netd: Use common BPF helper for C program

am: 14ba88881f

Change-Id: I8001088a9d0cc054d7fef2887fc8ef7bbf9e76cc
This commit is contained in:
Joel Fernandes (Google) 2019-02-14 13:52:14 -08:00 committed by android-build-merger
commit 1b9a6c52e1
3 changed files with 65 additions and 38 deletions

View file

@ -0,0 +1,49 @@
/* Common BPF helpers to be used by all BPF programs loaded by Android */
#include <linux/bpf.h>
#include <stdbool.h>
#include <stdint.h>
/* place things in different elf sections */
#define SEC(NAME) __attribute__((section(NAME), used))
/*
* Helper functions called from eBPF programs written in C. These are
* implemented in the kernel sources.
*/
/* generic functions */
static void* (*bpf_map_lookup_elem)(void* map, void* key) = (void*) BPF_FUNC_map_lookup_elem;
static int (*bpf_map_update_elem)(void* map, void* key, void* value,
unsigned long long flags) = (void*) BPF_FUNC_map_update_elem;
static int (*bpf_map_delete_elem)(void* map, void* key) = (void*) BPF_FUNC_map_delete_elem;
static int (*bpf_probe_read)(void* dst, int size, void* unsafe_ptr) = (void*) BPF_FUNC_probe_read;
static unsigned long long (*bpf_ktime_get_ns)(void) = (void*) BPF_FUNC_ktime_get_ns;
static int (*bpf_trace_printk)(const char* fmt, int fmt_size, ...) = (void*) BPF_FUNC_trace_printk;
static unsigned long long (*bpf_get_current_pid_tgid)(void) = (void*) BPF_FUNC_get_current_pid_tgid;
static unsigned long long (*bpf_get_current_uid_gid)(void) = (void*) BPF_FUNC_get_current_uid_gid;
static unsigned long long (*bpf_get_smp_processor_id)(void) = (void*) BPF_FUNC_get_smp_processor_id;
/* networking */
static uint64_t (*bpf_get_socket_cookie)(struct __sk_buff* skb) = (void*)BPF_FUNC_get_socket_cookie;
static uint32_t (*bpf_get_socket_uid)(struct __sk_buff* skb) = (void*)BPF_FUNC_get_socket_uid;
static int (*bpf_skb_load_bytes)(struct __sk_buff* skb, int off, void* to,
int len) = (void*)BPF_FUNC_skb_load_bytes;
/*
* Map structure to be used by Android eBPF C programs. The Android eBPF loader
* uses this structure from eBPF object to create maps at boot time.
*
* The eBPF C program should define structure in the maps section using
* SEC("maps") otherwise it will be ignored by the eBPF loader.
*
* For example:
* struct bpf_map_def SEC("maps") mymap { .type=... , .key_size=... }
*/
struct bpf_map_def {
unsigned int type;
unsigned int key_size;
unsigned int value_size;
unsigned int max_entries;
unsigned int map_flags;
unsigned int pad1;
unsigned int pad2;
};

View file

@ -43,17 +43,17 @@ int xt_bpf_ingress_prog(struct __sk_buff* skb) {
SEC("skfilter/whitelist/xtbpf")
int xt_bpf_whitelist_prog(struct __sk_buff* skb) {
uint32_t sock_uid = get_socket_uid(skb);
uint32_t sock_uid = bpf_get_socket_uid(skb);
if (is_system_uid(sock_uid)) return BPF_MATCH;
uint8_t* whitelistMatch = find_map_entry(&uid_owner_map, &sock_uid);
uint8_t* whitelistMatch = bpf_map_lookup_elem(&uid_owner_map, &sock_uid);
if (whitelistMatch) return *whitelistMatch & HAPPY_BOX_MATCH;
return BPF_NOMATCH;
}
SEC("skfilter/blacklist/xtbpf")
int xt_bpf_blacklist_prog(struct __sk_buff* skb) {
uint32_t sock_uid = get_socket_uid(skb);
uint8_t* blacklistMatch = find_map_entry(&uid_owner_map, &sock_uid);
uint32_t sock_uid = bpf_get_socket_uid(skb);
uint8_t* blacklistMatch = bpf_map_lookup_elem(&uid_owner_map, &sock_uid);
if (blacklistMatch) return *blacklistMatch & PENALTY_BOX_MATCH;
return BPF_NOMATCH;
}
@ -75,7 +75,7 @@ int inet_socket_create(struct bpf_sock* sk) {
* run time. See UserHandle#isSameApp for detail.
*/
uint32_t appId = (gid_uid & 0xffffffff) % PER_USER_RANGE;
uint8_t* internetPermission = find_map_entry(&uid_permission_map, &appId);
uint8_t* internetPermission = bpf_map_lookup_elem(&uid_permission_map, &appId);
if (internetPermission) return *internetPermission & ALLOW_SOCK_CREATE;
return NO_PERMISSION;
}

