bpf.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrii Nakryiko <andriin@fb.com>
To: <bpf@vger.kernel.org>, <netdev@vger.kernel.org>, <ast@fb.com>,
	<daniel@iogearbox.net>
Cc: <andrii.nakryiko@gmail.com>, <kernel-team@fb.com>,
	Andrii Nakryiko <andriin@fb.com>, Hao Luo <haoluo@google.com>,
	Arnaldo Carvalho de Melo <acme@kernel.org>,
	Song Liu <songliubraving@fb.com>,
	Quentin Monnet <quentin@isovalent.com>
Subject: [RFC PATCH bpf-next 2/8] libbpf: add support for extracting kernel symbol addresses
Date: Fri, 12 Jun 2020 15:31:44 -0700	[thread overview]
Message-ID: <20200612223150.1177182-3-andriin@fb.com> (raw)
In-Reply-To: <20200612223150.1177182-1-andriin@fb.com>

Add support for another (in addition to existing Kconfig) special kind of
externs in BPF code, kernel symbol externs. Such externs allow BPF code to
"know" kernel symbol address and either use it for comparisons with kernel
data structures (e.g., struct file's f_op pointer, to distinguish different
kinds of file), or, with the help of bpf_probe_user_kernel(), to follow
pointers and read data from global variables. Kernel symbol addresses are
found through /proc/kallsyms, which should be present in the system.

Currently, such kernel symbol variables are typeless: they have to be defined
as `extern const void <symbol>` and the only operation you can do (in C code)
with them is to take its address. Such extern should reside in a special
section '.ksyms'. bpf_helpers.h header provides __ksym macro for this. Strong
vs weak semantics stays the same as with Kconfig externs. If symbol is not
found in /proc/kallsyms, this will be a failure for strong (non-weak) extern,
but will be defaulted to 0 for weak externs.

If the same symbol is defined multiple times in /proc/kallsyms, then it will
be error if any of the associated addresses differs. In that case, address is
ambiguous, so libbpf falls on the side of caution, rather than confusing user
with randomly chosen address.

In the future, once kernel is extended with variables BTF information, such
ksym externs will be supported in a typed version, which will allow BPF
program to read variable's contents directly, similarly to how it's done for
fentry/fexit input arguments.

Signed-off-by: Andrii Nakryiko <andriin@fb.com>
---
 tools/lib/bpf/bpf_helpers.h |   1 +
 tools/lib/bpf/btf.h         |   5 ++
 tools/lib/bpf/libbpf.c      | 138 ++++++++++++++++++++++++++++++++++--
 3 files changed, 139 insertions(+), 5 deletions(-)

diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h
index f67dce2af802..a510d8ed716f 100644
--- a/tools/lib/bpf/bpf_helpers.h
+++ b/tools/lib/bpf/bpf_helpers.h
@@ -75,5 +75,6 @@ enum libbpf_tristate {
 };
 
 #define __kconfig __attribute__((section(".kconfig")))
+#define __ksym __attribute__((section(".ksyms")))
 
 #endif
diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h
index 70c1b7ec2bd0..06cd1731c154 100644
--- a/tools/lib/bpf/btf.h
+++ b/tools/lib/bpf/btf.h
@@ -168,6 +168,11 @@ static inline bool btf_kflag(const struct btf_type *t)
 	return BTF_INFO_KFLAG(t->info);
 }
 
+static inline bool btf_is_void(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_UNKN;
+}
+
 static inline bool btf_is_int(const struct btf_type *t)
 {
 	return btf_kind(t) == BTF_KIND_INT;
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index b1a36c965fc0..451fb0b5615b 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -285,6 +285,7 @@ struct bpf_struct_ops {
 #define BSS_SEC ".bss"
 #define RODATA_SEC ".rodata"
 #define KCONFIG_SEC ".kconfig"
+#define KSYMS_SEC ".ksyms"
 #define STRUCT_OPS_SEC ".struct_ops"
 
 enum libbpf_map_type {
@@ -329,6 +330,7 @@ struct bpf_map {
 
 enum extern_type {
 	EXT_UNKNOWN,
+	EXT_KSYM,
 	EXT_KCFG,
 };
 
@@ -357,6 +359,9 @@ struct extern_desc {
 			int data_off;
 			bool is_signed;
 		} kcfg;
+		struct {
+			unsigned long addr;
+		} ksym;
 	};
 };
 
@@ -2812,9 +2817,25 @@ static int cmp_externs(const void *_a, const void *_b)
 	return strcmp(a->name, b->name);
 }
 
