf[64]; + + skel = bpf_iter_bpf_sk_storage_map__open_and_load(); + if (CHECK(!skel, "bpf_iter_bpf_sk_storage_map__open_and_load", + "skeleton open_and_load failed\n")) + return; + + map_fd = bpf_map__fd(skel->maps.sk_stg_map); + num_sockets = ARRAY_SIZE(sock_fd); + for (i = 0; i < num_sockets; i++) { + sock_fd[i] = socket(AF_INET6, SOCK_STREAM, 0); + if (CHECK(sock_fd[i] < 0, "socket", "errno: %d\n", errno)) + goto out; + + val = i + 1; + expected_val += val; + + err = bpf_map_update_elem(map_fd, &sock_fd[i], &val, + BPF_NOEXIST); + if (CHECK(err, "map_update", "map_update failed\n")) + goto out; + } + + opts.map_fd = map_fd; + link = bpf_program__attach_iter(skel->progs.dump_bpf_sk_storage_map, &opts); + if (CHECK(IS_ERR(link), "attach_iter", "attach_iter failed\n")) + goto out; + + iter_fd = bpf_iter_create(bpf_link__fd(link)); + if (CHECK(iter_fd < 0, "create_iter", "create_iter failed\n")) + goto free_link; + + /* do some tests */ + while ((len = read(iter_fd, buf, sizeof(buf))) > 0) + ; + if (CHECK(len < 0, "read", "read failed: %s\n", strerror(errno))) + goto close_iter; + + /* test results */ + if (CHECK(skel->bss->ipv6_sk_count != num_sockets, + "ipv6_sk_count", "got %u expected %u\n", + skel->bss->ipv6_sk_count, num_sockets)) + goto close_iter; + + if (CHECK(skel->bss->val_sum != expected_val, + "val_sum", "got %u expected %u\n", + skel->bss->val_sum, expected_val)) + goto close_iter; + +close_iter: + close(iter_fd); +free_link: + bpf_link__destroy(link); +out: + for (i = 0; i < num_sockets; i++) { + if (sock_fd[i] >= 0) + close(sock_fd[i]); + } + bpf_iter_bpf_sk_storage_map__destroy(skel); +} + void test_bpf_iter(void) { if (test__start_subtest("btf_id_or_null")) @@ -839,4 +909,6 @@ void test_bpf_iter(void) test_bpf_array_map(); if (test__start_subtest("bpf_percpu_array_map")) test_bpf_percpu_array_map(); + if (test__start_subtest("bpf_sk_storage_map")) + test_bpf_sk_storage_map(); } diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_bpf_sk_storage_map.c b/tools/testing/selftests/bpf/progs/bpf_iter_bpf_sk_storage_map.c new file mode 100644 index 000000000000..6b70ccaba301 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/bpf_iter_bpf_sk_storage_map.c @@ -0,0 +1,34 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2020 Facebook */ +#include "bpf_iter.h" +#include "bpf_tracing_net.h" +#include +#include + +char _license[] SEC("license") = "GPL"; + +struct { + __uint(type, BPF_MAP_TYPE_SK_STORAGE); + __uint(map_flags, BPF_F_NO_PREALLOC); + __type(key, int); + __type(value, int); +} sk_stg_map SEC(".maps"); + +__u32 val_sum = 0; +__u32 ipv6_sk_count = 0; + +SEC("iter/bpf_sk_storage_map") +int dump_bpf_sk_storage_map(struct bpf_iter__bpf_sk_storage_map *ctx) +{ + struct sock *sk = ctx->sk; + __u32 *val = ctx->value; + + if (sk == (void *)0 || val == (void *)0) + return 0; + + if (sk->sk_family == AF_INET6) + ipv6_sk_count++; + + val_sum += *val; + return 0; +} -- cgit v1.2.3 From 9efcc4ad7a15ea50550c53fbf62457c309216051 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Thu, 23 Jul 2020 11:41:24 -0700 Subject: selftests/bpf: Add a test for out of bound rdonly buf access If the bpf program contains out of bound access w.r.t. a particular map key/value size, the verification will be still okay, e.g., it will be accepted by verifier. But it will be rejected during link_create time. A test is added here to ensure link_create failure did happen if out of bound access happened. $ ./test_progs -n 4 ... #4/23 rdonly-buf-out-of-bound:OK ... Signed-off-by: Yonghong Song Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/20200723184124.591700-1-yhs@fb.com --- tools/testing/selftests/bpf/prog_tests/bpf_iter.c | 22 ++++++++++++++ .../selftests/bpf/progs/bpf_iter_test_kern5.c | 35 ++++++++++++++++++++++ 2 files changed, 57 insertions(+) create mode 100644 tools/testing/selftests/bpf/progs/bpf_iter_test_kern5.c (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c index ffbbeb9fa268..d95de80b1851 100644 --- a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c +++ b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c @@ -20,6 +20,7 @@ #include "bpf_iter_bpf_array_map.skel.h" #include "bpf_iter_bpf_percpu_array_map.skel.h" #include "bpf_iter_bpf_sk_storage_map.skel.h" +#include "bpf_iter_test_kern5.skel.h" static int duration; @@ -865,6 +866,25 @@ out: bpf_iter_bpf_sk_storage_map__destroy(skel); } +static void test_rdonly_buf_out_of_bound(void) +{ + DECLARE_LIBBPF_OPTS(bpf_iter_attach_opts, opts); + struct bpf_iter_test_kern5 *skel; + struct bpf_link *link; + + skel = bpf_iter_test_kern5__open_and_load(); + if (CHECK(!skel, "bpf_iter_test_kern5__open_and_load", + "skeleton open_and_load failed\n")) + return; + + opts.map_fd = bpf_map__fd(skel->maps.hashmap1); + link = bpf_program__attach_iter(skel->progs.dump_bpf_hash_map, &opts); + if (CHECK(!IS_ERR(link), "attach_iter", "unexpected success\n")) + bpf_link__destroy(link); + + bpf_iter_test_kern5__destroy(skel); +} + void test_bpf_iter(void) { if (test__start_subtest("btf_id_or_null")) @@ -911,4 +931,6 @@ void test_bpf_iter(void) test_bpf_percpu_array_map(); if (test__start_subtest("bpf_sk_storage_map")) test_bpf_sk_storage_map(); + if (test__start_subtest("rdonly-buf-out-of-bound")) + test_rdonly_buf_out_of_bound(); } diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_test_kern5.c b/tools/testing/selftests/bpf/progs/bpf_iter_test_kern5.c new file mode 100644 index 000000000000..e3a7575e81d2 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/bpf_iter_test_kern5.c @@ -0,0 +1,35 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2020 Facebook */ +#include "bpf_iter.h" +#include +#include + +char _license[] SEC("license") = "GPL"; + +struct key_t { + int a; + int b; + int c; +}; + +struct { + __uint(type, BPF_MAP_TYPE_HASH); + __uint(max_entries, 3); + __type(key, struct key_t); + __type(value, __u64); +} hashmap1 SEC(".maps"); + +__u32 key_sum = 0; + +SEC("iter/bpf_map_elem") +int dump_bpf_hash_map(struct bpf_iter__bpf_map_elem *ctx) +{ + void *key = ctx->key; + + if (key == (void *)0) + return 0; + + /* out of bound access w.r.t. hashmap1 */ + key_sum += *(__u32 *)(key + sizeof(struct key_t)); + return 0; +} -- cgit v1.2.3 From d4b4dd6ce7709c2d2fe56dcfc15074ee18505bcb Mon Sep 17 00:00:00 2001 From: Song Liu Date: Thu, 23 Jul 2020 11:06:46 -0700 Subject: libbpf: Print hint when PERF_EVENT_IOC_SET_BPF returns -EPROTO The kernel prevents potential unwinder warnings and crashes by blocking BPF program with bpf_get_[stack|stackid] on perf_event without PERF_SAMPLE_CALLCHAIN, or with exclude_callchain_[kernel|user]. Print a hint message in libbpf to help the user debug such issues. Signed-off-by: Song Liu Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/20200723180648.1429892-4-songliubraving@fb.com --- tools/lib/bpf/libbpf.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'tools') diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index a05aa7e2bab6..e51479d60285 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -7833,6 +7833,9 @@ struct bpf_link *bpf_program__attach_perf_event(struct bpf_program *prog, pr_warn("program '%s': failed to attach to pfd %d: %s\n", bpf_program__title(prog, false), pfd, libbpf_strerror_r(err, errmsg, sizeof(errmsg))); + if (err == -EPROTO) + pr_warn("program '%s': try add PERF_SAMPLE_CALLCHAIN to or remove exclude_callchain_[kernel|user] from pfd %d\n", + bpf_program__title(prog, false), pfd); return ERR_PTR(err); } if (ioctl(pfd, PERF_EVENT_IOC_ENABLE, 0) < 0) { -- cgit v1.2.3 From 1da4864c2b20f81afbb18b2a0b914d0c776331fc Mon Sep 17 00:00:00 2001 From: Song Liu Date: Thu, 23 Jul 2020 11:06:47 -0700 Subject: selftests/bpf: Add callchain_stackid This tests new helper function bpf_get_stackid_pe and bpf_get_stack_pe. These two helpers have different implementation for perf_event with PEB entries. Signed-off-by: Song Liu Signed-off-by: Alexei Starovoitov Acked-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20200723180648.1429892-5-songliubraving@fb.com --- .../selftests/bpf/prog_tests/perf_event_stackmap.c | 116 +++++++++++++++++++++ .../selftests/bpf/progs/perf_event_stackmap.c | 59 +++++++++++ 2 files changed, 175 insertions(+) create mode 100644 tools/testing/selftests/bpf/prog_tests/perf_event_stackmap.c create mode 100644 tools/testing/selftests/bpf/progs/perf_event_stackmap.c (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/perf_event_stackmap.c b/tools/testing/selftests/bpf/prog_tests/perf_event_stackmap.c new file mode 100644 index 000000000000..72c3690844fb --- /dev/null +++ b/tools/testing/selftests/bpf/prog_tests/perf_event_stackmap.c @@ -0,0 +1,116 @@ +// SPDX-License-Identifier: GPL-2.0 +// Copyright (c) 2020 Facebook +#define _GNU_SOURCE +#include +#include +#include +#include "perf_event_stackmap.skel.h" + +#ifndef noinline +#define noinline __attribute__((noinline)) +#endif + +noinline int func_1(void) +{ + static int val = 1; + + val += 1; + + usleep(100); + return val; +} + +noinline int func_2(void) +{ + return func_1(); +} + +noinline int func_3(void) +{ + return func_2(); +} + +noinline int func_4(void) +{ + return func_3(); +} + +noinline int func_5(void) +{ + return func_4(); +} + +noinline int func_6(void) +{ + int i, val = 1; + + for (i = 0; i < 100; i++) + val += func_5(); + + return val; +} + +void test_perf_event_stackmap(void) +{ + struct perf_event_attr attr = { + /* .type = PERF_TYPE_SOFTWARE, */ + .type = PERF_TYPE_HARDWARE, + .config = PERF_COUNT_HW_CPU_CYCLES, + .precise_ip = 2, + .sample_type = PERF_SAMPLE_IP | PERF_SAMPLE_BRANCH_STACK | + PERF_SAMPLE_CALLCHAIN, + .branch_sample_type = PERF_SAMPLE_BRANCH_USER | + PERF_SAMPLE_BRANCH_NO_FLAGS | + PERF_SAMPLE_BRANCH_NO_CYCLES | + PERF_SAMPLE_BRANCH_CALL_STACK, + .sample_period = 5000, + .size = sizeof(struct perf_event_attr), + }; + struct perf_event_stackmap *skel; + __u32 duration = 0; + cpu_set_t cpu_set; + int pmu_fd, err; + + skel = perf_event_stackmap__open(); + + if (CHECK(!skel, "skel_open", "skeleton open failed\n")) + return; + + err = perf_event_stackmap__load(skel); + if (CHECK(err, "skel_load", "skeleton load failed: %d\n", err)) + goto cleanup; + + CPU_ZERO(&cpu_set); + CPU_SET(0, &cpu_set); + err = pthread_setaffinity_np(pthread_self(), sizeof(cpu_set), &cpu_set); + if (CHECK(err, "set_affinity", "err %d, errno %d\n", err, errno)) + goto cleanup; + + pmu_fd = syscall(__NR_perf_event_open, &attr, -1 /* pid */, + 0 /* cpu 0 */, -1 /* group id */, + 0 /* flags */); + if (pmu_fd < 0) { + printf("%s:SKIP:cpu doesn't support the event\n", __func__); + test__skip(); + goto cleanup; + } + + skel->links.oncpu = bpf_program__attach_perf_event(skel->progs.oncpu, + pmu_fd); + if (CHECK(IS_ERR(skel->links.oncpu), "attach_perf_event", + "err %ld\n", PTR_ERR(skel->links.oncpu))) { + close(pmu_fd); + goto cleanup; + } + + /* create kernel and user stack traces for testing */ + func_6(); + + CHECK(skel->data->stackid_kernel != 2, "get_stackid_kernel", "failed\n"); + CHECK(skel->data->stackid_user != 2, "get_stackid_user", "failed\n"); + CHECK(skel->data->stack_kernel != 2, "get_stack_kernel", "failed\n"); + CHECK(skel->data->stack_user != 2, "get_stack_user", "failed\n"); + +cleanup: + perf_event_stackmap__destroy(skel); +} diff --git a/tools/testing/selftests/bpf/progs/perf_event_stackmap.c b/tools/testing/selftests/bpf/progs/perf_event_stackmap.c new file mode 100644 index 000000000000..25467d13c356 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/perf_event_stackmap.c @@ -0,0 +1,59 @@ +// SPDX-License-Identifier: GPL-2.0 +// Copyright (c) 2020 Facebook +#include "vmlinux.h" +#include + +#ifndef PERF_MAX_STACK_DEPTH +#define PERF_MAX_STACK_DEPTH 127 +#endif + +typedef __u64 stack_trace_t[PERF_MAX_STACK_DEPTH]; +struct { + __uint(type, BPF_MAP_TYPE_STACK_TRACE); + __uint(max_entries, 16384); + __uint(key_size, sizeof(__u32)); + __uint(value_size, sizeof(stack_trace_t)); +} stackmap SEC(".maps"); + +struct { + __uint(type, BPF_MAP_TYPE_PERCPU_ARRAY); + __uint(max_entries, 1); + __type(key, __u32); + __type(value, stack_trace_t); +} stackdata_map SEC(".maps"); + +long stackid_kernel = 1; +long stackid_user = 1; +long stack_kernel = 1; +long stack_user = 1; + +SEC("perf_event") +int oncpu(void *ctx) +{ + stack_trace_t *trace; + __u32 key = 0; + long val; + + val = bpf_get_stackid(ctx, &stackmap, 0); + if (val > 0) + stackid_kernel = 2; + val = bpf_get_stackid(ctx, &stackmap, BPF_F_USER_STACK); + if (val > 0) + stackid_user = 2; + + trace = bpf_map_lookup_elem(&stackdata_map, &key); + if (!trace) + return 0; + + val = bpf_get_stack(ctx, trace, sizeof(stack_trace_t), 0); + if (val > 0) + stack_kernel = 2; + + val = bpf_get_stack(ctx, trace, sizeof(stack_trace_t), BPF_F_USER_STACK); + if (val > 0) + stack_user = 2; + + return 0; +} + +char LICENSE[] SEC("license") = "GPL"; -- cgit v1.2.3 From 346938e9380cc0b2ad8e2566389cdc570386fe22 Mon Sep 17 00:00:00 2001 From: Song Liu Date: Thu, 23 Jul 2020 11:06:48 -0700 Subject: selftests/bpf: Add get_stackid_cannot_attach This test confirms that BPF program that calls bpf_get_stackid() cannot attach to perf_event with precise_ip > 0 but not PERF_SAMPLE_CALLCHAIN; and cannot attach if the perf_event has exclude_callchain_kernel. Signed-off-by: Song Liu Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/20200723180648.1429892-6-songliubraving@fb.com --- .../bpf/prog_tests/get_stackid_cannot_attach.c | 91 ++++++++++++++++++++++ 1 file changed, 91 insertions(+) create mode 100644 tools/testing/selftests/bpf/prog_tests/get_stackid_cannot_attach.c (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/get_stackid_cannot_attach.c b/tools/testing/selftests/bpf/prog_tests/get_stackid_cannot_attach.c new file mode 100644 index 000000000000..d884b2ed5bc5 --- /dev/null +++ b/tools/testing/selftests/bpf/prog_tests/get_stackid_cannot_attach.c @@ -0,0 +1,91 @@ +// SPDX-License-Identifier: GPL-2.0 +// Copyright (c) 2020 Facebook +#include +#include "test_stacktrace_build_id.skel.h" + +void test_get_stackid_cannot_attach(void) +{ + struct perf_event_attr attr = { + /* .type = PERF_TYPE_SOFTWARE, */ + .type = PERF_TYPE_HARDWARE, + .config = PERF_COUNT_HW_CPU_CYCLES, + .precise_ip = 1, + .sample_type = PERF_SAMPLE_IP | PERF_SAMPLE_BRANCH_STACK, + .branch_sample_type = PERF_SAMPLE_BRANCH_USER | + PERF_SAMPLE_BRANCH_NO_FLAGS | + PERF_SAMPLE_BRANCH_NO_CYCLES | + PERF_SAMPLE_BRANCH_CALL_STACK, + .sample_period = 5000, + .size = sizeof(struct perf_event_attr), + }; + struct test_stacktrace_build_id *skel; + __u32 duration = 0; + int pmu_fd, err; + + skel = test_stacktrace_build_id__open(); + if (CHECK(!skel, "skel_open", "skeleton open failed\n")) + return; + + /* override program type */ + bpf_program__set_perf_event(skel->progs.oncpu); + + err = test_stacktrace_build_id__load(skel); + if (CHECK(err, "skel_load", "skeleton load failed: %d\n", err)) + goto cleanup; + + pmu_fd = syscall(__NR_perf_event_open, &attr, -1 /* pid */, + 0 /* cpu 0 */, -1 /* group id */, + 0 /* flags */); + if (pmu_fd < 0 && (errno == ENOENT || errno == EOPNOTSUPP)) { + printf("%s:SKIP:cannot open PERF_COUNT_HW_CPU_CYCLES with precise_ip > 0\n", + __func__); + test__skip(); + goto cleanup; + } + if (CHECK(pmu_fd < 0, "perf_event_open", "err %d errno %d\n", + pmu_fd, errno)) + goto cleanup; + + skel->links.oncpu = bpf_program__attach_perf_event(skel->progs.oncpu, + pmu_fd); + CHECK(!IS_ERR(skel->links.oncpu), "attach_perf_event_no_callchain", + "should have failed\n"); + close(pmu_fd); + + /* add PERF_SAMPLE_CALLCHAIN, attach should succeed */ + attr.sample_type |= PERF_SAMPLE_CALLCHAIN; + + pmu_fd = syscall(__NR_perf_event_open, &attr, -1 /* pid */, + 0 /* cpu 0 */, -1 /* group id */, + 0 /* flags */); + + if (CHECK(pmu_fd < 0, "perf_event_open", "err %d errno %d\n", + pmu_fd, errno)) + goto cleanup; + + skel->links.oncpu = bpf_program__attach_perf_event(skel->progs.oncpu, + pmu_fd); + CHECK(IS_ERR(skel->links.oncpu), "attach_perf_event_callchain", + "err: %ld\n", PTR_ERR(skel->links.oncpu)); + close(pmu_fd); + + /* add exclude_callchain_kernel, attach should fail */ + attr.exclude_callchain_kernel = 1; + + pmu_fd = syscall(__NR_perf_event_open, &attr, -1 /* pid */, + 0 /* cpu 0 */, -1 /* group id */, + 0 /* flags */); + + if (CHECK(pmu_fd < 0, "perf_event_open", "err %d errno %d\n", + pmu_fd, errno)) + goto cleanup; + + skel->links.oncpu = bpf_program__attach_perf_event(skel->progs.oncpu, + pmu_fd); + CHECK(!IS_ERR(skel->links.oncpu), "attach_perf_event_exclude_callchain_kernel", + "should have failed\n"); + close(pmu_fd); + +cleanup: + test_stacktrace_build_id__destroy(skel); +} -- cgit v1.2.3 From d4a89c1eb81431479664029bcdec593dbf23385f Mon Sep 17 00:00:00 2001 From: YiFei Zhu Date: Thu, 23 Jul 2020 23:47:41 -0500 Subject: selftests/bpf: Add test for CGROUP_STORAGE map on multiple attaches This test creates a parent cgroup, and a child of that cgroup. It attaches a cgroup_skb/egress