1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
|
// SPDX-License-Identifier: GPL-2.0
/* Copyright (c) 2019 Facebook */
#include <linux/bpf.h>
#include <netinet/in.h>
#include <stdbool.h>
#include <bpf/bpf_helpers.h>
#include <bpf/bpf_endian.h>
enum bpf_addr_array_idx {
ADDR_SRV_IDX,
ADDR_CLI_IDX,
__NR_BPF_ADDR_ARRAY_IDX,
};
enum bpf_result_array_idx {
EGRESS_SRV_IDX,
EGRESS_CLI_IDX,
INGRESS_LISTEN_IDX,
__NR_BPF_RESULT_ARRAY_IDX,
};
enum bpf_linum_array_idx {
EGRESS_LINUM_IDX,
INGRESS_LINUM_IDX,
__NR_BPF_LINUM_ARRAY_IDX,
};
struct {
__uint(type, BPF_MAP_TYPE_ARRAY);
__uint(max_entries, __NR_BPF_ADDR_ARRAY_IDX);
__type(key, __u32);
__type(value, struct sockaddr_in6);
} addr_map SEC(".maps");
struct {
__uint(type, BPF_MAP_TYPE_ARRAY);
__uint(max_entries, __NR_BPF_RESULT_ARRAY_IDX);
__type(key, __u32);
__type(value, struct bpf_sock);
} sock_result_map SEC(".maps");
struct {
__uint(type, BPF_MAP_TYPE_ARRAY);
__uint(max_entries, __NR_BPF_RESULT_ARRAY_IDX);
__type(key, __u32);
__type(value, struct bpf_tcp_sock);
} tcp_sock_result_map SEC(".maps");
struct {
__uint(type, BPF_MAP_TYPE_ARRAY);
__uint(max_entries, __NR_BPF_LINUM_ARRAY_IDX);
__type(key, __u32);
__type(value, __u32);
} linum_map SEC(".maps");
struct bpf_spinlock_cnt {
struct bpf_spin_lock lock;
__u32 cnt;
};
struct {
__uint(type, BPF_MAP_TYPE_SK_STORAGE);
__uint(map_flags, BPF_F_NO_PREALLOC);
__type(key, int);
__type(value, struct bpf_spinlock_cnt);
} sk_pkt_out_cnt SEC(".maps");
struct {
__uint(type, BPF_MAP_TYPE_SK_STORAGE);
__uint(map_flags, BPF_F_NO_PREALLOC);
__type(key, int);
__type(value, struct bpf_spinlock_cnt);
} sk_pkt_out_cnt10 SEC(".maps");
static bool is_loopback6(__u32 *a6)
{
return !a6[0] && !a6[1] && !a6[2] && a6[3] == bpf_htonl(1);
}
static void skcpy(struct bpf_sock *dst,
const struct bpf_sock *src)
{
dst->bound_dev_if = src->bound_dev_if;
dst->family = src->family;
dst->type = src->type;
dst->protocol = src->protocol;
dst->mark = src->mark;
dst->priority = src->priority;
dst->src_ip4 = src->src_ip4;
dst->src_ip6[0] = src->src_ip6[0];
dst->src_ip6[1] = src->src_ip6[1];
dst->src_ip6[2] = src->src_ip6[2];
dst->src_ip6[3] = src->src_ip6[3];
dst->src_port = src->src_port;
dst->dst_ip4 = src->dst_ip4;
dst->dst_ip6[0] = src->dst_ip6[0];
dst->dst_ip6[1] = src->dst_ip6[1];
dst->dst_ip6[2] = src->dst_ip6[2];
dst->dst_ip6[3] = src->dst_ip6[3];
dst->dst_port = src->dst_port;
dst->state = src->state;
}
static void tpcpy(struct bpf_tcp_sock *dst,
const struct bpf_tcp_sock *src)
{
dst->snd_cwnd = src->snd_cwnd;
dst->srtt_us = src->srtt_us;
dst->rtt_min = src->rtt_min;
dst->snd_ssthresh = src->snd_ssthresh;
dst->rcv_nxt = src->rcv_nxt;
dst->snd_nxt = src->snd_nxt;
dst->snd_una = src->snd_una;
dst->mss_cache = src->mss_cache;
dst->ecn_flags = src->ecn_flags;
dst->rate_delivered = src->rate_delivered;
dst->rate_interval_us = src->rate_interval_us;
dst->packets_out = src->packets_out;
dst->retrans_out = src->retrans_out;
dst->total_retrans = src->total_retrans;
dst->segs_in = src->segs_in;
dst->data_segs_in = src->data_segs_in;
dst->segs_out = src->segs_out;
dst->data_segs_out = src->data_segs_out;
