COE 205 Lab 2: Introduction to Assembly Language Programming (PDF)

COE 205 Lab Manual Lab 2: Introduction to Assembly Language Programming - page 16 Lab 2: Introduction to Assembly Language Programming Contents 2.1. Intel IA-32 Processor Architecture 2.2. Basic Program Execution Registers 2.3. FLAT Memory Model and Protected-Addres...

COE 205 Lab Manual Lab 2: Introduction to Assembly Language Programming - page 16 Lab 2: Introduction to Assembly Language Programming Contents 2.1. Intel IA-32 Processor Architecture 2.2. Basic Program Execution Registers 2.3. FLAT Memory Model and Protected-Address Mode 2.4. FLAT Memory Program Template 2.5. Writing a Program using the FLAT Memory Model 2.6. Editing, Assembling, Linking, and Debugging Programs 2.1 Intel IA-32 Processor Architecture Intel introduced the 8086 processor in 1979. It has a 20-bit address bus and a 16-bit data bus. The maximum physical memory that this processor can access is 220 bytes or 1MB. This processor uses segmentation to address memory. The maximum size of a segment is restricted to 216 bytes or 64 KB, since all registers were 16 bits at that time. With the advent of 32-bit processors, starting with the 80386 in 1985, and continuing up with the latest Pentium family of processors, Intel has introduced a 32-bit architecture known as IA-32. This family of processors allows the use of 32-bit addresses that can address up to 4 Gigabytes of memory. These 32-bit processors remove the limitations of the earlier 16-bit 8086 processor. 2.2 Basic Program Execution Registers There are eight 32-bit general-purpose registers (EAX, EBX, ECX, EDX, ESI, EDI, EBP, ESP), a 32-bit register that holds the processor flags (EFLAGS), and a 32-bit instruction pointer (EIP). These registers are shown below. EAX ESI EBX EDI ECX EBP EDX ESP EFLAGS EIP The general-purpose registers are primarily used for arithmetic and data movement. Each register can be addressed as either a single 32-bit value or a 16-bit value. Some 16-bit registers (AX, BX, CX, and DX) can be also addressed as two separate 8-bit values. For example, AX can be addressed as AH and AL, as shown below. Prepared by Dr. Muhamed Mudawar © KFUPM - Revised: August 2006 COE 205 Lab Manual Lab 2: Introduction to Assembly Language Programming - page 17 32-bit (IA-32) Registers 32-bit (IA-32) Registers 8-bit high 8-bit low 16-bit (8086) AX EAX AH AL ESI SI BX EDI DI EBX BH BL EBP BP CX ESP SP ECX CH CL EIP IP DX EDX DH DL EFLAGS FLAGS Some general-purpose registers have specialized uses. EAX is called the extended accumulator and is used by multiplication and division instructions. ECX is the extended counter register and is used as a loop counter. ESP is the extended stack pointer and is used to point to the top of the stack. EBP is the extended base pointer and is used to access function parameters and local variables on the stack. ESI is called the extended source index and EDI is called the extended destination index. They are used by string instructions. We will learn more about these instructions later during the semester. The EIP register is called the extended instruction pointer and contains the address of the next instruction to be executed. EFLAGS is the extended flags register that consists of individual bits that either control the operation of the CPU or reflect the outcome of some CPU operations. We will learn more about these flags later. 2.3 FLAT Memory Model and Protected-Address Mode The FLAT memory model is used in 32-bit operating systems (like Windows XP) running on a 32-bit processor. Each program addresses up to 4GB of memory. All code and data go into a single 32-bit (4-GigaByte) address space. Linear 32-bit addresses are used in each program to access the program instructions and data in memory. 32-bit address EBP STACK ESP Flat address Unused 32-bit address space of a ESI program EDI DATA (up to 4 GB) 32-bit address EIP CODE Prepared by Dr. Muhamed Mudawar © KFUPM - Revised: August 2006 COE 205 Lab Manual Lab 2: Introduction to Assembly Language Programming - page 18 Rather than using real memory addresses, a program uses virtual addresses. These virtual addresses are mapped onto real (physical) addresses by the operating system through a scheme called paging. The processor translates virtual addresses into real addresses as the program is running. With virtual memory, the processor runs in protected mode. This means that each program can access only the memory that was assigned to it by the operating system and cannot access the memory of other programs. 