program that simply counts packets, to a global variable (ARRAY map), and to a CGROUP_STORAGE map. The program is first attached to the parent cgroup only, then to parent and child. The test cases sends a message within the child cgroup, and because the program is inherited across parent / child cgroups, it will trigger the egress program for both the parent and child, if they exist. The program, when looking up a CGROUP_STORAGE map, uses the cgroup and attach type of the attachment parameters; therefore, both attaches uses different cgroup storages. We assert that all packet counts returns what we expects. Signed-off-by: YiFei Zhu Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/5a20206afa4606144691c7caa0d1b997cd60dec0.1595565795.git.zhuyifei@google.com --- .../selftests/bpf/prog_tests/cg_storage_multi.c | 161 +++++++++++++++++++++ .../bpf/progs/cg_storage_multi_egress_only.c | 30 ++++ 2 files changed, 191 insertions(+) create mode 100644 tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c create mode 100644 tools/testing/selftests/bpf/progs/cg_storage_multi_egress_only.c (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c b/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c new file mode 100644 index 000000000000..e90e0547d759 --- /dev/null +++ b/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c @@ -0,0 +1,161 @@ +// SPDX-License-Identifier: GPL-2.0-only + +/* + * Copyright 2020 Google LLC. + */ + +#include +#include +#include + +#include "cg_storage_multi_egress_only.skel.h" + +#define PARENT_CGROUP "/cgroup_storage" +#define CHILD_CGROUP "/cgroup_storage/child" + +static int duration; + +static bool assert_storage(struct bpf_map *map, const char *cgroup_path, + __u32 expected) +{ + struct bpf_cgroup_storage_key key = {0}; + __u32 value; + int map_fd; + + map_fd = bpf_map__fd(map); + + key.cgroup_inode_id = get_cgroup_id(cgroup_path); + key.attach_type = BPF_CGROUP_INET_EGRESS; + if (CHECK(bpf_map_lookup_elem(map_fd, &key, &value) < 0, + "map-lookup", "errno %d", errno)) + return true; + if (CHECK(value != expected, + "assert-storage", "got %u expected %u", value, expected)) + return true; + + return false; +} + +static bool assert_storage_noexist(struct bpf_map *map, const char *cgroup_path) +{ + struct bpf_cgroup_storage_key key = {0}; + __u32 value; + int map_fd; + + map_fd = bpf_map__fd(map); + + key.cgroup_inode_id = get_cgroup_id(cgroup_path); + key.attach_type = BPF_CGROUP_INET_EGRESS; + if (CHECK(bpf_map_lookup_elem(map_fd, &key, &value) == 0, + "map-lookup", "succeeded, expected ENOENT")) + return true; + if (CHECK(errno != ENOENT, + "map-lookup", "errno %d, expected ENOENT", errno)) + return true; + + return false; +} + +static bool connect_send(const char *cgroup_path) +{ + bool res = true; + int server_fd = -1, client_fd = -1; + + if (join_cgroup(cgroup_path)) + goto out_clean; + + server_fd = start_server(AF_INET, SOCK_DGRAM, NULL, 0, 0); + if (server_fd < 0) + goto out_clean; + + client_fd = connect_to_fd(server_fd, 0); + if (client_fd < 0) + goto out_clean; + + if (send(client_fd, "message", strlen("message"), 0) < 0) + goto out_clean; + + res = false; + +out_clean: + close(client_fd); + close(server_fd); + return res; +} + +static void test_egress_only(int parent_cgroup_fd, int child_cgroup_fd) +{ + struct cg_storage_multi_egress_only *obj; + struct bpf_link *parent_link = NULL, *child_link = NULL; + bool err; + + obj = cg_storage_multi_egress_only__open_and_load(); + if (CHECK(!obj, "skel-load", "errno %d", errno)) + return; + + /* Attach to parent cgroup, trigger packet from child. + * Assert that there is only one run and in that run the storage is + * parent cgroup's storage. + * Also assert that child cgroup's storage does not exist + */ + parent_link = bpf_program__attach_cgroup(obj->progs.egress, + parent_cgroup_fd); + if (CHECK(IS_ERR(parent_link), "parent-cg-attach", + "err %ld", PTR_ERR(parent_link))) + goto close_bpf_object; + err = connect_send(CHILD_CGROUP); + if (CHECK(err, "first-connect-send", "errno %d", errno)) + goto close_bpf_object; + if (CHECK(obj->bss->invocations != 1, + "first-invoke", "invocations=%d", obj->bss->invocations)) + goto close_bpf_object; + if (assert_storage(obj->maps.cgroup_storage, PARENT_CGROUP, 1)) + goto close_bpf_object; + if (assert_storage_noexist(obj->maps.cgroup_storage, CHILD_CGROUP)) + goto close_bpf_object; + + /* Attach to parent and child cgroup, trigger packet from child. + * Assert that there are two additional runs, one that run with parent + * cgroup's storage and one with child cgroup's storage. + */ + child_link = bpf_program__attach_cgroup(obj->progs.egress, + child_cgroup_fd); + if (CHECK(IS_ERR(child_link), "child-cg-attach", + "err %ld", PTR_ERR(child_link))) + goto close_bpf_object; + err = connect_send(CHILD_CGROUP); + if (CHECK(err, "second-connect-send", "errno %d", errno)) + goto close_bpf_object; + if (CHECK(obj->bss->invocations != 3, + "second-invoke", "invocations=%d", obj->bss->invocations)) + goto close_bpf_object; + if (assert_storage(obj->maps.cgroup_storage, PARENT_CGROUP, 2)) + goto close_bpf_object; + if (assert_storage(obj->maps.cgroup_storage, CHILD_CGROUP, 1)) + goto close_bpf_object; + +close_bpf_object: + bpf_link__destroy(parent_link); + bpf_link__destroy(child_link); + + cg_storage_multi_egress_only__destroy(obj); +} + +void test_cg_storage_multi(void) +{ + int parent_cgroup_fd = -1, child_cgroup_fd = -1; + + parent_cgroup_fd = test__join_cgroup(PARENT_CGROUP); + if (CHECK(parent_cgroup_fd < 0, "cg-create-parent", "errno %d", errno)) + goto close_cgroup_fd; + child_cgroup_fd = create_and_get_cgroup(CHILD_CGROUP); + if (CHECK(child_cgroup_fd < 0, "cg-create-child", "errno %d", errno)) + goto close_cgroup_fd; + + if (test__start_subtest("egress_only")) + test_egress_only(parent_cgroup_fd, child_cgroup_fd); + +close_cgroup_fd: + close(child_cgroup_fd); + close(parent_cgroup_fd); +} diff --git a/tools/testing/selftests/bpf/progs/cg_storage_multi_egress_only.c b/tools/testing/selftests/bpf/progs/cg_storage_multi_egress_only.c new file mode 100644 index 000000000000..ec0165d07105 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/cg_storage_multi_egress_only.c @@ -0,0 +1,30 @@ +// SPDX-License-Identifier: GPL-2.0-only + +/* + * Copyright 2020 Google LLC. + */ + +#include +#include +#include +#include +#include + +struct { + __uint(type, BPF_MAP_TYPE_CGROUP_STORAGE); + __type(key, struct bpf_cgroup_storage_key); + __type(value, __u32); +} cgroup_storage SEC(".maps"); + +__u32 invocations = 0; + +SEC("cgroup_skb/egress") +int egress(struct __sk_buff *skb) +{ + __u32 *ptr_cg_storage = bpf_get_local_storage(&cgroup_storage, 0); + + __sync_fetch_and_add(ptr_cg_storage, 1); + __sync_fetch_and_add(&invocations, 1); + + return 1; +} -- cgit v1.2.3 From 9e5bd1f7633bc1c3c8b25496eedfeced6d2675ff Mon Sep 17 00:00:00 2001 From: YiFei Zhu Date: Thu, 23 Jul 2020 23:47:42 -0500 Subject: selftests/bpf: Test CGROUP_STORAGE map can't be used by multiple progs The current assumption is that the lifetime of a cgroup storage is tied to the program's attachment. The storage is created in cgroup_bpf_attach, and released upon cgroup_bpf_detach and cgroup_bpf_release. Because the current semantics is that each attachment gets a completely independent cgroup storage, and you can have multiple programs attached to the same (cgroup, attach type) pair, the key of the CGROUP_STORAGE map, looking up the map with this pair could yield multiple storages, and that is not permitted. Therefore, the kernel verifier checks that two programs cannot share the same CGROUP_STORAGE map, even if they have different expected attach types, considering that the actual attach type does not always have to be equal to the expected attach type. The test creates a CGROUP_STORAGE map and make it shared across two different programs, one cgroup_skb/egress and one /ingress. It asserts that the two programs cannot be both loaded, due to verifier failure from the above reason. Signed-off-by: YiFei Zhu Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/30a6b0da67ae6b0296c4d511bfb19c5f3d035916.1595565795.git.zhuyifei@google.com --- .../selftests/bpf/prog_tests/cg_storage_multi.c | 43 +++++++++++++++++---- .../testing/selftests/bpf/progs/cg_storage_multi.h | 13 +++++++ .../bpf/progs/cg_storage_multi_egress_ingress.c | 45 ++++++++++++++++++++++ .../bpf/progs/cg_storage_multi_egress_only.c | 9 +++-- 4 files changed, 99 insertions(+), 11 deletions(-) create mode 100644 tools/testing/selftests/bpf/progs/cg_storage_multi.h create mode 100644 tools/testing/selftests/bpf/progs/cg_storage_multi_egress_ingress.c (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c b/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c index e90e0547d759..1c7653423698 100644 --- a/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c +++ b/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c @@ -8,7 +8,10 @@ #include #include +#include "progs/cg_storage_multi.h" + #include "cg_storage_multi_egress_only.skel.h" +#include "cg_storage_multi_egress_ingress.skel.h" #define PARENT_CGROUP "/cgroup_storage" #define CHILD_CGROUP "/cgroup_storage/child" @@ -16,10 +19,10 @@ static int duration; static bool assert_storage(struct bpf_map *map, const char *cgroup_path, - __u32 expected) + struct cgroup_value *expected) { struct bpf_cgroup_storage_key key = {0}; - __u32 value; + struct cgroup_value value; int map_fd; map_fd = bpf_map__fd(map); @@ -29,8 +32,8 @@ static bool assert_storage(struct bpf_map *map, const char *cgroup_path, if (CHECK(bpf_map_lookup_elem(map_fd, &key, &value) < 0, "map-lookup", "errno %d", errno)) return true; - if (CHECK(value != expected, - "assert-storage", "got %u expected %u", value, expected)) + if (CHECK(memcmp(&value, expected, sizeof(struct cgroup_value)), + "assert-storage", "storages differ")) return true; return false; @@ -39,7 +42,7 @@ static bool assert_storage(struct bpf_map *map, const char *cgroup_path, static bool assert_storage_noexist(struct bpf_map *map, const char *cgroup_path) { struct bpf_cgroup_storage_key key = {0}; - __u32 value; + struct cgroup_value value; int map_fd; map_fd = bpf_map__fd(map); @@ -86,6 +89,7 @@ out_clean: static void test_egress_only(int parent_cgroup_fd, int child_cgroup_fd) { struct cg_storage_multi_egress_only *obj; + struct cgroup_value expected_cgroup_value; struct bpf_link *parent_link = NULL, *child_link = NULL; bool err; @@ -109,7 +113,9 @@ static void test_egress_only(int parent_cgroup_fd, int child_cgroup_fd) if (CHECK(obj->bss->invocations != 1, "first-invoke", "invocations=%d", obj->bss->invocations)) goto close_bpf_object; - if (assert_storage(obj->maps.cgroup_storage, PARENT_CGROUP, 1)) + expected_cgroup_value = (struct cgroup_value) { .egress_pkts = 1 }; + if (assert_storage(obj->maps.cgroup_storage, + PARENT_CGROUP, &expected_cgroup_value)) goto close_bpf_object; if (assert_storage_noexist(obj->maps.cgroup_storage, CHILD_CGROUP)) goto close_bpf_object; @@ -129,9 +135,13 @@ static void test_egress_only(int parent_cgroup_fd, int child_cgroup_fd) if (CHECK(obj->bss->invocations != 3, "second-invoke", "invocations=%d", obj->bss->invocations)) goto close_bpf_object; - if (assert_storage(obj->maps.cgroup_storage, PARENT_CGROUP, 2)) + expected_cgroup_value = (struct cgroup_value) { .egress_pkts = 2 }; + if (assert_storage(obj->maps.cgroup_storage, + PARENT_CGROUP, &expected_cgroup_value)) goto close_bpf_object; - if (assert_storage(obj->maps.cgroup_storage, CHILD_CGROUP, 1)) + expected_cgroup_value = (struct cgroup_value) { .egress_pkts = 1 }; + if (assert_storage(obj->maps.cgroup_storage, + CHILD_CGROUP, &expected_cgroup_value)) goto close_bpf_object; close_bpf_object: @@ -141,6 +151,20 @@ close_bpf_object: cg_storage_multi_egress_only__destroy(obj); } +static void test_egress_ingress(int parent_cgroup_fd, int child_cgroup_fd) +{ + struct cg_storage_multi_egress_ingress *obj; + + /* Cannot load both programs due to verifier failure: + * "only one cgroup storage of each type is allowed" + */ + obj = cg_storage_multi_egress_ingress__open_and_load(); + CHECK(obj || errno != EBUSY, + "skel-load", "errno %d, expected EBUSY", errno); + + cg_storage_multi_egress_ingress__destroy(obj); +} + void test_cg_storage_multi(void) { int parent_cgroup_fd = -1, child_cgroup_fd = -1; @@ -155,6 +179,9 @@ void test_cg_storage_multi(void) if (test__start_subtest("egress_only")) test_egress_only(parent_cgroup_fd, child_cgroup_fd); + if (test__start_subtest("egress_ingress")) + test_egress_ingress(parent_cgroup_fd, child_cgroup_fd); + close_cgroup_fd: close(child_cgroup_fd); close(parent_cgroup_fd); diff --git a/tools/testing/selftests/bpf/progs/cg_storage_multi.h b/tools/testing/selftests/bpf/progs/cg_storage_multi.h new file mode 100644 index 000000000000..a0778fe7857a --- /dev/null +++ b/tools/testing/selftests/bpf/progs/cg_storage_multi.h @@ -0,0 +1,13 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ + +#ifndef __PROGS_CG_STORAGE_MULTI_H +#define __PROGS_CG_STORAGE_MULTI_H + +#include + +struct cgroup_value { + __u32 egress_pkts; + __u32 ingress_pkts; +}; + +#endif diff --git a/tools/testing/selftests/bpf/progs/cg_storage_multi_egress_ingress.c b/tools/testing/selftests/bpf/progs/cg_storage_multi_egress_ingress.c new file mode 100644 index 000000000000..9ce386899365 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/cg_storage_multi_egress_ingress.c @@ -0,0 +1,45 @@ +// SPDX-License-Identifier: GPL-2.0-only + +/* + * Copyright 2020 Google LLC. + */ + +#include +#include +#include +#include +#include + +#include "progs/cg_storage_multi.h" + +struct { + __uint(type, BPF_MAP_TYPE_CGROUP_STORAGE); + __type(key, struct bpf_cgroup_storage_key); + __type(value, struct cgroup_value); +} cgroup_storage SEC(".maps"); + +__u32 invocations = 0; + +SEC("cgroup_skb/egress") +int egress(struct __sk_buff *skb) +{ + struct cgroup_value *ptr_cg_storage = + bpf_get_local_storage(&cgroup_storage, 0); + + __sync_fetch_and_add(&ptr_cg_storage->egress_pkts, 1); + __sync_fetch_and_add(&invocations, 1); + + return 1; +} + +SEC("cgroup_skb/ingress") +int ingress(struct __sk_buff *skb) +{ + struct cgroup_value *ptr_cg_storage = + bpf_get_local_storage(&cgroup_storage, 0); + + __sync_fetch_and_add(&ptr_cg_storage->ingress_pkts, 1); + __sync_fetch_and_add(&invocations, 1); + + return 1; +} diff --git a/tools/testing/selftests/bpf/progs/cg_storage_multi_egress_only.c b/tools/testing/selftests/bpf/progs/cg_storage_multi_egress_only.c index ec0165d07105..44ad46b33539 100644 --- a/tools/testing/selftests/bpf/progs/cg_storage_multi_egress_only.c +++ b/tools/testing/selftests/bpf/progs/cg_storage_multi_egress_only.c @@ -10,10 +10,12 @@ #include #include +#include "progs/cg_storage_multi.h" + struct { __uint(type, BPF_MAP_TYPE_CGROUP_STORAGE); __type(key, struct bpf_cgroup_storage_key); - __type(value, __u32); + __type(value, struct cgroup_value); } cgroup_storage SEC(".maps"); __u32 invocations = 0; @@ -21,9 +23,10 @@ __u32 invocations = 0; SEC("cgroup_skb/egress") int egress(struct __sk_buff *skb) { - __u32 *ptr_cg_storage = bpf_get_local_storage(&cgroup_storage, 0); + struct cgroup_value *ptr_cg_storage = + bpf_get_local_storage(&cgroup_storage, 0); - __sync_fetch_and_add(ptr_cg_storage, 1); + __sync_fetch_and_add(&ptr_cg_storage->egress_pkts, 1); __sync_fetch_and_add(&invocations, 1); return 1; -- cgit v1.2.3 From 3573f384014f51fd5289df0e8369b63ae7fdc244 Mon Sep 17 00:00:00 2001 From: YiFei Zhu Date: Thu, 23 Jul 2020 23:47:44 -0500 Subject: selftests/bpf: Test CGROUP_STORAGE behavior on shared egress + ingress This mirrors the original egress-only test. The cgroup_storage is now extended to have two packet counters, one for egress and one for ingress. We also extend to have two egress programs to test that egress will always share with other egress origrams in the same cgroup. The behavior of the counters are exactly the same as the original egress-only test. The test is split into two, one "isolated" test that when the key type is struct bpf_cgroup_storage_key, which contains the attach type, programs of different attach types will see different storages. The other, "shared" test that when the key type is u64, programs of different attach types will see the same storage if they are attached to the same cgroup. Signed-off-by: YiFei Zhu Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/c756f5f1521227b8e6e90a453299dda722d7324d.1595565795.git.zhuyifei@google.com --- .../selftests/bpf/prog_tests/cg_storage_multi.c | 265 +++++++++++++++++++-- .../bpf/progs/cg_storage_multi_egress_ingress.c | 45 ---- .../bpf/progs/cg_storage_multi_isolated.c | 57 +++++ .../selftests/bpf/progs/cg_storage_multi_shared.c | 57 +++++ 4 files changed, 354 insertions(+), 70 deletions(-) delete mode 100644 tools/testing/selftests/bpf/progs/cg_storage_multi_egress_ingress.c create mode 100644 