NASM addition program - linux

I am a developer who uses high level languages, learning assembly language in my spare time. Please see the NASM program below:
section .data
section .bss
section .text
global main
main:
mov eax,21
mov ebx,9
add eax,ebx
mov ecx,eax
mov eax,4
mov ebx,1
mov edx,4
int 0x80
push ebp
mov ebp,esp
mov esp,ebp
pop ebp
ret
Here are the commands I use:
ian#ubuntu:~/Desktop/NASM/Program4$ nasm -f elf -o asm.o SystemCalls.asm
ian#ubuntu:~/Desktop/NASM/Program4$ gcc -o program asm.o
ian#ubuntu:~/Desktop/NASM/Program4$ ./program
I don't get any errors, however nothing is printed to the terminal. I used the following link to ensure the registers contained the correct values: http://docs.cs.up.ac.za/programming/asm/derick_tut/syscalls.html

You'll have to convert the integer value to a string to be able to print it with sys_write (syscall 4). The conversion could be done like this (untested):
; Converts the integer value in EAX to a string in
; decimal representation.
; Returns a pointer to the resulting string in EAX.
int_to_string:
mov byte [buffer+9],0 ; add a string terminator at the end of the buffer
lea esi,[buffer+9]
mov ebx,10 ; divisor
int_to_string_loop:
xor edx,edx ; clear edx prior to dividing edx:eax by ebx
div ebx ; EAX /= 10
add dl,'0' ; take the remainder of the division and convert it from 0..9 -> '0'..'9'
dec esi ; store it in the buffer
mov [esi],dl
test eax,eax
jnz int_to_string_loop ; repeat until EAX==0
mov eax,esi
ret
buffer: resb 10

programming in assembly requires a knowledge of ASCII codes and a some basic conversion routines. example: hexadecimal to decimal, decimal to hexadecimal are good routines to keep somewhere on some storage.
No registers can be printed as they are, you have to convert (a lot).
To be a bit more helpfull:
ASCII 0 prints nothing but some text editors (kate in kde linux) will show something on screen (a square or ...). In higher level language like C and C++ is it used to indicate NULL pointers and end of strings.
Usefull to calculate string lengths too.
10 is end of line. depending Linux or Windows there will be a carriage return (Linux) too or not (Windows/Dos).
13 is carriage return
1B is the ESC key (Linux users will now more about this)
255 is a hard return, I never knew why it is good for but it must have its purpose.
check http://www.asciitable.com/ for the entire list.

Convert the integer value to a string.
Here i have used macros pack and unpack to convert integers to string and macro unpack to do the vice-versa
%macro write 2
mov eax, 4
mov ebx, 1
mov ecx, %1
mov edx, %2
int 80h
%endmacro
%macro read 2
mov eax,3
mov ebx,0
mov ecx,%1
mov edx,%2
int 80h
%endmacro
%macro pack 3 ; 1-> string ,2->length ,3->variable
mov esi, %1
mov ebx,0
%%l1:
cmp byte [esi], 10
je %%exit
imul ebx,10
movzx edx,byte [esi]
sub edx,'0'
add ebx,edx
inc esi
jmp %%l1
%%exit:
mov [%3],ebx
%endmacro
%macro unpack 3 ; 1-> string ,2->length ,3->variable
mov esi, %1
mov ebx,0
movzx eax, byte[%3]
mov byte[%2],0
cmp eax, 0
jne %%l1
mov byte[%2],1
push eax
jmp %%exit2
%%l1:
mov ecx,10
mov edx,0
div ecx
add edx,'0'
push edx
inc byte[%2]
cmp eax, 0
je %%exit2
jmp %%l1
%%exit2:
movzx ecx,byte[%2]
%%l2:
pop edx
mov [esi],dl
inc esi
loop %%l2
%endmacro
section .data ; data section
msg1: db "First number : " ;
len1: equ $-msg1 ;
msg2: db "Second number : " ;
len2: equ $-msg2 ;
msg3: db "Sum : " ;
len3: equ $-msg3 ;
ln: db 10
lnl: equ $-ln
var1: resb 10
var2: resb 10
str1: resb 10
str2: resb 10
ans: resb 10
ansvar: resb 10
ansl: db ''
l1: db ''
l2: db ''
section.text ;code
global _start
_start:
write msg1,len1
read str1,10
pack str1,l1,var1
write msg2,len2
read str2,10
pack str2,l2,var2
mov al,[var1]
add al,[var2]
mov [ansvar],al
unpack ans,ansl,ansvar
write msg3,len3
write ans,10
write ln,lnl
mov ebx,0 ; exit code, 0=normal
mov eax,1 ; exit command to kernel
int 0x80 ; interrupt 80 hex, call kernel
To assembler, link and run:
nasm -f elf add.asm
ld -s -o add add.o
./add

