R_OPERATION_CTRL behavior ambiguity

[Poofjunior]

In the R_OPERATION_CTRL register specification, what should the defined behavior be in this case?

• When the DUMP bit is asserted from a PC-to-Device Harp Write message, does the device issue one reply per register, or a single reply where the payload contains the serialized register contents of all registers?
  • Does this reply (or replies) include just core registers or both core and app registers?

[filcarv]

The device issues one reply per register (i.e., one Harp Message per register) for all registers, including the core ones.

[Poofjunior]

Ok, it would be great to get this noted in the docs somehow, or just merge #6 .

[bruno-f-cruz]

We unfortunately lost #6 with the previous merge we did on this repo. Can you re-do the PR to explicitly state this behavior in the protocol? Thanks!