tools/testing/selftests/bpf/progs/cg_storage_multi_isolated.c create mode 100644 tools/testing/selftests/bpf/progs/cg_storage_multi_shared.c (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c b/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c index 1c7653423698..c67d8c076a34 100644 --- a/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c +++ b/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c @@ -11,25 +11,23 @@ #include "progs/cg_storage_multi.h" #include "cg_storage_multi_egress_only.skel.h" -#include "cg_storage_multi_egress_ingress.skel.h" +#include "cg_storage_multi_isolated.skel.h" +#include "cg_storage_multi_shared.skel.h" #define PARENT_CGROUP "/cgroup_storage" #define CHILD_CGROUP "/cgroup_storage/child" static int duration; -static bool assert_storage(struct bpf_map *map, const char *cgroup_path, +static bool assert_storage(struct bpf_map *map, const void *key, struct cgroup_value *expected) { - struct bpf_cgroup_storage_key key = {0}; struct cgroup_value value; int map_fd; map_fd = bpf_map__fd(map); - key.cgroup_inode_id = get_cgroup_id(cgroup_path); - key.attach_type = BPF_CGROUP_INET_EGRESS; - if (CHECK(bpf_map_lookup_elem(map_fd, &key, &value) < 0, + if (CHECK(bpf_map_lookup_elem(map_fd, key, &value) < 0, "map-lookup", "errno %d", errno)) return true; if (CHECK(memcmp(&value, expected, sizeof(struct cgroup_value)), @@ -39,17 +37,14 @@ static bool assert_storage(struct bpf_map *map, const char *cgroup_path, return false; } -static bool assert_storage_noexist(struct bpf_map *map, const char *cgroup_path) +static bool assert_storage_noexist(struct bpf_map *map, const void *key) { - struct bpf_cgroup_storage_key key = {0}; struct cgroup_value value; int map_fd; map_fd = bpf_map__fd(map); - key.cgroup_inode_id = get_cgroup_id(cgroup_path); - key.attach_type = BPF_CGROUP_INET_EGRESS; - if (CHECK(bpf_map_lookup_elem(map_fd, &key, &value) == 0, + if (CHECK(bpf_map_lookup_elem(map_fd, key, &value) == 0, "map-lookup", "succeeded, expected ENOENT")) return true; if (CHECK(errno != ENOENT, @@ -90,9 +85,12 @@ static void test_egress_only(int parent_cgroup_fd, int child_cgroup_fd) { struct cg_storage_multi_egress_only *obj; struct cgroup_value expected_cgroup_value; + struct bpf_cgroup_storage_key key; struct bpf_link *parent_link = NULL, *child_link = NULL; bool err; + key.attach_type = BPF_CGROUP_INET_EGRESS; + obj = cg_storage_multi_egress_only__open_and_load(); if (CHECK(!obj, "skel-load", "errno %d", errno)) return; @@ -113,11 +111,13 @@ static void test_egress_only(int parent_cgroup_fd, int child_cgroup_fd) if (CHECK(obj->bss->invocations != 1, "first-invoke", "invocations=%d", obj->bss->invocations)) goto close_bpf_object; + key.cgroup_inode_id = get_cgroup_id(PARENT_CGROUP); expected_cgroup_value = (struct cgroup_value) { .egress_pkts = 1 }; if (assert_storage(obj->maps.cgroup_storage, - PARENT_CGROUP, &expected_cgroup_value)) + &key, &expected_cgroup_value)) goto close_bpf_object; - if (assert_storage_noexist(obj->maps.cgroup_storage, CHILD_CGROUP)) + key.cgroup_inode_id = get_cgroup_id(CHILD_CGROUP); + if (assert_storage_noexist(obj->maps.cgroup_storage, &key)) goto close_bpf_object; /* Attach to parent and child cgroup, trigger packet from child. @@ -135,13 +135,15 @@ static void test_egress_only(int parent_cgroup_fd, int child_cgroup_fd) if (CHECK(obj->bss->invocations != 3, "second-invoke", "invocations=%d", obj->bss->invocations)) goto close_bpf_object; + key.cgroup_inode_id = get_cgroup_id(PARENT_CGROUP); expected_cgroup_value = (struct cgroup_value) { .egress_pkts = 2 }; if (assert_storage(obj->maps.cgroup_storage, - PARENT_CGROUP, &expected_cgroup_value)) + &key, &expected_cgroup_value)) goto close_bpf_object; + key.cgroup_inode_id = get_cgroup_id(CHILD_CGROUP); expected_cgroup_value = (struct cgroup_value) { .egress_pkts = 1 }; if (assert_storage(obj->maps.cgroup_storage, - CHILD_CGROUP, &expected_cgroup_value)) + &key, &expected_cgroup_value)) goto close_bpf_object; close_bpf_object: @@ -151,18 +153,228 @@ close_bpf_object: cg_storage_multi_egress_only__destroy(obj); } -static void test_egress_ingress(int parent_cgroup_fd, int child_cgroup_fd) +static void test_isolated(int parent_cgroup_fd, int child_cgroup_fd) { - struct cg_storage_multi_egress_ingress *obj; + struct cg_storage_multi_isolated *obj; + struct cgroup_value expected_cgroup_value; + struct bpf_cgroup_storage_key key; + struct bpf_link *parent_egress1_link = NULL, *parent_egress2_link = NULL; + struct bpf_link *child_egress1_link = NULL, *child_egress2_link = NULL; + struct bpf_link *parent_ingress_link = NULL, *child_ingress_link = NULL; + bool err; - /* Cannot load both programs due to verifier failure: - * "only one cgroup storage of each type is allowed" + obj = cg_storage_multi_isolated__open_and_load(); + if (CHECK(!obj, "skel-load", "errno %d", errno)) + return; + + /* Attach to parent cgroup, trigger packet from child. + * Assert that there is three runs, two with parent cgroup egress and + * one with parent cgroup ingress, stored in separate parent storages. + * Also assert that child cgroup's storages does not exist */ - obj = cg_storage_multi_egress_ingress__open_and_load(); - CHECK(obj || errno != EBUSY, - "skel-load", "errno %d, expected EBUSY", errno); + parent_egress1_link = bpf_program__attach_cgroup(obj->progs.egress1, + parent_cgroup_fd); + if (CHECK(IS_ERR(parent_egress1_link), "parent-egress1-cg-attach", + "err %ld", PTR_ERR(parent_egress1_link))) + goto close_bpf_object; + parent_egress2_link = bpf_program__attach_cgroup(obj->progs.egress2, + parent_cgroup_fd); + if (CHECK(IS_ERR(parent_egress2_link), "parent-egress2-cg-attach", + "err %ld", PTR_ERR(parent_egress2_link))) + goto close_bpf_object; + parent_ingress_link = bpf_program__attach_cgroup(obj->progs.ingress, + parent_cgroup_fd); + if (CHECK(IS_ERR(parent_ingress_link), "parent-ingress-cg-attach", + "err %ld", PTR_ERR(parent_ingress_link))) + goto close_bpf_object; + err = connect_send(CHILD_CGROUP); + if (CHECK(err, "first-connect-send", "errno %d", errno)) + goto close_bpf_object; + if (CHECK(obj->bss->invocations != 3, + "first-invoke", "invocations=%d", obj->bss->invocations)) + goto close_bpf_object; + key.cgroup_inode_id = get_cgroup_id(PARENT_CGROUP); + key.attach_type = BPF_CGROUP_INET_EGRESS; + expected_cgroup_value = (struct cgroup_value) { .egress_pkts = 2 }; + if (assert_storage(obj->maps.cgroup_storage, + &key, &expected_cgroup_value)) + goto close_bpf_object; + key.attach_type = BPF_CGROUP_INET_INGRESS; + expected_cgroup_value = (struct cgroup_value) { .ingress_pkts = 1 }; + if (assert_storage(obj->maps.cgroup_storage, + &key, &expected_cgroup_value)) + goto close_bpf_object; + key.cgroup_inode_id = get_cgroup_id(CHILD_CGROUP); + key.attach_type = BPF_CGROUP_INET_EGRESS; + if (assert_storage_noexist(obj->maps.cgroup_storage, &key)) + goto close_bpf_object; + key.attach_type = BPF_CGROUP_INET_INGRESS; + if (assert_storage_noexist(obj->maps.cgroup_storage, &key)) + goto close_bpf_object; - cg_storage_multi_egress_ingress__destroy(obj); + /* Attach to parent and child cgroup, trigger packet from child. + * Assert that there is six additional runs, parent cgroup egresses and + * ingress, child cgroup egresses and ingress. + * Assert that egree and ingress storages are separate. + */ + child_egress1_link = bpf_program__attach_cgroup(obj->progs.egress1, + child_cgroup_fd); + if (CHECK(IS_ERR(child_egress1_link), "child-egress1-cg-attach", + "err %ld", PTR_ERR(child_egress1_link))) + goto close_bpf_object; + child_egress2_link = bpf_program__attach_cgroup(obj->progs.egress2, + child_cgroup_fd); + if (CHECK(IS_ERR(child_egress2_link), "child-egress2-cg-attach", + "err %ld", PTR_ERR(child_egress2_link))) + goto close_bpf_object; + child_ingress_link = bpf_program__attach_cgroup(obj->progs.ingress, + child_cgroup_fd); + if (CHECK(IS_ERR(child_ingress_link), "child-ingress-cg-attach", + "err %ld", PTR_ERR(child_ingress_link))) + goto close_bpf_object; + err = connect_send(CHILD_CGROUP); + if (CHECK(err, "second-connect-send", "errno %d", errno)) + goto close_bpf_object; + if (CHECK(obj->bss->invocations != 9, + "second-invoke", "invocations=%d", obj->bss->invocations)) + goto close_bpf_object; + key.cgroup_inode_id = get_cgroup_id(PARENT_CGROUP); + key.attach_type = BPF_CGROUP_INET_EGRESS; + expected_cgroup_value = (struct cgroup_value) { .egress_pkts = 4 }; + if (assert_storage(obj->maps.cgroup_storage, + &key, &expected_cgroup_value)) + goto close_bpf_object; + key.attach_type = BPF_CGROUP_INET_INGRESS; + expected_cgroup_value = (struct cgroup_value) { .ingress_pkts = 2 }; + if (assert_storage(obj->maps.cgroup_storage, + &key, &expected_cgroup_value)) + goto close_bpf_object; + key.cgroup_inode_id = get_cgroup_id(CHILD_CGROUP); + key.attach_type = BPF_CGROUP_INET_EGRESS; + expected_cgroup_value = (struct cgroup_value) { .egress_pkts = 2 }; + if (assert_storage(obj->maps.cgroup_storage, + &key, &expected_cgroup_value)) + goto close_bpf_object; + key.attach_type = BPF_CGROUP_INET_INGRESS; + expected_cgroup_value = (struct cgroup_value) { .ingress_pkts = 1 }; + if (assert_storage(obj->maps.cgroup_storage, + &key, &expected_cgroup_value)) + goto close_bpf_object; + +close_bpf_object: + bpf_link__destroy(parent_egress1_link); + bpf_link__destroy(parent_egress2_link); + bpf_link__destroy(parent_ingress_link); + bpf_link__destroy(child_egress1_link); + bpf_link__destroy(child_egress2_link); + bpf_link__destroy(child_ingress_link); + + cg_storage_multi_isolated__destroy(obj); +} + +static void test_shared(int parent_cgroup_fd, int child_cgroup_fd) +{ + struct cg_storage_multi_shared *obj; + struct cgroup_value expected_cgroup_value; + __u64 key; + struct bpf_link *parent_egress1_link = NULL, *parent_egress2_link = NULL; + struct bpf_link *child_egress1_link = NULL, *child_egress2_link = NULL; + struct bpf_link *parent_ingress_link = NULL, *child_ingress_link = NULL; + bool err; + + obj = cg_storage_multi_shared__open_and_load(); + if (CHECK(!obj, "skel-load", "errno %d", errno)) + return; + + /* Attach to parent cgroup, trigger packet from child. + * Assert that there is three runs, two with parent cgroup egress and + * one with parent cgroup ingress. + * Also assert that child cgroup's storage does not exist + */ + parent_egress1_link = bpf_program__attach_cgroup(obj->progs.egress1, + parent_cgroup_fd); + if (CHECK(IS_ERR(parent_egress1_link), "parent-egress1-cg-attach", + "err %ld", PTR_ERR(parent_egress1_link))) + goto close_bpf_object; + parent_egress2_link = bpf_program__attach_cgroup(obj->progs.egress2, + parent_cgroup_fd); + if (CHECK(IS_ERR(parent_egress2_link), "parent-egress2-cg-attach", + "err %ld", PTR_ERR(parent_egress2_link))) + goto close_bpf_object; + parent_ingress_link = bpf_program__attach_cgroup(obj->progs.ingress, + parent_cgroup_fd); + if (CHECK(IS_ERR(parent_ingress_link), "parent-ingress-cg-attach", + "err %ld", PTR_ERR(parent_ingress_link))) + goto close_bpf_object; + err = connect_send(CHILD_CGROUP); + if (CHECK(err, "first-connect-send", "errno %d", errno)) + goto close_bpf_object; + if (CHECK(obj->bss->invocations != 3, + "first-invoke", "invocations=%d", obj->bss->invocations)) + goto close_bpf_object; + key = get_cgroup_id(PARENT_CGROUP); + expected_cgroup_value = (struct cgroup_value) { + .egress_pkts = 2, + .ingress_pkts = 1, + }; + if (assert_storage(obj->maps.cgroup_storage, + &key, &expected_cgroup_value)) + goto close_bpf_object; + key = get_cgroup_id(CHILD_CGROUP); + if (assert_storage_noexist(obj->maps.cgroup_storage, &key)) + goto close_bpf_object; + + /* Attach to parent and child cgroup, trigger packet from child. + * Assert that there is six additional runs, parent cgroup egresses and + * ingress, child cgroup egresses and ingress. + */ + child_egress1_link = bpf_program__attach_cgroup(obj->progs.egress1, + child_cgroup_fd); + if (CHECK(IS_ERR(child_egress1_link), "child-egress1-cg-attach", + "err %ld", PTR_ERR(child_egress1_link))) + goto close_bpf_object; + child_egress2_link = bpf_program__attach_cgroup(obj->progs.egress2, + child_cgroup_fd); + if (CHECK(IS_ERR(child_egress2_link), "child-egress2-cg-attach", + "err %ld", PTR_ERR(child_egress2_link))) + goto close_bpf_object; + child_ingress_link = bpf_program__attach_cgroup(obj->progs.ingress, + child_cgroup_fd); + if (CHECK(IS_ERR(child_ingress_link), "child-ingress-cg-attach", + "err %ld", PTR_ERR(child_ingress_link))) + goto close_bpf_object; + err = connect_send(CHILD_CGROUP); + if (CHECK(err, "second-connect-send", "errno %d", errno)) + goto close_bpf_object; + if (CHECK(obj->bss->invocations != 9, + "second-invoke", "invocations=%d", obj->bss->invocations)) + goto close_bpf_object; + key = get_cgroup_id(PARENT_CGROUP); + expected_cgroup_value = (struct cgroup_value) { + .egress_pkts = 4, + .ingress_pkts = 2, + }; + if (assert_storage(obj->maps.cgroup_storage, + &key, &expected_cgroup_value)) + goto close_bpf_object; + key = get_cgroup_id(CHILD_CGROUP); + expected_cgroup_value = (struct cgroup_value) { + .egress_pkts = 2, + .ingress_pkts = 1, + }; + if (assert_storage(obj->maps.cgroup_storage, + &key, &expected_cgroup_value)) + goto close_bpf_object; + +close_bpf_object: + bpf_link__destroy(parent_egress1_link); + bpf_link__destroy(parent_egress2_link); + bpf_link__destroy(parent_ingress_link); + bpf_link__destroy(child_egress1_link); + bpf_link__destroy(child_egress2_link); + bpf_link__destroy(child_ingress_link); + + cg_storage_multi_shared__destroy(obj); } void test_cg_storage_multi(void) @@ -179,8 +391,11 @@ void test_cg_storage_multi(void) if (test__start_subtest("egress_only")) test_egress_only(parent_cgroup_fd, child_cgroup_fd); - if (test__start_subtest("egress_ingress")) - test_egress_ingress(parent_cgroup_fd, child_cgroup_fd); + if (test__start_subtest("isolated")) + test_isolated(parent_cgroup_fd, child_cgroup_fd); + + if (test__start_subtest("shared")) + test_shared(parent_cgroup_fd, child_cgroup_fd); close_cgroup_fd: close(child_cgroup_fd); diff --git a/tools/testing/selftests/bpf/progs/cg_storage_multi_egress_ingress.c b/tools/testing/selftests/bpf/progs/cg_storage_multi_egress_ingress.c deleted file mode 100644 index 9ce386899365..000000000000 --- a/tools/testing/selftests/bpf/progs/cg_storage_multi_egress_ingress.c +++ /dev/null @@ -1,45 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only - -/* - * Copyright 2020 Google LLC. - */ - -#include -#include -#include -#include -#include - -#include "progs/cg_storage_multi.h" - -struct { - __uint(type, BPF_MAP_TYPE_CGROUP_STORAGE); - __type(key, struct bpf_cgroup_storage_key); - __type(value, struct cgroup_value); -} cgroup_storage SEC(".maps"); - -__u32 invocations = 0; - -SEC("cgroup_skb/egress") -int egress(struct __sk_buff *skb) -{ - struct cgroup_value *ptr_cg_storage = - bpf_get_local_storage(&cgroup_storage, 0); - - __sync_fetch_and_add(&ptr_cg_storage->egress_pkts, 1); - __sync_fetch_and_add(&invocations, 1); - - return 1; -} - -SEC("cgroup_skb/ingress") -int ingress(struct __sk_buff *skb) -{ - struct cgroup_value *ptr_cg_storage = - bpf_get_local_storage(&cgroup_storage, 0); - - __sync_fetch_and_add(&ptr_cg_storage->ingress_pkts, 1); - __sync_fetch_and_add(&invocations, 1); - - return 1; -} diff --git a/tools/testing/selftests/bpf/progs/cg_storage_multi_isolated.c b/tools/testing/selftests/bpf/progs/cg_storage_multi_isolated.c new file mode 100644 index 000000000000..a25373002055 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/cg_storage_multi_isolated.c @@ -0,0 +1,57 @@ +// SPDX-License-Identifier: GPL-2.0-only + +/* + * Copyright 2020 Google LLC. + */ + +#include +#include +#include +#include +#include + +#include "progs/cg_storage_multi.h" + +struct { + __uint(type, BPF_MAP_TYPE_CGROUP_STORAGE); + __type(key, struct bpf_cgroup_storage_key); + __type(value, struct cgroup_value); +} cgroup_storage SEC(".maps"); + +__u32 invocations = 0; + +SEC("cgroup_skb/egress/1") +int egress1(struct __sk_buff *skb) +{ + struct cgroup_value *ptr_cg_storage = + bpf_get_local_storage(&cgroup_storage, 0); + + __sync_fetch_and_add(&ptr_cg_storage->egress_pkts, 1); + __sync_fetch_and_add(&invocations, 1); + + return 1; +} + +SEC("cgroup_skb/egress/2") +int egress2(struct __sk_buff *skb) +{ + struct cgroup_value *ptr_cg_storage = + bpf_get_local_storage(&cgroup_storage, 0); + + __sync_fetch_and_add(&ptr_cg_storage->egress_pkts, 1); + __sync_fetch_and_add(&invocations, 1); + + return 1; +} + +SEC("cgroup_skb/ingress") +int ingress(struct __sk_buff *skb) +{ + struct cgroup_value *ptr_cg_storage = + bpf_get_local_storage(&cgroup_storage, 0); + + __sync_fetch_and_add(&ptr_cg_storage->ingress_pkts, 1); + __sync_fetch_and_add(&invocations, 1); + + return 1; +} diff --git a/tools/testing/selftests/bpf/progs/cg_storage_multi_shared.c b/tools/testing/selftests/bpf/progs/cg_storage_multi_shared.c new file mode 100644 index 000000000000..a149f33bc533 