linux-kernel - [PATCH bpf-next v1 1/4] bpf: Support reporting BPF htab map's used size for monitoring

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20221105025146.238209-2-horenchuang@bytedance.com>
Date:   Sat,  5 Nov 2022 02:51:43 +0000
From:   "Ho-Ren (Jack) Chuang" <horenchuang@...edance.com>
To:     Alexei Starovoitov <ast@...nel.org>,
        Alexei Starovoitov <alexei.starovoitov@...il.com>,
        Hao Luo <haoluo@...gle.com>, Jiri Olsa <jolsa@...nel.org>,
        Jiri Olsa <olsajiri@...il.com>,
        Andrii Nakryiko <andrii@...nel.org>,
        Daniel Borkmann <daniel@...earbox.net>,
        John Fastabend <john.fastabend@...il.com>,
        Martin KaFai Lau <martin.lau@...ux.dev>,
        Song Liu <song@...nel.org>, Yonghong Song <yhs@...com>,
        KP Singh <kpsingh@...nel.org>,
        Stanislav Fomichev <sdf@...gle.com>,
        Quentin Monnet <quentin@...valent.com>,
        Mykola Lysenko <mykolal@...com>, Shuah Khan <shuah@...nel.org>,
        Nathan Chancellor <nathan@...nel.org>,
        Nick Desaulniers <ndesaulniers@...gle.com>,
        Tom Rix <trix@...hat.com>,
        Joanne Koong <joannelkoong@...il.com>,
        Kui-Feng Lee <kuifeng@...com>,
        Lorenzo Bianconi <lorenzo@...nel.org>,
        Maxim Mikityanskiy <maximmi@...dia.com>,
        Hao Xiang <hao.xiang@...edance.com>,
        Punit Agrawal <punit.agrawal@...edance.com>,
        Yifei Ma <yifeima@...edance.com>,
        Xiaoning Ding <xiaoning.ding@...edance.com>,
        bpf@...r.kernel.org
Cc:     Ho-Ren Chuang <horenc@...edu>,
        Ho-Ren Chuang <horenchuang@...edance.com>,
        linux-kernel@...r.kernel.org, linux-kselftest@...r.kernel.org,
        llvm@...ts.linux.dev
Subject: [PATCH bpf-next v1 1/4] bpf: Support reporting BPF htab map's used size for monitoring

Expose BPF htab map's used size by counting accessed or allocated/freed
elements to userspace.

Leverage the htab->count value for both preallocated and
dynamically allocated maps. Expose the value to a new field
"used_entries" in a userspace struct bpf_map_info  to allow monitoring.
Support hash table type (BPF_MAP_TYPE_HASH).

Signed-off-by: Ho-Ren (Jack) Chuang <horenchuang@...edance.com>
---
 include/linux/bpf.h      |  1 +
 include/uapi/linux/bpf.h |  1 +
 kernel/bpf/hashtab.c     | 19 +++++++++++++++++++
 kernel/bpf/syscall.c     |  2 ++
 4 files changed, 23 insertions(+)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 9e7d46d16032..82ee14139b69 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -97,6 +97,7 @@ struct bpf_map_ops {
 	int (*map_pop_elem)(struct bpf_map *map, void *value);
 	int (*map_peek_elem)(struct bpf_map *map, void *value);
 	void *(*map_lookup_percpu_elem)(struct bpf_map *map, void *key, u32 cpu);
+	u32 (*map_get_used_elem)(struct bpf_map *map);
 
 	/* funcs called by prog_array and perf_event_array map */
 	void *(*map_fd_get_ptr)(struct bpf_map *map, struct file *map_file,
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 17f61338f8f8..63659368cf0e 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -6215,6 +6215,7 @@ struct bpf_map_info {
 	__u32 id;
 	__u32 key_size;
 	__u32 value_size;
+	__u32 used_entries;
 	__u32 max_entries;
 	__u32 map_flags;
 	char  name[BPF_OBJ_NAME_LEN];
diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index ed3f8a53603b..bc9c00b92e57 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -913,6 +913,7 @@ static void free_htab_elem(struct bpf_htab *htab, struct htab_elem *l)
 	if (htab_is_prealloc(htab)) {
 		check_and_free_fields(htab, l);
 		__pcpu_freelist_push(&htab->freelist, &l->fnode);
+		dec_elem_count(htab);
 	} else {
 		dec_elem_count(htab);
 		htab_elem_free(htab, l);
@@ -994,6 +995,7 @@ static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key,
 			if (!l)
 				return ERR_PTR(-E2BIG);
 			l_new = container_of(l, struct htab_elem, fnode);
+			inc_elem_count(htab);
 		}
 	} else {
 		if (is_map_full(htab))
@@ -2186,6 +2188,22 @@ static int bpf_for_each_hash_elem(struct bpf_map *map, bpf_callback_t callback_f
 	return num_elems;
 }
 
+u32 htab_map_get_used_elem(struct bpf_map *map)
+{
+	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
+
+	/* The elem count may temporarily go beyond the max after
+	 * inc_elem_count() but before dec_elem_count().
+	 */
+	if (htab->use_percpu_counter)
+		return min_t(u32, htab->map.max_entries,
+				percpu_counter_sum(&htab->pcount) +
+							atomic_read(&htab->count));
+	else
+		return min_t(u32, htab->map.max_entries,
+							atomic_read(&htab->count));
+}
+
 BTF_ID_LIST_SINGLE(htab_map_btf_ids, struct, bpf_htab)
 const struct bpf_map_ops htab_map_ops = {
 	.map_meta_equal = bpf_map_meta_equal,
@@ -2202,6 +2220,7 @@ const struct bpf_map_ops htab_map_ops = {
 	.map_seq_show_elem = htab_map_seq_show_elem,
 	.map_set_for_each_callback_args = map_set_for_each_callback_args,
 	.map_for_each_callback = bpf_for_each_hash_elem,
+	.map_get_used_elem = htab_map_get_used_elem,
 	BATCH_OPS(htab),
 	.map_btf_id = &htab_map_btf_ids[0],
 	.iter_seq_info = &iter_seq_info,
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 7b373a5e861f..ea4828bb22ac 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -4203,6 +4203,8 @@ static int bpf_map_get_info_by_fd(struct file *file,
 	info.map_flags = map->map_flags;
 	info.map_extra = map->map_extra;
 	memcpy(info.name, map->name, sizeof(map->name));
+	if (map->ops->map_get_used_elem)
+		info.used_entries = map->ops->map_get_used_elem(map);
 
 	if (map->btf) {
 		info.btf_id = btf_obj_id(map->btf);
-- 
Ho-Ren (Jack) Chuang