Re: [Xc3sprog-users] Bringing FreeBSD support to xc3sprog.

[Wojciech A. K. <wk...@Fr...>]

On Tue, Feb 09, 2010 at 12:49:34PM +0100, Uwe Bonnes wrote:
> >>>>> "Wojciech" == Wojciech A Koszek <wk...@Fr...> writes:
> 
>     Wojciech> On Mon, Feb 08, 2010 at 09:43:19PM +0100, Uwe Bonnes wrote:
>     >> >>>>> "Wojciech" == Wojciech A Koszek <wk...@Fr...> writes:
>     >> 
>     Wojciech> xc3sprog project (c) 2005-2010 Uwe Bonnes
>     Wojciech> bo...@el... WWW:
>     >>  The work was started by Andrew, others dropped in and now most
>     >> contributions are mine. But that doesn't give me the whole copyright.
> 
>     Wojciech> I haven't had a chance to exactly track the version history
>     Wojciech> and because of not being a copyright expert. Maybe:
> 
>     Wojciech> 	XC3SPROG (c) 2005-2010 xc3sprog project $Rev$ <OS>
> 
>     Wojciech> Or whatever else. As long as I can easily distinguish that a
>     Wojciech> user calls the right program, it's ok.
> 
> Do you have a suggestion for the <OS> printout logic. Otherwise I tried to
> implement your suggestions from this mail in rev 434. I tried to be as
> explicit as with the error mess 

Uwe,

Was something wrong with the patch?

You may want to obtain rev 432 for which my patch was addressed and
simply apply it and change printf's. I use uname(3) there and I think
it can be used here as well. "FreeBSD" and "Linux" should be enough.
Ideally (as I mentioned earlier), your cmake-magic knowledge could be
deployed, so that, except pure OS specific string returned from
uname(3), you'll also have specific information about cross
compilation.

Usage is much clearer in my patch. Moreover, it lists -c option, which
is missing right now.

The reason why I removed verbose notifications as I think they're
uncessary, and exceed 80 lines. From r435:

-------------------------------------------------------------------------------
"Feedback on success/failure/enhancement requests: http://sourceforge.net/mail/?group_id=170565"
"Check Sourceforge for updates: http://sourceforge.net/projects/xc3sprog/develop for updates"

Maybe:

  "In case of problems: http://sf.net/mail/?group_id=170565"

And probably only with -v...

I think logic present after your patch isn't yet right -- parallel
port constructor is always called, which means that my logic circuit
for the CNC machine I build here may suddenly, by accident, respond
correctly to the "probe" sequence..

My opinion is that under no condition xc3sprog should touch anything,
unless stated otherwise.  So, IMO (ideal case) the right logic would
be:

if XC3SPROG_CABLE set
	Use $XC3SPROG_CABLE.
else
	if -c had some hints
		Yes, use cable passed from -c.
	else
		Inform what to do.

Where:

- $XC3SPROG_CABLE == "pp:/path/to/dev" -> parallel port backend with
  /path/to/dev device.

- $XC3SPROG_CABLE == "ftdi:vid=<VID>,pid=<PID>,option=..." -> FTDI options

Or similar. This would let you to stay away from getenv()s in IOParport..

>     Wojciech> This 'usb_control_msg(0x42.1 Unknown error' appears here as
>     Wojciech> well.  BTW, a friend of mine confirmed that he can program his
>     Wojciech> S3ESK board under his FreeBSD 8 system without a problem with
>     Wojciech> some of my random notes.
>     >>  I'll look again at the 0x42.1 message again. What Xilinx dobgle is
>     >> that, what firmware file loaded?
> 
>     Wojciech> I have S3ESK board so far. I use its embedded USB JTAG cable
>     Wojciech> with this firmware:
> 
> I disabled the code so long. It doesn't get work for me and causes problems
> for you.

Thanks!

-- 
Wojciech A. Koszek
wk...@Fr...
http://FreeBSD.czest.pl/~wkoszek/