This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Ethernet ring topology

Other Parts Discussed in Thread: AM3357, TLK105

Admitted noobie to networking so this may be a bit pedantic:

I am planning to use an AM3357 Processor as a node in an Ethernet network. Because of our application environment, reliability is going to be reached by using a ring topology (not a daisy chain). Other pieces of background information:

(1) This is a Powerlink network so collisions are impossible

(2) Very low data rates

(3) Could do redundant circuits but I am keeping that off the table right now for other reasons

(4) There is only one ring - no risk of other networks creating rings

So, the actual question:

I understand that the AM3357 can support 2 Ethernet ports. I would like to use one as the primary transmitter and the other to receive to those packets (no receive = broken network). If the network is broken, both ports switch to primary send (I'll use a TLK105 and force the MDIX feature to switch Rx/Tx). Network "healing" can be detected by collisions (since there shouldn't be any) any the ports can switch back to its primary receive mode.

Is this a reasonable implementation under SYS/BIOS. Basically the application handles the network state and the mode for the second port. All the driver does is collect the received frames - the app can compare them with the sent (remember, very slow data rates). 

Addendum to the original question: can I use the PRUs as independent MACs for each port?