commit Bart's patch to fix CP/M call psp:05h compatibility call along with some comments to explain the call trampoline

git-svn-id: https://svn.code.sf.net/p/freedos/svn/kernel/trunk@1372 6ac86273-5f31-0410-b378-82cca8765d1b
This commit is contained in:
Kenneth J Davis 2009-05-01 04:25:27 +00:00
parent 497a2337ef
commit 240bcc30e2
2 changed files with 68 additions and 58 deletions

View File

@ -65,14 +65,22 @@ typedef struct {
UWORD ps_exit; /* 00 CP/M-like exit point: int 20 */
UWORD ps_size; /* 02 segment of first byte beyond */
/* memory allocated to program */
BYTE ps_fill1; /* 04 single char fill */
BYTE ps_fill1; /* 04 single char fill=0 */
/* CP/M-like entry point */
UBYTE ps_farcall; /* 05 far call opcode */
/* CP/M-like entry point */
/* offsets 5-9 are a far call to absolute address 0:000Ch
coded so that CP/M apps can do a near call to psp:5, does a
far call but ensures word at offset 6 is size of COM file
e.g. FEF0h by using 1MB wrap around address 0F01D:FEF0
(jmp code stored at 0:000C should be duplicated in HMA FFFF:00D0)
Note: MS-DOS has value as FEEE which wraps to 0:00BEh */
UBYTE ps_farcall; /* 05 far call opcode */
VOID(FAR ASMCFUNC * ps_reentry) (void); /* 06 re-entry point */
intvec ps_isv22, /* 0a terminate address */
ps_isv23, /* 0e break address */
ps_isv24; /* 12 critical error address */
intvec ps_isv22, /* 0a terminate address */
ps_isv23, /* 0e break address */
ps_isv24; /* 12 critical error address */
UWORD ps_parent; /* 16 parent psp segment */
UBYTE ps_files[20]; /* 18 file table - 0xff is unused */
UWORD ps_environ; /* 2c environment paragraph */

View File

@ -28,10 +28,10 @@
; $Id$
;
%include "segs.inc"
%include "segs.inc"
%include "stacks.inc"
segment HMA_TEXT
segment HMA_TEXT
extern _int21_syscall
extern _int21_service
extern _int2526_handler
@ -50,7 +50,7 @@ segment HMA_TEXT
extern _Int21AX:wrt DGROUP
extern _DGROUP_
extern _DGROUP_
global reloc_call_cpm_entry
global reloc_call_int20_handler
@ -65,9 +65,14 @@ segment HMA_TEXT
; VOID FAR
; cpm_entry(iregs UserRegs)
;
; This one is a strange one. The call is to psp:0005h but it returns to the
; function after the call. What we do is convert it to a normal call and
; fudge the stack to look like an int 21h call.
; For CP/M compatibility allow a program to invoke any DOS API function
; between 0 and 24h by doing a near call to psp:0005h which embeds a far call
; to absolute address 0:00C0h (int vector 30h & 31h) or FFFF:00D0 (hma).
; Note: int 31h is also used for DPMI
; Upon entry the stack has a near return offset (desired return address) and
; far return seg:offset (useless return to data at offset 0ah after far call
; in psp). We convert it to a normal call and correct the stack to appear same
; as if invoked via an int 21h call including proper return address.
;
reloc_call_cpm_entry:
; Stack is:
@ -75,10 +80,7 @@ reloc_call_cpm_entry:
; psp seg
; 000ah
;
push bp ; trash old return address
mov bp,sp
xchg bp,[2+bp]
pop bp
add sp, 2 ; remove unneeded far return offset 0ah
pushf ; start setting up int 21h stack
;
; now stack is
@ -107,12 +109,12 @@ reloc_call_cpm_entry:
; psp seg (alias .COM cs)
; return offset
;
cmp cl,024h
jbe cpm_error
cmp cl,024h ; restrict calls to functions 0-24h
ja cpm_error
mov ah,cl ; get the call # from cl to ah
jmp reloc_call_int21_handler ; do the system call
cpm_error: mov al,0
iret
iret ; cleanup stack and return to caller
;
; interrupt zero divide handler:
@ -123,20 +125,20 @@ cpm_error: mov al,0
; int20_handler(iregs UserRegs)
;
print_hex: mov cl, 12
print_hex: mov cl, 12
hex_loop:
mov ax, dx
shr ax, cl
and al, 0fh
cmp al, 10
sbb al, 69h
das
mov bx, 0070h
mov ah, 0eh
int 10h
sub cl, 4
jae hex_loop
ret
mov ax, dx
shr ax, cl
and al, 0fh
cmp al, 10
sbb al, 69h
das
mov bx, 0070h
mov ah, 0eh
int 10h
sub cl, 4
jae hex_loop
ret
divide_by_zero_message db 0dh,0ah,'Interrupt divide by zero, stack:',0dh,0ah,0
@ -159,27 +161,27 @@ zero_message_loop:
jmp short zero_message_loop
zero_done:
mov bp, sp
xor si, si ; print 13 words of stack for debugging LUDIV etc.
stack_loop:
mov bp, sp
xor si, si ; print 13 words of stack for debugging LUDIV etc.
stack_loop:
mov dx, [bp+si]
call print_hex
mov al, ' '
int 10h
inc si
inc si
cmp si, 13*2
jb stack_loop
mov al, 0dh
int 10h
mov al, 0ah
int 10h
call print_hex
mov al, ' '
int 10h
inc si
inc si
cmp si, 13*2
jb stack_loop
mov al, 0dh
int 10h
mov al, 0ah
int 10h
mov ax,04c7fh ; terminate with errorlevel 127
int 21h
sti
thats_it: hlt
jmp short thats_it ; it might be command.com that nukes
sti
thats_it: hlt
jmp short thats_it ; it might be command.com that nukes
invalid_opcode_message db 0dh,0ah,'Invalid Opcode at ',0
@ -299,7 +301,7 @@ int21_1:
; I don't know who needs that, but ... (TE)
;
mov word [_user_r+2],ss
mov word [_user_r],bp ; store and init
mov word [_user_r],bp ; store and init
;
; Decide which stack to run on.
@ -377,9 +379,9 @@ int21_exit_nodec:
%ifdef WATCOM
sub bp, 4 ; for fs and gs only
%else
sub bp, 6 ; high parts of eax, ebx or ecx, edx
sub bp, 6 ; high parts of eax, ebx or ecx, edx
%endif
%endif
%endif
cli
mov ss,si
@ -387,7 +389,7 @@ int21_exit_nodec:
int21_ret:
Restore386Registers
POP$ALL
POP$ALL
;
; ... and return.
@ -466,8 +468,8 @@ int2526:
Protect386Registers
push dx
push cx ; save user stack
push dx
push cx ; save user stack
push dx ; SS:SP -> user stack
push cx
@ -475,8 +477,8 @@ int2526:
call _int2526_handler
add sp, byte 6
pop cx
pop dx ; restore user stack
pop cx
pop dx ; restore user stack
Restore386Registers