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/cg_storage_multi_shared.c @@ -0,0 +1,57 @@ +// SPDX-License-Identifier: GPL-2.0-only + +/* + * Copyright 2020 Google LLC. + */ + +#include +#include +#include +#include +#include + +#include "progs/cg_storage_multi.h" + +struct { + __uint(type, BPF_MAP_TYPE_CGROUP_STORAGE); + __type(key, __u64); + __type(value, struct cgroup_value); +} cgroup_storage SEC(".maps"); + +__u32 invocations = 0; + +SEC("cgroup_skb/egress/1") +int egress1(struct __sk_buff *skb) +{ + struct cgroup_value *ptr_cg_storage = + bpf_get_local_storage(&cgroup_storage, 0); + + __sync_fetch_and_add(&ptr_cg_storage->egress_pkts, 1); + __sync_fetch_and_add(&invocations, 1); + + return 1; +} + +SEC("cgroup_skb/egress/2") +int egress2(struct __sk_buff *skb) +{ + struct cgroup_value *ptr_cg_storage = + bpf_get_local_storage(&cgroup_storage, 0); + + __sync_fetch_and_add(&ptr_cg_storage->egress_pkts, 1); + __sync_fetch_and_add(&invocations, 1); + + return 1; +} + +SEC("cgroup_skb/ingress") +int ingress(struct __sk_buff *skb) +{ + struct cgroup_value *ptr_cg_storage = + bpf_get_local_storage(&cgroup_storage, 0); + + __sync_fetch_and_add(&ptr_cg_storage->ingress_pkts, 1); + __sync_fetch_and_add(&invocations, 1); + + return 1; +} -- cgit v1.2.3 From dc8698cac7aada9b61a612cb819341d84591163e Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Tue, 21 Jul 2020 23:46:00 -0700 Subject: libbpf: Add support for BPF XDP link Sync UAPI header and add support for using bpf_link-based XDP attachment. Make xdp/ prog type set expected attach type. Kernel didn't enforce attach_type for XDP programs before, so there is no backwards compatiblity issues there. Also fix section_names selftest to recognize that xdp prog types now have expected attach type. Signed-off-by: Andrii Nakryiko Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/20200722064603.3350758-8-andriin@fb.com --- tools/include/uapi/linux/bpf.h | 10 +++++++++- tools/lib/bpf/libbpf.c | 9 ++++++++- tools/lib/bpf/libbpf.h | 2 ++ tools/lib/bpf/libbpf.map | 1 + tools/testing/selftests/bpf/prog_tests/section_names.c | 2 +- 5 files changed, 21 insertions(+), 3 deletions(-) (limited to 'tools') diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index 828c2f6438f2..e1ba4ae6a916 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -230,6 +230,7 @@ enum bpf_attach_type { BPF_CGROUP_INET_SOCK_RELEASE, BPF_XDP_CPUMAP, BPF_SK_LOOKUP, + BPF_XDP, __MAX_BPF_ATTACH_TYPE }; @@ -242,6 +243,7 @@ enum bpf_link_type { BPF_LINK_TYPE_CGROUP = 3, BPF_LINK_TYPE_ITER = 4, BPF_LINK_TYPE_NETNS = 5, + BPF_LINK_TYPE_XDP = 6, MAX_BPF_LINK_TYPE, }; @@ -614,7 +616,10 @@ union bpf_attr { struct { /* struct used by BPF_LINK_CREATE command */ __u32 prog_fd; /* eBPF program to attach */ - __u32 target_fd; /* object to attach to */ + union { + __u32 target_fd; /* object to attach to */ + __u32 target_ifindex; /* target ifindex */ + }; __u32 attach_type; /* attach type */ __u32 flags; /* extra flags */ } link_create; @@ -4064,6 +4069,9 @@ struct bpf_link_info { __u32 netns_ino; __u32 attach_type; } netns; + struct { + __u32 ifindex; + } xdp; }; } __attribute__((aligned(8))); diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index e51479d60285..54830d603fee 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -6915,7 +6915,8 @@ static const struct bpf_sec_def section_defs[] = { BPF_XDP_DEVMAP), BPF_EAPROG_SEC("xdp_cpumap/", BPF_PROG_TYPE_XDP, BPF_XDP_CPUMAP), - BPF_PROG_SEC("xdp", BPF_PROG_TYPE_XDP), + BPF_EAPROG_SEC("xdp", BPF_PROG_TYPE_XDP, + BPF_XDP), BPF_PROG_SEC("perf_event", BPF_PROG_TYPE_PERF_EVENT), BPF_PROG_SEC("lwt_in", BPF_PROG_TYPE_LWT_IN), BPF_PROG_SEC("lwt_out", BPF_PROG_TYPE_LWT_OUT), @@ -8281,6 +8282,12 @@ bpf_program__attach_netns(struct bpf_program *prog, int netns_fd) return bpf_program__attach_fd(prog, netns_fd, "netns"); } +struct bpf_link *bpf_program__attach_xdp(struct bpf_program *prog, int ifindex) +{ + /* target_fd/target_ifindex use the same field in LINK_CREATE */ + return bpf_program__attach_fd(prog, ifindex, "xdp"); +} + struct bpf_link * bpf_program__attach_iter(struct bpf_program *prog, const struct bpf_iter_attach_opts *opts) diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h index c6813791fa7e..9924385462ab 100644 --- a/tools/lib/bpf/libbpf.h +++ b/tools/lib/bpf/libbpf.h @@ -257,6 +257,8 @@ LIBBPF_API struct bpf_link * bpf_program__attach_cgroup(struct bpf_program *prog, int cgroup_fd); LIBBPF_API struct bpf_link * bpf_program__attach_netns(struct bpf_program *prog, int netns_fd); +LIBBPF_API struct bpf_link * +bpf_program__attach_xdp(struct bpf_program *prog, int ifindex); struct bpf_map; diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map index 6f0856abe299..ca49a6a7e5b2 100644 --- a/tools/lib/bpf/libbpf.map +++ b/tools/lib/bpf/libbpf.map @@ -286,6 +286,7 @@ LIBBPF_0.1.0 { bpf_map__set_value_size; bpf_map__type; bpf_map__value_size; + bpf_program__attach_xdp; bpf_program__autoload; bpf_program__is_sk_lookup; bpf_program__set_autoload; diff --git a/tools/testing/selftests/bpf/prog_tests/section_names.c b/tools/testing/selftests/bpf/prog_tests/section_names.c index 713167449c98..8b571890c57e 100644 --- a/tools/testing/selftests/bpf/prog_tests/section_names.c +++ b/tools/testing/selftests/bpf/prog_tests/section_names.c @@ -35,7 +35,7 @@ static struct sec_name_test tests[] = { {-EINVAL, 0}, }, {"raw_tp/", {0, BPF_PROG_TYPE_RAW_TRACEPOINT, 0}, {-EINVAL, 0} }, - {"xdp", {0, BPF_PROG_TYPE_XDP, 0}, {-EINVAL, 0} }, + {"xdp", {0, BPF_PROG_TYPE_XDP, BPF_XDP}, {0, BPF_XDP} }, {"perf_event", {0, BPF_PROG_TYPE_PERF_EVENT, 0}, {-EINVAL, 0} }, {"lwt_in", {0, BPF_PROG_TYPE_LWT_IN, 0}, {-EINVAL, 0} }, {"lwt_out", {0, BPF_PROG_TYPE_LWT_OUT, 0}, {-EINVAL, 0} }, -- cgit v1.2.3 From fe48230cf2ae27c9e3b96d29908e22e2926fd1ab Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Tue, 21 Jul 2020 23:46:01 -0700 Subject: selftests/bpf: Add BPF XDP link selftests Add selftest validating all the attachment logic around BPF XDP link. Test also link updates and get_obj_info() APIs. Signed-off-by: Andrii Nakryiko Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/20200722064603.3350758-9-andriin@fb.com --- tools/testing/selftests/bpf/prog_tests/xdp_link.c | 137 ++++++++++++++++++++++ tools/testing/selftests/bpf/progs/test_xdp_link.c | 12 ++ 2 files changed, 149 insertions(+) create mode 100644 tools/testing/selftests/bpf/prog_tests/xdp_link.c create mode 100644 tools/testing/selftests/bpf/progs/test_xdp_link.c (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_link.c b/tools/testing/selftests/bpf/prog_tests/xdp_link.c new file mode 100644 index 000000000000..52cba6795d40 --- /dev/null +++ b/tools/testing/selftests/bpf/prog_tests/xdp_link.c @@ -0,0 +1,137 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2020 Facebook */ +#include +#include +#include "test_xdp_link.skel.h" + +#define IFINDEX_LO 1 + +void test_xdp_link(void) +{ + __u32 duration = 0, id1, id2, id0 = 0, prog_fd1, prog_fd2, err; + DECLARE_LIBBPF_OPTS(bpf_xdp_set_link_opts, opts, .old_fd = -1); + struct test_xdp_link *skel1 = NULL, *skel2 = NULL; + struct bpf_link_info link_info; + struct bpf_prog_info prog_info; + struct bpf_link *link; + __u32 link_info_len = sizeof(link_info); + __u32 prog_info_len = sizeof(prog_info); + + skel1 = test_xdp_link__open_and_load(); + if (CHECK(!skel1, "skel_load", "skeleton open and load failed\n")) + goto cleanup; + prog_fd1 = bpf_program__fd(skel1->progs.xdp_handler); + + skel2 = test_xdp_link__open_and_load(); + if (CHECK(!skel2, "skel_load", "skeleton open and load failed\n")) + goto cleanup; + prog_fd2 = bpf_program__fd(skel2->progs.xdp_handler); + + memset(&prog_info, 0, sizeof(prog_info)); + err = bpf_obj_get_info_by_fd(prog_fd1, &prog_info, &prog_info_len); + if (CHECK(err, "fd_info1", "failed %d\n", -errno)) + goto cleanup; + id1 = prog_info.id; + + memset(&prog_info, 0, sizeof(prog_info)); + err = bpf_obj_get_info_by_fd(prog_fd2, &prog_info, &prog_info_len); + if (CHECK(err, "fd_info2", "failed %d\n", -errno)) + goto cleanup; + id2 = prog_info.id; + + /* set initial prog attachment */ + err = bpf_set_link_xdp_fd_opts(IFINDEX_LO, prog_fd1, XDP_FLAGS_REPLACE, &opts); + if (CHECK(err, "fd_attach", "initial prog attach failed: %d\n", err)) + goto cleanup; + + /* validate prog ID */ + err = bpf_get_link_xdp_id(IFINDEX_LO, &id0, 0); + CHECK(err || id0 != id1, "id1_check", + "loaded prog id %u != id1 %u, err %d", id0, id1, err); + + /* BPF link is not allowed to replace prog attachment */ + link = bpf_program__attach_xdp(skel1->progs.xdp_handler, IFINDEX_LO); + if (CHECK(!IS_ERR(link), "link_attach_fail", "unexpected success\n")) { + bpf_link__destroy(link); + /* best-effort detach prog */ + opts.old_fd = prog_fd1; + bpf_set_link_xdp_fd_opts(IFINDEX_LO, -1, XDP_FLAGS_REPLACE, &opts); + goto cleanup; + } + + /* detach BPF program */ + opts.old_fd = prog_fd1; + err = bpf_set_link_xdp_fd_opts(IFINDEX_LO, -1, XDP_FLAGS_REPLACE, &opts); + if (CHECK(err, "prog_detach", "failed %d\n", err)) + goto cleanup; + + /* now BPF link should attach successfully */ + link = bpf_program__attach_xdp(skel1->progs.xdp_handler, IFINDEX_LO); + if (CHECK(IS_ERR(link), "link_attach", "failed: %ld\n", PTR_ERR(link))) + goto cleanup; + skel1->links.xdp_handler = link; + + /* validate prog ID */ + err = bpf_get_link_xdp_id(IFINDEX_LO, &id0, 0); + if (CHECK(err || id0 != id1, "id1_check", + "loaded prog id %u != id1 %u, err %d", id0, id1, err)) + goto cleanup; + + /* BPF prog attach is not allowed to replace BPF link */ + opts.old_fd = prog_fd1; + err = bpf_set_link_xdp_fd_opts(IFINDEX_LO, prog_fd2, XDP_FLAGS_REPLACE, &opts); + if (CHECK(!err, "prog_attach_fail", "unexpected success\n")) + goto cleanup; + + /* Can't force-update when BPF link is active */ + err = bpf_set_link_xdp_fd(IFINDEX_LO, prog_fd2, 0); + if (CHECK(!err, "prog_update_fail", "unexpected success\n")) + goto cleanup; + + /* Can't force-detach when BPF link is active */ + err = bpf_set_link_xdp_fd(IFINDEX_LO, -1, 0); + if (CHECK(!err, "prog_detach_fail", "unexpected success\n")) + goto cleanup; + + /* BPF link is not allowed to replace another BPF link */ + link = bpf_program__attach_xdp(skel2->progs.xdp_handler, IFINDEX_LO); + if (CHECK(!IS_ERR(link), "link_attach_fail", "unexpected success\n")) { + bpf_link__destroy(link); + goto cleanup; + } + + bpf_link__destroy(skel1->links.xdp_handler); + skel1->links.xdp_handler = NULL; + + /* new link attach should succeed */ + link = bpf_program__attach_xdp(skel2->progs.xdp_handler, IFINDEX_LO); + if (CHECK(IS_ERR(link), "link_attach", "failed: %ld\n", PTR_ERR(link))) + goto cleanup; + skel2->links.xdp_handler = link; + + err = bpf_get_link_xdp_id(IFINDEX_LO, &id0, 0); + if (CHECK(err || id0 != id2, "id2_check", + "loaded prog id %u != id2 %u, err %d", id0, id1, err)) + goto cleanup; + + /* updating program under active BPF link works as expected */ + err = bpf_link__update_program(link, skel1->progs.xdp_handler); + if (CHECK(err, "link_upd", "failed: %d\n", err)) + goto cleanup; + + memset(&link_info, 0, sizeof(link_info)); + err = bpf_obj_get_info_by_fd(bpf_link__fd(link), &link_info, &link_info_len); + if (CHECK(err, "link_info", "failed: %d\n", err)) + goto cleanup; + + CHECK(link_info.type != BPF_LINK_TYPE_XDP, "link_type", + "got %u != exp %u\n", link_info.type, BPF_LINK_TYPE_XDP); + CHECK(link_info.prog_id != id1, "link_prog_id", + "got %u != exp %u\n", link_info.prog_id, id1); + CHECK(link_info.xdp.ifindex != IFINDEX_LO, "link_ifindex", + "got %u != exp %u\n", link_info.xdp.ifindex, IFINDEX_LO); + +cleanup: + test_xdp_link__destroy(skel1); + test_xdp_link__destroy(skel2); +} diff --git a/tools/testing/selftests/bpf/progs/test_xdp_link.c b/tools/testing/selftests/bpf/progs/test_xdp_link.c new file mode 100644 index 000000000000..eb93ea95d1d8 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/test_xdp_link.c @@ -0,0 +1,12 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2020 Facebook */ +#include +#include + +char LICENSE[] SEC("license") = "GPL"; + +SEC("xdp/handler") +int xdp_handler(struct xdp_md *xdp) +{ + return 0; +} -- cgit v1.2.3 From 70cfab1d871c771bd1963dd898f4f9b2731590c2 Mon Sep 17 00:00:00 2001 From: Quentin Monnet Date: Fri, 24 Jul 2020 10:06:17 +0100 Subject: tools, bpftool: Skip type probe if name is not found For probing program and map types, bpftool loops on type values and uses the relevant type name in prog_type_name[] or map_type_name[]. To ensure the name exists, we exit from the loop if we go over the size of the array. However, this is not enough in the case where the arrays have "holes" in them, program or map types for which they have no name, but not at the end of the list. This is currently the case for BPF_PROG_TYPE_LSM, not known to bpftool and which name is a null string. When probing for features, bpftool attempts to strlen() that name and segfaults. Let's fix it by skipping probes for "unknown" program and map types, with an informational message giving the numeral value in that case. Fixes: 93a3545d812a ("tools/bpftool: Add name mappings for SK_LOOKUP prog and attach type") Reported-by: Paul Chaignon Signed-off-by: Quentin Monnet Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20200724090618.16378-2-quentin@isovalent.com --- tools/bpf/bpftool/feature.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'tools') diff --git a/tools/bpf/bpftool/feature.c b/tools/bpf/bpftool/feature.c index 1cd75807673e..a43a6f10b564 100644 --- a/tools/bpf/bpftool/feature.c +++ b/tools/bpf/bpftool/feature.c @@ -504,6 +504,10 @@ probe_prog_type(enum bpf_prog_type prog_type, bool *supported_types, supported_types[prog_type] |= res; + if (!prog_type_name[prog_type]) { + p_info("program type name not found (type %d)", prog_type); + return; + } maxlen = sizeof(plain_desc) - strlen(plain_comment) - 1; if (strlen(prog_type_name[prog_type]) > maxlen) { p_info("program type name too long"); @@ -533,6 +537,10 @@ probe_map_type(enum bpf_map_type map_type, const char *define_prefix, * check required for unprivileged users */ + if (!map_type_name[map_type]) { + p_info("map type name not found (type %d)", map_type); + return; + } maxlen = sizeof(plain_desc) - strlen(plain_comment) - 1; if (strlen(map_type_name[map_type]) > maxlen) { p_info("map type name too long"); -- cgit v1.2.3 From 9a97c9d2af5ca798377342debf7f0f44281d050e Mon Sep 17 00:00:00 2001 From: Quentin Monnet Date: Fri, 24 Jul 2020 10:06:18 +0100 Subject: tools, bpftool: Add LSM type to array of prog names Assign "lsm" as a printed name for BPF_PROG_TYPE_LSM in bpftool, so that it can use it when listing programs loaded on the system or when probing features. Signed-off-by: Quentin Monnet Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20200724090618.16378-3-quentin@isovalent.com --- tools/bpf/bpftool/prog.c | 1 + 1 file changed, 1 insertion(+) (limited to 'tools') diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c index 3e6ecc6332e2..158995d853b0 100644 --- a/tools/bpf/bpftool/prog.c +++ b/tools/bpf/bpftool/prog.c @@ -59,6 +59,7 @@ const char * const prog_type_name[] = { [BPF_PROG_TYPE_TRACING] = "tracing", [BPF_PROG_TYPE_STRUCT_OPS] = "struct_ops", [BPF_PROG_TYPE_EXT] = "ext", + [BPF_PROG_TYPE_LSM] = "lsm", [BPF_PROG_TYPE_SK_LOOKUP] = "sk_lookup", }; -- cgit v1.2.3 From e1613b5714ee6c186c9628e9958edf65e9d9cddd Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Mon, 27 Jul 2020 15:47:15 -0700 Subject: bpf: Fix bpf_ringbuf_output() signature to return long Due to bpf tree fix merge, bpf_ringbuf_output() signature ended up with int as a return type, while all other helpers got converted to returning long. So fix it in bpf-next now. Fixes: b0659d8a950d ("bpf: Fix definition of bpf_ringbuf_output() helper in UAPI comments") Signed-off-by: Andrii Nakryiko Signed-off-by: Daniel Borkmann Acked-by: Song Liu Link: https://lore.kernel.org/bpf/20200727224715.652037-1-andriin@fb.com --- include/uapi/linux/bpf.h | 2 +- tools/include/uapi/linux/bpf.