Computer Organization & Assembly Language Programming

Size: px

Start display at page:

Download "Computer Organization & Assembly Language Programming"

August Pitts
6 years ago
Views:

1 Computer Organization & Assembly Language Programming CSE 2312 Lecture 11 Introduction of Assembly Language 1

2 Assembly Language Translation The Assembly Language layer is implemented by translation rather than by interpretation Translator is the programs that convert a user s program written in some language to another language Source Language: the language in which the original program is written Target Language: the one which it is converted Translation is used when a processor (either hardware or an interpreter) is available for the target language but not for the source language. Correct translation will give precisely the same results as the execution of the source program Two distinct steps Generating an equivalent program in the target language Executing the newly generated program. 2

Introduction Translators They can be roughly divided into two groups, depending on the

When the source language is essentially a symbolic representation for a numerical machine

3 Introduction Translators They can be roughly divided into two groups, depending on the relation between the source language and the target language. When the source language is essentially a symbolic representation for a numerical machine language, the translator is called an assembler and the source language is called an assembly language. When the source language is a high-level language such as Java or C and the target language is either a numerical machine language or a symbolic representation for one, the translator is called a compiler. 3

4 What is Assembly Language? One-to-one mapping A pure assembly language is a language in which each statement produces exactly one machine instruction. There is a one-to-one correspondence between machine instructions and statements in the assembly program. Fully accessing instructions in machines The assembly programmer has access to all instructions available on the target machine. The high-level language programmer does not. Everything that can be done in machine language can be done in assembly language, but many instructions, registers, and similar features are not available for the high-level language programmer to use. Running on only one family machine The assembly language program only can run on one family machines while a program written in high level language can potentially run on many different machines 4

5 Assembly Language vs. Machine Language Programming Comparing with that in machine language (in hexadecimal), programming in assembly language is much easier The use of symbolic names and symbolic addresses instead of binary or octal ones makes an enormous difference. Symbolic Name Most people can remember that the abbreviations for add, subtract, multiply, and divide are ADD, SUB, MUL, and DIV, but few can remember the corresponding numerical values the machine uses. The assembly language programmer need only remember the symbolic names because the assembler translates them to the machine instructions. Symbolic Address The assembly language programmer can give symbolic names to memory locations. The assembler translates them into correct numerical values. The machine language programmer must always work with the numerical values of the addresses. 5

6 Example Codes Machine Language Code Assembly Language Code 6

7 Why Use Assembly Language? Using assembly language is difficult Assembly language programming is difficult. It is hard to write errorfree long programs. Writing a program in assembly language takes much longer than writing the same program in a high-level language. It also takes much longer to debug and is much harder to maintain. Why Use? Two reasons: performance and access to the machine. First, an expert assembly language programmer can often produce code that is much smaller and much faster than a highlevel language programmer can. For some applications, speed and size are critical. Second, some procedures need complete access to the hardware, something usually impossible in high-level languages. For example, the device controllers in many embedded real-time systems fall into this. 7

8 Example Comparisons Assume, it requires 10 programmer-years to write some program in a high-level language. It requires 100 sec to execute a certain typical benchmark. A benchmark is a test program used to compare computers, compilers, etc. Writing the whole program in assembly language might require 50 programmer-years, and the final program might run the benchmark in about 33 sec, a speedup of 3 times. Observations 10% of the total program accounts for 90% of the execution time. For a 100-sec job, 90 sec is spent in this critical 10% and 10 sec is spent in the remaining 90% of the program. The critical 10% can now be improved by rewriting it in assembly language. This process is called tuning Tuning Additional 5 programmer years are needed to rewrite the critical procedures but their execution time is reduced from 90 sec to 30 sec Mixed high-level & Assembly language: 15 programmer years and 40 seconds 8

9 Example (cont d) 9

10 Exercise Compare three strategies with respect to programming time and execution time For a program, 2% of the code accounts for 50% of the execution time. Assume that it would take 100 man-months to write it in C, and that assembly code is 10 times slower to write and 4 times more efficient. a. Entire program in C. b. Entire program in assembler. c. First all in C, then the key 2% rewritten in assembler. Solution The three cases are as follows: a. Programming = 100 months, execution time = T. b. Programming = 1000 months, execution time = 0.25T. c. Programming = X10 = 118 months, execution time = 5T/8. 10

11 Format Assembly Language Statement Assembly language statements have four parts: a label field, an operation (opcode) field, an operands field, and a comments field. Label Field Labels are used to provide symbolic names for memory addresses and needed on executable statements so that the statements can be branched to. They are also needed for data words to permit the data stored there to be accessible by symbolic name. If a statement is labeled, the label (usually) begins in column 1. Opcode Field The opcode field contains a symbolic abbreviation for the opcode The choice of an appropriate name is just a matter of taste, and different assembly language designers often make different choices. The designers of the Intel assembler decided to use MOV for both loading a register from memory and storing a register into memory. The designers of the Motorola assembler chose MOVE for both of them These choices have nothing to do with the underlying machine. 11

