H MACRO-64_Assembler_for_OpenVMS_AXP_Systems__________$ Reference Manual - Order Number: AA-PT9KA-TE ! November 1992 A This document describes the MACRO-64 assembly language. G Revision/Update Information: This is a new manual. H Software Version: MACRO-64 Assembler forE OpenVMS AXP Systems = Version 1.0 I Operating System and Version: OpenVMS AXP Version 1.0 1 Digital Equipment Corporation * Maynard, Massachusetts N ________________________________________________________________ November 1992 C The information in this document is subject to change H without notice and should not be construed as a commitmentA by Digital Equipment Corporation. Digital Equipment G Corporation assumes no responsibility for any errors that * may appear in this document. H The software described in this document is furnished underI a license and may be used or copied only in accordance with ( the terms of such license. E No responsibility is assumed for the use or reliability F of software on equipment that is not supplied by Digital@ Equipment Corporation or its affiliated companies. 3 ⌐ Digital Equipment Corporation 1992. " All Rights Reserved. E The postpaid Reader's Comments forms at the end of this D document request your critical evaluation to assist in- preparing future documentation. ? The following are trademarks of Digital Equipment B Corporation: Alpha AXP, AXP, DEC, DEC 4000, DECchip,A DECwindows, Digital, OpenVMS, OpenVMS AXP, VAX, VAX @ DOCUMENT, VMS, the AXP logo, and the DIGITAL logo. 7 The following are third-party trademarks: F Ada[R] is a registered trademark of the U.S. Government,' Ada Joint Program Office. I This document was prepared using VAX DOCUMENT, Version 2.1. F _________________________________________________________________ F Contents F Preface................................................... xi 1 Introduction F 1.1 Language Features................................ 1-1F 1.1.1 Alpha AXP Native Mode Instructions ............ 1-1F 1.1.2 Direct Assignment Statements .................. 1-2F 1.1.3 Assembler Directives .......................... 1-2F 1.1.3.1 General Assembler Directives................ 1-2F 1.1.3.2 Macro Directives............................ 1-2F 1.2 Invoking MACRO-64 on OpenVMS Systems............. 1-3F 1.2.1 Command Line Qualifiers ....................... 1-3 0 2 Components of MACRO-64 Source Statements F 2.1 Source Statement Format.......................... 2-1F 2.1.1 Label Field ................................... 2-3F 2.1.2 Operator Field ................................ 2-4F 2.1.3 Operand Field ................................. 2-4F 2.1.4 Comment Field ................................. 2-5F 2.2 Character Set.................................... 2-5F 2.3 Numbers.......................................... 2-8F 2.3.1 Integers ...................................... 2-9F 2.3.2 Floating-Point Numbers ........................ 2-9F 2.4 Quoted Literals.................................. 2-10F 2.5 Symbols.......................................... 2-11F 2.5.1 Permanent Symbols ............................. 2-11F 2.5.2 User-Defined Symbols and Macro Names .......... 2-12F 2.5.3 Determining Symbol Values ..................... 2-13F 2.5.3.1 Using Symbols in the Operator Field......... 2-14F 2.5.3.2 Using Symbols in the Operand Field.......... 2-14F 2.6 Temporary Labels Within Source Code.............. 2-15F 2.7 Label Addresses ................................. 2-17 F iii : 2.7.1 Label Addresses, Optimization, and CodeI Alignment...................................... 2-18 ? 2.7.2 Label Addresses and Automatic Data Alignment I ............................................... 2-19 I 2.8 Terms and Expressions............................ 2-22 I 2.9 Unary Operators for Terms and Expressions........ 2-25 I 2.9.1 Radix Control Operators ....................... 2-26 I 2.9.2 Numeric Complement Operator ................... 2-27 I 2.10 Binary Operators................................. 2-28 I 2.10.1 Arithmetic Shift Operator ..................... 2-29 I 2.10.2 Logical AND Operator .......................... 2-29 I 2.10.3 Logical Inclusive OR Operator ................. 2-29 I 2.10.4 Logical Exclusive OR Operator ................. 2-29 I 2.11 Direct Assignment Statements..................... 2-30 I 2.12 Current Location Counter......................... 2-31 $ 3 MACRO-64 Lexical Operator I 3.1 Processing with Lexical Operators................ 3-1 I 3.2 Lexical Operator Syntax.......................... 3-1 I 3.3 Numeric Symbols and Lexical String Symbols....... 3-4 I 3.4 Lexical Substitution Operator.................... 3-4 I 3.5 Lexical Escape Operator.......................... 3-5 I 3.6 Using Lexical Operators.......................... 3-7 I 3.7 Lexical Operators................................ 3-11 I %EDIT............................................ 3-12 I %ELEMENT......................................... 3-14 I %EXTRACT......................................... 3-15 I %INTEGER......................................... 3-16 I %LENGTH.......................................... 3-17 I %LOCATE.......................................... 3-18 I %REPEAT.......................................... 3-19 I %STRING.......................................... 3-20 I %TIME............................................ 3-21 I %TYPE............................................ 3-22
iv 4 Macro Arguments I 4.1 Using Macro Arguments............................ 4-2 I 4.2 Using Default Values............................. 4-3 I 4.3 Using Keyword Arguments.......................... 4-3 I 4.4 Using String Arguments........................... 4-5 I 4.5 Argument Concatenation........................... 4-9 I 4.6 Passing Numeric Values of Symbols................ 4-9 I 4.7 Using Created Temporary Labels................... 4-10 ( 5 MACRO-64 Assembler Directives I 5.1 Program Sections (Psects)........................ 5-1 I 5.2 Automatic Data Alignment......................... 5-3 I 5.3 Directives....................................... 5-4 I .ADDRESS......................................... 5-9 I .ALIGN........................................... 5-10 I .ASCIC........................................... 5-13 I .ASCID........................................... 5-14 I .ASCII........................................... 5-16 I .ASCIZ........................................... 5-17 I .BASE............................................ 5-18 I .BEGIN_EXACT..................................... 5-22 I .BLKx............................................ 5-24 I .BYTE............................................ 5-27 I .CODE_ADDRESS.................................... 5-28 I .D_FLOATING...................................... 5-29 I .DISABLE......................................... 5-31 I .ELSE............................................ 5-32 I .ENABLE.......................................... 5-34 I .END............................................. 5-44 I .ENDC............................................ 5-46 I .ENDM............................................ 5-47 I .ENDR............................................ 5-48 I .END_EXACT....................................... 5-49 I .ERROR........................................... 5-50 I .EVEN............................................ 5-51 I .EXTERNAL........................................ 5-52 I .F_FLOATING...................................... 5-53 I .G_FLOATING...................................... 5-55 I v I .IDENT........................................... 5-56 I .IF.............................................. 5-57 I .IF_x............................................ 5-63 I .IIF............................................. 5-67 I .INCLUDE......................................... 5-69 I .INSTRUCTION..................................... 5-70 I .IRP............................................. 5-72 I .IRPC............................................ 5-75 I .LIBRARY......................................... 5-78 I .LINKAGE_PAIR.................................... 5-80 I .LIST............................................ 5-82 I .LOCAL_CODE_ADDRESS.............................. 5-83 I .LOCAL_LINKAGE_PAIR.............................. 5-84 I .LOCAL_PROCEDURE_DESCRIPTOR...................... 5-85 I .LONG............................................ 5-87 I .MACRO........................................... 5-88 I .MCALL........................................... 5-92 I .MDELETE......................................... 5-93 I .MEXIT........................................... 5-94 I .NARG............................................ 5-97 I .NCHR............................................ 5-99 I .NLIST........................................... 5-101 I .NOSHOW.......................................... 5-102 I .OCTA............................................ 5-103 I .ODD............................................. 5-105 I .PACKED.......................................... 5-106 I .PAGE............................................ 5-107 I .PRINT........................................... 5-108 I .PROCEDURE_DESCRIPTOR............................ 5-109 I .PSECT........................................... 5-111 I .QUAD............................................ 5-118 I .REPEAT.......................................... 5-119 I .RESTORE_PSECT................................... 5-122 I .SAVE_PSECT...................................... 5-124 I .S_FLOATING...................................... 5-125 I .SHOW............................................ 5-126 I .SIGNED_BYTE..................................... 5-129
vi I .SIGNED_WORD..................................... 5-130 I .SUBTITLE........................................ 5-131 I .T_FLOATING...................................... 5-132 I .TITLE........................................... 5-133 I .WARN............................................ 5-135 I .WEAK............................................ 5-136 I .WORD............................................ 5-138 # 6 MACRO-64 Supplied Macros I 6.1 MACRO-64 Supplied Library........................ 6-1 I 6.2 Routines and Lexical Scope....................... 6-2 I 6.2.1 Routines and Program Sections ................. 6-2 I 6.3 Using Macros to Control Program Sections......... 6-3 I 6.3.1 Defining Program Sections ..................... 6-3 I 6.3.2 Using Macro Specific Symbols .................. 6-4 I 6.3.3 Defining Procedure Type ....................... 6-5 I 6.3.4 Using Macros in Prologue Sections ............. 6-5 I 6.3.5 Using Macros in Epilogue Sections ............. 6-6 I 6.4 Programming Examples Using Supplied Macros....... 6-6 I 6.5 Using the $CALL Macro............................ 6-8 I 6.5.1 Using $CALL in Source Code .................... 6-9 I 6.6 Programming Considerations....................... 6-11 I 6.6.1 Making Multiple Calls From the Same Routine ... 6-11 I 6.6.2 Non-Standard Linkage .......................... 6-11 I 6.6.3 Routine Restrictions .......................... 6-11 I 6.7 Macro Descriptions............................... 6-12 I 6.7.1 Syntax Rules .................................. 6-12 I $BEGIN_EPILOGUE.................................. 6-13 I $CALL............................................ 6-15 I $CODE_SECTION ................................... 6-22 I $DATA_SECTION.................................... 6-23 I $END_EPILOGUE.................................... 6-24 I $END_PROLOGUE.................................... 6-26 I $END_ROUTINE..................................... 6-27 I $LINKAGE_PAIR.................................... 6-28 I $LINKAGE_SECTION................................. 6-30 I $OPDEF........................................... 6-31 I .PACKED.......................................... 6-37 I $PROCEDURE_DESCRIPTOR............................ 6-40 I vii I $RETURN.......................................... 6-41 I $ROUTINE......................................... 6-42 : A MACRO-64 Alpha AXP Architecture Quick Reference I A.1 Register Usage Conventions....................... A-3 I A.2 Instruction Operand Notation..................... A-4PI A.3 Instruction Qualifier Notation................... A-5 I A.4 F-P Control Register (FPCR) Format............... A-5 I A.5 Decodable Pseudo-Operations...................... A-7 I A.6 Common Architecture Opcodes in Numerical Order... A-9oI A.7 OpenVMS PALcode Instruction Summary.............. A-16 I A.8 PALcode Opcodes in Numerical Order............... A-20 I A.9 Common Architecture Instructions................. A-23 $ B Programming with MACRO-64 I B.1 OpenVMS Calling Standard......................... B-1 I B.2 Accessing Memory with Base Registers............. B-2tI B.3 Data Structures.................................. B-6_I B.3.1 Procedure Descriptor .......................... B-6hI B.3.2 Signature Block ............................... B-7 I B.3.3 Linkage Pair .................................. B-7lI B.4 Types of Routines................................ B-9eI B.4.1 Routine Capabilities .......................... B-10dI B.4.2 Entry Prologue and Exit Epilogue Sequences .... B-11hI B.5 Example Program.................................. B-12nI B.6 Establishing Self-Addressability................. B-13oI B.7 Optimization and Automatic Alignment............. B-15 I B.7.1 Automatic Data Alignment ...................... B-15rI B.7.1.1 Controlling Data Alignment.................. B-16pI B.7.1.2 Directives for Automatic Data Alignment..... B-16oI B.7.2 Automatic Code Label Alignment ................ B-18rI B.7.3 Scheduling Optimization ....................... B-18eI B.7.4 Peephole Optimization ......................... B-18qI B.7.5 Using MACRO-64 for Performance ................ B-19,I B.7.6 Viewing Optimization Results .................. B-20 viii l , " C Using LSE with MACRO-64 I C.1 Invoking LSE..................................... C-1eI C.2 Running Diagnostics.............................. C-1 % D Differences from VAX MACROi I D.1 Assembler Features in MACRO-64................... D-1_I D.2 VAX MACRO Features Not Present in MACRO-64....... D-6 E Error Messages Index Examples 8 3-1 Lexical Processing Without the EscapeI Operator....................................... 3-6. I 3-2 Lexical Processing With Escape Operator ....... 3-6o I 3-3 Using Lexical Operators ....................... 3-7. I 3-4 Source Statements After Macro Expansion ....... 3-10. I 6-1 Program Using Supplied Macros ................. 6-6 I 6-2 Program Using $CALL ........................... 6-10 I B-1 Routine Without Linkage Pairs ................. B-8 I B-2 Routine With Linkage Pairs .................... B-9o I B-3 Establishing a Base Address ................... B-14a I B-4 Enabling and Disabling Data Alignment ......... B-16. Figures. I A-1 Data Structures ............................... A-12 I A-2 Instruction Formats ........................... A-2 I ix2 r Tables I 2-1 Using Tab Stops in Statement Fields ........... 2-2. 6 2-2 Special Characters Used in MACRO-64I Statements..................................... 2-7. I 2-3 Summary of Unary Operators .................... 2-25. I 2-4 Summary of Binary Operators ................... 2-28. I 3-1 Summary of MACRO-64 Lexicals .................. 3-11. I 3-2 %TYPE Attributes .............................. 3-22- I 5-1 Summary of General Assembler Directives ....... 5-4 I 5-2 Summary of Macro Directives ................... 5-7 I 5-3 .ENABLE and .DISABLE Symbolic Arguments ....... 5-34 ; 5-4 Condition Tests for Conditional Assembly I Directives..................................... 5-58 I 5-5 Program Section Attributes .................... 5-112 I 5-6 Default Program Section Attributes ............ 5-115 I 5-7 .SHOW and .NOSHOW Symbolic Arguments .......... 5-1262 I 6-1 ARGS Arguments ................................ 6-17. I A-1 Register Usage Conventions for OpenVMS AXP .... A-3 I A-2 Instruction Operand Notation .................. A-4t I A-3 Instruction Qualifier Notation ................ A-5r I A-4 F-P Control Register (FPCR) Format ............ A-5l I A-5 Decodable Pseudo-Operations ................... A-7. ; A-6 Common Architecture Opcodes in NumericaliI Order.......................................... A-9g I A-7 OpenVMS Unprivileged PALcode Instructions ..... A-16c I A-8 OpenVMS Privileged PALcode Instructions ....... A-17l I A-9 PALcode Opcodes in Numerical Order ............ A-20g I A-10 Common Architecture Instructions .............. A-23i I B-1 Frame Attributes .............................. B-11p I B-2 Directives Using Automatic Alignment .......... B-16 x 2 I _________________________________________________________________e I Prefacee C This manual describes the MACRO-64 assembly language.. Intended Audience H This manual is intended for all programmers writing MACRO-D 64 programs. Before reading this manual, you should beH familiar with assembly language programming, the Alpha AXPH architecture and instruction set, and the operating system( in use on your AXP system. Document Structure I This manual contains the following chapters and appendices:. B o Chapter 1 introduces the features of the MACRO-64 language. F o Chapter 2 describes the format of the MACRO-64 source statements. G o Chapter 3 describes the MACRO-64 lexical operators. It H also discusses differences and similarities between theB VAX MACRO string operators and those of MACRO-64. D o Chapter 4 describes the arguments used with macros. C o Chapter 5 describes the MACRO-64 general assembler C directives and the directives used in defining and " expanding macros. H o Chapter 6 describes macros provided with the (MACRO-64) Assembler.. G o Appendix A provides a quick reference to the Alpha AXP.G architecture as it relates to the MACRO-64 programming.C language. For more information, refer to the Alpha.' Architecture Handbook. I xis b G o Appendix B provides information on how to program more : effectively using the MACRO-64 assembler. F o Appendix C explains how to invoke the LSE editor with' the MACRO-64 language.D G o Appendix D describes the differences between VAX MACROA' and MACRO-64 features.. H o Appendix E describes the error messages produced by the$ MACRO-64 assembler. Related Manuals. I For more information on MACRO-64, see the following manuals.( in this documentation set: H o MACRO-64 Assembler for OpenVMS AXP Systems Installation Guide I o MACRO-64 Assembler for OpenVMS AXP Systems Release Notes. I For more information on the Alpha AXP architecture, see the. following: 4 o Alpha Architecture Reference Manual , o Alpha Architecture Handbook < o Alpha AXP Achitecture Quick Reference Guide ) o OpenVMS Calling Standard. Conventions. F In this manual, every use of Alpha VMS means the OpenVMSB AXP operating system, every use of VAX VMS means theF OpenVMS VAX operating system, and every use of VMS meansG both the OpenVMS AXP operating system and the OpenVMS VAX. operating system.4 @ The following conventions are used in this manual: G Ctrl/x A sequence such as Ctrl/x indicates that.F you must hold down the key labeled CtrlH while you press another key or a pointing- device button.R xii. . . F PF1 x A sequence such as PF1 x indicates thatC you must first press and release the.F key labeled PF1, then press and releaseG another key or a pointing device button.. D <Return> In examples, a key name enclosed in aD box indicates that you press a key onH the keyboard. (In text, a key name is not2 enclosed in a box.) @ . . . A horizontal ellipsis in examples= indicates one of the following.- possibilities: D o Additional optional arguments in a> statement have been omitted. D o The preceding item or items can be= repeated one or more times.. I o Additional parameters, values, or other.= information can be entered.N I A vertical ellipsis indicates the omissionSF . of items from a code example or commandD . format; the items are omitted becauseH . they are not important to the topic being) discussed.. B ( ) In format descriptions, parenthesesE indicate that, if you choose more than.G one option, you must enclose the choices.. in parentheses. H [ ] In format descriptions, brackets indicateE optional elements. You can choose one, E none, or all of the options. (Brackets.@ are not optional, however, in theB syntax of a directory name in a VMSF file specification, or in the syntax ofI a substring specification in an assignment.* statement.) F { } In format descriptions, braces surroundE a required choice of options; you must @ choose one of the options listed. I xiii. . . H boldface text Boldface text represents the introductionH of a new term or the name of an argument,9 an attribute, or a reason.. F Boldface text is also used to show userF input in online versions of the manual. ? italic text Italic text emphasizes important.D information, indicates variables, andD indicates complete titles of manuals.F Italic text also represents informationD that can vary in system messages (forG example, Internal error number), command.G lines (for example, /PRODUCER=name), and.: command parameters in text. F UPPERCASE TEXT Uppercase text indicates a command, theH name of a routine, the name of a file, orG the abbreviation for a system privilege. G - A hyphen in code examples indicates that F additional arguments to the request areA provided on the line that follows.p A numbers All numbers in text are assumed tonB be decimal, unless otherwise noted.C Nondecimal radixes-binary, octal, or D hexadecimal-are explicitly indicated. D mouse The term mouse refers to any pointingD device, such as a mouse, a puck, or a& stylus. G MB1, MB2, MB3 MB1 indicates the left mouse button, MB2 E indicates the middle mouse button, andcI MB3 indicates the right mouse button. (The E buttons can be redefined by the user.)h I PB1, PB2, PB3, PB1, PB2, PB3, and PB4 indicate buttons on$( PB4 the puck. H SB, SB SB and SB indicate buttons on the stylus. F VMS, OpenVMS, The terms VMS, OpenVMS, and OpenVMS AXPB OpenVMS AXP refer to the same operating system. H MACRO-64 MACRO-64 refers to MACRO-64 Assembler for3 OpenVMS AXP Systems.. xiv. . . I 1 I _________________________________________________________________ I Introduction G MACRO-64 Assembler for OpenVMS AXP Systems is an assembly B language for programming Alpha AXP computers. SourceI programs written in MACRO-64 are translated into object (or G binary) code by the MACRO-64 assembler, which produces an.I object module and, optionally, a listing file. The features.H of the language are introduced in this chapter, as well as3 instructions to invoke the assembler.N 1.1 Language Features F MACRO-64 source programs consist of a sequence of sourceC statements. These source statements may be any of the. following: 3 o Alpha AXP native-mode instructions. - o Direct assignment statements % o Assembler directivesi 0 1.1.1 Alpha AXP Native Mode Instructions G Instructions manipulate data. They perform such functionsED as addition, data conversion, and transfer of control.G Instructions are usually followed in the source statementaG by operands, which can be any kind of data needed for theoC operation of the instruction. For a brief descriptiontF of the Alpha instruction set, see Appendix A. For a moreG detailed description of the Alpha native mode instruction5E set, see the Alpha Architecture Reference Manual or the * Alpha Architecture Handbook. I 1-1. 1 Introduction 1.1 Language Features. * 1.1.2 Direct Assignment Statements D Direct assignment statements equate symbols to values.G For more information on direct assignment statements, see. Section 2.11. " 1.1.3 Assembler Directives I Directives guide the assembly process and provide tools for.G using the instructions. For more information on assembler.( directives, see Chapter 5. D There are two classes of assembler directives: general8 assembler directives and macro directives. , 1.1.3.1 General Assembler Directives E General assembler directives can be used to perform thee# following operations: > o Store data or reserve memory for data storage H o Control the alignment of parts of the program in memory H o Specify the methods of accessing the sections of memory4 in which the program will be stored 3 o Specify a procedure in the programm D o Specify the way in which symbols will be referenced F o Specify that a part of the program is to be assembled. only under certain conditions > o Display informational and diagnostic messages I o Control the assembler options that are used to interpret.# the source programP 1.1.3.2 Macro Directives C Macro directives are used to define macros and repeat.C blocks. They allow you to repeat identical or similartI sequences of source statements throughout a program without H rewriting those sequences. Use of macros and repeat blocksG helps minimize programmer errors and speeds the debuggingo process. 1-2 I IntroductionAI 1.2 Invoking MACRO-64 on OpenVMS SystemsP 0 1.2 Invoking MACRO-64 on OpenVMS Systems F To invoke MACRO-64, use the following MACRO command with4 the /ALPHA_AXP command line qualifier: 2 MACRO/ALPHA_AXP file-spec[, . . . ] D Note that you must specify the /ALPHA_AXP command lineA qualifier before any other command line parameters,.1 qualifiers, or file specifications.t ! file-spec[, . . . ]. F If you do not specify a file type for an input file, the; assembler uses the default file type of .M64.. G You can specify one or more source files to be assembled.P? To assemble files individually, separate the file E specifications with commas. To concatenate and assembleiA the files as a single input file, separate the file 1 specifications with plus signs (+). H Command line qualifiers control special assembler options.G Assembler options can apply to the entire MACRO/ALPHA_AXP.F command line, or to the individual file being assembled.E When the qualifier follows the MACRO/ALPHA_AXP command, E it applies to all files listed. For more information on.9 command line qualifiers, see Section 1.2.1.i % 1.2.1 Command Line Qualifiers. I This section describes the command qualifiers used with the & MACRO/ALPHA_AXP command. # /[NO]ALIGNMENT=optioni I Controls the alignment of code and data. Valid options are:A I ___________________________________________________________uI Option____Function_________________________________________ B code Alignment of certain branch target labels. I data______Natural_alignment_of_data_items._________________N B If you omit the qualifier from the command line, theG default option is /NOALIGNMENT=(code, data). If more thani I 1-3m c n Introduction0 1.2 Invoking MACRO-64 on OpenVMS Systems F one option is specified, the options must be enclosed in3 parentheses and separated by a comma. $ /[NO]DEBUG[=(options)] 9 Specifies DEBUG support. Valid options are:g I ___________________________________________________________OI Option____Function_________________________________________n ; symbol Generates debug symbol information. 8 traceback Generates traceback information. > all Generates all above debug information. I none______Generates_no_debug_information.__________________ H The default qualifier is /NODEBUG. The default when /DEBUG: is specified with no keywords is /DEBUG=all. ) /[NO]DIAGNOSTIC[=file-spec]P D Controls whether diagnostics are created and stored inE the specified optional file. If a file specification is C not supplied, the assembler creates a diagnostic file B using the same name as the source file. For example,C if you use a source file named XXX.M64, the assembler F creates a diagnostic file named XXX.DIA. You can use theE the diagnostic file with other Digital layered products C including, but not limited to, the Language-Sensitive Editor (LSE).c 5 The default qualifier is /NODIAGNOSTIC.d $ /ENVIRONMENT=[NO]FLOAT E Controls whether the assembler generates floating-point.I instructions when optimizing code and performing code-labele alignment. I Currently, the only floating-point instruction generated bytH the assembler during optimization and alignment processingC is FNOP, the floating-point no-operation instruction. E If you specify /ENVIRONMENT=NOFLOAT, the assembler doesoE not generate any floating-point instructions as part of2C optimization and alignment processing. Floating-pointe 1-4 I IntroductioniI 1.2 Invoking MACRO-64 on OpenVMS SystemsO F instructions that you specify in your source program are unaffected.h /LIBRARY > Searches macro libraries in the following order: B 1. The library designated by the /LIBRARY qualifier. ) 2. The .LIBRARY directives.C H 3. The MACRO64.MLB library. The assembler searches for theF MACRO64.MLB macro library in the following locations:I MACRO64$LIBRARY, ALPHA$LIBRARY, and finally SYS$LIBRARY.r D 4. The STARLET.MLB library. The assembler searches forC the the STARLET.MLB macro library in the following G locations: MACRO64$LIBRARY, ALPHA$LIBRARY, and finally SYS$LIBRARY.i F In addition, you can place the macro library definitionsE in the listing file by using the command line qualifieri /SHOW=LIBRARY. # /[NO]LIST[=file-spec]a B Controls whether a listing is created and optionallyI provides an output file specification for the listing file. H Do not use wildcard characters in this file specification. E If you issue the MACRO/ALPHA_AXP command interactively,eG /NOLIST is the default. The assembler sends output to theiI current output device rather than to a listing file. If you G execute the MACRO/ALPHA_AXP command in a batch job, /LIST is the default.m G If you do not specify a file specification, the assemblerhF creates a listing file using the same name as the sourceH file. For example, if you use a source file named XXX.M64,A the assembler creates a listing file named XXX.LIS. I 1-5 t Introduction0 1.2 Invoking MACRO-64 on OpenVMS Systems /[NO]MACHINE_CODEe E Produces a binary machine code listing after the source A text if a listing file is requested. The default is /NOMACHINE_CODE. /NAMES=case_option F Specifies the alphabetic casing of identifiers in source1 code statements. Valid options are: I ___________________________________________________________ I Option____Function_________________________________________ D upper_ Converts all identifiers to upper alphabetic case case.q D lower_ Converts all identifiers to lower alphabetic case case.. I as_is Causes all identifiers to remain in the case used I __________in_source_statements.____________________________ F If you use the /NAMES option in a command line, you mustF supply a case_option. If you omit the qualifier from theD command line, the default option is /NAMES=upper_case. % /[NO]OBJECT[=file-spec]x G Controls whether an object file is created and optionally @ provides a file specification. Do not use wildcard4 characters in this file specification. G If you do not specify a file specification, the assembler F creates an object file using the same name as the sourceH file. For example, if you use a source file named XXX.M64,A the assembler creates an object file named XXX.OBJ. / The default qualifier is /OBJECT. + /[NO]OPTIMIZE[=(option-list)]e 1-6 I Introduction I 1.2 Invoking MACRO-64 on OpenVMS Systems H Specifies optional assembler optimizations. Valid items in" the option-list are: I ___________________________________________________________ I Option____Function_________________________________________d 9 schedule Specifies instruction scheduling. I peephole__Specifies_peepholing.____________________________ A Specifying /OPTIMIZE with no options is the same asp7 specifying /OPTIMIZE=(peephole,schedule).o G The default qualifier is /NOOPTIMIZE. See Section B.7 for,+ information on optimizations. " /[NO]SHOW=(item,...) A Modifies the output listing file. This qualifier isyG meaningful only when /LIST is specified. Valid items are:t I ___________________________________________________________aI Option___________Function__________________________________ 6 expansions Shows macro expansions. C conditionals Shows sections of code conditionally ' skipped.t 8 include Shows all .INCLUDE files. I library__________Shows_macro_library_modules.______________o 5 The default item is /SHOW=conditionals. ) /[NO]WARNINGS=(option-list)o F Controls the severity level of messages and diagnostics. Valid options are: I ___________________________________________________________ I Option______Function_______________________________________ 4 warnings Display/suppress warnings. : informationaDisplay/suppress informationals. F all Display/suppress warnings and informational. I none________Display/suppress_nothing.______________________ I 1-7 a Introduction0 1.2 Invoking MACRO-64 on OpenVMS Systems The defaultnG qualifier is /WARNINGS=(warnings,informationals). If more G than one option is specified, options must be enclosed in / parentheses separated by a comma.o 1-8 I 2tI _________________________________________________________________ I Components of MACRO-64 Source Statementse ? A source program consists of a sequence of source I statements that the assembler interprets and processes, one H at a time, generating object code or performing a specificF assembly-time process. A source statement can occupy oneB source line or can extend onto several source lines. G This chapter describes the format of the source statement + and the following components:B o Character set o Numbers o Symbols o Local labelse & o Terms and expressions + o Unary and binary operatorse - o Direct assignment statementsB ) o Current location counter # 2.1 Source Statement Format C MACRO-64 source statements can have a maximum of four ! fields, as follows:t A o Label field-Symbolically defines a location in af program.s D o Operator field-Specifies the action to be performedF by the statement; can be an instruction, an assembler, directive, or a macro call. G o Operand field-Contains the instruction operands or the_F assembler directive arguments or the macro arguments. I 2-1 0 Components of MACRO-64 Source Statements# 2.1 Source Statement Formato C o Comment field-Contains a comment that explains theAB meaning of the statement; does not affect program execution.c D Although statement fields can be separated by either aE space or a tab stop, Digital recommends that you formatsC statements with the Tab key to insure consistency andl% clarity. See Table 2-1. I Table_2-1_Using_Tab_Stops_in_Statement_Fields______________ " Column in Which FieldI Field______Begins_____Tab_Stops_to_Reach_Column____________i % Label 1 0s % Operator 9 1 % Operand 17 2 I Comment____41_________5____________________________________ E The following example shows a typical source statement.t S EXP: .BLKL 50 ; Table stores expected valuesh 0 Rules for Coding Source Statements E The following rules apply for coding source statements: E o You can continue a single statement on several lineseE by using a hyphen (-) as the last nonblank character F before the comment field, or at the end of line (when& there is no comment). C o In most cases, a statement can be continued at any-C point. If a symbol name is continued and the first.B character on the second line is a tab or a blank,E the symbol name is terminated at that character. For ? information on using symbols, see Section 2.5. E o Blank lines are legal, but they have no significance C in the source program except that they terminate a continued line. C The following sections describe each of the statement. fields in detail.l 2-2l d c I Components of MACRO-64 Source Statements I 2.1 Source Statement Formati 2.1.1 Label Field I A label is a user-defined symbol that identifies a locationrE in the program. The symbol is assigned a value equal totH the location counter where the label occurs. The followingD rules apply when coding source statements with labels: F o If a statement contains a label, the label must be in- the first field on the line. H o The user-defined symbol name can be up to 31 charactersI long and can contain any alphanumeric character, as well H as the underscore (_), dollar sign ($), and period (.) characters. E o If a label extends past column 7, Digital recommendsPG you place it on a line by itself so that the followingeG operator field can start in column 9 of the next line.t G o A label is terminated by a colon (:) or a double colons (::). I In the following source statement, EXP: is the label field.c S EXP: .BLKL 50 ; Table stores expected values B See Section 2.5.2 for a description of the rules for0 forming user-defined symbol names. . There are three types of labels: H o Global labels-Can be referenced by other object modules; and are defined using a double colon (::).R G o Local labels-Can be referenced only within the current A module and are defined using a single colon (:).e C o Temporary labels-Can be referenced only within thefF bounds of either two local labels, two global labels,@ two psects, or within the bounds of a temporaryC label scope, as defined by .ENABLE LOCAL_BLOCK andtC .DISABLE LOCAL_BLOCK. Temporary labels are defined D using one to five digits, followed by a dollar signE and a single colon ($:). See Section 2.6 for further information.o I 2-3 e e 0 Components of MACRO-64 Source Statements# 2.1 Source Statement Format S EXP: .BLKL 50 ; EXP is a local label used topZ ; identify a 50-word block of storageU DATA:: .BLKW 25 ; DATA is a global label used to Z ; identify a 25-word block of storageW 10$: .BLKW 5 ; 10$ is a temporary label used to T ; identify a five word block of? ; storage. 2.1.2 Operator Field H The operator field specifies the action to be performed byF the statement. This field can contain an instruction, anG assembler directive, or a macro call. If the operator is: G o An instruction, MACRO-64 generates the binary code foruH that instruction in the object module. The instructions* are listed in Appendix A. G o A directive, MACRO-64 performs certain control actions I or processing operations during source program assembly. E The assembler directives are described in Chapter 5.e F o A macro call, MACRO-64 expands the macro. Macro callsD are described in Chapter 4 and in Chapter 5 (.MACROH directive). Macros supplied with MACRO-64 are described in Chapter 6. H Use either a space or a tab stop to terminate the operatorH field; however, Digital recommends you use the tab stop to+ terminate the operator field.b F In the following source statement, .BLKL is the operator field. S EXP: .BLKL 50 ; Table stores expected valuesd 2.1.3 Operand Field_ H The operand field can contain operands for instructions orG arguments for either assembler directives or macro calls. H Operands for instructions identify the memory locations orG the registers that are used by the machine operation. ThelG operand field for a specific instruction must contain the G number and type of operands required by that instruction.c I Arguments for a directive must meet the format requirementshG of that directive. Chapter 5 describes the directives andc, the format of their arguments. 2-4t l r I Components of MACRO-64 Source Statements I 2.1 Source Statement FormatO G Operands for a macro must meet the requirements specifiedsH in the macro definition. See the description of the .MACRO% directive in Chapter 5.c C Use a comma (,) to separate operands for instructionslB and directives. (See Section 2.8 for a discussion of expressions.)l H The semicolon that starts the comment field terminates theI operand field. If a line does not have a comment field, thelA operand field is terminated by the end of the line.s B In the following source statement example, 50 is theA operand field supplied to the operand field, .BLKL. S EXP: .BLKL 50 ; Table stores expected values2 2.1.4 Comment Fieldy H The comment field contains text that explains the function@ of the statement. You can use comments to describeC algorithms, reasons for selecting particular methods, G and parameters passed to routines. Comments do not affect 7 assembly processing or program execution.M G The comment field must be preceded by a semicolon (;) and4I can contain any printable ASCII character. It is terminatedB% by the end of the line.B I A comment can appear on a line by itself. If a comment doesrG not fit on one line, it can be continued on the next, but E the continuation must be preceded by another semicolon. F In the following source statement example, "Table storesC expected values", is the comment field. Note that theu4 comment field begins with a semicolon. S EXP: .BLKL 50 ; Table stores expected valuess 2.2 Character Sety D When coding source statements, you need to be aware ofF what characters are acceptable to the assembler, and howF the assembler interprets them. The following numbers and7 characters are accepted by the assembler.b C o The letters of the alphabet, A to Z, uppercase andaB lowercase. By default, the assembler converts all I 2-5 f 0 Components of MACRO-64 Source Statements 2.2 Character Seti H lowercase letters to uppercase. This means it considersC lowercase letters equivalent to uppercase letters. D The assembler can operate in a case-sensitive mode.G In case sensitive mode, the assembler does not convert G lowercase letters to uppercase letters. On OpenVMS and I OpenVMS AXP systems, case sensitive mode can be selectedeG from the command line with the /NAMES=as_is qualifier.s # o The digits 0 to 9.M < o The special characters listed in Table 2-2. 2-6e a e I Components of MACRO-64 Source Statements_I 2.2 Character Set_ I Table_2-2_Special_Characters_Used_in_MACRO-64_Statements___e " CharacterI ___CharacteName________Function____________________________t ? _ Underscore Character in symbol names.s ? $ Dollar Character in symbol names._ sign_ G . Period Character in symbol names, currentmI location counter, and decimal point.l < : Colon Local label terminator. = :: Double Global label terminator. colon I = Equal sign Local direct assignment operator andfG macro keyword argument terminator.f G == Double Global direct assignment operator.s# equal signj 6 Tab Field terminator. 6 Space Field terminator. = # Number Literal value indicator.J sign ? @ At sign Arithmetic shift operator.) H , Comma Field, operand, and item separator. = ; Semicolon Comment field indicator. H + Plus sign Unary plus operator, and arithmetic7 addition operator. E - Minus sign Unary minus operator, arithmetic_C or subtraction operator, and line_< hyphen continuation indicator. H * Asterisk Arithmetic multiplication operator. B / Slash Arithmetic division operator. : & Ampersand Logical AND operator. I ! Exclamation Logical inclusive OR operator point./ point I (continued on next page)e I 2-7 0 Components of MACRO-64 Source Statements 2.2 Character Set_ C Table 2-2 (Cont.) Special Characters Used in MACRO-64 I __________________Statements_______________________________ " CharacterI ___CharacteName________Function____________________________ E \ Backslash Logical exclusive OR and numeric B conversion indicator in macro/ arguments. G ^ Circumflex Unary operators and macro argumentG/ delimiter.v D ( ) Parentheses Displacement and register fieldI delimiter in an instruction operand._D Argument delimiter to a lexical. operator. D <> Angle Argument or expression grouping0 brackets delimiters. E ? Question Created local label indicator inp5 mark macro arguments. A ' Apostrophe Macro argument concatenation / indicator.C > " Double Quoted literal delimiter. quote H % Percent Delimits the beginning of a lexical. sign operator. G (space) Space or Separates source statement fields. B tab Spaces within expressions are7 (tab) otherwise ignored. A , Comma Separates symbolic arguments_G within the operand field. MultipleCE expressions in the operand fieldoI _______________________must_be_separated_by_commas.