| |
The Human Code Machine
Registered: Sep 2005
Posts: 112 |
Dasm rorg directive
Hi, I'm looking for a DASM cross assembler version newer than 2.12 for DOS/Win with correctly working rorg directive. All newer versions I found on the net can't assemble the lower attached test source. I found it ages ago on one of the famous Fred Fish disks and I don't want to switch to another one :( I tried to contact one of the new maintainers, but I got no response. Anybody here has a working version or knows the dasm sourcecode a bit and has time to fix it?
processor 6502
org $c000
Main subroutine
sei
jsr Test
jmp nmi
Test subroutine
ldx #nmi_end-nmi_start-1
.loop lda nmi_start,x
sta nmi,x
dex
bpl .loop
rts
nmi_start subroutine
RORG $02
nmi subroutine
.1 INC $d020
jmp .1
rend
nmi_end
|
|
| |
Danzig
Registered: Jun 2002
Posts: 440 |
switch to kickassembler and explore a whole new world for you :D sorry, not the advise you expected but really: give it a try! |
| |
iAN CooG
Registered: May 2002
Posts: 3197 |
Quote: Hi, I'm looking for a DASM cross assembler version newer than 2.12 for DOS/Win with correctly working rorg directive. All newer versions I found on the net can't assemble the lower attached test source. I found it ages ago on one of the famous Fred Fish disks and I don't want to switch to another one :( I tried to contact one of the new maintainers, but I got no response. Anybody here has a working version or knows the dasm sourcecode a bit and has time to fix it?
processor 6502
org $c000
Main subroutine
sei
jsr Test
jmp nmi
Test subroutine
ldx #nmi_end-nmi_start-1
.loop lda nmi_start,x
sta nmi,x
dex
bpl .loop
rts
nmi_start subroutine
RORG $02
nmi subroutine
.1 INC $d020
jmp .1
rend
nmi_end
Grab my version, I just compiled it fine.
http://iancoog.altervista.org/
org $c000
Main subroutine
sei
jsr Test
jmp nmi
Test subroutine
ldx #nmi_end-nmi_start-1
.loop lda nmi_start,x
sta nmi,x
dex
bpl .loop
rts
nmi_start subroutine
RORG $02
nmi subroutine
.1 inc $d020
jmp .1
REND
nmi_end
dasm "x.s" -o"x.prg" -v2 -l"x.lst"
------- FILE x.s LEVEL 1 PASS 3
1 c000 org $c000
2 c000
3 c000 Main subroutine
4 c000 78 sei
5 c001 20 07 c0 jsr Test
6 c004 4c 02 00 jmp nmi
7 c007
8 c007 Test subroutine
9 c007 a2 05 ldx #nmi_end-nmi_start-1
10 c009 bd 12 c0 .loop lda nmi_start,x
11 c00c 95 02 sta nmi,x
12 c00e ca dex
13 c00f 10 f8 bpl .loop
14 c011 60 rts
15 c012
16 c012
17 c012 nmi_start subroutine
18 c012
19 c012 0000 RORG $02
20 c012 0000 nmi subroutine
21 c012 0002 ee 20 d0 .1 inc $d020
22 c015 0005 4c 02 00 jmp .1
23 c018 REND
24 c018
25 c018 nmi_end
|
| |
tlr
Registered: Sep 2003
Posts: 1790 |
It doesn't assemble in dasm V2.20.10, but not due to the rorg directive.
It rather fails because it assumes "sta nmi,x" is 3 bytes long in the first pass and in the second pass nmi_start has then moved one byte.
It probably behaves wrong in this sense.
If you force it to long by .wx it works. I think you can force it to zp too, but I don't remember how. |
| |
iAN CooG
Registered: May 2002
Posts: 3197 |
Quote: It doesn't assemble in dasm V2.20.10, but not due to the rorg directive.
It rather fails because it assumes "sta nmi,x" is 3 bytes long in the first pass and in the second pass nmi_start has then moved one byte.
It probably behaves wrong in this sense.
If you force it to long by .wx it works. I think you can force it to zp too, but I don't remember how.
the other dasms around sux. mine rulz. end of the story.
:D |
| |
chatGPZ
Registered: Dec 2001
Posts: 11386 |
any version of dasm sucks arse, fact =P |
| |
iAN CooG
Registered: May 2002
Posts: 3197 |
Quote: any version of dasm sucks arse, fact =P
hmpf. you're not even worth a reply
OMFG I replied anyway!1 /o\ |
| |
The Human Code Machine
Registered: Sep 2005
Posts: 112 |
Thx iAN CooG! I'll give it a try. I think DASM is still a good assembler and I like the way it handles local labels and I didn't find an assembler which offers this the same way. Perhaps I was too lazy to look deep enough... |
| |
chatGPZ
Registered: Dec 2001
Posts: 11386 |
:o)
does yours give some half sane error messages at least? i have to use dasm for work atm and its inability to even print meaningful errors drives my nuts =)
thcm: many assemblers support "cheap" local labels :) dasm really doesnt have any special features at all. |
| |
The Human Code Machine
Registered: Sep 2005
Posts: 112 |
@Groepaz: For me the SUBROUTINE is special enough and I like the simplicity and readability of my good old trusted DASM assembler. I don't like the high level stuff like the script stuff used in kickassembler. I also don't like the way local labels are handled, but that's just my personal taste. Thx everybody for the help. |
| |
iAN CooG
Registered: May 2002
Posts: 3197 |
Quote: :o)
does yours give some half sane error messages at least? i have to use dasm for work atm and its inability to even print meaningful errors drives my nuts =)
thcm: many assemblers support "cheap" local labels :) dasm really doesnt have any special features at all.
Huh, depends on what do you mean by sane =)
Any example of weird messages?