12 Example Format Computation of N = I + J. (a) Pentium 4. 12

13 Example Format Computation of N = I + J. (b) Motorola 680x0.. 13

14 Pseudoinstructions Commands to the assembler itself are called pseudoinstructions. 14

15 Pseudoinstructions (cont d) Some of the pseudoinstructions available in the Pentium 4 assembler (MASM).. 15

16 Macro Macro definition A macro definition is a way to give a name to a piece of text. After a macro has been defined, the programmer can write the macro name instead of the piece of program. Basic parts in macro definition A macro header giving the name of the macro being defined The text comprising the body of the macro A pseudo instruction marking the end of the definition (e.g., ENDM). Macro call and expansion When the assembler encounters a macro definition, it saves it in a macro definition table for subsequent use. From that point on, whenever the name of the macro appears as an opcode, the assembler replaces it by the macro body. The use of a macro name as an opcode is known as a macro call and its replacement by the macro body is called macro expansion. 16

17 Macro Example Assembly language code for interchanging P and Q twice. (a) Without a macro. (b) With a macro.. 17

18 Macro Expansion Macro expansion occurs during the assembly process and not during execution of the program. This point is important. Looking only at the machine language program, it is impossible to tell whether or not any macros were involved in its generation. The reason is that once macro expansion has been completed, the macro definitions are discarded by the assembler. Assembly Process: Two Passes We can think of the assembly process as taking place in two passes. On Pass one, all the macro definitions are saved and the macro calls expanded On pass two, the resulting text is processed as though it was in the original program. The source program is read in and is then transformed into another program from which all macro definitions have been removed, and in which all macro calls have been replaced by their bodies. The resulting output, an assembly language program containing no macros at all, is then fed into the assembler. 18

19 Macro vs. Procedure Both used to repeat sequences of instruction within a program. However, macro calls should not be confused with procedure calls. The basic difference is that a macro call is an instruction to the assembler to replace the macro name with the macro body. A procedure call is a machine instruction that is inserted into the object program and that will later be executed to call the procedure. Procedure has the disadvantage of requiring a procedure call instruction and a return instruction to be executed every time a sequence is needed. This procedure call overhead may significantly slow the program down. Macros provide an easy and efficient solution to this problem 19

20 Macro vs. Procedure (cont d) 20

21 Macro vs. Procedure (cont d) 21

22 Macros with Parameters Nearly identical sequences of statements. (a) Without a macro. (b) With a macro. Formal parameters: P1 and P2 Actual parameters: P, Q, R, S 22

23 Advanced Features Why? These advanced feature will make life easier for the assembly language programmer. Label duplication Suppose that a macro contains a conditional branch instruction and a label that is branched to. If the macro is called two or more times, the label will be duplicated, causing an assembly error. One solution is to have the programmer supply a different label on each call as a parameter. MASM allows a label to be declared LOCAL, with the assembler automatically generating a different label on each expansion of the macro. Macros can call other macros Macros can call other macros, including themselves. If a macro is recursive, that is, it calls itself, it must pass itself a parameter that is changed on each expansion and the macro must test the parameter and terminate the recursion when it reaches a certain value. Otherwise the assembler can be put into an infinite loop. If this happens, the assembler must be killed explicitly by the user. 23

24 Macro Implementation Two functions To implement a macro facility, an assembler must be able to perform two functions: save macro definitions and expand macro calls. Maintaining a Table The assembler must maintain a table of all macro names. Each name has a pointer to its stored definition so that it can be retrieved when needed. Some assemblers have a separate table for macro names and some have a combined opcode table in which all machine instructions, pseudoinstructions, and macro names are kept. Encountering a macro definition When a macro definition is encountered, a table entry is made giving the name of the macro, the number of formal parameters, and a pointer to another table: the macro definition table, where the macro body will be kept. A list of the formal parameters is also constructed at this time for use in processing the definition. The macro body is then read and stored in the macro definition table. Formal parameters occurring within the body are indicated by some special symbol. When a macro is called, the assembler temporarily stops reading input from the input device and starts reading from the stored macro body instead. 24

25 Exercise True or False Macro expansion occurs during the execution of the program. An expert on machine language programming may tell whether or not any macros were involved from the corresponding machine code. P and Q are formal parameters of macro FFF 25

The A ssembly Assembly Language Level Chapter 7 1

The A ssembly Assembly Language Level Chapter 7 1 The Assembly Language Level Chapter 7 1 Contemporary Multilevel Machines A six-level l computer. The support method for each level is indicated below it.2 Assembly Language Level a) It is implemented by