________s 2.3 Numbersp D Numbers can be integers or floating-point numbers. TheA following sections describe these types of numbers.t 2-8y e I Components of MACRO-64 Source Statements I 2.3 Numbersn 2.3.1 Integers ? Integers can be used in any expression, including > expressions in operands and in direct assignment= statements. For information on expressions, seed Section 2.8. Format snn seG An optional sign: plus sign (+) for positive numbers (the ? default), or minus sign (-) for negative numbers. nnB A string of numeric characters that is legal for the specified radix. G MACRO-64 interprets all integers in the source program asmF decimal unless the number is preceded by a radix controlH operator. For more information on radix control operators, see Section 2.9.1. I Integers must be in the range of -263 to +263-1 for signed D data or in the range of 0 to 264-1 for unsigned data. G Negative numbers must be preceded by a minus sign; MACRO-mG 64 translates such numbers into two's complement form. Inf: positive numbers, the plus sign is optional. $ 2.3.2 Floating-Point Numbers I A floating-point number can be used in the .DOUBLE, .FLOAT,fE .F_FLOATING, .D_FLOATING, .G_FLOATING, .S_FLOATING, andDI .T_FLOATING directives. For more information on directives,eF see Chapter 5. A floating-point number cannot be used inI an expression or with a unary or binary operator except the F unary plus and unary minus. For information on unary and= binary operators, Section 2.9 and Section 2.10._ I A floating-point number can be specified with or without an exponent.r I 2-9 o n 0 Components of MACRO-64 Source Statements 2.3 Numberso Format 5 Floating-point number without exponent: snn snn.nn snn. 2 Floating-point number with exponent: snnEsnn snn.nnEsnn snn.Esnn si An optional sign. nn@ A string of decimal digits in the range of 0 to 9. G The decimal point can appear anywhere to the right of the I first digit. Note that a floating-point number cannot start I with a decimal point because MACRO-64 will treat the numbernG as a user-defined symbol. For information on user-definedb) symbols, see Section 2.5.2.a 2.4 Quoted Literalst D A quoted literal is a string of characters enclosed inD double quotes (" "). Use the following guidelines when8 specifying characters in a quoted literal: E o Any character except null, carriage return, and form 3 feed can appear within the string. H o To include a double quote or backslash in a string, you6 must precede it with a backslash (\). C o To specify an arbitrary character, you can specifytE "\xhh", where each h represents a single hexadecimalu0 digit. For example, the string: ! "AB\\CD\"EF\x47"l 3 contains the following characters:i AB\CD"EFG G Also note that the assembler does not convert the case of < alphabetic characters within a quoted literal. 2-10 b n I Components of MACRO-64 Source StatementsiI 2.4 Quoted Literalso F Quoted literals can be continued over several lines. UseC the hyphen (-) as the line continuation character and.A delimit the string with double quotes. For example:m @ .ASCII "Strings are delimited with double-quotes."< .ASCII "The backslash is an escape character."S .ASCII "Use two backslashes (\\) to represent the back-slash itself."aV .ASCII "Hexidecimal escape sequences use lower or upper X: \x00 or \X00"W .ASCII "Precede a double quote with a backslash (\") to embed the quote."bD .ASCII "Strings can be continued onto multiple lines -& just as any other line." 2.5 Symbolss H You use symbols in MACRO-64 source statements to representG an instruction, directive, register name, or value. Three G types of symbols can be used in MACRO-64 source programs:cG permanent symbols, user-defined symbols, and macro names.d 2.5.1 Permanent Symbolsc H Permanent symbols consist of instruction mnemonics, MACRO-C 64 directives, and register names. You do not need tolI define directives before you use them in the operator field D of an MACRO-64 source statement. Also, you do not needG to define instruction mnemonics and register names before 7 using them in the instruction statements.a A You can express the 32 general registers and the 32 F floating point registers of the Alpha AXP processor in a( source program as follows: I ___________________________________________________________ RegisterI Name____Description________________________________________o ) R0 General register 0. ) R1 General register 1.r . . . . . .a I 2-11; e f 0 Components of MACRO-64 Source Statements 2.5 Symbols I ___________________________________________________________ RegisterI Name____Description________________________________________ F R29 or General register 29 or frame pointer. If you useI FP R29 as a framer pointer, Digital recommends you usenH the name FP. If you use R29 as a general register,> Digital recommends you use the name R29. F R30 or General register 30 or stack pointer. If you useI SP R30 as a stack pointer, the name SP is recommended;aH if you use R30 as a general register, the name R30% is recommended. * R31 General register 31. 0 F0 Floating point register 0. . .a . . . .a I F31_____Floating_point_register_31.________________________n B Registers are reserved names which cannot be used or' redefined in any context.i 2 2.5.2 User-Defined Symbols and Macro Names E You can use symbols to define labels, or you can equate H them to a specific value by a direct assignment statement.C (See Section 2.11.) These symbols can also be used inhC expressions. For more information on expressions, see Section 2.8. E Use the following rules to create user-defined symbols:h E o Use alphanumeric characters, underscores (_), dollarrA signs ($), and periods (.). Any other character ' terminates the symbol.t D o The first character of a symbol cannot be a number. G o The symbol must be no more than 31 characters long andc must be unique. 8 o The symbol must not be a register name. 2-12 u I Components of MACRO-64 Source StatementsiI 2.5 Symbolso I o The symbol cannot be one of the following conditional or " macro directives: 8 .ELSE .ENDC .ENDM C .ENDR .IF .IF_FALSE (.IFF)e 7 .IF_TRUE (.IFT) .IF_TRUE_FALSE .IIFd) (.IFTF)O 8 .INCLUDE .IRP .IRPC 9 .LIBRARY .MACRO .MCALL 8 .MDELETE .MEXIT .NARG ) .NCHAR .REPEAT 1 In addition, by Digital convention:s E o The dollar sign ($) is reserved for names defined bydE Digital. This convention ensures that a user-definedeA name (that does not have a dollar sign) will notsE conflict with a Digital-defined name (that does have a dollar sign). D o Do not use the period (.) in any global symbol nameF because many languages, such as FORTRAN, do not allow) periods in symbol names.a B Macro names follow the same rules and conventions asE user-defined symbols. See the description of the .MACROpI directive in Chapter 5 for more information on macro names.nH User-defined symbols and macro names do not conflict; thatI is, the same name can be used for a user-defined symbol andh a macro. ' 2.5.3 Determining Symbol Values F The value of a symbol depends on its use in the program.I MACRO-64 uses a different method to determine the values of I symbols in the operator field than it uses to determine the 5 values of symbols in the operand field., I 2-13 u 0 Components of MACRO-64 Source Statements 2.5 SymbolsB 3 2.5.3.1 Using Symbols in the Operator Fielde F A symbol in the operator field can be either a permanentD symbol or a macro name. MACRO-64 searches for a symbol0 definition in the following order: 1. Register names: R0-R31 FP SP F0-F31 2 2. Macro and conditional directives: 4 .ELSE .ENDC .ENDM ? .ENDR .IF .IF_FALSE (.IFF) 3 .IF_TRUE .IF_TRUE_FALSE .IIF4$ (.IFT) (.IFTF) 4 .INCLUDE .IRP .IRPC 5 .LIBRARY .MACRO .MCALLe 4 .MDELETE .MEXIT .NARG $ .NCHAR .REPEAT / 3. Previously defined macro namesi F 4. Permanent symbols (instructions and other directives) E This search order allows most permanent symbols, exceptsH registers, conditional directives and macro directives, toF be redefined as macro names. If a symbol in the operatorH field is not defined as a macro or a permanent symbol, the2 assembler displays an error message. 2 2.5.3.2 Using Symbols in the Operand Field I A symbol in the operand field must be either a user-defined : numeric symbol, a label, or a register name. I User-defined numeric symbols and labels can be either localsF (internal) symbols or global (external) symbols. WhetherG numeric symbols and labels are local or global depends on . their use in the source program. D A local numeric symbol or label can be referenced onlyD in the module in which it is defined. If local numericB symbols or labels with the same names are defined inF different modules, the symbols and labels are completely 2-14 i o I Components of MACRO-64 Source Statements I 2.5 Symbolsi G independent. The definition of a global numeric symbol ormF label, however, can be referenced from any module in the program. I MACRO-64 treats all user-defined numeric symbols and labelseI as local unless you explicitly declare them to be global by ) doing one of the following:l H o Use the double colon (::) in defining a label. For more: information on labels, see Section 2.1.1. F o Use the double equal sign (==) in a direct assignmentE statement. For more information on direct assignment+. statements, see Section 2.11. E o Use the .WEAK directive. For more information on the , .WEAK directive, see .WEAK. G User-defined lexical string symbols can only be used withoC the lexical string operators. For more information onaE lexical string operators, see Chapter 3. You can defineoG a macro using the same name as a previously defined localaH numeric symbol, global numeric symbol, or a lexical stringH symbol. However, you cannot define a lexical string symbol7 and a numeric symbol using the same name. H In addition, you cannot use the same name for both a localH and global numeric symbol. Nor can you use the same symbolD name for both a numeric symbol (local or global) and a& label (local or global). / 2.6 Temporary Labels Within Source Code_ G Use temporary labels to identify addresses within a block of source code. Format nnnnn$:_ nnnnn ; A decimal integer in the range of 1 to 65535. I In most cases, you can use temporary labels in the same way F you use other labels that you define; however, there are some differences:o H o Temporary labels cannot be referenced outside the block; of source code in which they are declared. I 2-15 0 Components of MACRO-64 Source Statements/ 2.6 Temporary Labels Within Source Code G o Temporary labels can be redeclared in another block of source code.l D o Temporary labels that occur within a psect with theF MIX or NOMIX attribute, do not appear in the debuggerI symbol table and thus cannot be accessed by the symbolic I debugger. For more information on psects, see Chapter 5. ? o Temporary labels cannot be used in the .END or G .PROCEDURE_DESCRIPTOR directives. For more information 6 on the .END directive, see Chapter 5. A By convention, temporary labels are positioned like B statement labels: left-justified in the source text.D Although temporary labels can appear in the program inI any order, by convention, the temporary labels in any block C of source code should be in increasing numeric order. B Temporary labels are useful as branch addresses whenD you use the address only within the block. You can useH temporary labels to distinguish between addresses that areI referenced only in a small block of code and addresses thattG are referenced elsewhere in the module. A disadvantage oftI temporary labels is that their numeric names do not provide G any indication of their purpose. Consequently, you should G not use temporary labels to label sequences of statementsrG that are logically unrelated; user-defined symbols shouldo be used instead. C Digital recommends that users create temporary labels E only in the range of 0$ to 29999$ because the assembler D automatically creates temporary labels in the range ofI 30000$ to 65535$ for use in macros. For more information oni0 temporary labels, see Section 4.7. E The temporary label block in which a temporary label is = valid is delimited by the following statements:p 7 o A user-defined label: global or local.e G o A .PSECT directive. For more information on the .PSECT * directive, see Chapter 5. D An example showing the correct and incorrect use of* temporary labels follows: 2-16 e m I Components of MACRO-64 Source StatementssI 2.6 Temporary Labels Within Source Codes & A: ADDQ R1, R2, R37 BEQ R3, 2$ ; correct use & MULQ R2, R3, R4I 2$: ADDQ R1, R4, R5 ; definition of temporary label 3 B: BNE R5, 2$ ; illegali& C: SUBQ R2, R4, R6 E In this example, 2$ is a temporary label, defined inrF the block between A: and B:. The forward reference inF the second instruction is properly resolved. The lineC labeled B: also references 2$, but the label B hastE already closed the range. The temporary label 2$ can F be used later in the program, but its definition must; appear within the same block as the label.t H o The .ENABLE and .DISABLE directives, which can extend aH local label block beyond user-defined labels and .PSECTD directives. For more information on the .ENABLE and4 .DISABLE directives, see Chapter 5. G A temporary label block is usually delimited by two user-oH defined labels. However, the .ENABLE LOCAL_BLOCK directiveC starts a local block that is terminated by one of thea following: 7 o A second .ENABLE LOCAL_BLOCK directiveu E o A .DISABLE LOCAL_BLOCK directive followed by a user-i4 defined label or a .PSECT directive @ Temporary labels can be preserved with the context> of the program section in which they are defined? for future reference. For more information on the F .SAVE_PSECT [LOCAL_BLOCK] and .RESTORE_PSECT directives, see Chapter 5. 2.7 Label Addresses ? In the absence of optimization and automatic data H alignment, label addresses are defined to be the psect andG offset of the current location counter (see Section 2.11)cB at the point where the label is defined. When eitherG optimization or automatic data alignment are enabled, theeH following additional considerations apply. See Section B.7G for information on optimizations and automatic alignment. I 2-17o t h 0 Components of MACRO-64 Source Statements 2.7 Label Addressesi ? 2.7.1 Label Addresses, Optimization, and Code Alignment F Optimization and code alignment can affect the addressesD assigned to labels defined in psects that have the EXEG and NOMIX attributes. Optimization and code alignment are H disabled by default, and can be enabled with the /OPTIMIZEH and /ALIGNMENT command-line qualifiers (see Section 1.2.1)G and the .ENABLE directive (see Section 5.1) . In general,"G the assembler assigns the psect and offset of the currentkD location counter prior to optimization or alignment ofD code labels. However, the assembler adjusts referencesD to labels in branch instructions to the address of theI label after optimization and code alignment processing. ThetF assembler does not adjust references to labels where theH label reference occurs in an expression with more than one; term. The following example illustrates this.e - .psect CODE, exe, nomixsU BSR R0, 10$ ; R0 -> 10$ (post-optimization)eT 10$: LDA R1, 20$-10$(R0) ; R1 -> 20$ (pre-optimization)" JMP (R1) [...] 20$: D In the example above, the assembler adjusts the targetF address of the BSR instruction to be the location of 10$E after optimization and code alignment have taken place.uE Thus, the branch to 10$ functions as expected. However, I when processing the LDA instruction, the assembler computeslF the offset between 20$ and 10$ prior to optimization andH code alignment. Thus, the address of 20$ that is stored inI R1 is the address prior to optimization and code alignment. G Depending on the sequence of instructions between 10$ and_E 20$, the address before optimization and code alignment E may differ from the address after optimization and code H alignment and the JMP instruction may not transfer control& to the expected address. > Note also that the assembler only performs post-G optimization adjustment of label addresses when the label_> is the only term in the expression. For example: 2-18 c t I Components of MACRO-64 Source StatementsiH 2.7 Label Addresses - .psect CODE, exe, nomix ' .base R27,LINKAGEa0 LDQ R26, ROUTINE1_ADDR( JSR R26, (R26)0 LDQ R26, ROUTINE2_ADDR( JSR R26, (R26)! RET R28r NOP NOP ROUTINE1:R# RET (R26)d ROUTINE2: # RET (R26) + .psect LINKAGE, noexe LINKAGE: ROUTINE1_ADDR:' .address ROUTINE1_ ROUTINE2_ADDR:) .address ROUTINE2+0 E In the example above, the assembler adjusts the addressiH stored with the .ADDRESS ROUTINE1 directive to the addressF of label ROUTINE1 after optimization and code alignment.F However, since the expression in the .ADDRESS ROUTINE2+0D directive is not a single term, the assembler adds theH offset 0 and the address of ROUTINE2 prior to optimizationI and code alignment and stores the result. Since the addresseB stored is the address prior to optimization and codeD alignment, the second JSR instruction may not transfer6 control to the address that is expected. : 2.7.2 Label Addresses and Automatic Data Alignment ? Automatic data alignment can affect the addressesoE assigned to labels in psects that have the NOEXE or MIX B attributes. Automatic data alignment is enabled withE the .ENABLE ALIGN_DATA directive or the /ALIGNMENT=dataOD command line option. For more information on automaticE alignment, see Section 5.2. For more information on the 1 .ENABLE directive, see Section 5.3. I 2-19 0 Components of MACRO-64 Source Statements 2.7 Label AddressesF C A label that occurs in a statment with a data storage.G directive is assigned the psect and offset of the storage B allocated by the data storage directive. If the dataI storage directive requires automatic alignment, the address A is assigned to the label after automatic alignment.n D The same is true of labels that occur in statements byE themselves and that are followed by a data directive iniC a subsequent statement. However, if a label occurs inoC a statement by itself and is followed by an assemblerhE directive that is not a data storage directive, a macrohA directive, or a conditional directive, the label isnC assigned the psect and offset of the current locationi7 counter prior to any automatic alignment.s F Note that the assembler only assigns addresses to labelsB after alignment under the conditions described aboveE and only with automatic alignment. If you place a labelaH before a .ALIGN directive that manually aligns the currentD location counter, the assembler assigns the address ofE the label prior to performing the manual alignment. ThemD following example illustrates the interaction of label> address assignment and automatic data alignment: & .enable align_data% .psect data,noexe X .byte 1 ; The byte is stored in psect data at offset 0U A: .print "Not aligned" ; Any non-macro, non-conditional statement, T ; including this .PRINT directive preventsA ; A from being automatically aligned -- A is = ; assigned offset 1bR B: ; B is automatically aligned to offset 4R C: .long 2 ; C is automatically aligned to offset 4] D: .align 0 ; The .ALIGN global directive prevents D from being \ ; automatically aligned -- D is assigned offset 8S E: .octa 3 ; E is automatically aligned to offset 16 C Automatic data alignment and label-address assignment D can be an important consideration when calculating theF difference between two labels. For example, consider theG following macro, which stores a string preceded by a word 0 that contains the string's length: 2-20 I Components of MACRO-64 Source Statements H 2.7 Label Addresses ; .macro VARYING_STRING STRING ?L1, ?L2 ) .word L2-L1p- L1: .ASCII "STRING"i L2: * .ENDM VARYING_STRING I If an invocation of the VARYING_STRING macro is followed byeI a data storage directive that requires automatic alignment, H the VARYING_STRING macro will not store the correct string" length. For example: . .psect DATA, noexe, octa( .enable align_dataK VARYING_STRING <Time for something sweet!> ; 25 byteso F: .octa 4 D In this example, the intention is to make the L2 labelC generated by the VARYING_STRING macro be assigned the C offset 27, 2 for the .WORD directive, plus 25 for thelB .ASCII directive. Instead, it is assigned the offsetC 32 along with the label F because the .OCTA directive C requires automatic alignment to offset 32. Therefore, F the string length is incorrectly stored as 30 rather 25.C To make this macro work as desired, you must include, F in the macro definition, a macro directive that is not aE data storage, macro, or conditional directive after thetD generated label L2. As shown below, .ALIGN 0 is a goodH choice as it reflects the idea that the preceding label is not aligned. 4 .macro VARYING_STRING STRING ?L1, ?L2 .word L2-L1=& L1: .ASCII "STRING" L2: .align 0n# .ENDM VARYING_STRING B With this change, the generated label L2 is assignedC the offset 27 before the assembler performs automaticeB data alignment for the .OCTA directive. As a result,F the VARYING_STRING macro works as desired and stores the* correct string length of 25. I 2-21e m s 0 Components of MACRO-64 Source Statements! 2.8 Terms and Expressionsi ! 2.8 Terms and Expressionse 1 A term can be any of the following: o A number. " o A numeric symbol. o A label. F o The current location counter. For more information on@ the current location counter, see Section 2.12. F o Any of the previously noted items preceded by a unary operator. G For more information on unary operators, see Section 2.9.e G MACRO-64 evaluates terms as quadword (8-byte) values. ThetH current location counter (.) has the value of the location: counter at the start of the current operand. I MACRO-64 considers unary operators part of a term and thus,sI performs the action indicated by a unary operator before iteA performs the action indicated by a binary operator.o D Expressions are combinations of terms joined by binaryF operators and evaluated as quadword (8-byte) values. ForE more information on binary operators, see Section 2.10.mD MACRO-64 evaluates expressions from left to right withC no operator precedence rules. However, angle brackets E (<>) can be used to change the order of evaluation. AnyrF part of an expression that is enclosed in angle bracketsF is first evaluated to a single value, which is then usedE in evaluating the complete expression. For example, themG expressions A*B+C and A*<B+C> are different. In the firstnB case, A and B are multiplied and then C added to theD product. In the second case, B and C are added and theE sum is multiplied by A. Angle brackets can also be used H to apply a unary operator to an entire expression, such as -<A+B>. A Expressions fall into four categories: relocatable, I absolute, external (global), and complex. You can determine < the type of expression by the following rules: C o An expression is relocatable if its value is fixed F relative to the start of the program section in which 2-22 e t I Components of MACRO-64 Source StatementslI 2.8 Terms and Expressionsi H it appears. The current location counter is relocatable2 in a relocatable program section. G o An expression is absolute if its value is an assembly- A time constant. An expression whose terms are allhF numbers is absolute. An expression that consists of aE relocatable term minus another relocatable term fromtF the same program section is absolute, because such anA expression reduces to an assembly-time constant.e G o An expression is external if it is not complex, and itoI contains one or more symbols that are not defined in the current module. I o An expression is complex if it contains a relocatable orlI external term or sub-expression that is operated upon bylF an operator other than the plus sign (+) or the minusG sign (-). An expression is also complex if it contains I more than one term or sub-expression that is relocatableyI or external. An exception to this rule is the difference G between two relocatable sub-expressions or terms where I both relocatable values occur in the same psect. In this 2 case, the expression is absolute. F Complex expressions are constrained to use the following form: FORMAT term operator term term? Term is a relocatable or external sub-expression. operatorD Operator is any of the MACRO-64 operators described in Section 2.10.e D Neither term can itself be a complex subexpression. IfF you specify a complex expression that does not match theC correct form, the assembler issues a diagnostic errorlD message indicating that the expression is too complex.F Note also that the assembler does not attempt to reorderD expressions to make your expressions match the correctI form. For example, the following expression is too complex:P .external E1, E2F- .quad E1+5+E2+6 ; too complex I 2-23e l 0 Components of MACRO-64 Source Statements! 2.8 Terms and ExpressionsL H Because the assembler has no precedence rules, it attemptsH to evaluate the above expression as <<<E1+5>+E2>+6>. SinceH <<E1+5>+E2> is itself a complex term, <<<E1+5>+E2>+6> doesG not match the above form and is too complex. However, youeE can regroup the expression using angle brackets (<>) to 1 match the required form as follows:e .external E1, E2r> .quad <E1+5>+<E2+6> ; legal complex expression G In this case, both <E1+5> and <E2+6> are simple, external.H terms. Since none of the terms are complex, the expressionD matches the correct form and the assembler accepts the! complex expression. A Any type of expression can be used in most MACRO-64 D statements, but restrictions are placed on expressions- used in the following contexts:n ! o .BASE directive. 5 o .BLKx storage allocation directives.7 ; o .IF conditional assembly block directives.a 1 o .REPEAT repeat block directives. . o Direct assignment statements. H o Lexical string operator arguments. For more informationC on direct assignment statements, see Section 2.11.n H See Chapter 5 for descriptions of the directives listed in! the preceding list.d I Expressions used in these contexts can contain only symbolsiH or labels that have been previously defined in the current module.P B The .BASE directive accepts expressions that containE external symbols previously declared with the .EXTERNAL G directive. The other contexts previously described cannot G accept expressions that contain external symbols. SymbolsgG defined later in the current module cannot be used in anye of these contexts. D Expressions in the .IF conditional directives, .REPEATA conditional directives, and lexical string operatoreD arguments are relocatable. However, expressions in the0 .BLKx directives must be absolute. 2-24 e I Components of MACRO-64 Source StatementseI 2.8 Terms and Expressions A Note that expressions cannot contain floating-point A numbers. The floating-point data storage directivesRG accept constant values. They do not accept floating-pointtE expressions. For more information on the floating-points5 data storage directives, see Chapter 5. A The following example shows the use of expressions: I A = 2*100 ; 2*100 is an absolute expression L .BLKB A+50 ; A+50 is an absolute expression andI ; contains no undefined symbols < LAB: .BLKW A ; LAB is relocatableD HALF = LAB+<A/2> ; LAB+<A/2> is a relocatableF ; expression and contains no= ; undefined symbolsdL LAB2: .BLKB LAB2-LAB ; LAB2-LAB is an absolute expressionM ; and contains no undefined symbolsl 5 2.9 Unary Operators for Terms and Expressions C A unary operator modifies a term or an expression andaB indicates the action to be performed on that term orF expression. Expressions modified by unary operators mustH be enclosed in angle brackets. You can use unary operatorsB to indicate whether a term or expression is positiveC or negative. If unary plus or minus is not specified, E the default value is plus. In addition, unary operators I perform radix conversion and numeric control operations, as G described in the following sections. Table 2-3 summarizes(" the unary operators. I Table_2-3_Summary_of_Unary_Operators_______________________ Unary OperatorI OperatorName________Example___Operation____________________6 I + Plus sign +A Results in the positive value 1 of A.a C - Minus sign -A Results in the negativeaD (two's complement) value1 of A.h I (continued on next page) I 2-25e o n 0 Components of MACRO-64 Source Statements5 2.9 Unary Operators for Terms and Expressionsf I Table_2-3_(Cont.)_Summary_of_Unary_Operators_______________e Unary OperatorI OperatorName________Example___Operation____________________o D \ Value of \symbol Indicates that the valueC Escape of the symbol should be F used. In a string literal,I indicates an escape sequence.d8 For example:5 "Bob\X0A"i H ^A or ASCII ^A/ABCD/ Specifies an ASCII constant. ^a H ^B or Binary ^B11000111Specifies that 11000111 is a: ^b binary number. C ^D or Decimal ^D127 Specifies that 127 is a ; ^d decimal number. I ^O or Octal ^O34 Specifies that 34 is an octale3 ^o number.n D ^X or Hexadecimal ^XFCF9 Specifies that FCF9 is a? ^x hexadecimal number.r I ^C or Complement ^C24 Produces the one's complementcI ^c____________________________value_of_24_(decimal)._______m % 2.9.1 Radix Control Operatorse F Radix control operators determine the radix of a term orG expression. MACRO-64 accepts terms or expressions in foursI different radixes: binary, decimal, octal, and hexadecimal.l+ The default radix is decimal. Formats ! ^Bnn ^Dnn ^Onn ^Xnn nnC A string of characters that is legal in the specifiedtD radix. The following are the legal characters for each radix: 2-26 I Components of MACRO-64 Source Statements I 2.9 Unary Operators for Terms and Expressions I ___________________________________________________________eI Format__Radix_Name__Legal_Characters_______________________ ) ^Bnn Binary 0 and 1 ( ^Dnn Decimal 0 to 9 ( ^Onn Octal 0 to 7 I ^Xnn____Hexadecimal_0_to_9_and_A_to_F______________________d C Radix control operators can be included in the source H program anywhere a numeric value is legal. A radix controlF operator affects only the term immediately following it,G causing that term to be evaluated in the specified radix. For example: < .WORD ^B00001101 ; Binary radixG .WORD ^D123 ; Decimal radix (default) ; .WORD ^O47 ; Octal radix: I Do not place spaces or tabs between the circumflex (^), the D radix specifier (B, D, O, or X), or the numeric value. ) 2.9.2 Numeric Complement Operator H The complement operator (^C) produces the one's complement% of the specified value.i Format ^Cterm termG Any term or expression. If an expression is specified, it 1 must be enclosed in angle brackets. D MACRO-64 evaluates the term or expression as an 8-byte, value before complementing it. For example: > .LONG ^C^XFF ; Produces FFFFFF00 (hex)= .LONG ^C25 ; Produces complement of0: ; 25 (dec) which is7 ; FFFFFFE6 (hex) I 2-27c V I 0 Components of MACRO-64 Source Statements 2.10 Binary Operators 2.10 Binary Operators F In contrast to unary operators, binary operators specifyB actions to be performed on two terms or expressions.F Expressions can be enclosed in angle brackets to specifyF the order of evaluation. Table 2-4 summarizes the binary operators. I Table_2-4_Summary_of_Binary_Operators______________________ Binary OperatorI ___OperatoName__________Example_Operation__________________e 6 + Plus sign A+B Addition 9 - Minus sign A-B Subtraction < * Asterisk A*B Multiplication 6 / Slash A/B Division > @ At sign A@B Arithmetic shift C & Ampersand A&B Logical AND (product)t > ! Exclamation A!B Logical OR (sum) pointi I ___\______Backslash_____A\B_____Logical_XOR_(difference)___m @ All binary operators have equal priority. Terms orD expressions can be grouped for evaluation by enclosingH them in angle brackets. The enclosed terms and expressionsI are evaluated first, and remaining operations are performede. from left to right. For example: . .LONG 1+2*3 ; Equals 9. .LONG 1+<2*3> ; Equals 7 C Note that a 64-bit result is returned from all binary C operations. If you use the 64-bit result in a context D requiring less than 64 bits, only the lower order bitsE of the result are used. If the truncation causes a lossaH of significance in a data storage directive, the assembler( displays an error message. C The following sections describe the arithmetic shift, F logical AND, logical inclusive OR, and logical exclusive OR operators.R 2-28 n I Components of MACRO-64 Source Statements I 2.10 Binary Operators. ( 2.10.1 Arithmetic Shift Operator C Use the arithmetic shift operator (@) to perform left.G and right arithmetic shifts of arithmetic quantities. ThesF first argument is shifted left or right by the number ofD bit positions that you specify in the second argument.G If the second argument is positive, the first argument isRI shifted left and the low-order bits are set to zero. If thenH second argument is negative, the first argument is shiftedG right and the high-order bits are set to the value of thesB original high-order bit (the sign bit). For example: L .LONG ^B101@4 ; Yields 1010000 (binary)H .LONG 1@2 ; Yields 100 (binary) A = 4oJ .LONG 1@A ; Yields 10000 (binary)D .LONG ^X1234@-A ; Yields 123(hex) # 2.10.2 Logical AND Operatora G The logical AND operator (&) takes the logical AND of twod$ operands. For example: A = ^B1010 B = ^B1100D .LONG A&B ; Yields 1000 (binary) , 2.10.3 Logical Inclusive OR Operator E The logical inclusive OR operator (!) takes the logicalh8 inclusive OR of two operands. For example: A = ^B1010 B = ^B1100D .LONG A!B ; Yields 1110 (binary) , 2.10.4 Logical Exclusive OR Operator E The logical exclusive OR operator (\) takes the logicale9 exclusive OR of two arguments. For example:o A = ^B1010 B = ^B1100D .LONG A\B ; Yields 0110 (binary) I 2-29n 0 Components of MACRO-64 Source Statements) 2.11 Direct Assignment Statementsa ) 2.11 Direct Assignment Statementsr A A direct assignment statement equates a symbol to a F specific value. Unlike a symbol that you use as a label,H you can redefine a symbol defined with a direct assignment2 statement as many times as you want. Formatsn symbol=expression symbol==expression # symbol=quoted-literalo symbol$ A user-defined symbol. expressionG An expression that does not contain any undefined symbolsnF or forward references. For more information on terms andE expressions, see Section 2.8. The result must be eitheruB an absolute or relocatable value, whose value can beH determined at the current point in the assembly. This form' defines a numeric symbol.n E The format with a single equal sign (=) defines a localmB symbol, and the format with a double equal sign (==)G defines a global symbol. For more information about locals4 and global symbols, see Section 2.5.3. A The following three syntactic rules apply to directx$ assignment statements: B o An equal sign (=) or double equal sign (==) mustE separate the symbol from the expression that defineseD its value. Spaces preceding or following the directA assignment operators have no significance in the ! resulting value. B o Only one symbol can be defined in a single direct& assignment statement. H o A direct assignment statement can be followed only by a comment field. 2-30 i I Components of MACRO-64 Source Statements I 2.11 Direct Assignment Statementso G For best results, Digital recommends you place the symbol F in a direct assignment statement in the label field. For example: B A == 1 ; The symbol 'A' is globally@ ; equated to the value 1 A B = A@5 ; The symbol 'B' is equated ; ; to 1@5 or 32(dec) A C = 127*10 ; The symbol 'C' is equated 6 ; to 1270(dec) A D = ^X100/^X10 ; The symbol 'D' is equated 4 ; to 10(hex) quoted-literalI A literal within double quotes. This form defines a lexical I string symbol. Lexical string symbols can only be used witht' lexical string operators.. C For more information on lexical string operators, see> Chapter 3. % 2.12 Current Location Counter. B The current location counter is a counter kept by an? assembler to determine the address assigned to ansB instruction or constant that is being assembled. TheF symbol for the current location counter, the period (.),G represents the address of the current byte. MACRO-64 setsfC the current location counter to zero at the beginningrF of the assembly and at the beginning of each new program section. F Every MACRO-64 source statement that allocates memory inC the object module increments the value of the currenthD location counter by the number of bytes allocated. ForE example, the directive, .LONG 0, increments the currentFG location counter by 4. However, with the exception of theaI special form described below, a direct assignment statement G does not increase the current location counter because no " memory is allocated. I 2-31o h i 0 Components of MACRO-64 Source Statements% 2.12 Current Location Counters C The current location counter can be explicitly set byeD a special form of the direct assignment statement. TheH location counter can be either incremented or decremented.I This method of setting the location counter is often usefuliF when defining data structures. Data storage areas shouldI not be reserved by explicitly setting the location counter;eH use the .BLKx directives. For more information on the .BLK' directive, see Chapter 5.i Format .=expression expressionB An expression that does not contain any undefined orD external symbols. For more information on expressions, see Section 2.8. F In a relocatable program section, the expression must beF relocatable; that is, the expression must be relative toG an address in the current program section. It also may benD relative to the current location counter. For example: > . = .+40 ; Moves location counter forward D When a program section that you defined in the currentF module is continued, the current location counter is setG to the last value of the current location counter in thats program section. E In a program section with the EXE and NOMIX attributes:B 8 o The location counter cannot be changed. A o If optimization is enabled, the location counter H represents the location prior to optimization. For more@ information on optimization, see Section 2.7.1. 