Message-Id: <20241030111533.907289-1-xukuohai@huaweicloud.com>
Date: Wed, 30 Oct 2024 19:15:33 +0800
From: Xu Kuohai <xukuohai@...weicloud.com>
To: bpf@...r.kernel.org,
	netdev@...r.kernel.org
Cc: Alexei Starovoitov <ast@...nel.org>,
	Daniel Borkmann <daniel@...earbox.net>,
	Andrii Nakryiko <andrii@...nel.org>,
	Martin KaFai Lau <martin.lau@...ux.dev>,
	Eduard Zingerman <eddyz87@...il.com>,
	Yonghong Song <yonghong.song@...ux.dev>,
	Kui-Feng Lee <thinker.li@...il.com>
Subject: [PATCH bpf-next] bpf: Add kernel symbol for struct_ops trampoline

From: Xu Kuohai <xukuohai@...wei.com>

Without kernel symbols for struct_ops trampoline, the unwinder may
produce unexpected stacktraces.

For example, the x86 ORC and FP unwinders check if an IP is in kernel
text by verifying the presence of the IP's kernel symbol. When a
struct_ops trampoline address is encountered, the unwinder stops due
to the absence of symbol, resulting in an incomplete stacktrace that
consists only of direct and indirect child functions called from the
trampoline.

The arm64 unwinder is another example. While the arm64 unwinder can
proceed across a struct_ops trampoline address, the corresponding
symbol name is displayed as "unknown", which is confusing.

Thus, add kernel symbol for struct_ops trampoline. The name is
bpf_trampoline_<PROG_NAME>, where PROG_NAME is the name of the
struct_ops prog linked to the trampoline.

Fixes: 85d33df357b6 ("bpf: Introduce BPF_MAP_TYPE_STRUCT_OPS")
Signed-off-by: Xu Kuohai <xukuohai@...wei.com>
---
 include/linux/bpf.h         |  3 +-
 kernel/bpf/bpf_struct_ops.c | 65 +++++++++++++++++++++++++++++++++++++
 kernel/bpf/dispatcher.c     |  3 +-
 kernel/bpf/trampoline.c     |  9 +++--
 4 files changed, 76 insertions(+), 4 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index c3ba4d475174..46f8d6c1a55c 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1402,7 +1402,8 @@ int arch_prepare_bpf_dispatcher(void *image, void *buf, s64 *funcs, int num_func
 void bpf_dispatcher_change_prog(struct bpf_dispatcher *d, struct bpf_prog *from,
 				struct bpf_prog *to);
 /* Called only from JIT-enabled code, so there's no need for stubs. */
-void bpf_image_ksym_add(void *data, unsigned int size, struct bpf_ksym *ksym);
+void bpf_image_ksym_init(void *data, unsigned int size, struct bpf_ksym *ksym);
+void bpf_image_ksym_add(struct bpf_ksym *ksym);
 void bpf_image_ksym_del(struct bpf_ksym *ksym);
 void bpf_ksym_add(struct bpf_ksym *ksym);
 void bpf_ksym_del(struct bpf_ksym *ksym);
diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
index fda3dd2ee984..7bd93c716c87 100644
--- a/kernel/bpf/bpf_struct_ops.c
+++ b/kernel/bpf/bpf_struct_ops.c
@@ -38,6 +38,9 @@ struct bpf_struct_ops_map {
 	 * that stores the func args before calling the bpf_prog.
 	 */
 	void *image_pages[MAX_TRAMP_IMAGE_PAGES];
+	u32 ksyms_cnt;
+	/* ksyms for bpf trampolines */
+	struct bpf_ksym *ksyms;
 	/* The owner moduler's btf. */
 	struct btf *btf;
 	/* uvalue->data stores the kernel struct
@@ -586,6 +589,33 @@ int bpf_struct_ops_prepare_trampoline(struct bpf_tramp_links *tlinks,
 	return 0;
 }
 
