linux/tools/testing/selftests/bpf/network_helpers.h

281 lines
7.6 KiB
C
Raw Permalink Normal View History

/* SPDX-License-Identifier: GPL-2.0 */
#ifndef __NETWORK_HELPERS_H
#define __NETWORK_HELPERS_H
#include <arpa/inet.h>
#include <sys/socket.h>
#include <sys/types.h>
#include <linux/types.h>
typedef __u16 __sum16;
#include <linux/if_ether.h>
#include <linux/if_packet.h>
#include <linux/if_tun.h>
#include <linux/ip.h>
#include <linux/ipv6.h>
#include <linux/ethtool.h>
#include <linux/sockios.h>
#include <linux/err.h>
#include <netinet/tcp.h>
#include <netinet/udp.h>
#include <bpf/bpf_endian.h>
#include <net/if.h>
#include <stdio.h>
#define MAGIC_VAL 0x1234
#define NUM_ITER 100000
#define VIP_NUM 5
#define MAGIC_BYTES 123
struct network_helper_opts {
int timeout_ms;
int proto;
/* +ve: Passed to listen() as-is.
* 0: Default when the test does not set
* a particular value during the struct init.
* It is changed to 1 before passing to listen().
* Most tests only have one on-going connection.
* -ve: It is changed to 0 before passing to listen().
* It is useful to force syncookie without
* changing the "tcp_syncookies" sysctl from 1 to 2.
*/
int backlog;
int (*post_socket_cb)(int fd, void *opts);
void *cb_opts;
};
/* ipv4 test vector */
struct ipv4_packet {
struct ethhdr eth;
struct iphdr iph;
struct tcphdr tcp;
} __packed;
extern struct ipv4_packet pkt_v4;
/* ipv6 test vector */
struct ipv6_packet {
struct ethhdr eth;
struct ipv6hdr iph;
struct tcphdr tcp;
} __packed;
extern struct ipv6_packet pkt_v6;
int settimeo(int fd, int timeout_ms);
int start_server_str(int family, int type, const char *addr_str, __u16 port,
const struct network_helper_opts *opts);
bpf: selftests: A few improvements to network_helpers.c This patch makes a few changes to the network_helpers.c 1) Enforce SO_RCVTIMEO and SO_SNDTIMEO This patch enforces timeout to the network fds through setsockopt SO_RCVTIMEO and SO_SNDTIMEO. It will remove the need for SOCK_NONBLOCK that requires a more demanding timeout logic with epoll/select, e.g. epoll_create, epoll_ctrl, and then epoll_wait for timeout. That removes the need for connect_wait() from the cgroup_skb_sk_lookup.c. The needed change is made in cgroup_skb_sk_lookup.c. 2) start_server(): Add optional addr_str and port to start_server(). That removes the need of the start_server_with_port(). The caller can pass addr_str==NULL and/or port==0. I have a future tcp-hdr-opt test that will pass a non-NULL addr_str and it is in general useful for other future tests. "int timeout_ms" is also added to control the timeout on the "accept(listen_fd)". 3) connect_to_fd(): Fully use the server_fd. The server sock address has already been obtained from getsockname(server_fd). The sockaddr includes the family, so the "int family" arg is redundant. Since the server address is obtained from server_fd, there is little reason not to get the server's socket type from the server_fd also. getsockopt(server_fd) can be used to do that, so "int type" arg is also removed. "int timeout_ms" is added. 4) connect_fd_to_fd(): "int timeout_ms" is added. Some code is also refactored to connect_fd_to_addr() which is shared with connect_to_fd(). 5) Preserve errno: Some callers need to check errno, e.g. cgroup_skb_sk_lookup.c. Make changes to do it more consistently in save_errno_close() and log_err(). Signed-off-by: Martin KaFai Lau <kafai@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Yonghong Song <yhs@fb.com> Link: https://lore.kernel.org/bpf/20200702004852.2103003-1-kafai@fb.com
2020-07-01 17:48:52 -07:00
int start_server(int family, int type, const char *addr, __u16 port,
int timeout_ms);