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'tools') diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index e1ba4ae6a916..eb5e0c38eb2c 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -3241,7 +3241,7 @@ union bpf_attr { * Return * The id is returned or 0 in case the id could not be retrieved. * - * int bpf_ringbuf_output(void *ringbuf, void *data, u64 size, u64 flags) + * long bpf_ringbuf_output(void *ringbuf, void *data, u64 size, u64 flags) * Description * Copy *size* bytes from *data* into a ring buffer *ringbuf*. * If **BPF_RB_NO_WAKEUP** is specified in *flags*, no notification diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index e1ba4ae6a916..eb5e0c38eb2c 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -3241,7 +3241,7 @@ union bpf_attr { * Return * The id is returned or 0 in case the id could not be retrieved. * - * int bpf_ringbuf_output(void *ringbuf, void *data, u64 size, u64 flags) + * long bpf_ringbuf_output(void *ringbuf, void *data, u64 size, u64 flags) * Description * Copy *size* bytes from *data* into a ring buffer *ringbuf*. * If **BPF_RB_NO_WAKEUP** is specified in *flags*, no notification -- cgit v1.2.3 From 363885d7c62e293fb093c7c355bf5f05fa0a25a9 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Mon, 27 Jul 2020 16:33:45 -0700 Subject: selftests/bpf: Add new bpf_iter context structs to fix build on old kernels Add bpf_iter__bpf_map_elem and bpf_iter__bpf_sk_storage_map to bpf_iter.h. Fixes: 3b1c420bd882 ("selftests/bpf: Add a test for bpf sk_storage_map iterator") Fixes: 2a7c2fff7dd6 ("selftests/bpf: Add test for bpf hash map iterators") Signed-off-by: Andrii Nakryiko Signed-off-by: Daniel Borkmann Acked-by: Yonghong Song Link: https://lore.kernel.org/bpf/20200727233345.1686358-1-andriin@fb.com --- tools/testing/selftests/bpf/progs/bpf_iter.h | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) (limited to 'tools') diff --git a/tools/testing/selftests/bpf/progs/bpf_iter.h b/tools/testing/selftests/bpf/progs/bpf_iter.h index 17db3bac518b..c196280df90d 100644 --- a/tools/testing/selftests/bpf/progs/bpf_iter.h +++ b/tools/testing/selftests/bpf/progs/bpf_iter.h @@ -11,6 +11,8 @@ #define tcp6_sock tcp6_sock___not_used #define bpf_iter__udp bpf_iter__udp___not_used #define udp6_sock udp6_sock___not_used +#define bpf_iter__bpf_map_elem bpf_iter__bpf_map_elem___not_used +#define bpf_iter__bpf_sk_storage_map bpf_iter__bpf_sk_storage_map___not_used #include "vmlinux.h" #undef bpf_iter_meta #undef bpf_iter__bpf_map @@ -22,6 +24,8 @@ #undef tcp6_sock #undef bpf_iter__udp #undef udp6_sock +#undef bpf_iter__bpf_map_elem +#undef bpf_iter__bpf_sk_storage_map struct bpf_iter_meta { struct seq_file *seq; @@ -78,3 +82,17 @@ struct udp6_sock { struct udp_sock udp; struct ipv6_pinfo inet6; } __attribute__((preserve_access_index)); + +struct bpf_iter__bpf_map_elem { + struct bpf_iter_meta *meta; + struct bpf_map *map; + void *key; + void *value; +}; + +struct bpf_iter__bpf_sk_storage_map { + struct bpf_iter_meta *meta; + struct bpf_map *map; + struct sock *sk; + void *value; +}; -- cgit v1.2.3 From ca5cd355b7f0372da0d50fce5b12a3367e417290 Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Tue, 28 Jul 2020 07:39:02 -0700 Subject: bpf, selftests: use :: 1 for localhost in tcp_server.py Using localhost requires the host to have a /etc/hosts file with that specific line in it. By default my dev box did not, they used ip6-localhost, so the test was failing. To fix remove the need for any /etc/hosts and use ::1. I could just add the line, but this seems easier. Signed-off-by: John Fastabend Signed-off-by: Daniel Borkmann Acked-by: Song Liu Acked-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/159594714197.21431.10113693935099326445.stgit@john-Precision-5820-Tower --- tools/testing/selftests/bpf/tcp_client.py | 2 +- tools/testing/selftests/bpf/tcp_server.py | 2 +- tools/testing/selftests/bpf/test_netcnt.c | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'tools') diff --git a/tools/testing/selftests/bpf/tcp_client.py b/tools/testing/selftests/bpf/tcp_client.py index a53ed58528d6..bfff82be3fc1 100755 --- a/tools/testing/selftests/bpf/tcp_client.py +++ b/tools/testing/selftests/bpf/tcp_client.py @@ -34,7 +34,7 @@ serverPort = int(sys.argv[1]) # create active socket sock = socket.socket(socket.AF_INET6, socket.SOCK_STREAM) try: - sock.connect(('localhost', serverPort)) + sock.connect(('::1', serverPort)) except socket.error as e: sys.exit(1) diff --git a/tools/testing/selftests/bpf/tcp_server.py b/tools/testing/selftests/bpf/tcp_server.py index 0ca60d193bed..42ab8882f00f 100755 --- a/tools/testing/selftests/bpf/tcp_server.py +++ b/tools/testing/selftests/bpf/tcp_server.py @@ -38,7 +38,7 @@ serverSocket = None # create passive socket serverSocket = socket.socket(socket.AF_INET6, socket.SOCK_STREAM) -try: serverSocket.bind(('localhost', 0)) +try: serverSocket.bind(('::1', 0)) except socket.error as msg: print('bind fails: ' + str(msg)) diff --git a/tools/testing/selftests/bpf/test_netcnt.c b/tools/testing/selftests/bpf/test_netcnt.c index c1da5404454a..7a68c9069639 100644 --- a/tools/testing/selftests/bpf/test_netcnt.c +++ b/tools/testing/selftests/bpf/test_netcnt.c @@ -82,9 +82,9 @@ int main(int argc, char **argv) } if (system("which ping6 &>/dev/null") == 0) - assert(!system("ping6 localhost -c 10000 -f -q > /dev/null")); + assert(!system("ping6 ::1 -c 10000 -f -q > /dev/null")); else - assert(!system("ping -6 localhost -c 10000 -f -q > /dev/null")); + assert(!system("ping -6 ::1 -c 10000 -f -q > /dev/null")); if (bpf_prog_query(cgroup_fd, BPF_CGROUP_INET_EGRESS, 0, NULL, NULL, &prog_cnt)) { -- cgit v1.2.3 From 12e6196fb15953605be54ac9320ac54371aecab7 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Tue, 28 Jul 2020 15:18:01 -0700 Subject: selftests/bpf: Test bpf_iter buffer access with negative offset Commit afbf21dce668 ("bpf: Support readonly/readwrite buffers in verifier") added readonly/readwrite buffer support which is currently used by bpf_iter tracing programs. It has a bug with incorrect parameter ordering which later fixed by Commit f6dfbe31e8fa ("bpf: Fix swapped arguments in calls to check_buffer_access"). This patch added a test case with a negative offset access which will trigger the error path. Without Commit f6dfbe31e8fa, running the test case in the patch, the error message looks like: R1_w=rdwr_buf(id=0,off=0,imm=0) R10=fp0 ; value_sum += *(__u32 *)(value - 4); 2: (61) r1 = *(u32 *)(r1 -4) R1 invalid (null) buffer access: off=-4, size=4 With the above commit, the error message looks like: R1_w=rdwr_buf(id=0,off=0,imm=0) R10=fp0 ; value_sum += *(__u32 *)(value - 4); 2: (61) r1 = *(u32 *)(r1 -4) R1 invalid rdwr buffer access: off=-4, size=4 Signed-off-by: Yonghong Song Signed-off-by: Daniel Borkmann Acked-by: Song Liu Link: https://lore.kernel.org/bpf/20200728221801.1090406-1-yhs@fb.com --- tools/testing/selftests/bpf/prog_tests/bpf_iter.c | 13 +++++++++++++ .../selftests/bpf/progs/bpf_iter_test_kern6.c | 21 +++++++++++++++++++++ 2 files changed, 34 insertions(+) create mode 100644 tools/testing/selftests/bpf/progs/bpf_iter_test_kern6.c (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c index d95de80b1851..4ffefdc1130f 100644 --- a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c +++ b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c @@ -21,6 +21,7 @@ #include "bpf_iter_bpf_percpu_array_map.skel.h" #include "bpf_iter_bpf_sk_storage_map.skel.h" #include "bpf_iter_test_kern5.skel.h" +#include "bpf_iter_test_kern6.skel.h" static int duration; @@ -885,6 +886,16 @@ static void test_rdonly_buf_out_of_bound(void) bpf_iter_test_kern5__destroy(skel); } +static void test_buf_neg_offset(void) +{ + struct bpf_iter_test_kern6 *skel; + + skel = bpf_iter_test_kern6__open_and_load(); + if (CHECK(skel, "bpf_iter_test_kern6__open_and_load", + "skeleton open_and_load unexpected success\n")) + bpf_iter_test_kern6__destroy(skel); +} + void test_bpf_iter(void) { if (test__start_subtest("btf_id_or_null")) @@ -933,4 +944,6 @@ void test_bpf_iter(void) test_bpf_sk_storage_map(); if (test__start_subtest("rdonly-buf-out-of-bound")) test_rdonly_buf_out_of_bound(); + if (test__start_subtest("buf-neg-offset")) + test_buf_neg_offset(); } diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_test_kern6.c b/tools/testing/selftests/bpf/progs/bpf_iter_test_kern6.c new file mode 100644 index 000000000000..1c7304f56b1e --- /dev/null +++ b/tools/testing/selftests/bpf/progs/bpf_iter_test_kern6.c @@ -0,0 +1,21 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2020 Facebook */ +#include "bpf_iter.h" +#include + +char _license[] SEC("license") = "GPL"; + +__u32 value_sum = 0; + +SEC("iter/bpf_map_elem") +int dump_bpf_hash_map(struct bpf_iter__bpf_map_elem *ctx) +{ + void *value = ctx->value; + + if (value == (void *)0) + return 0; + + /* negative offset, verifier failure. */ + value_sum += *(__u32 *)(value - 4); + return 0; +} -- cgit v1.2.3 From 4fb5f94911405b6d2645d4384c2ae1215bfc6a76 Mon Sep 17 00:00:00 2001 From: Stanislav Fomichev Date: Tue, 28 Jul 2020 17:31:04 -0700 Subject: selftests/bpf: Verify socket storage in cgroup/sock_{create, release} Augment udp_limit test to set and verify socket storage value. That should be enough to exercise the changes from the previous patch. Signed-off-by: Stanislav Fomichev Signed-off-by: Daniel Borkmann Acked-by: Song Liu Link: https://lore.kernel.org/bpf/20200729003104.1280813-2-sdf@google.com --- tools/testing/selftests/bpf/progs/udp_limit.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) (limited to 'tools') diff --git a/tools/testing/selftests/bpf/progs/udp_limit.c b/tools/testing/selftests/bpf/progs/udp_limit.c index 8429b22525a7..165e3c2dd9a3 100644 --- a/tools/testing/selftests/bpf/progs/udp_limit.c +++ b/tools/testing/selftests/bpf/progs/udp_limit.c @@ -6,14 +6,28 @@ int invocations = 0, in_use = 0; +struct { + __uint(type, BPF_MAP_TYPE_SK_STORAGE); + __uint(map_flags, BPF_F_NO_PREALLOC); + __type(key, int); + __type(value, int); +} sk_map SEC(".maps"); + SEC("cgroup/sock_create") int sock(struct bpf_sock *ctx) { + int *sk_storage; __u32 key; if (ctx->type != SOCK_DGRAM) return 1; + sk_storage = bpf_sk_storage_get(&sk_map, ctx, 0, + BPF_SK_STORAGE_GET_F_CREATE); + if (!sk_storage) + return 0; + *sk_storage = 0xdeadbeef; + __sync_fetch_and_add(&invocations, 1); if (in_use > 0) { @@ -31,11 +45,16 @@ int sock(struct bpf_sock *ctx) SEC("cgroup/sock_release") int sock_release(struct bpf_sock *ctx) { + int *sk_storage; __u32 key; if (ctx->type != SOCK_DGRAM) return 1; + sk_storage = bpf_sk_storage_get(&sk_map, ctx, 0, 0); + if (!sk_storage || *sk_storage != 0xdeadbeef) + return 0; + __sync_fetch_and_add(&invocations, 1); __sync_fetch_and_add(&in_use, -1); return 1; -- cgit v1.2.3 From dfdb0d93e5bc351af5b286ae9c630d3cf869b810 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Wed, 29 Jul 2020 16:56:58 +0800 Subject: selftests/bpf: Add xdpdrv mode for test_xdp_redirect This patch add xdpdrv mode for test_xdp_redirect.sh since veth has support native mode. After update here is the test result: # ./test_xdp_redirect.sh selftests: test_xdp_redirect xdpgeneric [PASS] selftests: test_xdp_redirect xdpdrv [PASS] Signed-off-by: Hangbin Liu Signed-off-by: Daniel Borkmann Acked-by: Song Liu Acked-by: William Tu Link: https://lore.kernel.org/bpf/20200729085658.403794-1-liuhangbin@gmail.com --- tools/testing/selftests/bpf/test_xdp_redirect.sh | 84 +++++++++++++++--------- 1 file changed, 52 insertions(+), 32 deletions(-) (limited to 'tools') diff --git a/tools/testing/selftests/bpf/test_xdp_redirect.sh b/tools/testing/selftests/bpf/test_xdp_redirect.sh index c4b17e08d431..dd80f0c84afb 100755 --- a/tools/testing/selftests/bpf/test_xdp_redirect.sh +++ b/tools/testing/selftests/bpf/test_xdp_redirect.sh @@ -10,52 +10,72 @@ # | xdp forwarding | # ------------------ -cleanup() +ret=0 + +setup() { - if [ "$?" = "0" ]; then - echo "selftests: test_xdp_redirect [PASS]"; - else - echo "selftests: test_xdp_redirect [FAILED]"; - fi - set +e + local xdpmode=$1 + + ip netns add ns1 + ip netns add ns2 + + ip link add veth1 index 111 type veth peer name veth11 netns ns1 + ip link add veth2 index 222 type veth peer name veth22 netns ns2 + + ip link set veth1 up + ip link set veth2 up + ip -n ns1 link set dev veth11 up + ip -n ns2 link set dev veth22 up + + ip -n ns1 addr add 10.1.1.11/24 dev veth11 + ip -n ns2 addr add 10.1.1.22/24 dev veth22 +} + +cleanup() +{ ip link del veth1 2> /dev/null ip link del veth2 2> /dev/null ip netns del ns1 2> /dev/null ip netns del ns2 2> /dev/null } -ip link set dev lo xdpgeneric off 2>/dev/null > /dev/null -if [ $? -ne 0 ];then - echo "selftests: [SKIP] Could not run test without the ip xdpgeneric support" - exit 0 -fi -set -e - -ip netns add ns1 -ip netns add ns2 +test_xdp_redirect() +{ + local xdpmode=$1 -trap cleanup 0 2 3 6 9 + setup -ip link add veth1 index 111 type veth peer name veth11 -ip link add veth2 index 222 type veth peer name veth22 + ip link set dev veth1 $xdpmode off &> /dev/null + if [ $? -ne 0 ];then + echo "selftests: test_xdp_redirect $xdpmode [SKIP]" + return 0 + fi -ip link set veth11 netns ns1 -ip link set veth22 netns ns2 + ip -n ns1 link set veth11 $xdpmode obj xdp_dummy.o sec xdp_dummy &> /dev/null + ip -n ns2 link set veth22 $xdpmode obj xdp_dummy.o sec xdp_dummy &> /dev/null + ip link set dev veth1 $xdpmode obj test_xdp_redirect.o sec redirect_to_222 &> /dev/null + ip link set dev veth2 $xdpmode obj test_xdp_redirect.o sec redirect_to_111 &> /dev/null -ip link set veth1 up -ip link set veth2 up + ip netns exec ns1 ping -c 1 10.1.1.22 &> /dev/null + local ret1=$? + ip netns exec ns2 ping -c 1 10.1.1.11 &> /dev/null + local ret2=$? -ip netns exec ns1 ip addr add 10.1.1.11/24 dev veth11 -ip netns exec ns2 ip addr add 10.1.1.22/24 dev veth22 + if [ $ret1 -eq 0 -a $ret2 -eq 0 ]; then + echo "selftests: test_xdp_redirect $xdpmode [PASS]"; + else + ret=1 + echo "selftests: test_xdp_redirect $xdpmode [FAILED]"; + fi -ip netns exec ns1 ip link set dev veth11 up -ip netns exec ns2 ip link set dev veth22 up + cleanup +} -ip link set dev veth1 xdpgeneric obj test_xdp_redirect.o sec redirect_to_222 -ip link set dev veth2 xdpgeneric obj test_xdp_redirect.o sec redirect_to_111 +set -e +trap cleanup 2 3 6 9 -ip netns exec ns1 ping -c 1 10.1.1.22 -ip netns exec ns2 ping -c 1 10.1.1.11 +test_xdp_redirect xdpgeneric +test_xdp_redirect xdpdrv -exit 0 +exit $ret -- cgit v1.2.3 From 80546ac4586c0bd326aa7ce80f076646db02bcd0 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Tue, 28 Jul 2020 21:50:56 -0700 Subject: selftests/bpf: Don't destroy failed link Check that link is NULL or proper pointer before invoking bpf_link__destroy(). Not doing this causes crash in test_progs, when cg_storage_multi selftest fails. Fixes: 3573f384014f ("selftests/bpf: Test CGROUP_STORAGE behavior on shared egress + ingress") Signed-off-by: Andrii Nakryiko Signed-off-by: Daniel Borkmann Acked-by: Song Liu Link: https://lore.kernel.org/bpf/20200729045056.3363921-1-andriin@fb.com --- .../selftests/bpf/prog_tests/cg_storage_multi.c | 42 ++++++++++++++-------- 1 file changed, 28 insertions(+), 14 deletions(-) (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c b/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c index c67d8c076a34..643dfa35419c 100644 --- a/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c +++ b/tools/testing/selftests/bpf/prog_tests/cg_storage_multi.c @@ -147,8 +147,10 @@ static void test_egress_only(int parent_cgroup_fd, int child_cgroup_fd) goto close_bpf_object; close_bpf_object: - bpf_link__destroy(parent_link); - bpf_link__destroy(child_link); + if (!IS_ERR(parent_link)) + bpf_link__destroy(parent_link); + if (!IS_ERR(child_link)) + bpf_link__destroy(child_link); cg_storage_multi_egress_only__destroy(obj); } @@ -262,12 +264,18 @@ static void test_isolated(int parent_cgroup_fd, int child_cgroup_fd) goto close_bpf_object; close_bpf_object: - bpf_link__destroy(parent_egress1_link); - bpf_link__destroy(parent_egress2_link); - bpf_link__destroy(parent_ingress_link); - bpf_link__destroy(child_egress1_link); - bpf_link__destroy(child_egress2_link); - bpf_link__destroy(child_ingress_link); + if (!IS_ERR(parent_egress1_link)) + bpf_link__destroy(parent_egress1_link); + if (!IS_ERR(parent_egress2_link)) + bpf_link__destroy(parent_egress2_link); + if (!IS_ERR(parent_ingress_link)) + bpf_link__destroy(parent_ingress_link); + if (!IS_ERR(child_egress1_link)) + bpf_link__destroy(child_egress1_link); + if (!IS_ERR(child_egress2_link)) + bpf_link__destroy(child_egress2_link); + if (!IS_ERR(child_ingress_link)) + bpf_link__destroy(child_ingress_link); cg_storage_multi_isolated__destroy(obj); } @@ -367,12 +375,18 @@ static void test_shared(int parent_cgroup_fd, int child_cgroup_fd) goto close_bpf_object; close_bpf_object: - bpf_link__destroy(parent_egress1_link); - bpf_link__destroy(parent_egress2_link); - bpf_link__destroy(parent_ingress_link); - bpf_link__destroy(child_egress1_link); - bpf_link__destroy(child_egress2_link); - bpf_link__destroy(child_ingress_link); + if (!IS_ERR(parent_egress1_link)) + bpf_link__destroy(parent_egress1_link); + if (!IS_ERR(parent_egress2_link)) + bpf_link__destroy(parent_egress2_link); + if (!IS_ERR(parent_ingress_link)) + bpf_link__destroy(parent_ingress_link); + if (!IS_ERR(child_egress1_link)) + bpf_link__destroy(child_egress1_link); + if (!IS_ERR(child_egress2_link)) + bpf_link__destroy(child_egress2_link); + if (!IS_ERR(child_ingress_link)) + bpf_link__destroy(child_ingress_link); cg_storage_multi_shared__destroy(obj); } -- cgit v1.2.3 From a6599abdeac30063baf89df166068b20758e0e86 Mon Sep 17 00:00:00 2001 From: Jakub Sitnicki Date: Thu, 30 Jul 2020 14:53:25 +0200 Subject: selftests/bpf: Omit nodad flag when adding addresses to loopback Setting IFA_F_NODAD flag for IPv6 addresses to add to loopback is unnecessary. Duplicate Address Detection does not happen on loopback device. Also, passing 'nodad' flag to 'ip address' breaks libbpf CI, which runs in an environment with BusyBox implementation of 'ip' command, that doesn't understand this flag. Fixes: 0ab5539f8584 ("selftests/bpf: Tests for BPF_SK_LOOKUP attach point") Reported-by: Andrii Nakryiko Signed-off-by: Jakub Sitnicki Signed-off-by: Daniel Borkmann Tested-by: Andrii Nakryiko Acked-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20200730125325.1869363-1-jakub@cloudflare.com --- tools/testing/selftests/bpf/prog_tests/sk_lookup.