Related

Segmentation Fault while replacing a character x86 NASM on Linux

I'm getting a segmentation fault after writing mov [gtt+4], byte '>' where my variable is defined as gtt: db "I'm ! than 10".
I feel like it might be an error when assembling the program.
I'm using the commands: nasm -f elf file.asm -o file.o, ld -m elf_i386 file.o -o file
I have tried to use elf32 rather than elf but it doesn't make a difference.
The purpose of my program is to loop 2 times, multiplying whatever is in ebx by 2 each time. Afterwards, it will compare if ecx > 10 and displaying the corresponding string. I'm trying to get it to replace the "!" with either "<" or ">".
I'm using Ubuntu, and I'm working in x86 assembly.
Here is my full code snippet:
section .data:
starti: db "Starting value: 1", 0x0a
startil equ $-starti
gtt: db "I'm > than 10!", 0x0a
gttml equ $-gtt
section .text:
_start:
mov eax, 4
mov ebx, 1
mov ecx, starti
mov edx, startil
int 0x80
mov ebx, 1 ; Start value!
mov ecx, 2 ; Number of iterations
label:
add ebx, ebx ; Add 2*ebx
dec ecx ; ecx -= 1
cmp ecx, 0 ; If ecx > 0, loop again
jg label
cmp ebx, 10 ; If cbx > 10, jump to gtt-g
jg gttg
mov [gtt+4], byte '<'
mov eax, 4 ; If cbx < 10, output <
mov ebx, 1
mov ecx, gtt
mov edx, gttml
int 0x80
mov eax, 1 ; Exit gracefully
mov ebx, 0
int 0x80
gttg:
mov [gtt+4], byte '>'
mov eax, 4 ; Display gtt
mov ebx, 1
mov ecx, gtt
mov edx, gttml
int 0x80
mov eax, 1 ; Exit gracefully
mov ebx, 0
int 0x80
Fix
Thanks to Jester for this, "you must not use colons after section names, they are not labels. As it is, the colon is included in the name and hence the assembler doesn't recognize them as standard sections and doesn't apply the expected attributes meaning your data is read only. TL;DR: use section .data and section .text without a trailing".