int *start_reuseport_server(int family, int type, const char *addr_str,
__u16 port, int timeout_ms,
unsigned int nr_listens);
int start_server_addr(int type, const struct sockaddr_storage *addr, socklen_t len,
const struct network_helper_opts *opts);
void free_fds(int *fds, unsigned int nr_close_fds);
int client_socket(int family, int type,
const struct network_helper_opts *opts);
int connect_to_addr(int type, const struct sockaddr_storage *addr, socklen_t len,
const struct network_helper_opts *opts);
int connect_to_addr_str(int family, int type, const char *addr_str, __u16 port,
const struct network_helper_opts *opts);
bpf: selftests: A few improvements to network_helpers.c This patch makes a few changes to the network_helpers.c 1) Enforce SO_RCVTIMEO and SO_SNDTIMEO This patch enforces timeout to the network fds through setsockopt SO_RCVTIMEO and SO_SNDTIMEO. It will remove the need for SOCK_NONBLOCK that requires a more demanding timeout logic with epoll/select, e.g. epoll_create, epoll_ctrl, and then epoll_wait for timeout. That removes the need for connect_wait() from the cgroup_skb_sk_lookup.c. The needed change is made in cgroup_skb_sk_lookup.c. 2) start_server(): Add optional addr_str and port to start_server(). That removes the need of the start_server_with_port(). The caller can pass addr_str==NULL and/or port==0. I have a future tcp-hdr-opt test that will pass a non-NULL addr_str and it is in general useful for other future tests. "int timeout_ms" is also added to control the timeout on the "accept(listen_fd)". 3) connect_to_fd(): Fully use the server_fd. The server sock address has already been obtained from getsockname(server_fd). The sockaddr includes the family, so the "int family" arg is redundant. Since the server address is obtained from server_fd, there is little reason not to get the server's socket type from the server_fd also. getsockopt(server_fd) can be used to do that, so "int type" arg is also removed. "int timeout_ms" is added. 4) connect_fd_to_fd(): "int timeout_ms" is added. Some code is also refactored to connect_fd_to_addr() which is shared with connect_to_fd(). 5) Preserve errno: Some callers need to check errno, e.g. cgroup_skb_sk_lookup.c. Make changes to do it more consistently in save_errno_close() and log_err(). Signed-off-by: Martin KaFai Lau <kafai@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Yonghong Song <yhs@fb.com> Link: https://lore.kernel.org/bpf/20200702004852.2103003-1-kafai@fb.com
2020-07-01 17:48:52 -07:00
int connect_to_fd(int server_fd, int timeout_ms);
int connect_to_fd_opts(int server_fd, const struct network_helper_opts *opts);
bpf: selftests: A few improvements to network_helpers.c This patch makes a few changes to the network_helpers.c 1) Enforce SO_RCVTIMEO and SO_SNDTIMEO This patch enforces timeout to the network fds through setsockopt SO_RCVTIMEO and SO_SNDTIMEO. It will remove the need for SOCK_NONBLOCK that requires a more demanding timeout logic with epoll/select, e.g. epoll_create, epoll_ctrl, and then epoll_wait for timeout. That removes the need for connect_wait() from the cgroup_skb_sk_lookup.c. The needed change is made in cgroup_skb_sk_lookup.c. 2) start_server(): Add optional addr_str and port to start_server(). That removes the need of the start_server_with_port(). The caller can pass addr_str==NULL and/or port==0. I have a future tcp-hdr-opt test that will pass a non-NULL addr_str and it is in general useful for other future tests. "int timeout_ms" is also added to control the timeout on the "accept(listen_fd)". 