+static int find_int_btf_id(const struct btf *btf)
+{
+	const struct btf_type *t;
+	int i, n;
+
+	n = btf__get_nr_types(btf);
+	for (i = 1; i <= n; i++) {
+		t = btf__type_by_id(btf, i);
+
+		if (btf_is_int(t) && btf_int_bits(t) == 32)
+			return i;
+	}
+
+	return 0;
+}
+
 static int bpf_object__collect_externs(struct bpf_object *obj)
 {
-	struct btf_type *sec, *kcfg_sec = NULL;
+	struct btf_type *sec, *kcfg_sec = NULL, *ksym_sec = NULL;
 	const struct btf_type *t;
 	struct extern_desc *ext;
 	int i, n, off;
@@ -2895,6 +2916,17 @@ static int bpf_object__collect_externs(struct bpf_object *obj)
 				pr_warn("kconfig extern '%s' type is unsupported\n", ext_name);
 				return -ENOTSUP;
 			}
+		} else if (strcmp(sec_name, KSYMS_SEC) == 0) {
+			const struct btf_type *vt;
+
+			ksym_sec = sec;
+			ext->type = EXT_KSYM;
+
+			vt = skip_mods_and_typedefs(obj->btf, t->type, NULL);
+			if (!btf_is_void(vt)) {
+				pr_warn("ksym extern '%s' is not typeless (void)\n", ext_name);
+				return -ENOTSUP;
+			}
 		} else {
 			pr_warn("unrecognized extern section '%s'\n", sec_name);
 			return -ENOTSUP;
@@ -2908,6 +2940,43 @@ static int bpf_object__collect_externs(struct bpf_object *obj)
 	/* sort externs by type, for kcfg ones also by (align, size, name) */
 	qsort(obj->externs, obj->nr_extern, sizeof(*ext), cmp_externs);
 
+	/* for .ksyms section, we need to turn all externs into allocated
+	 * variables in BTF to pass kernel verification; we do this by
+	 * pretending that each extern is a 8-byte variable
+	 */
+	if (ksym_sec) {
+		/* find existing 4-byte integer type in BTF to use for fake
+		 * extern variables in DATASEC
+		 */
+		int int_btf_id = find_int_btf_id(obj->btf);
+
+		for (i = 0; i < n; i++) {
+			pr_debug("extern #%d (ksym): symbol %d, name %s\n",
+				 i, ext->sym_idx, ext->name);
+		}
+
+		sec = ksym_sec;
+		n = btf_vlen(sec);
+		for (i = 0, off = 0; i < n; i++, off += sizeof(int)) {
+			struct btf_var_secinfo *vs = btf_var_secinfos(sec) + i;
+			struct btf_type *vt;
+
+			vt = (void *)btf__type_by_id(obj->btf, vs->type);
+			ext_name = btf__name_by_offset(obj->btf, vt->name_off);
+			ext = find_extern_by_name(obj, ext_name);
+			if (!ext) {
+				pr_warn("failed to find extern definition for BTF var '%s'\n",
+					ext_name);
+				return -ESRCH;
+			}
+			btf_var(vt)->linkage = BTF_VAR_GLOBAL_ALLOCATED;
+			vt->type = int_btf_id;
+			vs->offset = off;
+			vs->size = sizeof(int);
+		}
+		sec->size = off;
+	}
+
 	if (kcfg_sec) {
 		sec = kcfg_sec;
 		/* for kcfg externs calculate their offsets within a .kconfig map */
@@ -5009,9 +5078,14 @@ bpf_program__relocate(struct bpf_program *prog, struct bpf_object *obj)
 			break;
 		case RELO_EXTERN:
 			ext = &obj->externs[relo->sym_off];
-			insn[0].src_reg = BPF_PSEUDO_MAP_VALUE;
-			insn[0].imm = obj->maps[obj->kconfig_map_idx].fd;
-			insn[1].imm = ext->kcfg.data_off;
+			if (ext->type == EXT_KCFG) {
+				insn[0].src_reg = BPF_PSEUDO_MAP_VALUE;
+				insn[0].imm = obj->maps[obj->kconfig_map_idx].fd;
+				insn[1].imm = ext->kcfg.data_off;
+			} else /* EXT_KSYM */ {
+				insn[0].imm = (__u32)ext->ksym.addr;
+				insn[1].imm = ext->ksym.addr >> 32;
+			}
 			break;
 		case RELO_CALL:
 			err = bpf_program__reloc_text(prog, obj, relo);
@@ -5630,10 +5704,57 @@ static int bpf_object__sanitize_maps(struct bpf_object *obj)
 	return 0;
 }
 
