Monday, October 31, 2011

macro processor

Posted by Unknown at 10:37 AM


Macro  Processor
Macro  Instruction 
  • A macro  instruction (macro) is simply a notational  convenience for the programmer.
  • A macro represents a commonly used group of statements in the source program.
  • The macro processor replaces each macro instruction with the corresponding group of source statements.
    • This operation is called “expanding the macro”
  • Using macros allows a programmer to write a shorthand version of a program.
  • For example, before calling a subroutine, the contents of all registers may need to be stored. This routine work can be done using a macro.
Machine  Independent 
  • The functions  of a macro processor essentially involve  the substitution of one group of lines  for another. Normally, the processor performs  no analysis of the text it handles.
  • The meaning of these statements are of no concern during macro expansion.
  • Therefore, the design of a macro processor generally is machine independent.
  • Macro mostly are used un assembler language programming. However, it can also be used in high-level programming languages such as C or C++.
Basic  Functions 
  • Macro definition
    • The two directive MACRO and MEND are used in macro definition.
    • The macro’s name appears before the MACRO directive.
    • The macro’s parameters appear after the MACRO directive.
    • Each parameter begins with ‘&’
    • Between MACRO and MEND is the body of the macro. These are the statements that will be generated as the expansion of the macro definition.
Basic  Functions 
  • Macro expansion  (or invocation)
    • Give the name of the macro to be expanded and the arguments to be used in expanding the macro.
    • Each macro invocation statement will be expanded into the statements that form the body of the macro, with arguments from the macro invocation substituted for the parameters in the macro prototype.
    • The arguments and parameters are associated with one another according to their positions.
      • The first argument corresponds to the first parameter, and so on.
Macro  Program Example 
Macro  definition 
Avoid  the use of labels in a macro
Macro  definition 
Avoid  the use of labels in a macro
Macro  invocations
Expanded  Macro Program
Retain  Labels on Expanded Macro 
  • The label  on the macro invocation statement CLOOP  has been retained as a label on the  first statement generated in the macro  expansion.
  • This allows the programmer to use a macro instruction in exactly the same way as an assembler language mnemonic.
Differences  between Macro and Subroutine 
  • After macro  processing, the expanded file can be used  as input to the assembler.
  • The statements generated from the macro expansions will be assembled exactly as though they had been written directly by the programmer.
  • The differences between macro invocation and subroutine call
    • The statements that form the body of the macro are generated each time a macro is expanded.
    • Statements in a subroutine appear only once, regardless of how many times the subroutine is called.
Avoid  Uses of Labels in Macro 
  • In RDBUFF  and WRBUFF macros, many program-counter relative  addressing instructions are used to avoid  the uses of labels in a macro.
    • For example,   JLT  * - 19
  • This is to avoid generating duplicate labels when the same macro is expanded multiple time at different places in the program. (will be treated as error by the assembler)
  • Later on, we will present a method which allows a programmer to use labels in a macro definition.
Two-Pass  Macro Processor 
  • Like an  assembler or a loader, we can design  a two-pass macro processor in which all  macro definitions are processed during the  first pass, and all macro invocation statements  are expanded during the second pass.
  • However, such a macro processor cannot allow the body of one macro instruction to contain definitions of other macros.
    • Because all macros would have to be defined during the first pass before any macro invocations were expanded.
Macro  Containing Macro Example
Macro  Containing Macro Example 
  • MACROS contains  the definitions of RDBUFF and WRBUFF which  are written in SIC instructions.
  • MACROX contains the definitions of RDBUFF and WRBUFF which are written in SIC/XE instructions.
  • A program that is to be run on SIC system could invoke MACROS whereas a program to be run on SIC/XE can invoke MACROX.
  • Defining MACROS or MACROX does not define RDBUFF and WRBUFF. These definitions are processed only when an invocation of MACROS or MACROX is expanded.
One-Pass  Macro Processor 
  • A one-pass  macro processor that alternate between macro  definition and macro expansion is able  to handle “macro in macro”.
  • However, because of the one-pass structure, the definition of a macro must appear in the source program before any statements that invoke that macro.
    • This restriction is reasonable.
Data  Structures 
  • DEFTAB
    • Store the definition statements of macros
    • Comment lines are omitted.
    • References to the macro instruction parameters are converted to a positional notation for efficiency in substituting arguments.
  • NAMTAB
    • Store macro names, which serves an index to DEFTAB
    • Contain pointers to the beginning and end of the definition
  • ARGTAB
    • Used during the expansion of macro invocations.
    • When a macro invocation statement is encountered, the arguments are stored in this table according to their position in the argument list.
Data  Structures Snapshot
Algorithm 
  • Procedure DEFINE
    • Called when the beginning of a macro definition is recognized. Make appropriate entries in DEFTAB and NAMTAB.
  • Procedure EXPAND
    • Called to set up the argument values in ARGTAB and expand a macro invocation statement
  • Procedure GETLINE
    • Get the next line to be processed
Handle  Macro in Macro 