3) connect_to_fd(): Fully use the server_fd. The server sock address has already been obtained from getsockname(server_fd). The sockaddr includes the family, so the "int family" arg is redundant. Since the server address is obtained from server_fd, there is little reason not to get the server's socket type from the server_fd also. getsockopt(server_fd) can be used to do that, so "int type" arg is also removed. "int timeout_ms" is added. 4) connect_fd_to_fd(): "int timeout_ms" is added. Some code is also refactored to connect_fd_to_addr() which is shared with connect_to_fd(). 5) Preserve errno: Some callers need to check errno, e.g. cgroup_skb_sk_lookup.c. Make changes to do it more consistently in save_errno_close() and log_err(). Signed-off-by: Martin KaFai Lau <kafai@fb.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Yonghong Song <yhs@fb.com> Link: https://lore.kernel.org/bpf/20200702004852.2103003-1-kafai@fb.com
2020-07-01 17:48:52 -07:00
int connect_fd_to_fd(int client_fd, int server_fd, int timeout_ms);
int fastopen_connect(int server_fd, const char *data, unsigned int data_len,
int timeout_ms);
selftests/bpf: Tests for BPF_SK_LOOKUP attach point Add tests to test_progs that exercise: - attaching/detaching/querying programs to BPF_SK_LOOKUP hook, - redirecting socket lookup to a socket selected by BPF program, - failing a socket lookup on BPF program's request, - error scenarios for selecting a socket from BPF program, - accessing BPF program context, - attaching and running multiple BPF programs. Run log: bash-5.0# ./test_progs -n 70 #70/1 query lookup prog:OK #70/2 TCP IPv4 redir port:OK #70/3 TCP IPv4 redir addr:OK #70/4 TCP IPv4 redir with reuseport:OK #70/5 TCP IPv4 redir skip reuseport:OK #70/6 TCP IPv6 redir port:OK #70/7 TCP IPv6 redir addr:OK #70/8 TCP IPv4->IPv6 redir port:OK #70/9 TCP IPv6 redir with reuseport:OK #70/10 TCP IPv6 redir skip reuseport:OK #70/11 UDP IPv4 redir port:OK #70/12 UDP IPv4 redir addr:OK #70/13 UDP IPv4 redir with reuseport:OK #70/14 UDP IPv4 redir skip reuseport:OK #70/15 UDP IPv6 redir port:OK #70/16 UDP IPv6 redir addr:OK #70/17 UDP IPv4->IPv6 redir port:OK #70/18 UDP IPv6 redir and reuseport:OK #70/19 UDP IPv6 redir skip reuseport:OK #70/20 TCP IPv4 drop on lookup:OK #70/21 TCP IPv6 drop on lookup:OK #70/22 UDP IPv4 drop on lookup:OK #70/23 UDP IPv6 drop on lookup:OK #70/24 TCP IPv4 drop on reuseport:OK #70/25 TCP IPv6 drop on reuseport:OK #70/26 UDP IPv4 drop on reuseport:OK #70/27 TCP IPv6 drop on reuseport:OK #70/28 sk_assign returns EEXIST:OK #70/29 sk_assign honors F_REPLACE:OK #70/30 sk_assign accepts NULL socket:OK #70/31 access ctx->sk:OK #70/32 narrow access to ctx v4:OK #70/33 narrow access to ctx v6:OK #70/34 sk_assign rejects TCP established:OK #70/35 sk_assign rejects UDP connected:OK #70/36 multi prog - pass, pass:OK #70/37 multi prog - drop, drop:OK #70/38 multi prog - pass, drop:OK #70/39 multi prog - drop, pass:OK #70/40 multi prog - pass, redir:OK #70/41 multi prog - redir, pass:OK #70/42 multi prog - drop, redir:OK #70/43 multi prog - redir, drop:OK #70/44 multi prog - redir, redir:OK #70 sk_lookup:OK Summary: 1/44 PASSED, 0 SKIPPED, 0 FAILED Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/20200717103536.397595-16-jakub@cloudflare.com
2020-07-17 12:35:36 +02:00
int make_sockaddr(int family, const char *addr_str, __u16 port,
struct sockaddr_storage *addr, socklen_t *len);
char *ping_command(int family);
int get_socket_local_port(int sock_fd);
int get_hw_ring_size(char *ifname, struct ethtool_ringparam *ring_param);
int set_hw_ring_size(char *ifname, struct ethtool_ringparam *ring_param);
int open_tuntap(const char *dev_name, bool need_mac);
struct nstoken;
/**
* open_netns() - Switch to specified network namespace by name.
