patrick/zephyr
1
0
Fork 0
Code Issues Pull requests Actions 7 Packages Projects Releases Wiki Activity

net: shell: Use the ICMP API for ping command

Browse source 
Change the ping command to use the ICMP API.

Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
This commit is contained in:
Jukka Rissanen 2023-09-21 17:25:47 +03:00 committed by Carles Cufí
parent 19273087f2
commit e3fb634e7a
1 changed files with 93 additions and 80 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

173
subsys/net/ip/net_shell.c
View file

@ -33,6 +33,8 @@ LOG_MODULE_REGISTER(net_shell, LOG_LEVEL_DBG);
#include <zephyr/net/ethernet_mgmt.h>
#endif /* CONFIG_NET_L2_ETHERNET */
#include <zephyr/net/icmp.h>
#include "route.h"
#include "icmpv6.h"
#include "icmpv4.h"
@ -4305,7 +4307,12 @@ static int cmd_net_nbr(const struct shell *sh, size_t argc, char *argv[])
static struct ping_context {
struct k_work_delayable work;
struct net_addr addr;
struct net_icmp_ctx icmp;
union {
struct sockaddr_in addr4;
struct sockaddr_in6 addr6;
struct sockaddr addr;
};
struct net_if *iface;
const struct shell *sh;
@ -4322,27 +4329,15 @@ static void ping_done(struct ping_context *ctx);
#if defined(CONFIG_NET_NATIVE_IPV6)
static enum net_verdict handle_ipv6_echo_reply(struct net_pkt *pkt,
struct net_ipv6_hdr *ip_hdr,
struct net_icmp_hdr *icmp_hdr);
static struct net_icmpv6_handler ping6_handler = {
.type = NET_ICMPV6_ECHO_REPLY,
.code = 0,
.handler = handle_ipv6_echo_reply,
};
static inline void remove_ipv6_ping_handler(void)
{
net_icmpv6_unregister_handler(&ping6_handler);
}
static enum net_verdict handle_ipv6_echo_reply(struct net_pkt *pkt,
struct net_ipv6_hdr *ip_hdr,
struct net_icmp_hdr *icmp_hdr)
static int handle_ipv6_echo_reply(struct net_icmp_ctx *ctx,
struct net_pkt *pkt,
struct net_icmp_ip_hdr *hdr,
struct net_icmp_hdr *icmp_hdr,
void *user_data)
{
NET_PKT_DATA_ACCESS_CONTIGUOUS_DEFINE(icmp_access,
struct net_icmpv6_echo_req);
struct net_ipv6_hdr *ip_hdr = hdr->ipv6;
struct net_icmpv6_echo_req *icmp_echo;
uint32_t cycles;
char time_buf[16] = { 0 };
@ -4350,14 +4345,14 @@ static enum net_verdict handle_ipv6_echo_reply(struct net_pkt *pkt,
icmp_echo = (struct net_icmpv6_echo_req *)net_pkt_get_data(pkt,
&icmp_access);
if (icmp_echo == NULL) {
return -NET_DROP;
return -EIO;
}
net_pkt_skip(pkt, sizeof(*icmp_echo));
if (net_pkt_remaining_data(pkt) >= sizeof(uint32_t)) {
if (net_pkt_read_be32(pkt, &cycles)) {
return -NET_DROP;
return -EIO;
}
cycles = k_cycle_get_32() - cycles;
@ -4394,35 +4389,36 @@ static enum net_verdict handle_ipv6_echo_reply(struct net_pkt *pkt,
}
net_pkt_unref(pkt);
return NET_OK;
return 0;
}
#else
#define remove_ipv6_ping_handler()
static int handle_ipv6_echo_reply(struct net_icmp_ctx *ctx,
struct net_pkt *pkt,
struct net_icmp_ip_hdr *hdr,
struct net_icmp_hdr *icmp_hdr,
void *user_data)
{
ARG_UNUSED(ctx);
ARG_UNUSED(pkt);
ARG_UNUSED(hdr);
ARG_UNUSED(icmp_hdr);
ARG_UNUSED(user_data);
return -ENOTSUP;
}
#endif /* CONFIG_NET_IPV6 */