I have to remember that I stripped from my build any other processor, it's 6502 only, so no need to write processor 6502 on top anymore -it's simply ignored - so if you use it for some microcontroller coding with different cpu it's no use for you. I only need coding in asm for the C64 =) |
| |
chatGPZ
Registered: Dec 2001
Posts: 11386 |
well the one i use ("official" 2.20.11) doesnt really give any errors at all. for example when you make a simple typo somewhere, it just says assembling failed, doesnt give a line number or whatever, VERY annoying =P
and yeah well, i use it for 6502 :)
|
| |
iAN CooG
Registered: May 2002
Posts: 3197 |
Quote: well the one i use ("official" 2.20.11) doesnt really give any errors at all. for example when you make a simple typo somewhere, it just says assembling failed, doesnt give a line number or whatever, VERY annoying =P
and yeah well, i use it for 6502 :)
oh no, that would be unusable for sure.
My version is based on 2.20.07 but I can assure it gives errors and line numbers even with 0 verboseness. But it's been so modified that has nothing to spare with the original source, after 4 years of mods.
-v and -l are there for extra verboseness. I use -v 2 and -lname.lst all the time
|
| |
chatGPZ
Registered: Dec 2001
Posts: 11386 |
gotta try it then i guess :)
|
| |
Frantic
Registered: Mar 2003
Posts: 1648 |
iAN: Is your mod available somewhere?
//FTC |
| |
Mace
Registered: May 2002
Posts: 1799 |
Frantic, check message nr. 3 in this thread.
You'll find a link to iAN's website ;) |
| |
Martin Piper
Registered: Nov 2007
Posts: 722 |
This works in ACME:
*=$c000
sei
jsr Test
jmp nmi
Test
ldx #nmi_end-nmi_start-1
.loop lda nmi_start,x
sta+1 nmi,x ; +1 forces 8 bit addressing mode
dex
bpl .loop
rts
nmi_start
!pseudopc $02 {
nmi
.1 INC $d020
jmp .1
rend
}
nmi_end
|
| |
chatGPZ
Registered: Dec 2001
Posts: 11386 |
urks, i tried your version ian.... and unfortunately i have to say that it sucks even more than the official one. the program counter seems to get confused under certain circumstances :/
747 9ffa ; BANK 1 Vectors at $fffa
748 9ffa org $9ffa
749 9ffa a0 ff .byte.b $a0,$ff ; NMI
750 9ffc ac ff .byte.b $ac,$ff ; Reset
751 9ffe ac ff .byte.b $ac,$ff ; IRQ
results in:
0000:1FD0 | D0 8D 21 D0 A0 00 B9 72 FF 99 00 02 C8 D0 F7 4C
0000:1FE0 | 00 02 A0 FF AC FF AC FF 4C 12 80 4C 12 80 FF 43
0000:1FF0 | 42 4D A0 00 B9 3B A8 99 00 40 C8 D0 F7 A5 06 F0
the very same file assembled with official 2.20.11 gives the expected (correct) output. so now i can choose between useful error messages and correct output. "great" =P
|
| |
iAN CooG
Registered: May 2002
Posts: 3197 |
Quote: urks, i tried your version ian.... and unfortunately i have to say that it sucks even more than the official one. the program counter seems to get confused under certain circumstances :/
747 9ffa ; BANK 1 Vectors at $fffa
748 9ffa org $9ffa
749 9ffa a0 ff .byte.b $a0,$ff ; NMI
750 9ffc ac ff .byte.b $ac,$ff ; Reset
751 9ffe ac ff .byte.b $ac,$ff ; IRQ
results in:
0000:1FD0 | D0 8D 21 D0 A0 00 B9 72 FF 99 00 02 C8 D0 F7 4C
0000:1FE0 | 00 02 A0 FF AC FF AC FF 4C 12 80 4C 12 80 FF 43
0000:1FF0 | 42 4D A0 00 B9 3B A8 99 00 40 C8 D0 F7 A5 06 F0
the very same file assembled with official 2.20.11 gives the expected (correct) output. so now i can choose between useful error messages and correct output. "great" =P
wtf?
org $9ffa
.byte $a0,$ff ; NMI
.byte $ac,$ff ; Reset
.byte $ac,$ff ; IRQ
this assembles just fine. Probably there is more you are not showing? |
| |
chatGPZ
Registered: Dec 2001
Posts: 11386 |
ofcourse, this is from a large project (mmcr bios). i have to admit that i am not really in the mood to cut it down into a simple testcase right now :) |
| |
WVL
Registered: Mar 2002
Posts: 902 |
Ehr hold on..
Are you saying you have the sources to the mmcr bios... also for the built-in sidplayer? Willing to share (if you are allowed?) so we can fix that thing? |
| |
chatGPZ
Registered: Dec 2001
Posts: 11386 |
i can't share - but i *am* fixing it, just right now =) |
| |
Ninja
Registered: Jan 2002
Posts: 411 |
gpz: Hint: You know that RR development was the main reason for Doc Bacardi to create DreamAss ;) |
| |
iAN CooG
Registered: May 2002
Posts: 3197 |
Quote: ofcourse, this is from a large project (mmcr bios). i have to admit that i am not really in the mood to cut it down into a simple testcase right now :)
I fear some reversed org addresses can cause errors.
Make sure the lowest org is declared 1st.
Also raising the number of build passes can help resolving labels declared later in the source. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11386 |
Quote:gpz: Hint: You know that RR development was the main reason for Doc Bacardi to create DreamAss ;)
hehe well, i don't really feel like converting the source to another assembler :) (and then i'd choose ca65 =P) |
| |
Ninja
Registered: Jan 2002
Posts: 411 |
:D
Well, as long as there is a new binary for all of us at the end of the day. Oh. and what WVL said! ;)
|