2-32 x I 3 I _________________________________________________________________ I MACRO-64 Lexical Operator D This chapter describes the MACRO-64 lexical operators. - 3.1 Processing with Lexical Operatorsn D Lexical operator processing is performed on all source@ lines and macro expansion lines prior to any otherF assembler processing. Thus, macro invocations, assemblerA directives, and instructions are subject to lexicaloG operator processing prior to normal assembler processing.d C Lexical operators are recognized and processed withinrH string literals. Lexical operator processing is suppressedF during macro registration in order that lexical operatorB evaluation may occur during macro expansion. LexicalC operator evaluation is also suppressed for a range of E text that is conditionally excluded with one of the .IF_I directives. In addition, lexical operator processing is not ) performed within a comment. # 3.2 Lexical Operator Syntax F You invoke a lexical string operator with a percent signH followed by the lexical operator name, a left parentheses,B a list of arguments separated by commas, and a rightH parentheses. The following example illustrates the lexical operator syntax: - .print "%EDIT(<Fred>,<upcase>)"n G Spaces are allowed between syntax elements in the general I lexical operator syntax. For example, the following syntax,s+ including spaces, is allowed.r 2 .print "%EDIT ( <Fred> , <upcase> )" I 3-1a t ! MACRO-64 Lexical Operator # 3.2 Lexical Operator Syntax E Spaces are also allowed between the opening and closing C percent signs in a lexical substitution operator. SeeeA Section 3.4 for information on lexical substitution operators. . .print "% lexical_symbol_name %" F Spaces are not allowed between the pair of percent signsG indicating a lexical escape operator. See Section 3.5 for 6 information on lexical escape operators. I Lexical operator arguments can be specified in the same way ! as macro arguments: H o A numeric symbol name preceded by a backslash (\). ThisF construct results in the decimal value of the numeric; symbol, as shown in the following example:h \N F o Any string of characters surrounded by left and rightC angle brackets, as shown in the following example: <Foo bar thud>a D You can nest angle brackets (<>). See the following example:o <<X+7>*17> C o Any string of characters surrounded by a delimitersG specified after a caret character (^). You cannot nesta7 delimiters. See the following example:m ^%Foo bar thud% H o Any undelimited string of characters not separated by aH space, tab, form feed, comma, equal sign, semicolon, or8 end of line. See the following example: A+B+C F In addition to the formats allowed for a macro argument,A lexical operator arguments may also be specified ase follows: E o An undelimited string of characters may also containeF a string of characters enclosed within left and rightG parenthesis. The characters between the left and right I parenthesis may contain space, tab, or comma delimiters. + See the following example:_ 3-2_ d I MACRO-64 Lexical Operator I 3.2 Lexical Operator Syntax 16( R27 ) I o You can use a lexical operator as an argument to anothertD lexical operator as shown in the following example: F %EXTRACT( %LOCATE($,X), %LENGTH(X) - %LOCATE($,X) ,X) I Except for the %TYPE lexical operator, a string symbol name F supplied as a lexical operator argument is replaced with- the value of the string symbol. G Each lexical operator accepts a given number of argumentsuD and each argument has a specific type. There are threeF different types of arguments: string, integer, and name. C o A string argument can be any arbitrary sequence of characters. G o An integer argument must be an absolute or relocatableiE expression that can be resolved at that point in theeF assembly. A relocatable expression represents a psectB and an offset within that psect. If you specify aD relocatable expression for an integer argument, theC assembler uses only the value of the offset withincA the psect. The offset value is determined before @ optimization and code alignment, but after data alignment. I o The name argument type is used only by the %TYPE lexical F operator. The %TYPE lexical operator accepts the nameE of a numeric symbol, string symbol, label, psect, or E a permanent symbol as its argument. Unlike the other H lexical operators, if a string symbol name is specifiedH as an argument to %TYPE, the value of the string symbolF is not substituted for its name. Instead, information, about the name is returned. E If you omit a string argument, the default is the empty E string. An empty string is a string with no characters.nC If you omit an integer argument or specify an illegal G expression, the default value is zero. The assembler doesiH not issue diagnostic messages for illegal expressions usedD as aguments to lexical operators. If you omit the nameF argument or specify an illegal name to the %TYPE lexical3 operator, %TYPE returns a zero value.k I 3-3e o h ! MACRO-64 Lexical Operator 6 3.3 Numeric Symbols and Lexical String Symbols 6 3.3 Numeric Symbols and Lexical String Symbols F Lexical string symbols are similar in concept and syntaxI to numeric symbols. MACRO-64 supports numeric symbols using # the following syntax:9 6 numeric_symbol_name = numeric_expression @ MACRO-64 supports lexical string symbols using the following syntax:t = string_symbol_name = "any string of characters" A The assembler differentiates between numeric symbolsI assignment and lexical string symbol assignment as follows:t G o In a lexical string symbol assignment, a quoted string.: literal must appear after the equal sign. I o A lexical string symbol value is specified by the quotedc string literal. G Note that the quotes are not included in the value of thetG lexical string symbol. You cannot use the same name for a = lexical string symbol, numeric symbol or label.i G Like numeric symbols, lexical string symbols are assemblypH time variables. Once you have assigned a string value to aG lexical string symbol, you can reassign a different valuee2 to the symbol later in the assembly. H You can use lexical string symbols as arguments to lexicalD operators. In particular, you can use a lexical stringH symbol as an argument to the lexical substitution operatorC (%) or the %STRING lexical operator to substitute thedI value of the lexical string symbol at any point in the text of your program. ) 3.4 Lexical Substitution Operator H You can use the lexical substitution operator at any pointF in your program to cause the assembler to substitute theB value of a lexical string symbol for the name of theF symbol. The lexical substitution operator is the percentF sign (%). Place the lexical substitution operator to theE left and right of the name of the lexical string symbol 5 that you wish to subsitute, as follows: 3-4 N I MACRO-64 Lexical OperatorrI 3.4 Lexical Substitution Operatora %lexsym_name% For example: 6 HORSES = "All the king's horses"3 MEN = "all the king's men" . .print "%HORSES% and %MEN%" C The above example defines two lexical string symbols:hD HORSES and MEN. The third statement displays a messageF at assembly time. The text of the message specifies thatD the value of the HORSES and MEN lexical string symbolsG be subsituted as indicated. After lexical processing, the ) third statement appears as:t H .print "All the king's horses and all the king's men" # 3.5 Lexical Escape Operator H It is possible to defer the processing of a lexical stringI operator by using the lexical escape operator, which is theeH percent sign (%). Since all lexical string operators beginD with a percent sign, the effect of placing two percentI signs before the name of the lexical string operator deferslH the evaluation of the lexical string operator. If you wantE to defer processing of a lexical substitution operator, G place two percent signs to the left and two percent signsF= to the right of the lexical string symbol name.e B This can be useful when you want the evaluation of aE lexical string operator that you have used in a default I macro argument to occur during macro expansion, rather thanyE during macro definition. Lexical operator processing is F suppressed during macro registration. Therefore, lexicalF operator processing is automatically deferred within theF body of a macro. However, the .MACRO directive line thatF begins the macro definition is subject to normal lexicalD operator processing. Sometimes you may need to use theI value of a lexical string symbol as the default for a macrorD argument, but you need to use the value of the lexicalG string symbol that is current when the macro expands, notfH when the macro is defined. Example 3-1 shows an example of8 this, though not using an escape operator: I 3-5c b o ! MACRO-64 Lexical Operatort# 3.5 Lexical Escape Operator H Example 3-1 Lexical Processing Without the Escape Operator + CODE_PSECT_NAME = "CODE1"rG .macro CODE_PSECT PSECT_NAME=%string(CODE_PSECT_NAME)e$ .psect PSECT_NAME" .endm CODE_PSECT CODE_PSECT+ CODE_PSECT_NAME = "CODE2"1 CODE_PSECT F Example 3-1 does not process correctly for the following reasons: E o The lexical operator in the .MACRO directive line is H processed when the macro is defined, not when the macro expands.i @ o The CODE_PSECT macro always defaults to settingA the psect to the CODE1 psect because the default C for the PSECT_NAME argument will be set to CODE1, > not %string(CODE_PSECT_NAME). This is becauseE %string(CODE_PSECT_NAME) is evaluated when the CODE_n= PSECT macro is defined, not when it expands.o F Example 3-2 is similar to Example 3-1 except it uses the& lexical escape operator: A Example 3-2 Lexical Processing With Escape Operatora + CODE_PSECT_NAME = "CODE1" H .macro CODE_PSECT PSECT_NAME=%%string(CODE_PSECT_NAME)" .psect PSECT_NAME" .endm CODE_PSECT CODE_PSECT+ CODE_PSECT_NAME = "CODE2"t CODE_PSECT H Example 3-2 processes correctly for the following reasons: I o Lexical operator processing of %%string(CODE_PSECT_NAME)oF is deferred when the CODE_PSECT macro is defined. TheG default value for the PSECT_NAME argument is stored ass* %string(CODE_PSECT_NAME). A o During macro expansion, %string(CODE_PSECT_NAME)xD is evaluated, resulting in the current value of theB CODE_PSECT_NAME lexical string symbol as desired. 3-6o d r I MACRO-64 Lexical OperatoruI 3.6 Using Lexical Operators # 3.6 Using Lexical Operators H This section contains a programming example showing sourceH statements using lexical operators and the same statementsD after macro expansion and lexical operator processing. 1 Example 3-3 Using Lexical Operatorsn D ; Macro to load a general or floating register. .macro LOAD REG LOCATION L ; If the location doesn't specify a base register,L ; load relative to the procedure descriptor PD off# ; of R27. K .if lt, %locate(<(>, LOCATION), %length(LOCATION) 7 ; base register specified . LOC = "LOCATION" .elseh; ; base register not specifieds6 LOC = "LOCATION-PD(R27)" .endc F ; Use LDQ if a general register is specifiedD .if ne, <%type(REG) & MACRO64$TYPE_GENREG>4 LDQ REG,%string(LOC) I (continued on next page): I 3-7 i ! MACRO-64 Lexical Operatori# 3.6 Using Lexical Operatorsn 9 Example 3-3 (Cont.) Using Lexical Operatorst .else K ; Use LDx if a floating register is specified G .if ne <%type(REG) & MACRO64$TYPE_FLTREG> I ; Use the appropriate LDx floating load H ; depending on the first letter of the5 ; location to load..1 FTYPES = "FGST" N .if ge, %locate(%extract(0,1,LOC),FTYPES), -5 %length(FTYPES)_. .error -F "Unknown floating data type: %extract(0,1,LOC)", .mexit' .endc K LD%extract(0,1,LOCATION) REG,%string(LOC)o# .else * .error -N "First argument must be a general or floating register"# .endcs .endc .endm LOAD$ .psect D,noexe? pd: .procedure_descriptor my_routine,my_entry ? ; Fill in procedure descriptor details...d BUILD_TIME:o, .asciz "%time()"5 K: .quad %repeat(9,<42,>)42o* GPI: .g_floating 3.14159- CIRCLE: .address FAC_CIRCLE TX = 0l TY = 8 TRADIUS = 16" .psect X,exe MY_ENTRY: LOAD R1,K% LOAD R2,CIRCLE + LOAD F13,TRADIUS(R2) # LOAD F11,GPI ! LOAD F10,Kt LOAD K I (continued on next page) 3-8e l i I MACRO-64 Lexical Operator I 3.6 Using Lexical Operatorse 9 Example 3-3 (Cont.) Using Lexical Operatorst I 3-9 ! MACRO-64 Lexical Operator # 3.6 Using Lexical OperatorsA A Example 3-4 Source Statements After Macro Expansion $ .psect D,noexe? pd: .procedure_descriptor my_routine,my_entryc? ; Fill in procedure descriptor details...o BUILD_TIME:t9 .asciz " 9-JUL-1992 17:56:20"e@ K: .quad 42,42,42,42,42,42,42,42,42,42* GPI: .g_floating 3.14159- CIRCLE: .address FAC_CIRCLEo TX = 0n TY = 8 TRADIUS = 16" .psect X,exe MY_ENTRY: ( LDQ R1,K-PD(R27)- LDQ R2,CIRCLE-PD(R27)a+ LDT F13,TRADIUS(R2)u) LDG F11,GPI-PD(R27)r< .error "Unknown floating data type: K"T .error "First argument must be a general or floating register" 3-10 f o I MACRO-64 Lexical OperatorsI 3.7 Lexical Operators 3.7 Lexical Operators D This section describes the MACRO-64 lexical operators,G their return values, and their arguments. Table 3-1 lists I the lexicals and gives a brief summary of their function. ArG decimal return value results in the decimal digits of theaG numeric value being substituted for the lexical operator.c I Table_3-1_Summary_of_MACRO-64_Lexicals_____________________ ReturnI Lexical___Type______Function_______________________________a C %EDIT String Lexical operator for editing text * strings. D %ELEMENT String Lexical operator for extracting anB element from a list of elements. C %EXTRACT String Lexical operator for extracting anF range of characters from a string of- characters. G %INTEGER Decimal Lexical operator to convert the value F of an expression to a decimal value. F %LENGTH Decimal Lexical operator for determining the5 length of a string.x H %LOCATE Decimal Lexical operator for locating a stringD of text within a another string of' text. B %REPEAT String Lexical operator for repeating aE specified string x number of times. D %STRING String Lexical operator for obtaining theC value of a lexical string symbol. I %TIME String Lexical operator for obtaining the date @ and time of the assembly unit. @ %TYPE Decimal Lexical operator for obtainingI ____________________information_about_a_name.______________e I 3-11s r " MACRO-64 Lexical Operators %EDITt I _________________________________________________________________r %EDIT 8 Lexical operator for editing text strings. Format & %EDIT (string1,string2) Arguments string1nI The first argument, of type string, specifies the string to be edited. string2TF The second argument, of type string, specifies a list of4 edits to perform, separated by commas. Descriptionc C %EDIT is modeled after the OpenVMS DCL F$EDIT lexical D function. It is used to perform one or more edits on aG specified string. %EDIT processes the string of argumentsrD from left to right. %EDIT gives precedence to the last@ argument. %EDIT gives precedence to uppercase over lowercase. B The list of edits may contain any combination of the! following elements:c I ___________________________________________________________ I Element_____Function_______________________________________ 6 COLLAPSE Removes all tabs and spaces. I COMPRESS Replaces multiple tabs and spaces with a single space. D LOWERCASE Changes uppercase characters to lowercase. G TRIM Removes leading and trailing spaces and tabs. I UPCASE______Changes_lowercase_characters_to_uppercase._____ 3-12 b I MACRO-64 Lexical OperatorsgI %EDIT Examples Example 1 K .PRINT "%EDIT(< Fred Smith >, <TRIM,COLLAPSE,UPCASE>)" F After lexical processing, the statement apears as the following:u ' .PRINT "FREDSMITH" Example 2 @ .PRINT "%EDIT(<AbCdEfG>,<upcase,lowercase>)@ .PRINT "%EDIT(<AbCdEfG>,<lowercase,upcase>) I The first source statement produces the string "abcdefg"aD and the second source statement produces the stringA "ABCDEFG". Each of the edits in the edit list isr6 performed in sequence, left to right. I 3-13t g m " MACRO-64 Lexical Operators %ELEMENT I _________________________________________________________________o %ELEMENT E Lexical operator for extracting elements from a list ofd elements. Format 1 %ELEMENT (integer,string1,string2) Argumentsm integer.H The first argument, of type integer, is the element number; to extract. The first element is number zero.i string1 F The second argument, of type string, is the delimiter or0 delimiters that separate elements. string2u@ The third argument, of type string, is the list of elements. Descriptions G %ELEMENT is modeled after OpenVMS DCL's F$ELEMENT lexicaleG function. It is used to extract one element from a stringtF of elements. Note that unlike F$ELEMENT, you may specifyE multiple delimiters. The result is the specified stringsF element. If the specified element number is greater thanH the number of elements in the list, the delimiter argument is returned. Examples ; .PRINT "%ELEMENT (2, <+-*/>, JOE+FRED- TOM*BILL/ERIC)" A After lexical processing, the statement appears as: ! .PRINT "TOM"o 3-14 x I MACRO-64 Lexical OperatorshI %EXTRACT I _________________________________________________________________e %EXTRACT H Lexical operator for extracting a range of characters from% a string of characters.e Format 2 %EXTRACT (integer1,integer2,string) Argumentsa integer1I The first argument, of type integer, is the offset at which G to begin the extraction. The first character is at offsetn zero.e integer2D The second argument, of type integer, is the number of$ characters to extract. stringD The third argument, of type string, is the string from. which to extract the characters. Descriptiona I %EXTRACT is modeled after VAX MACRO's %EXTRACT macro stringtD operator and OpenVMS DCL's F$EXTRACT lexical function.I %EXTRACT is used to extract a specified range of charactersp from a string. Examples < .PRINT "%EXTRACT(3,4,ABCDEFGHIJKLMNOP)" A After lexical processing, the statement appears as:e " .PRINT "DEFG" I 3-15 d " MACRO-64 Lexical Operators %INTEGER I _________________________________________________________________a %INTEGER H Lexical operator for converting the value of an expression! to a decimal value.t Format ! %INTEGER (integer) ArgumentsM integer H The single argument, of type integer, is the expression to be converted.n Descriptione G %INTEGER is modeled after OpenVMS DCL's F$INTEGER lexical H function. It is used to convert the value of an expressionB to a decimal value. The result is its decimal value.H %INTEGER can also be used convert a relocatable expression# to an absolute value.h Examples 2 .PRINT "%INTEGER (<<X+7>*17>) A After lexical processing, if X has the value 3, the-( statement would appear as: ! .PRINT "170"x 3-16 E p I MACRO-64 Lexical OperatorscI %LENGTHc I _________________________________________________________________ %LENGTHO F Lexical operator for determining the length of a string. Format %LENGTH (string)g Argumentsn stringE The single argument, of type string, is the string fromp1 which the length is to be computed.c Descriptionp G %LENGTH is modeled after VAX MACRO's %LENGTH macro string C operator and OpenVMS DCL's F$LENGTH lexical function. F %LENGTH is used to determine the length of a string. TheG result is the length of the string expressed as a decimal number.A Examples < .PRINT "%LENGTH(<The quick brown fox>)" A After lexical processing, the statement appears as:x .PRINT "19" I 3-17 " MACRO-64 Lexical Operators %LOCATES I _________________________________________________________________ %LOCATEd C Lexical operator for locating a string of text withinT% another string of text.E Format ( %LOCATE (string1,string2) Arguments string1lI The first argument, of type string, is the string for whiche %LOCATE searches. string2 I The second argument, of type string, is the string in which & the search is performed. Descriptionx G %LOCATE is modeled after VAX MACRO's %LOCATE macro stringgC operator and OpenVMS DCL's F$LOCATE lexical function.lB %LOCATE is used to locate one string within another.C The value returned is the decimal offset to the first F occurrence of the first string within the second string.E The offset to the first character is zero. If the firstcI string cannot be found within the second string, the lengthg/ of the second string is returned. Examples < .PRINT "%LOCATE (DEF,ABCDEFGHIJKLMNOP)" A After lexical processing, the statement appears as:l .PRINT "3" 3-18 L I MACRO-64 Lexical OperatorseI %REPEATr I _________________________________________________________________i %REPEATL A Lexical operator for repeating a specified string ag( specified number of times. Format ' %REPEAT (integer,string) Arguments integerpI The first argument, of type integer, is the number of times H to repeat the string. If you specify a negative value, the, string is repeated zero times. stringF The second argument, of type string, is the string to be repeated. Description H %REPEAT is used to repeat the specified string a specified number of times. Examples L .PRINT "Never, %REPEAT (3, <ever, >)touch that button!" A After lexical processing, the statement appears as: I .PRINT "Never, ever, ever, ever, touch that button!" I 3-19p E & " MACRO-64 Lexical Operators %STRING I _________________________________________________________________e %STRINGl C Lexical operator for obtaining the value of a lexical string symbol. Format %STRING (string) Argumentsa stringD The single argument is of type string. If the argumentD is the name of lexical string symbol, the value of theH lexical string symbol is returned. Otherwise, the argument$ is returned unchanged. Description E %STRING is modeled after OpenVMS DCL's F$STRING lexical H function. %STRING is generally used to obtain the value ofI a lexical string symbol, but it can be used with any stringe argument. H The lexical substitution operator described in Section 3.4* performs a similar function. Examples 2 FOO = "All the king's horses"* .PRINT "%STRING(FOO)" A After lexical processing, the statement appears as:c 3 .PRINT "All the king's horses"% 3-20 : I MACRO-64 Lexical Operators I %TIME I _________________________________________________________________ %TIME E Lexical operator for obtaining the date and time of theD assembly unit. Format %TIME () Description A %TIME is modeled after OpenVMS DCL's F$TIME lexicaloD function. %TIME is used to obtain the time and date ofH the assembly unit. There are no arguments. The result is aG string specifying the date and time of the assembly unit. Examples % .PRINT "%TIME()" A After lexical processing, the statement appears as: 2 .PRINT " 8-OCT-1991 13:17:57" I 3-21r " MACRO-64 Lexical Operators %TYPEx I _________________________________________________________________u %TYPEr F Lexical operator for obtaining information about a name. Format %TYPE (name) Arguments nameA The single argument is of type name. Information is2@ returned about the name specified in the argument. DescriptionL A %TYPE is modeled after OpenVMS DCL's F$TYPE lexical C function. %TYPE is used to obtain information about aYF name. The value returned is a numeric value with certainE bit positions, either zero or one, depending on whetheruD the specified name has the corresponding attribute. AsC described elsewhere, he decimal digits of the numericiC value are substituted for the %TYPE lexical operator. D Table 3-2 shows the symbolic names that are predefined! for each attribute: I Table_3-2_%TYPE_Attributes_________________________________ I Symbolic_Name______________Attribute_______________________ G MACRO64$TYPE_SYMBOL Name is a numeric symbol name.i G MACRO64$TYPE_PROC_DESC Name is a procedure descriptore. name. 9 MACRO64$TYPE_LABEL Name is a label.i B MACRO64$TYPE_EXTERN Name is an external name. = MACRO64$TYPE_WEAK Name is a weak name.O 9 MACRO64$TYPE_PSECT Name is a psect. > MACRO64$TYPE_MACRO Name is a macro name. I (continued on next page)i 3-22 I MACRO-64 Lexical OperatorstI %TYPE I Table_3-2_(Cont.)_%TYPE_Attributes_________________________n I Symbolic_Name______________Attribute_______________________ H MACRO64$TYPE_STRING Name is a lexical string symbol. name. ; MACRO64$TYPE_OPCODE Name is an opcode. = MACRO64$TYPE_DIR Name is a directive.d D MACRO64$TYPE_GENREG Name is a general register. I MACRO64$TYPE_FLTREG________Name_is_a_floating_register.____ E A given name may have zero, one, or several attributes. Examples % .macro IS_GR ARG D .IF equal, %TYPE(ARG) & <MACRO64$TYPE_GENREG>> .PRINT "ARG is not a general register" .ENDC .endm IS_GR IS_GR F11 I Initially, the first line of the IS_GR macro expands as thet following: B .IF equal, <%TYPE(F11) & MACRO64$TYPE_GENREG> A After lexical processing, the statement appears as: < .IF equal, <8192 & MACRO64$TYPE_GENREG> I In this example, 8192 is the attribute value for a floating C point register. This value could change in subsequentII releases. Use only the predefined attribute masks described_F in Table 3-2. Since the attribute for a general registerF MACRO64$TYPE_GENREG is 4096, the expression evaluates as zero. 1 <8192 & MACRO64$TYPE_GENREG>r I 3-23 I 4 I _________________________________________________________________ I Macro Arguments G By using macros, you can use a single source statement toeD insert a sequence of source statements into a program. F A macro definition contains the source statements of theD macro. The macro definition may have formal arguments.H These formal arguments can be used throughout the sequenceH of source statements within the definition. When the macroH is called, the formal arguments are replaced by the actual. arguments within the macro call. G The macro call is a single source statement consisting of H the macro name, optionally followed by arguments. When theI macro is called, the assembler replaces the line containingcD the macro call with the source statements in the macroC definition. The assembler replaces any occurrences oftF formal arguments in the macro definition with the actualD arguments specified in the macro call. This process is) called the macro expansion. A By default, macro expansions are not printed in the G assembly listing. To print the macro expansions, you mustEH specify the /SHOW=expansions qualifier and argument in theF command line. Note that the examples of macro expansionsH used in this chapter are listed as they would appear using: the /SHOW=expansions argument and qualifier. A Use .SHOW with a symbolic argument of EXPANSIONS inuD the source text of a program to specify the listing of expansions.o C MACRO-64 provides specific directives for controllingeG macro execution. For information on these directives, see Table 5-2. I The remainder of this chapter describes macro arguments and ' created temporary labels.R I 4-1_ _ _ Macro Arguments_! 4.1 Using Macro Arguments ! 4.1 Using Macro Argumentsm D Macros have two types of arguments: actual and formal.I Actual arguments are the text given in the macro call after F the name of the macro. Formal arguments are specified byI name in the macro definition; that is, after the macro name F in the .MACRO directive. Actual arguments in macro callsH and formal arguments in macro definitions can be separated- by commas (,), tabs, or spaces. E The number of actual arguments in the macro call can benE less than or equal to the number of formal arguments in H the macro definition. If the number of actual arguments isH greater than the number of formal arguments, the assembler( displays an error message. D Formal and actual arguments normally maintain a strictI positional relationship. That is, the first actual argument C in a macro call replaces all occurrences of the first B formal argument in the macro definition. This strictE positional relationship can be overridden by the use of F keyword arguments. For more information on using keyword) arguments, see Section 4.3. E An example of a macro definition using formal argumentsa follows: + .MACRO STORE ARG1,ARG2,ARG3 G .LONG ARG1 ; ARG1 is first argument_G .WORD ARG3 ; ARG3 is third argumenttH .BYTE ARG2 ; ARG2 is second argument .ENDM STORE @ The following two examples show possible calls and9 expansions of the macro previously defined:r ; STORE 3,2,1 ; Macro call D .LONG 3 ; 3 is first argumentD .WORD 1 ; 1 is third argumentE .BYTE 2 ; 2 is second argumentc ; STORE X,X-Y,Z ; Macro callfD .LONG X ; X is first argumentD .WORD Z ; Z is third argumentG .BYTE X-Y ; X-Y is second argumente 4-2T T x I Macro ArgumentsaI 4.2 Using Default Values 4.2 Using Default Values E Default values are values that are defined in the macro B definition. They are used when no value for a formal6 argument is specified in the macro call. E Default values are specified in the .MACRO directive as follows: 2 formal-argument-name = default-value H An example of a macro definition specifying default values follows: 6 .MACRO STORE ARG1=12,ARG2=0,ARG3=1000 .LONG ARG1 .WORD ARG3 .BYTE ARG2 .ENDM STORE, B The following three examples show possible calls and9 expansions of the macro defined previously: = STORE ; No arguments suppliedt .LONG 12 .WORD 1000 .BYTE 0e C STORE ,5,X ; Last two arguments suppliedu .LONG 12 .WORD X .BYTE 5 ? STORE 1 ; First argument supplied7 .LONG 1 .WORD 1000 .BYTE 0 # 4.3 Using Keyword Arguments A Keyword arguments allow a macro call to specify the6G arguments in any order. In this case, the macro call musteG specify the same formal argument names that appear in the I macro definition. Keyword arguments are useful when a macro_H definition has more formal arguments than necessary in the call.e I 4-3 Macro Argumentsi# 4.3 Using Keyword Arguments, G In any one macro call, it is good practice to specify the I arguments as either all positional arguments or all keywordrD arguments. For example, the following macro definition( specifies three arguments: , .MACRO STORE ARG1,ARG2,ARG3 .LONG ARG1 .WORD ARG3 .BYTE ARG2 .ENDM STOREA C The following macro call specifies keyword arguments: 3 STORE ARG3=27+5/4,ARG2=5,ARG1=SYMBL .LONG SYMBL .WORD 27+5/4 .BYTE 5 G Because the keywords are specified in the macro call, the3H arguments in the macro call need not be given in the order7 they were listed in the macro definition._ E Positional and keyword arguments may be mixed. Usually,rG positional arguments are placed before keyword arguments:e 3 .MACRO STORE ARG1,ARG2,ARG3=27+5/4g .LONG ARG1 .BYTE ARG2 .WORD 27+5/4 .ENDM STOREe F ________________________ Note ________________________ A Keyword arguments are not counted when positionalgE arguments are parsed. This means that when positionaloA and keyword arguments are used in the same macro,cC one argument can be specified twice. The last value 3 specified for the argument is used.n F ______________________________________________________ 4-4 I Macro ArgumentsfI 4.4 Using String Argumentsn " 4.4 Using String Arguments E If an actual argument is a string containing charactersxD that the assembler interprets as separators (such as aC tab, space, or comma), the string must be enclosed bylC delimiters. String delimiters for macro arguments are B usually paired angle brackets (<>). A quoted literalC enclosed in double quotes ("") is also a valid string argument. 9 The assembler also interprets any character_F (except A, B, C, D, O, or X) after an initial circumflexG (^) as a delimiter. Note that ^B, ^D, ^O, and ^X are used I as radix control operators rather than argument delimiters.uD ^A is used as the ASCII operator and ^C is used as theH complement operator. To pass an angle bracket as part of aG string, you can use the circumflex form of the delimiter.z F The following are examples of delimited macro arguments: * <HAVE THE SUPPLIES RUN OUT?>" <LAB: CLR R4> F "A quoted literal is taken as a single parameter value." 2 ^%ARGUMENT IS <LAST,FIRST> FOR CALL%* ^?EXPRESSION IS <5+3>*<4+2>? H In the last two examples, the initial circumflex indicatesF that the percent sign (%) and question mark (?) are theC delimiters. Note that only the left-hand delimiter is ' preceded by a circumflex. D The assembler interprets a string argument enclosed byF delimiters as one actual argument and associates it withE one formal argument. If a string argument that contains E separator characters is not enclosed by delimiters, themH assembler interprets it as successive actual arguments and= associates it with successive formal arguments.n H For example, the following macro definition has one formal argument:f ' .MACRO DOUBLE_ASCII STRNGs .ASCII "STRNG"e .ASCII "STRNG"n" .ENDM DOUBLE_ASCII I 4-5 Macro Argumentsu" 4.4 Using String Arguments H The following two macro calls demonstrate actual arguments* with and without delimiters: & DOUBLE_ASCII <A B C D E>! .ASCII "A B C D E"l! .ASCII "A B C D E" % DOUBLE_ASCII A B C D EsE %MACRO64-E-TOOMNYARGS, Too many arguments in macro callf I Note that the assembler interprets the second macro call asRI having five actual arguments instead of one actual argument with spaces. B When a macro is called, the assembler removes normalG delimiters around a string before associating it with the_G formal arguments. However, a quoted literal within double H quotes is treated as a single token and retains its double quote delimiters. F If a string contains a semicolon (;), the string must beG enclosed by delimiters, otherwise the semicolon will mark D the start of the comment field. Further, if the stringG contains a semicolon, you cannot continue the line unless - the string is a quoted literal.a ? Macro invocations can be nested, that is, a macro%H definition can contain a call to another macro. If, withinG a macro definition, another macro is called and is passed F a string argument, you must delimit the argument so thatD the entire string is passed to the second macro as one argument. C The following macro definition contains a call to the_1 DOUBLE_ASCII macro defined earlier:p 6 .MACRO CNTDA LAB1,LAB2,STR_ARGJ LAB1: .BYTE LAB2-LAB1-1 ; Length of 2*stringO DOUBLE_ASCII <STR_ARG> ; Call DOUBLE_ASCII macro LAB2:t$ .ENDM CNTDA C Note that the argument in the call to DOUBLE_ASCII isDH enclosed in angle brackets even though it does not containF any separator characters. The argument is thus delimitedF because it is a formal argument in the definition of theF macro CNTDA and will be replaced with an actual argument4 that may contain separator characters. 4-6o h u I Macro Arguments I 4.4 Using String Arguments H The following example calls the macro CNTDA, which in turn+ calls the macro DOUBLE_ASCII:b . CNTDA ST,FIN,<LEARN YOUR ABC'S>& ST: .BYTE FIN-ST-15 DOUBLE_ASCII <LEARN YOUR ABC'S> 0 .ASCII "LEARN YOUR ABC'S"0 .ASCII "LEARN YOUR ABC'S" FIN: > In addition to nested macro invocations, you canB also nest macro definitions. That is, you can define< one macro within another. In this example, the< INNER_MACRO_DEF macro is not defined until the< OUTER_MACRO_DEF macro is invoked and expanded: , .macro OUTER_MACRO_DEF0 .macro INNER_MACRO_DEF! ... / .endm INNER_MACRO_DEF + .endm OUTER_MACRO_DEF @ You can use this capability to define a macro that redefines itself:_ " .macro SETUP A = 75 B = 92 C = 87 D = 0 ! E = -12 F = 42& .macro SETUP: ; Setup is done - do nothing% .endm SETUPm! .endm SETUP B In this example, the SETUP macro defines a number ofI assembly constants. After the SETUP macro has been expanded H once, its work is done. Subsequent expansions of the setupH macro need not take any action. Therefore, the SETUP macroG redefines itself to a macro whose expansion includes only C a comment statement. As described elsewhere, when you D redefine a macro, the original version of the macro isI automatically deleted. If that macro is currently expanding H (as would be the case with the SETUP macro above), the newG definition is immediately associated with the macro name.E I 4-7f t Macro Arguments " 4.4 Using String Arguments G However, the old definition is retained until all pending_G expansions complete normally. When all pending expansions F complete, the old version of the macro is deleted. Thus,G the SETUP macro may be invoked any number of times in therH assembly unit. Since the first expansion redefines itself,G the expansion of the SETUP macro has no effect other than + the first time it is invoked. I Another way to pass string arguments in nested macros is to > enclose the macro argument in nested delimiters. F ________________________ NOTE ________________________ C Each time you use the delimited argument in a macrogA call, the assembler removes the outermost pair ofe@ delimiters before associating it with the formalC argument. This method is not recommended because ituD requires that you know how deeply a macro is nested. F ______________________________________________________ H The following macro definition also contains a call to the! DOUBLE_ASCII macro:g 5 .MACRO CNTDA2 LAB1,LAB2,STR_ARGrI LAB1: .BYTE LAB2-LAB1-1 ; Length of 2*stringcN DOUBLE_ASCII STR_ARG ; Call DOUBLE_ASCII macro LAB2:e# .ENDM CNTDA2i G Note that the argument in the call to DOUBLE_ASCII is notg) enclosed in angle brackets.u ; The following example calls the macro CNTDA2:r > CNTDA2 BEG,TERM,<<MIND YOUR P'S AND Q'S>>I BEG: .BYTE TERM-BEG-1 ; Length of 2*stringeT DOUBLE_ASCII <MIND YOUR P'S AND Q'S> ; Call DOUBLE_ASCII macro4 .ASCII "MIND YOUR P'S AND Q'S"4 .ASCII "MIND YOUR P'S AND Q'S" TERM: I Note that even though the call to DOUBLE_ASCII in the macro G definition is not enclosed in delimiters, the call in theaG expansion is enclosed because the call to CNTDA2 containsp; nested delimiters around the string argument. 4-8 R 4 I Macro ArgumentsrI 4.5 Argument Concatenation " 4.5 Argument Concatenation F The argument concatenation operator, the apostrophe ('),I concatenates a macro argument with constant text or another_I argument. Apostrophes can either precede or follow a formal 0 argument name in the macro source. C If an apostrophe precedes the argument name, the text C before the apostrophe is concatenated with the actual G argument when the macro is expanded. For example, if ARG1 F is a formal argument associated with the actual argument= TEST, then ABCDE'ARG1 is expanded to ABCDETEST.l D If an apostrophe follows the formal argument name, theH actual argument is concatenated with the text that followsG the apostrophe when the macro is expanded. The apostrophet< itself does not appear in the macro expansion. C To concatenate two arguments, separate the two formal H arguments with two successive apostrophes. Two apostrophesI are needed because each concatenation operation discards ant, apostrophe from the expansion. F An example of a macro definition that uses concatenation follows: ) .MACRO CONCAT A,B A''B: .WORD 0f" .ENDM CONCAT @ Note that two successive apostrophes are used when= concatenating the two formal arguments A and B. ? An example of a macro call and expansion follows:e ! CONCAT X,Yr XY: .WORD 0: - 4.6 Passing Numeric Values of Symbols H When a symbol is specified as an actual argument, the nameD of the symbol, not the numeric value of the symbol, isH passed to the macro. The value of the symbol can be passedI by inserting a backslash (\) before the symbol in the macrosH call. The assembler passes the characters representing theG decimal value of the symbol to the macro. For example, if G the symbol COUNT has a value of 2 and the actual argument I 4-9 B Macro Argumentsi- 4.6 Passing Numeric Values of Symbols- G specified is \COUNT, the assembler passes the string 2 to H the macro; it does not pass the name of the symbol, COUNT. I Passing numeric values of symbols is especially useful withiH the apostrophe (') concatenation operator for creating new symbols. I An example of a macro definition for passing numeric values4! of symbols follows: # .MACRO WORD n WORD'n: .WORD nh .ENDM WORD G The following example shows a possible call and expansion . of the macro previously defined: 6 X = 1 ; Start counting at 1 WORD \X WORD1: .WORD 1 * 4.7 Using Created Temporary Labels H Temporary labels are often very useful in macros. AlthoughH you can create a macro definition that specifies temporary? labels within it, these temporary labels might beTI duplicated elsewhere in the temporary label block, possibly I causing errors. However, the assembler can create temporaryiG labels in the macro expansion that will not conflict withsE other temporary labels. These labels are called created temporary labels. C Created temporary labels range from 30000$ to 65535$.LD Each time the assembler creates a new temporary label,D it increments the numeric part of the label name by 1.I Consequently, no user-defined temporary labels should be in , the range of 30000$ to 65535$. G A created temporary label is specified by a question markuF (?) in front of the formal argument name. When the macroF is expanded, the assembler creates a new temporary labelC if the corresponding actual argument is blank. If the G corresponding actual argument is specified, the assembler F substitutes the actual argument for the formal argument. 4-10 I Macro Arguments I 4.7 Using Created Temporary Labels F The following example is a macro definition specifying a& created temporary label: 6 .MACRO POSITIVE ARG1,?L1% BGE ARG1,L1 ' NEGQ ARG1,ARG11& L1: .ENDM POSITIVE C The following three calls and expansions of the macrolI defined previously show both created temporary labels and a3+ user-defined temporary label:N " POSITIVE R0' BGE R0,30000$ # NEGQ R0,R0d 30000$:3 " POSITIVE R5' BGE R5,30001$ # NEGQ R5,R5r 30001$: & POSITIVE R7,10$$ BGE R7,10$# NEGQ R7,R7e 10$: H For more information on temporary labels, see Section 2.6. I 4-11 _ I 5oI _________________________________________________________________s I MACRO-64 Assembler Directives ? MACRO-64 directives enable you to control certaineE aspects of the assembly process. This chapter describes A the MACRO-64 assembler directives. It describes the_E various functional categories of directives, as well as I a detailed description of each directive. It also discusses B the effects of automatic data alignment on assembler directives. % 5.1 Program Sections (Psects)t F MACRO-64 allows you to divide your program into sectionsI called psects using the .PSECT directive. Psects are usefullI for organizing your program, and for low-level control overrE the linking process. More importantly, each psect falls 9 into one of the following three categories:d G o CODE psects can contain only instructions. It containsrH no data. Psects in this category have the EXE and NOMIX" psect attributes. B o DATA psects can contain only data. It contains noE instructions. Psects in this category have the NOEXE & and NOMIX attributes. B o MIXED psects can contain instructions or data, orF both. Psects in this category have the MIX attribute.C In addition, they may have either the EXE or NOEXEs attribute.: 2 MACRO-64 categorizes psects because: I o There is a significant performance compromise associated F with mixing instructions and data in the same programC section within the Alpha AXP architecture. This isnG because the Alpha AXP architecture typically maintainshB separate memory caches for instructions and data. I 5-1d b % MACRO-64 Assembler Directivesn% 5.1 Program Sections (Psects) D o If you mix instructions and data, it is likely thatG instructions will migrate into the data cache and that I data will migrate into the instruction cache. While this E situation still yields correct results, the benefits C of the instruction and data caches are diminished. E Placing data in the instruction stream can also have H detrimental effects on the instruction pipeline and theH multiple instruction-issue capabilities that most Alpha$ AXP systems employ. I Since a code psect can contain only instructions and cannotaF contain arbitrary data, instructions you place in a CODEE psect can be analyzed by the assembler optimizer and by D the symbolic debugger. Since a mixed psect can containF arbitrary data as well as instructions, instructions youF place in a mixed psect are not analyzed by the assemblerA optimizer or by the symbolic debugger. Instead, theaI assembler internally converts instructions in a mixed psectf3 to an equivalent data representation. I Because of the compromises associated with mixed psects, bybG default the assembler creates psects with the NOMIX psectiF attribute. If you need to place data in a psect that hasH the EXE attribute, or if you need to place instructions inI a psect that has the NOEXE attribute, you must also specifyaD the MIX attribute in the psect's definition. Note thatG unlike the other psect attributs, the MIX psect attributehI is an assembly-time attribute. The MIX psect attribute doessG not appear in the psect definitions in your object modulei9 and it does not affect the linking process. C All assembler directives and instructions can be usedaF within mixed psects. While many assembler directives canD be used within both code and data psects, data storageG directives cannot be used in code psects and instructionssB and instruction storage directives cannot be used inC data psects. If you place instructions or instruction F storage directives in a data psect, or if you place dataH storage directives in a code psect, the assembler issues aH diagnostic message. Section 5.3 indicates the applicibilty< of each directive within code and data psects. 5-2 D I MACRO-64 Assembler DirectivesDI 5.1 Program Sections (Psects) B In summary, code psects may contain only storage forI instructions and storage created by instruction directives. B Data psects may contain only storage created by dataE directives. Mixed psects may contain either storage forhE instructions or data or both. There are no restrictionsaF on the use of data directives in a mixed psect. However,F the assembler converts instructions you place in a mixedH psect to a data representation. Therefore, instructions inF a mixed psect are not analyzed as instructions by eitherH the assembler optimizer or the symbolic debugger. For moreB infomation on the .PSECT directive, see Section 5.3. $ 5.2 Automatic Data Alignment E The assembler can optionally align your data on naturalnD boundaries. While disabled by default, this feature isE enabled with the /ALIGNMENT=data command line qualifier F or the .ENABLE ALIGN_DATA directive (see Section 1.2 andC Section 5.3. A natural boundary is an address that iscH evenly divisibly by the size of the scalar data type beingI accessed. The MACRO-64 automatic data alignment feature can_H automatically align word, longword, quadword, and octawordE data to a natural boundary. Accessing data on a naturaltD boundary can be significantly faster than an unalignedG access. For more information on accessing data on aligned G and unaligned boundaries, refer to the Alpha Architecture Reference Manual. C When the MACRO-64 automatic data alignment feature is B enabled, the assembler automatically aligns all dataB storage directives to a natural boundary. To acheiveH alignment, the assembler pads with zero bytes as necessaryE prior to allocating the storage for the data directive. A Labels that occur before the data storage directivenE are defined to be the address of the data storage after E alignment. For more information, see Section 2.7.1. ThenG directive descriptions in Section 5.3 indicate those datamD storage directives that are affected by automatic data alignment. I 5-3, e w % MACRO-64 Assembler Directiveso 5.3 Directives 5.3 Directives E The general assembler directives provide facilities for I performing eleven types of functions. Table 5-1 lists these F types of functions and their directives. Some directivesE are only applicable within data psects (psects with the D NOEXE and NOMIX attributes). Other directives are onlyD applicable within code psects (psects with the EXE andE NOMIX attributes). All directives are applicable within H psects which contain either data and code, or both (psectsA with the MIX attribute). For information on the MIX F assembly-time psect and any associated restrictions, seeF the description of the .PSECT directive in this chapter. F Table 5-1 shows how each directive can be applied within" data or code psects. I Table_5-1_Summary_of_General_Assembler_Directives__________ I Category____________Directives[1]_________Psect_Application ; Listing control .TITLE All 4 directives .SUBTITLE (.SBTTL)( .IDENT' .LIST ( .NLIST' .SHOW ) .NOSHOW ' .PAGE ; Message display .PRINT Allf' directives .WARNS( .ERROR ; Assembler option .ENABLE (.ENABL) AllA3 directives .DISABLE (.DSABL)l I [1]The_alternate_form,_if_any,_is_given_in_parentheses.