Why doesn't DevPac recognize vectors: BLK.L 6

rcman · 30 January 2020, 19:33

I am sure this have been posted before. While trying to do a compile it says that

vectors: BLK.L 6

Instruction is not recognized.

This as well

sprite2: ;Address must be 64-bit aligned
BLK.B (SPRITE_DY+2)*16,0

Thanks
RC

Galahad/FLT · 30 January 2020, 20:13

change blk.l to dc.l

change blk.b to dc.b

rcman · 30 January 2020, 20:22

Quote:

Originally Posted by Galahad/FLT

change blk.l to dc.l

change blk.b to dc.b

Thank you so much!

RC

jotd · 30 January 2020, 20:40

errr I think you mean DS.L, and DS.B. DC.L and DC.B just declare one value, not the number in argument.

ross · 30 January 2020, 20:58

I think Phill meant dcb.l (or .b)

meynaf · 30 January 2020, 20:59

BLK is really neither DC (single constant) nor DS (define storage). It is a constant block (equivalent to DS if the constant value is 0).
For Devpac you can use DCB directive instead. Assemblers such as Phxass accept both directives.

But it's probably not what is meant in the provided code example, where DS would work just fine.

jotd · 30 January 2020, 21:01

using dc.b or dc.l would assemble, but would fail with strange errors because the data size will be too small. The joys of assembly...

rcman · 30 January 2020, 21:09

Okay I fixed the one problem but I cannot use ds.b here it gives me an Illedgal Operand and if I use dc.b it gives me Out of Range 8 Bit

sprite2: ;Address must be 64-bit aligned
dc.b (SPRITE_DY+2)*16,0
sprite3: ;Address must be 64-bit aligned
dc.b (SPRITE_DY+2)*16,0
sprite4: ;Address must be 64-bit aligned
dc.b (SPRITE_DY+2)*16,0
sprite5: ;Address must be 64-bit aligned
dc.b (SPRITE_DY+2)*16,0
sprite6: ;Address must be 64-bit aligned
dc.b (SPRITE_DY+2)*16,0
sprite7: ;Address must be 64-bit aligned
dc.b (SPRITE_DY+2)*16,0
spriteVoid: ;Address must be 64-bit aligned
dc.b 16,0
palette:
DC.W $0000, $0000

RC

jotd · 30 January 2020, 21:16

dc.b is INCORRECT. Fortunately it doesn't work because the size is higher than 255 and the assembler detects an overflow (dc.b is made to declare constants: for instance if I want to define a table with 1,2,3,4 I just do dc.b 1,2,3,4). You can't reserve 200 bytes with dc.b, unless you copy and paste 0 200 times.

Use dcb.b as noted in other posts.

phx · 30 January 2020, 21:17

See above. Either write

        dcb.b   (SPRITE_DY+2)*16,0

or

        ds.b    (SPRITE_DY+2)*16

"DS.B <n>" is effectively the same as "DCB.B <n>,0".

And make sure not to start the "DCB/DS" directive at the first column, otherwise it is regarded as a label.

Galahad/FLT · 30 January 2020, 22:04

Quote:

Originally Posted by ross

I think Phill meant dcb.l (or .b)

He did, but he's a dick!

jotd · 30 January 2020, 22:32

a typo that could have cost millions of $ in debugging shame on you Phil

30 January 2020, 19:33	#1
rcman Long time Amiga Owner Join Date: Jun 2017 Location: Uxbridge Canada Age: 57 Posts: 182	Why doesn't DevPac recognize vectors: BLK.L 6 I am sure this have been posted before. While trying to do a compile it says that vectors: BLK.L 6 Instruction is not recognized. This as well sprite2: ;Address must be 64-bit aligned BLK.B (SPRITE_DY+2)*16,0 Thanks RC

