poke function ....

Dony dony@willowglen.com.sg
Sun Sep 20 19:21:00 GMT 1998


Art Berggreen wrote:
> 
> >
> >Hii all,
> >Right now, I'm trying to porting 80x86 code to M68K.
> >In Intel, there is a function: pokeb(base_addr,offset,data)
> >and using Borland C Compiler will translated like this :
> >
> >       /* Assembly Routine to do a pokeb                */
> >       /* bp points at stack                            */
> >       /* stack + 0xc = data                            */
> >       /* stack + 0xa = register offset from base addr. */
> >       /* es - set to be segment value (CS)             */
> >       /* si - set to be register offset                */
> >       /* al - data                                     */
> >
> >       mov ax,[base_addr]
> >       mov es,ax
> >       mov al,[bp+0x0C]
> >       mov si,[bp+0x0A]
> >       mov es:[si],al
> >
> >I'm a high level programmer, and I don't really understand
> >where this 0x0c and 0x0a come from, and how could in 0x0c
> >is data, and in 0x0a is register offset??
> >Can somebody explain step by step what the assembler
> >trying to do??
> >Is there anybody M68K programmer know any C function to replace
> >pokeb, peekb function in M68K??
> 
> I don't know the 80x86 machine architecture or 80x86 C calling
> interface, but I'll make a few guesses.
> 
> It looks like pokeb() takes a variable holding a segment number,
> an offset into that segment, and data to write into memory ("poke").
> 
> >       mov ax,[base_addr]
> >       mov es,ax
> 
> Loads the content of the "base_address" variable into register es.
> 
> >       mov al,[bp+0x0C]
> 
> Loads the value of "data" which is being passed on the stack at a stack
> offset of 0x0c into register al.
> 
> >       mov si,[bp+0x0A]
> 
> Loads the value of "offset" which is being passed on the stack at a stack
> offset of 0x0a into register si.
> 
> >       mov es:[si],al
> 
> Stores the data in register al into memory at the location pointed to
> by the segment in register es and the offset in register si.
> 
> Since the m68k is not a segmented memory machine, the pokeb() function
> wouldn't need a base and offset, but in the interest of following the
> same model, an m68k routine could look like:
> 
> /* pokeb(char *base_addr, int offset, char data) */
>         .global pokeb
> pokeb:
>         move.l  4(%sp),%a0      /* get base_addr */
>         add.l   8(%sp),%a0      /* add offset */
>         move.b  15(%sp),(%a0)   /* move data byte */
>         rts
> 
> Art

Hii,
Can anybody tell me how the argument will be passed into stack for
function pokeb(char *base_addr, int offset, char data)??
Is it like below??

data highest byte (3rd arg, 4th byte)		->	sp+15
data high byte (3rd arg, 3rd byte)		->	sp+14
data low byte (3rd arg, 2nd byte)		->	sp+13
data lowest byte (3rd arg, 1st byte)		->	sp+12
offset highest byte (2nd arg, 4th byte)		->	sp+11
offset high byte (2nd arg, 3rd byte)		->	sp+10
offset low byte (2nd arg, 2nd byte)		->	sp+9
offset lowest byte (2nd arg, 1 byte)		->	sp+8
base_addr highest byte (1st arg, 4th byte)	->	sp+7
base_addr high byte (1st arg, 3rd byte)		->	sp+6
base_addr low byte (1st arg, 2nd byte)		->	sp+5
base_addr lowest byte (1st arg, 1st byte)	->	sp+4
sp highest byte					->	sp+3
sp high byte					->	sp+2
sp low byte					->	sp+1
sp lowest byte					->	sp+0

If it is the case, the assembler pokeb suppose to be like this :

		move.l  4(%sp),%a0      /* get base_addr */
		swap	%a0		/* exchange 16 bit halves, so that later on */
					/* address will be base_addr+offset */
		add.l   8(%sp),%a0      /* add offset */
		move.b  15(%sp),(%a0)   /* move data byte */
		rts


-- 
Dony
Email : dony@willowglen.com.sg



More information about the crossgcc mailing list