New tool for DragonDOS disk images!

[rolfmichelsen]

The OS-9 images in the archive should be good candidates, here https://archive.worldofdragon.org/brows ... 20Ltd/OS-9.

If I recall correctly, the Microdeal Mr Dig disk also loaded using a boot sector. I'm travelling this week and not able to test. Can try when I return home...

[sixxie]

Blockdown has a boot sector

[robcfg]

Hehe, thank you very much!

[robcfg]

A couple of questions.

Which sector is loaded when you type BOOT? LSN 2? Looking at Ciaran's Blockdown disk image on a hex editor it looks like the boot sector is the 3rd sector (or Track 0, side 0, sector 2).

I see on the DragonDOS manual that the boot sector is loaded at 9728 and executed at 9730. Are those decimal addresses? I see that the reason for the 2-byte offset in the addresses is that these two bytes contain the characters 'OS' which I guess is a way to check that it's a valid boot sector. Is that right?

[sixxie]

Which sector is loaded when you type BOOT? LSN 2? Looking at Ciaran's Blockdown disk image on a hex editor it looks like the boot sector is the 3rd sector (or Track 0, side 0, sector 2).

I see on the DragonDOS manual that the boot sector is loaded at 9728 and executed at 9730. Are those decimal addresses? I see that the reason for the 2-byte offset in the addresses is that these two bytes contain the characters 'OS' which I guess is a way to check that it's a valid boot sector. Is that right?

Yes, sectors 3 to 18 inclusive of track 0 (so LSNs 2..17) are loaded into memory starting from $2600 (9728 dec), and execution starts at $2602 (9730 dec) following the "OS" signature.

[robcfg]

Ah, I see! Thanks for the info!

So, this looks like your boot sector code?

Code: Select all

2602: 8E 26 2A        LDX       #$262A
2605: B6 A0 00        LDA       $A000
2608: 84 20           ANDA      #$20
260A: 27 03           BEQ       $260F
260C: 8E 26 3A        LDX       #$263A
260F: CE 02 DC        LDU       #$02DC
2612: DF A6           STU       <$A6
2614: A6 80           LDA       ,X+
2616: A7 C0           STA       ,U+
2618: 26 FA           BNE       $2614
261A: 7F 26 00        CLR       $2600
261D: AE E4           LDX       ,S
261F: EC 01           LDD       1,X
2622: 10 83 83 44     CMPD      #$8344
2625: 26 02           BNE       $2629
2627: 32 62           LEAS      2,S
2629: 39              RTS 

Followed by what seems to be the tokenized version of RUN"BLOCKD.BIN"

[sixxie]

looks good, here's the original

Code: Select all

2600                                  org $2600
2600                                  setdp 0         ; assumed
 
2600 4F53             boot_flag       fcc /OS/
 
2602 8E262A           boot_exec       ldx #run_dragon
2605 B6A000                           lda $a000       ; [$A000] points to ROM1 in CoCos
2608 8420                             anda #$20
260A 2703                             beq 10F         ; dragon
260C 8E263A                           ldx #run_coco2
260F CE02DC           10              ldu #$02dc
2612 DFA6                             stu $00a6
2614 A680             20              lda ,x+
2616 A7C0                             sta ,u+
2618 26FA                             bne 20B
261A 7F2600                           clr boot_flag   ; needed for coco
                                      ; work around DOSplus issue where BOOT does not
                                      ; expect to return
261D AEE4                             ldx ,s
261F EC01                             ldd 1,x
2621 10838344                         cmpd #$8344     ; looks like jump to error handler?
2625 2602                             bne 30F
2627 3262                             leas 2,s
2629 39               30              rts
 
262A 3A               run_dragon      fcc /:/
262B 8F22424C4F434B44                 fcc $8f,/"BLOCKD.BIN":/,0
     2E42494E223A00
 
263A 3A               run_coco2       fcc /:/
263B D34D22424C4F434B                 fcc $d3,/M"BLOCKD.BIN":/,$a2,0
     442E42494E223AA2
     00

[robcfg]

Niiiiiiice!

So, would a button on the main interface, named something like "View Boot track", leading to the disassembly of the content of sectors 3 to 18 be the way to go?

Then, If you want to view/edit individual sectors, I could import the hex view I did for my DragonBlock util and add the needed fields to select side, track and sector.

[Azerpy]

Hello Rob
I use the V1.0.6 version to extract and insert BAS ad BIN files, very useful.
I can easyly edit them with HxD editor.
So I do with the FLEXplorer tool for flex files.

I notice that your tool can read the DDos files without the vdk header, which is not necessary.

I have seen 2 little bugs (its seems to me)
- the last byte is not disassembled
- in the column "bytes", the lenght is the file lengh + 9

Regards. Gérard

[robcfg]

Hi Gérard, thank you for your nice feedback!

I notice that your tool can read the DDos files without the vdk header, which is not necessary.

For me it's useful because I generate a lot of data myself which then can go to a disk, a tape or a rom file, so it doesn't have a header by default.

Regarding the bugs:

- the last byte is not disassembled

You're right, I'll take a look at it.

- in the column "bytes", the lenght is the file lengh + 9

That's not a bug, but a feature

Jokes aside, the 9 extra bytes are the DragonDOS header, and if you type DIR on your Dragon, you'll see that DragonDOS shows the file size plus the header size, so I decided to leave like that for consistency. On top of that, if you extract binary files to edit them, you'll want also the header, so the size would be actually correct.

Cheers,
Rob