  • When a  macro definition is being entered into  DEFTAB, the normal approach is to continue  until an MEND directive is reached.
  • This will not work for “macro in macro” because the MEND first encountered (for the inner macro) will prematurely end the definition of the outer macro.
  • To solve this problem, a counter LEVEL is used to keep track of the level of macro definitions. A MEND will end the definition of the macro currently being processed only when LEVEL is 0.
    • This is very much like matching left and right parentheses when scanning an arithmetic expression.
Algorithm  Pseudo Code
Machine  Independent Features
Concatenation  of Macro Parameters 
  • Most macro  processors allow parameters to be concatenated  with other character stings.
  • E.g., to flexibly and easily generate the variables XA1, XA2, XA3, …, or XB1, XB2, XB3, “A” or “B” can be input as an argument. We just need to concatenate “X”, the argument, and the “1” , “2”, “3” .. together.
Concatenation  Problem 
  • Suppose that  the parameter to such a macro instruction  is named &ID, the body of the  macro definition may contain a statement  like LDA X&ID1, in which &ID is  concatenated after the string “X” and  before the string “1”.
  • The problem is that the end of the parameter is not marked. Thus X&ID1 may mean “X” + ID + “1” or “X” + ID1.
  • To avoid this ambiguity, a special concatenation operator -> is used. The new form becomes X&ID->1. Of course, -> will not appear in the macro expansion.
Concatenation  Example
Generation  of Unique Labels 
  • Previously  we see that, without special processing,  if labels are used in macro definition,  we may encounter the “duplicate labels”  problem if a macro is invocated multiple  time.
  • To generate unique labels for each macro invocation, when writing macro definition, we must begin a label with $.
  • During macro expansion, the $ will be replaced with $xx, where xx is a two-character alphanumeric counter of the number of macro instructions expanded.
    • XX will start from AA, AB, AC,…..
Unique  Labels Macro Definition
Unique  Labels Macro Expansion
Conditional  Macro Expansion 
  • So far,  when a macro instruction is invoked, the  same sequence of statements are used to  expand the macro.
  • Here, we allow conditional assembly to be used.
    • Depending on the arguments supplied in the macro invocation, the sequence of statements generated for a macro expansion can be modified.
  • Conditional macro expansion can be very useful. It can generate code that is suitable for a particular application.
Conditional  Macro Example 
  • In the  following example, the values of &EOR  and &MAXLTH parameters are used to  determine which parts of a macro definition  need to be generated.
  • There are some macro-time control structures introduced for doing conditional macro expansion:
    • IF- ELSE-ENDIF
    • WHILE-ENDW
  • Macro-time variables can also be used to store values that are used by these macro-time control structures.
    • Used to store the boolean expression evaluation result
    • A variable that starts with & but not defined in the parameter list is treated as a macro-time variable.
Conditional  macro control structure 
Macro  time variable
Conditional  macro expansion 1
Conditional  macro expansion 2
Conditional  macro expansion 3
Conditional  Macro Implementation 
  • The assembler  maintains a symbol table that contains  the values of all macro-time variables  used.
  • Entries in this table are made or modified when SET statements are processed.
  • When an IF statement is encountered during the expansion of a macro, the specified boolean expression is evaluated.
    • If the value of this expression is TRUE, the macro processor continues to process until it encounters the next ELSE or ENDIF.
      • If ELSE is encountered, then skips to ENDIF
    • Otherwise, the assembler skips to ELSE and continues to process until it reaches ENDIF.
Conditional  Macro Example 
Macro  processor function
Conditional  Macro Expansion v.s. Conditional Jump Instructions  
  • The testing  of Boolean expression in IF statements  occurs at the time macros are expanded. 
  • By the time the program is assembled, all such decisions have been made.
  • There is only one sequence of source statements during program execution.
  • In contrast, the COMPR instruction test data values during program execution. The sequence of statements that are executed during program execution may be different in different program executions.
Keyword  Macro Parameters 
  • So far,  all macro instructions use positional parameters.
    • If an argument is to be omitted, the macro invocation statement must contain a null argument to maintain the correct argument positions.
    • E.g., GENER ,,DIRECT,,,,,,3.
  • If keyword parameters are used, each argument value is written with a keyword that names the corresponding parameters.
    • Arguments thus can appear in any order.
    • Null arguments no longer need to be used.
    • E.g., GENER TYPE=DIRECT, CHANNEL=3
  • Keyword parameter method can make a program easier to read than the positional method.
Keyword  Macro Example 
Can specify  default values
Keyword  parameters
Design  Options
Recursive  Macro Expansion 
  • If we  want to allow a macro to be invoked  in a macro definition, the already presented  macro processor implementation cannot be used.
  • This is because the EXPAND routine is recursively called but the variable used by it (e.g., EXPANDING) is not saved across these calls.
  • It is easy to solve this problem if we use a programming language that support recursive functions. (e.g., C or C++).
Recursive  Macro Example
Recursive  Macro Example 
For easy  implementation, we require that RDCHAR macro 
be defined  before it is used in RDBUFF macro.  This
requirement  is very reasonable.

 

0 comments:

Post a Comment

Subscribe to: Post Comments (Atom)