libbpf: Add opts-based attach/detach/query API for tcx

Extend libbpf attach opts and add a new detach opts API so this can be used
to add/remove fd-based tcx BPF programs. The old-style bpf_prog_detach() and
bpf_prog_detach2() APIs are refactored to reuse the new bpf_prog_detach_opts()
internally.

The bpf_prog_query_opts() API got extended to be able to handle the new
link_ids, link_attach_flags and revision fields.

For concrete usage examples, see the extensive selftests that have been
developed as part of this series.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20230719140858.13224-4-daniel@iogearbox.netSigned-off-by: Alexei Starovoitov <ast@kernel.org>

tools/lib/bpf/bpf.c

@@ -629,57 +629,91 @@ int bpf_prog_attach(int prog_fd, int target_fd, enum bpf_attach_type type,
 	return bpf_prog_attach_opts(prog_fd, target_fd, type, &opts);
 }
 
-int bpf_prog_attach_opts(int prog_fd, int target_fd,
-			  enum bpf_attach_type type,
+int bpf_prog_attach_opts(int prog_fd, int target, enum bpf_attach_type type,
 			 const struct bpf_prog_attach_opts *opts)
 {
-	const size_t attr_sz = offsetofend(union bpf_attr, replace_bpf_fd);
+	const size_t attr_sz = offsetofend(union bpf_attr, expected_revision);
+	__u32 relative_id, flags;
+	int ret, relative_fd;
 	union bpf_attr attr;
-	int ret;
 
 	if (!OPTS_VALID(opts, bpf_prog_attach_opts))
 		return libbpf_err(-EINVAL);
 
+	relative_id = OPTS_GET(opts, relative_id, 0);
+	relative_fd = OPTS_GET(opts, relative_fd, 0);
+	flags = OPTS_GET(opts, flags, 0);
+
+	/* validate we don't have unexpected combinations of non-zero fields */
+	if (relative_fd && relative_id)
+		return libbpf_err(-EINVAL);
+
 	memset(&attr, 0, attr_sz);
-	attr.target_fd	   = target_fd;
+	attr.target_fd		= target;
 	attr.attach_bpf_fd	= prog_fd;
 	attr.attach_type	= type;
-	attr.attach_flags  = OPTS_GET(opts, flags, 0);
-	attr.replace_bpf_fd = OPTS_GET(opts, replace_prog_fd, 0);
+	attr.replace_bpf_fd	= OPTS_GET(opts, replace_fd, 0);
+	attr.expected_revision	= OPTS_GET(opts, expected_revision, 0);
+
+	if (relative_id) {
+		attr.attach_flags = flags | BPF_F_ID;
+		attr.relative_id  = relative_id;
+	} else {
+		attr.attach_flags = flags;
+		attr.relative_fd  = relative_fd;
+	}
 
 	ret = sys_bpf(BPF_PROG_ATTACH, &attr, attr_sz);
 	return libbpf_err_errno(ret);
 }
 
-int bpf_prog_detach(int target_fd, enum bpf_attach_type type)
+int bpf_prog_detach_opts(int prog_fd, int target, enum bpf_attach_type type,
+			 const struct bpf_prog_detach_opts *opts)
 {
-	const size_t attr_sz = offsetofend(union bpf_attr, replace_bpf_fd);
+	const size_t attr_sz = offsetofend(union bpf_attr, expected_revision);
+	__u32 relative_id, flags;
+	int ret, relative_fd;
 	union bpf_attr attr;
-	int ret;
 
-	memset(&attr, 0, attr_sz);
-	attr.target_fd	 = target_fd;
-	attr.attach_type = type;
+	if (!OPTS_VALID(opts, bpf_prog_detach_opts))
+		return libbpf_err(-EINVAL);
 
-	ret = sys_bpf(BPF_PROG_DETACH, &attr, attr_sz);
-	return libbpf_err_errno(ret);
-}
+	relative_id = OPTS_GET(opts, relative_id, 0);
+	relative_fd = OPTS_GET(opts, relative_fd, 0);
+	flags = OPTS_GET(opts, flags, 0);
 
-int bpf_prog_detach2(int prog_fd, int target_fd, enum bpf_attach_type type)
-{
-	const size_t attr_sz = offsetofend(union bpf_attr, replace_bpf_fd);
-	union bpf_attr attr;
-	int ret;
+	/* validate we don't have unexpected combinations of non-zero fields */
+	if (relative_fd && relative_id)
+		return libbpf_err(-EINVAL);
 
 	memset(&attr, 0, attr_sz);
-	attr.target_fd	 = target_fd;
+	attr.target_fd		= target;
 	attr.attach_bpf_fd	= prog_fd;
 	attr.attach_type	= type;
+	attr.expected_revision	= OPTS_GET(opts, expected_revision, 0);
+
+	if (relative_id) {
+		attr.attach_flags = flags | BPF_F_ID;
+		attr.relative_id  = relative_id;
+	} else {
+		attr.attach_flags = flags;
+		attr.relative_fd  = relative_fd;
+	}
 
 	ret = sys_bpf(BPF_PROG_DETACH, &attr, attr_sz);
 	return libbpf_err_errno(ret);
 }
 
