michaelh/zephyr
1
0
Fork 0
Code Releases Activity

net: tcp2: Switch from k_timer to k_delayed_work

Browse source 
k_timer callback is executed from ISR context, which isn't
currenty compatible with Zephyr's shell implementation.

This problem was found while testing TCP2 client side.

Signed-off-by: Oleg Zhurakivskyy <oleg.zhurakivskyy@intel.com>
This commit is contained in:
Oleg Zhurakivskyy 2020-03-06 13:43:43 +02:00 • committed by Jukka Rissanen
commit d6f1937fbb
2 changed files with 12 additions and 14 deletions
Download patch file Download diff file Expand all files Collapse all files

23
subsys/net/ip/tcp2.c
View file

@ -256,8 +256,8 @@ static void tcp_send_queue_flush(struct tcp *conn)
{
struct net_pkt *pkt;
if (is_timer_subscribed(&conn->send_timer)) {
k_timer_stop(&conn->send_timer);
if (k_delayed_work_remaining_get(&conn->send_timer)) {
k_delayed_work_cancel(&conn->send_timer);
}
while ((pkt = tcp_slist(&conn->send_queue, get,
@ -341,9 +341,9 @@ int net_tcp_unref(struct net_context *context)
return ref_count;
}
static void tcp_send_process(struct k_timer *timer)
static void tcp_send_process(struct k_work *work)
{
struct tcp *conn = k_timer_user_data_get(timer);
struct tcp *conn = CONTAINER_OF(work, struct tcp, send_timer);
struct net_pkt *pkt = tcp_slist(&conn->send_queue, peek_head,
struct net_pkt, next);
@ -367,15 +367,15 @@ static void tcp_send_process(struct k_timer *timer)
next) : tcp_pkt_clone(pkt);
tcp_send(pkt);
if (forget == false && is_timer_subscribed(
&conn->send_timer) == false) {
if (forget == false && !k_delayed_work_remaining_get(
&conn->send_timer)) {
conn->send_retries = tcp_retries;
conn->in_retransmission = true;
}
}
if (conn && conn->in_retransmission) {
k_timer_start(&conn->send_timer, K_MSEC(tcp_rto), K_NO_WAIT);
k_delayed_work_submit(&conn->send_timer, K_MSEC(tcp_rto));
}
}
@ -383,7 +383,7 @@ static void tcp_send_timer_cancel(struct tcp *conn)
{
NET_ASSERT(conn->in_retransmission == true, "Not in retransmission");
k_timer_stop(&conn->send_timer);
k_delayed_work_cancel(&conn->send_timer);
{
struct net_pkt *pkt = tcp_slist(&conn->send_queue, get,
@ -396,7 +396,7 @@ static void tcp_send_timer_cancel(struct tcp *conn)
conn->in_retransmission = false;
} else {
conn->send_retries = tcp_retries;
k_timer_start(&conn->send_timer, K_MSEC(tcp_rto), K_NO_WAIT);
k_delayed_work_submit(&conn->send_timer, K_MSEC(tcp_rto));
}
}
@ -752,7 +752,7 @@ static void tcp_out(struct tcp *conn, u8_t flags, ...)
sys_slist_append(&conn->send_queue, &pkt->next);
tcp_send_process(&conn->send_timer);
tcp_send_process((struct k_work *)&conn->send_timer);
out:
return;
}
@ -787,8 +787,7 @@ static struct tcp *tcp_conn_alloc(void)
sys_slist_init(&conn->rsv_bufs);
k_timer_init(&conn->send_timer, tcp_send_process, NULL);
k_timer_user_data_set(&conn->send_timer, conn);
k_delayed_work_init(&conn->send_timer, tcp_send_process);
tcp_conn_ref(conn);

3
subsys/net/ip/tcp2_priv.h
View file

@ -7,7 +7,6 @@
#include "tp.h"
#define is(_a, _b) (strcmp((_a), (_b)) == 0)
#define is_timer_subscribed(_t) (k_timer_remaining_get(_t))
#define th_seq(_x) ntohl((_x)->th_seq)
#define th_ack(_x) ntohl((_x)->th_ack)
@ -156,7 +155,7 @@ struct tcp { /* TCP connection */
u16_t win;
struct tcp_win *rcv;
struct tcp_win *snd;
struct k_timer send_timer;
struct k_delayed_work send_timer;
sys_slist_t send_queue;
bool in_retransmission;
size_t send_retries;