[bpf-next,v1,11/13] tools/libbpf: Add bpf_program__attach_lsm
diff mbox series

Message ID 20191220154208.15895-12-kpsingh@chromium.org
State New
Headers show
Series
  • MAC and Audit policy using eBPF (KRSI)
Related show

Commit Message

KP Singh Dec. 20, 2019, 3:42 p.m. UTC
From: KP Singh <kpsingh@google.com>

Add functionality in libbpf to attach eBPF program to LSM hooks.

Signed-off-by: KP Singh <kpsingh@google.com>
---
 tools/lib/bpf/libbpf.c   | 127 +++++++++++++++++++++++++++++++++++++--
 tools/lib/bpf/libbpf.h   |   2 +
 tools/lib/bpf/libbpf.map |   1 +
 3 files changed, 126 insertions(+), 4 deletions(-)

Comments

Andrii Nakryiko Dec. 24, 2019, 6:44 a.m. UTC | #1
On Fri, Dec 20, 2019 at 7:42 AM KP Singh <kpsingh@chromium.org> wrote:
>
> From: KP Singh <kpsingh@google.com>
>
> Add functionality in libbpf to attach eBPF program to LSM hooks.
>
> Signed-off-by: KP Singh <kpsingh@google.com>
> ---
>  tools/lib/bpf/libbpf.c   | 127 +++++++++++++++++++++++++++++++++++++--
>  tools/lib/bpf/libbpf.h   |   2 +
>  tools/lib/bpf/libbpf.map |   1 +
>  3 files changed, 126 insertions(+), 4 deletions(-)
>
> diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
> index b0b27d8e5a37..ab2b23b4f21f 100644
> --- a/tools/lib/bpf/libbpf.c
> +++ b/tools/lib/bpf/libbpf.c
> @@ -5122,8 +5122,8 @@ int libbpf_prog_type_by_name(const char *name, enum bpf_prog_type *prog_type,
>         return -ESRCH;
>  }
>
> -static inline int __btf__typdef_with_prefix(struct btf *btf, const char *name,
> -                                           const char *prefix)
> +static inline int __btf__type_with_prefix(struct btf *btf, const char *name,
> +                                         const char *prefix)

Please do this rename in a patch that introduced this function, there
is no need to split such changes between two patches. See also my
request to rename and generalize it a bit.

>  {
>
>         size_t prefix_len = strlen(prefix);
> @@ -5149,9 +5149,9 @@ int libbpf_find_vmlinux_btf_id(const char *name,
>         }
>

[...]

>
> +
> +static int bpf_link__destroy_lsm(struct bpf_link *link)
> +{
> +       struct bpf_link_lsm *ll = container_of(link, struct bpf_link_lsm, link);

struct bpf_link link being a first field is a requirement for
bpf_link, so you don't need container_of, just cast link to your type.

> +       char errmsg[STRERR_BUFSIZE];
> +       int ret;
> +
> +       ret = bpf_prog_detach2(ll->prog_fd, ll->hook_fd, BPF_LSM_MAC);
> +       if (ret < 0) {
> +               ret = -errno;
> +               pr_warn("failed to detach from hook: %s\n",
> +                       libbpf_strerror_r(ret, errmsg, sizeof(errmsg)));
> +               return ret;
> +       }
> +       close(ll->hook_fd);
> +       return 0;
> +}
> +
> +static const char *__lsm_hook_name(const char *title)
> +{
> +
> +       int i;
> +
> +       if (!title)
> +               return ERR_PTR(-EINVAL);
> +
> +       for (i = 0; i < ARRAY_SIZE(section_names); i++) {

section_names have been renamed to section_defs a while ago, please rebase

> +               if (section_names[i].prog_type != BPF_PROG_TYPE_LSM)
> +                       continue;
> +

[...]
James Morris Jan. 8, 2020, 6:24 p.m. UTC | #2
On Fri, 20 Dec 2019, KP Singh wrote:

> From: KP Singh <kpsingh@google.com>
> 
> Add functionality in libbpf to attach eBPF program to LSM hooks.
> 
> Signed-off-by: KP Singh <kpsingh@google.com>


Reviewed-by: James Morris <jamorris@linux.microsoft.com>

Patch
diff mbox series

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index b0b27d8e5a37..ab2b23b4f21f 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -5122,8 +5122,8 @@  int libbpf_prog_type_by_name(const char *name, enum bpf_prog_type *prog_type,
 	return -ESRCH;
 }
 
-static inline int __btf__typdef_with_prefix(struct btf *btf, const char *name,
-					    const char *prefix)
+static inline int __btf__type_with_prefix(struct btf *btf, const char *name,
+					  const char *prefix)
 {
 
 	size_t prefix_len = strlen(prefix);
@@ -5149,9 +5149,9 @@  int libbpf_find_vmlinux_btf_id(const char *name,
 	}
 
 	if (attach_type == BPF_TRACE_RAW_TP)
-		err = __btf__typdef_with_prefix(btf, name, BTF_TRACE_PREFIX);
+		err = __btf__type_with_prefix(btf, name, BTF_TRACE_PREFIX);
 	else if (attach_type == BPF_LSM_MAC)
-		err = __btf__typdef_with_prefix(btf, name, BTF_LSM_PREFIX);
+		err = __btf__type_with_prefix(btf, name, BTF_LSM_PREFIX);
 	else
 		err = btf__find_by_name_kind(btf, name, BTF_KIND_FUNC);
 
