> From the spec:
> 
> The case when ``length of \field{write_buf}''=0, and at the same time,
> ``length of \field{read_buf}''=0 doesn't make any sense.
> 
> I mentioned this in my first reply and to my understanding I did not get
> a reply that this has changed meanwhile.
> 

Also, this code as mentioned before:

> +             if (!msgs[i].len)
> +                     break;

I hope this can extended in the future to allow zero-length messages. If
this is impossible we need to set an adapter quirk instead.