Files
linux/tools/testing/selftests/bpf/progs/connect4_prog.c
Martin KaFai Lau c075c9c4af selftests/bpf: Remove the bpf_tcp_helpers.h usages from other non tcp-cc tests
The patch removes the remaining bpf_tcp_helpers.h usages in the
non tcp-cc networking tests. It either replaces it with bpf_tracing_net.h
or just removed it because the test is not actually using any
kernel sockets. For the later, the missing macro (mainly SOL_TCP) is
defined locally.

An exception is the test_sock_fields which is testing
the "struct bpf_sock" type instead of the kernel sock type.
Whenever "vmlinux.h" is used instead, it hits a verifier
error on doing arithmetic on the sock_common pointer:

; return !a6[0] && !a6[1] && !a6[2] && a6[3] == bpf_htonl(1); @ test_sock_fields.c:54
21: (61) r2 = *(u32 *)(r1 +28)        ; R1_w=sock_common() R2_w=scalar(smin=0,smax=umax=0xffffffff,var_off=(0x0; 0xffffffff))
22: (56) if w2 != 0x0 goto pc-6       ; R2_w=0
23: (b7) r3 = 28                      ; R3_w=28
24: (bf) r2 = r1                      ; R1_w=sock_common() R2_w=sock_common()
25: (0f) r2 += r3
R2 pointer arithmetic on sock_common prohibited

Hence, instead of including bpf_tracing_net.h, the test_sock_fields test
defines a tcp_sock with one lsndtime field in it.

Another highlight is, in sockopt_qos_to_cc.c, the tcp_cc_eq()
is replaced by bpf_strncmp(). tcp_cc_eq() was a workaround
in bpf_tcp_helpers.h before bpf_strncmp had been added.

The SOL_IPV6 addition to bpf_tracing_net.h is needed by the
test_tcpbpf_kern test.

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20240509175026.3423614-10-martin.lau@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2024-05-09 11:13:12 -07:00