*
* Returns token with which to restore the original namespace
* using close_netns().
*/
struct nstoken *open_netns(const char *name);
void close_netns(struct nstoken *token);
int send_recv_data(int lfd, int fd, uint32_t total_bytes);
int make_netns(const char *name);
int remove_netns(const char *name);
/**
* append_tid() - Append thread ID to the given string.
*
* @str: string to extend
* @sz: string's size
*
* 8 characters are used to append the thread ID (7 digits + '\0')
*
* Returns -1 on errors, 0 otherwise
*/
int append_tid(char *str, size_t sz);
static __u16 csum_fold(__u32 csum)
{
csum = (csum & 0xffff) + (csum >> 16);
csum = (csum & 0xffff) + (csum >> 16);
return (__u16)~csum;
}
static __wsum csum_partial(const void *buf, int len, __wsum sum)
{
__u16 *p = (__u16 *)buf;
int num_u16 = len >> 1;
int i;
for (i = 0; i < num_u16; i++)
sum += p[i];
return sum;
}
static inline __sum16 build_ip_csum(struct iphdr *iph)
{
__u32 sum = 0;
__u16 *p;
iph->check = 0;
p = (void *)iph;
sum = csum_partial(p, iph->ihl << 2, 0);
return csum_fold(sum);
}
/**
* csum_tcpudp_magic - compute IP pseudo-header checksum
*
* Compute the IPv4 pseudo header checksum. The helper can take a
* accumulated sum from the transport layer to accumulate it and directly
* return the transport layer
*
* @saddr: IP source address
* @daddr: IP dest address
* @len: IP data size
* @proto: transport layer protocol
* @csum: The accumulated partial sum to add to the computation
*
* Returns the folded sum
*/
static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
__u32 len, __u8 proto,
__wsum csum)
{
__u64 s = csum;
s += (__u32)saddr;
s += (__u32)daddr;
s += htons(proto + len);
s = (s & 0xffffffff) + (s >> 32);
s = (s & 0xffffffff) + (s >> 32);
return csum_fold((__u32)s);
}
/**
* csum_ipv6_magic - compute IPv6 pseudo-header checksum
*
* Compute the ipv6 pseudo header checksum. The helper can take a
* accumulated sum from the transport layer to accumulate it and directly
* return the transport layer
*
* @saddr: IPv6 source address
* @daddr: IPv6 dest address
* @len: IPv6 data size
* @proto: transport layer protocol
* @csum: The accumulated partial sum to add to the computation
*
* Returns the folded sum
*/
static inline __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
const struct in6_addr *daddr,
__u32 len, __u8 proto,
__wsum csum)
{
__u64 s = csum;
int i;
for (i = 0; i < 4; i++)
s += (__u32)saddr->s6_addr32[i];
for (i = 0; i < 4; i++)
s += (__u32)daddr->s6_addr32[i];
s += htons(proto + len);
s = (s & 0xffffffff) + (s >> 32);
s = (s & 0xffffffff) + (s >> 32);
return csum_fold((__u32)s);
}
/**
* build_udp_v4_csum - compute UDP checksum for UDP over IPv4
*
* Compute the checksum to embed in UDP header, composed of the sum of IP
* pseudo-header checksum, UDP header checksum and UDP data checksum
* @iph IP header
* @udph UDP header, which must be immediately followed by UDP data
*
* Returns the total checksum
*/
static inline __sum16 build_udp_v4_csum(const struct iphdr *iph,
const struct udphdr *udph)
{
unsigned long sum;
sum = csum_partial(udph, ntohs(udph->len), 0);
return csum_tcpudp_magic(iph->saddr, iph->daddr, ntohs(udph->len),
IPPROTO_UDP, sum);
}
/**
* build_udp_v6_csum - compute UDP checksum for UDP over IPv6
*
* Compute the checksum to embed in UDP header, composed of the sum of IPv6
* pseudo-header checksum, UDP header checksum and UDP data checksum
* @ip6h IPv6 header
* @udph UDP header, which must be immediately followed by UDP data
*
* Returns the total checksum
*/
static inline __sum16 build_udp_v6_csum(const struct ipv6hdr *ip6h,