Segmentation fault in my assembly program

I am getting a segmentation fault from this simple starting program.
I am using Ubuntu 16.10 and kdbg for debugging. Affter reaching starting __int 80h__, it stops moving to the next line.
section .bss ; section containing uninitialized data
BUFFLEN equ 1024 ; length of buffer
Buff: resb BUFFLEN ; text buffer itself
section .data ; section containing initialzed data
section .text ; secttion containing code
global _start ; linker needs to find the entry point!
_start:
nop ; this no-op keeps gdb happy
; read buffer full of text form stdin:
read:
mov eax, 3 ; specify sys_read call
mov ebx, 0 ; specify file descriptor 0 : standard input
mov ecx, Buff ; pass offset of the buffer to read to
mov edx, BUFFLEN ; pass number of bytes to be read at one pass
int 80h ; call sys_read to fill the buffer
mov esi,eax ; copy sys_read return value for safekeeping
cmp eax, 0 ; if eax = 0 , sys_read reached EOF on stdin
je Done ; jump if Equal ( to o, form compare)
; set up the register for the process buffer step:
mov ecx, esi ; place the number of bytes read into ecx
mov ebp, Buff ; pace address of buffer into ebp
dec ebp ; adjust the count to offset
; go through the buffer and cnvert lowercase to uppercase characters:
Scan:
cmp byte [ebp+ecx], 61h ; test input char agaisnst lowercase 'a'
jb Next ; if Below 'a' in ASCII, not lowercase
cmp byte [ebp+ecx], 7Ah ; test against lowercase 'z'
ja Next
sub byte [ebx+ecx], 20h ; subtract 20h to give uppercase..
Next:
dec ecx ; Decrement counter
jnz Scan ; if characters reamin, loop back
; Write the buffer full of processed text to stdout:
Write:
mov eax,4 ; Specify sys_write call
mov ebx, 1 ; Specify file descriptor 1 : stdout
mov ecx, Buff ; pass the offset of the buffer
mov edx, esi ; pass the # of bytes of data in the buffer
int 80h ; make sys_write kernel call
jmp read ; loop back and load another buffer full
Done:
mov eax, 1 ; Code for Exit sys_call
mov ebx, 0 ; return code of Zero
int 80h
I used these commands:
nasm -f elf -g -F stabs uppercaser1.asm
ld -m elf_i386 -o uppercaser1 uppercaser1.o
./uppercaser < inputflie
I think this code is generally public so I am posting with that in mind.
You should use the code as a guide to understanding what may be wrong in your code; however, it does not conform to any coding standard so it is pointless just to copy and paste and turn it in as an assignment; assuming that is the case.
You will never increase your assembly programming skills by merely doing a copy/paste.
lsb_release -a
...
Description: Ubuntu 16.04.3 LTS
nasm -f elf32 -g uppercase1.s -o uppercase1.o && ld -m elf_i386 uppercase1.o -o uppercase1
section .bss
Buff resb 1
section .data
section .text
global _start
_start:
nop
Read:
mov eax, 3 ; read syscall
mov ebx, 0 ; stdin
mov ecx, Buff ; pass address of the buffer to read to
mov edx, 1 ; read one char or one byte
int 0x80 ;
cmp eax, 0 ; if syscall returns returns 0
je Exit ;
cmp byte [Buff], 0x61 ; lower case a
jb Write ; jump if byte is below a in ASCII chart
cmp byte [Buff], 0x7a ; lower case z
ja Write ; jump if byte is above z in ASCII chart
sub byte [Buff], 0x20 ; changes the value in the buffer to an uppercase char
Write:
mov eax, 4 ; write syscall
mov ebx, 1 ; stdout
mov ecx, Buff ; what to print
mov edx, 1 ; length is one byte - each char is a byte
int 0x80
jmp Read ; go back to Read
Exit:
mov eax, 1
mov ebx, 0
int 0x80
Sample output:
david#ubuntuserver00A:~/asm$ ./uppercase1 < uppercase1.s
SECTION .BSS
BUFF RESB 1
SECTION .DATA
SECTION .TEXT
GLOBAL _START
_START:
NOP
READ:
MOV EAX, 3 ; READ SYSCALL
MOV EBX, 0 ; STDIN
MOV ECX, BUFF ; PASS ADDRESS OF THE BUFFER TO READ TO
MOV EDX, 1 ; READ ONE CHAR OR ONE BYTE
INT 0X80 ;
CMP EAX, 0 ; IF SYSCALL RETURNS RETURNS 0
JE EXIT ;
CMP BYTE [BUFF], 0X61 ; LOWER CASE A
JB WRITE ; JUMP IF BYTE IS BELOW A IN ASCII CHART
CMP BYTE [BUFF], 0X7A ; LOWER CASE Z
JA WRITE ; JUMP IF BYTE IS ABOVE Z IN ASCII CHART
SUB BYTE [BUFF], 0X20 ; CHANGES THE VALUE IN THE BUFFER TO AN UPPERCASE CHAR
WRITE:
MOV EAX, 4 ; WRITE SYSCALL
MOV EBX, 1 ; STDOUT
MOV ECX, BUFF ; WHAT TO PRINT
MOV EDX, 1 ; LENGTH IS ONE BYTE - EACH CHAR IS A BYTE
INT 0X80
JMP READ ; GO BACK TO READ
EXIT:
MOV EAX, 1
MOV EBX, 0
INT 0X80