2.4 FLAT Memory Model Program Template Writing an assembly language program is a complicated task, particularly for a beginner. We will simplify this task by hiding those details that are irrelevant. We will use the following template for writing FLAT memory programs. This template consists of three types of statements: executable instructions, assembler directives, and macros. Executable instructions generate machine code for the processor to execute at runtime. Assembler directives provide information to the assembler while translating the program. Macros are shorthand for a sequence of instructions, directives, or even other macros. We will learn more about instructions, directives, and macros throughout the semester. TITLE FLAT Memory Program Template (template.asm) ; Program Description: ; Author: ; Date Created: ; Last Modified:.686.MODEL FLAT, STDCALL.STACK INCLUDE Irvine32.inc ; (insert symbol definitions here).DATA ; (insert variables here).CODE main PROC ; (insert executable instructions here) exit ; exit to operating system main ENDP ; (insert additional procedures here) END main The first line of an assembly language program is the TITLE line. This line is optional. It contains a brief heading of the program and the disk file name. The next few lines are line comments. They begin with a semicolon (;) and terminate with the end of the line. They are ignored and not processed by the assembler. However, they are used to document a program and are of prime importance to the assembly language programmer, because assembly language code is not easy to read or understand. Insert comments at the beginning of a program to describe the program, its author, the date when it was first written and the date when it was last modified. You need also comments to document your data and your code. Prepared by Dr. Muhamed Mudawar © KFUPM - Revised: August 2006 COE 205 Lab Manual Lab 2: Introduction to Assembly Language Programming - page 19 The.MODEL is a directive that specifies the memory configuration for the assembly language program. For our purposes, the FLAT memory model will be used. The.686 is a processor directive used before the.MODEL FLAT directive to provide access to the 32-bit instructions and registers available in the Pentium Processor. The STDCALL directive tells the assembler to use standard conventions for names and procedure calls. The.STACK is a directive that tells the assembler to define a stack for the program. The size of the stack can be optionally specified by this directive. The stack is required for procedure calls. We will learn more about the stack and procedures later during the semester. The.DATA is a directive that defines an area in memory for the program data. The program's variables should be defined under this directive. The assembler will allocate storage for these variables and initialize their locations in memory. The.CODE is a directive defines the code section of a program. The code is a sequence of assembly language instructions. These instructions are translated by the assembler into machine code and placed in the code area in memory. The INCLUDE directive causes the assembler to include code from another file. We will include Irvine32.inc that specifies basic input and output procedures provided by the book author Kip Irvine, and that can be used to simplify programming. These procedures are defined in the Irvine32.lib library that we will link to the programs that we will write. Under the code segment, we can define any number of procedures. As a convention, we will define the first procedure to be the main procedure. This procedure is defined using the PROC and ENDP directives: main PROC... main ENDP The exit at the end of the main procedure is used to terminate the execution of the program and exit to the operating system. Note that exit is a macro. It is defined in Irvine32.inc and provides a simple way to terminate a program. The END is a directive that marks the last line of the program. It also identifies the name (main) of the program’s startup procedure, where program execution begins. 2.5 Writing a Program using the FLAT Memory Model The following program adds and subtracts integers. You may open it using any text editor. TITLE Add and Subtract (addsub.asm) ; This program adds and subtracts integers.686.MODEL flat, stdcall.STACK INCLUDE Irvine32.inc.code main PROC mov eax, 60000h ; EAX = 60000h add eax, 80000h ; EAX = EAX + 80000h sub eax, 20000h ; EAX = EAX - 20000h exit main ENDP END main Prepared by Dr. Muhamed Mudawar © KFUPM - Revised: August 2006 COE 205 Lab Manual Lab 2: Introduction to Assembly Language Programming - page 20 2.5.1 Lab Work: Guessing the Value of the EAX Register The constant values 60000h, 80000h, and 20000h are in hexadecimal. Guess and write the values of the EAX register in the above program after the add and sub (subtract) instructions: Value of EAX in hexadecimal after add = Value of EAX in hexadecimal after sub = 2.5.2 Instructions The above program uses 3 instructions. The mov instruction is used for moving data. The add instruction is used for adding two value, and the sub instruction is used for subtraction. An instruction is a statement executed by the processor at runtime after the program has been loaded into memory and started. An instruction contains four basic parts: Label: Instruction Operand(s) ; Comment (optional) Mnemonic A label is an identifier that acts as a place marker for an instruction. It must end with a colon (:) character. The assembler assigns an address to each instruction. A label placed just before an instruction implies the instruction address. Labels are often used as targets of jump instructions. An instruction mnemonic is a short word that identifies the operation carried out by the instruction when it is executed at runtime. Instruction mnemonics have useful names, such as mov, add, sub, jmp (jumping to a target instruction), and call (calling a procedure). An instruction can have between zero and three operands, each of which can be a register, memory operand, constant expression, or an I/O port. In the above program (addsub.asm), the mov, add, and sub instructions used two operands. The first operand specified the destination, which was the eax register. The second operand was a constant integer value. An instruction can be terminated with an optional comment. The comment starts with a semicolon (;) and terminates with the end of line. A comment at the end of an instruction can be used as a short description and clarification for the use of that instruction. 2.6 Editing, Assembling, Linking, and Debugging Programs The process of editing, assembling, linking, and debugging programs is shown below. You will learn a lot about this cycle during this semester. The editor is used to write new programs and to make changes to existing ones. Once a program is written, it can be assembled and linked using the ML.exe program. Alternatively, it can be assembled only using the ML.exe program and linked using the LINK32.exe program. The assembler translates the source assembly (.asm) file into an equivalent object (.obj) file in machine language. As an option, the assembler can also produce a listing (.lst) file. This file shows the work of the assembler and contains the assembly source code, offset addresses, translated machine code, and a symbol table. The linker combines one or more object (.obj) files produced by the assembler with one or more link library (.lib) files to produce the executable program (.exe) file. In addition, the linker can produce an optional (.map) file. A map file contains information about the program being linked. It contains a list of segment groups in the program, a list of public symbols, and the address of the program's entry point. Prepared by Dr. Muhamed Mudawar © KFUPM - Revised: August 2006 COE 205 Lab Manual Lab 2: Introduction to Assembly Language Programming - page 21 Edit prog.asm Assemble library.lib prog.obj prog.lst Link prog.exe prog.map Debug Run Once the executable program is generated, it can be executed and/or debugged. A debugger allows you to trace the execution of a program and examine and/or modify the content of registers and memory. With a debugger, you will be able to discover your errors in the program. Once these errors are discovered, you will make the necessary changes in the source assembly program. You will go back to the editor to make these changes, assemble, link, and debug the program again. This cycle repeats until the program starts functioning properly, and correct results are produced. 2.6.1 Lab Work: Using ML, LINK32, and MAKE32 Commands We will use the Command Prompt to assemble and link a 32-bit program. Type the following commands after changing the directory to one containing the addsub.asm program. Make sure the environment variables are set properly (as explained in Lab 1). Prepared by Dr. Muhamed Mudawar © KFUPM - Revised: August 2006 COE 205 Lab Manual Lab 2: Introduction to Assembly Language Programming - page 22 ml –c –Zi –Fl –coff addsub.asm link32 addsub.obj kernel32.lib /subsystem:console /debug /map The ml command assembles addsub.asm. The –c option is used to assemble only without linking. This will generate addsub.obj object file. The –Zi option will add debugging information and the –Fl option will generate the listing file addsub.lst. You may examine the addsub.lst file to understand how the assembler has translated the source assembly language file into an object file. The –coff option tells the assembler to generate a COFF object file (Common Object File Format) used for 32-bit programs. The link32 command is a 32-bit linker used to link addsub.obj file to the kernel32.lib library. This generates addsub.exe, a 32-bit executable file. The /subsystem:console option specifies the console as being the subsystem. The /debug option tells the linker to generate debugging information in the executable file, and the /map options generates a.map file. To save typing, we can write a batch file to assemble and link a 32-bit program. This batch file is already written for you and exits under the MASM installation directory and is called make32.bat. Type make32 addsub to assemble and link the addsub.asm program. 