+int bpf_prog_detach(int target_fd, enum bpf_attach_type type)
+{
+	return bpf_prog_detach_opts(0, target_fd, type, NULL);
+}
+
+int bpf_prog_detach2(int prog_fd, int target_fd, enum bpf_attach_type type)
+{
+	return bpf_prog_detach_opts(prog_fd, target_fd, type, NULL);
+}
+
 int bpf_link_create(int prog_fd, int target_fd,
 		    enum bpf_attach_type attach_type,
 		    const struct bpf_link_create_opts *opts)
@@ -841,8 +875,7 @@ int bpf_iter_create(int link_fd)
 	return libbpf_err_errno(fd);
 }
 
-int bpf_prog_query_opts(int target_fd,
-			enum bpf_attach_type type,
+int bpf_prog_query_opts(int target, enum bpf_attach_type type,
 			struct bpf_prog_query_opts *opts)
 {
 	const size_t attr_sz = offsetofend(union bpf_attr, query);
@@ -853,18 +886,20 @@ int bpf_prog_query_opts(int target_fd,
 		return libbpf_err(-EINVAL);
 
 	memset(&attr, 0, attr_sz);
-
-	attr.query.target_fd	= target_fd;
+	attr.query.target_fd		= target;
 	attr.query.attach_type		= type;
 	attr.query.query_flags		= OPTS_GET(opts, query_flags, 0);
-	attr.query.prog_cnt	= OPTS_GET(opts, prog_cnt, 0);
+	attr.query.count		= OPTS_GET(opts, count, 0);
 	attr.query.prog_ids		= ptr_to_u64(OPTS_GET(opts, prog_ids, NULL));
+	attr.query.link_ids		= ptr_to_u64(OPTS_GET(opts, link_ids, NULL));
 	attr.query.prog_attach_flags	= ptr_to_u64(OPTS_GET(opts, prog_attach_flags, NULL));
+	attr.query.link_attach_flags	= ptr_to_u64(OPTS_GET(opts, link_attach_flags, NULL));
 
 	ret = sys_bpf(BPF_PROG_QUERY, &attr, attr_sz);
 
 	OPTS_SET(opts, attach_flags, attr.query.attach_flags);
-	OPTS_SET(opts, prog_cnt, attr.query.prog_cnt);
+	OPTS_SET(opts, revision, attr.query.revision);
+	OPTS_SET(opts, count, attr.query.count);
 
 	return libbpf_err_errno(ret);
 }

tools/lib/bpf/bpf.h

@@ -312,21 +312,67 @@ LIBBPF_API int bpf_obj_get(const char *pathname);
 LIBBPF_API int bpf_obj_get_opts(const char *pathname,
 				const struct bpf_obj_get_opts *opts);
 
+LIBBPF_API int bpf_prog_attach(int prog_fd, int attachable_fd,
+			       enum bpf_attach_type type, unsigned int flags);
+LIBBPF_API int bpf_prog_detach(int attachable_fd, enum bpf_attach_type type);
+LIBBPF_API int bpf_prog_detach2(int prog_fd, int attachable_fd,
+				enum bpf_attach_type type);
+
 struct bpf_prog_attach_opts {
 	size_t sz; /* size of this struct for forward/backward compatibility */
-	unsigned int flags;
+	__u32 flags;
+	union {
 		int replace_prog_fd;
+		int replace_fd;
+	};
+	int relative_fd;
+	__u32 relative_id;
+	__u64 expected_revision;
+	size_t :0;
 };
-#define bpf_prog_attach_opts__last_field replace_prog_fd
+#define bpf_prog_attach_opts__last_field expected_revision
 
-LIBBPF_API int bpf_prog_attach(int prog_fd, int attachable_fd,
-			       enum bpf_attach_type type, unsigned int flags);
-LIBBPF_API int bpf_prog_attach_opts(int prog_fd, int attachable_fd,
+struct bpf_prog_detach_opts {
+	size_t sz; /* size of this struct for forward/backward compatibility */
+	__u32 flags;
+	int relative_fd;
+	__u32 relative_id;
+	__u64 expected_revision;
+	size_t :0;
+};
+#define bpf_prog_detach_opts__last_field expected_revision
+
+/**
+ * @brief **bpf_prog_attach_opts()** attaches the BPF program corresponding to
+ * *prog_fd* to a *target* which can represent a file descriptor or netdevice
+ * ifindex.
+ *
+ * @param prog_fd BPF program file descriptor
+ * @param target attach location file descriptor or ifindex
+ * @param type attach type for the BPF program
+ * @param opts options for configuring the attachment
+ * @return 0, on success; negative error code, otherwise (errno is also set to
+ * the error code)
+ */
+LIBBPF_API int bpf_prog_attach_opts(int prog_fd, int target,
 				    enum bpf_attach_type type,
 				    const struct bpf_prog_attach_opts *opts);
-LIBBPF_API int bpf_prog_detach(int attachable_fd, enum bpf_attach_type type);
-LIBBPF_API int bpf_prog_detach2(int prog_fd, int attachable_fd,
-				enum bpf_attach_type type);
+
+/**
+ * @brief **bpf_prog_detach_opts()** detaches the BPF program corresponding to
+ * *prog_fd* from a *target* which can represent a file descriptor or netdevice
+ * ifindex.
+ *
+ * @param prog_fd BPF program file descriptor
+ * @param target detach location file descriptor or ifindex
+ * @param type detach type for the BPF program
+ * @param opts options for configuring the detachment
+ * @return 0, on success; negative error code, otherwise (errno is also set to
+ * the error code)
+ */
+LIBBPF_API int bpf_prog_detach_opts(int prog_fd, int target,
+				    enum bpf_attach_type type,
+				    const struct bpf_prog_detach_opts *opts);
 