____ I (continued on next page) 5-4 R S I MACRO-64 Assembler DirectivesrI 5.3 DirectivesB I Table_5-1_(Cont.)_Summary_of_General_Assembler_Directives__ I Category____________Directives[1]_________Psect_Application F Data storage .BYTE NOEXE (or MIX)' directives .WORDi. .SIGNED_BYTE. .SIGNED_WORD' .LONGt* .ADDRESS/ .CODE_ADDRESS 5 .LOCAL_CODE_ADDRESS ' .OCTAc' .QUADa( .ASCII( .ASCIC( .ASCID( .ASCIZ6 .F_FLOATING (.FLOAT)- .D_FLOATINGm+ (.DOUBLE)e- .G_FLOATING - .S_FLOATINGh- .T_FLOATINGa D .INSTRUCTION EXE (or MIX) ; Location control .ALIGN Alln. directives .BEGIN_EXACT, .END_EXACT I [1]The_alternate_form,_if_any,_is_given_in_parentheses.____ I (continued on next page)o I 5-5e e o % MACRO-64 Assembler Directives 5.3 Directives I Table_5-1_(Cont.)_Summary_of_General_Assembler_Directives__ I Category____________Directives[1]_________Psect_Applications F NOEXE (or MIX)' .EVEN & .ODD' .BLKA ' .BLKBt' .BLKDh' .BLKFo' .BLKG ' .BLKLf' .BLKO ' .BLKQ ' .BLKSl' .BLKTi' .BLKW ; Program sectioning .PSECT Alls5 directives .SAVE_PSECT (.SAVE)y0 .RESTORE_PSECT, (.RESTORE) ; Symbol control .EXTERNAL (.EXTRN) Alln' directives .WEAKt ; Conditional .IF Allb2 assembly block .IF_FALSE (.IFF)' directives .ENDC ' .ELSE 1 .IF_TRUE (.IFT)n0 .IF_TRUE_FALSE) (.IFTF) & .IIF ; Source inclusion .INCLUDE All I [1]The_alternate_form,_if_any,_is_given_in_parentheses.____h I (continued on next page)l 5-6n a e I MACRO-64 Assembler DirectivesoI 5.3 Directivesh I Table_5-1_(Cont.)_Summary_of_General_Assembler_Directives__ I Category____________Directives[1]_________Psect_Application: ; Linkage control .BASE All ; .END All F .LINKAGE_PAIR NOEXE (or MIX)F .LOCAL_LINKAGE_PAIR NOEXE (or MIX)- .PROCEDURE_ , DESCRIPTOR3 .LOCAL_PROCEDURE_iI [1]The_alternate_form,SifIany,_is_given_in_parentheses.____e I ___________________________________________________________a I The macro directives provide facilities for performing fivesG categories of functions. Table 5-2 lists these categories I and their associated directives. See Chapter 4 for detailed B information on macro arguments and string operators. I Table_5-2_Summary_of_Macro_Directives______________________q E Application in EXE I Category_________Directives[1]_______or_NOEXE_Psects_______e 6 Macro .MACRO All$ definition .ENDM' directives .LIBRARYl% .MCALL 6 Macro deletion .MDELETE All directives 6 Macro exit .MEXIT All directive 6 Repeat block .REPEAT (.REPT) All# directives .IRP $ .IRPC$ .ENDR 6 Counting .NARG All$ directive .NCHRI [1]The_alternate_form,_if_any,_is_given_in_parentheses.____ I ___________________________________________________________G F The remainder of this chapter describes both the generalD assembler directives and the macro directives, showingF their formats and giving examples of their use. For ease I 5-7 % MACRO-64 Assembler Directives 5.3 Directives H of reference, the directives are presented in alphabetical order. 5-8$ I Assembler Directives I .ADDRESSe I _________________________________________________________________ .ADDRESS ' Address storage directive Format $ .ADDRESS address-list Parameters address-listH A list of symbols or expressions, separated by commas (,),5 which MACRO-64 interprets as addresses. DescriptionD G .ADDRESS stores successive quadwords (8 bytes) containingtI addresses in the object module. Digital recommends that you H use .ADDRESS rather than .QUAD for storing address data to; provide additional information to the linker.l Notes E o This directive can only be used within data or mixed A psects (psects that have either the NOEXE or MIX D attributes). For more information, see Section 5.1. G o If automatic data alignment is enabled, this directive G aligns the current (64-bit) boundary before allocating storage.g F o You can define a 32-bit address item using macros and2 the .LONG directive. For example: ' .macro address_32 item .long itemn! .endm address_32a Examples .ADDRESS A,B,C I 5-9 T s Assembler Directives .ALIGN I _________________________________________________________________ .ALIGN 2 Location counter alignment directive Format / .ALIGN integer [,fill-specifier]h / .ALIGN keyword [,fill-specifier]n Parameters integer E An integer in the range 0 to 9. The location counter isaH aligned at an address that is the value of 2 raised to the# power of the integer.g keyword G One of five keywords that specify the alignment boundary.uG The location counter is aligned to an address that is thel4 next multiple of the following values: I ___________________________________________________________ I Keyword___Size_(in_Bytes)__________________________________i BYTE 20= 1s WORD 21= 2I LONG 22= 4i QUAD 23= 8 OCTA 24= 16I ___________________________________________________________t [,fill-specifier] F Any expression that resolves to an assembly time integerI value containing no forward references. The filling is doneP4 per byte, regardless of the alignment. Descriptione @ .ALIGN aligns the location counter to the boundary: specified by either an integer or a keyword. 5-10 s s I Assembler DirectivesnI .ALIGNE Notes I o If .ALIGN is specified in a psect with the EXE and NOMIXcI attributes, the fill specifier is ignored. The assembler H aligns the psect to the requested boundary padding with* NOP or FNOP instructions. E o If .ALIGN is specified in a psect that does not have E the EXE attribute and a fill_specifier is specified, I the assembler aligns the psect to the requested boundaryeI padding, with byte locations using the fill-specifier assB the initial value for the generated byte padding. D o If the fill specifier expression encounters a valueD that is too large to fit in a boundary specified byH the keyword, the data is truncated and an informational& message is displayed. G o The alignment that you specify in .ALIGN cannot exceed B the alignment of the program section in which theH alignment is attempted (see the description of .PSECT).G For example, if you are using the BYTE program sectioneG alignment and you specify .ALIGN with a word or largereD alignment, the assembler displays an error message. Examples Example 1 5 .PSECT A,QUAD ; Begin at quadword 0 B::.BYTE 4 ; Data is byte0 .ALIGN QUAD ; Next data is9 C::.WORD 6 ; also quadword alignedo Example 2 * .PSECT A,EXE,NOMIX,OCTA, L1::TRAPB: ; offset 0; .ALIGN OCTA ; NOP padding bytes 4..15g. TRAPB: ; offset 16 I 5-11 - Assembler Directives .ALIGN Example 3 , .PSECT A,NOEXE,NOMIX,OCTA4 L1:.WORD 5 ; byte offset 0..1@ .ALIGN QUAD,2 ; fill specifier initial value7 ; of 2 for bytes 2..7s5 .WORD 6 ; byte offsets 8..9c 5-12 I Assembler DirectiveseI .ASCICb I _________________________________________________________________e .ASCIC 4 Counted ASCII string storage directive Format $ .ASCIC quoted-literal Parameters quoted-literal; An ASCII string delimited with double quotes.o Descriptionu F .ASCIC performs the same function as .ASCII, except thatE .ASCIC inserts a count byte before the string data. TheaH count byte contains the length of the string in bytes. TheI length given includes any bytes of non-printable charactersSF specified using the backslash (\) operator, but excludesI the count byte. See Section 2.4 for a description of how to & specify quoted literals. @ .ASCIC is useful in copying text because the count< indicates the length of the text to be copied. Notesa E o This directive can only be used within data or mixedbA psects (psects that have either the NOEXE or MIX D attributes). For more information, see Section 5.1. B o This directive also accepts VAX MACRO syntax. SeeC VAX MACRO and Instruction Set Reference Manual forn details. Examples C .ASCIC "MY STRING" ; In the listing, this becomes:t- ; .BYTE 9 8 ; .ASCII \MY STRING\ I 5-13 Assembler Directives .ASCID I _________________________________________________________________ .ASCID > String-descriptor ASCII string storage directive Format $ .ASCID quoted-literal Parameters quoted-literalC A ASCII string delimited with double quotes. For more @ information on how to specify quoted literals, see Section 2.4. Description H .ASCID performs the same function as the .ASCII directive,C except that .ASCID inserts a string descriptor beforeeD the string data. The descriptor format is identical toF that defined for OpenVMS AXP and OpenVMS VAX. The string2 descriptor has the following format: 6 +---------------+------------+6 | Information | Length |6 +---------------+------------+6 | Pointer |6 +----------------------------+ Notes > o The string-length field is two bytes in size. B o Descriptor information (2 bytes) is always set to ^X010E. E o Position-independent 32-bit pointer to the string (4aG bytes). String descriptors are used in calling certain ! system routines.G F o If natural alignment is enabled (using .ENABLE ALIGN_G DATA), the descriptor is quadword aligned. This allowsoH you to access the entire descriptor (2 data words and a; longword address) on a quadword alignment. 5-14 I Assembler Directives I .ASCID E o This directive can only be used within data or mixed A psects (psects that have either the NOEXE or MIX D attributes). For more information, see Section 5.1. B o This directive also accepts VAX MACRO syntax. SeeC VAX MACRO and Instruction Set Reference Manual forn details. Examples Example 1 M .DESCR1: .ASCID "ARGUMENT FOR CALL" ; String descriptor Example 2 U .DESCR2: .ASCID "SECOND ARGUMENT" ; Another string descriptorB I 5-15_ _ _ Assembler Directives .ASCII I _________________________________________________________________R .ASCII , ASCII string storage directive Format $ .ASCII quoted-literal Parameters quoted-literal; An ASCII string delimited with double quotes. Description D .ASCII stores the ASCII value of each character in theF ASCII string or the value of each byte expression in theG next available byte. See Section 2.4 for a description of - how to specify quoted literals. Notes E o This directive can only be used within data or mixed A psects (psects that have either the NOEXE or MIXED attributes). For more information, see Section 5.1. B o This directive also accepts VAX MACRO syntax. SeeC VAX MACRO and Instruction Set Reference Manual for details.o Examples " .ASCII "MY STRING" 5-16 I Assembler Directives_I .ASCIZ I _________________________________________________________________ .ASCIZ < Zero-terminated ASCII string storage directive Format $ .ASCIZ quoted-literal Parameters quoted-literal; An ASCII string delimited with double quotes.t Descriptionp F .ASCIZ performs the same function as .ASCII, except thatF .ASCIZ appends a null byte as the final character of theC string. When a list or text string is created with an F .ASCIZ directive, you need only perform a search for theI null character in the last byte to determine the end of the I string. See Section 2.4 for a description of how to specify quoted literals. Notes E o This directive can only be used within data or mixed A psects (psects that have either the NOEXE or MIX D attributes). For more information, see Section 5.1. B o This directive also accepts VAX MACRO syntax. SeeC VAX MACRO and Instruction Set Reference Manual forR details.c Examples 5 .ASCIZ "MY STRING" ; Equivalent toA? ; .ASCII "MY STRING \x00"b I 5-17 D Assembler Directives .BASE I __________________________________________________________________ .BASE % Base register directive Format * .BASE Rn [,base_expression] Parameters RnD One of the base registers, R0 through R30, FP, and SP. base_expression H The base address, which is optional, and can be one of the following: ' o An absolute expressionr ) o A relocatable expression ' o An external expression B An expression must not contain forward references orG implicit external symbols. An implicitly defined external_B symbol is a symbol that the assembler defaults to anH external symbol. This occurs when the assembler encountersI references to a symbol, but does not encounter a definition H for the symbol or an .EXTERNAL directive that declares the symbol. Description F The .BASE directive is used to inform the assembler thatA a specified base register contains a specified baseiF address. Later in your program, the assembler allows youG to implicitly reference the specified base register. WheniH the assembler knows which base addresses are stored in oneH or more base registers, it can convert an expression to anH offset from one of the base registers previously specifiedH in a .BASE directive. .BASE provides a convenient and moreC readable short hand for accessing memory and constant E values using base registers. .BASE also makes it easierDF for you to change your register assignments if you later modify your code. 5-18 n I Assembler Directives I .BASE E The base expression is optional. If the base expression E is specified, this base address value is assumed by thePH assembler to be in the specified register, Rn. If the baseG expression is omitted, the contents of the specified baseiD register, Rn, is considered undefined until a new base> expression is associated with the base register. E R31 is defined to always contain zero, according to the_D architecture definition. Therefore, R31 is known to beC a predefined base register containing zero. For everycF assembly, the assembler assumes the following statement: .BASE R31, 0 C Because the contents of R31 cannot change, you cannot - specify a base address for R31. E You can use the .BASE directive to implicitly referenceaH base registers. You can also automatically compute offsetsE from a base address known to be in a register to a base 9 address you use in an instruction argument. B Most of the memory format Alpha AXP instructions areG defined such that one of their arguments must have a base_F register and an offset. If the assembler encounters onlyI an expression with no base register, the assembler attemptsAE to find a base register that contains a base address ortH constant within a 16-bit signed offset of the value of theI expression. If it finds such a base register, the assemblertD computes an offset that when added to the value of theG base register, results in a value equal to the expressionn4 specified in the instruction argument. I 5-19 e Assembler Directives .BASEs Examples Example 1 + .EXTERNAL COMM_AREA 1 + .BASE R1, COMM_AREA 2e/ CURR_LINE = COMM_AREA + 0a/ CURR_COLUMN = COMM_AREA + 4 / CURR_MODE = COMM_AREA + 8oD LDA R4, 17 ; LDA R4, 17(R31) 3D LDL R2, CURR_LINE ; LDL R2, 0(R1) 4? LDL R3, CURR_COLUMN ; LDL R3, 4(R1)2? STL R4, CURR_MODE ; STL R4, 8(R1)a ? 1 This statement declares an external symbol, @ COMM_AREA. COMM_AREA is a global symbol thatC represents the base address of a three-longwordNI communication area that is used by different routines_# in the program. B 2 This statement informs the assembler that baseE register R1 contains the base address, COMM_AREA, I of this communication area. The next three statements C define variables within the communication area. @ 3 The first instruction shows how you can loadI registers with constant values in the range -32768 to H +32767 by implicitly using R31 as the base register. @ 4 The last three statements show how the .BASEE directive allows you to implicitly reference base_H registers and automatically compute offsets. In eachI of these instructions, the second argument is definedE= to require an offset and a base register. F Since no base register is specified, the assemblerG attempts to imply the base register and compute the_I offset based upon information given in previous .BASEi directives.i H In the last three instructions, the address argumentC is within -327678 to +32767 of the base address F known to be in R1 (i.e., COMM_AREA). Therefore, R1H is selected as the base register. The assembler alsoE computes the correct offset from the base address E known to be in R1 to the address specified in the ) instruction argument. 5-20 I Assembler Directives I .BASE Example 2 G The assembler performs a sequential search through the I list of possible base registers, R0 through R31. It uses I the first definition possible if multiple base registers ( are valid. For example: " .BASE R5, 300 :" LDQ R10, 100 F The assembler outputs the LDQ instruction as follows: ' LDQ R10, -200(R5)p G Both R31 and R5 are defined as base registers that canbF be used in constructing the instruction argument. R31H always contains zero. In this example, R5 is also knownD to contain the constant 300. The assembler uses theD first base register, starting at R0 and progressingC to R31, which provides a known value within -32768 E to +32767 of the specified argument value. Since the F assembler considers R5 before it considers R31, R5 is& used rather than R31. I 5-21 e Assembler Directives .BEGIN_EXACT I _________________________________________________________________D .BEGIN_EXACT / Exact instruction block directive Format .BEGIN_EXACT Description F An exact instruction block suppresses code optimizations? (SCHEDULE, PEEPHOLE, ALIGN_CODE, and ALIGN_DATA). C regardless if these optimizations are enabled for the F assembly unit. Unlike .ENABLE and .DISABLE, which can beF used to enable or disable specific optimizations for theE entire assembly unit, .BEGIN_EXACT and .END_EXACT allowUC you to suppress optimization for a specified range of D instructions. Instructions outside the specified rangeC remain subject to any optimizations you have enabled. Notes G o This directive cannot appear in a psect with the NOEXE & and NOMIX attributes. E o Although this directive is accepted by the assembler G in a psect with the MIX attribute, it has no effect in G these psects since no code optimizations are in affect for MIX psects. G o .BEGIN_EXACT must be paired with a matching .END_EXACT 6 to close the exact instruction block. F o .BEGIN_EXACT and .END_EXACT instruction blocks can beD nested. The outermost level of the .BEGIN_EXACT andB matching .END_EXACT directives delimit the actualF exact instruction block from which code optimizationsD are suppressed. Nesting .BEGIN_EXACT and .END_EXACTF instruction blocks can be useful in macro definitionsH where the macro expansion requires an exact instructionI sequence. Nested .BEGIN_EXACT and .END_EXACT instructionuH blocks allow a macro to be invoked both from within and5 without the exact instruction block. 5-22 e I Assembler Directives I .BEGIN_EXACT Examples G The following example shows an instruction sequence priort to optimization: , addf f7, f8, f9 ; 1, addf f2, f3, f4 ; 2, addl r5, r6, r7 ; 3, addl r8, r9, r10 ; 4 H The assembler optimizes the previous example to a sequence< similar to the following instruction sequence: :, addf f7, f8, f9 ; 1, addl r5, r6, r7 ; 3, addf f2, f3, f4 ; 2, addl r8, r9, r10 ; 4 : D If you choose to suppress optimization in the previous= example, enclose the four instructions with the E .BEGIN_EXACT and .END_EXACT directives, as shown in the following example: ! .BEGIN_EXACT , addf f7, f8, f9 ; 1, addf f2, f3, f4 ; 2, addl r5, r6, r7 ; 3, addl r8, r9, r10 ; 4 .END_EXACTe I 5-23 Assembler Directives .BLKx I _________________________________________________________________- .BLKx 1 Block storage allocation directives| Format ! .BLKA [expression]+ ! .BLKB [expression] ! .BLKD [expression] ! .BLKF [expression]o ! .BLKG [expression] ! .BLKL [expression] ! .BLKO [expression]t ! .BLKQ [expression] ! .BLKS [expression] ! .BLKT [expression] ! .BLKW [expression] Parameters expressionD An integer expression specifying the amount of storageE to be allocated. All the symbols in the expression mustaA be defined at the current point in the assembly andiC the expression must be an absolute expression. If the E expression is omitted, a default value of 1 is assumed. Description E MACRO-64 has the following 11 block storage directives: I ___________________________________________________________tI DirectiveReserves_Storage_for:______Bytes_Allocated________e I .BLKA Addresses (quadwords) 8 * value of expressionS E .BLKB Byte data Value of expression 5-24 t s I Assembler Directives I .BLKxa I ___________________________________________________________ I DirectiveReserves_Storage_for:______Bytes_Allocated________ I .BLKD Double-precision 8 * value of expression_* floating-point data" (quadwords) I .BLKF Single-precision 4 * value of expression * floating-point data" (longwords) I .BLKG G_floating data 8 * value of expression " (quadwords) I .BLKL Longword data 4 * value of expressione ? .BLKO Octaword data 16 * value ofi< expression I .BLKQ Quadword data 8 * value of expression I .BLKS S_floating data 4 * value of expressiont" (longwords) I .BLKT T_floating data 8 * value of expression " (quadwords) I .BLKW____Word_data__________________2_*_value_of_expressions H Each directive reserves storage for a different data type.G The value of the expression determines the number of data E items for which MACRO-64 reserves storage. For example, H .BLKL 4 reserves storage for 4 longwords of data and .BLKBF 2 reserves storage for 2 bytes of data. The total numberG of bytes reserved is equal to the length of the data type_0 times the value of the expression. Notes B o If automatic data alignment is enabled, the .BLKxE directives align the current location counter to onelA of the alignments listed in the following table: I 5-25c t Assembler Directives .BLKxi I ________________________________________________________aI DirectivAlignment_______________________________________h 3 .BLKA Quadword (64-bit) boundary .BLKB None 3 .BLKD Quadword (64-bit) boundaryh 3 .BLKF Longword (32-bit) boundary 3 .BLKG Quadword (64-bit) boundary 3 .BLKL Longword (32-bit) boundarye 4 .BLKO Octaword (128-bit) boundary 3 .BLKQ Quadword (64-bit) boundary 3 .BLKS Longword (32-bit) boundarys 3 .BLKT Quadword (64-bit) boundaryS I .BLKW___Word_(16-bit)_boundary__________________________ G o These directives can only be used within psects having I either the NOEXE or MIX attributes. For more information , on psects, see Section 5.1. Examples " .PSECT A,NOEXEJ B:: .BLKW 10 ; 10 words (20 bytes) of storageM .BLKQ 5 ; 5 quadwords (40 bytes) of storage G .BLKW ; 1 word (2 bytes) of storagee 5-26 n I Assembler DirectivessI .BYTEt I _________________________________________________________________ .BYTEl $ Byte storage directive Format $ .BYTE expression-list Parameters expression-listl? One or more expressions separated by commas. Each E expression is first evaluated as a quadword expression; G then the value of the expression is truncated to fit in a I byte. The value of each expression should be in the range 0 G to 255 for unsigned data or in the range -128 to +127 for signed data. Descriptionh D .BYTE generates successive bytes of binary data in the object module. Notess @ o The assembler displays a warning message if the> expression is outside the range -128 to -255. I o The assembler will truncate the most significant bits ofsH an integer or external value that is too large to store in eight bits.t E o This directive can only be used within data or mixed.A psects (psects that have either the NOEXE or MIX D attributes). For more information, see Section 5.1. Examples < A: .BYTE 5 ; Stores 5 in a byte I 5-27 - Assembler Directives .CODE_ADDRESS I _________________________________________________________________ .CODE_ADDRESSS , Code address storage directive Format & .CODE_ADDRESS name-list Parameters name-list B A list of symbols separated by commas. These symbolsB should reference either a procedure descriptor name,H such as a routine name, or an externally defined procedure descriptor.i Descriptiona F .CODE_ADDRESS causes the code addresses of the specifiedG identifiers to be placed at the current psect and currentoI location counter. The specified identifier should referencer: a procedure descriptor defined in the image. Notes E o This directive can only be used within data or mixed3A psects (psects that have either the NOEXE or MIXfD attributes). For more information, see Section 5.1. G o If automatic data alignment is enabled, this directiveuB aligns the current location counter to a quadword= (64-bit) boundary before allocating storage. Examples .CODE_ADDRESS Al 5-28 I Assembler DirectivesiI .D_FLOATINGe I _________________________________________________________________a .D_FLOATINGn .DOUBLEo . Floating-point storage directive Format 5 .D_FLOATING floating-point-number-lists 1 .DOUBLE floating-point-number-list Parameters ( floating-point-number-listE A comma-separated list of floating-point constants. ThefF constants cannot contain any operators except unary plusD or unary minus. For more information on floating-point) numbers, see Section 2.3.2. Description @ .D_FLOATING evaluates the specified floating-pointD constants and stores the results in the object module.G .D_FLOATING generates 64-bit, double-precision, floating- H point data (1 bit of sign, 8 bits of exponent, and 55 bitsB of fraction). See the description of .F_FLOATING forD information on storing single-precision floating-pointG numbers and the descriptions of .G_FLOATING, .S_FLOATING, F and .T_FLOATING for descriptions of other floating-point constants. Notes C o Double-precision floating-point numbers are alwaysh rounded.r > o The alternate form of .D_FLOATING is .DOUBLE. E o This directive can only be used within data or mixed A psects (psects that have either the NOEXE or MIXeD attributes). For more information, see Section 5.1. G o If automatic data alignment is enabled, this directive B aligns the current location counter to a quadword= (64-bit) boundary before allocating storage.u I 5-29u i h Assembler Directives .D_FLOATINGt D o The Alpha AXP architecture supports only conversion? operations for the D floating-point data type.s Examples $ .D_FLOATING 3.1E+02 5-30 e c I Assembler DirectivesiI .DISABLEm I _________________________________________________________________ .DISABLE .DSABL 3 Disable assembler functions directive Format % .DISABLE argument-listn # .DSABL argument-list Parameters argument-list G One or more of the symbolic arguments listed in Table 5-3oF under the description of .ENABLE. You can use either theB long or the short form of the symbolic arguments. IfF you specify multiple arguments, separate them by commas, spaces, or tabs. Description I .DISABLE disables the specified assembler function. See the : description of .ENABLE for more information. 7 The alternate form of .DISABLE is .DSABL. I 5-31 s f Assembler Directives .ELSE I _________________________________________________________________ .ELSEv 2 Conditional assembly block directive Format .ELSE Description @ A conditional assembly block is a series of sourceF statements that is assembled only if a certain conditionE is met. .IF starts the conditional block and .ENDC ends G the conditional block; each .IF must have a correspondingyD .ENDC. The .IF directive contains a condition test andC one or two arguments. The condition test specified isfF applied to the arguments. If the test is met, all MACRO-G 64 statements between .IF and .ELSE are assembled. If the7I test is not met, the statements between .ELSE and .ENDC arec assembled. F Conditional blocks can be nested; that is, a conditionalD block can be inside another conditional block. In thisE case, the statements in the inner conditional block arevG assembled only if the condition is met for both the outer_H and inner block. For more information, see the description# of the .IF directive. Notes ? o You cannot use the .ELSE directive in the samen: conditional block as the .IF_x directive. @ o The .ELSE directive is similar to the .IF_FALSEI directive. However, you can only use .ELSE once within a G conditional block. .IF_FALSE can be used any number of . times in a conditional block. 5-32 e s I Assembler DirectivesiI .ELSEs Examples @ An example of a conditional assembly directive is: H .IF EQUAL ALPHA+1 ; Assemble block if ALPHA+1=0. .X .aD .ELSE ; Assemble when .IF=false. .e .i .ENDCa I 5-33 Assembler Directives .ENABLE. I _________________________________________________________________ .ENABLE .ENABL 2 Enable assmebler functions directive Format $ .ENABLE argument-list # .ENABL argument-list Parameters argument-listfI One or more of the symbolic arguments listed in Table 5-3 .IG You can use either the long form or the short form of the ! symbolic arguments. C If you specify multiple arguments, separate them with & commas, spaces, or tabs. I Table_5-3_.ENABLE_and_.DISABLE_Symbolic_Arguments__________o * Default! Short I Long_Form_____Form___ConditioFunction______________________ I (continued on next page) 5-34 I Assembler Directives I .ENABLE I Table_5-3_(Cont.)_.ENABLE_and_.DISABLE_Symbolic_Arguments__; * Default! ShortpI Long_Form_____Form___ConditioFunction______________________l D ALIGN_CODE DisabledThe code alignment optionG aligns certain branch target7A labels. The ALIGN_CODE E option is disabled for the F assembly unit if any one ofA the following is true: B o /NOALIGNMENT=code isF specified on the command3 line. D o .DISABLE ALIGN_CODE isE specified in the source.6 program. H o The /ALIGNMENT=code option; is defaulted._ > See Section B.7 forG information on optimizations C and automatic alignment. I (continued on next page)[ I 5-35B x Assembler Directives .ENABLE I Table_5-3_(Cont.)_.ENABLE_and_.DISABLE_Symbolic_Arguments__ * Default! Short I Long_Form_____Form___ConditioFunction______________________ G ALIGN_DATA DisabledWhen ALIGN_DATA is disabled, F the data storage directivesF put each succeeding item onB the next byte boundary. F When ALIGN_DATA is enabled,F the data storage directivesF put each succeeding item onF natural boundaries (such asD words on word boundaries,@ longwords on longwordG boundaries) adding pad bytes 8 as necessary. E Accessing data on anything_H other than natural boundariesG usually incurs a significant ? performance penalty.a D You can enable or disableD the ALIGN_DATA option forF specific ranges within yourH program. For more informationG on automatic data alignment, E see Section 5.2. Also, see F Section B.7 for informationI on optimizations and automaticf5 alignment. I (continued on next page) 5-36 I Assembler Directives I .ENABLE I Table_5-3_(Cont.)_.ENABLE_and_.DISABLE_Symbolic_Arguments__h * Default! Short I Long_Form_____Form___ConditioFunction______________________v I FLOAT Enabled Controls whether the assemblereC generates floating-pointeG instructions when optimizing I code and performing code-labela5 alignment.i H Currently, the only floating-F point instruction generatedB by the assembler duringE optimization and alignment_B processing is FNOP, theF floating-point no-operationF instruction. If you specifyH .DISABLE FLOAT, the assembler@ does not generate anyF floating-point instructionsF as part of optimization and@ alignment processing. D The initial value of thisE option is specified by thebI /ENVIRONMENT=[NO]FLOAT command F line option. The last valueI of the FLOAT option at the endtI of assembly determines whether H FLOAT is enabled or DISABLED. I (continued on next page) I 5-37 - Assembler Directives .ENABLE I Table_5-3_(Cont.)_.ENABLE_and_.DISABLE_Symbolic_Arguments__ * Default! Short_I Long_Form_____Form___ConditioFunction______________________e F GLOBAL GBL Enabled When GLOBAL is enabled, theF assembler implicitly treatsE any undefined symbol as aniE external reference definedlD in another module. If theE GLOBAL option is disabled,2A the assembler issues a H warning and implicitly treatsE the undefined symbol as an H external reference assumed toH be defined in another module.G The last value of the GLOBALsH option at the end of assemblyH determines whether the GLOBALI option is enabled or disabled.v I (continued on next page)n 5-38 I Assembler Directives I .ENABLEe I Table_5-3_(Cont.)_.ENABLE_and_.DISABLE_Symbolic_Arguments___ * Default! ShortdI Long_Form_____Form___ConditioFunction______________________m G LOCAL_BLOCK LSB DisabledUsed to override the defaultiG assembler behavior to definefF a temporary label block. (AD temporary label is of theE form n$ where n representsdG a number.) A temporary labelhE block is usually delimitedfG by two user defined local orfF global labels. However, theH .ENABLE LOCAL_BLOCK directiveG defines the start of a blockiG that is terminated by one ofe9 the following: E o A second .ENABLE LOCAL_y> BLOCK directive. D o A .DISABLE LOCAL_BLOCKC directive followed by E a user-defined local oreG global label, or a .PSECT_8 directive. I (continued on next page)D I 5-39a e s Assembler Directives .ENABLE I Table_5-3_(Cont.)_.ENABLE_and_.DISABLE_Symbolic_Arguments__o * Default! ShortnI Long_Form_____Form___ConditioFunction______________________i H PEEPHOLE DisabledPeephole optimization reducesB the strength of certainF instructions and eliminatesG instructions where possible.8A The PEEPHOLE option is I disabled for the assembly unit-I if any one of the following isF0 true: E o /NOOPTIMIZE=PEEPHOLE is F specified on the command3 line. B o .DISABLE PEEPHOLE isE specified in the source 6 program. D o The PEEPHOLE option is8 defaulted. > See Section B.7 forG information on optimizationsoC and automatic alignment. I (continued on next page)N 5-40 e r I Assembler DirectivesdI .ENABLE I Table_5-3_(Cont.)_.ENABLE_and_.DISABLE_Symbolic_Arguments__ * Default! ShortlI Long_Form_____Form___ConditioFunction______________________ A SCHEDULE DisabledInstruction scheduling_@ optimization reordersI instructions to more optimally B utilize the instructionA pipeline. The SCHEDULE E option is disabled for the F assembly unit if any one ofA the following is true: E o /NOOPTIMIZE=SCHEDULE is F specified on the command3 line. B o .DISABLE SCHEDULE isE specified in the source 6 program. D o The SCHEDULE option is8 defaulted. > See Section B.7 forG information on optimizations I _____________________________and_automatic_alignment.______e Descriptiono F .ENABLE enables the specified assembly function. .ENABLED and its negative form, .DISABLE, control the following" assembler functions: , o Creating local label blocks I o Specifying that undefined symbol references are external referencesh E o Enabling or disabling specific optimizations for thef assembly unit I 5-41 e Assembler Directives .ENABLE F You can enable one or more specific optimization optionsA with either the .ENABLE directive, or the /OPTIMIZEtF command-line qualifier, or both. For more information onG command line qualifiers, see Section 1.2. See Section B.7tA for information on optimizations. If you explicitlymH disable one or more specific optimization options with theI .DISABLE directive, those optimization options are disablednA regardless of the command-line options you specify.v Notes 9 The alternate form of .ENABLE is .ENABL. Examples Example 1 C The following example shows the ALIGN_DATA option:u .PSECT A, NOEXE 6 .ENABLE ALIGN_DATA ; Align on natural8 ; natural boundaries% A: .BYTE 1 ; 7 B: .QUAD 1000 ; B is allocated at : ; a natural boundary -; ; specifically at A + 7 % .DISABLE ALIGN_DATA ;o% C: .BYTE 2 ; 7 D: .QUAD 1001 ; D is allocated at = ; an unaligned boundary - 8 ; specifically C + 1 Example 2 H The following example shows the GLOBAL option disabled: .DISABLE GLOBAL ? .ADDRESS X ; Assembler issues a warning_ .END 5-42 _ I Assembler DirectivesrI .ENABLE Example 3 C The following example shows the LOCAL_BLOCK optionr enabled:e ) .ENABLE LOCAL_BLOCK $ .PSECT A,NOEXE A1::I 5$: .PROCEDURE_DESCRIPTOR PROC_1 ; Temporary label 5$e .blkb 32 A2::Q .address 5$ ; By default the declaration_Q ; of A2 would have ended therP ; temporary label block andP ; made this reference to 5$V ; illegal. However, this defaultS ; behavior has been overridden Y ; by the use of .ENABLE LOCAL_BLOCK.o * .DISABLE LOCAL_BLOCK .END Example 4 I The following example shows an unoptimized and optimized D list of instructions with the SCHEDULE and PEEPHOLE! options enabled: ) .ENABLE PEEPHOLE,SCHEDULE 3 .psect A,EXE,QUAD ; unoptimized TRAPB A::ADDF F1,F2,F3 ADDF F4,F5,F6 ADDL R1,R2,R3 ADDL R4,R5,R6 G This example shows the optimized list of instructions: ) .ENABLE PEEPHOLE,SCHEDULEs1 .psect A,EXE,QUAD ; optimizedc A::ADDF F1,F2,F3 ADDL R1,R2,R3 ADDF F4,F5,F6 ADDL R4,R5,R6 I 5-43u . Assembler Directives .END I _________________________________________________________________ .END , Assembly termination directive Format .END [label]c Parameters label F The procedure descriptor name that specifies the routineC (called the transfer address) where program execution 0 begins. This argument is optional. Description D .END terminates the source program. No additional textH should occur beyond this point in the current source file,H or in any additional source files specified in the commandH line for this assembly. If any additional text does occur,D the assembler ignores it. The additional text does notF appear in the listing file nor does it affect the object file. Notes F o When an executable image consisting of several objectA modules is linked, only one object module should D be terminated by an .END directive that specifies aE transfer address. All other object modules should be D terminated by .END directives that do not specify aI transfer address. If an executable image contains either G no transfer address or more than one transfer address, 6 the linker displays an error message. D o For more information, see the .PROCEDURE_DESCRIPTOR directive. 5-44 e I Assembler Directives I .END Examples . . .u? .PROCEDURE_DESCRIPTOR TRANSFER1,code_address_T1 .n . . L .END TRANSFER1 ; TRANSFER1 is module transfer address I 5-45p r Assembler Directives .ENDC I _________________________________________________________________ .ENDCn ' End conditional directive Format .ENDCi Description G .ENDC terminates the conditional range started by the .IF H directive. See the description of .IF for more information and examples.x 5-46 I Assembler Directives I .ENDMn I _________________________________________________________________ .ENDM , End macro definition directive Format ! .ENDM [macro-name]_ Parameters macro-nameI The name of the macro whose definition is to be terminated.gI The macro name is optional; if specified, it must match thecF name defined in the matching .MACRO directive. The macroG name should be specified so that the assembler can detecti6 any improperly nested macro definitions. Descriptiont H .ENDM terminates the macro definition. See the descriptionH of .MACRO for an example of the use of an .ENDM directive. Notes H If .ENDM is encountered outside a macro definition, the5 assembler displays an error message. I 5-47i y Assembler Directives .ENDR I _________________________________________________________________e .ENDRy ( End repeat range directive Format .ENDR Description G .ENDR indicates the end of a repeat range. It must be the C final statement of every repeat block. A repeat block D consists of any range of text beginning with the .IRP,H .IRPC, or .REPEAT directive. For more information, see theH description for the .IRP, .IRPC, or .REPEAT directives for9 examples of how to use the .ENDR directive. Notesh D If .ENDR is encountered outside a repeat block, the5 assembler displays an error message. 5-48 I Assembler Directives I .END_EXACT I _________________________________________________________________ .END_EXACT 3 End exact instruction block directive Format .END_EXACT Description H .END_EXACT delimits the end of an exact instruction block.E An exact instruction block suppresses the optimizations I SCHEDULE and PEEPHOLE, and the alignment options ALIGN_CODE D and ALIGN_DATA for the specified range of instructionsB regardless if code optimizations are enabled for the assembly unit. G For more information on the .END_EXACT directive, see the H description of the .BEGIN_EXACT directive in this chapter. I 5-49 Assembler Directives .ERROR I _________________________________________________________________ .ERROR Error directiver Format $ .ERROR quoted-literal Parameters quoted-literalF A string of characters, between a pair of double quotes,G displayed during assembly. For more information on quoted ( literals, see Section 2.4. Description H .ERROR causes the assembler to display an error message onH the terminal or batch log file and in the listing file (if there is one). D Using .ERROR prevents an output object file from being produced.E Notes I o .PRINT, .WARN, and .ERROR are directives used to display B messages. You can use them to display informationE indicating unexpected or important conditions within_ the assembly. B o This directive also accepts VAX MACRO syntax. SeeC VAX MACRO and Instruction Set Reference Manual for details. Examples * .ERROR "Illegal Arguments" ^lG %MACRO64-E-GENERROR, Generated ERROR: Illegal Argumentsn@ at line number 3 in file DISK$:[TEST]ERROR.M64;2 5-50 I Assembler Directives I .EVEN I _________________________________________________________________ .EVEN 7 Even location counter alignment directive Format .EVENc Description B .EVEN ensures that the current value of the locationF counter is even by adding 1 if the current value is odd.G If the current value is already even, no action is taken. Notes E o This directive can only be used within data or mixedBA psects (psects that have either the NOEXE or MIXtD attributes). For more information, see Section 5.1. I 5-51 Assembler Directives .EXTERNAL I _________________________________________________________________ .EXTERNAL .EXTRN 1 External symbol attribute directives Format $ .EXTERNAL symbol-list ! .EXTRN symbol-listS Parameters symbol-list : A list of symbol names, separated by commas. Descriptionm @ .EXTERNAL indicates that the specified symbols areC external; that is, the symbols are defined in another object module. Notesn ; The alternate form of .EXTERNAL is .EXTRN.l Examples D .EXTERNAL B ; B is defined in another module . .i . ? A:: .ADDRESS B ; Its address is stored here 5-52 n f I Assembler Directives I .F_FLOATING I _________________________________________________________________ .F_FLOATING .FLOAT . Floating-point storage directive Format 5 .F_FLOATING floating-point-number-lists 0 .FLOAT floating-point-number-list Parameters ( floating-point-number-listI A list of one or more floating-point constants separated by I commas. For more information on floating-point numbers, see G Section 2.3.2. The constants cannot contain any operators / except unary plus or unary minus.i Description @ .F_FLOATING evaluates the specified floating-pointF constant(s) and stores the results in the object module.G .F_FLOATING generates 32-bit, single-precision, floating- H point data (1 bit of sign, 8 bits of exponent, and 23 bitsE of fractional significance). See the description of .D_ B FLOATING for information on storing double-precisionB floating-point constants and the descriptions of .G_F FLOATING, S_FLOATING, and T_FLOATING for descriptions of- other floating-point constants. Noteso = o The alternate form of .F_FLOATING is .FLOAT.n E o This directive can only be used within data or mixedtA psects (psects that have either the NOEXE or MIX.D attributes). For more information, see Section 5.1. G o If automatic data alignment is enabled, this directiveiG aligns the current location counter to a longword (32- 9 bit) boundary before allocating storage.O I 5-53 t c Assembler Directives .F_FLOATINGe Examples & .F_FLOATING 1.0,3.0E+2 5-54 I Assembler DirectiveshI .G_FLOATINGe I _________________________________________________________________ .G_FLOATING 0 G_floating-point storage directive Format 5 .G_FLOATING floating-point-number-liste Parameters ( floating-point-number-listB A comma-separated list of one or more floating-pointH constants. For more information on floating-point numbers,A see Section 2.3.2. The constants cannot contain anyu9 operators except unary plus or unary minus.r Description @ .G_FLOATING evaluates the specified floating-pointH constants and stores the results in the object module. .G_G FLOATING generates 64-bit data (1 bit of sign, 11 bits ofsH exponent, and 52 bits of fraction). See the description ofE .D_FLOATING for information on storing double-precisionoB floating-point constants and the descriptions of .F_F FLOATING, S_FLOATING, and T_FLOATING for descriptions of- other floating-point constants.m Notest E o This directive can only be used within data or mixedmA psects (psects that have either the NOEXE or MIXpD attributes). For more information, see Section 5.1. G o If automatic data alignment is enabled, this directivetB aligns the current location counter to a quadword= (64-bit) boundary before allocating storage.s Examples # .G_FLOATING 2.0E-3 I 5-55 Assembler Directives .IDENT I _________________________________________________________________ .IDENT & Identification directive Format $ .IDENT quoted-literal Parameters quoted-literalE A 1- to 31-character string, within double quotes, thattF identifies the module, such as a string that specifies aF version number. For more information on quoted literals, see Section 2.4. Descriptione G .IDENT provides a means of identifying the object module. E This identification is in addition to the name assigned F to the object module with .TITLE. A character string canE be specified in .IDENT to label the object module. This E string is printed in the header of the listing file and 0 also appears in the object module. Notest F If a source module contains more than one .IDENT, theF last directive given establishes the character stringE that forms part of the object module identification.m Examples $ .IDENT "Module Name" 5-56 a e I Assembler Directives I .IF I _________________________________________________________________ .IF 2 Conditional assembly block directive Format ( .IF condition argument(s) . . . rangeo . . . .ENDCh Parameters condition D A specified condition that must be met if the block isC to be included in the assembly. The condition must bepD separated from the argument by a comma, space, or tab.F Table 5-4 lists the conditions that can be tested by the. conditional assembly directives. argument(s)pB One or more symbolic arguments or expressions of the? specified conditional test. If the argument is an3B expression, it cannot contain any undefined symbols.F The assembler converts relocatable arguments to absoluteD arguments by discarding the relocatable portion of theH expression and using only the offset from the beginning of@ the psect. Arguments must be separated by a comma. range H The block of source code that is conditionally included in the assembly.E I 5-57 Assembler Directives .IFc I Table_5-4_Condition_Tests_for_Conditional_Assembly_Directives____ Q ConditionrL Condition Complement Number ThatQ Condition Argument of Assembles M Test Test Type Arguments Blocku/ Short Shortf LongI Form_______Form___Long_Form_______Form___________________________ S EQUAL EQ NOT_EQUAL NE Expression 1 or Expression-bL 2[1] 1 isP equal toS expression- P 2 or notP equal toS expression-oJ 2. S GREATER GT LESS_EQUAL LE Expression 1 or Expression- L 2[1] 1 isO greater L thanS expression- L 2 orL lessO than or P equal toS expression- J 2. I [1]If_the_second_argument_is_omitted,_comparison_is_made_against__
