💾 Archived View for mirrors.apple2.org.za › archive › apple.cabi.net › Z80.CPM › A2.CPM.files › apun… captured on 2023-01-29 at 08:39:32.
-=-=-=-=-=-=-
The unspool program by Gary P. Novosielski is a very useful routine,
particularly for those with slow printers. Apple owners will never
get beyond the message "Error in system table" without a little added
massage. Gary aimed me in the right direction, any errors are mine.
The big problem is that standard CP/M 2.2 has ( & unspool expects) 16
jump vectors st the start of BIOS, counting from the warm boot vector.
Microsoft ("in their infinite wisdom")only saw fit to provide standard
implementations for 14 of these. Unspool uses number 15, LISTST:, which
checks list device ready status, returning 00 in a if not ready, 0ffh
in a if ready. {Implementing vector 16 must wait for another day.}
First, Unspool must be modified to only expect 15 jumps. At (about)
line 104, NVECTS equ 16 must be changed from 16 to 15, so that Unspool
will be happy with just 15 vectors.
Next, you must add and implement vector No. 15. This vector should be
found at address 242dh in cpm56.com (or 0da2dh in 56k CP/M BIOS, or,
according to my manual, at 0aa2dh in 44k CP/M BIOS; all future address
references will be to the image in cpm56.com). Occupying that location
now is [xor a, ret] af c9 00, which is a simulated and jumpless LISTST,
always returning 00 in a. At this address you will have to place a jump
to where you will want to place LISTST; I would suggest 2483h as this
location, hence use DDT to install c3 83 da. Naturally, the jump address
must be the actual CP/M address, not the address it occupies in
cpm56.com, that is da not 24 high-order byte. Note that all these changes
in cpm56.com will be incorporated in CP/M by running cpm56.com, but
remember after this to reestablish any special items you have used
CONFIGIO to install, such as printer drivers, keyboard redefinitions
or undefinitions, terminal configurations, etc., as cpm56.com returns
these all to their default values.
Finally you have to write your own LISTST code, and find a place to
put it. The batches of 00's starting at 2433h are a series of 16
byte disk parameter headers, 6 in all for disk drives a: through f:.
If, like me, you will never need drive f:, you can use that location
(16 bytes in length) for LISTST. The location is 2483h through
2492h. A LISTST (in z-80 code) for the CCS-7728 parallel printer
card in slot 1 is given below, it is only 10 bytes (11 if you con-
vert to 8080 code by changing the relative jump). Note that the
status code of the 7728 is exactly the opposite of that expected by
LISTST, hence the xor a and cpl codes to get things right.
;This is a is a simple implementation of LISTST for the
;ccs-7728 printer card in slot 1
.z80
E090 status equ 0e090h ;status port, ccs-7728
;in slot 1
0000' aseg
org 0da83h ;loc. of DPH for
;drive f:
DA83 3A E090 ld a,(status) ;printer status to a
DA86 B7 or a ;set flags
DA87 28 02 jr z,ready ;if zero, jump to
;next routine
DA89 AF xor a ;set a to 00h, meaning
;not ready
DA8A C9 ret ;and return
DA8B 2F ready: cpl ;since a was 0,
;complement = 0ffh
DA8C C9 ret ;a=ff shows lst: ready
;for char.
end 0da83h
That's all there is to it, folks, sit back and enjoy printing while you
compute!!
Bob Thompson
Reply messages via Hyde Park or Calamity Cliffs RCPM, Chicago, please.