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c index 9bbd2b2b7630..379da6f10ee9 100644 --- a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c +++ b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c @@ -1290,8 +1290,8 @@ static void run_tests(struct test_sk_lookup *skel) static int switch_netns(void) { static const char * const setup_script[] = { - "ip -6 addr add dev lo " EXT_IP6 "/128 nodad", - "ip -6 addr add dev lo " INT_IP6 "/128 nodad", + "ip -6 addr add dev lo " EXT_IP6 "/128", + "ip -6 addr add dev lo " INT_IP6 "/128", "ip link set dev lo up", NULL, }; -- cgit v1.2.3 From 50450fc716c1a570ee8d8bfe198ef5d3cfca36e4 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Wed, 29 Jul 2020 16:21:48 -0700 Subject: libbpf: Make destructors more robust by handling ERR_PTR(err) cases Most of libbpf "constructors" on failure return ERR_PTR(err) result encoded as a pointer. It's a common mistake to eventually pass such malformed pointers into xxx__destroy()/xxx__free() "destructors". So instead of fixing up clean up code in selftests and user programs, handle such error pointers in destructors themselves. This works beautifully for NULL pointers passed to destructors, so might as well just work for error pointers. Suggested-by: Song Liu Signed-off-by: Andrii Nakryiko Signed-off-by: Daniel Borkmann Acked-by: Song Liu Link: https://lore.kernel.org/bpf/20200729232148.896125-1-andriin@fb.com --- tools/lib/bpf/btf.c | 4 ++-- tools/lib/bpf/btf_dump.c | 2 +- tools/lib/bpf/libbpf.c | 9 ++++----- 3 files changed, 7 insertions(+), 8 deletions(-) (limited to 'tools') diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c index c9e760e120dc..ded5b29965f9 100644 --- a/tools/lib/bpf/btf.c +++ b/tools/lib/bpf/btf.c @@ -386,7 +386,7 @@ __s32 btf__find_by_name_kind(const struct btf *btf, const char *type_name, void btf__free(struct btf *btf) { - if (!btf) + if (IS_ERR_OR_NULL(btf)) return; if (btf->fd >= 0) @@ -1025,7 +1025,7 @@ static int btf_ext_parse_hdr(__u8 *data, __u32 data_size) void btf_ext__free(struct btf_ext *btf_ext) { - if (!btf_ext) + if (IS_ERR_OR_NULL(btf_ext)) return; free(btf_ext->data); free(btf_ext); diff --git a/tools/lib/bpf/btf_dump.c b/tools/lib/bpf/btf_dump.c index e1c344504cae..cf711168d34a 100644 --- a/tools/lib/bpf/btf_dump.c +++ b/tools/lib/bpf/btf_dump.c @@ -183,7 +183,7 @@ void btf_dump__free(struct btf_dump *d) { int i, cnt; - if (!d) + if (IS_ERR_OR_NULL(d)) return; free(d->type_states); diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index 54830d603fee..b9f11f854985 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -6504,7 +6504,7 @@ void bpf_object__close(struct bpf_object *obj) { size_t i; - if (!obj) + if (IS_ERR_OR_NULL(obj)) return; if (obj->clear_priv) @@ -7690,7 +7690,7 @@ int bpf_link__destroy(struct bpf_link *link) { int err = 0; - if (!link) + if (IS_ERR_OR_NULL(link)) return 0; if (!link->disconnected && link->detach) @@ -8502,7 +8502,7 @@ void perf_buffer__free(struct perf_buffer *pb) { int i; - if (!pb) + if (IS_ERR_OR_NULL(pb)) return; if (pb->cpu_bufs) { for (i = 0; i < pb->cpu_cnt; i++) { @@ -9379,8 +9379,7 @@ void bpf_object__detach_skeleton(struct bpf_object_skeleton *s) for (i = 0; i < s->prog_cnt; i++) { struct bpf_link **link = s->progs[i].link; - if (!IS_ERR_OR_NULL(*link)) - bpf_link__destroy(*link); + bpf_link__destroy(*link); *link = NULL; } } -- cgit v1.2.3 From 1acf8f90ea7ee59006d0474275922145ac291331 Mon Sep 17 00:00:00 2001 From: Jerry Crunchtime Date: Fri, 31 Jul 2020 17:08:01 +0200 Subject: libbpf: Fix register in PT_REGS MIPS macros The o32, n32 and n64 calling conventions require the return value to be stored in $v0 which maps to $2 register, i.e., the register 2. Fixes: c1932cd ("bpf: Add MIPS support to samples/bpf.") Signed-off-by: Jerry Crunchtime Signed-off-by: Daniel Borkmann Acked-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/43707d31-0210-e8f0-9226-1af140907641@web.de --- tools/lib/bpf/bpf_tracing.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'tools') diff --git a/tools/lib/bpf/bpf_tracing.h b/tools/lib/bpf/bpf_tracing.h index 58eceb884df3..eebf020cbe3e 100644 --- a/tools/lib/bpf/bpf_tracing.h +++ b/tools/lib/bpf/bpf_tracing.h @@ -215,7 +215,7 @@ struct pt_regs; #define PT_REGS_PARM5(x) ((x)->regs[8]) #define PT_REGS_RET(x) ((x)->regs[31]) #define PT_REGS_FP(x) ((x)->regs[30]) /* Works only with CONFIG_FRAME_POINTER */ -#define PT_REGS_RC(x) ((x)->regs[1]) +#define PT_REGS_RC(x) ((x)->regs[2]) #define PT_REGS_SP(x) ((x)->regs[29]) #define PT_REGS_IP(x) ((x)->cp0_epc) @@ -226,7 +226,7 @@ struct pt_regs; #define PT_REGS_PARM5_CORE(x) BPF_CORE_READ((x), regs[8]) #define PT_REGS_RET_CORE(x) BPF_CORE_READ((x), regs[31]) #define PT_REGS_FP_CORE(x) BPF_CORE_READ((x), regs[30]) -#define PT_REGS_RC_CORE(x) BPF_CORE_READ((x), regs[1]) +#define PT_REGS_RC_CORE(x) BPF_CORE_READ((x), regs[2]) #define PT_REGS_SP_CORE(x) BPF_CORE_READ((x), regs[29]) #define PT_REGS_IP_CORE(x) BPF_CORE_READ((x), cp0_epc) -- cgit v1.2.3 From 4939b2847d26c025e2e2118744226967f239a1ac Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Fri, 31 Jul 2020 15:09:14 -0700 Subject: bpf, selftests: Use single cgroup helpers for both test_sockmap/progs Nearly every user of cgroup helpers does the same sequence of API calls. So push these into a single helper cgroup_setup_and_join. The cases that do a bit of extra logic are test_progs which currently uses an env variable to decide if it needs to setup the cgroup environment or can use an existingi environment. And then tests that are doing cgroup tests themselves. We skip these cases for now. Signed-off-by: John Fastabend Signed-off-by: Alexei Starovoitov Acked-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/159623335418.30208.15807461815525100199.stgit@john-XPS-13-9370 --- tools/testing/selftests/bpf/cgroup_helpers.c | 23 ++++++++++++++++++++++ tools/testing/selftests/bpf/cgroup_helpers.h | 1 + tools/testing/selftests/bpf/get_cgroup_id_user.c | 14 ++----------- tools/testing/selftests/bpf/test_cgroup_storage.c | 17 +--------------- tools/testing/selftests/bpf/test_dev_cgroup.c | 15 ++------------ tools/testing/selftests/bpf/test_netcnt.c | 17 ++-------------- .../selftests/bpf/test_skb_cgroup_id_user.c | 8 +------- tools/testing/selftests/bpf/test_sock.c | 8 +------- tools/testing/selftests/bpf/test_sock_addr.c | 8 +------- tools/testing/selftests/bpf/test_sock_fields.c | 14 +++---------- tools/testing/selftests/bpf/test_socket_cookie.c | 8 +------- tools/testing/selftests/bpf/test_sockmap.c | 18 ++--------------- tools/testing/selftests/bpf/test_sysctl.c | 8 +------- tools/testing/selftests/bpf/test_tcpbpf_user.c | 8 +------- tools/testing/selftests/bpf/test_tcpnotify_user.c | 8 +------- 15 files changed, 43 insertions(+), 132 deletions(-) (limited to 'tools') diff --git a/tools/testing/selftests/bpf/cgroup_helpers.c b/tools/testing/selftests/bpf/cgroup_helpers.c index 0fb910df5387..033051717ba5 100644 --- a/tools/testing/selftests/bpf/cgroup_helpers.c +++ b/tools/testing/selftests/bpf/cgroup_helpers.c @@ -290,3 +290,26 @@ free_mem: free(fhp); return ret; } + +int cgroup_setup_and_join(const char *path) { + int cg_fd; + + if (setup_cgroup_environment()) { + fprintf(stderr, "Failed to setup cgroup environment\n"); + return -EINVAL; + } + + cg_fd = create_and_get_cgroup(path); + if (cg_fd < 0) { + fprintf(stderr, "Failed to create test cgroup\n"); + cleanup_cgroup_environment(); + return cg_fd; + } + + if (join_cgroup(path)) { + fprintf(stderr, "Failed to join cgroup\n"); + cleanup_cgroup_environment(); + return -EINVAL; + } + return cg_fd; +} diff --git a/tools/testing/selftests/bpf/cgroup_helpers.h b/tools/testing/selftests/bpf/cgroup_helpers.h index d64bb8957090..5fe3d88e4f0d 100644 --- a/tools/testing/selftests/bpf/cgroup_helpers.h +++ b/tools/testing/selftests/bpf/cgroup_helpers.h @@ -9,6 +9,7 @@ __FILE__, __LINE__, clean_errno(), ##__VA_ARGS__) +int cgroup_setup_and_join(const char *path); int create_and_get_cgroup(const char *path); int join_cgroup(const char *path); int setup_cgroup_environment(void); diff --git a/tools/testing/selftests/bpf/get_cgroup_id_user.c b/tools/testing/selftests/bpf/get_cgroup_id_user.c index e8da7b39158d..b8d6aef99db4 100644 --- a/tools/testing/selftests/bpf/get_cgroup_id_user.c +++ b/tools/testing/selftests/bpf/get_cgroup_id_user.c @@ -58,20 +58,10 @@ int main(int argc, char **argv) int exit_code = 1; char buf[256]; - err = setup_cgroup_environment(); - if (CHECK(err, "setup_cgroup_environment", "err %d errno %d\n", err, - errno)) + cgroup_fd = cgroup_setup_and_join(TEST_CGROUP); + if (CHECK(cgroup_fd < 0, "cgroup_setup_and_join", "err %d errno %d\n", cgroup_fd, errno)) return 1; - cgroup_fd = create_and_get_cgroup(TEST_CGROUP); - if (CHECK(cgroup_fd < 0, "create_and_get_cgroup", "err %d errno %d\n", - cgroup_fd, errno)) - goto cleanup_cgroup_env; - - err = join_cgroup(TEST_CGROUP); - if (CHECK(err, "join_cgroup", "err %d errno %d\n", err, errno)) - goto cleanup_cgroup_env; - err = bpf_prog_load(file, BPF_PROG_TYPE_TRACEPOINT, &obj, &prog_fd); if (CHECK(err, "bpf_prog_load", "err %d errno %d\n", err, errno)) goto cleanup_cgroup_env; diff --git a/tools/testing/selftests/bpf/test_cgroup_storage.c b/tools/testing/selftests/bpf/test_cgroup_storage.c index 655729004391..d946252a25bb 100644 --- a/tools/testing/selftests/bpf/test_cgroup_storage.c +++ b/tools/testing/selftests/bpf/test_cgroup_storage.c @@ -74,22 +74,7 @@ int main(int argc, char **argv) goto out; } - if (setup_cgroup_environment()) { - printf("Failed to setup cgroup environment\n"); - goto err; - } - - /* Create a cgroup, get fd, and join it */ - cgroup_fd = create_and_get_cgroup(TEST_CGROUP); - if (cgroup_fd < 0) { - printf("Failed to create test cgroup\n"); - goto err; - } - - if (join_cgroup(TEST_CGROUP)) { - printf("Failed to join cgroup\n"); - goto err; - } + cgroup_fd = cgroup_setup_and_join(TEST_CGROUP); /* Attach the bpf program */ if (bpf_prog_attach(prog_fd, cgroup_fd, BPF_CGROUP_INET_EGRESS, 0)) { diff --git a/tools/testing/selftests/bpf/test_dev_cgroup.c b/tools/testing/selftests/bpf/test_dev_cgroup.c index d850fb9076b5..804dddd97d4c 100644 --- a/tools/testing/selftests/bpf/test_dev_cgroup.c +++ b/tools/testing/selftests/bpf/test_dev_cgroup.c @@ -33,21 +33,10 @@ int main(int argc, char **argv) goto out; } - if (setup_cgroup_environment()) { - printf("Failed to load DEV_CGROUP program\n"); - goto err; - } - - /* Create a cgroup, get fd, and join it */ - cgroup_fd = create_and_get_cgroup(TEST_CGROUP); + cgroup_fd = cgroup_setup_and_join(TEST_CGROUP); if (cgroup_fd < 0) { printf("Failed to create test cgroup\n"); - goto err; - } - - if (join_cgroup(TEST_CGROUP)) { - printf("Failed to join cgroup\n"); - goto err; + goto out; } /* Attach bpf program */ diff --git a/tools/testing/selftests/bpf/test_netcnt.c b/tools/testing/selftests/bpf/test_netcnt.c index 7a68c9069639..a7b9a69f4fd5 100644 --- a/tools/testing/selftests/bpf/test_netcnt.c +++ b/tools/testing/selftests/bpf/test_netcnt.c @@ -58,22 +58,9 @@ int main(int argc, char **argv) goto out; } - if (setup_cgroup_environment()) { - printf("Failed to load bpf program\n"); - goto err; - } - - /* Create a cgroup, get fd, and join it */ - cgroup_fd = create_and_get_cgroup(TEST_CGROUP); - if (cgroup_fd < 0) { - printf("Failed to create test cgroup\n"); + cgroup_fd = cgroup_setup_and_join(TEST_CGROUP); + if (cgroup_fd < 0) goto err; - } - - if (join_cgroup(TEST_CGROUP)) { - printf("Failed to join cgroup\n"); - goto err; - } /* Attach bpf program */ if (bpf_prog_attach(prog_fd, cgroup_fd, BPF_CGROUP_INET_EGRESS, 0)) { diff --git a/tools/testing/selftests/bpf/test_skb_cgroup_id_user.c b/tools/testing/selftests/bpf/test_skb_cgroup_id_user.c index 356351c0ac28..4a64306728ab 100644 --- a/tools/testing/selftests/bpf/test_skb_cgroup_id_user.c +++ b/tools/testing/selftests/bpf/test_skb_cgroup_id_user.c @@ -160,16 +160,10 @@ int main(int argc, char **argv) exit(EXIT_FAILURE); } - if (setup_cgroup_environment()) - goto err; - - cgfd = create_and_get_cgroup(CGROUP_PATH); + cgfd = cgroup_setup_and_join(CGROUP_PATH); if (cgfd < 0) goto err; - if (join_cgroup(CGROUP_PATH)) - goto err; - if (send_packet(argv[1])) goto err; diff --git a/tools/testing/selftests/bpf/test_sock.c b/tools/testing/selftests/bpf/test_sock.c index 52bf14955797..9613f7538840 100644 --- a/tools/testing/selftests/bpf/test_sock.c +++ b/tools/testing/selftests/bpf/test_sock.c @@ -464,16 +464,10 @@ int main(int argc, char **argv) int cgfd = -1; int err = 0; - if (setup_cgroup_environment()) - goto err; - - cgfd = create_and_get_cgroup(CG_PATH); + cgfd = cgroup_setup_and_join(CG_PATH); if (cgfd < 0) goto err; - if (join_cgroup(CG_PATH)) - goto err; - if (run_tests(cgfd)) goto err; diff --git a/tools/testing/selftests/bpf/test_sock_addr.c b/tools/testing/selftests/bpf/test_sock_addr.c index 0358814c67dc..b8c72c1d9cf7 100644 --- a/tools/testing/selftests/bpf/test_sock_addr.c +++ b/tools/testing/selftests/bpf/test_sock_addr.c @@ -1638,16 +1638,10 @@ int main(int argc, char **argv) exit(err); } - if (setup_cgroup_environment()) - goto err; - - cgfd = create_and_get_cgroup(CG_PATH); + cgfd = cgroup_setup_and_join(CG_PATH); if (cgfd < 0) goto err; - if (join_cgroup(CG_PATH)) - goto err; - if (run_tests(cgfd)) goto err; diff --git a/tools/testing/selftests/bpf/test_sock_fields.c b/tools/testing/selftests/bpf/test_sock_fields.c index f0fc103261a4..6c9f269c396d 100644 --- a/tools/testing/selftests/bpf/test_sock_fields.c +++ b/tools/testing/selftests/bpf/test_sock_fields.c @@ -421,19 +421,11 @@ int main(int argc, char **argv) struct bpf_object *obj; struct bpf_map *map; - err = setup_cgroup_environment(); - CHECK(err, "setup_cgroup_environment()", "err:%d errno:%d", - err, errno); - - atexit(cleanup_cgroup_environment); - /* Create a cgroup, get fd, and join it */ - cgroup_fd = create_and_get_cgroup(TEST_CGROUP); - CHECK(cgroup_fd == -1, "create_and_get_cgroup()", + cgroup_fd = cgroup_setup_and_join(TEST_CGROUP); + CHECK(cgroup_fd < 0, "cgroup_setup_and_join()", "cgroup_fd:%d errno:%d", cgroup_fd, errno); - - err = join_cgroup(TEST_CGROUP); - CHECK(err, "join_cgroup", "err:%d errno:%d", err, errno); + atexit(cleanup_cgroup_environment); err = bpf_prog_load_xattr(&attr, &obj, &egress_fd); CHECK(err, "bpf_prog_load_xattr()", "err:%d", err); diff --git a/tools/testing/selftests/bpf/test_socket_cookie.c b/tools/testing/selftests/bpf/test_socket_cookie.c index 15653b0e26eb..154a8fd2a48d 100644 --- a/tools/testing/selftests/bpf/test_socket_cookie.c +++ b/tools/testing/selftests/bpf/test_socket_cookie.c @@ -191,16 +191,10 @@ int main(int argc, char **argv) int cgfd = -1; int err = 0; - if (setup_cgroup_environment()) - goto err; - - cgfd = create_and_get_cgroup(CG_PATH); + cgfd = cgroup_setup_and_join(CG_PATH); if (cgfd < 0) goto err; - if (join_cgroup(CG_PATH)) - goto err; - if (run_test(cgfd)) goto err; diff --git a/tools/testing/selftests/bpf/test_sockmap.c b/tools/testing/selftests/bpf/test_sockmap.c index 78789b27e573..9b6fb00dc7a0 100644 --- a/tools/testing/selftests/bpf/test_sockmap.c +++ b/tools/testing/selftests/bpf/test_sockmap.c @@ -1963,23 +1963,9 @@ int main(int argc, char **argv) } if (!cg_fd) { - if (setup_cgroup_environment()) { - fprintf(stderr, "ERROR: cgroup env failed\n"); - return -EINVAL; - } - - cg_fd = create_and_get_cgroup(CG_PATH); - if (cg_fd < 0) { - fprintf(stderr, - "ERROR: (%i) open cg path failed: %s\n", - cg_fd, strerror(errno)); + cg_fd = cgroup_setup_and_join(CG_PATH); + if (cg_fd < 0) return cg_fd; - } - - if (join_cgroup(CG_PATH)) { - fprintf(stderr, "ERROR: failed to join cgroup\n"); - return -EINVAL; - } cg_created = 1; } diff --git a/tools/testing/selftests/bpf/test_sysctl.c b/tools/testing/selftests/bpf/test_sysctl.c index d196e2a4a6e0..a20a919244c0 100644 --- a/tools/testing/selftests/bpf/test_sysctl.c +++ b/tools/testing/selftests/bpf/test_sysctl.c @@ -1619,16 +1619,10 @@ int main(int argc, char **argv) int cgfd = -1; int err = 0; - if (setup_cgroup_environment()) - goto err; - - cgfd = create_and_get_cgroup(CG_PATH); + cgfd = cgroup_setup_and_join(CG_PATH); if (cgfd < 0) goto err; - if (join_cgroup(CG_PATH)) - goto err; - if (run_tests(cgfd)) goto err; diff --git a/tools/testing/selftests/bpf/test_tcpbpf_user.c