COE 205 Lab 3: Defining Data and Symbolic Constants PDF
Document Details
Uploaded by RightfulVibraphone
Dr. Muhamed Mudawar
Tags
Related
- Assembly Language Programming ACSL PDF
- Lecture 02 Introduction to Assembly Language PDF
- Lecture 2 Assembly Language Programming 2024-2025 PDF
- CSC 204 - Assembly Language - Chapter 04 - Fall 2024-2025 PDF
- CSC 204 Assembly Language - Chapter 02 - Fall 2024 - PDF
- CPE 409 Assembly Language Programming PDF
Summary
This document is a lab manual, containing information on data types used in x86 assembly language programming, specifically MASM data types, showing examples, and how to use the Windows Debugger. It is intended to be used for educational purposes.
Full Transcript
COE 205 Lab Manual Lab 3: Defining Data and Symbolic Constants - page 25 Lab 3: Defining Data and Symbolic Constants Contents 3.1. MASM Data Types 3.2. Defining Integer Data 3.3. Watching Variables using the Windows Debugger 3.4. Multiple Initialize...
COE 205 Lab Manual Lab 3: Defining Data and Symbolic Constants - page 25 Lab 3: Defining Data and Symbolic Constants Contents 3.1. MASM Data Types 3.2. Defining Integer Data 3.3. Watching Variables using the Windows Debugger 3.4. Multiple Initializers, Defining Strings, and the DUP operator 3.5. Watching Memory using the Windows Debugger 3.6. Data-Related Operators 3.7. Symbolic Constants and the EQU and = directives 3.8. Viewing Symbolic Constants in the Listing (.lst) File 3.1 MASM Data Types MASM defines various intrinsic data types, each of which describes a set of values that can be assigned to variables and expressions of the given type. The following table lists these data types. The first 9 data types are used to define integer data, while the last 3 are used to define real data according to the IEEE standard real number formats. Type Usage BYTE 8-bit unsigned integer (range 0 to 255) SBYTE 8-bit signed integer (range –128 to +127) WORD 16-bit unsigned integer (range 0 to 65535 = 216 – 1) SWORD 16-bit signed integer (range –32768 to +32767 = 215 – 1) DWORD 32-bit unsigned integer (Double Word: range 0 to 4,294,967,295 = 232 – 1) SDWORD 32-bit signed integer (range –2,147,483,648 to +2,147,483,647 = 231 – 1) FWORD 48-bit integer (FAR pointer in protected mode) QWORD 64-bit integer (Quad Word) TBYTE 80-bit (Ten Byte) integer REAL4 32-bit (4-byte) IEEE floating-point number REAL8 64-bit (8-byte) IEEE double-precision floating-point number REAL10 80-bit (10-byte) IEEE extended-precision floating-point number Data definition has the following syntax: [name] directive initializer [,initializer]... For more information, refer to your textbook or class notes. Prepared by Dr. Muhamed Mudawar © KFUPM – Revised August 2006 COE 205 Lab Manual Lab 3: Defining Data and Symbolic Constants - page 26 3.2 Lab Work: Defining Integer Data The following program demonstrates integer data definition under the.DATA section. You may open and view this program in ConTEXT or any other text editor. Assemble and link this program to produce the IntegerDef.exe executable file. TITLE Integer Data Definitions (File:IntegerDef.asm) ; Examples Demonstrating Integer Data Definition.686.MODEL flat, stdcall.STACK INCLUDE Irvine32.inc.data ; ----------------- Byte Values --------------------- byte1 BYTE 'A' ; 'A' = 65 = 41h byte2 BYTE 0 ; smallest unsigned byte value byte3 BYTE 255 ; largest unsigned byte value byte4 SBYTE -128 ; smallest signed byte value byte5 SBYTE +127 ; largest signed byte value ; byte6 BYTE ? ; uninitialized ; ----------------- Word Values --------------------- word1 WORD 65535 ; largest unsigned word value word2 SWORD -32768 ; smallest signed word value word3 WORD ? ; uninitialized ; --------------- DoubleWord Values ----------------- dword1 DWORD 0FFFFFFFFh ; largest unsigned value in hex dword2 SDWORD -2147483648 ; smallest signed value in decimal ; --------------- QuadWord Value -------------------- quad1 QWORD 0123456789ABCDEFh.code main PROC ; No instructions to execute exit main ENDP END main 3.3 Lab Work: Watching Variables using the Windows Debugger Now run the Windows debugger to watch the variables and the memory content. You may run the debugger from ConTEXT Tools menu (if it is properly configured) or from the command prompt by typing: windbg –QY –G IntegerDef.exe 3.3.1 Lab Work: Watch Window Open the Watch window (from the View menu). Insert the variable byte1 and byte4 under the Name column as shown below. To add a variable to the Watch list, click in the first empty cell in the Name column, enter the name of this variable, and press ENTER. You can watch variables in any order. It does not have to be the same order declared in the source program. Prepared by Dr. Muhamed Mudawar © KFUPM – Revised August 2006 COE 205 Lab Manual Lab 3: Defining Data and Symbolic Constants - page 27 Observe the Value of these variables under the Value column. You can also view the type and memory addresses of these variables by pressing on the Typecast and Locations buttons. Observe that all the memory location addresses are in hexadecimal. The value of variables can be in hexadecimal or decimal depending on whether the type is unsigned or not. The windows debugger uses different type names other than the ones used in MASM. The type BYTE becomes unsigned char and SBYTE becomes char. The type WORD becomes unsigned short and SWORD becomes short. The type DWORD becomes unsigned long, SDWORD becomes long, and the type QWORD becomes int64. Now, let us change the Type of byte1 from unsigned char to char. This can be done easily by clicking on the type and editing its value. Observe that the value of byte1 now appears in decimal (65) after it appeared in hexadecimal (0x41). Similarly, let us change the type of byte4 from char to unsigned char as shown below. Observe that changing the type of a variable to unsigned (or vice versa) does NOT change the value of the variable. The value is still the same and is stored in binary in the memory of the computer. It is only how we view the same value as being either in signed decimal or hexadecimal. To delete a Name from the watch list, select that name and press the delete button to delete the name. The line will disappear from the watch list as soon as click outside. Now insert all the variable names in the watch list and observe their values and locations. Fill in the following table showing the Location (address) of these variables in the data segment in memory, as well as their values in hex and in decimal. Observe that the variables occupy locations at successive addresses in memory, according to their order of appearance in the program. Name Location (hex) Value (hex) Value (decimal) byte1 00404000 'A' = 41h 65 byte2 byte3 byte4 byte5 byte6 word1 Prepared by Dr. Muhamed Mudawar © KFUPM – Revised August 2006 COE 205 Lab Manual Lab 3: Defining Data and Symbolic Constants - page 28 word2 word3 dword1 dword2 quad1 What is the total number of bytes allocated for data?..................................................... 3.4 Lab Work: Multiple Initializers, Defining Strings, and the DUP Operator You can create arrays of bytes, words, double words, etc., either by explicitly using multiple initializers or by using the DUP (Duplicate) operator. Multiple initializers are separated by commas and are used to initialize each element of the array with an explicit number. The DUP operator generates a repeated storage allocation, using a constant expression as a counter. The initializers and the DUP operator can be combined together and can be nested. You can also create string data definition by enclosing a sequence of characters in quotation marks (either single quotes or double quotes can be used). Strings are commonly terminated with a null character, a byte containing the value 0. We will follow the convention of terminating all strings with a null char. TITLE Multiple Initializers (MultipleInitializers.asm) ; Examples showing multiple initializers and the DUP operator.686.MODEL flat, stdcall.STACK INCLUDE Irvine32.inc ; ----------------- Byte Values ----------------.data list1 BYTE 10, 32, 41h, 00100010b list2 BYTE 0Ah, 20h, 'A', 22h array1 BYTE 8 DUP(0) ; 8 bytes initialized to 0 greeting BYTE "Good afternoon",0 ; ----------------- Word Values --------------------- myList WORD 1,2,3,4 ; array of words ; --------------- DoubleWord Values -------------- array2 DWORD 4 DUP(01234567h).code main PROC ; No instructions to execute exit main ENDP END main Prepared by Dr. Muhamed Mudawar © KFUPM – Revised August 2006 COE 