Hi. I'm working with HyperLink and am transferring data between two EVMs. I've modified the HyperLink example program available with the MCSDK so that data is only being written on one board (Tx), and is being read on the other (Rx). I've gone through the HyperLink UG, and all other training required for HyperLink. I had 4 questions regarding the address mapping used in the program:
1) The getWindow() function in the example program (called inside AddrMap()) returns the address window of HyperLink. I've noticed that when I run the program, it always returns an address of 0x40000000, which is the starting address of the memory set aside for HyperLink. Will this function always return 0x40000000? If not, what exactly is it returning?
2) According to documentation, the rxSegVal values are part of a look-up table that is used on the Rx side for address translation. The rxSegVal values in the example program are calculated from the address of dataBufferSrc. However, as the example program is meant to run on two EVMs simultaneously, two calculations of rxSegVal take place - on the local (Tx) and the remote (Rx) device. Whenever I run the program, the 2 values turn out to be the same. But which one is being used for address translation on the Rx side?
3) Similarly, is the value of dataBufferViaHlink sent over the HyperLink cable to be used for address translation?
4) I'm aiming to use the HyperLink program to transfer data between boards and subsequently read that data. However, the address at which the data is located on the Rx side is dependent on the dataBufferViaHlink value, which in turn is dependent on the address on the Tx side where the original data is stored. Hence, my program now involves the user entering the address on the Rx side for reading. Is there any way to fix the address of the data on the Tx side and therefore, the address of the data written on the Rx side?
