I’m interested in implementing an isochronous transfer mode with the XEM6310-LX150. I understand this is possibly (see ticket #25432 details below):
Using our API, I think you’ll find that you can accomplish what you need. Essentially, you can use the Pipe transfers in a manner that is quite similar to how isochronous transfers work. The FPGA will provide buffering as you require. If for some reason the host is unable to keep up, the FPGA can just drop a frame or some data…
My question is how to setup this mode? Also, what indication is there (to the application) that a frame has been lost? If I understand correctly, normally packet loss is handled below the hood by the protocol itself, and there is effectively no data loss, as retransmits occur automatically (wire transfer failures), correct?