#if defined(CONFIG_NET_NATIVE_IPV4)
static enum net_verdict handle_ipv4_echo_reply(struct net_pkt *pkt,
struct net_ipv4_hdr *ip_hdr,
struct net_icmp_hdr *icmp_hdr);
static struct net_icmpv4_handler ping4_handler = {
.type = NET_ICMPV4_ECHO_REPLY,
.code = 0,
.handler = handle_ipv4_echo_reply,
};
static inline void remove_ipv4_ping_handler(void)
{
net_icmpv4_unregister_handler(&ping4_handler);
}
static enum net_verdict handle_ipv4_echo_reply(struct net_pkt *pkt,
struct net_ipv4_hdr *ip_hdr,
struct net_icmp_hdr *icmp_hdr)
static int handle_ipv4_echo_reply(struct net_icmp_ctx *ctx,
struct net_pkt *pkt,
struct net_icmp_ip_hdr *hdr,
struct net_icmp_hdr *icmp_hdr,
void *user_data)
{
NET_PKT_DATA_ACCESS_CONTIGUOUS_DEFINE(icmp_access,
struct net_icmpv4_echo_req);
struct net_ipv4_hdr *ip_hdr = hdr->ipv4;
uint32_t cycles;
struct net_icmpv4_echo_req *icmp_echo;
char time_buf[16] = { 0 };
@ -4430,14 +4426,14 @@ static enum net_verdict handle_ipv4_echo_reply(struct net_pkt *pkt,
icmp_echo = (struct net_icmpv4_echo_req *)net_pkt_get_data(pkt,
&icmp_access);
if (icmp_echo == NULL) {
return -NET_DROP;
return -EIO;
}
net_pkt_skip(pkt, sizeof(*icmp_echo));
if (net_pkt_remaining_data(pkt) >= sizeof(uint32_t)) {
if (net_pkt_read_be32(pkt, &cycles)) {
return -NET_DROP;
return -EIO;
}
cycles = k_cycle_get_32() - cycles;
@ -4468,11 +4464,23 @@ static enum net_verdict handle_ipv4_echo_reply(struct net_pkt *pkt,
}
net_pkt_unref(pkt);
return NET_OK;
return 0;
}
#else
#define remove_ipv4_ping_handler()
static int handle_ipv4_echo_reply(struct net_icmp_ctx *ctx,
struct net_pkt *pkt,
struct net_icmp_ip_hdr *hdr,
struct net_icmp_hdr *icmp_hdr,
void *user_data)
{
ARG_UNUSED(ctx);
ARG_UNUSED(pkt);
ARG_UNUSED(hdr);
ARG_UNUSED(icmp_hdr);
ARG_UNUSED(user_data);
return -ENOTSUP;
}
#endif /* CONFIG_NET_IPV4 */
static int parse_arg(size_t *i, size_t argc, char *argv[])
@ -4506,8 +4514,7 @@ static int parse_arg(size_t *i, size_t argc, char *argv[])
static void ping_cleanup(struct ping_context *ctx)
{
remove_ipv6_ping_handler();
remove_ipv4_ping_handler();
(void)net_icmp_cleanup_ctx(&ctx->icmp);
shell_set_bypass(ctx->sh, NULL);
}
@ -4525,6 +4532,7 @@ static void ping_work(struct k_work *work)
struct ping_context *ctx =
CONTAINER_OF(dwork, struct ping_context, work);
const struct shell *sh = ctx->sh;
struct net_icmp_ping_params params;
int ret;
ctx->sequence++;
@ -4535,26 +4543,18 @@ static void ping_work(struct k_work *work)
return;
}
if (ctx->addr.family == AF_INET6) {
ret = net_icmpv6_send_echo_request(ctx->iface,
&ctx->addr.in6_addr,
sys_rand32_get(),
ctx->sequence,
ctx->tos,
ctx->priority,
NULL,
ctx->payload_size);
} else {
ret = net_icmpv4_send_echo_request(ctx->iface,
&ctx->addr.in_addr,
sys_rand32_get(),
ctx->sequence,
ctx->tos,
ctx->priority,
NULL,
ctx->payload_size);
}
params.identifier = sys_rand32_get();
params.sequence = ctx->sequence;
params.tc_tos = ctx->tos;
params.priority = ctx->priority;
params.data = NULL;
params.data_size = ctx->payload_size;