205 Lab Manual Lab 3: Defining Data and Symbolic Constants - page 29 Now open MultipleInitializers.asm and assemble and link the file. You can use the make32 batch file from the command prompt or from the ConTEXT editor’s Tools menu. 3.5 Lab Work: Watching Memory using the Windows Debugger Now run the Windows debugger to view the memory content. You may open the debugger from the ConTEXT editor’s Tools menu (if it is properly configured) or from the command prompt by typing: windbg –QY –G MultipleInitializers.exe 3.5.1 Lab Work: Memory Window From the View menu, select Memory to open the Memory Window. This window will allow you to watch the content of memory. Under Virtual enter list1, the name of the first variable. The virtual address of list1 is 00404000 (in hexadecimal) as shown in the first column of the Memory Window. We call this address a virtual address because it is not a real address. The Windows Operating System maps virtual addresses onto real addresses, but we are not concerned here about the details. In the Memory window shown below, the Display format is shown as Byte, and the Byte values in memory appear in hexadecimal. You may resize the Memory window so that exactly 16 bytes are displayed on each line. The virtual address goes by increments of 10h = 16 bytes. It is also possible to write the virtual address of the variable in the Virtual address box. For example, we could have written 00404000 in the virtual address box, but it is easier to refer to memory by name, rather than by address. The addresses are still listed in the first column of the Memory Window in hexadecimal. Observe also that printable characters are displayed on the right. For example, the byte at address 00404010 is 47h = G, and the byte at address 00404013 is 64h = d. You can view memory starting at any address or at any variable. For example, if you want to view myList then type the variable name in the Virtual address box. Now answer the following: What is the virtual address (in hexadecimal) of myList?.................................................... What is the virtual address (in hexadecimal) of array2?.................................................... How many bytes are allocated for myList?......................................................................... How many bytes are allocated for array2?......................................................................... What is the byte value (in hex) at virtual address 00404018?............................................ What is the byte value (in hex) at virtual address 00404032?............................................ Prepared by Dr. Muhamed Mudawar © KFUPM – Revised August 2006 COE 205 Lab Manual Lab 3: Defining Data and Symbolic Constants - page 30 3.5.2 Little Endian Order The variable myList is an array of words and each word occupies 2 bytes of memory. The first element of myList has the value 1 and occupies 2 bytes of memory. The least significant byte has value 01h and the most significant byte has value 00h. The least significant byte is stored at address 00404020h (low byte address) and the most significant byte is stored at address 00404021h (high byte address) as shown in the above Memory window. Similarly, observe the byte order of the elements of array2. Each element of array2 is a double word and occupies 4 bytes of memory. Each element has a value 01234567h, where the least significant byte 67h is stored at the first byte address, while the most significant byte 01h is stored at the last byte address. This byte ordering, from least significant byte to most significant byte, is called Little Endian order. This byte ordering is used by the Intel processors to store values that occupy more than one byte. 3.5.3 Lab Work: Changing the Display Format The Byte display format is not convenient to view arrays of words or double words. Let us now change the Display format to Short Hex to have a WORD view of myList in hexadecimal, as shown below. The Short Unsigned display format can be used to view an unsigned WORD array in decimal, rather than in hex. The Short display format can be used to view the elements of an SWORD (Signed WORD) array in decimal. For array2, which is an array of double words, use the Long Hex display format