Re: 13-sector Disk ][ boot code won't boot

[pausch@saaf.se (Paul Schlyter)]

In article <KlhPi.6739$4V6.6681@newssvr14.news.prodigy.net>,
Chris Mosher <chris@mosher.mine.nu> wrote:

> I found the old Disk ][ ROMs that are for booting the 13-sector DOS 3.2 and 
> DOS 3.1 disks on asimov 
> ftp://ftp.apple.asimov.net/pub/apple_II/unsorted/13-sector-disk-ii-roms.zip
> and I thought I try using them to boot the DOS 3.1 nib image
> (such as DOS 3.1 Master-fixed.d13.gz in the same unsorted directory).
> Long story short... it wouldn't boot; it just printed ERR. I disassembled 
> the boot code and I
> am confused by it. It find the address header D5 AA B5 and sector 0 just 
> fine, and it reads in the
> encoded bytes, but the denibblizing routine doesn't seem to be doing the 
> right thing. Why?
> I compared it to the denibblizing routine in DOS 3.1 image, and to the 
> BOOT13 program
> (on the DOS 3.3 System Master), which both work fine, and this routine is 
> different.

On 13 sector disks, the nibblizing of the very first sector was different
than for the other sectors - presumably, this was to be able to squeeze
that different denibblizing code into the 256 byte boot PROM.

On 16 sector disks the nibblizing of all sectors are the same - the
very first sector is not different in that respect.

If you compare the 16-sector denibblizing code with the 13-sector
denibblizing code, you'll see that the 13-sector code is longer.
That's why a special nibblizing for the very first sector was
needed for 13-sector but not 16-sector disks.

> Does anybody have any idea what is going on here?
> Disassembly of the data-read and denibblizing routines follows:
> C6A1-   A0 9A       LDY   #$9A <-- read first 154 enc'd bytes
> C6A3-   84 3C       STY   $3C
> C6A5-   BC 8C C0    LDY   $C08C,X
> C6A8-   10 FB       BPL   $C6A5
> C6AA-   59 00 08    EOR   $0800,Y
> C6AD-   A4 3C       LDY   $3C
> C6AF-   88          DEY
> C6B0-   99 00 08    STA   $0800,Y
> C6B3-   D0 EE       BNE   $C6A3
> C6B5-   84 3C       STY   $3C      <--- read remaining 256 enc'd bytes
> C6B7-   BC 8C C0    LDY   $C08C,X
> C6BA-   10 FB       BPL   $C6B7
> C6BC-   59 00 08    EOR   $0800,Y
> C6BF-   A4 3C       LDY   $3C
> C6C1-   91 26       STA   ($26),Y <-- note 26-27 contains $0300
> C6C3-   C8          INY
> C6C4-   D0 EF       BNE   $C6B5
> C6C6-   BC 8C C0    LDY   $C08C,X
> C6C9-   10 FB       BPL   $C6C6
> C6CB-   59 00 08    EOR   $0800,Y
> C6CE-   D0 8D       BNE   $C65D
> C6D0-   60          RTS      (actually RTSs to next instr C6D1):
> C6D1-   A8          TAY   <-- start of denib routine that doesn't work
> C6D2-   A2 00       LDX   #$00
> C6D4-   B9 00 08    LDA   $0800,Y
> C6D7-   4A          LSR
> C6D8-   3E CC 03    ROL   $03CC,X
> C6DB-   4A          LSR
> C6DC-   3E 99 03    ROL   $0399,X
> C6DF-   85 3C       STA   $3C
> C6E1-   B1 26       LDA   ($26),Y
> C6E3-   0A          ASL
> C6E4-   0A          ASL
> C6E5-   0A          ASL
> C6E6-   05 3C       ORA   $3C
> C6E8-   91 26       STA   ($26),Y
> C6EA-   C8          INY
> C6EB-   E8          INX
> C6EC-   E0 33       CPX   #$33
> C6EE-   D0 E4       BNE   $C6D4
> C6F0-   C6 2A       DEC   $2A
> C6F2-   D0 DE       BNE   $C6D2
> C6F4-   CC 00 03    CPY   $0300
> C6F7-   D0 03       BNE   $C6FC
> C6F9-   4C 01 03    JMP   $0301
> C6FC-   4C 2D FF    JMP   $FF2D  <-- print ERR
> 
> Chris Mosher
-- 
----------------------------------------------------------------
Paul Schlyter,  Grev Turegatan 40,  SE-114 38 Stockholm,  SWEDEN
e-mail:  pausch at stockholm dot bostream dot se
WWW:     http://stjarnhimlen.se/