 union bpf_iter_link_info; /* defined in up-to-date linux/bpf.h */
 struct bpf_link_create_opts {
@@ -495,13 +541,31 @@ struct bpf_prog_query_opts {
 	__u32 query_flags;
 	__u32 attach_flags; /* output argument */
 	__u32 *prog_ids;
-	__u32 prog_cnt; /* input+output argument */
+	union {
+		/* input+output argument */
+		__u32 prog_cnt;
+		__u32 count;
+	};
 	__u32 *prog_attach_flags;
+	__u32 *link_ids;
+	__u32 *link_attach_flags;
+	__u64 revision;
+	size_t :0;
 };
-#define bpf_prog_query_opts__last_field prog_attach_flags
+#define bpf_prog_query_opts__last_field revision
 
-LIBBPF_API int bpf_prog_query_opts(int target_fd,
-				   enum bpf_attach_type type,
+/**
+ * @brief **bpf_prog_query_opts()** queries the BPF programs and BPF links
+ * which are attached to *target* which can represent a file descriptor or
+ * netdevice ifindex.
+ *
+ * @param target query location file descriptor or ifindex
+ * @param type attach type for the BPF program
+ * @param opts options for configuring the query
+ * @return 0, on success; negative error code, otherwise (errno is also set to
+ * the error code)
+ */
+LIBBPF_API int bpf_prog_query_opts(int target, enum bpf_attach_type type,
 				   struct bpf_prog_query_opts *opts);
 LIBBPF_API int bpf_prog_query(int target_fd, enum bpf_attach_type type,
 			      __u32 query_flags, __u32 *attach_flags,

tools/lib/bpf/libbpf.c

@@ -118,6 +118,8 @@ static const char * const attach_type_name[] = {
 	[BPF_TRACE_KPROBE_MULTI]	= "trace_kprobe_multi",
 	[BPF_STRUCT_OPS]		= "struct_ops",
 	[BPF_NETFILTER]			= "netfilter",
+	[BPF_TCX_INGRESS]		= "tcx_ingress",
+	[BPF_TCX_EGRESS]		= "tcx_egress",
 };
 
 static const char * const link_type_name[] = {
@@ -8696,9 +8698,13 @@ static const struct bpf_sec_def section_defs[] = {
 	SEC_DEF("ksyscall+",		KPROBE,	0, SEC_NONE, attach_ksyscall),
 	SEC_DEF("kretsyscall+",		KPROBE, 0, SEC_NONE, attach_ksyscall),
 	SEC_DEF("usdt+",		KPROBE,	0, SEC_NONE, attach_usdt),
-	SEC_DEF("tc",			SCHED_CLS, 0, SEC_NONE),
-	SEC_DEF("classifier",		SCHED_CLS, 0, SEC_NONE),
-	SEC_DEF("action",		SCHED_ACT, 0, SEC_NONE),
+	SEC_DEF("tc/ingress",		SCHED_CLS, BPF_TCX_INGRESS, SEC_NONE), /* alias for tcx */
+	SEC_DEF("tc/egress",		SCHED_CLS, BPF_TCX_EGRESS, SEC_NONE),  /* alias for tcx */
+	SEC_DEF("tcx/ingress",		SCHED_CLS, BPF_TCX_INGRESS, SEC_NONE),
+	SEC_DEF("tcx/egress",		SCHED_CLS, BPF_TCX_EGRESS, SEC_NONE),
+	SEC_DEF("tc",			SCHED_CLS, 0, SEC_NONE), /* deprecated / legacy, use tcx */
+	SEC_DEF("classifier",		SCHED_CLS, 0, SEC_NONE), /* deprecated / legacy, use tcx */
+	SEC_DEF("action",		SCHED_ACT, 0, SEC_NONE), /* deprecated / legacy, use tcx */
 	SEC_DEF("tracepoint+",		TRACEPOINT, 0, SEC_NONE, attach_tp),
 	SEC_DEF("tp+",			TRACEPOINT, 0, SEC_NONE, attach_tp),
 	SEC_DEF("raw_tracepoint+",	RAW_TRACEPOINT, 0, SEC_NONE, attach_raw_tp),

tools/lib/bpf/libbpf.map

@@ -395,5 +395,6 @@ LIBBPF_1.2.0 {
 LIBBPF_1.3.0 {
 	global:
 		bpf_obj_pin_opts;
+		bpf_prog_detach_opts;
 		bpf_program__attach_netfilter;
 } LIBBPF_1.2.0;