@@ -5502,6 +5502,18 @@  struct bpf_link_fd {
 	int fd; /* hook FD */
 };
 
+/*
+ * The other attach types allow the link to be destroyed by using an ioctl or
+ * an operation on some file descriptor that describes the attachment. An LSM
+ * hook can have multiple programs attached to each hook, so the link needs to
+ * specify the program that must be detached when the link is destroyed.
+ */
+struct bpf_link_lsm {
+	struct bpf_link link;
+	int hook_fd;
+	int prog_fd;
+};
+
 static int bpf_link__destroy_perf_event(struct bpf_link *link)
 {
 	struct bpf_link_fd *l = (void *)link;
@@ -5876,6 +5888,113 @@  struct bpf_link *bpf_program__attach_trace(struct bpf_program *prog)
 	return (struct bpf_link *)link;
 }
 
+
+static int bpf_link__destroy_lsm(struct bpf_link *link)
+{
+	struct bpf_link_lsm *ll = container_of(link, struct bpf_link_lsm, link);
+	char errmsg[STRERR_BUFSIZE];
+	int ret;
+
+	ret = bpf_prog_detach2(ll->prog_fd, ll->hook_fd, BPF_LSM_MAC);
+	if (ret < 0) {
+		ret = -errno;
+		pr_warn("failed to detach from hook: %s\n",
+			libbpf_strerror_r(ret, errmsg, sizeof(errmsg)));
+		return ret;
+	}
+	close(ll->hook_fd);
+	return 0;
+}
+
+static const char *__lsm_hook_name(const char *title)
+{
+
+	int i;
+
+	if (!title)
+		return ERR_PTR(-EINVAL);
+
+	for (i = 0; i < ARRAY_SIZE(section_names); i++) {
+		if (section_names[i].prog_type != BPF_PROG_TYPE_LSM)
+			continue;
+
+		if (strncmp(title, section_names[i].sec,
+			    section_names[i].len)) {
+			pr_warn("title for a LSM prog must begin with '%s'\n",
+				section_names[i].sec);
+			return ERR_PTR(-EINVAL);
+		}
+
+		return title + section_names[i].len;
+	}
+
+	pr_warn("could not find section information for BPF_PROG_TYPE_LSM\n");
+	return ERR_PTR(-ESRCH);
+}
+
+struct bpf_link *bpf_program__attach_lsm(struct bpf_program *prog)
+{
+	char hook_path[PATH_MAX] = "/sys/kernel/security/bpf/";
+	const char *title, *hook_name;
+	char errmsg[STRERR_BUFSIZE];
+	int prog_fd, target_fd, ret;
+	struct bpf_link_lsm *link;
+
+	title = bpf_program__title(prog, false);
+	if (IS_ERR(title)) {
+		pr_warn("could not determine title of the program\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	hook_name = __lsm_hook_name(title);
+	if (IS_ERR(hook_name)) {
+		pr_warn("could not determine LSM hook name from title '%s'\n",
+			title);
+		return ERR_PTR(-EINVAL);
+	}
+
+	prog_fd = bpf_program__fd(prog);
+	if (prog_fd < 0) {
+		pr_warn("program '%s': can't attach before loaded\n", title);
+		return ERR_PTR(-EINVAL);
+	}
+
+	link = malloc(sizeof(*link));
+	if (!link)
+		return ERR_PTR(-ENOMEM);
+	link->link.destroy = &bpf_link__destroy_lsm;
+
+	/* Attach the BPF program to the given hook */
+	strncat(hook_path, hook_name,
+		sizeof(hook_path) - (strlen(hook_path) + 1));
+	target_fd = open(hook_path, O_RDWR);
+	if (target_fd < 0) {
+		ret = -errno;
+		pr_warn("program '%s': failed to open to hook '%s': %s\n",
+			title, hook_path,
+			libbpf_strerror_r(ret, errmsg, sizeof(errmsg)));
+		return ERR_PTR(ret);
+	}
+
+	ret = bpf_prog_attach(prog_fd, target_fd, BPF_LSM_MAC,
+			      BPF_F_ALLOW_OVERRIDE);
+	if (ret < 0) {
+		ret = -errno;
+		pr_warn("program '%s': failed to attach to hook '%s': %s\n",
+			title, hook_name,
+			libbpf_strerror_r(ret, errmsg, sizeof(errmsg)));
+		goto error;
+	}
+
+	link->hook_fd = target_fd;
+	link->prog_fd = prog_fd;
+	return &link->link;
+
+error:
+	close(target_fd);
+	return ERR_PTR(ret);
+}
+
 enum bpf_perf_event_ret
 bpf_perf_event_read_simple(void *mmap_mem, size_t mmap_size, size_t page_size,
 			   void **copy_mem, size_t *copy_size,
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 9cd69d602c82..655f27ad6ece 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -256,6 +256,8 @@  bpf_program__attach_raw_tracepoint(struct bpf_program *prog,
 
 LIBBPF_API struct bpf_link *
 bpf_program__attach_trace(struct bpf_program *prog);
+LIBBPF_API struct bpf_link *
+bpf_program__attach_lsm(struct bpf_program *prog);
 struct bpf_insn;
 
 /*
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index 3d396149755d..5d64ba9b2a43 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -213,4 +213,5 @@  LIBBPF_0.0.7 {
 	global:
 		bpf_program__is_lsm;
 		bpf_program__set_lsm;
+		bpf_program__attach_lsm;
 } LIBBPF_0.0.6;