to display the values. Observe that each element of array2 occupies 4 bytes of memory. We can also use the Long and Long Unsigned display formats to display the numbers as signed or unsigned decimal, rather than hex. Using the Long Hex format for array2, write the address and the four double word values of array2, as they appear in the Memory window: Other display formats are also available and can be explored. The Quad display formats can be used to display QWORD (8-byte) numbers either signed or unsigned, in decimal or in hex. The Real display formats can be used to display floating-point numbers of different sizes. Prepared by Dr. Muhamed Mudawar © KFUPM – Revised August 2006 COE 205 Lab Manual Lab 3: Defining Data and Symbolic Constants - page 31 3.6 Lab Work: Data Related Operators In the following program, you will learn about TYPE, LENGTHOF, SIZEOF, OFFSET, and PTR operators. These operators are not instructions and are not executed by the processor. Instead, they are only processed by the assembler during assembly time. TYPE Size (in bytes) of each element in an array. LENGTHOF Number of elements in an array. SIZEOF Number of bytes used by an array initializer. OFFSET Virtual address of a variable. PTR Used to override a variable's default size. For more details about these operators, refer to the lecture notes or your textbook. Open Operators.asm using ConTEXT or any other text editor and assemble and link the file. TITLE Operators (File: Operators.asm) ; Demonstration of TYPE, LENGTHOF, SIZEOF, OFFSET, and PTR operators.686.MODEL flat, stdcall.STACK INCLUDE Irvine32.inc.data byte1 BYTE 10,20,30,40 array1 WORD 30 DUP(?),0,0 array2 WORD 5 DUP(3 DUP(?)) array3 DWORD 01234567h,2,3,4 digitStr BYTE '12345678',0 myArray BYTE 10h,20h,30h,40h,50h,60h,70h,80h,90h.code main PROC ; Demonstrating TYPE operator mov al, TYPE byte1 mov bl, TYPE array1 mov cl, TYPE array3 mov dl, TYPE digitStr ; Demonstrating LENGTHOF operator mov eax, LENGTHOF array1 mov ebx, LENGTHOF array2 mov ecx, LENGTHOF array3 mov edx, LENGTHOF digitStr ; Demonstrating SIZEOF operator mov eax, SIZEOF array1 mov ebx, SIZEOF array2 mov ecx, SIZEOF array3 mov edx, SIZEOF digitStr ; Demonstrating OFFSET operator mov eax, OFFSET byte1 mov ebx, OFFSET array1 mov ecx, OFFSET array2 mov edx, OFFSET array3 mov esi, OFFSET digitStr mov edi, OFFSET myArray Prepared by Dr. Muhamed Mudawar © KFUPM – Revised August 2006 COE 205 Lab Manual Lab 3: Defining Data and Symbolic Constants - page 32 ; Demonstrating PTR operator mov al, BYTE PTR array3 mov bx, WORD PTR array3 mov cx, WORD PTR myArray mov edx, DWORD PTR myArray exit main ENDP END main 3.6.1 Lab Work: Tracing Program Execution and Watching Registers Run the Windows debugger from the ConTEXT editor, or from the command prompt by typing: windbg –QY –G Operators.exe. From the View menu, select Registers to open the Registers Window. Customize the registers to appear in the order that you want. Press the Customize button and enter: al bl cl dl ax bx cx dx eax ebx ecx edx esi edi We only care to view these registers. The rest is not important. The list can be customized differently for different programs. You can also make this register window always floating or you may dock it, by selecting the option in the upper drop-down menu. Now place the cursor at the beginning of the main procedure and press F7 to start debugging the main procedure. You may also place the cursor at any instruction and press F7 to run the program until that instruction. This is useful if you don't want to view the execution of each instruction separately. Now press F10 to watch the execution of each instruction separately in the main procedure, and to view the changes in the registers at each step. These value changes appear in red. The next instruction in the source window is highlighted. This instruction is not executed yet, but will be executed the next time that you press F10. Try first to guess and understand the values of the registers in the above program. Write these values in decimal for the TYPE, LENGTHOF, and SIZEOF operators, and in hexadecimal for the OFFSET and PTR operators. By default, all register values appear in hexadecimal. You will have to convert them to decimal for the LENGTHOF and SIZEOF