Il dom 28 dic 2025, 18:19 Peter Maydell <[email protected]> ha scritto:
> But an inline memcpy() is hard to read and easy to get wrong: > we have a pointer-cast and some pointer arithmetic going on here. > What we want is to express our intent: "I am doing a load/store > of N bytes which are in the host byte order and which might not > be aligned". That's what the _he_p() functions are all for. > Yes, I think (hope :)) that Phil wants to define ldn_he_p/stn_he_p as a memcpy instead of a switch statement. Paolo > thanks > -- PMM > >
