[Coco] Color Forth Help Request

[Stephen Pereira]

Hi Art,

Thanks a million for your replies!  Greatly appreciated.

Here is my code:

*
* BRUTE FORCE ROUTINE TO LOAD SCREENS 1 THRU 8
* USING THE DECB LOADM ROUTINE
* JAN 2015, SMP
* ERROR CHECK & ADVICE FROM BRETT GORDON
*
	FCB	$84,'L,'S,'C,'R+$80
	FDB	SSCR-7
LSCR	FDB	*+2
	PSHS	CC,A,B,DP,X,Y,U	SAVE ALL JUST IN CASE
	LDX	#$094C	CLEAR FILE NAME LOCATION
	LDA	#32
L3	STA	,X+
	CMPX	#$0954
	BNE	L3
	LDX	#$094C	FILE NAME LOCATION
L4	JSR	$A1B1	GET KEY IN A
	TFR	A,B	SAVE KEY IN B
	PSHS	B,X,Y
	JSR	[$A002]	ECHO KEY
	PULS	Y,X,B
	TFR	B,A	GET KEY IN A
	CMPA	#13	CR?
	BEQ	NXT2	CR ENDS LOOP
	STA	,X+	STORE KEY IN FILE NAME
	CMPX	#$0954	DONE IF 8 KEYS
	BNE	L4
NXT2	LDX	#$0954	FILE EXTENSION LOCATION
	LDA	#$42	B
	STA	,X+
	LDA	#$49	I
	STA	,X+
	LDA	#$4E	N
	STA	,X
	LDD	#$0200	FILE TYPE = 2, ASCII FLAG = 0
	STD	$0957
	LDX	#$0000
	STX	#$00D3  OFFSET = 0
	JSR	$CA07	OPEN NEXT AVAIL FILE FOR INPUT
	LDD	#RET2	RETURN ADDR FOR LOADM RTS
	PSHS	D
	JMP	$CFE3	LOADM
RET2	PULS	U,Y,X,DP,B,A,CC	RESTORE ALL
	JMP	NEXT	RETURN TO FORTH

This code works fine on a stock CoCo 3, as well as in VCC with a 6809 CPU.  It is not elegant, and I don’t think I tried any fancy stuff.  Don’t worry about the first few lines of code - they are for linking into Color Forth, and that seem to be fine.

As far as I can tell, this code does not initiate any disk activity when the CPU is a 6309, the machine simply goes off into never-never-land.

Thanks, in advance, for whatever observations or advice you may have to offer.

smp
--
Stephen M. Pereira
Bedford, NH  03110
KB1SXE


> On Feb 20, 2015, at 5:54 PM, Arthur Flexser <flexser at fiu.edu> wrote:
> 
> The only time that I'm aware of when the 6309 in emulation mode does not
> behave exactly like a 6809 is when it encounters an illegal opcode, where
> the 6309 vectors to an error location.  (The 6809 will just skip the
> illegal opcode, usually.)  So my guess would be that that's what's
> happening somewhere in Color Forth.  The 6309 error vector, not having been
> initialized to anything in particular, looks like it winds up in a section
> of the ROM CLOAD code.
> 
> If the 6309 pushes registers enroute to the error vector (I'm not sure if
> it does or not, but this would be documented), you could tell what location
> the error is coming from by pointing the error vector to a routine that
> could look at the stack contents and tell you what the PC register
> contained at the time of the error.
> 
> Art
> 
> On Fri, Feb 20, 2015 at 5:20 PM, Stephen Pereira <spereira1952 at comcast.net>
> wrote:
> 
>> Hello again, all,
>> 
>> I am happy to say that I am one of the lucky ones who was able to get one
>> of the last CoCo 3 systems from Cloud-9 - I got a fully tricked out CoCo 3
>> with a Pro-Tektor, a 6309 processor, and the Triad 512K RAM upgrade (Thanks
>> a million for a great machine, Mark!).
>> 
>> I discovered a problem (for me):  My Color Forth code does not run
>> properly on this machine!  What I found is that I can LOADM”CC-FORTH” and
>> EXEC just fine.  I can issue Color Forth commands (although I have not
>> attempted every command) and things seem OK.  I can issue my added DIR or
>> BASIC commands, and those works well.  Then I try to load a file of Forth
>> screens with my LSCR command.  Upon execution of that command, the CoCo
>> goes off into never-never-land.  Usually, the machine simply goes off and
>> never returns.  It does not access the disk, it just disappears.  Sometimes
>> I get a beautiful screen of scrambled graphics, but not usually.
>> 
>> I had the bright idea to try all this on my VCC emulator.  The 512k RAM
>> does not make any difference.  Color Forth runs just fine.  However, when I
>> change the CPU to be a 6309, VCC fails in exactly the same way as I
>> described above on my real system.  Actually, on VCC, when I execute the
>> LSCR command with a filename, I get a blank green screen with an S in the
>> upper left hand corner, as if the machine is trying to load a cassette
>> file.  Since I removed all the cassette file code from the Color Forth
>> source code, I am puzzled why this screen appears.
>> 
>> I am puzzled by this altogether.  Why should the 6309 not work like the
>> 6809, if it is supposed to be code compatible?
>> 
>> I would like to hear from anyone who may be able to assist me in figuring
>> out why my code that works fine on a CoCo with a 6809 does not work on a
>> CoCo with a 6309.  Am I doomed to purchasing a 6890 and installing that
>> instead?
>> 
>> Thanks very much for listening!
>> 
>> --
>> Stephen M. Pereira
>> Bedford, NH  03110
>> KB1SXE
>> 
>> 
>> --
>> Coco mailing list
>> Coco at maltedmedia.com
>> https://pairlist5.pair.net/mailman/listinfo/coco
>> 
> 
> -- 
> Coco mailing list
> Coco at maltedmedia.com
> https://pairlist5.pair.net/mailman/listinfo/coco