chore: call bpf_program__attach_iter from go (#304)

This changes libbpfgo to call bpf_program__attach_iter directly from
go instead of doing this via a C wrapper. What this means is that
we maintained opts creation in C, but the actual libbpf attach call is
done in go.

This creates two new functions in libbpfgo.h as helpers:
- bpf_iter_attach_opts_new()
- bpf_iter_attach_opts_free()

libbpfgo.go

@@ -1613,11 +1613,16 @@ func (p *BPFProg) AttachIter(opts IterOpts) (*BPFLink, error) {
 	tid := C.uint(opts.Tid)
 	pid := C.uint(opts.Pid)
 	pidFd := C.uint(opts.PidFd)
-	link, errno := C.bpf_prog_attach_iter(p.prog, mapFd, cgroupIterOrder, cgroupFd, cgroupId, tid, pid, pidFd)
+	cOpts, errno := C.bpf_iter_attach_opts_new(mapFd, cgroupIterOrder, cgroupFd, cgroupId, tid, pid, pidFd)
+	if cOpts == nil {
+		return nil, fmt.Errorf("failed to create iter_attach_opts to program %s: %w", p.name, errno)
+	}
+	defer C.bpf_iter_attach_opts_free(cOpts)
+
+	link, errno := C.bpf_program__attach_iter(p.prog, cOpts)
 	if link == nil {
 		return nil, fmt.Errorf("failed to attach iter to program %s: %w", p.name, errno)
 	}
-
 	eventName := fmt.Sprintf("iter-%s-%d", p.name, opts.MapFd)
 	bpfLink := &BPFLink{
 		link:      link,

libbpfgo.h

@@ -125,24 +125,43 @@ int bpf_prog_detach_cgroup_legacy(
   return syscall(__NR_bpf, BPF_PROG_DETACH, &attr, sizeof(attr));
 }
 
-struct bpf_link *bpf_prog_attach_iter(struct bpf_program *prog, __u32 map_fd,
-                                      enum bpf_cgroup_iter_order order,
-                                      __u32 cgroup_fd, __u64 cgroup_id,
-                                      __u32 tid, __u32 pid, __u32 pid_fd) {
-  DECLARE_LIBBPF_OPTS(bpf_iter_attach_opts, opts);
-  union bpf_iter_link_info linfo;
-  memset(&linfo, 0, sizeof(linfo));
-  linfo.map.map_fd = map_fd;
-  linfo.cgroup.order = order;
-  linfo.cgroup.cgroup_fd = cgroup_fd;
-  linfo.cgroup.cgroup_id = cgroup_id;
-  linfo.task.tid = tid;
-  linfo.task.pid = pid;
-  linfo.task.pid_fd = pid_fd;
-  opts.link_info = &linfo;
-  opts.link_info_len = sizeof(linfo);
-
-  return bpf_program__attach_iter(prog, &opts);
+struct bpf_iter_attach_opts *
+bpf_iter_attach_opts_new(__u32 map_fd, enum bpf_cgroup_iter_order order,
+                         __u32 cgroup_fd, __u64 cgroup_id, __u32 tid, __u32 pid,
+                         __u32 pid_fd) {
+  union bpf_iter_link_info *linfo;
+  linfo = calloc(1, sizeof(*linfo));
+  if (!linfo)
+    return NULL;
+
+  linfo->map.map_fd = map_fd;
+  linfo->cgroup.order = order;
+  linfo->cgroup.cgroup_fd = cgroup_fd;
+  linfo->cgroup.cgroup_id = cgroup_id;
+  linfo->task.tid = tid;
+  linfo->task.pid = pid;
+  linfo->task.pid_fd = pid_fd;
+
+  struct bpf_iter_attach_opts *opts;
+  opts = calloc(1, sizeof(*opts));
+  if (!opts) {
+    free(linfo);
+    return NULL;
+  }
+
+  opts->sz = sizeof(*opts);
+  opts->link_info_len = sizeof(*linfo);
+  opts->link_info = linfo;
+
+  return opts;
+}
+
+void bpf_iter_attach_opts_free(struct bpf_iter_attach_opts *opts) {
+  if (!opts)
+    return;
+
+  free(opts->link_info);
+  free(opts);
 }
 
 struct bpf_object *open_bpf_object(char *btf_file_path, char *kconfig_path,