Assembly won't print digits

I'm trying to learn assembly. I wanted to write a simple program that counted to 20 and printed out the numbers. I know you have to subtract ascii '0' from a ascii representation of a number to turn it into it's digit, but my implementation just refuses to work. I still get 123456789:;<=>?#ABCD
Here is my code.
section .bss
num resb 1
section .text
global _start
_start:
mov eax, '1'
mov ebx, 1 ; Filehandler 1 = stdout
mov ecx, 20 ; The number we're counting to
mov edx, 1 ; Size of a number in bytes
l1:
mov [num], eax ; Put eax into the value of num
mov eax, 4 ; Put 4 into eax (write)
push ecx ; Save ecx on the stack
mov ecx, num ; print num
int 0x80 ; Do the print
pop ecx ; Bring ecx back from the stack
mov eax, [num] ; Put the value of num into eax
sub eax, '0' ; Convert to digit
inc eax ; Increment eax
add eax, '0' ; Convert back to ascii
loop l1
mov eax,1 ; System call number (sys_exit)
int 0x80 ; Call kernel
Can anyone see what the problem is? I'm totally hitting a brick wall. I'm using nasm to compile and ld to link.

Binary representation in processor's registers in Nasm

I would like to ask about process of put instructions into registers. For example: we want to overwrite count '50' into EBX (in ASCII '50' is count '2').
EBX consists of 32 bits. When we put '50' into it, it will be arranged as binary represent, yes? (0000000 | 00000000 | 00000000 | 00110010). Have a right? What happens with bits, when we place a string into register?
EAX holds 32 bits which Intel calls "integer". The programmer - and sometimes the assembler - decides how to interpret these bits. If you load EAX with the number 50 (not the string '50')
mov eax, 50
the assembler decides to generate a machine instruction that loads the 50 in a manner, that you can read it as number 50 in a binary system:
00000000000000000000000000110010
Try out, what the assembler does if you feed it with a string:
GLOBAL _start
SECTION .bss
outstr resb 40
SECTION .data
_start:
mov eax, 'Four' ; Load EAX with a string
call int2bin ; Convert it to a binary string in outstr
mov byte [edi], 10 ; Add a line feed
inc edi ; Increment the pointer
mov eax, 4 ; SYS_WRITE
mov ebx, 1 ; STDOUT
mov ecx, outstr ; Pointer to output buffer
mov edx, edi ; Count of bytes to send:
sub edx, outstr ; EDX = EDI (offset returned from int2bin) - offset of output buffer
int 0x80 ; Call kernel
mov eax, 1 ; SYS_EXIT
xor ebx, ebx ; Returncode: 0 (ok)
int 0x80 ; Call kernel
int2bin: ; Converts an integer in EAX to a binary string in outstr
mov edi, outstr ; Pointer to a string
mov ecx, 32 ; Loop counter
.LL1:
test cl, 0b111 ; CL%8 = 0 ?
jnz .F ; No: skip the next instructions
mov Byte [edi], ' ' ; Store a space
inc edi ; and increment the pointer
.F:
shl eax, 1 ; The leftmost bit into carry flag
setc dl ; Carry flag into DL
