Re: [hatari-devel] SCC-B support

[Thomas Huth <th.huth@xxxxxxxxx>]

Am Sun, 30 Dec 2018 20:43:10 +0200
schrieb Eero Tamminen <oak@xxxxxxxxxxxxxx>:

> Hi,
> 
> On 12/29/18 8:50 AM, Thomas Huth wrote:
> > Am Fri, 28 Dec 2018 23:10:40 +0200
> > schrieb Eero Tamminen <oak@xxxxxxxxxxxxxx>:  
> >> I've tried the Hatari SCC-B support a bit when testing EmuTOS
> >> serial output support.
> >>
> >> Some comments:
> >> * SCC-B doesn't have separate files for input and output like
> >>     normal MFP serial does (in case they're unidirectional pipes)
> >> * Unlike MFP serial, SCC serial doesn't create the output file,
> >>     instead it fails if it doesn't already exist  
> > 
> > Well, yes, that code from Aranym was only meant to be used with
> > real /dev/ttyS* devices, not with normal files or pipes. I only
> > added some hack so far that it is possible to open normal files for
> > output, so it is usable with the "make test" regression tester,
> > too... But I agree with you, this should eventually become similar
> > to --rs232-out/in and --midi-out/in, too. I'll see what I can
> > do...  
> > >> * Although SCC-B is TT / Falcon specific, it gets still  
> >>     initialized for other machine types  
> > 
> > That should be easy to fix... I pushed a patch now.  
> 
> Verified.
> 
> 
> If I've understood correctly:
> * ST / MegaST / STE: 1 MFP serial port
> * MegaSTE & TT: 1 MFP, 2x SCC serial ports
> * Falcon: 2 SCC serial ports

As far as I know, the Falcon has only one serial port. SCC-A is
hard-wired to the "LAN" connector here (while it can be switched to a
normal serial connector for the TT and Mega-STE). The code that I've
copied from Aranym does not fully support the SCC-A channel yet, so not
sure whether it makes sense to provide a "--scc-a" parameter already...
Let's focus on SCC-B for now which is way more important, since it is
the only serial port of the Falcon.

 Thomas