dst->lost_out = src->lost_out;
dst->sacked_out = src->sacked_out;
dst->bytes_received = src->bytes_received;
dst->bytes_acked = src->bytes_acked;
}
#define RETURN { \
linum = __LINE__; \
bpf_map_update_elem(&linum_map, &linum_idx, &linum, 0); \
return 1; \
}
SEC("cgroup_skb/egress")
int egress_read_sock_fields(struct __sk_buff *skb)
{
struct bpf_spinlock_cnt cli_cnt_init = { .lock = 0, .cnt = 0xeB9F };
__u32 srv_idx = ADDR_SRV_IDX, cli_idx = ADDR_CLI_IDX, result_idx;
struct bpf_spinlock_cnt *pkt_out_cnt, *pkt_out_cnt10;
struct sockaddr_in6 *srv_sa6, *cli_sa6;
struct bpf_tcp_sock *tp, *tp_ret;
struct bpf_sock *sk, *sk_ret;
__u32 linum, linum_idx;
linum_idx = EGRESS_LINUM_IDX;
sk = skb->sk;
if (!sk || sk->state == 10)
RETURN;
sk = bpf_sk_fullsock(sk);
if (!sk || sk->family != AF_INET6 || sk->protocol != IPPROTO_TCP ||
!is_loopback6(sk->src_ip6))
RETURN;
tp = bpf_tcp_sock(sk);
if (!tp)
RETURN;
srv_sa6 = bpf_map_lookup_elem(&addr_map, &srv_idx);
cli_sa6 = bpf_map_lookup_elem(&addr_map, &cli_idx);
if (!srv_sa6 || !cli_sa6)
RETURN;
if (sk->src_port == bpf_ntohs(srv_sa6->sin6_port))
result_idx = EGRESS_SRV_IDX;
else if (sk->src_port == bpf_ntohs(cli_sa6->sin6_port))
result_idx = EGRESS_CLI_IDX;
else
RETURN;
sk_ret = bpf_map_lookup_elem(&sock_result_map, &result_idx);
tp_ret = bpf_map_lookup_elem(&tcp_sock_result_map, &result_idx);
if (!sk_ret || !tp_ret)
RETURN;
skcpy(sk_ret, sk);
tpcpy(tp_ret, tp);
if (result_idx == EGRESS_SRV_IDX) {
/* The userspace has created it for srv sk */
pkt_out_cnt = bpf_sk_storage_get(&sk_pkt_out_cnt, sk, 0, 0);
pkt_out_cnt10 = bpf_sk_storage_get(&sk_pkt_out_cnt10, sk,
0, 0);
} else {
pkt_out_cnt = bpf_sk_storage_get(&sk_pkt_out_cnt, sk,
&cli_cnt_init,
BPF_SK_STORAGE_GET_F_CREATE);
pkt_out_cnt10 = bpf_sk_storage_get(&sk_pkt_out_cnt10,
sk, &cli_cnt_init,
BPF_SK_STORAGE_GET_F_CREATE);
}
if (!pkt_out_cnt || !pkt_out_cnt10)
RETURN;
/* Even both cnt and cnt10 have lock defined in their BTF,
* intentionally one cnt takes lock while one does not
* as a test for the spinlock support in BPF_MAP_TYPE_SK_STORAGE.
*/
pkt_out_cnt->cnt += 1;
bpf_spin_lock(&pkt_out_cnt10->lock);
pkt_out_cnt10->cnt += 10;
bpf_spin_unlock(&pkt_out_cnt10->lock);
RETURN;
}
SEC("cgroup_skb/ingress")
int ingress_read_sock_fields(struct __sk_buff *skb)
{
__u32 srv_idx = ADDR_SRV_IDX, result_idx = INGRESS_LISTEN_IDX;
struct bpf_tcp_sock *tp, *tp_ret;
struct bpf_sock *sk, *sk_ret;
struct sockaddr_in6 *srv_sa6;
__u32 linum, linum_idx;
linum_idx = INGRESS_LINUM_IDX;
sk = skb->sk;
if (!sk || sk->family != AF_INET6 || !is_loopback6(sk->src_ip6))
RETURN;
srv_sa6 = bpf_map_lookup_elem(&addr_map, &srv_idx);
if (!srv_sa6 || sk->src_port != bpf_ntohs(srv_sa6->sin6_port))
RETURN;
if (sk->state != 10 && sk->state != 12)
RETURN;
sk = bpf_get_listener_sock(sk);
if (!sk)
RETURN;
tp = bpf_tcp_sock(sk);
if (!tp)
RETURN;
sk_ret = bpf_map_lookup_elem(&sock_result_map, &result_idx);
tp_ret = bpf_map_lookup_elem(&tcp_sock_result_map, &result_idx);
if (!sk_ret || !tp_ret)
RETURN;
skcpy(sk_ret, sk);
tpcpy(tp_ret, tp);
RETURN;
}
char _license[] SEC("license") = "GPL";
|
