lwip-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lwip-users] lwip + modbus TCP + UDP


From: goldsimon
Subject: Re: [lwip-users] lwip + modbus TCP + UDP
Date: Mon, 31 Dec 2018 15:58:38 +0100
User-agent: K-9 Mail for Android


Pablo Ledergerber wrote:
>
>hello,
>
>I'm working on a project for a new sensor with ethernet connectivity, 
>modbus TCP for data and a kind of discovery function with udp packets 
>for identificationdata. During the sensor is working well, there are 
>sometime packet retransmission and after a cupple of minutes or when I 
>send new data over UDP to the sensor the software got stuck. This is
>the 
>first time for me that I use the lwip stack in a project. I've tried to
>
>set the parameters into the lwipopts.h file without success.
>In this function get stuck:
>
>tcp_fasttmr(void)
>{
>   struct tcp_pcb *pcb;
>   ++tcp_timer_ctr;
>tcp_fasttmr_start:
>   pcb = tcp_active_pcbs;
>   while (pcb != NULL) {
>     if (pcb->last_timer != tcp_timer_ctr) {  <------ exactly here!

Looks like a threading issue to me.

>Enclosed you find a wireshark data file. My setup is the following:
>- xmc4400 from Infineon
>- KSZ8081 Phy from Microchip
>- RTOS: CMSIS
>- Freemodbus Library for Modbus TCP
>- lwip 2.2
>The freemodbus library uses the TCP Raw library and the upd discovery 
>function the netconn api. We have one thread for modbusTCP,

Having a thread for modbusTCP when that library uses the raw (callback) API 
looks like a threading violation.

Have you read the docs at http://www.nongnu.org/lwip/ (especially the chapter 
"common pitfalls")?

Regards,
Simon

> one for the
>
>sensor data (ADC) and one for the discovery function (UDP Netconn API).
>
>I tried to implement the netconn API for TCP into the freemodbus
>library 
>without success. I don't know exactly if I'm having a memory leak or 
>another problem.
>I'm a bit lost in the jungle and I would appreciate a lot if somebody 
>could me lead to the right way, so that I could solve this problem. 
>Please don't excitate to ask me. I have done a lot of things and right 
>now I don't know exactly which information is important to find a
>solution.
>Thank you in advance.
>
>Pablo Ledergerber



reply via email to

[Prev in Thread] Current Thread [Next in Thread]