const struct udphdr *udph)
{
unsigned long sum;
sum = csum_partial(udph, ntohs(udph->len), 0);
return csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, ntohs(udph->len),
IPPROTO_UDP, sum);
}
selftests/bpf: Add traffic monitor functions. Add functions that capture packets and print log in the background. They are supposed to be used for debugging flaky network test cases. A monitored test case should call traffic_monitor_start() to start a thread to capture packets in the background for a given namespace and call traffic_monitor_stop() to stop capturing. (Or, option '-m' implemented by the later patches.) lo In IPv4 127.0.0.1:40265 > 127.0.0.1:55907: TCP, length 68, SYN lo In IPv4 127.0.0.1:55907 > 127.0.0.1:40265: TCP, length 60, SYN, ACK lo In IPv4 127.0.0.1:40265 > 127.0.0.1:55907: TCP, length 60, ACK lo In IPv4 127.0.0.1:55907 > 127.0.0.1:40265: TCP, length 52, ACK lo In IPv4 127.0.0.1:40265 > 127.0.0.1:55907: TCP, length 52, FIN, ACK lo In IPv4 127.0.0.1:55907 > 127.0.0.1:40265: TCP, length 52, RST, ACK Packet file: packets-2173-86-select_reuseport:sockhash_IPv4_TCP_LOOPBACK_test_detach_bpf-test.log #280/87 select_reuseport/sockhash IPv4/TCP LOOPBACK test_detach_bpf:OK The above is the output of an example. It shows the packets of a connection and the name of the file that contains captured packets in the directory /tmp/tmon_pcap. The file can be loaded by tcpdump or wireshark. This feature only works if libpcap is available. (Could be found by pkg-config) Acked-by: Stanislav Fomichev <sdf@fomichev.me> Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com> Link: https://lore.kernel.org/r/20240815053254.470944-2-thinker.li@gmail.com Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
2024-08-14 22:32:49 -07:00
struct tmonitor_ctx;
typedef int (*tm_print_fn_t)(const char *format, va_list args);
selftests/bpf: Add traffic monitor functions. Add functions that capture packets and print log in the background. They are supposed to be used for debugging flaky network test cases. A monitored test case should call traffic_monitor_start() to start a thread to capture packets in the background for a given namespace and call traffic_monitor_stop() to stop capturing. (Or, option '-m' implemented by the later patches.) lo In IPv4 127.0.0.1:40265 > 127.0.0.1:55907: TCP, length 68, SYN lo In IPv4 127.0.0.1:55907 > 127.0.0.1:40265: TCP, length 60, SYN, ACK lo In IPv4 127.0.0.1:40265 > 127.0.0.1:55907: TCP, length 60, ACK lo In IPv4 127.0.0.1:55907 > 127.0.0.1:40265: TCP, length 52, ACK lo In IPv4 127.0.0.1:40265 > 127.0.0.1:55907: TCP, length 52, FIN, ACK lo In IPv4 127.0.0.1:55907 > 127.0.0.1:40265: TCP, length 52, RST, ACK Packet file: packets-2173-86-select_reuseport:sockhash_IPv4_TCP_LOOPBACK_test_detach_bpf-test.log #280/87 select_reuseport/sockhash IPv4/TCP LOOPBACK test_detach_bpf:OK The above is the output of an example. It shows the packets of a connection and the name of the file that contains captured packets in the directory /tmp/tmon_pcap. The file can be loaded by tcpdump or wireshark. This feature only works if libpcap is available. (Could be found by pkg-config) Acked-by: Stanislav Fomichev <sdf@fomichev.me> Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com> Link: https://lore.kernel.org/r/20240815053254.470944-2-thinker.li@gmail.com Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
2024-08-14 22:32:49 -07:00
#ifdef TRAFFIC_MONITOR
struct tmonitor_ctx *traffic_monitor_start(const char *netns, const char *test_name,
const char *subtest_name);