203 lines
4.5 KiB
C

// SPDX-License-Identifier: GPL-2.0
// Copyright (c) 2018 Facebook
#include <string.h>
#include <linux/stddef.h>
#include <linux/bpf.h>
#include <linux/in.h>
#include <linux/in6.h>
#include <linux/tcp.h>
#include <linux/if.h>
#include <errno.h>
#include <bpf/bpf_helpers.h>
#include <bpf/bpf_endian.h>
#define SRC_REWRITE_IP4 0x7f000004U
#define DST_REWRITE_IP4 0x7f000001U
#define DST_REWRITE_PORT4 4444
#ifndef TCP_CA_NAME_MAX
#define TCP_CA_NAME_MAX 16
#endif
#ifndef TCP_NOTSENT_LOWAT
#define TCP_NOTSENT_LOWAT 25
#endif
#ifndef IFNAMSIZ
#define IFNAMSIZ 16
#endif
#ifndef SOL_TCP
#define SOL_TCP 6
#endif
__attribute__ ((noinline)) __weak
int do_bind(struct bpf_sock_addr *ctx)
{
struct sockaddr_in sa = {};
sa.sin_family = AF_INET;
sa.sin_port = bpf_htons(0);
sa.sin_addr.s_addr = bpf_htonl(SRC_REWRITE_IP4);
if (bpf_bind(ctx, (struct sockaddr *)&sa, sizeof(sa)) != 0)
return 0;
return 1;
}
static __inline int verify_cc(struct bpf_sock_addr *ctx,
char expected[TCP_CA_NAME_MAX])
{
char buf[TCP_CA_NAME_MAX];
int i;
if (bpf_getsockopt(ctx, SOL_TCP, TCP_CONGESTION, &buf, sizeof(buf)))
return 1;
for (i = 0; i < TCP_CA_NAME_MAX; i++) {
if (buf[i] != expected[i])
return 1;
if (buf[i] == 0)
break;
}
return 0;
}
static __inline int set_cc(struct bpf_sock_addr *ctx)
{
char reno[TCP_CA_NAME_MAX] = "reno";
char cubic[TCP_CA_NAME_MAX] = "cubic";
if (bpf_setsockopt(ctx, SOL_TCP, TCP_CONGESTION, &reno, sizeof(reno)))
return 1;
if (verify_cc(ctx, reno))
return 1;
if (bpf_setsockopt(ctx, SOL_TCP, TCP_CONGESTION, &cubic, sizeof(cubic)))
return 1;
if (verify_cc(ctx, cubic))
return 1;
return 0;
}
static __inline int bind_to_device(struct bpf_sock_addr *ctx)
{
char veth1[IFNAMSIZ] = "test_sock_addr1";
char veth2[IFNAMSIZ] = "test_sock_addr2";
char missing[IFNAMSIZ] = "nonexistent_dev";
char del_bind[IFNAMSIZ] = "";
if (bpf_setsockopt(ctx, SOL_SOCKET, SO_BINDTODEVICE,
&veth1, sizeof(veth1)))
return 1;
if (bpf_setsockopt(ctx, SOL_SOCKET, SO_BINDTODEVICE,
&veth2, sizeof(veth2)))
return 1;
if (bpf_setsockopt(ctx, SOL_SOCKET, SO_BINDTODEVICE,
&missing, sizeof(missing)) != -ENODEV)
return 1;
if (bpf_setsockopt(ctx, SOL_SOCKET, SO_BINDTODEVICE,
&del_bind, sizeof(del_bind)))
return 1;
return 0;
}
static __inline int set_keepalive(struct bpf_sock_addr *ctx)
{
int zero = 0, one = 1;
if (bpf_setsockopt(ctx, SOL_SOCKET, SO_KEEPALIVE, &one, sizeof(one)))
return 1;
if (ctx->type == SOCK_STREAM) {
if (bpf_setsockopt(ctx, SOL_TCP, TCP_KEEPIDLE, &one, sizeof(one)))
return 1;
if (bpf_setsockopt(ctx, SOL_TCP, TCP_KEEPINTVL, &one, sizeof(one)))
return 1;
if (bpf_setsockopt(ctx, SOL_TCP, TCP_KEEPCNT, &one, sizeof(one)))
return 1;
if (bpf_setsockopt(ctx, SOL_TCP, TCP_SYNCNT, &one, sizeof(one)))
return 1;
if (bpf_setsockopt(ctx, SOL_TCP, TCP_USER_TIMEOUT, &one, sizeof(one)))
return 1;
}
if (bpf_setsockopt(ctx, SOL_SOCKET, SO_KEEPALIVE, &zero, sizeof(zero)))
return 1;
return 0;
}
static __inline int set_notsent_lowat(struct bpf_sock_addr *ctx)
{
int lowat = 65535;
if (ctx->type == SOCK_STREAM) {
if (bpf_setsockopt(ctx, SOL_TCP, TCP_NOTSENT_LOWAT, &lowat, sizeof(lowat)))
return 1;
}
return 0;
}
SEC("cgroup/connect4")
int connect_v4_prog(struct bpf_sock_addr *ctx)
{
struct bpf_sock_tuple tuple = {};
struct bpf_sock *sk;
/* Verify that new destination is available. */
memset(&tuple.ipv4.saddr, 0, sizeof(tuple.ipv4.saddr));
memset(&tuple.ipv4.sport, 0, sizeof(tuple.ipv4.sport));
tuple.ipv4.daddr = bpf_htonl(DST_REWRITE_IP4);
tuple.ipv4.dport = bpf_htons(DST_REWRITE_PORT4);
/* Bind to device and unbind it. */
if (bind_to_device(ctx))
return 0;
if (set_keepalive(ctx))
return 0;
if (set_notsent_lowat(ctx))
return 0;
if (ctx->type != SOCK_STREAM && ctx->type != SOCK_DGRAM)
return 0;
else if (ctx->type == SOCK_STREAM)
sk = bpf_sk_lookup_tcp(ctx, &tuple, sizeof(tuple.ipv4),
BPF_F_CURRENT_NETNS, 0);
else
sk = bpf_sk_lookup_udp(ctx, &tuple, sizeof(tuple.ipv4),
BPF_F_CURRENT_NETNS, 0);
if (!sk)
return 0;
if (sk->src_ip4 != tuple.ipv4.daddr ||
sk->src_port != DST_REWRITE_PORT4) {
bpf_sk_release(sk);
return 0;
}
bpf_sk_release(sk);
/* Rewrite congestion control. */
if (ctx->type == SOCK_STREAM && set_cc(ctx))
return 0;
/* Rewrite destination. */
ctx->user_ip4 = bpf_htonl(DST_REWRITE_IP4);
ctx->user_port = bpf_htons(DST_REWRITE_PORT4);
return do_bind(ctx) ? 1 : 0;
}
char _license[] SEC("license") = "GPL";