0. I (continued on next page)o 5-58 m I Assembler DirectivestI .IFd B Table 5-4 (Cont.) Condition Tests for Conditional AssemblyI __________________Directives_____________________________________ Q Condition L Condition Complement Number ThatQ Condition Argument of Assembles M Test Test Type Arguments Block / Short Shorta LongI Form_______Form___Long_Form_______Form___________________________i S LESS_ LT GREATER_EQUAL GE Expression 1 or Expression-aL THAN 2[1] 1 isL lessL thanS expression- L 2 orO greater O than or P equal toS expression-_J 2. N DEFINED DF NOT_DEFINED NDF Symbolic 1 SymbolJ isO definedrN or notP defined. P BLANK B NOT_BLANK NB Macro 1 ArgumentP is blankN or notN blank. Q IDENTICAL IDN DIFFERENT DIF Macro 2 ArgumentsDK areNO identi-_N cal orO differ- L ent.I [1]If_the_second_argument_is_omitted,_comparison_is_made_against_
0.I _________________________________________________________________A I 5-59o m t Assembler Directives .IFu Descriptiono @ A conditional assembly block is a series of sourceG statements that are assembled only if a certain condition I is met. A .IF starts the conditional block and a .ENDC ends G the conditional block; each .IF must have a correspondingvD .ENDC. The .IF directive contains a condition test andC one or two arguments. The condition test specified is H applied to the arguments. If the test is met, all MACRO-64I statements between .IF and .ENDC are assembled. If the test G is not met, the statements are not assembled. Optionally,iF you can use the .ELSE directive (or a combination of theG .IFF, .IFT, and .IFTF directives) to specify an alternate F series of statements to assemble if the test is not met. F Conditional blocks can be nested; that is, a conditionalD block can be inside another conditional block. In thisE case, the statements in the inner conditional block are,G assembled only if the condition is met for both the outera and inner block. Notes I o The assembler displays an error message if the following G directives occur outside a conditional assembly block:MC .ENDC, .ELSE, .IF_FALSE, .IF_TRUE, .IF_TRUE_FALSE.r D o MACRO-64 permits a nesting depth of 100 conditionalH assembly levels. If a statement attempts to exceed thisE nesting level depth, the assembler displays an error message.e G o The effect of logical expressions can only be achieved G by using several levels of .IF directives. See Example 5. F o Lowercase string arguments are converted to uppercaseG before being compared, unless the string is surrounded E by double quotes or /NAMES=as_is is specified on the command line. I o The assembler displays an error message if .IF specifiesnH any of the following: a condition test other than thoseF in Table 5-4, an illegal argument, or a null argument/ specified in an .IF directive.t 5-60 d t I Assembler Directives I .IF Examples Example 1 C An example of a conditional assembly directive is: K .IF EQUAL ALPHA+1 ; Assemble block if ALPHA+1=0. Do_K . ; not assemble if ALPHA+1 not=0R . .n .ENDCe Example 2 = Nested conditional directives take the form:l + .IF condition argument(s)y+ .IF condition argument(s)m .a .t . .ENDCt .ENDCa Example 3 H The following conditional directives can govern whether= assembly of the specified range is to occur:N ! .IF DEFINED SYM1 ! .IF DEFINED SYM2 .t . . .ENDC .ENDC Example 4 C In this example, if the outermost condition is not2C satisfied, no deeper level of evaluation of nestedAB conditional statements within the program occurs.F Therefore, both SYM1 and SYM2 must be defined for the1 specified range to be assembled.F I An alternate series of statements can be specified using .ELSE: I 5-61i p t Assembler Directives .IF B .IF EQUAL A,B ; Assemble if A is equal to BF .ELSE ; Assemble if A is not equal to B .ENDCn Example 5 H The following example demonstrates the use of .ELSE and nesting:s C .IF LESS_THAN X,Y ; Assemble if X is less than YpC .IF DEFINED Z ; Assemble if Z is defined andl8 ; X is less than YG .ELSE ; Assemble if Z is not defined and 8 ; X is less than Y .ENDCR .ELSE ; Assemble if X is greater than or equal to YH .IF DEFINED Z ; Assemble if Z is defined and X isB ; greater than or equal to Y .ENDC .ENDCn 5-62 f _ I Assembler DirectivesaI .IF_x I _________________________________________________________________ .IF_x 6 Subconditional assembly block directives Format .IF_FALSEt .IF_TRUE .IF_TRUE_FALSE DescriptionO G For compatibility with VAX MACRO, MACRO-64 provides three53 directives for use within .IF blocks: I ___________________________________________________________EI Directive_______Function___________________________________ D .IF_FALSE If the condition of the assembly blockC tests false, the program includes the A source code following the .IF_FALSE_E directive and continuing up to the nextGG subconditional directive or to the end of = the conditional assembly block.e D .IF_TRUE If the condition of the assembly blockI tests true, the program includes the source G code following the .IF_TRUE directive and F continuing up to the next subconditionalH directive or to the end of the conditional- assembly block.l H .IF_TRUE_FALSE Regardless of whether the condition of theE assembly block tests true or false, the,F source code following the .IF TRUE_FALSEF directive (and continuing up to the nextG subconditional directive or to the end ofnI ________________the_assembly_block)_is_always_included.____n G The implied argument of a subconditional directive is the D condition test specified when the conditional assembly@ block was entered. A conditional or subconditionalE directive in a nested conditional assembly block is not I 5-63 Assembler Directives .IF_xd D evaluated if the preceding (or outer) condition in theC block is not satisfied (see Example 3 and Example 4). D A conditional block with a subconditional directive isI different from a nested conditional block. If the condition_I in the .IF is not met, the inner conditional blocks are notTI assembled, but a subconditional directive can cause a block to be assembled. Notes @ o If a subconditional directive appears outside aF conditional assembly block, the assembler displays an error message. E o The alternate forms of .IF_FALSE, .IF_TRUE, and .IF_ 6 TRUE_FALSE are .IFF, .IFT, and .IFTF. F o You cannot use .ELSE in the same conditional block as .IF_x.f Examples Example 1 3 Assume that symbol SYM is defined: 5-64 e h I Assembler DirectivesiI .IF_xo U .IF DEFINED SYM ; Tests TRUE since SYM is defined. T . ; Assembles the following code. . .O .IF_FALSE ; Tests FALSE since previous N . ; .IF was TRUE. Does notS . ; assemble the following code. .U .IF_TRUE ; Tests TRUE since SYM is defined. T . ; Assembles the following code. . .M .IF_TRUE_FALSE ; Assembles following code G . ; unconditionally.a . .U .IF_TRUE ; Tests TRUE since SYM is defined.dM . ; Assembles remainder ofmR . ; conditional assembly block. . .ENDC Example 2 I Assume that symbol X is defined and that symbol Y is not defined:f S .IF DEFINED X ; Tests TRUE since X is defined.iX .IF DEFINED Y ; Tests FALSE since Y is not defined.W .IF_FALSE ; Tests TRUE since Y is not defined. T . ; Assembles the following code. . .X .IF_TRUE ; Tests FALSE since Y is not defined.V . ; Does not assemble the following< . ; code. . .ENDC .ENDC I 5-65 Assembler Directives .IF_xe Example 3 I Assume that symbol A is defined and that symbol B is not defined:e S .IF DEFINED A ; Tests TRUE since A is defined. T . ; Assembles the following code. . .T .IF_FALSE ; Tests FALSE since A is defined.V . ; Does not assemble the following< . ; code. .Q .IF NOT_DEFINED B ; Nested conditional directive H . ; is not evaluated. . . .ENDC .ENDC Example 4 D Assume that symbol X is not defined but symbol Y is defined: X .IF DEFINED X ; Tests FALSE since X is not defined.V . ; Does not assemble the following< . ; code. .Q .IF DEFINED Y ; Nested conditional directive H . ; is not evaluated. . .J .IF_FALSE ; Nested subconditionalR . ; directive is not evaluated. . .J .IF_TRUE ; Nested subconditionalR . ; directive is not evaluated. . . .ENDC .ENDC 5-66 a n I Assembler Directives I .IIF I _________________________________________________________________ .IIF < Immediate conditional assembly block directive Format 7 .IIF condition [,]argument(s), statementi Parameters condition F One of the legal condition tests defined for conditionalH assembly blocks in Table 5-4 (see the description of .IF).E The condition must be separated from the arguments by anI comma, space, or tab. If the first argument can be a blank, G the condition must be separated from the arguments with a comma. argument(s)gI An expression or symbolic argument (described in Table 5-4)RF associated with the immediate conditional assembly blockD directive. If the argument is an expression, it cannotD contain any undefined symbols. For more information onC terms and expressions, see Section 2.8. The assembler E converts relocatable arguments to absolute arguments by F discarding the relocatable portion of the expression andH using only the offset from the beginning of the psect. TheH arguments must be separated from the statement by a comma. statement ? The statement to be assembled if the condition is satisfied. Description E .IIF provides a means of writing a one-line conditional @ assembly block. The condition to be tested and theH conditional assembly block are expressed completely withinD the line containing the .IIF directive. No terminating5 .ENDC statement is required or allowed. I 5-67 Assembler Directives .IIF Notes @ The assembler displays an error message if .IIFC specifies a condition test other than those listed F in Table 5-4, which the assembler considers valid, an6 illegal argument, or a null argument. Examples I In the following example, the symbol EXAM is defined within ! the source program: + .IIF DEFINED EXAM, BR ALPHAr : This directive generates the following code: BR ALPHA 5-68 i I Assembler Directives I .INCLUDEr I _________________________________________________________________. .INCLUDE + Include source file directivea Format & .INCLUDE quoted-literal Parameters quoted-literalF The name of the source file to be included within doubleB quotes. If a logical name exists that is the same asF the source file name, specify the .M64 file extension toG override the logical name. For more information on how tot7 specify quoted literals, see Section 2.4.b Description C .INCLUDE indicates that the current input source filevH should be suspended and the specified file should be used.F When that file ends, the original source stream resumes,2 starting at the line after .INCLUDE. Notes B The assembler issues an error message if the file* nesting level exceeds 50. Examples $ .INCLUDE "file1.m64" I 5-69o e y Assembler Directives .INSTRUCTION I _________________________________________________________________ .INSTRUCTION # Instruction directivee Format & .INSTRUCTION expression Parameters expressionC An absolute expression in the range of -2147483648 to H 2147483647. For more information on terms and expressions,D see Section 2.8. The expression cannot be relocatable,# external, or complex.s Description I The specified value is stored at the current location as an H instruction. You can use .INSTRUCTION to specify arbitraryC instructions. Similar to .LONG, .INSTRUCTION stores a E longword (32-bits) of data. Unlike .LONG, the assembler I considers .INSTRUCTION an instruction and allows its use in code psects. Notesa ? o This directive can only be used within code or E mixed psects (psects that have either the EXE or MIX D attributes). For more information, see Section 5.1. F o If automatic data alignment is enabled within a mixedB psect, this directive aligns the current location? counter to a longword (32-bit) boundary before $ allocating storage. I o This directive can be used to store arbitrary, longword,u? assembly-time constant data in a code section. 5-70 I Assembler Directives I .INSTRUCTIONm Examples .INSTRUCTION 7 I 5-71 Assembler Directives .IRP I _________________________________________________________________E .IRP 2 Indefinite repeat argument directive Format * .IRP symbol,<argument list> . . . range . . . .ENDR Parameters symbolF A formal argument that is successively replaced with theC specified actual arguments enclosed in angle brackets E (<>). If no formal argument is specified, the assembler ( displays an error message. <argument list>DG A list of actual arguments enclosed in angle brackets andtF used in expanding the indefinite repeat range. An actualF argument can consist of one or more characters. MultipleF arguments must be separated by a legal separator (comma,F space, or tab). If no actual arguments are specified, no action is taken. rangenC The block of source text to be repeated once for each E occurrence of an actual argument in the list. The range H can contain macro definitions and repeat ranges. .MEXIT isI legal within the range and causes the current and remaining ( repetitions to be aborted. 5-72 I Assembler DirectiveshI .IRPT Descriptioni D .IRP replaces a formal argument with successive actualG arguments specified in an argument list. This replacement C process occurs during the expansion of the indefinite G repeat block range. The .ENDR directive specifies the end of the range.E C .IRP is analogous to a macro definition with only one I formal argument. At each successive expansion of the repeat E block, this formal argument is replaced with successive D elements from the argument list. The directive and itsE range are coded in line within the source program. This C type of macro definition and its range do not require E calling the macro by name, as do other macros described_ in this section. D .IRP can appear either inside or outside another macroG definition, indefinite repeat block, or repeat block (see_C the description of .REPEAT). The rules for specifying G .IRP arguments are the same as those for specifying macroR arguments. Examples 1 The macro definition is as follows:_ : .macro CHECK_PROCEDURE_KIND PROCEDURE_KINDB OK = 0 ; Assume procedure_kind is unknownB .irp REFERENCE_KIND,BOUND,NULL, REGISTER,STACKI .if identical, <PROCEDURE_KIND>, <REFERENCE_KIND> < OK = 1 ; Procedure_kind is known< .mexit ; No need to look further .endc .endrd; .if eq, OK ; If unknown procedure kind G .error "Unknown procedure kind: PROCEDURE_KIND"i .endc * .endm CHECK_PROCEDURE_KIND - CHECK_PROCEDURE_KIND REGISTER + CHECK_PROCEDURE_KIND FOOZLEt I 5-73l Assembler Directives .IRP D The macro call and expansion of the previously defined" macro is as follows: - CHECK_PROCEDURE_KIND REGISTER B OK = 0 ; Assume procedure kind is unknown4 .if identical,<REGISTER>,<BOUND> .endc 3 .if identical,<REGISTER>,<NULL> .endcc7 .if identical,<REGISTER>,<REGISTER>_8 OK = 1 ; Procedure kind is known8 .mexit ; No need to look further; .if eq, OK ; If unknown procedure kind .endc + CHECK_PROCEDURE_KIND FOOZLE B OK = 0 ; Assume procedure kind is unknown2 .if identical,<FOOZLE>,<BOUND> .endc 1 .if identical,<FOOZLE>,<NULL>i .endc 5 .if identical,<FOOZLE>,<REGISTER>o .endc 2 .if identical,<FOOZLE>,<STACK> .endcd; .if eq, OK ; If unknown procedure kind ? .error "Unknown procedure kind: FOOZLE" .endco E In this example the CHECK_PROCEDURE_KIND macro uses thetI .IRP directive to iterate over a list of reference keywordseG to determine if its argument matches one of the referencepF keywords. If a match is not found, the macro displays an error message. 5-74 T a I Assembler Directives I .IRPC I _________________________________________________________________ .IRPC 3 Indefinite repeat character directive Format $ .IRPC symbol,<STRING> . . . range . . . .ENDR Parameters symbolF A formal argument that is successively replaced with theF specified characters enclosed in angle brackets (<>). IfH no formal argument is specified, the assembler displays an error message. <STRING>E A sequence of characters enclosed in angle brackets and C used in the expansion of the indefinite repeat range.eD Although the angle brackets are required only when theA string contains separating characters, their use is.) recommended for legibility.t range C The block of source text to be repeated once for eachUB occurrence of a character in the list. The range canD contain macro definitions and repeat ranges. .MEXIT is% legal within the range. I 5-75 Assembler Directives .IRPCm Description H .IRPC is similar to .IRP except that .IRPC permits single-G character substitution rather than argument substitution.pC On each iteration of the indefinite repeat range, thebH formal argument is replaced with each successive characterH in the specified string. The .ENDR directive specifies the end of the range.; D .IRPC is analogous to a macro definition with only oneE formal argument. At each expansion of the repeat block, I this formal argument is replaced with successive characters D from the actual argument string. The directive and itsF range are coded in line within the source program and do4 not require calling the macro by name. E .IRPC can appear either inside or outside another macro G definition, indefinite repeat block, or repeat block (see & description of .REPEAT). Examples 1 The macro definition is as follows: & .macro X_COUNT ARG! COUNT = 03& .irpc CH,<ARG>F .iif identical,<CH>,<X>, COUNT = COUNT + 1 .endr ! .endm X_COUNT ? The macro call and expansion of the macro defined;' previously is as follows: 5-76 I Assembler Directives I .IRPCe ' X_COUNT XXFOOXBARXX ! COUNT = 0 . .irpc CH,<XXFOOXBARXX>F .iif identical,<CH>,<X>, COUNT = COUNT + 1 .endr;E .iif identical,<X>,<X>, COUNT = COUNT + 1 E .iif identical,<X>,<X>, COUNT = COUNT + 1aE .iif identical,<F>,<X>, COUNT = COUNT + 1oE .iif identical,<O>,<X>, COUNT = COUNT + 1 E .iif identical,<O>,<X>, COUNT = COUNT + 1 E .iif identical,<X>,<X>, COUNT = COUNT + 1 E .iif identical,<B>,<X>, COUNT = COUNT + 1 E .iif identical,<A>,<X>, COUNT = COUNT + 1eE .iif identical,<R>,<X>, COUNT = COUNT + 1 E .iif identical,<X>,<X>, COUNT = COUNT + 1 E .iif identical,<X>,<X>, COUNT = COUNT + 1i 0 .print "%integer(COUNT)"7 %MACRO64-I-GENPRINT, Generated PRINT: 5 G This example uses the .IRPC directive to iterate over the C characters in the argument to the X_COUNT macro. Each D time an argument character is X, the variable COUNT isD incremented. After the X_COUNT macro has expanded, theE example uses the %INTEGER() lexical operator to display ! the value of COUNT.e I 5-77_ _ _ Assembler Directives .LIBRARY I _________________________________________________________________t .LIBRARY % Macro library directive Format 9 .LIBRARY quoted-literal1 [quoted-literal2]l Parameters quoted-literal1 E A string enclosed within double quotes that is the file H specification of a macro library. If a logical name existsG and it is the same as the macro library name, specify the H .MLB file extension to override the logical name. For more> information on quoted literals, see Section 2.4. quoted-literal2eC An optional string enclosed within double quotes thatoF specifies a search list of file specifications where theH assembler should look for the specified macro library. TheF assembler successively processes the search list in leftG to right order and attempts to locate the specified macromI library in each location specified in the search list untilnE the macro library is found. If the macro library is notcF found, the assembler issues a diagnostic message. If youE omit the second argument to the .LIBRARY directive, the B assembler uses the following search list by default: 1 o The current device and directoryf G o The device and directory specified by the logical nameo- MACRO64$LIBRARY (if defined)a G o The device and directory specified by the logical namen+ ALPHA$LIBRARY (if defined) G o The device and directory specified by the logical name SYS$LIBRARY ? Logical names may be defined as VMS search lists.t 5-78 I Assembler DirectivesFI .LIBRARY Descriptionw D .LIBRARY adds a name to the macro library list that isB searched whenever a .MCALL or an undefined opcode isD encountered. The libraries are searched in the reverseB order in which they were specified to the assembler. E If you omit any information from the macro-library-name G argument, default values are assumed. The device defaults I to you current default disk; the directory defaults to your G current default directory; the file type defaults to MLB. Examples & .LIBRARY "MY_MACROS" 1 .LIBRARY "PROJ__3 MACROS" "PROJ:[MACRO],PROJ:[DEVELOPMENT]" 2d I 1 The first statement adds the macro library MY_MACROS.MLBlE to the macro library list. The assembler first looks I for MY_MACROS.MLB in the current directory. If not found D there, the assembler next looks in the directory orE directories indicated by the MACRO64$LIBRARY logicalv@ name if it is defined. If not found there or ifC MACRO64$LIBRARY is not defined, the assembler nextoC looks in the directory or directories indicated byfD the ALPHA$LIBRARY logical name if it is defined. IfH not found there or if ALPHA$LIBRARY is not defined, theE assembler next looks in the directory or directoriesoH indicated by the SYS$LIBRARY logical name. If not foundB there, the assembler issues a diagnostic message. < 2 The second statement adds the macro libraryI PROJ_MACROS.MLB to the macro library list. The assembler D first looks for PROJ_MACROS.MLB in the PROJ:[MACRO]H directory. If not found there, the assembler next looksI in the PROJ:[DEVELOPMENT] directory. If not found there,m; the assembler issues a diagnostic message.e I 5-79f 1 8 Assembler Directives .LINKAGE_PAIR I _________________________________________________________________ .LINKAGE_PAIR Linkage directive Format ! .LINKAGE_PAIR nameo Parameters nameD The name of the procedure descriptor, possibly definedF in a different module of the routine to which linkage is required.) Description E .LINKAGE_PAIR causes a linkage pair to be stored at thelF current location counter. A linkage pair consists of twoF quadwords. The first quadword is the code entry point ofE the routine indicated by the specified identifier. (See C .CODE_ADDRESS.) The second quadword is the address of F the procedure descriptor of the routine indicated by theF specified identifier. The second quadword is also calledF the procedure value. The specified name should referenceF a procedure descriptor that is either defined within theC assembly unit or that becomes defined at the time thei program is linked. Notes E o This directive can only be used within data or mixedcA psects (psects that have either the NOEXE or MIX D attributes). For more information, see Section 5.1. G o If automatic data alignment is enabled, this directive I aligns the current location counter to an octaword (128- 9 bit) boundary before allocating storage. 5-80 _ _ I Assembler Directives I .LINKAGE_PAIR Examples K .LINKAGE_PAIR A ; Code address A followed by address of < ; procedure descriptor A I 5-81 Assembler Directives .LISTk I _________________________________________________________________ .LISTa Listing directive Format $ .LIST [argument-list] Parameters argument-listdH One or more of the symbolic arguments defined in Table 5-7I . You can use either the long form or the short form of the F arguments. If multiple arguments are specified, separate0 them with commas, spaces, or tabs. Descriptiona H .LIST is equivalent to .SHOW. See the description of .SHOW# for more information. 5-82 I Assembler DirectivesaI .LOCAL_CODE_ADDRESSe I _________________________________________________________________ .LOCAL_CODE_ADDRESSA 2 Local code address storage directive Format , .LOCAL_CODE_ADDRESS name-list Parameters name-listm@ A list of symbols separated by commas. Each symbolG references a procedure descriptor, defined in the currenth module.f Descriptione B .LOCAL_CODE_ADDRESS causes the code addresses of theE specified identifiers to be placed at the current psectoI and current location counter. The specified identifier mustoI reference a procedure descriptor defined within the module.h Notes E o This directive can only be used within data or mixed A psects (psects that have either the NOEXE or MIX D attributes). For more information, see Section 5.1. G o If automatic data alignment is enabled, this directive B aligns the current location counter to a quadword= (64-bit) boundary before allocating storage.e Examples + .PROCEDURE_DESCRIPTOR P1,C1 .BLKQ 1r; .LOCAL_CODE_ADDRESS P1 ; Code addressoG ; of P1...address of C1... > ; is stored here. I 5-83 u p Assembler Directives .LOCAL_LINKAGE_PAIR, I _________________________________________________________________I .LOCAL_LINKAGE_PAIR % Local linkage directiven Format ' .LOCAL_LINKAGE_PAIR nameo Parameters nameH The name of a procedure descriptor of the routine to whichE linkage is required. The specified procedure descriptor 4 must be defined in the current module. DescriptionC G .LOCAL_LINKAGE_PAIR causes a linkage pair to be stored at H the current location counter. A linkage pair consists of aG code address and the address of the specified identifier.eF The specified name must reference a procedure descriptor7 that is defined within the assembly unit.: Notes E o This directive can only be used within data or mixedeA psects (psects that have either the NOEXE or MIXSD attributes). For more information, see Section 5.1. G o If automatic data alignment is enabled, this directiveiI aligns the current location counter to an octaword (128-u9 bit) boundary before allocating storage.t Examples , .PROCEDURE_DESCRIPTOR P1,CA1 . . . .LOCAL_LINKAGE_ 5 PAIR P1 ; Code address CA1 followed byoP ; procedure descriptor address P1. 5-84 i I Assembler Directives I .LOCAL_PROCEDURE_DESCRIPTOR I _________________________________________________________________ # .LOCAL_PROCEDURE_DESCRIPTORk 5 Procedure descriptor labeling directivei Format ; .LOCAL_PROCEDURE_DESCRIPTOR pd-name, ca-nameC Parameters pd-name G The name of the procedure descriptor. This name can be up D to 31 characters long. It cannot be a temporary label. ca-namehB The name of the code address that corresponds to theC procedure descriptor. This name must be defined later F in the program as a label in a psect that has either theF EXE or MIX attribute, or both. This name can be up to 31> characters long. It cannot be a temporary label. Description B .LOCAL_PROCEDURE_DESCRIPTOR defines a bivalued localG identifier that is used to represent a local routine. The F first value is the procedure value, which is the addressG of the procedure descriptor. This value is defined as the E current location counter at the point where you use thetH .LOCAL_PROCEDURE_DESCRIPTOR directive. The second value isH the code address, which is the code entry-point address ofI the procedure. This value is defined by the second argument I to the .LOCAL_PROCEDURE_DESCRIPTOR directive. No storage isk allocated. Noteso G o See the VMS Calling Standard for a full description oft' procedure descriptors. H o You must specify .LOCAL_PROCEDURE_DESCRIPTOR before the2 code of the routine it describes. I 5-85 u n Assembler Directives# .LOCAL_PROCEDURE_DESCRIPTOR E o See Section 6.7 for a description of the $PROCEDURE_nE DESCRIPTOR and $ROUTINE library macros. These macros G define the procedure identifier and define the storageI. for the procedure descriptor. E o This directive can only be used within data or mixed A psects (psects that have either the NOEXE or MIX A attributes). For more information on psects, see Section 5.1.p G o If automatic data alignment is enabled, this directive.G aligns the current location counter to a quadword (64- H bit) boundary before defining the procedure identifier. Examples 2 .LOCAL_PROCEDURE_DESCRIPTOR LP1,C1 5-86 I Assembler DirectivesrI .LONG I _________________________________________________________________u .LONGr ( Longword storage directive Format $ .LONG expression-list Parameters expression-list : One or more expressions separated by commas. Description G .LONG generates successive longwords (4 bytes) of data in G the object module. The assembler truncates on the left of + an integer or external value.d Notes E o This directive can only be used within data or mixed A psects (psects that have either the NOEXE or MIX D attributes). For more information, see Section 5.3. G o If automatic data alignment is enabled, this directive G aligns the current location counter to a longword (32- 9 bit) boundary before allocating storage.< F o You can define a 32-bit address item using macros and2 the .LONG directive. For example: ' .macro address_32 item .long item ! .endm address_32> Examples = .LONG 4 ; Places 4 in 4 bytes of storage. I 5-87 Assembler Directives .MACRO I _________________________________________________________________ .MACRO ( Macro definition directive Format 7 .MACRO macro-name [formal-argument-list]N . . . range . . . ! .ENDM [macro name]i Parameters macro-nameG The name of the macro to be defined; this name can be anyN4 legal symbol up to 31 characters long. " formal-argument-listE The symbols, separated by commas, to be replaced by ther1 actual arguments in the macro call.i range D The source text to be included in the macro expansion. Descriptionl I .MACRO begins the definition of a macro. It gives the macrofG name and a list of formal arguments. The .MACRO directive H is followed by the source text to be included in the macroE expansion. The .ENDM directive specifies the end of the_ range. I Macro names do not conflict with user-defined symbols. BotheG a macro and a user-defined symbol can have the same name.e 5-88 a I Assembler DirectivesnI .MACROb B When the assembler encounters a .MACRO directive, itD adds the macro name to its macro name table and storesD the source text of the macro (up to the matching .ENDMG directive). No other processing occurs until the macro is expanded.e D The symbols in the formal argument list are associatedE with the macro name and are limited to the scope of the D definition of that macro. For this reason, the symbolsE that appear in the formal argument list can also appeard' elsewhere in the program. 4 For more information, see Section 4.1. Notesn I o If a macro has the same name as an Alpha AXP opcode, the G macro is used instead of the instruction. This feature > allows you to temporarily redefine an opcode. E o A macro can be redefined by using a .MACRO directive F with the same name as in a previous macro definition.G Therefore, the previous macro definition is implicitlyCD deleted and the new macro definition supersedes theI previous definition. See the .MDELETE directive for mored/ information on macro deletion.p F o A macro definition can be nested within another macroE definition. The inner macro is not defined until the ( outer macro is invoked. G o A macro invocation can be nested so that one macro can C invoke another. The assembler suports nested macrotC invocations to a depth of 1000. If a macro invokestH itself, either directly or indirectly, it is recursive.G Recursive macros must specify a basis-step in order to G avoid infinite recursion. A basis-step is a macro exit H condition that will eventually cause the macro to exit,C thus ending the recursion (see the example below). I 5-89u n e Assembler Directives .MACRO Examples Example 1 I This example shows how macro definitions and invocations[E may be nested. It also shows examples of the various E forms of parameter passing. See Section 4.1 for more 4 information on parameter passing. . $ .MACRO OP1 A,B=R4,?C& C: ADDL R2, B, R3 .MACRO OP'Br TRAPBi .ENDM OP'B& ADDL A, R2, R3 OP'B% .MDELETE OP'BL .ENDM OP1a C When OP1 is invoked "OP1 R0", the text expands to:i ' 33000$: ADDL R2, R4, R3B .MACRO OPR4f TRAPBf .ENDML' ADDL R0, R2, R3 OPR4% .MDELETE OPR4 G Processing this text will cause OPR4 to be expanded to / TRAPB; the final text will be:g & 33000$: ADDL R2, R4 R3' ADDL R0, R2, R3 TRAPBi 5-90 I Assembler DirectivesoI .MACROR Example 2 @ The following example shows macro redefinition: ! .MACRO INITIALIZE ; .MACRO INITIALIZE ;Redefine to nothing1# .ENDM INITIALIZEv X=0 Y=1 Z=-1_ .ENDM INITIALIZE C Note that while the redefined version of the macro F immediately supersedes the previous definition in anyF subsequent invocation, the invocation of the original2 definition expands to completion. Example 3 6 This example shows a recursive macro: " .MACRO FACTORIAL NA .IF EQUAL <N>,0 ;Basis step; stop at zerou F=1t .ELSE% FACTORIAL <N-1> ! F = F * <N>o .ENDC .ENDM FACTORIAL I 5-91 t Assembler Directives .MCALL I _________________________________________________________________e .MCALL " Macro call directive Format % .MCALL macro-name-list Parameters macro-name-list H A list of macros to be defined for this assembly. Separate* the macro names with commas. Description G .MCALL specifies the names of the system and user-defineduI macros that are required to assemble the source program buti1 are not defined in the source file. D If any named macro is not found upon completion of theH search (that is, if the macro is not defined in any of theH macro libraries), the assembler displays an error message. Notesi H Using the .MCALL directive is optional unless the macroF name is the same as an opcode or assembler directive.I The assembler automatically searches for a library macro G when it encounters an identifier that is not an opcode E or directive in the opcode field. If your macro name H is the same as an opcode or directive, you must use theH .MCALL directive. You can also use the .MCALL directiveE in your program to document which macros are used bye your program. Examples D .MCALL TRAPB ; Substitute macro in library for6 ; TRAPB instruction 5-92 I Assembler DirectiveseI .MDELETE I _________________________________________________________________v .MDELETE & Macro deletion directive Format ' .MDELETE macro-name-list Parameters macro-name-list G A list of macros whose definitions are to be deleted. You = may separate the macros with a comma or spaces. Description_ C .MDELETE deletes the definitions of specified macros.D D .MDELETE completely deletes the macro. If you delete aE macro that is currently expanding (such as a macro thateI deletes itself), the macro name is immediately removed from H the macro name table and the macro is marked for deletion.? When the macro finishes expanding, it is deleted.t Examples .MACRO FOO% .PRINT "In macro FOO" .ENDM FOO FOO .MDELETE FOO I 5-93r f d Assembler Directives .MEXIT I _________________________________________________________________p .MEXIT " Macro exit directive Format .MEXIT Description C .MEXIT terminates a macro expansion before the end of A the macro. Termination is the same as if .ENDM weredG encountered. You can also use the directive within repeat I blocks. .MEXIT is useful in conditional expansion of macros G and repeat blocks because it bypasses the complexities of I nested conditional directives and alternate assembly paths. Notes D o When .MEXIT occurs in a repeat block, the assemblerC terminates the current repetition of the range and B suppresses further expansion of the repeat range. I o When macros or repeat blocks are nested, .MEXIT exits to 4 the next higher level of expansion. H o If .MEXIT occurs outside a macro definition or a repeat@ block, the assembler displays an error message. Examples Example 1 ? The following macro definition uses the .MEXIToI directive to exit the current macro when it has finishedt: processing a particular kind of argument. 5-94 I Assembler DirectivesdI .MEXITf 2 .macro STORE REG, LOCATION3 .if identical,<REG>,<FP> 1 STQ REG, LOCATION & .mexit! .endcv4 .if identical,<REG>,<SP>1 STQ REG, LOCATION & .mexit! .endc C .if identical,<%extract(0,1,<REG>)>,<R> 1 STQ REG, LOCATIONa& .mexit! .endc C .if identical,<%extract(0,1,<REG>)>,<F> 1 STT REG, LOCATION & .mexit! .endcrH .error "Register argument is not a register"# .endm STORE Example 2 H In this example, the STORE macro (as defined in ExampleH 1) attempts to recognize its REG argument as either FP,H SP, an integer register (a register with R as its firstG letter), or a floating-point register (a register withEG F as its first letter). The following example show twon/ expansions of the STORE macro.c ' STORE R1, 0(SP) 3 .if identical,<R1>,<FP>.! .endc 3 .if identical,<R1>,<SP>o! .endciC .if identical,<%extract(0,1,<R1>)>, <R>a1 .if identical,<R>,<R> - STQ R1, 0(SP)n& .mexit I 5-95O U D Assembler Directives .MEXIT , STORE 24(SP), 16(SP)7 .if identical,<24(SP)>,<FP> ! .endc 7 .if identical,<24(SP)>,<SP> ! .endcuF .if identical,<%extract(0,1,<24(SP)>)>,<R>1 .if identical,<2>,<R> ! .endcdF .if identical, <%extract(0,1<24(SP)>)>,<F>1 .if identical,<2>,<F>.! .endceH .error "Register argument is not a register" E The first call of the STORE macro above stores R1 at I 0(SP). The STORE macro determines to do an integer storePG by recognizing the letter R as the first letter of theeH register name. Once it has done so, it abandons furtherC expansion of the macro using the .MEXIT directive.LB The second invocation attempts to store 24(SP) atE 16(SP). Since the STORE macro cannot identify 24(SP)UE as a legitimate register in any of the four forms itrF recognizes, the STORE macro does not attempt to storeF the REG argument, and does not abandon expansion withG the .MEXIT directive. Instead, the STORE macro expandsE1 and issues a diagnostic message.f 5-96 I Assembler Directives I .NARG I _________________________________________________________________s .NARG + Number of arguments directive Format .NARG symbol Parameters symbolF A symbol that is assigned a value equal to the number of5 positional arguments in the macro call. Description_ E .NARG determines the number of arguments in the currentr macro call.s H .NARG counts all the positional arguments specified in theI macro call, including null arguments (specified by adjacentrF commas). The value assigned to the specified symbol doesG not include any keyword arguments or any formal arguments ' that have default values.t Notese I If .NARG appears outside a macro, the assembler displays " an error message. Examples Example 1 1 The macro definition is as follows: I .MACRO CNT_ARG A1,A2,A3,A4,A5,A6,A7,A8,A9=DEF9,A10=DEF10 G .NARG COUNTER ; COUNTER is set to no. of ARGS @ .WORD COUNTER ; Store value of COUNTER .ENDM CNT_ARGf I 5-97 n a Assembler Directives .NARG Example 2. D The macro calls and expansions of the macro previously% defined are as follows: : CNT_ARG TEST,FIND,ANS ; COUNTER will = 3G .NARG COUNTER ; COUNTER is set to no. of ARGS @ .WORD COUNTER ; Store value of COUNTER : CNT_ARG ; COUNTER will = 0G .NARG COUNTER ; COUNTER is set to no. of ARGSn@ .WORD COUNTER ; Store value of COUNTER D CNT_ARG TEST,A2=SYMB2,A3=SY3 ; COUNTER will = 1G .NARG COUNTER ; COUNTER is set to no. of ARGS.