View file

@ -19,6 +19,7 @@
* program.
*/
#include <bpf_helpers.h>
#include <linux/bpf.h>
#include <linux/if.h>
#include <linux/if_ether.h>
@ -30,8 +31,6 @@
#include <stdint.h>
#include "netdbpf/bpf_shared.h"
#define SEC(NAME) __attribute__((section(NAME), used))
struct uid_tag {
uint32_t uid;
uint32_t tag;
@ -55,17 +54,6 @@ struct IfaceValue {
char name[IFNAMSIZ];
};
/* helper functions called from eBPF programs written in C */
static void* (*find_map_entry)(void* map, void* key) = (void*)BPF_FUNC_map_lookup_elem;
static int (*write_to_map_entry)(void* map, void* key, void* value,
uint64_t flags) = (void*)BPF_FUNC_map_update_elem;
static int (*delete_map_entry)(void* map, void* key) = (void*)BPF_FUNC_map_delete_elem;
static uint64_t (*get_socket_cookie)(struct __sk_buff* skb) = (void*)BPF_FUNC_get_socket_cookie;
static uint32_t (*get_socket_uid)(struct __sk_buff* skb) = (void*)BPF_FUNC_get_socket_uid;
static int (*bpf_skb_load_bytes)(struct __sk_buff* skb, int off, void* to,
int len) = (void*)BPF_FUNC_skb_load_bytes;
static uint64_t (*bpf_get_current_uid_gid)(void) = (void*)BPF_FUNC_get_current_uid_gid;
// This is defined for cgroup bpf filter only.
#define BPF_PASS 1
#define BPF_DROP 0
@ -83,16 +71,6 @@ static uint64_t (*bpf_get_current_uid_gid)(void) = (void*)BPF_FUNC_get_current_u
#define TCP_FLAG_OFF 13
#define RST_OFFSET 2
/* loader usage */
struct bpf_map_def {
unsigned int type;
unsigned int key_size;
unsigned int value_size;
unsigned int max_entries;
unsigned int map_flags;
unsigned int pad[2];
};
struct bpf_map_def SEC("maps") cookie_tag_map = {
.type = BPF_MAP_TYPE_HASH,
.key_size = sizeof(uint64_t),
@ -163,11 +141,11 @@ static __always_inline int is_system_uid(uint32_t uid) {
static __always_inline inline void bpf_update_stats(struct __sk_buff* skb, struct bpf_map_def* map,
int direction, void* key) {
struct stats_value* value;
value = find_map_entry(map, key);
value = bpf_map_lookup_elem(map, key);
if (!value) {
struct stats_value newValue = {};
write_to_map_entry(map, key, &newValue, BPF_NOEXIST);
value = find_map_entry(map, key);
bpf_map_update_elem(map, key, &newValue, BPF_NOEXIST);
value = bpf_map_lookup_elem(map, key);
}
if (value) {
if (direction == BPF_EGRESS) {
@ -221,7 +199,7 @@ static inline bool skip_owner_match(struct __sk_buff* skb) {
static __always_inline BpfConfig getConfig(uint32_t configKey) {
uint32_t mapSettingKey = configKey;
BpfConfig* config = find_map_entry(&configuration_map, &mapSettingKey);
BpfConfig* config = bpf_map_lookup_elem(&configuration_map, &mapSettingKey);
if (!config) {
// Couldn't read configuration entry. Assume everything is disabled.
return DEFAULT_CONFIG;
@ -239,7 +217,7 @@ static inline int bpf_owner_match(struct __sk_buff* skb, uint32_t uid) {
return BPF_PASS;
}
uint8_t* uidEntry = find_map_entry(&uid_owner_map, &uid);
uint8_t* uidEntry = bpf_map_lookup_elem(&uid_owner_map, &uid);
uint8_t uidRules = uidEntry ? *uidEntry : 0;
if ((enabledRules & DOZABLE_MATCH) && !(uidRules & DOZABLE_MATCH)) {
return BPF_DROP;
@ -263,7 +241,7 @@ static __always_inline inline void update_stats_with_config(struct __sk_buff* sk
}
static __always_inline inline int bpf_traffic_account(struct __sk_buff* skb, int direction) {
uint32_t sock_uid = get_socket_uid(skb);
uint32_t sock_uid = bpf_get_socket_uid(skb);
int match = bpf_owner_match(skb, sock_uid);
if ((direction == BPF_EGRESS) && (match == BPF_DROP)) {
// If an outbound packet is going to be dropped, we do not count that
@ -271,8 +249,8 @@ static __always_inline inline int bpf_traffic_account(struct __sk_buff* skb, int
return match;
}
uint64_t cookie = get_socket_cookie(skb);
struct uid_tag* utag = find_map_entry(&cookie_tag_map, &cookie);
uint64_t cookie = bpf_get_socket_cookie(skb);
struct uid_tag* utag = bpf_map_lookup_elem(&cookie_tag_map, &cookie);
uint32_t uid, tag;
if (utag) {
uid = utag->uid;
@ -284,11 +262,11 @@ static __always_inline inline int bpf_traffic_account(struct __sk_buff* skb, int
struct stats_key key = {.uid = uid, .tag = tag, .counterSet = 0, .ifaceIndex = skb->ifindex};
uint8_t* counterSet = find_map_entry(&uid_counterset_map, &uid);
uint8_t* counterSet = bpf_map_lookup_elem(&uid_counterset_map, &uid);
if (counterSet) key.counterSet = (uint32_t)*counterSet;
uint32_t mapSettingKey = CURRENT_STATS_MAP_CONFIGURATION_KEY;
uint8_t* selectedMap = find_map_entry(&configuration_map, &mapSettingKey);
uint8_t* selectedMap = bpf_map_lookup_elem(&configuration_map, &mapSettingKey);
if (!selectedMap) {
return match;
}