Re: [Nasm-users] code problems--relatively new

[Frank K. <fbk...@zy...>]

Tyler Littlefield wrote:
> hello,
> thanks; I'll do that.
> I'd origenally wanted to  do a strlen type thing, and then just put the 
> message or something in, and then run the call, so that I wouldn't have 
> to define messages for everything in the .data, if there would be an 
> efficient way of doing it.
> So, I would send "hello world." to the stack, then pop it off in the 
> write func and use the strlen for edx.

Something like?

global _start

section .data
	msg db "Hello world!", 10, 0

section .text
_start:
	nop
	push msg
	call print_zero_terminated_string
	add esp, 4
	...
	mov eax, 1
	int 80h

print_zero_terminated_string:
	mov ecx, [esp + 4] ; buffer in ecx
	xor edx, edx ; get length in edx
.getlen:
	cmp byte [ecx + edx], 0
	jz .gotlen
	inc edx
	jmp .getlen
.gotlen:
	mov ebx, 1 ; stdout
	mov eax, 4 ; __NR_write
	int 80h
	ret


Of course, Nasm will calculate the length for you...

msg db "Hello world", 10
msg_len equ $ - msg

...saving you having to walk down the string byte-by-byte looking for 
the zero. There's an algorithm around for scanning a string a dword at a 
time - worthwhile if you've got long strings, or a lot of them - but 
it's long and complicated. Fastest way is to not do it. If you're 
interfacing with C, you may be stuck with zero-terminated strings, but 
it isn't the most efficient data structure...

Or maybe that's not what you had in mind...

Best,
Frank