@ .WORD COUNTER ; Store value of COUNTERK ; Keyword arguments are not countedr : CNT_ARG ,SYMBL,, ; COUNTER will = 4G .NARG COUNTER ; COUNTER is set to no. of ARGS @ .WORD COUNTER ; Store value of COUNTERD ; Null arguments are counted 5-98 r x I Assembler Directives I .NCHRe I _________________________________________________________________i .NCHRc , Number of characters directive Format $ .NCHR symbol,<string> Parameters symbolF A symbol that is assigned a value equal to the number of; characters in the specified character string. <string>G A sequence of printable characters. Delimit the character F string with angle brackets (<>) (or a character precededB by a circumflex (^)) only if the specified characterF string contains a legal separator (comma, space, or tab)! or a semicolon (;).c Description F .NCHR determines the number of characters in a specifiedE character string. It can appear anywhere in an MACRO-64rF program and is useful in calculating the length of macro arguments. Notesh H You can use the %LENGTH lexical operator instead of the! .NCHR directive. Examples Example 1 4 The macro definition is as follows: G .MACRO CHAR MESS ; Define MACROpQ .NCHR CHRCNT,<MESS> ; Assign value to CHRCNT F .WORD CHRCNT ; Store valueK .ASCII "MESS" ; Store characters A .ENDM CHAR ; Finish I 5-99 d Assembler Directives .NCHR Example 2 G The macro calls and expansions of the macro previously ( defined are as follows: I CHAR <HELLO> ; CHRCNT will = 5.P .NCHR CHRCNT,<HELLO> ; Assign value to CHRCNTF .WORD CHRCNT ; Store valueK .ASCII "HELLO" ; Store characters O CHAR <14, 75.39 4> ; CHRCNT will = 12(dec) P .NCHR CHRCNT,<14, 75.39 4> ; Assign value to CHRCNTF .WORD CHRCNT ; Store valueK .ASCII "14, 75.39 4" ; Store charactersi 5-100g e c I Assembler Directives I .NLISTe I _________________________________________________________________ .NLIST ) Listing exclusion directiven Format % .NLIST [argument-list]x Parameters argument-listaG One or more of the symbolic arguments listed in Table 5-7 A . Use either the long form or the short form of the D arguments. If you specify multiple arguments, separate0 them with commas, spaces, or tabs. Description E .NLIST is equivalent to .NOSHOW. See the description of ) .SHOW for more information. I 5-101 Assembler Directives .NOSHOW I _________________________________________________________________ .NOSHOW ) Listing exclusion directive Format & .NOSHOW [argument-list] Parameters argument-list G One or more of the symbolic arguments listed in Table 5-7,F in the description of .SHOW. Use either the long form orF the short form of the arguments. If you specify multipleD arguments, separate them with commas, spaces, or tabs. Description @ .NOSHOW specifies listing control options. See the8 description of .SHOW for more information. 5-102 h 1 I Assembler Directives I .OCTA I _________________________________________________________________i .OCTAe ( Octaword storage directive Format $ .OCTA expression-list Parameters expression-list.G A list of constant values separated by commas. Each valueeG results in a 64-bit value being sign-extended to 128 bits.( and stored in an octaword. Description A .OCTA generates 128 bits (16 bytes) of binary data.N Notes H o The low quadword contains the specified constant value. E o The high quadword contains the sign extension of the E specified constant value. That is, the high quadwordeF contains zero if the specified value is positive, andH it contains all bits set to 1 if the specified value is negative. E o This directive can only be used within data or mixed A psects (psects that have either the NOEXE or MIX D attributes). For more information, see Section 5.1. G o If automatic data alignment is enabled, this directivenI aligns the current location counter to an octaword (128-d9 bit) boundary before allocating storage. I 5-103t t Assembler Directives .OCTAn Examples 9 .OCTA 0 ; OCTA 0rK .OCTA ^X01234ABCD5678F9 ; OCTA hex value specified D .OCTA VINTERVAL ; VINTERVAL has 64- bit value, B ; sign-extended 5-104d I Assembler Directives I .ODD I _________________________________________________________________o .ODD 6 Odd location counter alignment directive Format .ODD Description I .ODD ensures that the current value of the location counterBE is odd by adding 1 if the current value is even. If the ? current value is already odd, no action is taken. Notes E This directive can be used only within data or mixed_A psects (psects that have either the NOEXE or MIXED attributes). For more information, see Section 5.3. I 5-105w e f Assembler Directives .PACKEDe I _________________________________________________________________ .PACKEDs 5 Packed decimal string storage directiveM Format . .PACKED decimal-string[,symbol] Description G .PACKED is supplied as a library macro with MACRO-64. Forn. more information, see Chapter 6. 5-106n e c I Assembler DirectivesCI .PAGE I _________________________________________________________________ .PAGE % Page ejection directive Format .PAGEe DescriptionM C .PAGE forces a new page in the listing. The directive_F itself is also printed in the listing and begins the new page. @ .PAGE in a macro definition is ignored. The pagingD operation is performed only during macro expansion. IfE the .PAGE directive occurs during macro expansion, textYG beginning with the original macro invocation line appearsT' at the top of a new page.c Examples < .MACRO SKIP ; macro definition with .PAGE .PAGE ; .ENDM SKIP ; .PSECT A,NOEXE ; .BLKW 10 ;A SKIP ; In the listing file, a form feede7 ; will be inserted hereu I 5-107T i t Assembler Directives .PRINT I _________________________________________________________________t .PRINT ( Assembly message directive Format $ .PRINT quoted-literal Parameters quoted-literalG The string of characters enclosed in quotes are displayedhG when encountered during assembly. For more information on / quoted literals, see Section 2.4. Description E .PRINT causes the assembler to display an informational : message. The message consists of the string. Notesc F o .PRINT, .WARN, and .ERROR are directives that displayC messages. You can use these to display information E indicating unexpected or important conditions within the assembly. B o This directive also accepts VAX MACRO syntax. SeeC VAX MACRO and Instruction Set Reference Manual for details. Examples + .PRINT "Questionable usage" ^TH %MACRO64-I-GENPRINT, Generated PRINT: Questionable usage@ at line number 3 in file DISK$:[TEST]PRINT.M64;2 5-108 I Assembler Directives I .PROCEDURE_DESCRIPTOR I _________________________________________________________________E .PROCEDURE_DESCRIPTORh 5 Procedure descriptor labeling directive Format 5 .PROCEDURE_DESCRIPTOR pd-name, ca-name Parameters pd-nametG The name of the procedure descriptor. This name can be up(D to 31 characters long. It cannot be a temporary label. ca-name B The name of the code address that corresponds to theC procedure descriptor. This name must be defined later<F in the program as a label in a psect that has either theF EXE or MIX attribute, or both. This name can be up to 31> characters long. It cannot be a temporary label. Descriptioni H .PROCEDURE_DESCRIPTOR defines a bivalued global identifierC that is used to represent a global routine. The first C value is the procedure value, which is the address of D the procedure descriptor. This value is defined as theE current location counter at the point where you use the>E .PROCEDURE_DESCRITOR directive. The second value is the H code address, which is the code entry-point address of theE procedure. This value is defined by the second argument C to the .PROCEDURE_DESCRIPTOR directive. No storage is allocated. Notesd G o See the VMS Calling Standard for a full description ofi' procedure descriptors. G o You must specify .PROCEDURE_DESCRIPTOR before the codes- of the routine it describes.S I 5-109n t o Assembler Directives .PROCEDURE_DESCRIPTOR 9 o See Section 6.7 for a description of theiD $PROCEDURE_DESCRIPTOR and $ROUTINE library macros.H These macros define the procedure identifier and define: the storage for the procedure descriptor. E o This directive can only be used within data or mixedeA psects (psects that have either the NOEXE or MIX,C attributes). For more information, see Section 5.1a G o If automatic data alignment is enabled, this directiveiG aligns the current location counter to a quadword (64- H bit) boundary before defining the procedure identifier. 5-110r i I Assembler Directives I .PSECT I _________________________________________________________________m .PSECT * Program sectioning directive Format : .PSECT program-section-name[,argument-list] Parameters " program-section-nameF The name of the program section. For more information on& psects, see Section 5.1. argument-list F A list containing the program section attributes and theG program section alignment. Table 5-5 lists the attributes I and their functions. Table 5-6 lists the default attributes H and their opposites. Program sections are aligned when youH specify an integer in the range 0 to 16 or one of the fiveG keywords listed in the following table. If you specify an I integer, the program section is linked to begin at the next E virtual address that is a multiple of two raised to the_I power of the integer. If you specify a keyword, the programND section is linked to begin at the next virtual addressI that is a multiple of the corresponding value listed in the following table: I ___________________________________________________________bI Keyword_Size_(in_Bytes)____________________________________ BYTE 20= 1s WORD 21= 2 LONG 22= 4 QUAD 23= 8T OCTA 24= 16I ___________________________________________________________ " QUAD is the default. I 5-111 l 0 Assembler Directives .PSECT I Table_5-5_Program_Section_Attributes_____________________________f I AttributFunction_________________________________________________= H ABS Absolute-The program section has an absolute address. AnF absolute program section contributes no binary code toG the image, so its byte allocation request to the linker G is zero. You cannot store initial values in an absolute E program section with directives such as .BYTE, .WORD, E .LONG, QUAD. Usually the .BLKx directives are used in E conjunction with label definitions within an absolute C program section to define symbolic offsets within a D structure. Compare this attribute with its opposite, REL. C CON Concatenate-Program sections with the same name and_A attributes (including CON) from other modules areHC concatenated into one program section at link time. E Their contents are concatenated in the order in which G the linker acquires them. The allocated virtual addresseI space is the sum of the individual requested allocations. > Compare this attribute with its opposite, OVL. E EXE Executable-The program section contains instructions.sD This attribute provides the capability of separatingD instructions from read-only and read/write data. TheH linker uses this attribute in gathering program sectionsC and in verifying that the transfer address is in anaE executable program section. The assembler only allowswG you to place instructions in a program section that hastF either or both the EXE or MIX attributes. Compare this3 attribute with its opposite, NOEXE.n C GBL Global-Program sections that have the same name and F attributes will be combined at link time into a singleI program section even when the individual program sectionsfF are in different clusters. This attribute is specifiedG for FORTRAN COMMON block program sections. Compare this 1 attribute with its opposite, LCL.S G LCL Local-The program section is restricted to its cluster.A> Compare this attribute with its opposite, GBL. I (continued on next page)s 5-112s I Assembler DirectiveseI .PSECT: I Table_5-5_(Cont.)_Program_Section_Attributes_____________________ I AttributFunction_________________________________________________C A MIX Mix-The program section can contain both data andvH instructions. The MIX and NOMIX attributes are assembly-G time attributes which only affect assembler processing.CH The MIX and NOMIX attributes do not appear in the objectB module and do not affect linker processing. To mixI instructions and data in the same psect, you must specify I the MIX attribute. The NOMIX attribute is the default. If G you choose to use instructions in a NOEXE psect, or use I data directives in an EXE psect, you must specify the MIX I attribute. The following limitations apply when using the_ MIX attribute: E o Optimizations and alignment of code labels are notoH performed. Optimizations and code-label alignment areG not performed on instructions placed in a MIX psect,rE regardless of whether the .ENABLE/.DISABLE optionssF have been set, or if command line /OPTIMIZATION and? /ALIGNMENT=code options have been specified.m H o Limited debugging information is provided. No PC-lineI (program counter) correlation information is generatedo> for the instructions placed in a MIX psect. @ o The listing file includes only binary encodedF instructions. All instructions that are placed in aF MIX psect appear in the machine-code portion of theI listing file, in their binary encoded instruction form E as a data initializer to the .LONG data directive.a E There are no restrictions on data directives. CompareeA this attribute with its opposite, NOMIX. For morer- information, see Section 5.1. I NOEXE Not Executable-The program section contains data only; it H does not contain instructions. The assembler only allowsG you to place data allocations in a program section thateH has either or both the NOEXE and MIX attributes. Compare6 this attribute with its opposite, EXE. I (continued on next page)c I 5-113O Assembler Directives .PSECT I Table_5-5_(Cont.)_Program_Section_Attributes_____________________r I AttributFunction_________________________________________________a H NOMIX The default attribute when you use the .PSECT directive.> Compare this attribute with its opposite, MIX. G NOPIC Non-Position-Independent Content-The program section isdG assigned to a fixed location in virtual memory (when it I is in a shareable image). Compare this attribute with itsT opposite, PIC. A NORD Nonreadable-Reserved for future use. Compare this 0 attribute with its opposite, RD. H NOSHR No Share-The program section is reserved for private useI at execution time by the initiating process. Compare thisa1 attribute with its opposite, SHR. G NOWRT Nonwriteable-The contents of the program section cannotiI be altered (written into) at execution time. Compare thisE1 attribute with its opposite, WRT.f ? OVR Overlay-Program sections with the same name andiE attributes (including OVR) from other modules receive I the same relocatable base address in memory at link time.rD The allocated virtual address space is the requestedE allocation of the largest overlaying program section.b> Compare this attribute with its opposite, CON. G PIC Position-Independent Content-The program section can be I relocated; that is, it can be assigned to any memory areaOI (when it is in a shareable image). Compare this attribute ) with its opposite, NOPIC. H RD Readable-Reserved for future use. Compare this attribute( with its opposite, NORD. D REL Relocatable-The linker assigns the program section aE relocatable base address. The contents of the program H section can be code or data. Compare this attribute with" its opposite, ABS. I (continued on next page) 5-114 I Assembler DirectivesoI .PSECTI I Table_5-5_(Cont.)_Program_Section_Attributes_____________________ I AttributFunction_________________________________________________ D SHR Share-The program section can be shared at executionF time by multiple processes. This attribute is assignedH to a program section that can be linked into a shareableG image. Compare this attribute with its opposite, NOSHR. H WRT Write-The contents of the program section can be alteredH (written into) at execution time. Compare this attributeI ________with_its_opposite,_NOWRT.________________________________ I Table_5-6_Default_Program_Section_Attributes_______________ DefaultaI Attribute_____Opposite_Attribute___________________________p CON OVR ! EXE NOEXE LCL GBL NOMIX MIX NOPIC PIC NOSHR SHR RD NORD REL ABS_ I WRT___________NOWRT________________________________________a Description E .PSECT defines a program section and its attributes and I refers to a program section once it is defined. Use program + sections to do the following:e * o Develop modular programs. 1 o Separate instructions from data.i A o Allow different modules to access the same data. C o Protect read-only data and instructions from beingx modified. I o Identify sections of the object module to the linker and the debugger. I 5-115 G Assembler Directives .PSECT G o Control the order in which program sections are stored # in virtual memory.e C When the assembler encounters a .PSECT directive thatuD specifies a new program section name, it creates a newB program section and stores the name, attributes, andF alignment of the program section. The assembler includesG all data or instructions that follow the .PSECT directiveeH in that program section until it encounters another .PSECTG directive. The assembler starts all program sections at a - relative location counter of 0.r A The assembler does not automatically define programeG sections. Any code or data placed before the first .PSECT F directive in the source code produces an assembly error. A If the assembler encounters a .PSECT directive thateI specifies the name of a previously defined program section,eC it stores the new data or instructions after the lasteC entry in the previously defined program section, evenmE with program sections that have the OVR attribute. (OVR D program sections from separate modules are overlaid byH the linker. The OVR attribute does not affect how multipleD contributions to a psect are processed within a singleF assembly unit.) You need not re-list the attributes whenG continuing a program section, but any attributes that areDG listed must be the same as those previously in effect for F the program section. A continuation of a program sectionI cannot contain attributes conflicting with those specified, = or defaulted, in the original .PSECT directive._ H The attributes listed in the .PSECT directive describe theE contents of the program section. Except for the EXE andEF NOEXE attributes, the assembler does not check to ensureF that the contents of the program section actually adhereF to the attributes listed. However, the assembler and theE linker do check that all program sections with the samedF name have exactly the same attributes. The assembler andD linker display an error message if the program section, attributes are not consistent. D Program section names are independent of local symbol,B global symbol, and macro names. You can use the sameI symbolic name for a program section and for a local symbol,iF global symbol, or macro name. You may want to use unique4 names for clarity and maintainability. 5-116 I Assembler DirectivesnI .PSECTE Notes I o The .ALIGN directive cannot specify an alignment greater-H than that of the current program section; consequently,F .PSECT should specify the largest alignment needed in% the program section.e H o For efficiency of execution and ease of programming, anG alignment of quadword or larger is recommended for all : program sections that have quadword data. Examples 2 .PSECT A,QUAD,EXE ; Code psect I 5-117 Assembler Directives .QUADh I _________________________________________________________________ .QUADm ( Quadword storage directive Format $ .QUAD expression-list Parameters expression-lists: One or more expressions separated by commas. DescriptionE ? .QUAD generates 64 bits (8 bytes) of binary data.e Notes E o This directive can only be used within data or mixed A psects (psects that have either the NOEXE or MIX D attributes). For more information, see Section 5.1. G o If automatic data alignment is enabled, this directive B aligns the current location counter to a quadword= (64-bit) boundary before allocating storage. Examples A:: .QUAD 4 5-118_ _ _ I Assembler DirectivescI .REPEATe I _________________________________________________________________ .REPEAT .REPT. $ Repeat block directive Format .REPEAT expression . . . range . . . .ENDR Parameters expressionD An expression whose value controls the number of timesC the range is to be assembled within the program. WhenoF the expression is less than or equal to zero, the repeatE block is not assembled. The expression must be absolutetH or relocatable and must not contain any undefined symbols.H The assembler converts a relocatable value to the relative& offset within the psect. rangea@ The source text to be repeated the number of timesH specified by the value of the expression. The repeat blockC can contain macro definitions or other repeat blocks._E .MEXIT is legal within the range and causes the current_; and all succeeding repetitions to be aborted. I 5-119 A Assembler Directives .REPEAT_ Description C .REPEAT repeats a block of code a specified number of H times, in line with other source code. The .ENDR directive- specifies the end of the range.l Notest 8 The alternate form of .REPEAT is .REPT. Examples Example 1 @ The following macro definition uses the .REPEATF directive to store an ASCII string a specified number3 of times, followed by a zero byte:t * .MACRO COPIES STRING,NUM .REPEAT NUMo .ASCII "STRING" .ENDRr .BYTE 0e .ENDM COPIES Example 2 C The following macro call stores five copies of thebG string "ABCDEF". This example is divided into 4 parts:f " Macro invocation: " COPIES <ABCDEF>,5 7 Macro expansion of .REPEAT invocation:O .REPEAT 5o" .ASCII "ABCDEF" .ENDRd # .REPEAT expansion: " .ASCII "ABCDEF"" .ASCII "ABCDEF"" .ASCII "ABCDEF"" .ASCII "ABCDEF"" .ASCII "ABCDEF" 5-120 r e I Assembler Directives I .REPEAT ( End of macro expansion: .BYTE 0s Example 3 D The following macro call stores three copies of theH string "How Many Times". This example is divided into 4 parts: " Macro invocation: VARB = 3/ COPIES <How Many Times>,VARBb 7 Macro expansion of .REPEAT invocation: .REPEAT VARB* .ASCII "How Many Times" .ENDRs # .REPEAT expansion:p ( .ASCII "How Many Times"* .ASCII "How Many Times"* .ASCII "How Many Times" ( End of macro expansion: .BYTE 0n I 5-121 Assembler Directives .RESTORE_PSECT I _________________________________________________________________ .RESTORE_PSECT .RESTORE @ Restore previous program section context directive Format .RESTORE_PSECT .RESTORE Description G .RESTORE_PSECT retrieves the program section from the toprH of the program section context stack, an internal stack inF the assembler. If the stack is empty when .RESTORE_PSECTF is issued, the assembler displays an error message. WhenE .RESTORE_PSECT retrieves a program section, it restores G the current location counter to the value it had when thetG program section was saved. The maximum stack level is 50. F See the description of .SAVE_PSECT for more information. Notes B o The alternate form of .RESTORE_PSECT is .RESTORE. D o You cannot use .RESTORE_PSECT to overwrite previousH data storage initializations. In the following example,E MACRO-64 attempts to store "42" over "43" and fails,s+ resulting in a diagnostic. .PSECT Ac .SAVE PSECT .PSECT Am .QUAD 43h .RESTORE PSECT: .QUAD 42m 5-122o l l I Assembler DirectivesaI .RESTORE_PSECTr Examples ' .PSECT A,QUAD,NOEXEi A1: .WORD 5 A2: .QUAD 6/> .SAVE_PSECT ; Saves psect A context' .PSECT B,QUAD,NOEXEL B1: .WORD 6iB .RESTORE_PSECT ; Return A location counter A3: .WORD 5o ' .PSECT B,QUAD,NOEXEl 1$: .WORD 5 .SAVE LOCAL_4 BLOCK ; Saves psect B context and temporary6 ; label context " .PSECT C,NOEXE 1$: .WORD 6 C .RESTORE_PSECT ; Restores psect B and saves 6 ; label contextB .ADDRESS 1$ ; References the address ofC ; psect B temporary label 1$ I 5-123n n n Assembler Directives .SAVE_PSECTc I _________________________________________________________________ .SAVE_PSECTa .SAVE < Save current program section context directive Format ( .SAVE_PSECT [LOCAL_BLOCK] " .SAVE [LOCAL_BLOCK] Description G .SAVE_PSECT stores the current program section context on G the top of the program section context stack, an internal D assembler stack. It leaves the current program sectionF context in effect. The program section context stack canE hold up to 50 entries. Each entry includes the value ofSI the current location counter and the maximum value assignedIH to the location counter in the current program section. IfI the stack is full when .SAVE_PSECT is encountered, an errori occurs.i A If the LOCAL_BLOCK option is specified, the current E temporary label block is saved with the current program. section context. E .SAVE_PSECT and .RESTORE_PSECT are especially useful insI macros that define program sections. See the description ofe> .RESTORE_PSECT for an example using .SAVE_PSECT. NotesS < The alternate form of .SAVE_PSECT is .SAVE. 5-124 a r I Assembler DirectiveshI .S_FLOATINGa I _________________________________________________________________e .S_FLOATINGh G Single-precision IEEE floating-point arithmetic directiveh Format 5 .S_FLOATING floating-point-number-listi Parameters ( floating-point-number-listF A list of IEEE single-precision floating-point constantsI separated by commas. For more information on floating-point;) numbers, see Section 2.3.2.a Descriptionw @ .S_FLOATING evaluates the specified floating-pointD constants and stores the results in the object module.G .S_FLOATING generates 32-bit, single-precision, floating-,H point data (1 bit of sign, 8 bits of exponent, and 23 bitsE of fractional significance). See the description of .T_gB FLOATING for information on storing double-precisionA floating-point IEEE numbers and the descriptions of H .D_FLOATING, .F_FLOATING, and .G_FLOATING for descriptions. of other floating-point numbers. Notes E o This directive can only be used within data or mixed A psects (psects that have either the NOEXE or MIXeD attributes). For more information, see Section 5.3. G o If automatic data alignment is enabled, this directive G aligns the current location counter to a longword (32-p9 bit) boundary before allocating storage.o Examples ) .S_FLOATING 2.0,3.0,4.405 I 5-125 t h Assembler Directives .SHOW I _________________________________________________________________ .SHOWo .NOSHOW_ 8 Listing inclusion and exclusion directives Format $ .SHOW [argument-list] & .NOSHOW [argument-list] Parameters [argument-list] B Use one or more of the symbolic arguments defined inF Table 5-7. You can use either the long form or the shortG form of the arguments. If you specify multiple arguments, F you must separate them by commas. If any argument is notC specifically included in a listing control statement,_F the assembler assumes its default value (show or noshow), throughout the source program. I Table_5-7_.SHOW_and_.NOSHOW_Symbolic_Arguments___________________ Short I Long_Form_______Form______Default___Function_____________________t I CONDITIONALS CND Noshow Lists unsatisfied conditionale@ code associated withD the conditional assembly7 directives. H EXPANSIONS ME Noshow Lists macro and repeat range7 expansions.e > LIBRARY None Noshow Includes the macroD definitions in a library; in the listing. F INCLUDE None Noshow Lists include file text inI ____________________________________the_listing_file.____________i 5-126 l I Assembler Directives I .SHOWa Descriptionr F .SHOW and .NOSHOW specify listing control options in theE source text of a program. You can use .SHOW and .NOSHOW / with or without an argument list.f I When you use them with an argument list, .SHOW includes andtF .NOSHOW excludes the lines specified in Table 5-7. .SHOWF and .NOSHOW control the listing of the source lines thatH are in conditional assembly blocks (see the description ofH .IF), macros, and repeat blocks. When you use them withoutH arguments, these directives alter the listing level count.D The listing level count is initialized to 0. Each timeD .SHOW appears in a program, the listing level count isF incremented; Each time .NOSHOW appears in a program, the1 listing level count is decremented.o F When the listing level count is negative, the listing isH suppressed, unless the line contains an error. Conversely,C when the listing level count is positive, the listingpC is generated. When the count is 0, the line is eitheruI listed or suppressed, depending on the value of the listingt) control symbolic arguments.r Notese C o The listing level count allows macros to be listedtD selectively; a macro definition can specify .NOSHOWH at the beginning to decrement the listing count and canI specify .SHOW at the end to restore the listing count too$ its original value. G o The alternate forms of .SHOW and .NOSHOW are .LIST and .NLIST. E o The initial setting for each list .LIST/.SHOW optioneD is obtained from the command line setting using theF /SHOW qualifier. For more information on command line- qualifiers, see Section 1.2. I 5-127s h a Assembler Directives .SHOWo Examples H .NOSHOW ; Turn off listing file display. Counter < 0. . . .G .SHOW ; Turn on listing file display. Counter = 0.A< ; Value of .SHOW options are used. . . .A .SHOW ; Counter > 0. Listing file display is-E ; on for all options regardless of setting.l . . . 5-128r f e I Assembler DirectivesoI .SIGNED_BYTEt I _________________________________________________________________ .SIGNED_BYTE + Signed byte storage directive Format + .SIGNED_BYTE expression-list Parameters expression-listsG An expression or list of expressions separated by commas._G Each expression specifies a value to be stored. The valuee5 must be in the range -128 through +127.p Description G .SIGNED_BYTE generates successive bytes of binary data inrI the object module and performs signed range checking. Apart G from the range check, .SIGNED_BYTE is equivalent to .BYTE % for storage allocation.d Notesi E This directive can only be used within data or mixed A psects (psects that have either the NOEXE or MIX.D attributes). For more information, see Section 5.1. Examples .PSECTA,NOEXEc> .SIGNED_BYTE LABEL1-LABEL2 ; Data must fit= .SIGNED_BYTE -126 ; in a byte I 5-129 Assembler Directives .SIGNED_WORD I __________________________________________________________________ .SIGNED_WORD + Signed word storage directiveA Format + .SIGNED_WORD expression-list Parameters expression-list G An expression or list of expressions separated by commas. G Each expression specifies a value to be stored. The value ; must be in the range -32,768 through +32,767. Descriptions G .SIGNED_WORD generates successive words of binary data in I the object module and performs signed range checking. ApartsG from the range check, .SIGNED_WORD is equivalent to .WORD - in terms of storage allocation. Notesd E o This directive can only be used within data or mixedvA psects (psects that have either the NOEXE or MIXiD attributes). For more information, see Section 5.1. G o If automatic data alignment is enabled, this directivehG aligns the current location counter to a word (16-bit)n4 boundary before allocating storage. Examples " .PSECT $DATA,NOEXE$ .SIGNED_WORD -32766;9 .SIGNED_WORD 32769 ;causes assembly error 5-130 - I Assembler Directives I .SUBTITLEu I _________________________________________________________________i .SUBTITLEc .SBTTL ( Listing subtitle directive Format ' .SUBTITLE quoted-literalm $ .SBTTL quoted-literal Parameters quoted-literalH An ASCII string enclosed in quotes from 1 to 31 charactersI long; excess characters are truncated. For more information 2 on quoted literals, see Section 2.4. Description F .SUBTITLE causes the assembler to print the line of textI as the subtitle on the second line of each assembly listing D page. This subtitle text is printed on each page untilI altered by a subsequent .SUBTITLE directive in the program. Noteso ; o The alternate form of .SUBTITLE is .SBTTL.c oP B o This directive also accepts VAX MACRO syntax. SeeC VAX MACRO and Instruction Set Reference Manual fore details. Examples 5 .SUBTITLE "Terminal Display Routines"A I 5-131" Assembler Directives .T_FLOATING I _________________________________________________________________ .T_FLOATING G Double-precision IEEE floating-point arithmetic directive Format 5 .T_FLOATING floating-point-number-listo Parameters ( floating-point-number-listF A list of IEEE double-precision floating-point constantsI separated by commas. For more information on floating point ) numbers, see Section 2.3.2.b Descriptionc @ .T_FLOATING evaluates the specified floating-pointD constants and stores the results in the object module.G .T_FLOATING generates 64-bit, double-precision, floating-MI point data (1 bit of sign, 11 bits of exponent, and 52 bitsAE of fractional significance). See the description of .S_ B FLOATING for information on storing single-precisionE floating-point IEEE numbers and the descriptions of .D_ F FLOATING, F_FLOATING, and G_FLOATING for descriptions of+ other floating-point numbers._ Notes_ E o This directive can only be used within data or mixed A psects (psects that have either the NOEXE or MIX D attributes). For more information, see Section 5.1. G o If automatic data alignment is enabled, this directive B aligns the current location counter to a quadword= (64-bit) boundary before allocating storage.w Examples ( .T_FLOATING 4.5036,6.034 5-132 n I Assembler Directives I .TITLE I _________________________________________________________________h .TITLE % Listing title directive Format 3 .TITLE module-name ["listing-title"]T Parameters module-nameEI Either a quoted literal or an identifier that specifies the F module's title. For more information on quoted literals, see Section 2.4. "listing-title"cF Optional quoted literal that specifies a title to appearI within the first line of each listing output file. For moreS> information on quoted literals, see Section 2.4. Description 9 .TITLE assigns a name to the object module.a Notes ? o The module name specified with .TITLE bears no E relationship to the file specification of the object G module, as specified in the MACRO-64 command line. The_F object module name appears in the linker load map andH is also the module name that the debugger and librarian recognize.r I o If .TITLE is not specified, MACRO-64 assigns the default E name (.MAIN.) to the object module. If more than oneLI .TITLE directive is specified in the source program, the G last .TITLE directive encountered establishes the name . for the entire object module. B o This directive also accepts VAX MACRO syntax. SeeC VAX MACRO and Instruction Set Reference Manual forf details. I 5-133 Assembler Directives .TITLE Examples 0 .TITLE "MAIN" "Main Entry Point" 5-134_ _ _ I Assembler DirectivesoI .WARNC I _________________________________________________________________ .WARNS Warning directivem Format # .WARN quoted-literalm Parameters quoted-literalG The string of characters enclosed in quotes are displayedxG during assembly. For more information on quoted literals, see Section 2.4. Description H .WARN causes the assembler to display a warning message onG the terminal or in the batch log file, and in the listing % file (if there is one).E Notes, F o .PRINT, .WARN, and .ERROR are directives that displayB messages. You can use them to display information? indicating an unexpected or important assembly. condition. B o This directive also accepts VAX MACRO syntax. SeeC VAX MACRO and Instruction Set Reference Manual forS details.s Examples : .WARN "Illegal parameter value; 0 assumed" ^ %MACRO64-W- F GENWARN, Generated WARNING: Illegal parameter value; 0 assumed? at line number 3 in file DISK$:[TEST]WARN.M64;2 I 5-135_ _ _ Assembler Directives .WEAK I _________________________________________________________________m .WEAK - Weak symbol attribute directive Format .WEAK symbol-list Parameters symbol-list 8 A list of identifiers separated by commas. Description H .WEAK specifies symbols that are either defined externallyB in another module or defined globally in the currentD module. .WEAK suppresses any object library search for the symbol.S F When .WEAK specifies a symbol that is not defined in theF current module, the symbol is externally defined. If theE linker finds the symbol's definition in another module, E it uses that definition. If the linker does not find anaI external definition, the symbol has a value of zero and the,I linker does not report an error. The linker does not searchtI a library for the symbol, but if a module brought in from a H library for another reason contains the symbol definition,! the linker uses it. B When .WEAK specifies a symbol that is defined in theE current module, the symbol is considered to be globally G defined. However, if this module is inserted in an object C library, this symbol is not inserted in the library'sxG symbol table. Consequently, searching the library at link I time to resolve this symbol does not cause the module to be included.v 5-136 _ I Assembler Directives I .WEAKs Examples .WEAK A,B,C E A:: .WORD 5 ; A and B are weak global definitions B:: .QUAD 6 @ .ADDRESS C ; C is a weak external reference I 5-137 o Assembler Directives .WORDn I _________________________________________________________________ .WORDb $ Word storage directive Format $ .WORD expression-list Parameters expression-list_C One or more expressions separated by commas. For more_D information on terms and expressions, see Section 2.8. Description G .WORD generates successive words (2 bytes) of data in the object module. Notes A o The expression is first evaluated as a quadword, F then truncated to a word. The value of the expressionH should be in the range of -32,768 to +32,767 for signedE data or 0 to 65,535 for unsigned data. The assemblerwC displays an error if the high-order 6 bytes of theeD quadword expression have a value other than zero or ^XFFFFFFFFFFFF. E o The assembler truncates on the left of an integer or_ external value. 6 o Addresses are not allowed with .WORD. E o This directive can only be used within data or mixed A psects (psects that have either the NOEXE or MIX D attributes). For more information, see Section 5.1. G o If automatic data alignment is enabled, this directiveoG aligns the current location counter to a word (16-bit)g4 boundary before allocating storage. 5-138H i u I Assembler Directives I .WORD Examples .WORD 5,6,7e I 5-139n I 6nI _________________________________________________________________e I MACRO-64 Supplied Macrose E This chapter describes the library macros supplied withtG the MACRO-64 assembler. These macros make it easy for youeH to define and call routines. For a detailed description of* each macro, see Section 6.7. H This chapter also describes the $OPDEF macro, which allows- you to define your own opcodes.t % 6.1 MACRO-64 Supplied Librarye F The MACRO-64 assembler provides a library of macros thatI help you to program in conformance with the OpenVMS Calling F Standard. This library also allows you to define you ownH opcodes. This library, called MACRO64.MLB, is installed on6 your system with the MACRO-64 assembler. G The MACRO-64 assembler automatically adds the MACRO64.MLBLE library to the library list before it begins processingoG your source program. Consequently, you do not need to addt7 it manually using the .LIBRARY directive.n ? The following macros are supplied by the MACRO-64 Assembler. $BEGIN_EPILOGUE $CALL $CODE_SECTION $DATA_SECTION $END_EPILOGUE $END_PROLOGUE $END_ROUTINE $LINKAGE_PAIR! $LINKAGE_SECTION $OPDEFu& $PROCEDURE_DESCRIPTOR $RETURN I 6-1 . MACRO-64 Supplied Macros% 6.1 MACRO-64 Supplied Libraryl $ROUTINE G Also, Appendix B explains how to program more effectively G using the MACRO-64 assembly language within the framework . of the OpenVMS Calling Standard. & 6.2 Routines and Lexical Scope F The calling-standard macros use the concept of a single,H current, active routine. A routine is simply a programmingC entity that is associated with a procedure descriptor G that may be called, or is a main routine specified as the 1 transfer address of a linked image.a I Only one routine can be active or current at any given time E during assembly. If more than one routine is defined incG a single assembler source file, all items associated witheF the current routine, that is within the lexical scope ofF the routine, must be completed before making a differentF routine current. The lexical scope of one routine cannot; overlap the lexical scope of another routine. G A routine becomes current or comes into scope by invoking I the $ROUTINE macro with the appropriate arguments. $ROUTINEpH marks the beginning of the lexical scope of a routine. TheE complementary macro, $END_ROUTINE, marks the end of the . current routine's lexical scope. + 6.2.1 Routines and Program Sections G Routines have three types of associated program sections: E o Code section-Contains the executable instructions of E the routine. This section is typically read only and executable. B o Data section-Contains data accessed by a routine.B Typically, this is where variable data is stored.D This section is typically non-executable, readable, and writeable. ? o Linkage section-Contains a routine's procedure E descriptor and the necessary linkage information for G calling other routines, and for linkage to data not in H the linkage section, if any. Also, constant data may beF placed here. Typically, this section is read only and not executable. 6-2c k S I MACRO-64 Supplied Macros I 6.2 Routines and Lexical Scopei I The linkage section is considered a type of data section - with the following function:f F - Provides linkage information for calls made from a@ routine associated with the linkage section. H - Provides linkage information for data outside of the$ linkage section. I - Defines the associated routine's procedure descriptor = so that calls can be made to the routine. 