or dl, '0' ; Convert it to ASCII
mov [edi], dl ; Store it to outstr
inc edi ; Increment the pointer
loop .LL1 ; Loop ECX times
mov byte [edi], 0 ; Null termination if needed as C string (not needed here)
ret
Output:
01110010 01110101 01101111 01000110
NASM stored it backwards in EAX. The ASCII of leftmost character is stored in the rightmost byte of EAX, the second-to-last character is to be found in the second byte, and so on. Better to see when those bytes are printed as ASCII characters:
GLOBAL _start
SECTION .bss
outstr resb 40
SECTION .data
_start:
mov eax, 'Four' ; Load EAX with a string
call int2str ; Convert it to a binary string in outstr
mov byte [edi], 10 ; Add a line feed
inc edi ; Increment the pointer
mov eax, 4 ; SYS_WRITE
mov ebx, 1 ; STDOUT
mov ecx, outstr ; Pointer to output buffer
mov edx, edi ; Count of bytes to send:
sub edx, outstr ; EDX = EDI (offset returned from int2bin) - offset of output buffer
int 0x80 ; Call kernel
mov eax, 1 ; SYS_EXIT
xor ebx, ebx ; Returncode: 0 (ok)
int 0x80 ; Call kernel
int2str: ; Converts an integer in EAX to an ASCII string in outstr
mov edi, outstr ; Pointer to a string
mov ecx, 4 ; Loop counter
.LL1:
rol eax, 8
mov [edi], al ; Store it to outstr
inc edi ; Increment the pointer
loop .LL1 ; Loop ECX times
mov byte [edi], 0 ; Null termination if needed as C string (not needed here)
ret
Output:
ruoF
Both programs above show EAX in big endian order. This is the order you are familiar with looking at decimal numbers. The most significant digit is left and the least significant digit is right. However, EAX would be saved in memory or disk in little endian order, starting the sequence from the right with the least significant byte. Looking at the memory with a disassembler or debugger you would see 'F','o','u','r' as well as you had defined it in a .data section with db 'Four'. Therefore you'll get no difference when you load a register with a string, save it to memory and call the write routine of the kernel:
GLOBAL _start
SECTION .bss
outstr resb 40
SECTION .data
_start:
mov eax, 'Hell' ; Load EAX with the first part of the string
mov ebx, 'o wo' ; Load EBX with the second part
mov ecx, 'rld!' ; Load ECX with the third part
mov dword [outstr], eax ; Store the first part in outstr (little endian)
mov dword [outstr+4], ebx ; Append the second part
mov dword [outstr+8], ecx ; Append the third part
mov eax, 4 ; SYS_WRITE
mov ebx, 1 ; STDOUT
mov ecx, outstr ; Pointer to output buffer
mov edx, (3*4) ; Count of bytes to send (3 DWORD à 4 bytes)
int 0x80 ; Call kernel
mov eax, 1 ; SYS_EXIT
xor ebx, ebx ; Returncode: 0 (ok)
int 0x80 ; Call kernel
Output:
Hello world!
Please note: This behavior is made by the NASM programmers. Other assemblers might have a different behavior.

How should I work with dynamically-sized input in NASM Assembly?

