Why does my code crash AsmOne & AsmPro

oRBIT · 02 April 2021, 16:29

This is driving me nuts, doing some little testing here but it "crashes" both AsmOne and AsmPro in the same way during compilation.
** Unexpected end of file
** end of file
And usually a few moments later my entire WinUAE environment locks up so I need to "reboot" my virtual Amiga.

Anyone see anything obvious that can cause this crash? Do I need new glasses? Probably..

Would btw happily contribute to someone who wanted to continue to improve and bugfix AsmOne/AsmPro. Anyone up for the task?

Code:

PushToStack:	MACRO
		move.b	\1,([StackPtr])
		subq.b	#1,StackPtr+3
		ENDM

PullFromStack:	MACRO
		addq.b	#1,StackPtr+3
		move.b	([StackPtr]),\1
		ENDM

PushToStackW:	MACRO
		move.b	\1,([StackPtr])
		subq.b	#1,StackPtr+3
		lsr.w	#8,\1
		move.b	\1,([StackPtr])
		subq.b	#1,StackPtr+3
		ENDM

PullFromStackW:	MACRO
		addq.b	#1,StackPtr+3
		move.b	([StackPtr]),\1
		lsl.w	#8,\1
		addq.b	#1,StackPtr+3
		move.b	([StackPtr]),\1
	
		ENDM


start:		lea	mem,a0
.loop:		tst.w	a0
		beq.b	done
		addq.l	#1,a0
		bra.b	.loop

done:		move.l	#$100,d0
		move.l	a0,a1
		add.l	d0,a1
		move.l	a1,stackptr

		move.l	#$aabb,d1
		;PushToStackW d1

		;PullFromStackW d2		

		;move.l	#$a0,d1
		;PushToStack d1

		;move.l	#$b0,d1
		;PushToStack d1


		;PullFromStack d2
		;PullFromStack d3

		rts

StackPtr:	dc.l	0


mem:		blk.b	64*1024*2,0

Don_Adan · 02 April 2021, 16:43

For which CPU? I dont know if tst.w A0 exist (i never seen this in real), tst.l A0 for 68020+ exist. And you used addq.l #1,a0, for me logical is addq.l #2,a0.

oRBIT · 02 April 2021, 16:43

68020+
I suspect it's somehow related to the MACRO's...

Don_Adan · 02 April 2021, 16:48

Then remove all macros and check, if works later add one macro, next second macro etc

Don_Adan · 02 April 2021, 16:51

Btw. i dont coded long time, but perhaps i will use something like this:
move.l #mem+$fffe,D0
clr.w D0
move.l D0,A0

StingRay · 02 April 2021, 17:12

It's caused by the move.b ([StackPtr]),\1 line in MACRO PullFromStackW.
You have discovered yet another ASM-One/Pro MACRO bug, there are MANY more!

oRBIT · 02 April 2021, 19:29

Thanks alot StingRay!
You're not tempted taking up those old sources and fixing them up?

StingRay · 02 April 2021, 21:09

The only feasible approach would be a complete recode, everything else is bound to fail. ASM-Pro's source is available but a giant unmaintainable mess of disassembled ASM-One code, ASM-One's source is only available for the very early version which doesn't help much.

Many people, including myself, tried to fix bugs in ASM-Pro but everyone gave up sooner or later for obvious reasons.

The answer to your question is no, I'm afraid.

oRBIT · 02 April 2021, 21:11

Would be awesome with an updated assembler (that works similar to Asm-One/Asm-Pro). I guess it would be a pretty huge task though..

02 April 2021, 16:51	#5
Don_Adan Registered User Join Date: Jan 2008 Location: Warsaw/Poland Age: 55 Posts: 1,957	Btw. i dont coded long time, but perhaps i will use something like this: move.l #mem+$fffe,D0 clr.w D0 move.l D0,A0 Last edited by Don_Adan; 02 April 2021 at 17:02.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Why does this code crash?	kamelito	Coders. General	8	23 December 2019 18:38
LUCAS & FRANCES Boards, PAL Code & PCB Masks	Kai	Hardware mods	173	29 October 2017 18:35
'Harald Haardtand' whdload crash code 103	hippox89	support.FS-UAE	24	04 July 2017 04:40
ANSI control-code (?) bug in "all" AsmOne versions under Kick 2.0+	Photon	support.WinUAE	4	17 July 2011 19:29
Why nobody continue Asmone/AsmPro development ?	kamelito	Amiga scene	16	25 March 2007 15:08

02 April 2021, 16:43	#2
Don_Adan Registered User Join Date: Jan 2008 Location: Warsaw/Poland Age: 55 Posts: 1,957	For which CPU? I dont know if tst.w A0 exist (i never seen this in real), tst.l A0 for 68020+ exist. And you used addq.l #1,a0, for me logical is addq.l #2,a0.

02 April 2021, 16:43	#3
oRBIT Zone Friend Join Date: Apr 2006 Location: Gothenburg/Sweden Age: 48 Posts: 339	68020+ I suspect it's somehow related to the MACRO's...

02 April 2021, 16:48	#4
Don_Adan Registered User Join Date: Jan 2008 Location: Warsaw/Poland Age: 55 Posts: 1,957	Then remove all macros and check, if works later add one macro, next second macro etc

02 April 2021, 17:12	#6
StingRay move.l #$c0ff33,throat Join Date: Dec 2005 Location: Berlin/Joymoney Posts: 6,863	It's caused by the move.b ([StackPtr]),\1 line in MACRO PullFromStackW. You have discovered yet another ASM-One/Pro MACRO bug, there are MANY more!

02 April 2021, 19:29	#7
oRBIT Zone Friend Join Date: Apr 2006 Location: Gothenburg/Sweden Age: 48 Posts: 339	Thanks alot StingRay! You're not tempted taking up those old sources and fixing them up?

02 April 2021, 21:09	#8
StingRay move.l #$c0ff33,throat Join Date: Dec 2005 Location: Berlin/Joymoney Posts: 6,863	The only feasible approach would be a complete recode, everything else is bound to fail. ASM-Pro's source is available but a giant unmaintainable mess of disassembled ASM-One code, ASM-One's source is only available for the very early version which doesn't help much. Many people, including myself, tried to fix bugs in ASM-Pro but everyone gave up sooner or later for obvious reasons. The answer to your question is no, I'm afraid.

02 April 2021, 21:11	#9
oRBIT Zone Friend Join Date: Apr 2006 Location: Gothenburg/Sweden Age: 48 Posts: 339	Would be awesome with an updated assembler (that works similar to Asm-One/Asm-Pro). I guess it would be a pretty huge task though..

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)