ret = net_icmp_send_echo_request(&ctx->icmp,
ctx->iface,
&ctx->addr,
&params,
ctx);
if (ret != 0) {
PR_WARNING("Failed to send ping, err: %d", ret);
ping_done(ctx);
@ -4583,7 +4583,7 @@ static void ping_bypass(const struct shell *sh, uint8_t *data, size_t len)
}
}
static struct net_if *ping_select_iface(int id, struct net_addr *target)
static struct net_if *ping_select_iface(int id, struct sockaddr *target)
{
struct net_if *iface = net_if_get_by_index(id);
@ -4591,8 +4591,8 @@ static struct net_if *ping_select_iface(int id, struct net_addr *target)
goto out;
}
if (IS_ENABLED(CONFIG_NET_IPV4) && target->family == AF_INET) {
iface = net_if_ipv4_select_src_iface(&target->in_addr);
if (IS_ENABLED(CONFIG_NET_IPV4) && target->sa_family == AF_INET) {
iface = net_if_ipv4_select_src_iface(&net_sin(target)->sin_addr);
if (iface != NULL) {
goto out;
}
@ -4601,25 +4601,25 @@ static struct net_if *ping_select_iface(int id, struct net_addr *target)
goto out;
}
if (IS_ENABLED(CONFIG_NET_IPV6) && target->family == AF_INET6) {
if (IS_ENABLED(CONFIG_NET_IPV6) && target->sa_family == AF_INET6) {
struct net_nbr *nbr;
#if defined(CONFIG_NET_ROUTE)
struct net_route_entry *route;
#endif
iface = net_if_ipv6_select_src_iface(&target->in6_addr);
iface = net_if_ipv6_select_src_iface(&net_sin6(target)->sin6_addr);
if (iface != NULL) {
goto out;
}
nbr = net_ipv6_nbr_lookup(NULL, &target->in6_addr);
nbr = net_ipv6_nbr_lookup(NULL, &net_sin6(target)->sin6_addr);
if (nbr) {
iface = nbr->iface;
goto out;
}
#if defined(CONFIG_NET_ROUTE)
route = net_route_lookup(NULL, &target->in6_addr);
route = net_route_lookup(NULL, &net_sin6(target)->sin6_addr);
if (route) {
iface = route->iface;
goto out;
@ -4652,6 +4652,7 @@ static int cmd_net_ping(const struct shell *sh, size_t argc, char *argv[])
int tos = 0;
int payload_size = 4;
int priority = -1;
int ret;
for (size_t i = 1; i < argc; ++i) {
@ -4736,13 +4737,25 @@ static int cmd_net_ping(const struct shell *sh, size_t argc, char *argv[])
ping_ctx.payload_size = payload_size;
if (IS_ENABLED(CONFIG_NET_IPV6) &&
net_addr_pton(AF_INET6, host, &ping_ctx.addr.in6_addr) == 0) {
ping_ctx.addr.family = AF_INET6;
net_icmpv6_register_handler(&ping6_handler);
net_addr_pton(AF_INET6, host, &ping_ctx.addr6.sin6_addr) == 0) {
ping_ctx.addr6.sin6_family = AF_INET6;
ret = net_icmp_init_ctx(&ping_ctx.icmp, NET_ICMPV6_ECHO_REPLY, 0,
handle_ipv6_echo_reply);
if (ret < 0) {
PR_WARNING("Cannot initialize ICMP context for %s\n", "IPv6");
return 0;
}
} else if (IS_ENABLED(CONFIG_NET_IPV4) &&
net_addr_pton(AF_INET, host, &ping_ctx.addr.in_addr) == 0) {
ping_ctx.addr.family = AF_INET;
net_icmpv4_register_handler(&ping4_handler);
net_addr_pton(AF_INET, host, &ping_ctx.addr4.sin_addr) == 0) {
ping_ctx.addr4.sin_family = AF_INET;
ret = net_icmp_init_ctx(&ping_ctx.icmp, NET_ICMPV4_ECHO_REPLY, 0,
handle_ipv4_echo_reply);
if (ret < 0) {
PR_WARNING("Cannot initialize ICMP context for %s\n", "IPv4");
return 0;
}
} else {
PR_WARNING("Invalid IP address\n");
return 0;