void traffic_monitor_stop(struct tmonitor_ctx *ctx);
tm_print_fn_t traffic_monitor_set_print(tm_print_fn_t fn);
selftests/bpf: Add traffic monitor functions. Add functions that capture packets and print log in the background. They are supposed to be used for debugging flaky network test cases. A monitored test case should call traffic_monitor_start() to start a thread to capture packets in the background for a given namespace and call traffic_monitor_stop() to stop capturing. (Or, option '-m' implemented by the later patches.) lo In IPv4 127.0.0.1:40265 > 127.0.0.1:55907: TCP, length 68, SYN lo In IPv4 127.0.0.1:55907 > 127.0.0.1:40265: TCP, length 60, SYN, ACK lo In IPv4 127.0.0.1:40265 > 127.0.0.1:55907: TCP, length 60, ACK lo In IPv4 127.0.0.1:55907 > 127.0.0.1:40265: TCP, length 52, ACK lo In IPv4 127.0.0.1:40265 > 127.0.0.1:55907: TCP, length 52, FIN, ACK lo In IPv4 127.0.0.1:55907 > 127.0.0.1:40265: TCP, length 52, RST, ACK Packet file: packets-2173-86-select_reuseport:sockhash_IPv4_TCP_LOOPBACK_test_detach_bpf-test.log #280/87 select_reuseport/sockhash IPv4/TCP LOOPBACK test_detach_bpf:OK The above is the output of an example. It shows the packets of a connection and the name of the file that contains captured packets in the directory /tmp/tmon_pcap. The file can be loaded by tcpdump or wireshark. This feature only works if libpcap is available. (Could be found by pkg-config) Acked-by: Stanislav Fomichev <sdf@fomichev.me> Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com> Link: https://lore.kernel.org/r/20240815053254.470944-2-thinker.li@gmail.com Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
2024-08-14 22:32:49 -07:00
#else
static inline struct tmonitor_ctx *traffic_monitor_start(const char *netns, const char *test_name,
const char *subtest_name)
{
return NULL;
}
static inline void traffic_monitor_stop(struct tmonitor_ctx *ctx)
{
}
static inline tm_print_fn_t traffic_monitor_set_print(tm_print_fn_t fn)
{
return NULL;
}
selftests/bpf: Add traffic monitor functions. Add functions that capture packets and print log in the background. They are supposed to be used for debugging flaky network test cases. A monitored test case should call traffic_monitor_start() to start a thread to capture packets in the background for a given namespace and call traffic_monitor_stop() to stop capturing. (Or, option '-m' implemented by the later patches.) lo In IPv4 127.0.0.1:40265 > 127.0.0.1:55907: TCP, length 68, SYN lo In IPv4 127.0.0.1:55907 > 127.0.0.1:40265: TCP, length 60, SYN, ACK lo In IPv4 127.0.0.1:40265 > 127.0.0.1:55907: TCP, length 60, ACK lo In IPv4 127.0.0.1:55907 > 127.0.0.1:40265: TCP, length 52, ACK lo In IPv4 127.0.0.1:40265 > 127.0.0.1:55907: TCP, length 52, FIN, ACK lo In IPv4 127.0.0.1:55907 > 127.0.0.1:40265: TCP, length 52, RST, ACK Packet file: packets-2173-86-select_reuseport:sockhash_IPv4_TCP_LOOPBACK_test_detach_bpf-test.log #280/87 select_reuseport/sockhash IPv4/TCP LOOPBACK test_detach_bpf:OK The above is the output of an example. It shows the packets of a connection and the name of the file that contains captured packets in the directory /tmp/tmon_pcap. The file can be loaded by tcpdump or wireshark. This feature only works if libpcap is available. (Could be found by pkg-config) Acked-by: Stanislav Fomichev <sdf@fomichev.me> Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com> Link: https://lore.kernel.org/r/20240815053254.470944-2-thinker.li@gmail.com Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
2024-08-14 22:32:49 -07:00
#endif
#endif