+static void bpf_struct_ops_ksym_init(struct bpf_prog *prog, void *image,
+				     unsigned int size, struct bpf_ksym *ksym)
+{
+	INIT_LIST_HEAD_RCU(&ksym->lnode);
+	snprintf(ksym->name, KSYM_NAME_LEN, "bpf_trampoline_%s",
+		 prog->aux->ksym.name);
+	bpf_image_ksym_init(image, size, ksym);
+}
+
+static void bpf_struct_ops_map_ksyms_add(struct bpf_struct_ops_map *st_map)
+{
+	struct bpf_ksym *ksym = st_map->ksyms;
+	struct bpf_ksym *end = ksym + st_map->ksyms_cnt;
+
+	while (ksym != end && ksym->start)
+		bpf_image_ksym_add(ksym++);
+}
+
+static void bpf_struct_ops_map_ksyms_del(struct bpf_struct_ops_map *st_map)
+{
+	struct bpf_ksym *ksym = st_map->ksyms;
+	struct bpf_ksym *end = ksym + st_map->ksyms_cnt;
+
+	while (ksym != end && ksym->start)
+		bpf_image_ksym_del(ksym++);
+}
+
 static long bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key,
 					   void *value, u64 flags)
 {
@@ -601,6 +631,7 @@ static long bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key,
 	int prog_fd, err;
 	u32 i, trampoline_start, image_off = 0;
 	void *cur_image = NULL, *image = NULL;
+	struct bpf_ksym *ksym;
 
 	if (flags)
 		return -EINVAL;
@@ -640,6 +671,7 @@ static long bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key,
 	kdata = &kvalue->data;
 
 	module_type = btf_type_by_id(btf_vmlinux, st_ops_ids[IDX_MODULE_ID]);
+	ksym = st_map->ksyms;
 	for_each_member(i, t, member) {
 		const struct btf_type *mtype, *ptype;
 		struct bpf_prog *prog;
@@ -735,6 +767,11 @@ static long bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key,
 
 		/* put prog_id to udata */
 		*(unsigned long *)(udata + moff) = prog->aux->id;
+
+		/* init ksym for this trampoline */
+		bpf_struct_ops_ksym_init(prog, image + trampoline_start,
+					 image_off - trampoline_start,
+					 ksym++);
 	}
 
 	if (st_ops->validate) {
@@ -790,6 +827,8 @@ static long bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key,
 unlock:
 	kfree(tlinks);
 	mutex_unlock(&st_map->lock);
+	if (!err)
+		bpf_struct_ops_map_ksyms_add(st_map);
 	return err;
 }
 
@@ -883,6 +922,10 @@ static void bpf_struct_ops_map_free(struct bpf_map *map)
 	 */
 	synchronize_rcu_mult(call_rcu, call_rcu_tasks);
 
+	/* no trampoline in the map is running anymore, delete symbols */
+	bpf_struct_ops_map_ksyms_del(st_map);
+	synchronize_rcu();
+
 	__bpf_struct_ops_map_free(map);
 }
 
@@ -895,6 +938,19 @@ static int bpf_struct_ops_map_alloc_check(union bpf_attr *attr)
 	return 0;
 }
 
+static u32 count_func_ptrs(const struct btf *btf, const struct btf_type *t)
+{
+	int i;
+	u32 count;
+	const struct btf_member *member;
+
+	count = 0;
+	for_each_member(i, t, member)
+		if (btf_type_resolve_func_ptr(btf, member->type, NULL))
+			count++;
+	return count;
+}
+
 static struct bpf_map *bpf_struct_ops_map_alloc(union bpf_attr *attr)
 {
 	const struct bpf_struct_ops_desc *st_ops_desc;
@@ -905,6 +961,8 @@ static struct bpf_map *bpf_struct_ops_map_alloc(union bpf_attr *attr)
 	struct bpf_map *map;
 	struct btf *btf;
 	int ret;
+	size_t ksyms_offset;
+	u32 ksyms_cnt;
 
 	if (attr->map_flags & BPF_F_VTYPE_BTF_OBJ_FD) {
 		/* The map holds btf for its whole life time. */
@@ -951,6 +1009,11 @@ static struct bpf_map *bpf_struct_ops_map_alloc(union bpf_attr *attr)
 		 */
 		(vt->size - sizeof(struct bpf_struct_ops_value));
 