I'm trying to learn assembly with NASM on 64 bit Linux.
I managed to make a program that reads two numbers and adds them. The first thing I realized was that the program will only work with one-digit numbers (and results):
; Calculator
SECTION .data
msg1 db "Enter the first number: "
msg1len equ $-msg1
msg2 db "Enter the second number: "
msg2len equ $-msg2
msg3 db "The result is: "
msg3len equ $-msg3
SECTION .bss
num1 resb 1
num2 resb 1
result resb 1
SECTION .text
global main
main:
; Ask for the first number
mov EAX,4
mov EBX,1
mov ECX,msg1
mov EDX,msg1len
int 0x80
; Read the first number
mov EAX,3
mov EBX,1
mov ECX,num1
mov EDX,2
int 0x80
; Ask for the second number
mov EAX,4
mov EBX,1
mov ECX,msg2
mov EDX,msg2len
int 0x80
; Read the second number
mov EAX,3
mov EBX,1
mov ECX,num2
mov EDX,2
int 0x80
; Prepare to announce the result
mov EAX,4
mov EBX,1
mov ECX,msg3
mov EDX,msg3len
int 0x80
; Do the sum
; Store read values to EAX and EBX
mov EAX,[num1]
mov EBX,[num2]
; From ASCII to decimal
sub EAX,'0'
sub EBX,'0'
; Add
add EAX,EBX
; Convert back to EAX
add EAX,'0'
; Save the result back to the variable
mov [result],EAX
; Print result
mov EAX,4
mov EBX,1
mov ECX,result
mov EDX,1
int 0x80
As you can see, I reserve one byte for the first number, another for the second, and one more for the result. This isn't very flexible. I would like to make additions with numbers of any size.
How should I approach this?
First of all you are generating a 32-bit program, not a 64-bit program. This is no problem as Linux 64-bit can run 32-bit programs if they are either statically linked (this is the case for you) or the 32-bit shared libraries are installed.
Your program contains a real bug: You are reading and writing the "EAX" register from a 1-byte field in RAM:
mov EAX, [num1]
This will normally work on little-endian computers (x86). However if the byte you want to read is at the end of the last memory page of your program you'll get a bus error.
Even more critical is the write command:
mov [result], EAX
This command will overwrite 3 bytes of memory following the "result" variable. If you extend your program by additional bytes:
num1 resb 1
num2 resb 1
result resb 1
newVariable1 resb 1
You'll overwrite these variables! To correct your program you must use the AL (and BL) register instead of the complete EAX register:
mov AL, [num1]
mov BL, [num2]
...
mov [result], AL
Another finding in your program is: You are reading from file handle #1. This is the standard output. Your program should read from file handle #0 (standard input):
mov EAX, 3 ; read
mov EBX, 0 ; standard input
...
int 0x80
But now the answer to the actual question:
The C library functions (e.g. fgets()) use buffered input. Doing it like this would be a bit to complicated for the beginning so reading one byte at a time could be a possibility.
Thinking the way "how would I solve this problem using a high-level language like C". If you don't use libraries in your assembler program you can only use system calls (section 2 man pages) as functions (e.g. you cannot use "fgets()" but only "read()").
In your case a C program reading a number from standard input could look like this:
int num1;
char c;
...
num1 = 0;
while(1)
{
if(read(0,&c,1)!=1) break;
if(c=='\r' || c=='\n') break;
num1 = 10*num1 + c - '0';
}
Now you may think about the assembler code (I typically use GNU assembler, which has another syntax, so maybe this code contains some bugs):
c resb 1
num1 resb 4
...
; Set "num1" to 0
mov EAX, 0
mov [num1], EAX
; Here our while-loop starts
next_digit:
; Read one character
mov EAX, 3
mov EBX, 0
mov ECX, c
mov EDX, 1
int 0x80
; Check for the end-of-input
cmp EAX, 1
jnz end_of_loop
; This will cause EBX to be 0.
; When modifying the BL register the
; low 8 bits of EBX are modified.
; The high 24 bits remain 0.
; So clearing the EBX register before
; reading an 8-bit number into BL is
; a method for converting an 8-bit
; number to a 32-bit number!
xor EBX, EBX
; Load the character read into BL
; Check for "\r" or "\n" as input
mov BL, [c]
cmp BL, 10
jz end_of_loop
cmp BL, 13
jz end_of_loop
; read "num1" into EAX
mov EAX, [num1]
; Multiply "num1" with 10
mov ECX, 10
mul ECX
; Add one digit
sub EBX, '0'
add EAX, EBX
; write "num1" back
mov [num1], EAX
; Do the while loop again
jmp next_digit
; The end of the loop...
end_of_loop:
; Done
Writing decimal numbers with more digits is more difficult!

Resources