b/tools/testing/selftests/bpf/test_tcpbpf_user.c index 3ae127620463..74a9e49988b6 100644 --- a/tools/testing/selftests/bpf/test_tcpbpf_user.c +++ b/tools/testing/selftests/bpf/test_tcpbpf_user.c @@ -102,16 +102,10 @@ int main(int argc, char **argv) __u32 key = 0; int rv; - if (setup_cgroup_environment()) - goto err; - - cg_fd = create_and_get_cgroup(cg_path); + cg_fd = cgroup_setup_and_join(cg_path); if (cg_fd < 0) goto err; - if (join_cgroup(cg_path)) - goto err; - if (bpf_prog_load(file, BPF_PROG_TYPE_SOCK_OPS, &obj, &prog_fd)) { printf("FAILED: load_bpf_file failed for: %s\n", file); goto err; diff --git a/tools/testing/selftests/bpf/test_tcpnotify_user.c b/tools/testing/selftests/bpf/test_tcpnotify_user.c index f9765ddf0761..8549b31716ab 100644 --- a/tools/testing/selftests/bpf/test_tcpnotify_user.c +++ b/tools/testing/selftests/bpf/test_tcpnotify_user.c @@ -86,16 +86,10 @@ int main(int argc, char **argv) CPU_SET(0, &cpuset); pthread_setaffinity_np(pthread_self(), sizeof(cpu_set_t), &cpuset); - if (setup_cgroup_environment()) - goto err; - - cg_fd = create_and_get_cgroup(cg_path); + cg_fd = cgroup_setup_and_join(cg_path); if (cg_fd < 0) goto err; - if (join_cgroup(cg_path)) - goto err; - if (bpf_prog_load(file, BPF_PROG_TYPE_SOCK_OPS, &obj, &prog_fd)) { printf("FAILED: load_bpf_file failed for: %s\n", file); goto err; -- cgit v1.2.3 From 2e49527e52486dac910460b1b3f6fce6e21c6b48 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Fri, 31 Jul 2020 11:28:27 -0700 Subject: libbpf: Add bpf_link detach APIs Add low-level bpf_link_detach() API. Also add higher-level bpf_link__detach() one. Signed-off-by: Andrii Nakryiko Signed-off-by: Alexei Starovoitov Acked-by: Song Liu Acked-by: John Fastabend Link: https://lore.kernel.org/bpf/20200731182830.286260-3-andriin@fb.com --- tools/include/uapi/linux/bpf.h | 5 +++++ tools/lib/bpf/bpf.c | 10 ++++++++++ tools/lib/bpf/bpf.h | 2 ++ tools/lib/bpf/libbpf.c | 5 +++++ tools/lib/bpf/libbpf.h | 1 + tools/lib/bpf/libbpf.map | 2 ++ 6 files changed, 25 insertions(+) (limited to 'tools') diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index eb5e0c38eb2c..b134e679e9db 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -117,6 +117,7 @@ enum bpf_cmd { BPF_LINK_GET_NEXT_ID, BPF_ENABLE_STATS, BPF_ITER_CREATE, + BPF_LINK_DETACH, }; enum bpf_map_type { @@ -634,6 +635,10 @@ union bpf_attr { __u32 old_prog_fd; } link_update; + struct { + __u32 link_fd; + } link_detach; + struct { /* struct used by BPF_ENABLE_STATS command */ __u32 type; } enable_stats; diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c index e1bdf214f75f..eab14c97c15d 100644 --- a/tools/lib/bpf/bpf.c +++ b/tools/lib/bpf/bpf.c @@ -603,6 +603,16 @@ int bpf_link_create(int prog_fd, int target_fd, return sys_bpf(BPF_LINK_CREATE, &attr, sizeof(attr)); } +int bpf_link_detach(int link_fd) +{ + union bpf_attr attr; + + memset(&attr, 0, sizeof(attr)); + attr.link_detach.link_fd = link_fd; + + return sys_bpf(BPF_LINK_DETACH, &attr, sizeof(attr)); +} + int bpf_link_update(int link_fd, int new_prog_fd, const struct bpf_link_update_opts *opts) { diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h index 6d367e01d05e..28855fd5b5f4 100644 --- a/tools/lib/bpf/bpf.h +++ b/tools/lib/bpf/bpf.h @@ -178,6 +178,8 @@ LIBBPF_API int bpf_link_create(int prog_fd, int target_fd, enum bpf_attach_type attach_type, const struct bpf_link_create_opts *opts); +LIBBPF_API int bpf_link_detach(int link_fd); + struct bpf_link_update_opts { size_t sz; /* size of this struct for forward/backward compatibility */ __u32 flags; /* extra flags */ diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index b9f11f854985..7be04e45d29c 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -7748,6 +7748,11 @@ struct bpf_link *bpf_link__open(const char *path) return link; } +int bpf_link__detach(struct bpf_link *link) +{ + return bpf_link_detach(link->fd) ? -errno : 0; +} + int bpf_link__pin(struct bpf_link *link, const char *path) { int err; diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h index 9924385462ab..3ed1399bfbbc 100644 --- a/tools/lib/bpf/libbpf.h +++ b/tools/lib/bpf/libbpf.h @@ -229,6 +229,7 @@ LIBBPF_API int bpf_link__unpin(struct bpf_link *link); LIBBPF_API int bpf_link__update_program(struct bpf_link *link, struct bpf_program *prog); LIBBPF_API void bpf_link__disconnect(struct bpf_link *link); +LIBBPF_API int bpf_link__detach(struct bpf_link *link); LIBBPF_API int bpf_link__destroy(struct bpf_link *link); LIBBPF_API struct bpf_link * diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map index ca49a6a7e5b2..099863411f7d 100644 --- a/tools/lib/bpf/libbpf.map +++ b/tools/lib/bpf/libbpf.map @@ -273,6 +273,8 @@ LIBBPF_0.0.9 { LIBBPF_0.1.0 { global: + bpf_link__detach; + bpf_link_detach; bpf_map__ifindex; bpf_map__key_size; bpf_map__map_flags; -- cgit v1.2.3 From 90806ccc90bbd0150267a97ae4003269597a6a6c Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Fri, 31 Jul 2020 11:28:28 -0700 Subject: selftests/bpf: Add link detach tests for cgroup, netns, and xdp bpf_links Add bpf_link__detach() testing to selftests for cgroup, netns, and xdp bpf_links. Signed-off-by: Andrii Nakryiko Signed-off-by: Alexei Starovoitov Acked-by: Song Liu Acked-by: John Fastabend Link: https://lore.kernel.org/bpf/20200731182830.286260-4-andriin@fb.com --- .../testing/selftests/bpf/prog_tests/cgroup_link.c | 20 ++++++++- tools/testing/selftests/bpf/prog_tests/sk_lookup.c | 51 ++++++++++------------ tools/testing/selftests/bpf/prog_tests/xdp_link.c | 14 ++++++ tools/testing/selftests/bpf/testing_helpers.c | 14 ++++++ tools/testing/selftests/bpf/testing_helpers.h | 3 ++ 5 files changed, 73 insertions(+), 29 deletions(-) (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/cgroup_link.c b/tools/testing/selftests/bpf/prog_tests/cgroup_link.c index 6e04f8d1d15b..4d9b514b3fd9 100644 --- a/tools/testing/selftests/bpf/prog_tests/cgroup_link.c +++ b/tools/testing/selftests/bpf/prog_tests/cgroup_link.c @@ -2,6 +2,7 @@ #include #include "cgroup_helpers.h" +#include "testing_helpers.h" #include "test_cgroup_link.skel.h" static __u32 duration = 0; @@ -37,7 +38,8 @@ void test_cgroup_link(void) int last_cg = ARRAY_SIZE(cgs) - 1, cg_nr = ARRAY_SIZE(cgs); DECLARE_LIBBPF_OPTS(bpf_link_update_opts, link_upd_opts); struct bpf_link *links[ARRAY_SIZE(cgs)] = {}, *tmp_link; - __u32 prog_ids[ARRAY_SIZE(cgs)], prog_cnt = 0, attach_flags; + __u32 prog_ids[ARRAY_SIZE(cgs)], prog_cnt = 0, attach_flags, prog_id; + struct bpf_link_info info; int i = 0, err, prog_fd; bool detach_legacy = false; @@ -219,6 +221,22 @@ void test_cgroup_link(void) /* BPF programs should still get called */ ping_and_check(0, cg_nr); + prog_id = link_info_prog_id(links[0], &info); + CHECK(prog_id == 0, "link_info", "failed\n"); + CHECK(info.cgroup.cgroup_id == 0, "cgroup_id", "unexpected %llu\n", info.cgroup.cgroup_id); + + err = bpf_link__detach(links[0]); + if (CHECK(err, "link_detach", "failed %d\n", err)) + goto cleanup; + + /* cgroup_id should be zero in link_info */ + prog_id = link_info_prog_id(links[0], &info); + CHECK(prog_id == 0, "link_info", "failed\n"); + CHECK(info.cgroup.cgroup_id != 0, "cgroup_id", "unexpected %llu\n", info.cgroup.cgroup_id); + + /* First BPF program shouldn't be called anymore */ + ping_and_check(0, cg_nr - 1); + /* leave cgroup and remove them, don't detach programs */ cleanup_cgroup_environment(); diff --git a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c index 379da6f10ee9..c571584c00f5 100644 --- a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c +++ b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c @@ -34,6 +34,7 @@ #include "bpf_util.h" #include "cgroup_helpers.h" #include "network_helpers.h" +#include "testing_helpers.h" #include "test_sk_lookup.skel.h" /* External (address, port) pairs the client sends packets to. */ @@ -469,34 +470,10 @@ static int update_lookup_map(struct bpf_map *map, int index, int sock_fd) return 0; } -static __u32 link_info_prog_id(struct bpf_link *link) -{ - struct bpf_link_info info = {}; - __u32 info_len = sizeof(info); - int link_fd, err; - - link_fd = bpf_link__fd(link); - if (CHECK(link_fd < 0, "bpf_link__fd", "failed\n")) { - errno = -link_fd; - log_err("bpf_link__fd failed"); - return 0; - } - - err = bpf_obj_get_info_by_fd(link_fd, &info, &info_len); - if (CHECK(err, "bpf_obj_get_info_by_fd", "failed\n")) { - log_err("bpf_obj_get_info_by_fd"); - return 0; - } - if (CHECK(info_len != sizeof(info), "bpf_obj_get_info_by_fd", - "unexpected info len %u\n", info_len)) - return 0; - - return info.prog_id; -} - static void query_lookup_prog(struct test_sk_lookup *skel) { struct bpf_link *link[3] = {}; + struct bpf_link_info info; __u32 attach_flags = 0; __u32 prog_ids[3] = {}; __u32 prog_cnt = 3; @@ -534,18 +511,36 @@ static void query_lookup_prog(struct test_sk_lookup *skel) if (CHECK(prog_cnt != 3, "bpf_prog_query", "wrong program count on query: %u", prog_cnt)) goto detach; - prog_id = link_info_prog_id(link[0]); + prog_id = link_info_prog_id(link[0], &info); CHECK(prog_ids[0] != prog_id, "bpf_prog_query", "invalid program #0 id on query: %u != %u\n", prog_ids[0], prog_id); - prog_id = link_info_prog_id(link[1]); + CHECK(info.netns.netns_ino == 0, "netns_ino", + "unexpected netns_ino: %u\n", info.netns.netns_ino); + prog_id = link_info_prog_id(link[1], &info); CHECK(prog_ids[1] != prog_id, "bpf_prog_query", "invalid program #1 id on query: %u != %u\n", prog_ids[1], prog_id); - prog_id = link_info_prog_id(link[2]); + CHECK(info.netns.netns_ino == 0, "netns_ino", + "unexpected netns_ino: %u\n", info.netns.netns_ino); + prog_id = link_info_prog_id(link[2], &info); CHECK(prog_ids[2] != prog_id, "bpf_prog_query", "invalid program #2 id on query: %u != %u\n", prog_ids[2], prog_id); + CHECK(info.netns.netns_ino == 0, "netns_ino", + "unexpected netns_ino: %u\n", info.netns.netns_ino); + + err = bpf_link__detach(link[0]); + if (CHECK(err, "link_detach", "failed %d\n", err)) + goto detach; + + /* prog id is still there, but netns_ino is zeroed out */ + prog_id = link_info_prog_id(link[0], &info); + CHECK(prog_ids[0] != prog_id, "bpf_prog_query", + "invalid program #0 id on query: %u != %u\n", + prog_ids[0], prog_id); + CHECK(info.netns.netns_ino != 0, "netns_ino", + "unexpected netns_ino: %u\n", info.netns.netns_ino); detach: if (link[2]) diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_link.c b/tools/testing/selftests/bpf/prog_tests/xdp_link.c index 52cba6795d40..6f814999b395 100644 --- a/tools/testing/selftests/bpf/prog_tests/xdp_link.c +++ b/tools/testing/selftests/bpf/prog_tests/xdp_link.c @@ -131,6 +131,20 @@ void test_xdp_link(void) CHECK(link_info.xdp.ifindex != IFINDEX_LO, "link_ifindex", "got %u != exp %u\n", link_info.xdp.ifindex, IFINDEX_LO); + err = bpf_link__detach(link); + if (CHECK(err, "link_detach", "failed %d\n", err)) + goto cleanup; + + memset(&link_info, 0, sizeof(link_info)); + err = bpf_obj_get_info_by_fd(bpf_link__fd(link), &link_info, &link_info_len); + if (CHECK(err, "link_info", "failed: %d\n", err)) + goto cleanup; + CHECK(link_info.prog_id != id1, "link_prog_id", + "got %u != exp %u\n", link_info.prog_id, id1); + /* ifindex should be zeroed out */ + CHECK(link_info.xdp.ifindex != 0, "link_ifindex", + "got %u != exp %u\n", link_info.xdp.ifindex, 0); + cleanup: test_xdp_link__destroy(skel1); test_xdp_link__destroy(skel2); diff --git a/tools/testing/selftests/bpf/testing_helpers.c b/tools/testing/selftests/bpf/testing_helpers.c index 0af6337a8962..800d503e5cb4 100644 --- a/tools/testing/selftests/bpf/testing_helpers.c +++ b/tools/testing/selftests/bpf/testing_helpers.c @@ -64,3 +64,17 @@ int parse_num_list(const char *s, bool **num_set, int *num_set_len) return 0; } + +__u32 link_info_prog_id(const struct bpf_link *link, struct bpf_link_info *info) +{ + __u32 info_len = sizeof(*info); + int err; + + memset(info, 0, sizeof(*info)); + err = bpf_obj_get_info_by_fd(bpf_link__fd(link), info, &info_len); + if (err) { + printf("failed to get link info: %d\n", -errno); + return 0; + } + return info->prog_id; +} diff --git a/tools/testing/selftests/bpf/testing_helpers.h b/tools/testing/selftests/bpf/testing_helpers.h index 923b51762759..d4f8e749611b 100644 --- a/tools/testing/selftests/bpf/testing_helpers.h +++ b/tools/testing/selftests/bpf/testing_helpers.h @@ -1,5 +1,8 @@ /* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */ /* Copyright (C) 2020 Facebook, Inc. */ #include +#include +#include int parse_num_list(const char *s, bool **set, int *set_len); +__u32 link_info_prog_id(const struct bpf_link *link, struct bpf_link_info *info); -- cgit v1.2.3 From 0e8c7c07f090668566db2030a027a360ffd00938 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Fri, 31 Jul 2020 11:28:29 -0700 Subject: tools/bpftool: Add `link detach` subcommand Add ability to force-detach BPF link. Also add missing error message, if specified link ID is wrong. Signed-off-by: Andrii Nakryiko Signed-off-by: Alexei Starovoitov Acked-by: Song Liu Acked-by: John Fastabend Link: https://lore.kernel.org/bpf/20200731182830.286260-5-andriin@fb.com --- tools/bpf/bpftool/link.c | 37 ++++++++++++++++++++++++++++++++++++- 1 file changed, 36 insertions(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/bpf/bpftool/link.c b/tools/bpf/bpftool/link.c index 326b8fdf0243..1b793759170e 100644 --- a/tools/bpf/bpftool/link.c +++ b/tools/bpf/bpftool/link.c @@ -22,6 +22,8 @@ static const char * const link_type_name[] = { static int link_parse_fd(int *argc, char ***argv) { + int fd; + if (is_prefix(**argv, "id")) { unsigned int id; char *endptr; @@ -35,7 +37,10 @@ static int link_parse_fd(int *argc, char ***argv) } NEXT_ARGP(); - return bpf_link_get_fd_by_id(id); + fd = bpf_link_get_fd_by_id(id); + if (fd < 0) + p_err("failed to get link with ID %d: %s", id, strerror(errno)); + return fd; } else if (is_prefix(**argv, "pinned")) { char *path; @@ -316,6 +321,34 @@ static int do_pin(int argc, char **argv) return err; } +static int do_detach(int argc, char **argv) +{ + int err, fd; + + if (argc != 2) { + p_err("link specifier is invalid or missing\n"); + return 1; + } + + fd = link_parse_fd(&argc, &argv); + if (fd < 0) + return 1; + + err = bpf_link_detach(fd); + if (err) + err = -errno; + close(fd); + if (err) { + p_err("failed link detach: %s", strerror(-err)); + return 1; + } + + if (json_output) + jsonw_null(json_wtr); + + return 0; +} + static int do_help(int argc, char **argv) { if (json_output) { @@ -326,6 +359,7 @@ static int do_help(int argc, char **argv) fprintf(stderr, "Usage: %1$s %2$s { show | list } [LINK]\n" " %1$s %2$s pin LINK FILE\n" + " %1$s %2$s detach LINK\n" " %1$s %2$s help\n" "\n" " " HELP_SPEC_LINK "\n" @@ -341,6 +375,7 @@ static const struct cmd cmds[] = { { "list", do_show }, { "help", do_help }, { "pin", do_pin }, + { "detach", do_detach }, { 0 } }; -- cgit v1.2.3 From e85f99aa7760e74bb5a7e8515948f99c264a275f Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Fri, 31 Jul 2020 11:28:30 -0700 Subject: tools/bpftool: Add documentation and bash-completion for `link detach` Add info on link detach sub-command to man page. Add detach to bash-completion as well. Signed-off-by: Andrii Nakryiko Signed-off-by: Alexei Starovoitov Acked-by: Song Liu Acked-by: John Fastabend Date: Fri, 31 Jul 2020 13:49:57 -0700 Subject: selftests/bpf: Fix spurious test failures in core_retro selftest core_retro selftest uses BPF program that's triggered on sys_enter system-wide, but has no protection from some unrelated process doing syscall while selftest is running. This leads to occasional test failures with unexpected PIDs being returned. Fix that by filtering out all processes that are not test_progs process. Fixes: fcda189a5133 ("selftests/bpf: Add test relying only on CO-RE and no recent kernel features") Signed-off-by: Andrii Nakryiko Signed-off-by: Alexei Starovoitov Acked-by: John Fastabend Link: https://lore.kernel.org/bpf/20200731204957.2047119-1-andriin@fb.com --- tools/testing/selftests/bpf/prog_tests/core_retro.c | 8 ++++++-- tools/testing/selftests/bpf/progs/test_core_retro.c | 13 +++++++++++++ 2 files changed, 19 insertions(+), 2 deletions(-) (limited to 'tools') diff --git a/tools/testing/selftests/bpf/prog_tests/core_retro.c b/tools/testing/selftests/bpf/prog_tests/core_retro.c index 78e30d3a23d5..6acb0e94d4d7 100644 --- a/tools/testing/selftests/bpf/prog_tests/core_retro.c +++ b/tools/testing/selftests/bpf/prog_tests/core_retro.c @@ -6,7 +6,7 @@ void test_core_retro(void) { - int err, zero = 0, res, duration = 0; + int err, zero = 0, res, duration = 0, my_pid = getpid(); struct test_core_retro *skel; /* load program */ @@ -14,6 +14,10 @@ void test_core_retro(void) if (CHECK(!skel, "skel_load", "skeleton open/load failed\n")) goto out_close; + err = bpf_map_update_elem(bpf_map__fd(skel->maps.exp_tgid_map), &zero, &my_pid, 0); + if (CHECK(err, "map_update", "failed to set expected PID: %d\n", errno)) + goto out_close; + /* attach