2.6.2 Lab Work: Using the Windows Debugger At the Command Prompt, type: windbg –QY –G addsub.exe to run the Windows Debugger. Make sure that the installation directory for windbg.exe, which is typically C:\Program Files\Debugging Tools for Windows\, exists in the path variable. Prepared by Dr. Muhamed Mudawar © KFUPM - Revised: August 2006 COE 205 Lab Manual Lab 2: Introduction to Assembly Language Programming - page 23 Open the source file addsub.asm from the File menu if it is not already opened. Watch the registers by selecting Registers in the View menu or by pressing Alt+4. You can make the Registers window floating or you can dock it as shown above. You can customize the order of registers. Click on the Customize… button and type eax at the beginning to show the eax register on top of the list. You can customize the rest as shown. Place the cursor at the beginning of the main procedure and press F7 to start the execution of the main procedure as shown above. Press F10 to step through the execution of the main procedure. Observe the changes to the EAX register after executing each instruction. Make the necessary corrections to the values of EAX that you have guessed in Section 2.5.1. 2.6.3 Lab Work: Understanding Program Termination The exit at the end of the main procedure is a macro. It is expanded into a system call to the ExitProcess MS-Windows function that terminates the program. This function is defined in the kernel32 library. To have a better understanding of program termination, remove the exit at the end of the main procedure and replace it with the following two instructions: exit ⇒ push 0 call ExitProcess The push 0 pushes the number 0 on the stack and the call instruction is used to call the function ExitProcess. You can also replace exit with INVOKE ExitProcess, 0. The assembler will translate into: push 0 and call ExitProcess as shown above. There is no need to include the Irvine32.inc file, since the exit macro is no longer used. However, there is a need to declare that ExitProcess is an external function defined outside the addsub.asm program. We use the PROTO directive for this purpose. Therefore, remove the INCLUDE directive and replace it with the PROTO directive as shown below: INCLUDE Irvine32.inc ⇒ ExitProcess PROTO, ExitCode:DWORD The PROTO directive declares functions used by a program and defined outside the program file. This directive also specifies the parameters and types of a given function. Now that you have made these changes, assemble and link the program using the ML and LINK32 programs as explained in Section 2.6.1. Notice that addsub.obj is linked to the kernel32.lib because the ExitProcess function is defined in the kernel32 library. Open the debugger and trace the execution of the program. Prepared by Dr. Muhamed Mudawar © KFUPM - Revised: August 2006 COE 205 Lab Manual Lab 2: Introduction to Assembly Language Programming - page 24 Review Questions 1. Name all eight 32-bit general-purpose registers. 2. Name all eight 16-bit general-purpose registers. 3. Name all eight 8-bit general-purpose registers. 4. What special purpose does the EAX register serve? 5. What is the purpose of the EIP register? 6. What is the purpose of the ESP register? 7. In the FLAT memory model, how many bits are used to hold a memory address? 8. What is the meaning of the INCLUDE directive? 9. What does the.CODE directive identify? 10. Which directive begins a procedure and which directive ends it? 11. What is the purpose of the END directive? 12. What does the PROTO directive do? 13. What types of files are produced by the assembler? 14. What types of files are produced by the linker? Programming Exercises 1. Using the addsub program as a reference, write a program that moves four integers into the EAX, EBX, ECX, and EDX registers and then accumulates their sum into the EAX register. Trace the execution of the program and view the registers using the windows debugger. 2. Rewrite the above program using the 16-bit registers: AX, BX, CX, and DX. Prepared by Dr. Muhamed Mudawar © KFUPM - Revised: August 2006