+	st_map_size = round_up(st_map_size, sizeof(struct bpf_ksym));
+	ksyms_offset = st_map_size;
+	ksyms_cnt = count_func_ptrs(btf, t);
+	st_map_size += ksyms_cnt * sizeof(struct bpf_ksym);
+
 	st_map = bpf_map_area_alloc(st_map_size, NUMA_NO_NODE);
 	if (!st_map) {
 		ret = -ENOMEM;
@@ -958,6 +1021,8 @@ static struct bpf_map *bpf_struct_ops_map_alloc(union bpf_attr *attr)
 	}
 
 	st_map->st_ops_desc = st_ops_desc;
+	st_map->ksyms = (void *)st_map + ksyms_offset;
+	st_map->ksyms_cnt = ksyms_cnt;
 	map = &st_map->map;
 
 	st_map->uvalue = bpf_map_area_alloc(vt->size, NUMA_NO_NODE);
diff --git a/kernel/bpf/dispatcher.c b/kernel/bpf/dispatcher.c
index 70fb82bf1637..b77db7413f8c 100644
--- a/kernel/bpf/dispatcher.c
+++ b/kernel/bpf/dispatcher.c
@@ -154,7 +154,8 @@ void bpf_dispatcher_change_prog(struct bpf_dispatcher *d, struct bpf_prog *from,
 			d->image = NULL;
 			goto out;
 		}
-		bpf_image_ksym_add(d->image, PAGE_SIZE, &d->ksym);
+		bpf_image_ksym_init(d->image, PAGE_SIZE, &d->ksym);
+		bpf_image_ksym_add(&d->ksym);
 	}
 
 	prev_num_progs = d->num_progs;
diff --git a/kernel/bpf/trampoline.c b/kernel/bpf/trampoline.c
index 9f36c049f4c2..ecdd2660561f 100644
--- a/kernel/bpf/trampoline.c
+++ b/kernel/bpf/trampoline.c
@@ -115,10 +115,14 @@ bool bpf_prog_has_trampoline(const struct bpf_prog *prog)
 		(ptype == BPF_PROG_TYPE_LSM && eatype == BPF_LSM_MAC);
 }
 
-void bpf_image_ksym_add(void *data, unsigned int size, struct bpf_ksym *ksym)
+void bpf_image_ksym_init(void *data, unsigned int size, struct bpf_ksym *ksym)
 {
 	ksym->start = (unsigned long) data;
 	ksym->end = ksym->start + size;
+}
+
+void bpf_image_ksym_add(struct bpf_ksym *ksym)
+{
 	bpf_ksym_add(ksym);
 	perf_event_ksymbol(PERF_RECORD_KSYMBOL_TYPE_BPF, ksym->start,
 			   PAGE_SIZE, false, ksym->name);
@@ -377,7 +381,8 @@ static struct bpf_tramp_image *bpf_tramp_image_alloc(u64 key, int size)
 	ksym = &im->ksym;
 	INIT_LIST_HEAD_RCU(&ksym->lnode);
 	snprintf(ksym->name, KSYM_NAME_LEN, "bpf_trampoline_%llu", key);
-	bpf_image_ksym_add(image, size, ksym);
+	bpf_image_ksym_init(image, size, ksym);
+	bpf_image_ksym_add(ksym);
 	return im;
 
 out_free_image:
-- 
2.39.5

Powered by blists - more mailing lists