30 January 2020, 21:17	#10
phx Natteravn Join Date: Nov 2009 Location: Herford / Germany Posts: 2,544	See above. Either write dcb.b (SPRITE_DY+2)16,0 or ds.b (SPRITE_DY+2)16 "DS.B <n>" is effectively the same as "DCB.B <n>,0". And make sure not to start the "DCB/DS" directive at the first column, otherwise it is regarded as a label. Last edited by phx; 30 January 2020 at 21:18. Reason: EDIT: jotd has won! ;)

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
A600 doesn't recognize CF-cards	A500Fan	support.Hardware	17	01 March 2020 21:34
Doesn't Blitz2 recognize IDCMP_IDCMPUPDATE event?	peceha	Coders. Blitz Basic	4	18 April 2019 14:36
Winuae doesn't recognize Doubleclick with Win7	user	support.WinUAE	10	20 May 2018 22:40
iGame doesn't recognize two games...	Xenon	support.Apps	3	30 September 2009 18:54
GVP 040 Combo doesn't recognize SIMMs	astuermer	support.Hardware	9	28 May 2005 18:38

30 January 2020, 20:13	#2
Galahad/FLT Going nowhere Join Date: Oct 2001 Location: United Kingdom Age: 50 Posts: 9,017	change blk.l to dc.l change blk.b to dc.b

30 January 2020, 20:40	#4
jotd This cat is no more Join Date: Dec 2004 Location: FRANCE Age: 52 Posts: 8,369	errr I think you mean DS.L, and DS.B. DC.L and DC.B just declare one value, not the number in argument.

30 January 2020, 20:58	#5
ross Defendit numerus Join Date: Mar 2017 Location: Crossing the Rubicon Age: 54 Posts: 4,491	I think Phill meant dcb.l (or .b)

30 January 2020, 20:59	#6
meynaf son of 68k Join Date: Nov 2007 Location: Lyon / France Age: 51 Posts: 5,355	BLK is really neither DC (single constant) nor DS (define storage). It is a constant block (equivalent to DS if the constant value is 0). For Devpac you can use DCB directive instead. Assemblers such as Phxass accept both directives. But it's probably not what is meant in the provided code example, where DS would work just fine.

30 January 2020, 21:01	#7
jotd This cat is no more Join Date: Dec 2004 Location: FRANCE Age: 52 Posts: 8,369	using dc.b or dc.l would assemble, but would fail with strange errors because the data size will be too small. The joys of assembly...

30 January 2020, 21:09	#8
rcman Long time Amiga Owner Join Date: Jun 2017 Location: Uxbridge Canada Age: 57 Posts: 182	Okay I fixed the one problem but I cannot use ds.b here it gives me an Illedgal Operand and if I use dc.b it gives me Out of Range 8 Bit sprite2: ;Address must be 64-bit aligned dc.b (SPRITE_DY+2)16,0 sprite3: ;Address must be 64-bit aligned dc.b (SPRITE_DY+2)16,0 sprite4: ;Address must be 64-bit aligned dc.b (SPRITE_DY+2)16,0 sprite5: ;Address must be 64-bit aligned dc.b (SPRITE_DY+2)16,0 sprite6: ;Address must be 64-bit aligned dc.b (SPRITE_DY+2)16,0 sprite7: ;Address must be 64-bit aligned dc.b (SPRITE_DY+2)16,0 spriteVoid: ;Address must be 64-bit aligned dc.b 16,0 palette: DC.W $0000, $0000 RC

30 January 2020, 21:16	#9
jotd This cat is no more Join Date: Dec 2004 Location: FRANCE Age: 52 Posts: 8,369	dc.b is INCORRECT. Fortunately it doesn't work because the size is higher than 255 and the assembler detects an overflow (dc.b is made to declare constants: for instance if I want to define a table with 1,2,3,4 I just do dc.b 1,2,3,4). You can't reserve 200 bytes with dc.b, unless you copy and paste 0 200 times. Use dcb.b as noted in other posts.

30 January 2020, 22:32	#12
jotd This cat is no more Join Date: Dec 2004 Location: FRANCE Age: 52 Posts: 8,369	a typo that could have cost millions of $ in debugging shame on you Phil

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