Programming Forums > Application Development > Assembly

reynard082 · Dec 13th, 2004, 3:15 PM

I have to create a assembly program that does the Fibonacci series using recursion. I wrote the program in 16bit code using a subproc for the fibonacci recursion and debugged it until it works. Then I converted the code to 32bit code by adding the E's to the registers and adding a few lines to the top. I also doubled the pointer addresses that I use. This program also has to be complied with C++ code.

Like I said when the program runs in 16-bit mode, everything work fine. But when it's converted and complied w/ C++ code, when I input 1, it works fine. However when I put in 2 I get 4551668. However when I put in 3, i get 4551669. If i substract 4551668 from whatever number I get, it works. So here's my question - where does the 4551668 come from? Is there something i'm not taking into consideration when i'm debugging it?

any suggestion would be extremely helpful...

thanks
Matt

here's my code...

(Toggle Plain Text)

.386p
.model flat
.code

_fib proc near

	push ebp
	mov ebp,esp
	mov eax,[ebp+8]
  
	cmp eax,1
	je L2A
	cmp eax,0
	je L2B
	
	L1:

;fib(n-2)
	
                sub eax,2
 push eax
 CALL _fib 

 add esp,4

	;fib(n-1)

 mov eax,[ebp+8];reset N value
	
                sub eax,1
 push eax
 CALL _fib 

 add esp,4
 mov eax,edx

 jmp L2B
	
	L2A:
                add edx,1
 pop ebp
 jmp L2D
 
	L2B:
 pop ebp
 
	L2D:
 ret

_fib endp
END

my C++ code

(Toggle Plain Text)

#include <iostream>
using namespace std;

extern "C" unsigned int fib(int);

int main ()
	{
	int N;
	unsigned int result;

        cout<<"Enter a Postitive Integer: ";
	cin>>N;

        result=fib(N);
        cout<<N<<" fib number -> "<<result<<endl;

        cout<<"Enter a Postitive Integer: ";
	cin>>N;
        }
  
	return 0;
	}

saxman · Dec 25th, 2004, 5:59 PM

Can you post the original 16-bit code?

liquidsilver · Mar 11th, 2005, 3:02 PM

Maybe you cleared ax and not eax?

PS: In your code change Postitive > Positive

Christian_Rosenkreuz_777 · Mar 11th, 2005, 4:53 PM

They're compatible. If you want to take advantage of 32-bit registers, instead of ax write eax. I use 16-bit though because i'm used to it and I haven't needed more room.

liquidsilver · Mar 12th, 2005, 12:36 PM

Christian_Rosenkreuz_777: I thought that's what he did:

Quote:

Then I converted the code to 32bit code by adding the E's to the registers and adding a few lines to the top.

Dec 13th, 2004, 3:15 PM	#1
reynard082 Newbie Join Date: Dec 2004 Posts: 1 Rep Power: 0	I have to create a assembly program that does the Fibonacci series using recursion. I wrote the program in 16bit code using a subproc for the fibonacci recursion and debugged it until it works. Then I converted the code to 32bit code by adding the E's to the registers and adding a few lines to the top. I also doubled the pointer addresses that I use. This program also has to be complied with C++ code. Like I said when the program runs in 16-bit mode, everything work fine. But when it's converted and complied w/ C++ code, when I input 1, it works fine. However when I put in 2 I get 4551668. However when I put in 3, i get 4551669. If i substract 4551668 from whatever number I get, it works. So here's my question - where does the 4551668 come from? Is there something i'm not taking into consideration when i'm debugging it? any suggestion would be extremely helpful... thanks Matt here's my code... (Toggle Plain Text) .386p .model flat .code _fib proc near push ebp mov ebp,esp mov eax,[ebp+8] cmp eax,1 je L2A cmp eax,0 je L2B L1: ;fib(n-2) sub eax,2 push eax CALL _fib add esp,4 ;fib(n-1) mov eax,[ebp+8];reset N value sub eax,1 push eax CALL _fib add esp,4 mov eax,edx jmp L2B L2A: add edx,1 pop ebp jmp L2D L2B: pop ebp L2D: ret _fib endp END .386p .model flat .code _fib proc near push ebp mov ebp,esp mov eax,[ebp+8] cmp eax,1 je L2A cmp eax,0 je L2B L1: ;fib(n-2) sub eax,2 push eax CALL _fib add esp,4 ;fib(n-1) mov eax,[ebp+8];reset N value sub eax,1 push eax CALL _fib add esp,4 mov eax,edx jmp L2B L2A: add edx,1 pop ebp jmp L2D L2B: pop ebp L2D: ret _fib endp END my C++ code (Toggle Plain Text) #include <iostream> using namespace std; extern "C" unsigned int fib(int); int main () { int N; unsigned int result; cout<<"Enter a Postitive Integer: "; cin>>N; result=fib(N); cout<<N<<" fib number -> "<<result<<endl; cout<<"Enter a Postitive Integer: "; cin>>N; } return 0; } #include <iostream> using namespace std; extern "C" unsigned int fib(int); int main () { int N; unsigned int result; cout<<"Enter a Postitive Integer: "; cin>>N; result=fib(N); cout<<N<<" fib number -> "<<result<<endl; cout<<"Enter a Postitive Integer: "; cin>>N; } return 0; }

Mar 11th, 2005, 3:02 PM	#3
liquidsilver Newbie Join Date: Mar 2005 Location: South Africa Posts: 21 Rep Power: 0	Maybe you cleared ax and not eax? PS: In your code change Postitive > Positive __________________ Small is beautiful

Mar 11th, 2005, 4:53 PM	#4
Christian_Rosenkreuz_777 Newbie Join Date: Feb 2005 Location: Charlottesville, VA (rated #1 place to live) Posts: 25 Rep Power: 0	They're compatible. If you want to take advantage of 32-bit registers, instead of ax write eax. I use 16-bit though because i'm used to it and I haven't needed more room. __________________ Cavear Emptor

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Dec 25th, 2004, 5:59 PM	#2
saxman Newbie Join Date: Dec 2004 Posts: 7 Rep Power: 0	Can you post the original 16-bit code?

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)

Programming Forums > Application Development > Assembly

Converting 16bit Code To 32bit Code