probe */ err = test_core_retro__attach(skel); if (CHECK(err, "attach_kprobe", "err %d\n", err)) @@ -26,7 +30,7 @@ void test_core_retro(void) if (CHECK(err, "map_lookup", "failed to lookup result: %d\n", errno)) goto out_close; - CHECK(res != getpid(), "pid_check", "got %d != exp %d\n", res, getpid()); + CHECK(res != my_pid, "pid_check", "got %d != exp %d\n", res, my_pid); out_close: test_core_retro__destroy(skel); diff --git a/tools/testing/selftests/bpf/progs/test_core_retro.c b/tools/testing/selftests/bpf/progs/test_core_retro.c index 75c60c3c29cf..20861ec2f674 100644 --- a/tools/testing/selftests/bpf/progs/test_core_retro.c +++ b/tools/testing/selftests/bpf/progs/test_core_retro.c @@ -8,6 +8,13 @@ struct task_struct { int tgid; } __attribute__((preserve_access_index)); +struct { + __uint(type, BPF_MAP_TYPE_ARRAY); + __uint(max_entries, 1); + __type(key, int); + __type(value, int); +} exp_tgid_map SEC(".maps"); + struct { __uint(type, BPF_MAP_TYPE_ARRAY); __uint(max_entries, 1); @@ -21,6 +28,12 @@ int handle_sys_enter(void *ctx) struct task_struct *task = (void *)bpf_get_current_task(); int tgid = BPF_CORE_READ(task, tgid); int zero = 0; + int real_tgid = bpf_get_current_pid_tgid() >> 32; + int *exp_tgid = bpf_map_lookup_elem(&exp_tgid_map, &zero); + + /* only pass through sys_enters from test process */ + if (!exp_tgid || *exp_tgid != real_tgid) + return 0; bpf_map_update_elem(&results, &zero, &tgid, 0); -- cgit v1.2.3 From a278f3d8191228212c553a5d4303fa603214b717 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Thu, 30 Jul 2020 19:42:44 -0700 Subject: tools, build: Propagate build failures from tools/build/Makefile.build The '&&' command seems to have a bad effect when $(cmd_$(1)) exits with non-zero effect: the command failure is masked (despite `set -e`) and all but the first command of $(dep-cmd) is executed (successfully, as they are mostly printfs), thus overall returning 0 in the end. This means in practice that despite compilation errors, tools's build Makefile will return success. We see this very reliably with libbpf's Makefile, which doesn't get compilation error propagated properly. This in turns causes issues with selftests build, as well as bpftool and other projects that rely on building libbpf. The fix is simple: don't use &&. Given `set -e`, we don't need to chain commands with &&. The shell will exit on first failure, giving desired behavior and propagating error properly. Fixes: 275e2d95591e ("tools build: Move dependency copy into function") Signed-off-by: Andrii Nakryiko Signed-off-by: Daniel Borkmann Acked-by: Jiri Olsa Link: https://lore.kernel.org/bpf/20200731024244.872574-1-andriin@fb.com --- tools/build/Build.include | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/build/Build.include b/tools/build/Build.include index 9ec01f4454f9..585486e40995 100644 --- a/tools/build/Build.include +++ b/tools/build/Build.include @@ -74,7 +74,8 @@ dep-cmd = $(if $(wildcard $(fixdep)), # dependencies in the cmd file if_changed_dep = $(if $(strip $(any-prereq) $(arg-check)), \ @set -e; \ - $(echo-cmd) $(cmd_$(1)) && $(dep-cmd)) + $(echo-cmd) $(cmd_$(1)); \ + $(dep-cmd)) # if_changed - execute command if any prerequisite is newer than # target, or command line has changed -- cgit v1.2.3 From 041549b7b2c7811ec40e705c439211f00ade2dda Mon Sep 17 00:00:00 2001 From: Tianjia Zhang Date: Sun, 2 Aug 2020 19:15:40 +0800 Subject: tools, bpftool: Fix wrong return value in do_dump() In case of btf_id does not exist, a negative error code -ENOENT should be returned. Fixes: c93cc69004df3 ("bpftool: add ability to dump BTF types") Signed-off-by: Tianjia Zhang Signed-off-by: Daniel Borkmann Reviewed-by: Tobias Klauser Acked-by: Andrii Nakryiko Acked-by: John Fastabend Link: https://lore.kernel.org/bpf/20200802111540.5384-1-tianjia.zhang@linux.alibaba.com --- tools/bpf/bpftool/btf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/bpf/bpftool/btf.c b/tools/bpf/bpftool/btf.c index fc9bc7a23db6..37c091308714 100644 --- a/tools/bpf/bpftool/btf.c +++ b/tools/bpf/bpftool/btf.c @@ -596,7 +596,7 @@ static int do_dump(int argc, char **argv) goto done; } if (!btf) { - err = ENOENT; + err = -ENOENT; p_err("can't find btf with ID (%u)", btf_id); goto done; } -- cgit v1.2.3 From 94a1fedd63edb672933bef44ca9213937e377c05 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Sat, 1 Aug 2020 18:32:17 -0700 Subject: libbpf: Add btf__parse_raw() and generic btf__parse() APIs Add public APIs to parse BTF from raw data file (e.g., /sys/kernel/btf/vmlinux), as well as generic btf__parse(), which will try to determine correct format, currently either raw or ELF. Signed-off-by: Andrii Nakryiko Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20200802013219.864880-2-andriin@fb.com --- tools/lib/bpf/btf.c | 114 ++++++++++++++++++++++++++++++++--------------- tools/lib/bpf/btf.h | 5 ++- tools/lib/bpf/libbpf.map | 2 + 3 files changed, 83 insertions(+), 38 deletions(-) (limited to 'tools') diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c index ded5b29965f9..856b09a04563 100644 --- a/tools/lib/bpf/btf.c +++ b/tools/lib/bpf/btf.c @@ -562,6 +562,83 @@ done: return btf; } +struct btf *btf__parse_raw(const char *path) +{ + void *data = NULL; + struct btf *btf; + FILE *f = NULL; + __u16 magic; + int err = 0; + long sz; + + f = fopen(path, "rb"); + if (!f) { + err = -errno; + goto err_out; + } + + /* check BTF magic */ + if (fread(&magic, 1, sizeof(magic), f) < sizeof(magic)) { + err = -EIO; + goto err_out; + } + if (magic != BTF_MAGIC) { + /* definitely not a raw BTF */ + err = -EPROTO; + goto err_out; + } + + /* get file size */ + if (fseek(f, 0, SEEK_END)) { + err = -errno; + goto err_out; + } + sz = ftell(f); + if (sz < 0) { + err = -errno; + goto err_out; + } + /* rewind to the start */ + if (fseek(f, 0, SEEK_SET)) { + err = -errno; + goto err_out; + } + + /* pre-alloc memory and read all of BTF data */ + data = malloc(sz); + if (!data) { + err = -ENOMEM; + goto err_out; + } + if (fread(data, 1, sz, f) < sz) { + err = -EIO; + goto err_out; + } + + /* finally parse BTF data */ + btf = btf__new(data, sz); + +err_out: + free(data); + if (f) + fclose(f); + return err ? ERR_PTR(err) : btf; +} + +struct btf *btf__parse(const char *path, struct btf_ext **btf_ext) +{ + struct btf *btf; + + if (btf_ext) + *btf_ext = NULL; + + btf = btf__parse_raw(path); + if (!IS_ERR(btf) || PTR_ERR(btf) != -EPROTO) + return btf; + + return btf__parse_elf(path, btf_ext); +} + static int compare_vsi_off(const void *_a, const void *_b) { const struct btf_var_secinfo *a = _a; @@ -2951,41 +3028,6 @@ static int btf_dedup_remap_types(struct btf_dedup *d) return 0; } -static struct btf *btf_load_raw(const char *path) -{ - struct btf *btf; - size_t read_cnt; - struct stat st; - void *data; - FILE *f; - - if (stat(path, &st)) - return ERR_PTR(-errno); - - data = malloc(st.st_size); - if (!data) - return ERR_PTR(-ENOMEM); - - f = fopen(path, "rb"); - if (!f) { - btf = ERR_PTR(-errno); - goto cleanup; - } - - read_cnt = fread(data, 1, st.st_size, f); - fclose(f); - if (read_cnt < st.st_size) { - btf = ERR_PTR(-EBADF); - goto cleanup; - } - - btf = btf__new(data, read_cnt); - -cleanup: - free(data); - return btf; -} - /* * Probe few well-known locations for vmlinux kernel image and try to load BTF * data out of it to use for target BTF. @@ -3021,7 +3063,7 @@ struct btf *libbpf_find_kernel_btf(void) continue; if (locations[i].raw_btf) - btf = btf_load_raw(path); + btf = btf__parse_raw(path); else btf = btf__parse_elf(path, NULL); diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h index 491c7b41ffdc..f4a1a1d2b9a3 100644 --- a/tools/lib/bpf/btf.h +++ b/tools/lib/bpf/btf.h @@ -64,8 +64,9 @@ struct btf_ext_header { LIBBPF_API void btf__free(struct btf *btf); LIBBPF_API struct btf *btf__new(const void *data, __u32 size); -LIBBPF_API struct btf *btf__parse_elf(const char *path, - struct btf_ext **btf_ext); +LIBBPF_API struct btf *btf__parse(const char *path, struct btf_ext **btf_ext); +LIBBPF_API struct btf *btf__parse_elf(const char *path, struct btf_ext **btf_ext); +LIBBPF_API struct btf *btf__parse_raw(const char *path); LIBBPF_API int btf__finalize_data(struct bpf_object *obj, struct btf *btf); LIBBPF_API int btf__load(struct btf *btf); LIBBPF_API __s32 btf__find_by_name(const struct btf *btf, diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map index 099863411f7d..0c4722bfdd0a 100644 --- a/tools/lib/bpf/libbpf.map +++ b/tools/lib/bpf/libbpf.map @@ -293,5 +293,7 @@ LIBBPF_0.1.0 { bpf_program__is_sk_lookup; bpf_program__set_autoload; bpf_program__set_sk_lookup; + btf__parse; + btf__parse_raw; btf__set_fd; } LIBBPF_0.0.9; -- cgit v1.2.3 From 8526df04570f5698c97ac661ad1f2f35293557a7 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Sat, 1 Aug 2020 18:32:18 -0700 Subject: tools/bpftool: Use libbpf's btf__parse() API for parsing BTF from file Use generic libbpf API to parse BTF data from file, instead of re-implementing it in bpftool. Signed-off-by: Andrii Nakryiko Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20200802013219.864880-3-andriin@fb.com --- tools/bpf/bpftool/btf.c | 54 +------------------------------------------------ 1 file changed, 1 insertion(+), 53 deletions(-) (limited to 'tools') diff --git a/tools/bpf/bpftool/btf.c b/tools/bpf/bpftool/btf.c index 37c091308714..8ab142ff5eac 100644 --- a/tools/bpf/bpftool/btf.c +++ b/tools/bpf/bpftool/btf.c @@ -422,54 +422,6 @@ done: return err; } -static struct btf *btf__parse_raw(const char *file) -{ - struct btf *btf; - struct stat st; - __u8 *buf; - FILE *f; - - if (stat(file, &st)) - return NULL; - - f = fopen(file, "rb"); - if (!f) - return NULL; - - buf = malloc(st.st_size); - if (!buf) { - btf = ERR_PTR(-ENOMEM); - goto exit_close; - } - - if ((size_t) st.st_size != fread(buf, 1, st.st_size, f)) { - btf = ERR_PTR(-EINVAL); - goto exit_free; - } - - btf = btf__new(buf, st.st_size); - -exit_free: - free(buf); -exit_close: - fclose(f); - return btf; -} - -static bool is_btf_raw(const char *file) -{ - __u16 magic = 0; - int fd, nb_read; - - fd = open(file, O_RDONLY); - if (fd < 0) - return false; - - nb_read = read(fd, &magic, sizeof(magic)); - close(fd); - return nb_read == sizeof(magic) && magic == BTF_MAGIC; -} - static int do_dump(int argc, char **argv) { struct btf *btf = NULL; @@ -547,11 +499,7 @@ static int do_dump(int argc, char **argv) } NEXT_ARG(); } else if (is_prefix(src, "file")) { - if (is_btf_raw(*argv)) - btf = btf__parse_raw(*argv); - else - btf = btf__parse_elf(*argv, NULL); - + btf = btf__parse(*argv, NULL); if (IS_ERR(btf)) { err = -PTR_ERR(btf); btf = NULL; -- cgit v1.2.3 From f86ca3cffef153555a3f4755b3a44881d962754f Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Sat, 1 Aug 2020 18:32:19 -0700 Subject: tools/resolve_btfids: Use libbpf's btf__parse() API Instead of re-implementing generic BTF parsing logic, use libbpf's API. Also add .gitignore for resolve_btfids's build artifacts. Signed-off-by: Andrii Nakryiko Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20200802013219.864880-4-andriin@fb.com --- tools/bpf/resolve_btfids/.gitignore | 4 +++ tools/bpf/resolve_btfids/main.c | 58 +------------------------------------ 2 files changed, 5 insertions(+), 57 deletions(-) create mode 100644 tools/bpf/resolve_btfids/.gitignore (limited to 'tools') diff --git a/tools/bpf/resolve_btfids/.gitignore b/tools/bpf/resolve_btfids/.gitignore new file mode 100644 index 000000000000..a026df7dc280 --- /dev/null +++ b/tools/bpf/resolve_btfids/.gitignore @@ -0,0 +1,4 @@ +/FEATURE-DUMP.libbpf +/bpf_helper_defs.h +/fixdep +/resolve_btfids diff --git a/tools/bpf/resolve_btfids/main.c b/tools/bpf/resolve_btfids/main.c index 6956b6350cad..52d883325a23 100644 --- a/tools/bpf/resolve_btfids/main.c +++ b/tools/bpf/resolve_btfids/main.c @@ -403,62 +403,6 @@ static int symbols_collect(struct object *obj) return 0; } -static struct btf *btf__parse_raw(const char *file) -{ - struct btf *btf; - struct stat st; - __u8 *buf; - FILE *f; - - if (stat(file, &st)) - return NULL; - - f = fopen(file, "rb"); - if (!f) - return NULL; - - buf = malloc(st.st_size); - if (!buf) { - btf = ERR_PTR(-ENOMEM); - goto exit_close; - } - - if ((size_t) st.st_size != fread(buf, 1, st.st_size, f)) { - btf = ERR_PTR(-EINVAL); - goto exit_free; - } - - btf = btf__new(buf, st.st_size); - -exit_free: - free(buf); -exit_close: - fclose(f); - return btf; -} - -static bool is_btf_raw(const char *file) -{ - __u16 magic = 0; - int fd, nb_read; - - fd = open(file, O_RDONLY); - if (fd < 0) - return false; - - nb_read = read(fd, &magic, sizeof(magic)); - close(fd); - return nb_read == sizeof(magic) && magic == BTF_MAGIC; -} - -static struct btf *btf_open(const char *path) -{ - if (is_btf_raw(path)) - return btf__parse_raw(path); - else - return btf__parse_elf(path, NULL); -} - static int symbols_resolve(struct object *obj) { int nr_typedefs = obj->nr_typedefs; @@ -469,7 +413,7 @@ static int symbols_resolve(struct object *obj) struct btf *btf; __u32 nr; - btf = btf_open(obj->btf ?: obj->path); + btf = btf__parse(obj->btf ?: obj->path, NULL); err = libbpf_get_error(btf); if (err) { pr_err("FAILED: load BTF from %s: %s", -- cgit v1.2.3 From 21594c44083c375697d418729c4b2e4522cf9f70 Mon Sep 17 00:00:00 2001 From: Dmitry Yakunin Date: Mon, 3 Aug 2020 12:05:45 +0300 Subject: bpf: Allow to specify ifindex for skb in bpf_prog_test_run_skb Now skb->dev is unconditionally set to the loopback device in current net namespace. But if we want to test bpf program which contains code branch based on ifindex condition (eg filters out localhost packets) it is useful to allow specifying of ifindex from userspace. This patch adds such option through ctx_in (__sk_buff) parameter. Signed-off-by: Dmitry Yakunin Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20200803090545.82046-3-zeil@yandex-team.ru --- net/bpf/test_run.c | 22 ++++++++++++++++++++-- tools/testing/selftests/bpf/prog_tests/skb_ctx.c | 5 +++++ 2 files changed, 25 insertions(+), 2 deletions(-) (limited to 'tools') diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c index 736a5964fa95..99eb8c6c0fbc 100644 --- a/net/bpf/test_run.c +++ b/net/bpf/test_run.c @@ -327,6 +327,12 @@ static int convert___skb_to_skb(struct sk_buff *skb, struct __sk_buff *__skb) /* priority is allowed */ if (!range_is_zero(__skb, offsetofend(struct __sk_buff, priority), + offsetof(struct __sk_buff, ifindex))) + return -EINVAL; + + /* ifindex is allowed */ + + if (!range_is_zero(__skb, offsetofend(struct __sk_buff, ifindex), offsetof(struct __sk_buff, cb))) return -EINVAL; @@ -381,6 +387,7 @@ static void convert_skb_to___skb(struct sk_buff *skb, struct __sk_buff *__skb) __skb->mark = skb->mark; __skb->priority = skb->priority; + __skb->ifindex = skb->dev->ifindex; __skb->tstamp = skb->tstamp; memcpy(__skb->cb, &cb->data, QDISC_CB_PRIV_LEN); __skb->wire_len = cb->pkt_len; @@ -391,6 +398,8 @@ int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr, union bpf_attr __user *uattr) { bool is_l2 = false, is_direct_pkt_access = false; + struct net *net = current->nsproxy->net_ns; + struct net_device *dev = net->loopback_dev; u32 size = kattr->test.data_size_in; u32 repeat = kattr->test.repeat; struct __sk_buff *ctx = NULL; @@ -432,7 +441,7 @@ int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr, kfree(ctx); return -ENOMEM; } - sock_net_set(sk, current->nsproxy->net_ns); + sock_net_set(sk, net); sock_init_data(NULL, sk); skb = build_skb(data, 0); @@ -446,7 +455,14 @@ int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr, skb_reserve(skb, NET_SKB_PAD + NET_IP_ALIGN); __skb_put(skb, size); - skb->protocol = eth_type_trans(skb, current->nsproxy->net_ns->loopback_dev); + if (ctx && ctx->ifindex > 1) { + dev = dev_get_by_index(net, ctx->ifindex); + if (!dev) { + ret = -ENODEV; + goto out; + } + } + skb->protocol = eth_type_trans(skb, dev); skb_reset_network_header(skb); switch (skb->protocol) { @@ -502,6 +518,8 @@ int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr, ret = bpf_ctx_finish(kattr, uattr, ctx, sizeof(struct __sk_buff)); out: + if (dev && dev != net->loopback_dev) + dev_put(dev); kfree_skb(skb); bpf_sk_storage_free(sk); kfree(sk); diff --git a/tools/testing/selftests/bpf/prog_tests/skb_ctx.c b/tools/testing/selftests/bpf/prog_tests/skb_ctx.c index 7021b92af313..25de86af2d03 100644 --- a/tools/testing/selftests/bpf/prog_tests/skb_ctx.c +++ b/tools/testing/selftests/bpf/prog_tests/skb_ctx.c @@ -11,6 +11,7 @@ void test_skb_ctx(void) .cb[3] = 4, .cb[4] = 5, .priority = 6, + .ifindex = 1, .tstamp = 7, .wire_len = 100, .gso_segs = 8, @@ -92,6 +93,10 @@ void test_skb_ctx(void) "ctx_out_priority", "skb->priority == %d, expected %d\n", skb.priority, 7); + CHECK_ATTR(skb.ifindex != 1, + "ctx_out_ifindex", + "skb->ifindex == %d, expected %d\n", + skb.ifindex, 1); CHECK_ATTR(skb.tstamp != 8, "ctx_out_tstamp", "skb->tstamp == %lld, expected %d\n", -- cgit v1.2.3