operators. Prepared by Dr. Muhamed Mudawar © KFUPM – Revised August 2006 COE 205 Lab Manual Lab 3: Defining Data and Symbolic Constants - page 33 TYPE Operator al = bl = cl = dl = LENGTHOF Operator eax (decimal) = ecx (decimal) = ebx (decimal) = edx (decimal) = SIZEOF Operator eax (decimal) = ecx (decimal) = ebx (decimal) = edx (decimal) = OFFSET Operator eax (hex) = edx (hex) = ebx (hex) = esi (hex) = ecx (hex) = edi (hex) = PTR Operator al (hex) = cx (hex) = bx (hex) = edx (hex) = 3.7 Lab Work: Symbolic Constants and the EQU and = directives A symbolic constant is created by associating an identifier (a symbol) with either an integer expression or some text. Unlike a variable definition, which reserves storage, a symbolic constant does not use any storage. The value of a symbolic constant is defined by the assembler and does not change at run time. The EQU (Equal) directive associates a symbolic name with either an integer expression or some arbitrary text, according to the following formats: name EQU expression name EQU A symbol defined with EQU cannot be redefined in the same source code file. This prevents an existing symbol from being inadvertently assigned a new value. Unlike the EQU directive, the = directive can redefine a symbol any number of times. The = directive can associate a symbolic name with an integer expression only according to the following format. However, the = directive cannot associate a symbol with text. name = expression Prepared by Dr. Muhamed Mudawar © KFUPM – Revised August 2006 COE 205 Lab Manual Lab 3: Defining Data and Symbolic Constants - page 34 The following program illustrates the definition of symbolic constants: TITLE Symbolic Constants (File: Constants.asm) ; Demonstration of EQU and = directives.686.MODEL flat, stdcall.STACK INCLUDE Irvine32.inc.data Rows EQU 3 Cols EQU 3 Elements EQU Rows * Cols CR EQU 10 LF EQU 13 PromptText EQU matrix WORD Elements DUP(0) prompt BYTE PromptText COUNT = 10h COUNT = 100h COUNT = 1000h COUNT = SIZEOF matrix.code main PROC exit main ENDP END main Fill the table below, listing only the symbolic constants in the above Program and their values in hexadecimal. If a symbolic constant is redefined then it should be listed multiple times. Symbolic Constant Value (hexadecimal) What is the total number of bytes allocated for data?.......................................................... Prepared by Dr. Muhamed Mudawar © KFUPM – Revised August 2006 COE 205 Lab Manual Lab 3: Defining Data and Symbolic Constants - page 35 3.8 Lab Work: Viewing Symbolic Constants in the Listing (.lst) File Open a command prompt and type: ml –c –Zi –Fl –coff Constants.asm. The ML program is the assembler. It will generate the Constants.obj and Constants.lst files. The Constants.obj is the object file and contains the machine code. The Constants.lst file is a listing file produced optionally when you use the –Fl option with the ML command. Open the Constants.lst file and examine its contents. This file shows the work of the assembler and contains a copy of the assembly language source code, offset addresses, translated machine code, procedures, and symbols. The symbolic constants and their values are listed in this file. Examine the content of the Constants.lst file, and check the values of the symbolic constants. Notice that the symbolic constant values are listed in hexadecimal. Review Questions 1. Write a data declaration for an 8-bit unsigned integer variable. 2. Write a data declaration for a 32-bit signed integer variable. 3. Declare a 16-bit signed integer and initialize it with the smallest negative 16-bit number. 4. Declare an unsigned 16-bit integer variable wArray that uses three initializers. 5. Declare an uninitialized array of 50 unsigned 32-bit integers named dArray. 6. Declare a string variable containing the word “TEST” repeated 100 times. 7. Show the order of individual bytes in memory (lowest to highest) for the following: dvar DWORD 5012AB6Fh 8. Consider the following array declaration: myArray DWORD 30 DUP(5 DUP(0)) Define a symbolic constant Elements that calculates the number of elements in myArray. Define a symbolic constant Size that calculates the number of bytes in myArray. Prepared by Dr. Muhamed Mudawar © KFUPM – Revised August 2006