While looking through the arithmetics behind the procedure, I noticed that it could be shortened by changing the multiplier constant, please correct me if I'm wrong >,<:
_dwtoa proc dwValue, lpBuffer
push ebx
push edi
push esi
mov eax,[dwValue]
mov edi,[lpBuffer]
test eax,eax
jnz nonzero
mov byte ptr [edi],30h
jmp _dwx
nonzero:
jns signed
mov byte ptr [edi],2Dh
neg eax
inc edi
signed:
mov esi, 1999999Ah
xor ecx, ecx
.repeat
mov ebx, eax
mul esi
mov eax, edx
lea edx, [edx+edx*4]
add edx, edx
sub ebx, edx
or bl, 30h
mov [edi+ecx], bl
inc ecx
.until (!eax)
mov [edi+ecx],al
@@:
mov ah,[edi]
mov al,[edi+ecx-1]
mov [edi+ecx-1],ah
stosb
sub ecx,2
jnbe @B
_dwx:
pop esi
pop edi
pop ebx
ret
_dwtoa endp
hmm... not tested this one, but if i remember well, it generate few wrong values if you don't use thesvin's technic here... anyway this one is quite slow, especially for big values...
read this topic for a faster one : http://www.masm32.com/board/index.php?topic=8974.0
Thanks, i'll check it out =)