+static int bpf_object__read_kallsyms_file(struct bpf_object *obj)
+{
+	char sym_type, sym_name[256];
+	unsigned long sym_addr;
+	struct extern_desc *ext;
+	int ret, err = 0;
+	FILE *f;
+
+	f = fopen("/proc/kallsyms", "r");
+	if (!f) {
+		err = -errno;
+		pr_warn("failed to open /proc/kallsyms: %d\n", err);
+		return err;
+	}
+
+	while (true) {
+		ret = fscanf(f, "%lx %c %s%*[^\n]\n",
+			     &sym_addr, &sym_type, sym_name);
+		if (ret == EOF && feof(f))
+			break;
+		if (ret != 3) {
+			err = -EINVAL;
+			goto out;
+		}
+
+		ext = find_extern_by_name(obj, sym_name);
+		if (!ext || ext->type != EXT_KSYM)
+			continue;
+
+		if (ext->is_set && ext->ksym.addr != sym_addr) {
+			pr_warn("ksym extern '%s' resolution is ambiguous: 0x%lx or 0x%lx\n",
+				sym_name, ext->ksym.addr, sym_addr);
+			err = -EINVAL;
+			goto out;
+		}
+		if (!ext->is_set) {
+			ext->is_set = true;
+			ext->ksym.addr = sym_addr;
+			pr_debug("ksym extern %s=0x%lx\n", sym_name, sym_addr);
+		}
+	}
+
+out:
+	fclose(f);
+	return err;
+}
+
 static int bpf_object__resolve_externs(struct bpf_object *obj,
 				       const char *extra_kconfig)
 {
-	bool need_config = false;
+	bool need_config = false, need_kallsyms = false;
 	struct extern_desc *ext;
 	void *kcfg_data;
 	int err, i;
@@ -5663,6 +5784,8 @@ static int bpf_object__resolve_externs(struct bpf_object *obj,
 		} else if (ext->type == EXT_KCFG &&
 			   strncmp(ext->name, "CONFIG_", 7) == 0) {
 			need_config = true;
+		} else if (ext->type == EXT_KSYM) {
+			need_kallsyms = true;
 		} else {
 			pr_warn("unrecognized extern '%s'\n", ext->name);
 			return -EINVAL;
@@ -5686,6 +5809,11 @@ static int bpf_object__resolve_externs(struct bpf_object *obj,
 		if (err)
 			return -EINVAL;
 	}
+	if (need_kallsyms) {
+		err = bpf_object__read_kallsyms_file(obj);
+		if (err)
+			return -EINVAL;
+	}
 	for (i = 0; i < obj->nr_extern; i++) {
 		ext = &obj->externs[i];
 
-- 
2.24.1


  parent reply	other threads:[~2020-06-12 22:34 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-12 22:31 [RFC PATCH bpf-next 0/8] libbpf ksym support and bpftool show PIDs Andrii Nakryiko
2020-06-12 22:31 ` [RFC PATCH bpf-next 1/8] libbpf: generalize libbpf externs support Andrii Nakryiko
     [not found]   ` <CA+khW7hAYVdoQX5-j0z1iGEVZeww4BBu4NXzy5eS5OwDRYqe2w@mail.gmail.com>
2020-06-15 18:55     ` Andrii Nakryiko
2020-06-12 22:31 ` Andrii Nakryiko [this message]
     [not found]   ` <CA+khW7hFZzp_K_xydSFw0O3LYB22_fC=Z4wG7i9Si+phGHn4cQ@mail.gmail.com>
2020-06-15 19:08     ` [RFC PATCH bpf-next 2/8] libbpf: add support for extracting kernel symbol addresses Andrii Nakryiko
2020-06-16  8:05       ` Hao Luo
2020-06-17  1:24         ` Hao Luo
2020-06-17  1:36           ` Andrii Nakryiko
2020-06-18  7:53             ` Hao Luo
2020-06-12 22:31 ` [RFC PATCH bpf-next 3/8] selftests/bpf: add __ksym extern selftest Andrii Nakryiko
     [not found]   ` <CA+khW7jxdS1KRpk2syVGjDqbyn3wAd3Eh_LEMAEhkPUehuXMwg@mail.gmail.com>
2020-06-15 19:11     ` Andrii Nakryiko
2020-06-12 22:31 ` [RFC PATCH bpf-next 4/8] tools/bpftool: move map/prog parsing logic into common Andrii Nakryiko
2020-06-12 22:31 ` [RFC PATCH bpf-next 5/8] tools/bpftool: minimize bootstrap bpftool Andrii Nakryiko
2020-06-12 22:31 ` [RFC PATCH bpf-next 6/8] tools/bpftool: generalize BPF skeleton support and generate vmlinux.h Andrii Nakryiko
2020-06-12 22:31 ` [RFC PATCH bpf-next 7/8] libbpf: wrap source argument of BPF_CORE_READ macro in parentheses Andrii Nakryiko
2020-06-12 22:31 ` [RFC PATCH bpf-next 8/8] tools/bpftool: show PIDs with FDs open against BPF map/prog/link/btf Andrii Nakryiko
2020-06-13  3:45   ` Alexei Starovoitov
2020-06-13  5:57     ` Andrii Nakryiko
2020-06-13 22:14       ` Arnaldo Carvalho de Melo
2020-06-15  9:04         ` Toke Høiland-Jørgensen
2020-06-15  9:30           ` Quentin Monnet

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200612223150.1177182-3-andriin@fb.com \
    --to=andriin@fb.com \
    --cc=acme@kernel.org \
    --cc=andrii.nakryiko@gmail.com \
    --cc=ast@fb.com \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=haoluo@google.com \
    --cc=kernel-team@fb.com \
    --cc=netdev@vger.kernel.org \
    --cc=quentin@isovalent.com \
    --cc=songliubraving@fb.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).