hi,we use threadx to develop a video coding application and use net50 chip. The net50 chip is connected with one DSP chip and the DSP reset pin is connected with one net50 GPIO pin. Now we encounter 1 strange problems: 1) we use a thread to implement multicast funnction to transmit data. First I can access that chip by ping, but after about 15 hours testing I found that the chip can’t access by ping and the DSP is reseted in cycle. But what surprise me is the multicast task is alive. I guess the TCP/IP protocol is runned by os kernel. why though the TCP/IP is down(can’t access) other task can alive! Thanks in advance!
I have seen a similar case. I dont use a DSP, but also a NET50 chip with multicast and TCP communication. Same thing happened to be. Suddenly TCP layer just drops dead, but multicast is still working. If you discover some more information please post it. I will do the same if I get lucky and discover the problem.