4 - Defines constant or static data. 4 6.3 Using Macros to Control Program Sections D The following section explains how to use the MACRO-64G supplied macros to define and control the various program_' sections during assembly.T ' 6.3.1 Defining Program Sectionsi G Each of the three program section types can be referenced F by a corresponding macro. The following three macros areG used to refer to and define these three psects associated within a routine:x F o $CODE_SECTION-makes the routine's code psect current. F o $DATA_SECTION-makes the routine's data psect current. C o $LINKAGE_SECTION-makes the routine's linkage psects current. E To switch to one of the current routine's sections, you I invoke the corresponding macro. For example, after invoking H $ROUTINE, to place instructions into the current routine'sH code psect, you invoke the $CODE_SECTION macro. This makesD the code psect associated with the current routine theG current psect. Similarly, invoking $LINKAGE_SECTION makesiG the linkage psect associated with the current routine theo current psect. D You can also control the psect name and attributes forG each of the program sections by defining arguments to the_D $ROUTINE macro. For more information, see Section 6.7. I 6-3h i i MACRO-64 Supplied Macros4 6.3 Using Macros to Control Program Sections * 6.3.2 Using Macro Specific Symbols F The following symbols are defined by the $ROUTINE macro.A These symbols are useful while programming with thefE calling-standard macros to refer to particular data ande$ linkage section items: @ o $CS-The address of the beginning of the current( routine's code section. @ o $LS-The address of the beginning of the current+ routine's linkage section.g @ o $DS-The address of the beginning of the current( routine's data section. C o $DP-The linkage section address where the $ROUTINE F macro places the .ADDRESS $DS to enable access to theH data section from the linkage section (this variable is optional).n G o $SIZE-The size of the fixed stack area in bytes. $SIZEsC is defined using the value specified with the SIZEi argument. F o $RSA_OFFSET-The offset within the fixed stack area toI the register save area. $RSA_OFFSET is defined using the > value specified with the RSA_OFFSET argument. G o $RSA_END-The offset within the fixed stack area to thelE first byte beyond the end of the register save area.n H For more information, see the description for the $ROUTINE$ macro in this chapter. F The $CALL macro also defines the $STACK_ARG_SIZE symbol.E This symbol specifies the number of bytes used to storeEH arguments on the stack. This value is useful after callingH a routine that returns a value on the stack. In this case,F $CALL cannot remove the arguments from the stack becauseH you must first retrieve the returned value from the stack.G Subsequently, you can remove the arguments from the stacklI using the $STACK_ARG_SIZE symbol. For more information, seee7 the description of $CALL in this chapter.i 6-4l F m I MACRO-64 Supplied MacroseI 6.3 Using Macros to Control Program Sections % 6.3.3 Defining Procedure Typel @ The OpenVMS Calling Standard defines four types ofB routines; stack, register, null, and bound. For moreD information on types of routines, see Section B.4, and+ the OpenVMS Calling Standard.p ? You can define the routine type by using the KINDdE keyword argument with $ROUTINE or $PROCEDURE_DESCRIPTORiC macros. The validity and values of other $ROUTINE and C $PROCEDURE_DESCRIPTOR macro parameters are determinedIC by the type of routine being declared. For example, a B null procedure type has no stack size, therefore theC SIZE parameter is invalid and cannot be specified fortE a null procedure type. When using the KIND keyword withID the $ROUTINE or $PROCEDURE_DESCRIPTOR macros, note the# following exceptions: G o The $SIZE symbol defined by $ROUTINE is only valid forv- stack and register routines. I o The $RSA_OFFSET symbol defined by $ROUTINE is only valid % for a stack routine. / 6.3.4 Using Macros in Prologue Sections B With stack and register routines, $ROUTINE generates9 a standard prologue sequence if you specify : the STANDARD_PROLOGUE=TRUE keyword argument.D STANDARD_PROLOGUE=TRUE is the default for register and stack routines.t D Alternatively, you can code your own prologue sequenceC by specifying the STANDARD_PROLOGUE argument as FALSEeB and using the $END_PROLOGUE macro to mark the end ofC your prologue sequence. In this case, you may wish to C use the $SIZE and $RSA_OFFSET symbols to symbolicallypI specify the fixed stack size and register save area offset,l respectively. > For more information on the prologue sequence ofH instructions that must occur at the beginning of all stackF and register routines, see the OpenVMS Calling Standard. I 6-5 MACRO-64 Supplied Macros4 6.3 Using Macros to Control Program Sections / 6.3.5 Using Macros in Epilogue Sectionsd G With stack and register routines, you can use the $RETURNlD macro to generate an epilogue sequence. Alternatively,? you can code your own epilogue sequence using thedB $BEGIN_EPILOGUE and $END_EPILOGUE macros to mark the; beginning and end of your epilogue sequences. F The OpenVMS Calling Standard also describes the epilogueI sequence of instructions that must be executed every time a_> stack or register routine returns to its caller. 6 6.4 Programming Examples Using Supplied Macros I The following program examples show how to use the calling- I standard macros to define a routine, switch control between I psects, generate an epilogue sequence, and end the routine. 7 Example 6-1 Program Using Supplied Macros N $ROUTINE MAIN, KIND=STACK, - ; Stack routine kind 1N SAVED_REGS=<FP>, - ; Saves FP 2N SIZE=48 ; Stack size is 48 3 X $LINKAGE_SECTION ; Switch to the linkage psect. 4 I X: .long 6 ; X is a constantdI FP1_ADDR: ; FP1_ADDR -> FP1e" .address FP1 V $DATA_SECTION ; Switch to the data section 5I A:: .blkw 5 ; $DS points here B:: .blkw V $CODE_SECTION ; Switch to the code section 6K ; ($CS points here)e . . .W $RETURN ; Perform epilogue and return 7r W $END_ROUTINE MAIN ; Mark the end of the routine 8 G 1 $ROUTINE defines the routine MAIN. The routine type is H defined as a stack routine using the kind=stack keyword argument. 6-6 I MACRO-64 Supplied Macros I 6.4 Programming Examples Using Supplied Macros D 2 The keyword argument SAVED_REGS=<FP> adds the frameG pointer register to the list of registers saved on the F stack by the prologue code. The saved_regs keyword isG valid only for stack procedures. If you do not specifynI the FP register (R29) with this argument, it is assumed._ D 3 The keyword argument SIZE=48 defines the stack areaG in bytes. This argument is valid only for register and I stack routines. If you do not specify a stack size for aaI stack or register routine, $ROUTINE computes the minimumsF stack size required to accomodate the other arguments5 you specify or leave as the default. B 4 The LINKAGE_SECTION macro switches to the linkageF section. The $LS symbol created by the $ROUTINE macroC can be used to point to the address of the current @ routine's linkage section. $ROUTINE creates theE procedure descriptor and leaves the current locationlC counter within the linkage section just beyond thew& procedure descriptor. E 5 The DATA_SECTION macro switches to the data section.rI The $DS symbol created by the $ROUTINE macro can be used F to point to the address of the current routine's data section.y E 6 The CODE_SECTION macro switches to the code section. E The $CS symbol created by the $ROUTINE macro is usedaF to point to the address of the current routine's code section. @ 7 The $RETURN macro generates a standard epilogueG instruction sequence. You can use this macro only with-D stack or register routine defined with the $ROUTINE macro. E 8 The $END_ROUTINE macro marks the end of the routine. I 6-7 MACRO-64 Supplied Macros! 6.5 Using the $CALL Macro ! 6.5 Using the $CALL Macro G $CALL calls local or external routines previously defined C by the $ROUTINE macro or defined in another language._ E To call a routine using standard linkage, use the $CALL I macro. You invoke this macro from within the routine's codeb< section. $CALL performs the following actions: H o Searches a list of linkage pairs referenced in previousI invocations of the $CALL and $LINKAGE_PAIR macros withineH the calling routine. If a linkage pair is already foundB to exist on the list, $CALL uses the linkage pairF stored from the previous invocation. Otherwise, $CALLE stores the linkage pair of the called routine in thegF caller's linkage section and adds the linkage pair to# the caller's list.y C o Allocates stack space for arguments, if necessary.t F o Loads arguments specified with the ARGS argument intoG argument registers and onto the stack, as appropriate. H o Sets the arguments information register, R25, according? to the arguments specified with ARGS argument. H o Generates the following instruction sequence to performE the actual call based on the location of the linkage.I pair generated from above, and the value of Rls, linkage G register, which is assumed to point to the base of theP! linkage section:G T LDQ R26, code_addr_offset(Rls) ; load code addressQ LDQ R27, proc_desc_addr_offset(Rls) ; load procedure U ; descriptor address B ;V JSR R26, R26 ; Jump to the routineU ; saving the returnaR ; address in R26 F o Frees argument stack space, if any, and if the called= routine doesn't return a value on the stack.r G Like $ROUTINE, the $CALL macro invokes other macros toy) perform the above tasks.p 6-8 t I MACRO-64 Supplied MacrosdI 6.5 Using the $CALL Macro ? If you do not specify the Rls argument in youraF invocation of $CALL, $CALL assumes that you have usedH the .BASE directive to define a register that points toF the base address of your linkage section. That is, itF assumes that you have included a statement similar to the following: .BASE R27, $LS F This source statement defines the base address of theI linkage section to be R27, and to be associated with thetI macro symbol $LS. This source statement should be placedoC in your source code prior to the $CALL macro call., ( 6.5.1 Using $CALL in Source Code G Example 6-2 uses the source code from Example 6-1, except E it uses the $CALL macro to show how to call a local ands external routine.o I 6-9 e MACRO-64 Supplied Macros! 6.5 Using the $CALL Macron - Example 6-2 Program Using $CALLb . .y . T $CODE_SECTION ; Switch to the code sectionK ; ($CS points here)o .i .n . ) MOV R27,R2 1 ) .base R2, $LS 2e .o . . W $CALL SUB1 ; Call external routine SUB1 3dW $CALL SUB2, LOCAL=TRUE ; Call local routine SUB2 4 .s .f . U $RETURN ; Perform epilogue and returnrU $END_ROUTINE MAIN ; Mark the end of the routine C 1 The $LS symbol is defined to be the address of theeF procedure that is defined within the linkage section.H The calling routine places the address of the procedure? descriptor in R27 before making the call. This G guarantees that the address associated with the symboleC $LS is stored in R27 upon routine entry. Since the E information in R27 is erased during a standard call,u4 a copy is preserved in register R2. F 2 Register R2 now contains the address of our procedureE descriptor, which is the base address of our linkagepC section. Since $LS is defined to point to the baseeG address of the linkage section, the assembler computesrG the offsets within the linkage section using the .BASE directive.$ A 3 The $CALL macro calls the external routine SUB1. H 4 The $CALL macro calls the local routine SUB2. This callF uses the LOCAL=TRUE keyword argument to indicate that; routine SUB2 is defined within the module.x 6-10 n I MACRO-64 Supplied Macros I 6.6 Programming Considerations & 6.6 Programming Considerations H This section discusses some programming considerations youI need to be aware of when using the calling-standard macros.o 9 6.6.1 Making Multiple Calls From the Same Routine A The $CALL macro generates the following instructionh sequence:b Y LDQ R26, code_address_offset(Rls) ; load code address V LDQ R27, procedure_descriptor_address_offset(Rls) ; load procedureU ; descriptor R ; addressG JSR R26, R26 ;l C The contents of R26 and R27 are erased as a result ofdG using the $CALL macro. This is important since Rls in thetF previous sequence is typically R27. Thus, if you requireE subsequent access to your linkage section, such as whenaI making subsequent calls, you need to make a working copy ofsI R27 to use after the first call. Example 6-2 shows how thist technique is used. H Note that $CALL also overwrites the values in the argumentB registers and the scratch registers specified or theE default set by the SCRATCH_REGS argument, when you passv- arguments to the called routine " 6.6.2 Non-Standard Linkage E Under certain circumstances, there may be advantages inuI using a non-standard routine linkage. For more information,$> see Appendix B and the OpenVMS Calling Standard. " 6.6.3 Routine Restrictions E Different routine types have different capabilities and_B restrictions. For example, only a stack routine thatD specifies BASE_REG_IS_FP=TRUE can make standard calls.G For more information, see the description for $ROUTINE in this chapter. A For a description of the different capabilities and A restrictions associated with each routine type, see ; Section B.4 and the OpenVMS Calling Standard.r I 6-11t t MACRO-64 Supplied Macros 6.7 Macro Descriptions 6.7 Macro Descriptions A This section provides a detailed description of theo( following MACRO-64 macros: $BEGIN_EPILOGUE $CALL $CODE_SECTION $DATA_SECTION $END_EPILOGUE $END_PROLOGUE $END_ROUTINE $LINKAGE_PAIR! $LINKAGE_SECTION $OPDEF & $PROCEDURE_DESCRIPTOR $RETURN $ROUTINE 6.7.1 Syntax Rules ? You may use either positional or keyword argumentr@ association or a combination of the two with theseE macros. For positional association, the order of formalnH arguments is shown with the format of each macro. For moreE information on argument association, see Chapter 4. TheRF following syntax rules apply when invoking the assembler< using the command line qualifier /NAMES=as_is: H o When specifying macro names, you must use all uppercaseA or all lowercase characters. You cannot use botho1 uppercase and lowercase letters. I o When specifying keyword arguments, you must use the same I alphabetic case as the macro name it is associated with.rI If you use lowercase characters with the macro name, you$I must use lowercase characters with the keyword argument.aI If you use uppercase characters with the macro name, youyI must use uppercase characters with the keyword argument. 6-12 e c I MACRO-64 Macros I $BEGIN_EPILOGUET I _________________________________________________________________ $BEGIN_EPILOGUEd F Marks the beginning of an epilogue instruction sequence. Format $BEGIN_EPILOGUEe DescriptionO @ $BEGIN_EPILOGUE marks the beginning of an epilogueB instruction sequence that you code within a stack or? register routine defined with the $ROUTINE macro.O E At each point where a stack or register routine returnseC to its caller, the routine must perform a sequence of F operations to restore any saved registers and to performA stack frame management. This sequence is called the H epilogue and is described in detail in the OpenVMS Calling Standard. B You can use the $RETURN macro to generate a standardI epilogue instruction sequence for you, or you can code yoursG own sequence. If you code your own epilogue sequence, youiF must mark the beginning and end of the epilogue sequence@ with the $BEGIN_EPILOGUE and $END_EPILOGUE macros. Notesy A You must not use $BEGIN_EPILOGUE for an epiloguenC instruction sequence generated by $RETURN. $RETURNfI automatically invokes $BEGIN_EPILOGUE and $END_EPILOGUE.s I 6-13 MACRO-64 Macrost $BEGIN_EPILOGUE. Examples B $ROUTINE MUMBLE, KIND=REGISTER, SAVE_FP=R1 :e :a :s' $BEGIN_EPILOGUEcL MOV R1,FP ; Restore caller's frameF RET (R26) ; Return to caller% $END_EPILOGUEn : : : + $END_ROUTINE MUMBLE 6-14 I MACRO-64 Macros I $CALL I _________________________________________________________________ $CALLT / Issues a call to another routine.s Format 0 $CALL NAME=routine-being-called -5 [Rls=linkage-section-register] - 3 [LS=linkage-section-address] -t& [LOCAL=boolean] -+ [ARGS=argument-list] -$3 [SET_ARG_INFO=boolean-value] - 9 [STACK_RETURN_VALUE=boolean-value] - 6 [SCRATCH_REGS=scratch_reg-list] -* [TIE=boolean-value] -S [FUNC_RETURN={I64,D64,I32,U32,FF,FD,FG,FS,FT,FDC,FGC,FSC,FTC}] -7 [USES_VAX_ARGLIST=boolean-value] - : [SIGNATURE_BLOCK=signature_address] -2 [NONSTANDARD=boolean-value] - Parameters NAMEI The name of the routine to call. This argument is required.o RlsnF Linkage section register to use when generating the LDQ,G LDQ, JSR instruction sequence. This argument is optional.g E If Rls is omitted, $CALL assumes that you have issued avF .BASE directive prior to invoking $CALL that establishesD the value of a base register pointing into the linkageB section. If you omit the Rls argument and you do notG issue such a .BASE directive prior to invoking $CALL, theiE assembler issues the following error message during the $ $CALL macro expansion: T "Argument 2 invalid" The assembler failed to find a base registerO specified with a previous .BASE directive to form a registere5 expression of the form offset(Rn)"L I 6-15 b c MACRO-64 Macros $CALLc LSB LS is the address of the linkage section. If you useC $CALL within a routine defined by the $ROUTINE macro, C the LS argument defaults to the $LS symbol defined bykE $ROUTINE. If you use $CALL outside of a routine defined D by the $ROUTINE macro, there are two ways that you canD indicate the location of the linkage section to $CALL.B First, you can specify the LS argument to $CALL as aD relocatable address expression that indicates the baseH of the linkage section. In this case you must also specifyI the Rls argument. Second, you can specify both the linkage-aH section base register and the linkage section address in aH .BASE directive prior to invoking $CALL. In this case, you? must omit both the Rls and LS arguments to $CALL. G Digital recommends that you omit this argument if you use C $CALL within a routine defined by the $ROUTINE macro. LOCAL H A boolean value (TRUE or FALSE) that specifies whether theE routine to be called is local to the module or globally G visible. By default, $CALL generates a call to a globally F visible routine. To generate a call to a routine that is@ not globally visible, you must specify LOCAL=TRUE. ARGSA An optional list of arguments to pass to the calledrF routine. Enclose the argument list within angle bracketsD (<>), and separate the arguments with commas (,). YouE can use the following qualifiers with each argument youiG specify with the ARGS argument. Table 6-1 describes these G qualifiers. Each argument is an address expression (which I may include a register), followed by a qualifier. The table F also contains the argument type, the instruction used toE load the argument into a register, the instruction used'H to store the argument on the stack, and the encodings usedD in the Argument Information Register (R25) in the callH signature block when you specify TIE=TRUE. See the OpenVMSG Calling Standard for more information on these encodings.aG Note that arguments are only stored on the stack if there B are more than six arguments provided to the routine. 6-16 I MACRO-64 MacrosaI $CALLe I Table_6-1_ARGS_Arguments___________________________________i ; Argument RegnG Argument LOAD STORE AI Arg Mem Arg)I QualifieType______InstructInstructionodingSignaturSignaturea C /A Address LDA STQ I64 I64 I64 C /D D- LDG STG FD FD I64 floating C /F F- LDF STF FF FF I64 floating C /G G- LDG STG FG FG I64a floating C /L Longword LDL STQ I64 I32 I32 C /Q Quadword LDQ STQ I64 I64 I64 C /S S- LDS STS FS FS I64 floating C /T T- LDT STT FT FT I64t floating C /UL[1] Unsigned LDL STQ I64 U32 I64d$ longword /ZAP% #^xF0 I [1]Unsigned_32-bit_integers_are_normally_passed_using______a D the /L argument qualifier. Therefore, Digital does not@ recommend that you use the /UL argument qualifier.I ___________________________________________________________s SET_ARG_INFOC An optional argument to indicate whether $CALL shouldfG set the Argument Information (AI) register (R25) with theoG appropriate argument information or not. By default or ifsI you specify SET_ARG_INFO=TRUE, $CALL stores the appropriatemI argument information in R25. If specify SET_ARG_INFO=FALSE,sE $CALL does not affect R25. In this case, you must store I the appropriate information in R25 yourself before invokingl $CALL. STACK_RETURN_VALUEF An optional argument to indicate that the called routineE returns a value on the stack. By default, $CALL assumes D that the called routine does not return a value on theH stack. In this case, $CALL removes any arguments passed to I 6-17 MACRO-64 Macros $CALL G the called routine from the stack when the called routinec returns. E If the called routine returns a value on the stack, the B returned value is placed at a lower address than theD arguments on the stack. In this case, you must specifyH STACK_RETURN_VALUE=TRUE to prevent $CALL from removing theH arguments to the called routine from the stack and erasingI the value returned by the called routine. You must retrieve E the return value and remove it from the stack. Then you F can remove the arguments to the called routine using the6 $STACK_ARG_SIZE symbol defined by $CALL. SCRATCH_REGSI An optional list of scratch registers for $CALL to use when H processing arguments passed to the called routine with theG ARGS argument. If you pass more than six arguments to thewH called routine, $CALL may need to use scratch registers to process the call. @ By default, $CALL uses R0, R1, F0, and F1. You canE cause $CALL to use different scratch registers with the $ SCRATCH_REGS argument. F If you are passing integer arguments, you should specifyI at least one integer register. If you are passing floating-.H point arguments, you should specify at least one floating- point register.w D $CALL can process arguments to the called routine moreF efficiently if you specify two or more scratch registersG of the type or types appropriate to the arguments you aret passsing.s TIE D A boolean value (TRUE or FALSE) that specifies whetherF $CALL should generate a call sequence that is compatible@ with both native routines and the Translated ImageE Environment (TIE). By default, $CALL generates a fasternI call sequence that is only compatible with native routines. F If you specify TIE=TRUE, $CALL generates a call sequenceA that works with both native routines and translated F routines. If you are calling a VAX routine in a sharableF image that has been translated with the DECmigrate imageG translator, specify TIE=TRUE. If you are calling a native F routine, Digital recommends you default the TIE argument 6-18 u o I MACRO-64 MacrossI $CALLu I or specify TIE=FALSE. While $CALL generates a call sequence F that is compatible with native routines when you specifyB TIE=TRUE, that call sequence is slower than when you+ specify or default TIE=FALSE. FUNC_RETURN H An optional argument used to indicate the type of functionC return when you also specify TIE=TRUE. Note that this C argument is ignored unless you also specify TIE=TRUE.rD Specify one of I64, D64, I32, U32, FF, FD, FG, FS, FT,H FFC, FDC, FGC, FSC, or FTC. These values correspond to theG RASE$K_FR_* signature encodings described in table 3-7 inoF the VMS Calling Standard. If you specify TIE=TRUE and doF not specify a function return type with FUNC_RETURN, the2 default function return type is I64. F ________________________ Note ________________________ B Specification of the FUNC_RETURN argument does not@ in itself cause $ROUTINE to generate a procedureB signature block. However, if you specify either orC both the ARGLIST or USES_VAX_ARGLIST arguments, anySB value you specify with the FUNC_RETURN argument isA recorded in both the procedure descriptor and the * procedure signature block. F ______________________________________________________ USES_VAX_ARGLISTI An optional argument to indicate whether the called routinecB uses a VAX argument list. Note that this argument isI ignored unless you also specify TIE=TRUE. By default, $CALLeD assumes the called routine does not use a VAX argumentF list. Specify USES_VAX_ARGLIST=TRUE to indicate that the6 called routine uses a VAX argument list. SIGNATURE_BLOCK I An optional argument that you can use to supply the addressdB of the call signature block. Note that this argumentD is ignored unless you also specify TIE=TRUE. Note alsoC that you cannot specify a SIGNATURE_BLOCK argument inNE combination with either of the FUNC_RETURN or USES_VAX_NC ARGLIST arguments. By default, $CALL generates a call D signature block for you when you specify TIE=TRUE, andD you can in part control the contents of that signature I 6-19r MACRO-64 Macrosa $CALL H block with the FUNC_RETURN and USES_VAX_ARGLIST arguments.E If you wish to define your own call signature block, dooG not specify either of the FUNC_RETURN or USES_VAX_ARGLIST E arguments and supply the address of your call signaturem6 block with the SIGNATURE_BLOCK argument. NONSTANDARD D A boolean value (TRUE or FALSE) that specifies whetherF $CALL should suppress warning and informational messagesE concerning non-standard usage. By default, $CALL issuesk@ warning and informational messages to indicate youD are using $CALL in a way that violates the VMS CallingD Standard or in a way that requires special programmingE considerations. Specify NONSTANDARD=TRUE if you wish to & suppress these messages. Descriptiono E $CALL issues a call to another routine and performs thee following actions: H 1. Searches a list of linkage pairs referenced in previousF invocations of the $CALL and $LINKAGE_PAIR macros. IfF a linkage pair is already in the list, $CALL uses theF linkage pair from the previous invocation. Otherwise,G $CALL stores the linkage pair of the called routine in G the caller's linkage section and adds the linkage pairBH to the caller's list. If you use $CALL within a routineI defined with the $ROUTINE macro, $CALL and $LINKAGE_PAIRdB reset the list of linkage pairs for each routine. B 2. Allocates stack space for arguments if necessary. D 3. Generates instructions to load the arguments to the called routine. E 4. Sets the value in the argument information register, R25.a B 5. Generates the instruction sequence shown below toA perform the actual call based on the location oftE the linkage pair generated from Step 1 above and the I address specified or defaulted with the LS argument. ThenG register specified with the Rls argument is assumed toeI point to the base of the linkage section as shown in the # following example:u 6-20 N I I MACRO-64 MacrosNI $CALL_ \ LDQ R26, code_address_offset(Rls) ; load code addressY LDQ R27, procedure_descriptor_address_offset(Rls) ; load procedure X ; descriptorU ; address ! JSR R26, R26U F 6. Frees argument stack space, if any, and if the called> routine does not return a value on the stack. Examples Example 1 > $CALL SUB1, Rls=R13, LS=MY_LINKAGE_SECTION1 .BASE R13, MY_LINKAGE_SECTION $CALL SUB2 A $ROUTINE SUB3, KIND=STACK, SAVED_REGS=<R2,FP>4$ $LINKAGE_SECTION .QUAD 1 XX: ! $CODE_SECTION MOV R27, R2 " $CALL SUB4, R2! .BASE R2, $LS_ $CALL SUB5! $CALL SUB6, -a+ ARGS=<XX/A, R0/Q>, - - SCRATCH_REGS=<R22,R23> ! $CALL SUB7, --) ARGS=<1/A,2/A,3/A>a $RETURN % $END_ROUTINE SUB3 Example 2 @ $CALL FOO, ARGS=<A,B,C,D,E,F,G,H>, STACK_RETURN_ VALUE=TRUEX ; Fetch octaword return value from stack> LDQ R4, 0(SP) ; low quadword? LDQ R5, 8(SP) ; high quadword = LDA SP, $STACK_ARG_SIZE+16(SP) ; RESET STACKT I 6-21 MACRO-64 Macros- $CODE_SECTIONa I _________________________________________________________________i $CODE_SECTION D Switches control to the current routine's code section psect. Format $CODE_SECTION Description E $CODE_SECTION switches control to the current routine's ! code section psect. Examples $CODE_SECTIONa 6-22 o I I MACRO-64 MacroscI $DATA_SECTIONo I _________________________________________________________________ $DATA_SECTIONa D Switches control to the current routine's data section psect. Format $DATA_SECTION Descriptiont E $DATA_SECTION switches control to the current routine's ! data section psect. Examples $DATA_SECTION I 6-23d MACRO-64 Macros $END_EPILOGUEs I _________________________________________________________________e $END_EPILOGUE @ Marks the end of an epilogue instruction sequence. Format $END_EPILOGUEe Description D You can use the $END_EPILOGUE macro to mark the end ofE an epilogue instruction sequence that you code yourself A within a stack or register routine defined with the E $ROUTINE macro. At each point where a STACK or REGISTER G routine returns to its caller, the routine must perform a G sequence of operations to restore any saved registers and H to perform stack-frame management. This sequence is calledD the epilogue and is described in detail in the OpenVMS Calling Standard.e B You can use the $RETURN macro to generate a standardC epilogue instruction sequence for you. Alternatively, B you can code your own sequence. If you code your ownD epilogue sequence, you must mark the beginning and endC of the epilogue sequence with the $BEGIN_EPILOGUE ands1 $END_EPILOGUE macros, respectively.i F Note that you must not use $END_EPILOGUE for an epilogueH instruction sequence generated by $RETURN. $RETURN invokes8 $BEGIN_EPILOGUE and $END_EPILOGUE for you. B You may omit the invocation of $END_EPILOGUE if yourA epilogue sequence occurs at the end of the routine.iE The $END_ROUTINE macro invokes $END_EPILOGUE for you ifeE you invoke the $BEGIN_EPILOGUE macro without a matchingnD invocation of the $END_EPILOGUE macro. You must invokeE $END_EPILOGUE with epilogue sequences that occur in the % middle of your routine.i 6-24 I MACRO-64 MacrosAI $END_EPILOGUEc Examples B $ROUTINE MUMBLE, KIND=REGISTER, SAVE_FP=R1 : : :r' $BEGIN_EPILOGUEeL MOV R1,FP ; Restore caller's frameF RET (R26) ; Return to caller% $END_EPILOGUE : :g :_+ $END_ROUTINE MUMBLE I 6-25_ t t MACRO-64 Macrosr $END_PROLOGUE I _________________________________________________________________ $END_PROLOGUE ? Marks the end of a prologue instruction sequence.F Format $END_PROLOGUE6 Description ? $END_PROLOGUE marks the end of a routine prologue G instruction sequence that you code yourself. The prologueoG instruction sequence begins with the first instruction ofdE the routine. There can only be one prologue instruction H sequence in a routine. The prologue is described in detail. in the OpenVMS Calling Standard. D You invoke $END_PROLOGUE after the last instruction inD the routine prologue code. The last instruction in theI routine prologue is the one that updates FP (R29, the frame G pointer) and makes the frame become current. You must use G this macro when the routine type is stack or register and H you specify STANDARD_PROLOGUE=FALSE to the $ROUTINE macro. Notes_ H Do not use this macro when the $ROUTINE macro generatesH a standard prologue or when the routine type is null orI bound. For example, a standard prologue is generated for H you when you specify or leave the default of $ROUTINE's4 STANDARD_PROLOGUE argument to TRUE. Examples & MOV SP, FP% $END_PROLOGUEC 6-26 s s I MACRO-64 MacrosoI $END_ROUTINE I _________________________________________________________________ $END_ROUTINE ) Marks the end of a routine. Format / $END_ROUTINE [name=routine-name] Parameters nameE The name of the routine that is ended. This argument is I optional. If you specify a name, $END_ROUTINE verifies thatCH the name matches that which was specified with $ROUTINE toI begin the routine. If the name does not match, $END_ROUTINEuF issues a diagnostic message. There is no default. If youD omit the name, $END_ROUTINE does not verify a matching name.e DescriptionC E You must use this macro at the end of a routine that isuD defined with the $ROUTINE macro to delimit the end theF current routine and to perform final macro processing of the routine. Examples , $END_ROUTINE NAME=FOOZLE I 6-27 C MACRO-64 Macrost $LINKAGE_PAIRe I _________________________________________________________________e $LINKAGE_PAIR E Locates or defines a linkage pair in the linkage psect. Format 0 $LINKAGE_PAIR name, local=boolean Parameters nameB Name of the linkage pair to define. This argument is required.e localsH A boolean value (TRUE or FALSE) that specifies whether theG routine is defined within the module for not. The defaultaG is to store a linkage pair for a global routine. You mustnF specify LOCAL=TRUE to store a linkage pair for a routine+ that is not globally visible.c Descriptionl E You can invoke this macro to locate or define a linkagerB pair in the linkage psect. The linkage pair is addedD to a list for later use and retrieval by the $CALL andF $LINKAGE_PAIR macros. Thus, only the first invocation ofG $LINKAGE_PAIR (or $CALL) for a given routine to be called G results in a linkage pair being stored. $LINKAGE_PAIR and D $CALL use the same list of linkage pairs. The $ROUTINEH macro resets this list. $LINKAGE_PAIR restores the current$ psect when it is done. H $LINKAGE_PAIR defines the symbol $LP. This symbol value isH the address within the linkage section of the linkage pair0 for the specified routine to call. 6-28 i e I MACRO-64 Macros I $LINKAGE_PAIR Notes H Because the $CALL macro invokes the $LINKAGE_PAIR macroG for you, you do not need to use $LINKAGE_PAIR when you C are using $CALL. You may wish to use $LINKAGE_PAIRsC when you are not using $CALL or when you require ayG specific ordering or placement of linkage pairs withinn% the linkage section.T Examples $LINKAGE_ C PAIR SUB1 ; define linkage pair in linkage section LDQ R26, $LP" LDQ R27, $LP+8 JSR R26, R26 I 6-29 F MACRO-64 Macros $LINKAGE_SECTION I _________________________________________________________________t $LINKAGE_SECTION H $LINKAGE_SECTION switches control to the current routine's$ linkage section psect. Format $LINKAGE_SECTION Description_ H $LINKAGE_SECTION switches control to the current routine's$ linkage section psect. Examples $LINKAGE_SECTION 6-30 h I MACRO-64 MacrosfI $OPDEFo I _________________________________________________________________ $OPDEF % Used to define opcodes._ Format 0 $OPDEF MNEMONIC, FORMAT, ENCODING Parameters MNEMONICD MNEMONIC is the mnemonic name by which the instructionD is called. You may optionally specify a qualifier listF separated from the name by a slash (/). A qualifier listD is a sequence of one or more letters or digits with no/ intervening spaces or separators.t FORMAT7 FORMAT is one of the following arguments: I _________________________________________________________________.I Format__________________________Description______________________T > MEMORY Standard memory format5 instructions.e I MEMORY_FUNCTION Memory format instructions with ae6 function code. B JUMP Memory format instructionsI formatted like jump instructions. > BRANCH Standard branch format5 instructions. F OPERATE Standard operate instructions. G FLOATING_OPERATE Standard floating-point operate 5 instructions. F PAL Standard PALcode instructions. I [<]CUSTOM[=operand_type_list>]__Custom_format.___________________D G With the CUSTOM format, you may optionally specify a list G of the types of operands the instruction is to accept. If G you specify a list of operand types, you must enclose the H entire FORMAT argument within angle brackets, and you mustH specify the operand types in the order they are to be used I 6-31t s p MACRO-64 Macrosi $OPDEF I with the instruction. $OPDEF supports the following operands types: B IREGISTER Integer register, such as R5 or SP. C FREGISTER Floating-point register, such as F7. F LITERAL Integer literal, such as #123 or 32767. I INDIRECT Indirect integer register notation such as $ (R7). F DISPLACEMENT Indirect integer register notation withG an integer literal displacement, such asd( FOO(R12). G BRANCH_OFFSET Label or address expression, such as L1.n For example: 4 FORMAT=<CUSTOM=IREGISTER,DISPLACEMENT> H For a detailed description of instruction formats, see the2 Alpha Architecture Reference Manual. ENCODINGB ENCODING is the numeric encoding for the instructionF opcode. The default radix is decimal, as is the case forC all assembler constants. Prefix the value with ^X for G hexidecimal. Certain instruction formats allow multi-parta encoding: 6-32 I MACRO-64 Macros I $OPDEF I ___________________________________________________________ I Format________________Encoding_Description_________________ H MEMORY_FUNCTION Specify the base opcode, followed byI a dot, followed by the function code. 0 For example: 6 ENCODING=^X10.F000 G JUMP Specify the base opcode, optionally H followed by a dot, and the hardware-F hint bits optionally followed by aG dot and the software-hint bits. For , example: 5 ENCODING=^X1A.2.1a H OPERATE Specify the base opcode, followed byI a dot, followed by the function code. 0 For example: 4 ENCODING=^X12.3C& ex H FLOATING_OPERATE Specify the base opcode, followed byI a dot, followed by the function code. 0 For example: 5 ENCODING=^X17.02B G PAL Specify the base opcode, optionally F followed by a dot and the functionF code. Omit the function code for aH generic PAL instruction that acceptsG an integer-expression argument. ForR, example: 5 ENCODING=^X0.0080u1 ENCODING=^X19 I CUSTOM Specify a comma-separated list of bitiG ranges and values to place in thoseC< bit ranges. For example: J ENCODING = < 26:31=^X14, 21:25=%OP1, - Q ___________________________________16:20=%OP2.REG,_0:15=%OP2.DISP >_ B For CUSTOM format instructions, specify the ENCODINGI argument as a comma separated list of bit ranges and values I to place in those bit ranges. Enclose the list within angleE brackets. I 6-33s MACRO-64 Macros $OPDEF F Specify a bit range as start:end where start and end areH integer constant expressions. For a 1-bit bit range, startH and end are equal. Bit positions range from 0 to 31. PlaceF an equal sign (=) after the bit range specifier followedD by the value you wish to put in the bit range. You canF place either a constant integer expression or an operandG into the bit range. Start and end expressions and integerEB constant expressions must not reference any externalH symbols or symbols not yet defined in the assembly. $OPDEFI evaluates these expressions at the time that it defines the G instruction as opposed to when the defined instruction isA referenced.D H Operand names are of the form %OPn, where n is the ordinalH number of the operand as specified from left to right with" the FORMAT argument. E For IREGISTER, FREGISTER, and INDIRECT operands, $OPDEF I places the 5-bit register number into the bit positions you specify. F For LITERAL operands, $OPDEF places the literal value ofE the operand into the bit positions you specify. Operande@ literal values can be up to 32-bits long. The mostG significant bits are truncated if the size of the operandp> literal value exceeds the bit range you specify. G For DISPLACEMENT operands, $OPDEF defines two parts: a 5-pF bit register number and a displacement value that can beI up to 32-bits long. The most significant bits are truncatednI from the displacement value if the size of the displacement H value exceeds the bit range you specify. You can referenceI the register number by placing .REG after the operand name. E For example: %OP2.REG. Similarly, you can reference the I displacement value by placing .DISP after the operand name.t% For example: %OP2.DISP. B For BRANCH_OFFSET operands, $OPDEF stores the signedB longword offset between the next instruction and theE specified address in the bit positions you specify. The F address expression specified for a BRANCH_OFFSET operandI can be a backward or forward reference to a location withinnG the current psect. It cannot be an external address or aniG address in a different psect. The resulting offset can beUF up to 32-bits in size. If the size of the offset exceeds 6-34 I MACRO-64 Macros I $OPDEF F the bit range you specify, the most significant bits are truncated. G $OPDEF fills any bit positions you leave unspecified with zeros. Description F The $OPDEF macro can be used to define your own opcodes. H $OPDEF defines a macro using an unqualified version of theH mnemonic name you specify. When this macro is invoked withD the instruction qualifiers you specify when you defineD it with $OPDEF (if any), it expands to the instructionD representation you have defined with $OPDEF. Note thatG you can specify the qualfiers in any order as long as the < combination of qualifiers you use is the same. C Other uses of the mnemonic name remain vaild provided C you do not use the mnemonic name as a macro name in a B .MACRO directive. For instance, you can use the sameD mnemonic name in the opcode field with different or noC qualifiers. If the qualifiers (or absence thereof) dooB not match those specified in your $OPDEF instructionB definition, the macro defined by $OPDEF processes asH though you had not defined an instruction by that mnemonicD name. To do so, it expands to a single statement. ThisI expansion statement is identical to the mnemonic-name macro G invocation statement, except it is processed in a contexttB that prevents the mnemonic-name macro from expandingC recursively. Instead, the statement is processed as a G normal, MACRO-64 instruction statement. In this case, you H may notice references to the mnemonic-name macro expansionC in a MACAUXMSG diagnostic message, if the instructions( statement contains errors. C For instance, if you define a STQ/P instruction usingnG $OPDEF, you can still use the STQ instruction without theaI /P qualifier. If you do, and your STQ instruction statementUD contains an error, the assembler generates a MACAUXMSGB message indicating that the error occured during theF expansion of macro STQ. Aside from the fact that the STQI instruction is processed in the context of the expansion of I the STQ macro, $OPDEF's definition of the STQ/P instruction ? has no effect on your use of the STQ instruction._ I 6-35 MACRO-64 MacrosD $OPDEF Examples 3 $OPDEF MNEMONIC=BANG, FORMAT=PAL, -a( ENCODING=^X0.0099 6-36 h I MACRO-64 Macros I .PACKEDd I _________________________________________________________________N .PACKEDv 2 Packed decimal string storage macro. Format . .PACKED decimal-string[,symbol] Parameters decimal-stringH A decimal number from 0 to 31 digits long with an optional6 sign. Digits can be in the range 0 to 9. symbolG An optional symbol that is assigned a value equivalent to I the number of decimal digits in the string. The sign is not ! counted as a digit.o DescriptionI I .PACKED generates packed decimal data, two digits per byte._E Packed decimal data is useful in calculations requiringnE exact accuracy. It is operated on by the decimal stringN instructions.n G A packed decimal string is a contiguous sequence of byteskG in memory. It is specified by two attributes: the address E A of the first byte and a length L, which is the numberiF of digits in the string and not the length of the stringH in bytes. The bytes of a packed decimal string are dividedF into two, 4-bit fields (nibbles). Each nibble except theC low nibble (bits 3:0) of the last (highest-addressed) B byte must contain a decimal digit. The low nibble ofA the highest-addressed byte must contain a sign. TheeD representation for the digits and sign is indicated as follows: I 6-37L A P MACRO-64 Macrosi .PACKED I ___________________________________________________________ Digith orI Sign___Decimal__________Hexadecimal________________________I ' 0 0 0 ' 1 1 1E' 2 2 2 ' 3 3 3n' 4 4 4 ' 5 5 5o' 6 6 6e' 7 7 7 ' 8 8 8 ' 9 9 9 1 + 10,12,14, or 15 A,C,E, or FeI -______11_or_13_________B_or_D_____________________________c F The preferred sign representation is 12 for plus (+) andG 13 for minus (-). The length L is the number of digits in G the packed decimal string (not counting the sign); L must H be in the range 0 to 31. When the number of digits is odd,F the digits and the sign fit into a string of bytes whoseF length is defined by the following equation: L/2(integerF part only) + 1. When the number of digits is even, it isI required that an extra zero appear in the high nubble (bits H 7:4) of the first byte of the string. Again, the length in- bytes of the string is L/2 + 1. C The address A of the string specifies the byte of the_F string containing the most-significant digit in its highD nibble. Digits of decreasing significance are assignedF to increasing byte addresses and from high nibble to lowF nibble within a byte. Thus +123 has a length of 3 and is% represented as follows:o 7 4 3 0 +-----+------+ | 1 | 2 | :A +-----+------+" | 3 | 12 | :A+1 +-----+------+ 6-38 I MACRO-64 Macros I .PACKED_ D The packed decimal number -12 has a length of 2 and is% represented as follows: 7 4 3 0O +-----+------+ | 0 | 1 | :A +-----+------+" | 2 | 13 | :A+1 +-----+------+ Examples 9 .PACKED -12,PACKED_SIZED ; PACKED_a SIZE gets value of 2 .PACKED +500 .PACKED 0o6 .PACKED -0,SUM_SIZE ; SUM_ SIZE gets value of 1 I 6-39_ _ _ MACRO-64 Macrosm $PROCEDURE_DESCRIPTORc I _________________________________________________________________a $PROCEDURE_DESCRIPTOR E Defines a procedure descriptor structure at the current psect and offset.s Format # $PROCEDURE_DESCRIPTORn Description G The arguments for the $PROCEDURE_DESCRIPTOR macro are the H same as the $ROUTINE macro, with the following exceptions: 0 o The ENTRY argument is required. < o There are no CODE_SECTION, LINKAGE_SECTION,I DATA_SECTION, DATA_SECTION_POINTER, or STANDARD_PROLOGUEI arguments.n C o There is an additional END_PROLOGUE argument. ThisnD argument must be a label that you define at the endE of the routine's prologue sequence. This argument isaA required for stack and register procedure types. Notese 7 Because the $ROUTINE macro invokes thesH $PROCEDURE_DESCRIPTOR macro for you, you do not need toI use the $PROCEDURE_DESCRIPTOR macro if you use $ROUTINE.lG You may wish to use $PROCEDURE_DESCRIPTOR when you areo$ not using $ROUTINE. Examples 9 $PROCEDURE_DESCRIPTOR p1, - 9 KIND=NULL, -i4 ENTRY=p1_entry 6-40 I MACRO-64 Macros I $RETURNE I _________________________________________________________________ $RETURNe A Generates a standard epilogue instruction sequence. Format $RETURNn Description E Generates a standard epilogue instruction sequence whenhF used within a stack or register routine defined with theH $ROUTINE macro. The epilogue sequence generated by $RETURND restores any registers you specify with the SAVED_REGSF argument to $ROUTINE and performs stack frame managementF as necessary. You can use $RETURN whether you specify or7 default STANDARD_PROLOGUE as TRUE or not.e D You can use $RETURN any number of times within a givenI stack or register routine to affect a return to the calling^ routine. C You must not use the $BEGIN_EPILOGUE or $END_EPILOGUE C macros for an epilogue sequence generated by $RETURN. H $RETURN invokes $BEGIN_EPILOGUE and $END_EPILOGUE for you. Examples : $ROUTINE FOOZLE, KIND=REGISTER, SAVE_FP=R1 :_ :_ : $RETURN_# $END_ROUTINE FOOZLE_ I 6-41 MACRO-64 Macrosl $ROUTINE I _________________________________________________________________ $ROUTINE G Defines the current routine and creates a context for theo routine. Format + $ROUTINE NAME=routine name -r- ALIASES=alias names -t- LOCAL=boolean value - H STANDARD_PROLOGUE=boolean value ENTRY=code entry point - > CODE_SECTION=code section psect name -> DATA_SECTION=data section psect name -< DATA_SECTION_POINTER=boolean value -D LINKAGE_SECTION=linkage section psect name -* KIND=routine type-; HANDLER_REINVOKABLE=boolean value -L5 BASE_REG_IS_FP=boolean value-y2 REI_RETURN=boolean value -: STACK_RETURN_VALUE=boolean value -2 RSA_OFFSET=integer value -2 SAVE_FP=FP register name -> SAVE_RA=return address register name -, SIZE=numeric value -6 SAVED_REGS=list of registers -; HANDLER=exception handler address -eI HANDLER_DATA=data address for exception handler -i7 SYNCH_EXCEPTIONS=boolean value- 4 PROC_VALUE=procedure value -7 ENVIRONMENT=environment value - : FUNC_RETURN=function return type -4 ARGLIST=Argument type list -8 USES_VAX_ARGLIST=boolean value -C OVERRIDE_FLAGS=procedure descriptor flags - 9 DEFAULT_SIGNATURE=boolean value -55 COMMON_BASE=list of registers_ 6-42 2 S I MACRO-64 Macros I $ROUTINE Parameters NAMEH The name of the routine. This argument is required for all@ procedure kinds. There is no default. For example: NAME=FOOZLE ALIASES C List of alias names for the routine. This argument is D optional for all procedure types. There is no default. For example: D ALIASES=<FOO,BAR,FOOBAR,foo,bar,foobar,Foo,Bar,FooBar> LOCALpC Boolean value indicating whether the routine is localiD (TRUE) or externally visible (FALSE). This argument isI optional for all procedure kinds. The default is FALSE. Fori example: LOCAL=TRUE STANDARD_PROLOGUE D Specifies a boolean value to indicate whether $ROUTINEG should generate a standard, instruction prologue sequencehB at the beginning of the routine's code section. ThisC argument is optional and valid only with REGISTER andrA STACK procedures. If the procedure type is stack or E register , the default is TRUE and $ROUTINE generates aMI standard prologue sequence. The prologue sequence generateddB by $ROUTINE saves the registers you specify with theE SAVED_REGS argument and performs stack-frame management as necessary.i D If you also specify BASE_REG_IS_FP=FALSE, the standardI prologue sequence generated by $ROUTINE makes a copy of the I procedure descriptor address that is in R27 upon entry intonI R29 (FP). While you cannot change the value in R29 prior to I the epilogue, you can use R29 as a working, linkage-section E register. If you specify the STANDARD_PROLOGUE argument D as FALSE, you must code your own prologue sequence andH mark the end of the prologue with the $END_PROLOGUE macro.F Whether you specify or default STANDARD_PROLOGUE as TRUEG or not, you can generate a standard epilogue sequence fore I 6-43 MACRO-64 Macrosn $ROUTINE G stack and register procedures with the $RETURN macro. ForR example: % STANDARD_PROLOGUE=FALSEg ENTRY D The name of the code-entry point. This argument is theE code entry-point label that $ROUTINE defines for you atfC the beginning of the code section for the routine. IfdG this argument is omitted, $ROUTINE generates a label. Forn example: ENTRY=FOOZLE_ENTRY CODE_SECTIONE The psect name and attributes of the code section. ThishG argument is optional for all procedure kinds. If omitted, C the default is the name and attributes defined by the A $CODE$ lexical string symbol. If you specify a name D and attributes for the CODE_SECTION argument, $ROUTINEF redefines the $CODE$ lexical string symbol such that theH specified values become the new default. Initially, $CODE$$ is defined as follows: G $CODE$ = "$CODE$,OCTA,PIC,CON,REL,LCL,SHR,EXE,NORD,NOWRT"g H Since you must delimit the psect name and attributes usingC commas, be sure to enclose this argument within anglesG brackets to avoid having the assembler interpret the namexD and attributes as different arguments to $ROUTINE. For example: 3 CODE_SECTION=<MY_CODE,EXE,QUAD,NOWRT>d DATA_SECTIONE The psect name and attributes of the data section. ThiseG argument is optional for all procedure kinds. If omitted,nC the default is the name and attributes defined by thedA $DATA$ lexical string symbol. If you specify a namecD and attributes for the DATA_SECTION argument, $ROUTINEF redefines the $DATA$ lexical string symbol such that theH specified values become the new default. Initially, $DATA$$ is defined as follows: I $DATA$ = "$DATA$,OCTA,NOPIC,CON,REL,LCL,NOSHR,NOEXE,RD,WRT" 6-44 d o I MACRO-64 MacroseI $ROUTINEe H Since you must delimit the psect name and attributes usingC commas, be sure to enclose this argument within anglemG brackets to avoid having the assembler interpret the nameeD and attributes as different arguments to $ROUTINE. For example: 6 DATA_SECTION=<MY_DATA,NOEXE,QUAD,RD,WRT> " DATA_SECTION_POINTERF Boolean value indicating whether $ROUTINE should store aD pointer to the data section in the linkage section andI define $DP as the address of that pointer. This argument is I optional for all procedure kinds. The default is FALSE. Forr example: ' DATA_SECTION_POINTER=TRUEc G You can use the DATA_SECTION_POINTER argument as follows: F $routine TALLY_HO, data_section_pointer=TRUE' $data_sectionr! TALLY: .quad 0 ' $code_section V .base R27, $LS ; Inform assembler that R27->$LS? LDQ R1, $DP ; R1->$DSDS .base R1,$DS ;Inform assembler that R1-$DSMA LDQ R0, TALLY ; R0<-TALLY @ LDA R0, 1(R0) ; R0<-R0++A STQ R0, TALLY ; TALLY<-R0 > RET (R26) ; Return/ $end_routine TALLY_HO C In this example, the DATA_SECTION_POINTER argument is_G specified in order to obtain linkage to the data section.aH The first LDQ instruction loads R1 with the pointer to theG data section that $ROUTINE stores in the linkage section.iD The next three instructions increment the value in theF TALLY variable in the data section. Finally, the routine@ returns the incremented value to its caller in R0. LINKAGE_SECTIONaC The psect name and attributes of the linkage section.dC This argument is optional for all procedure kinds. IfkE omitted, the default is the name and attributes definedtH by the $LINK$ lexical string symbol. If you specify a nameG and attributes for the LINKAGE_SECTION argument, $ROUTINE F redefines the $LINK$ lexical string symbol such that the I 6-45 MACRO-64 Macrosh $ROUTINE H specified values become the new default. Initially, $LINK$$ is defined as follows: K $LINK$ = "$LINK$,OCTA,NOPIC,CON,REL,LCL,NOSHR,NOEXE,RD,NOWRT"b H Since you must delimit the psect name and attributes usingC commas, be sure to enclose this argument within angleiG brackets to avoid having the assembler interpret the nameiD and attributes as different arguments to $ROUTINE. For example: ; LINKAGE_SECTION=<MY_LINK,NOEXE,QUAD,RD,NOWRT> KINDG Specifies the kind of routine being defined. This must be I one of the following: STACK, REGISTER, NULL, or BOUND. This_H is an optional argument. The default is NULL. For example: KIND=STACK ! HANDLER_REINVOKABLE I Specifies a boolean value to indicate whether the condition F handler can be re-invoked. This argument is optional andH valid only with STACK and REGISTER procedures. It defaultsF to FALSE if no value is specified and the procedure kindI is STACK or REGISTER. Note that this argument is only valid H if a value is also specified for the HANDLER argument. For example: & HANDLER_REINVOKABLE=TRUE BASE_REG_IS_FPD Specifies a boolean value to indicate whether registerH R29 (FP) is used as the frame-pointer base register (TRUE)E or not (FALSE). If specified as FALSE, R29 must be used E to hold the address of the procedure descriptor (or the I address of the address of the procedure descriptor-refer to H the OpenVMS Calling Standard. Thus you may use R29 to holdE a working copy of the linkage-section address passed in F R27. In addition, your prologue and epilogue instructionC sequences can be shorter and more efficient. However, E you cannot not make standard calls to other routines ifsD BASE_REG_IS_FP is specified as FALSE. This argument isG optional and valid only with stack and register procedureiG kinds. It defaults to TRUE if the procedure type is stacky' or register. For example: 6-46 e t I MACRO-64 MacrosoI $ROUTINE " BASE_REG_IS_FP=FALSE REI_RETURNH Specifies a boolean value to indicate whether this routineI returns using an REI instruction. This argument is optionalAE and valid only with STACK, REGISTER, and NULL procedure I kinds. It defaults to FALSE if the procedure kind is STACK,e- REGISTER, or NULL. For example:f REI_RETURN=TRUE3 STACK_RETURN_VALUEG Specifies a boolean value indicating whether this routine E has a stack return value. This argument is optional andmH valid only with STACK, REGISTER, and NULL procedure kinds.B It defaults to FALSE if the procedure kind is STACK,, REGISTER or NULL. For example: % STACK_RETURN_VALUE=TRUE RSA_OFFSETE An integer value specifying the stack offset (in bytes) F of the register save area. This argument is optional andG valid only for STACK procedures. If you specify BASE_REG__G IS_FP as TRUE, the value you specify with RSA_OFFSET mustRG be at least 8. RSA_OFFSET defaults to 8 if BASE_REG_IS_FPu= is specified as TRUE, 0 otherwise. For example: RSA_OFFSET=32O SAVE_FP H The register that contains a copy of the value of FP (R29)B upon entry to this routine. The prologue instructionF sequence must copy FP to the register specified by SAVE_F FP and the epilogue instruction sequence(s) must restoreI FP from the register specified by SAVE_FP. This argument isaG required and only valid for REGISTER procedures. There isn& no default. For example: SAVE_FP=R1 SAVE_RA A The register that contains the return address. This > argument is optional and only valid for REGISTERI procedures. If SAVE_RA is not R26, the prologue instruction G sequence must copy R26 to the register specified by SAVE_ F RA and the epilogue instruction sequence(s) must use the I 6-47 MACRO-64 Macros $ROUTINE H return address stored in the register specified by SAVE_FPH to affect its return to caller. SAVE_RA defaults to R26 if: the procedure kind is REGISTER. For example: SAVE_RA=R22 SIZEF A numeric value that must be a multiple of 16 specifyingG the size of the fixed stack area in bytes. This parameter_F is valid only for REGISTER and STACK procedure kinds. ItD defaults to the minimum value possible given the otherE arguments you specify or default. $ROUTINE computes theeG amount of stack storage needed for the register save areaiG (if any) and defines the $RSA_END symbol to be the offsetcE of the first byte beyond the register save area. If youaH wish to allocate a fixed area of stack beyond the registerD save area, you can specify an expression with the SAZEF argument that includes the term $RSA_END plus the amountG of fixed stack storage you need for your application. Foru example: SIZE=$RSA_END+32 SAVED_REGSI A list of registers saved on the stack by the prologue codeeH of this routine. It is valid only for STACK procedures andG you must specify at least FP (R29) with this argument. ItEE defaults to FP (R29) for STACK procedures. For example: ) SAVED_REGS=<R5,FP,F2,F3,F4> HANDLER E The address of an exception handler. It is optional and C valid only for STACK and REGISTER procedure kinds. By H default, the procedure is defined not to have an exception# handler. For example:R HANDLER=MY_HANDLER HANDLER_DATAI The address of data for the specified handler, if any. ThisnH argument is optional and valid only for stack and registerB procedure kinds and has no default value. You cannotG specify a HANDLER_DATA argument if you do not specify thea, HANDLER argument. For example: * HANDLER_DATA=MY_HANDLER_DATA 6-48 o I MACRO-64 Macros I $ROUTINEA SYNCH_EXCEPTIONS@ An argument to indicate whether exceptions must beF syncronized or not. This argument is optional with STACKG and REGISTER routines and is not allowed with other kinds_H of routines. This argument defaults to TRUE if you specifyH an exception handler with the HANDLER argument. Otherwise,B it defaults to FALSE. When this argument is TRUE andE you specify or default STANDARD_PROLOGUE=TRUE, $ROUTINEsC generates a TRAPB instruction as part of the standard C prologue sequence. In addition, when this argument isAF true, the $RETURN macro generates a TRAPB instruction asH part of the standard epilogue sequence. When this argumentC is FALSE, neither $ROUTINE nor $RETURN generate TRAPBv instructions. PROC_VALUEE The procedure value of a bound procedure's parent. This F argument is required for BOUND procedures and is invalid9 for all other procedure kinds. For example: $ PROC_VALUE=PARENT_PROC ENVIRONMENT H Specifies an environment value. This parameter is optionalD and valid only for BOUND procedures. It has no default! value. For example: ENVIRONMENT=0 FUNC_RETURN B Specifies the function return type. This argument isG optional and valid for all procedure kinds. If specified,tF it must be one of the following: I64, D64, I32, U32, FF,F FD, FG, FS, FT, FFC, FDC, FGC, FSC, or FTC. These valuesD correspond to those listed in Table 3-7 of the OpenVMSC Calling Standard that have an additional "RASE$K_FR_" 7 prefix. There is no default. For example:B FUNC_RETURN=U32F ARGLISTOI Argument type list. This argument is optional and valid for G all procedure kinds. If the argument list contains one or G more elements, each of the first six elements must be one I of the following: I64, I32, U32, FF, FD, FG, FS, or FT. TheP I 6-49E MACRO-64 Macrosa $ROUTINE F seventh and subsequent arguments (if any) must be either I32 or I64.u D These values correspond to the PSIG$K_RA_* and MASE$K_B MA_* signature encodings in Table 3-6 of the OpenVMSH Calling Standard. There is no default. If you specify thisG argument, $ROUTINE generates a procedure signature block.O For example: * ARGLIST=<I64,I32,FF,FF,U32>. USES_VAX_ARGLISTF Specifies a boolean value indicating whether the routineI uses a VAX argument list. This argument is optional for alleH procedure kinds and defaults to FALSE. If you specify thisG argument, $ROUTINE generates a procedure signature block. For example: # USES_VAX_ARGLIST=TRUEl OVERRIDE_FLAGSF Specifies overriding flags for the PDSC$W_FLAGS field inE the procedure descriptor. This argument is optional andoD valid for all procedure kinds. However, it is requiredH for BOUND procedures when the parameter specified with theF PROC_VALUE argument is an external or forward reference./ There is no default. For example: ) OVERRIDE_FLAGS=PARENT_FLAGS6 DEFAULT_SIGNATURE4H Specifies a boolean value to indicate whether the standardI procedure signature is used. TRUE means to use the standardNC signature. It is optional and valid for all procedureoC kinds. The default is FALSE if you specify either theyC ARGLIST or USES_VAX_ARGLIST arguments. Otherwise, theiD default is TRUE. Note that this argument must be FALSEE or blank if you specify either the ARGLIST or USES_VAX_l ARGLIST arguments. $ DEFAULT_SIGNATURE=TRUE COMMON_BASEaD An argument to specify one or more base registers thatC are used in common with other routines. This argumentsE is optional for all routine kinds. By default, $ROUTINEsC invalidates any previous .BASE directives that may ber 6-50 r e I MACRO-64 MacrostI $ROUTINE$ F in effect when you invoke $ROUTINE. In this way, you areC prevented from inadvertently processing the second ortF subsequent routines in a module with .BASE directives inD effect that apply only to a previous routine. However,E you may wish to share a common base register assignmentbH between a number of routines. To do so, you could re-issueF the appropriate .BASE directive or directives after eachD invocation of $ROUTINE. Alternatively, you can specifyD one or more common base registers with the COMMON_BASED argument, and issue the appropriate .BASE directive orB directives only once at the beginning of the module.F Specify the value for the COMMON_BASE argument as a list0 of integer registers. For example: " COMMON_BASE=<R5,R13> F In this example, $ROUTINE invalidates any previous .BASEH directives except those for registers R5 and R13. PreviousG .BASE directives for registers R5 and R13 are unaffected. Description G $ROUTINE defines a routine, makes it the current routine, 1 and performs the following actions: H o Creates and associates a linkage section, code section,3 and data section with the routine.w F o Defines a procedure descriptor and optional signatureH block in accordance with the values of macro arguments. G o Optionally stores a pointer to the data section withinX% the linkage section.C C o Creates the following numeric and lexical symbols:h I 6-51 e e MACRO-64 Macrosn $ROUTINE I ________________________________________________________oI Symbol________Description_______________________________ H $CS Address of start of the current routine's, code section. H $DS Address of start of the current routine's, data section. C $DP Optional address of a pointer to the C current routine's data section. This D symbol has a value that is an addressG in the current routine's linkage section E at which the $ROUTINE macro has placedDG the address of the data section ($DS) as ' follows: & $DP = .+ .ADDRESS $DSu H $DP enables a programmer to access theG data area of the current routine froma6 its linkage section. G $LS Address of the current routine's linkagee' section. G $SIZE Size of fixed area of stack frame of the I current routine. This symbol is valid onlyh@ with STACK and REGISTER routines. E $RSA_OFFSET The offset within the fixed stack areaiH to the register save area. This symbol is> valid only with STACK routines. H $RSA_END The offset within the fixed stack area toG the the first byte beyond the end of theO; register save area (if any).b G $CODE$ A lexical string symbol that defines the H routine's code psect name and attributes. @ $DATA$ A lexical symbol that defines theH routine's data psect name and attributes. C $LINK$ A lexical string symbol that defines C the routine's linkage psect name and I ______________attributes._______________________________ F o Optionally generates a standard, instruction prologue? sequence at the beginning of the code section. 6-52 I MACRO-64 Macros I $ROUTINE E If you specify /NAMES=AS_IS on the command line, all C but the last three of these symbols are defined in F both complete uppercase and complete lowercase. TheseH symbols are intended for your use outside of the macrosE themselves. For example, the values of these numeric C symbols may be useful as a mnemonic when coding an I instruction with a register as in the following example: % lda SP,-$SIZE(SP)s G The last three symbols ,$CODE$, $DATA$, and $LINK$ areeI only defined in uppercase. They are used by the $ROUTINENF macro for the default code, data, and linkage sectionI psect names and attributes. You can define these symbolsaH prior to invoking $ROUTINE to alter the default program% sections as follows: B - $CODE$ = "MY_CODE,EXE,OCTA,SHR,NORD,NOWRT,GBL" B - $DATA$ = "MY_DATA,NOEXE,OCTA,NOSHR,RD,WRT,GBL" B - $LINK$ = "MY_LINK,NOEXE,OCTA,SHR,RD,NOWRT,GBL" G These statements cause $ROUTINE to use the above psectsD names and attributes by default. If you specify anyF of the CODE_SECTION, DATA_SECTION, or LINKAGE_SECTIONC arguments in your invocation of $ROUTINE, $ROUTINEtF uses the psect name and attributes specified with the argument. B In addition, $ROUTINE redefines the correspondingC $CODE$, $DATA$, or $LINK$ lexical string symbol to B the value you specify when you specify any of theI CODE_SECTION, DATA_SECTION, or LINKAGE_SECTION arguments with $ROUTINE. Examples 1 $ROUTINE MAIN1, KIND=NULL 1 $ROUTINE MAIN1, - 1 KIND=STACK, - 1 SIZE=48, -g1 SAVED_REGS=<R2,FP,F5>t I 6-53A t I AEI _________________________________________________________________ I MACRO-64 Alpha AXP Architecture Quick Referencee C This appendix provides figures and tables showing the E data types and addressing capabilities of the Alpha AXPoI architecture. The information is derived from the Alpha AXP H Achitecture Quick Reference Guide. Minor changes have beenE made to reflect the usage of the Alpha AXP architecturet+ that is specific to MACRO-64.. H For more information, see the Alpha Architecture Reference6 Manual and the OpenVMS Calling Standard. > Figure A-1 shows integer and floating point data? structures. Figure A-2 shows instruction formats. I A-1 h e 7 MACRO-64 Alpha AXP Architecture Quick Reference A-2 c e I MACRO-64 Alpha AXP Architecture Quick ReferenceeI A.1 Register Usage Conventionsa & A.1 Register Usage Conventions H Table A-1 lists the register usage conventions for OpenVMSF AXP. MACRO-64 recognizes FP and SP as register synonyms,H but does not recognize AI, RA, or PV as register synonyms. I Table_A-1_Register_Usage_Conventions_for_OpenVMS_AXP_____________o For OpenVMS:+ R0 Int func ret value6 R1 Scratch R2-R15 Saved! R16- Argumentu R21 R22- Scratch R24 - R25 AI Argument informationA' R26 RA Return address ( R27 PV Procedure value) R28 Volatile scratcho) R29 FP Stack frame basel& R30 SP Stack pointer R31 Zero:/ F0 F-P function ret valueI3 F1 F-P complex func ret valuei F2-F9 Saved F10- Scratch F15T! F16- Argumentl F22E F23- Scratch F30 I F31______________Zero____________________________________________r I A-3d r e 7 MACRO-64 Alpha AXP Architecture Quick ReferenceS( A.2 Instruction Operand Notation ( A.2 Instruction Operand Notation H Table A-2 shows the notation for instruction operands. The, notation format is as follows: - <name>.<access type><data type> I Table_A-2_Instruction_Operand_Notation___________________________f <name>:y disp Displacement field$ fnc PALcode function field6 Ra Integer register operand in the Ra field6 Rb Integer register operand in the Rb field5 #b Integer literal operand in the Rb fieldd6 Rc Integer register operand in the Rc field= Fa Floating-point register operand in the Ra field = Fb Floating-point register operand in the Rb field = Fc Floating-point register operand in the Rc field <access type>:C a The operand is used in an address calculation to form C an effective address. The data type code that followslE indicates the units of addressability (or scale factor)dF applied to this operand when the instruction is decoded.2 i The operand is an immediate literal.3 m The operand is both read and written.r' r The operand is read only.c( w The operand is write only. <data type>: b Byte f F_floating g G_floating l Longword q Quadword/ s IEEE single floating (S_floating)M/ t IEEE double floating (T_floating) w WordI x_____The_data_type_is_specified_by_the_instruction______________ A-4 I MACRO-64 Alpha AXP Architecture Quick ReferenceiI A.3 Instruction Qualifier Notationa * A.3 Instruction Qualifier Notation F Table A-3 shows the notation for instruction qualifiers. I Table_A-3_Instruction_Qualifier_Notation_________________________c I /Qualifier_Meaning_______________________________________________t # C Chopped roundinga# D Dynamic roundingu1 (mode determined by FPCR<DYN>) ( I Inexact result enable* M Minus infinity rounding- S Software completion enable , U Floating underflow enableI V__________Integer_overflow_enable_______________________________n . A.4 F-P Control Register (FPCR) Format F Table A-4 shows the format for the F-P Control Register. I Table_A-4_F-P_Control_Register_(FPCR)_Format_____________________r I Bits____Symbol___Meaning_________________________________________v . 63 SUM Bitwise OR of <57:52>; 62:60 RAZ Read as zero; ignored when written /IGN5 59:58 DYN IEEE rounding mode selected:r % 00 Choppedi, 01 Minus infinity- 10 Normal rounding + 11 Plus infinityTB 57 IOV Integer overflow of destination precision? 56 INE Floating mathematically inexact resultfC 55 UNF Floating underflow of destination exponent,B 54 OVF Floating overflow of destination exponent= 53 DZE Floating divide with divisor of zero 7 52 INV Floating invalid operand valuet I (continued on next page)l I A-5 7 MACRO-64 Alpha AXP Architecture Quick Reference . A.4 F-P Control Register (FPCR) Format I Table_A-4_(Cont.)_F-P_Control_Register_(FPCR)_Format_____________e I Bits____Symbol___Meaning_________________________________________F ; 51:0 RAZ Read as zero; ignored when written I ________/IGN_____________________________________________________ A-6 I MACRO-64 Alpha AXP Architecture Quick ReferencedI A.5 Decodable Pseudo-OperationsS ' A.5 Decodable Pseudo-Operations C Table A-5 lists decodable pseudo-operations and theirs- associated actual instructions.o I Table_A-5_Decodable_Pseudo-Operations____________________________S & Pseudo- ActualI Operation_______________Instruction______________________________n 6 BR target BR R31,target6 CLR Rx BIS R31,R31,Rx5 FABS Fx,Fy CPYS F31,Fx,Fyu6 FCLR Fx CPYS F31,F31,Fx4 FMOV Fx,Fy CPYS Fx,Fx,Fy4 FNEG Fx,Fy CPYSN Fx,Fx,Fy7 FNOP CPYS F31,F31,F31 7 MOV Lit,Rx LDA Rx,lit(R31) < MOV {Rx BIS R31,{Rx/Lit8},Ry /Lit8},Ryw4 MF_FPCR Fx MF_FPCR Fx,Fx,Fx4 MT_FPCR Fx MT_FPCR Fx,Fx,Fx5 NEGF Fx,Fy SUBF F31,Fx,Fy 5 NEGF/S Fx,Fy SUBF/S F31,Fx,Fy 5 NEGG Fx,Fy SUBG F31,Fx,Fy 5 NEGG/S Fx,Fy SUBG/S F31,Fx,Fyr; NEGL {Rx SUBL R31,{Rx/Lit},Ry /Lit8},Ryo; NEGL/V {Rx SUBL/V R31,{Rx/Lit},Rye /Lit8},Ryi; NEGQ {Rx SUBQ R31,{Rx/Lit},Ryu /Lit8},Ry ; NEGQ/V {Rx SUBQ/V R31,{Rx/Lit},RyE /Lit8},Ryo5 NEGS Fx,Fy SUBS F31,Fx,Fy 5 NEGS/SU Fx,Fy SUBS/SU F31,Fx,Fy=5 NEGS Fx,FY SUBS/SUI F31,Fx,Fyg /SUI5 NEGT Fx,Fy SUBT F31,Fx,Fys I (continued on next page)o I A-7. E r 7 MACRO-64 Alpha AXP Architecture Quick Reference ' A.5 Decodable Pseudo-Operations I Table_A-5_(Cont.)_Decodable_Pseudo-Operations____________________ & Pseudo- ActualI Operation_______________Instruction______________________________ 5 NEGT/SU Fx,Fy SUBT/SU F31,Fx,Fy 5 NEGT Fx,FY SUBT/SUI F31,Fx,Fy /SUI7 NOP BIS R31,R31,R31,; NOT {Rx ORNOT R31,{Rx/Lit},Rya /Lit8},Ryw; SEXTL {Rx/Lit},Ry ADDL R31,{Rx/Lit},Ry I UNOP____________________LDQ_U_______R31,0(Rx)____________________ A-8 m I MACRO-64 Alpha AXP Architecture Quick Reference I A.6 Common Architecture Opcodes in Numerical Ordere : A.6 Common Architecture Opcodes in Numerical Order @ Table A-6 lists the common architecture opcodes in numerical orderO I Table_A-6_Common_Architecture_Opcodes_in_Numerical_Order_________ I Opcode________________Opcode________________Opcode_______________3 C 00 CALL_PAL 11.26 CMOVNE 15.01E CVTDG @ /CB 01 OPC01 11.28 ORNOT 15.020 ADDG@ /CB 02 OPC02 11.40 XOR 15.021 SUBG@ /CB 03 OPC03 11.44 CMOVLT 15.022 MULG@ /CB 04 OPC04 11.46 CMOVGE 15.023 DIVG@ /CC 05 OPC05 11.48 EQV 15.02C CVTGF @ /CC 06 OPC06 11.64 CMOVLE 15.02D CVTGD_@ /CC 07 OPC07 11.66 CMOVGT 15.02F CVTGQ @ /CC 08 LDA 12.02 MSKBL 15.03C CVTQF @ /CC 09 LDAH 12.06 EXTBL 15.03E CVTQG @ /CB 0A OPC0A 12.0B INSBL 15.080 ADDFB 0B LDQ_U 12.12 MSKWL 15.081 SUBFB 0C OPC0C 12.16 EXTWL 15.082 MULFB 0D OPC0D 12.1B INSWL 15.083 DIVFC 0E OPC0E 12.22 MSKLL 15.09E CVTDG B 0F STQ_U 12.26 EXTLL 15.0A0 ADDGB 10.00 ADDL 12.2B INSLL 15.0A1 SUBGB 10.02 S4ADDL 12.30 ZAP 15.0A2 MULGB 10.09 SUBL 12.31 ZAPNOT 15.0A3 DIVGD 10.0B S4SUBL 12.32 MSKQL 15.0A5 CMPGEQD 10.0F CMPBGE 12.34 SRL 15.0A6 CMPGLTD 10.12 S8ADDL 12.36 EXTQL 15.0A7 CMPGLE I (continued on next page)t I A-9 I 7 MACRO-64 Alpha AXP Architecture Quick Referenceh: A.6 Common Architecture Opcodes in Numerical Order I Table_A-6_(Cont.)_Common_Architecture_Opcodes_in_Numerical_Order_l I Opcode________________Opcode________________Opcode_______________e C 10.1B S8SUBL 12.39 SLL 15.0AC CVTGF C 10.1D CMPULT 12.3B INSQL 15.0AD CVTGDaC 10.20 ADDQ 12.3C SRA 15.0AF CVTGQbC 10.22 S4ADDQ 12.52 MSKWH 15.0BC CVTQFsC 10.29 SUBQ 12.57 INSWH 15.0BE CVTQGsB 10.2B S4SUBQ 12.5A EXTWH 15.100 ADDFA /UCxB 10.2D CMPEQ 12.62 MSKLH 15.101 SUBFA /UCaB 10.32 S8ADDQ 12.67 INSLH 15.102 MULFA /UCsB 10.3B S8SUBQ 12.6A EXTLH 15.103 DIVFA /UC C 10.3D CMPULE 12.72 MSKQH 15.11E CVTDG A /UCAB 10.40 ADDL/V 12.77 INSQH 15.120 ADDGA /UCpB 10.49 SUBL/V 12.7A EXTQH 15.121 SUBGA /UClB 10.4D CMPLT 13.00 MULL 15.122 MULGA /UCiB 10.60 ADDQ/V 13.20 MULQ 15.123 DIVGA /UCyC 10.69 SUBQ/V 13.30 UMULH 15.12C CVTGF A /UCoC 10.6D CMPLE 13.40 MULL/V 15.12D CVTGDeA /UC C 11.00 AND 13.60 MULQ/V 15.12F CVTGQ A /VCCB 11.08 BIC 14 OPC14 15.180 ADDF@ /UB 11.14 CMOVLBS 15.000 ADDF/C 15.181 SUBF@ /UB 11.16 CMOVLBC 15.001 SUBF/C 15.182 MULF@ /UB 11.20 BIS 15.002 MULF/C 15.183 DIVF@ /U I (continued on next page)t A-10 I MACRO-64 Alpha AXP Architecture Quick ReferenceeI A.6 Common Architecture Opcodes in Numerical OrderO I Table_A-6_(Cont.)_Common_Architecture_Opcodes_in_Numerical_Order_ I Opcode________________Opcode________________Opcode_______________ C 11.24 CMOVEQ 15.003 DIVF/C 15.19E CVTDG @ /UD 15.1A0 ADDG/U 15.580 ADDF/SU 16.0A6 CMPTLTD 15.1A1 SUBG/U 15.581 SUBF/SU 16.0A7 CMPTLEC 15.1A2 MULG/U 15.582 MULF/SU 16.0AC CVTTS C 15.1A3 DIVG/U 15.583 DIVF/SU 16.0AF CVTTQ_C 15.1AC CVTGF/U 15.59E CVTDG/SU 16.0BC CVTQS C 15.1AD CVTGD/U 15.5A0 ADDG/SU 16.0BE CVTQT B 15.1AF CVTGQ/V 15.5A1 SUBG/SU 16.0C0 ADDS@ /DB 15.400 ADDF/SC 15.5A2 MULG/SU 16.0C1 SUBS@ /DB 15.401 SUBF/SC 15.5A3 DIVG/SU 16.0C2 MULS@ /DB 15.402 MULF/SC 15.5AC CVTGF/SU 16.0C3 DIVS@ /DB 15.403 DIVF/SC 15.5AD CVTGD/SU 16.0E0 ADDT@ /DB 15.41E CVTDG/SC 15.5AF CVTGQ/SV 16.0E1 SUBT@ /DB 15.420 ADDG/SC 16.000 ADDS/C 16.0E2 MULT@ /DB 15.421 SUBG/SC 16.001 SUBS/C 16.0E3 DIVT@ /DC 15.422 MULG/SC 16.002 MULS/C 16.0EC CVTTS @ /DC 15.423 DIVG/SC 16.003 DIVS/C 16.0EF CVTTQC@ /DC 15.42C CVTGF/SC 16.020 ADDT/C 16.0FC CVTQSa@ /DC 15.42D CVTGD/SC 16.021 SUBT/C 16.0FE CVTQT @ /DB 15.42F CVTGQ/SC 16.022 MULT/C 16.100 ADDSA /UCoB 15.480 ADDF/S 16.023 DIVT/C 16.101 SUBSA /UCo I (continued on next page)p I A-11 u 7 MACRO-64 Alpha AXP Architecture Quick Reference2: A.6 Common Architecture Opcodes in Numerical Order I Table_A-6_(Cont.)_Common_Architecture_Opcodes_in_Numerical_Order_ I Opcode________________Opcode________________Opcode________________ B 15.481 SUBF/S 16.02C CVTTS/C 16.102 MULSA /UC B 15.482 MULF/S 16.02F CVTTQ/C 16.103 DIVSA /UCnB 15.483 DIVF/S 16.03C CVTQS/C 16.120 ADDTA /UC B 15.49E CVTDG/S 16.03E CVTQT/C 16.121 SUBTA /UC B 15.4A0 ADDG/S 16.040 ADDS/M 16.122 MULTA /UCdB 15.4A1 SUBG/S 16.041 SUBS/M 16.123 DIVTA /UC C 15.4A2 MULG/S 16.042 MULS/M 16.12C CVTTSeA /UCpC 15.4A3 DIVG/S 16.043 DIVS/M 16.12F CVTTQ A /VC B 15.4A5 CMPGEQ/S 16.060 ADDT/M 16.140 ADDSA /UMeB 15.4A6 CMPGLT/S 16.061 SUBT/M 16.141 SUBSA /UM B 15.4A7 CMPGLE/S 16.062 MULT/M 16.142 MULSA /UM B 15.4AC CVTGF/S 16.063 DIVT/M 16.143 DIVSA /UM B 15.4AD CVTGD/S 16.06C CVTTS/M 16.160 ADDTA /UMlB 15.4AF CVTGQ/S 16.06F CVTTQ/M 16.161 SUBTA /UM B 15.500 ADDF/SUC 16.07C CVTQS/M 16.162 MULTA /UMrB 15.501 SUBF/SUC 16.07E CVTQT/M 16.163 DIVTA /UMnC 15.502 MULF/SUC 16.080 ADDS 16.16C CVTTS_A /UM_C 15.503 DIVF/SUC 16.081 SUBS 16.16F CVTTQaA /VMt I (continued on next page) A-12 o i I MACRO-64 Alpha AXP Architecture Quick ReferencenI A.6 Common Architecture Opcodes in Numerical Order_ I Table_A-6_(Cont.)_Common_Architecture_Opcodes_in_Numerical_Order_ I Opcode________________Opcode________________Opcode________________ B 15.51E CVTDG/SUC 16.082 MULS 16.180 ADDS@ /UB 15.520 ADDG/SUC 16.083 DIVS 16.181 SUBS@ /UB 15.521 SUBG/SUC 16.0A0 ADDT 16.182 MULS@ /UB 15.522 MULG/SUC 16.0A1 SUBT 16.183 DIVS@ /UB 15.523 DIVG/SUC 16.0A2 MULT 16.1A0 ADDT@ /UB 15.52C CVTGF/SUC 16.0A3 DIVT 16.1A1 SUBT@ /UB 15.52D CVTGD/SUC 16.0A4 CMPTUN 16.1A2 MULT@ /UB 15.52F CVTGQ/SVC 16.0A5 CMPTEQ 16.1A3 DIVT@ /UC 16.1AC CVTTS/U 16.5AF CVTTQ/SV 16.7BC CVTQS B /SUIC 16.1AF CVTTQ/V 16.5C0 ADDS/SUD 16.7BE CVTQT_B /SUIB 16.1C0 ADDS/UD 16.5C1 SUBS/SUD 16.7C0 ADDSC /SUID B 16.1C1 SUBS/UD 16.5C2 MULS/SUD 16.7C1 SUBSC /SUIDpB 16.1C2 MULS/UD 16.5C3 DIVS/SUD 16.7C2 MULSC /SUIDlB 16.1C3 DIVS/UD 16.5E0 ADDT/SUD 16.7C3 DIVSC /SUIDsB 16.1E0 ADDT/UD 16.5E1 SUBT/SUD 16.7E0 ADDTC /SUIDtB 16.1E1 SUBT/UD 16.5E2 MULT/SUD 16.7E1 SUBTC /SUID B 16.1E2 MULT/UD 16.5E3 DIVT/SUD 16.7E2 MULTC /SUID B 16.1E3 DIVT/UD 16.5EC CVTTS/SUD 16.7E3 DIVTC /SUID I (continued on next page) I A-13 7 MACRO-64 Alpha AXP Architecture Quick Reference : A.6 Common Architecture Opcodes in Numerical Order I Table_A-6_(Cont.)_Common_Architecture_Opcodes_in_Numerical_Order_ I Opcode________________Opcode________________Opcode_______________ C 16.1EC CVTTS/UD 16.5EF CVTTQ/SVD 16.7EC CVTTS C /SUID C 16.1EF CVTTQ/VD 16.6AC CVTST/S 16.7EF CVTTQ,C /SVID C 16.2AC CVTST 16.700 ADDS/SUIC 16.7FC CVTQS C /SUID C 16.500 ADDS/SUC 16.701 SUBS/SUIC 16.7FE CVTQTFC /SUIDEC 16.501 SUBS/SUC 16.702 MULS/SUIC 17.010 CVTLQEB 16.502 MULS/SUC 16.703 DIVS/SUIC 17.020 CPYSC 16.503 DIVS/SUC 16.720 ADDT/SUIC 17.021 CPYSN C 16.520 ADDT/SUC 16.721 SUBT/SUIC 17.022 CPYSE A 16.521 SUBT/SUC 16.722 MULT/SUIC 17.024 MT_cB FPCRA 16.522 MULT/SUC 16.723 DIVT/SUIC 17.025 MF_ B FPCRE 16.523 DIVT/SUC 16.72C CVTTS/SUIC 17.02A FCMOVEQ E 16.52C CVTTS/SUC 16.72F CVTTQ/SVIC 17.02B FCMOVNE3E 16.52F CVTTQ/SVC 16.73C CVTQS/SUIC 17.02C FCMOVLT,E 16.540 ADDS/SUM 16.73E CVTQT/SUIC 17.02D FCMOVGE E 16.541 SUBS/SUM 16.740 ADDS/SUIM 17.02E FCMOVLE E 16.542 MULS/SUM 16.741 SUBS/SUIM 17.02F FCMOVGTxC 16.543 DIVS/SUM 16.742 MULS/SUIM 17.030 CVTQL C 16.560 ADDT/SUM 16.743 DIVS/SUIM 17.130 CVTQL @ /VC 16.561 SUBT/SUM 16.760 ADDT/SUIM 17.530 CVTQL A /SV C 16.562 MULT/SUM 16.761 SUBT/SUIM 18.0000 TRAPB B 16.563 DIVT/SUM 16.762 MULT/SUIM 18.0400 EXCB@ 16.56C CVTTS/SUM 16.763 DIVT/SUIM 18.4000 MBA 16.56F CVTTQ/SVM 16.76C CVTTS/SUIM 18.4400 WMB C 16.580 ADDS/SU 16.76F CVTTQ/SVIM 18.8000 FETCH D 16.581 SUBS/SU 16.77C CVTQS/SUIM 18.A000 FETCH_? M B 16.582 MULS/SU 16.77E CVTQT/SUIM 18.C000 RPCC@ 16.583 DIVS/SU 16.780 ADDS/SUI 18.E000 RC@ 16.5A0 ADDT/SU 16.781 SUBS/SUI 18.F000 RS I (continued on next page) A-14 C I MACRO-64 Alpha AXP Architecture Quick Reference I A.6 Common Architecture Opcodes in Numerical Order I Table_A-6_(Cont.)_Common_Architecture_Opcodes_in_Numerical_Order_ I Opcode________________Opcode________________Opcode_______________E C 16.5A1 SUBT/SU 16.782 MULS/SUI 19 PAL19 A 16.5A2 MULT/SU 16.783 DIVS/SUI 1A.0 JMP A 16.5A3 DIVT/SU 16.7A0 ADDT/SUI 1A.1 JSR A 16.5A4 CMPTUN/SU 16.7A1 SUBT/SUI 1A.2 RET B 16.5A5 CMPTEQ/SU 16.7A2 MULT/SUI 1A.3 JSR_G COROUTINE5C 16.5A6 CMPTLT/SU 16.7A3 DIVT/SUI 1B PAL1B C 16.5A7 CMPTLE/SU 16.7AC CVTTS/SUI 1C OPC1C0C 16.5AC CVTTS/SU 16.7AF CVTTQ/SVI 1D PAL1D B 1E PAL1E 2A LDL_L 36 FBGEB 1F PAL1F 2B LDQ_L 37 FBGTB 20 LDF 2C STL 38 BLBCA 21 LDG 2D STQ 39 BEQ0A 22 LDS 2E STL_C 3A BLT1A 23 LDT 2F STQ_C 3B BLE2B 24 STF 30 BR 3C BLBSA 25 STG 31 FBEQ 3D BNE5A 26 STS 32 FBLT 3E BGE0A 27 STT 33 FBLE 3F BGT5) 28 LDL 34 BSR I 29______LDQ___________35______FBNE_______________________________ I A-15e e h 7 MACRO-64 Alpha AXP Architecture Quick Reference / A.7 OpenVMS PALcode Instruction Summaryp / A.7 OpenVMS PALcode Instruction Summary_ > Table A-7 lists the OpenVMS unprivileged PALcodeE instructions and Table A-8 lists the OpenVMS privileged # PALcode instructions. I Table_A-7_OpenVMS_Unprivileged_PALcode_Instructions______________ I Mnemonic______Opcode____Description______________________________s C AMOVRM 00.00A1 Atomic move from register to memory E AMOVRR 00.00A0 Atomic move from register to register * BPT 00.0080 Breakpoint( BUGCHK 00.0081 Bugcheck5 CHMK 00.0083 Change mode to kernel 8 CHME 00.0082 Change mode to executive9 CHMS 00.0084 Change mode to supervisor 3 CHMU 00.0085 Change mode to user 6 GENTRAP 00.00AA Generate software trap7 IMB 00.0086 I-stream memory barrier B INSQHIL 00.0087 Insert into longword queue at head+ interlocked B INSQHILR 00.00A2 Insert into longword queue at head4 interlocked residentB INSQHIQ 00.0089 Insert into quadword queue at head+ interlocked2B INSQHIQR 00.00A4 Insert into quadword queue at head4 interlocked residentB INSQTIL 00.0088 Insert into longword queue at tail+ interlockedUB INSQTILR 00.00A3 Insert into longword queue at tail4 interlocked residentB INSQTIQ 00.008A Insert into quadword queue at tail+ interlocked B INSQTIQR 00.00A5 Insert into quadword queue at tail4 interlocked resident@ INSQUEL 00.008B Insert entry into longword queueI INSQUEL/D 00.008D Insert entry into longword queue deferred @ INSQUEQ 00.008C Insert entry into quadword queueI INSQUEQ/D 00.008E Insert entry into quadword queue deferred 5 PROBER 00.008F Probe for read access1 I (continued on next page) A-16 I MACRO-64 Alpha AXP Architecture Quick Reference I A.7 OpenVMS PALcode Instruction Summary I Table_A-7_(Cont.)_OpenVMS_Unprivileged_PALcode_Instructions______ I Mnemonic______Opcode____Description______________________________ 6 PROBEW 00.0090 Probe for write access5 RD_PS 00.0091 Move processor status-3 READ_UNQ 00.009E Read unique context B REI 00.0092 Return from exception or interruptB REMQHIL 00.0093 Remove from longword queue at head+ interlockedoB REMQHILR 00.00A6 Remove from longword queue at head4 interlocked residentB REMQHIQ 00.0095 Remove from quadword queue at head+ interlocked0B REMQHIQR 00.00A8 Remove from quadword queue at head4 interlocked residentB REMQTIL 00.0094 Remove from longword queue at tail+ interlocked1B REMQTILR 00.00A7 Remove from longword queue at tail4 interlocked residentB REMQTIQ 00.0096 Remove from quadword queue at tail+ interlocked B REMQTIQR 00.00A9 Remove from quadword queue at tail4 interlocked resident@ REMQUEL 00.0097 Remove entry from longword queueI REMQUEL/D 00.0099 Remove entry from longword queue deferred @ REMQUEQ 00.0098 Remove entry from quadword queueI REMQUEQ/D 00.009A Remove entry from quadword queue deferred 9 RSCC 00.009D Read system cycle counter @ SWASTEN 00.009B Swap AST enable for current mode4 WRITE_UNQ 00.009F Write unique contextI WR_PS_SW______00.009C___Write_processor_status_software_field____ I Table_A-8_OpenVMS_Privileged_PALcode_Instructions________________C I Mnemonic_________Opcode____Description___________________________ . CFLUSH 00.0001 Cache flush2 CSERVE 00.0009 Console service/ DRAINA 00.0002 Drain aborts 1 HALT 00.0000 Halt processorC I (continued on next page) I A-17 7 MACRO-64 Alpha AXP Architecture Quick Reference / A.7 OpenVMS PALcode Instruction SummaryQ I Table_A-8_(Cont.)_OpenVMS_Privileged_PALcode_Instructions________5 I Mnemonic_________Opcode____Description___________________________ 9 LDQP 00.0003 Load quadword physicalFC MFPR_ASN 00.0006 Move from processor register ASN C MFPR_ESP 00.001E Move from processor register ESPDC MFPR_FEN 00.000B Move from processor register FEN C MFPR_IPL 00.000E Move from processor register IPL D MFPR_MCES 00.0010 Move from processor register MCESD MFPR_PCBB 00.0012 Move from processor register PCBBD MFPR_PRBR 00.0013 Move from processor register PRBRD MFPR_PTBR 00.0015 Move from processor register PTBRD MFPR_SCBB 00.0016 Move from processor register SCBBD MFPR_SISR 00.0019 Move from processor register SISRC MFPR_SSP 00.0020 Move from processor register SSP E MFPR_TBCHK 00.001A Move from processor register TBCHKUC MFPR_USP 00.0022 Move from processor register USPSD MFPR_VPTB 00.0029 Move from processor register VPTBE MFPR_WHAMI 00.003F Move from processor register WHAMI C MTPR_ASTEN 00.0026 Move to processor register ASTEN C MTPR_ASTSR 00.0027 Move to processor register ASTSR C MTPR_DATFX 00.002E Move to processor register DATFX5A MTPR_ESP 00.001F Move to processor register ESP A MTPR_FEN 00.000B Move to processor register FEN5B MTPR_IPIR 00.000D Move to processor register IPRIA MTPR_IPL 00.000E Move to processor register IPL5B MTPR_MCES 00.0011 Move to processor register MCESE MTPR_PERFMON 00.002B Move to processor register PERFMON4B MTPR_PRBR 00.0014 Move to processor register PRBRB MTPR_SCBB 00.0017 Move to processor register SCBBB MTPR_SIRR 00.0018 Move to processor register SIRRA MTPR_SSP 00.0021 Move to processor register SSP5B MTPR_TBIA 00.001B Move to processor register TBIAC MTPR_TBIAP 00.001C Move to processor register TBIAP4B MTPR_TBIS 00.001D Move to processor register TBISC MTPR_TBISD 00.0024 Move to processor register TBISDCC MTPR_TBISI 00.0025 Move to processor register TBISI I (continued on next page) A-18 6 0 I MACRO-64 Alpha AXP Architecture Quick Reference I A.7 OpenVMS PALcode Instruction Summary I Table_A-8_(Cont.)_OpenVMS_Privileged_PALcode_Instructions________C I Mnemonic_________Opcode____Description___________________________ A MTPR_USP 00.0023 Move to processor register USPEB MTPR_VPTB 00.002A Move to processor register VPTB: STQP 00.0004 Store quadword physical: SWPCTX 00.0005 Swap privileged contextI SWPPAL___________00.000A___Swap_PALcode_image____________________ I A-19 7 MACRO-64 Alpha AXP Architecture Quick Reference . A.8 PALcode Opcodes in Numerical Order . A.8 PALcode Opcodes in Numerical Order E Table A-9 lists the PALcode opcodes in numerical order._ I Table_A-9_PALcode_Opcodes_in_Numerical_Order_____________________e I Opcode(16)Opcode(10)OpenVMS______________________________________E 00.0000 00.0000 HALT" 00.0001 00.0001 CFLUSH" 00.0002 00.0002 DRAINA 00.0003 00.0003 LDQP 00.0004 00.0004 STQP" 00.0005 00.0005 SWPCTX$ 00.0006 00.0006 MFPR_ASN& 00.0007 00.0007 MTPR_ASTEN& 00.0008 00.0008 MTPR_ASTSR" 00.0009 00.0009 CSERVE" 00.000A 00.0010 SWPPAL$ 00.000B 00.0011 MFPR_FEN$ 00.000C 00.0012 MTPR_FEN% 00.000D 00.0013 MTPR_IPIR $ 00.000E 00.0014 MFPR_IPL$ 00.000F 00.0015 MTPR_IPL% 00.0010 00.0016 MFPR_MCES % 00.0011 00.0017 MTPR_MCESV% 00.0012 00.0018 MFPR_PCBB1% 00.0013 00.0019 MFPR_PRBR % 00.0014 00.0020 MTPR_PRBR % 00.0015 00.0021 MFPR_PTBRN% 00.0016 00.0022 MFPR_SCBB % 00.0017 00.0023 MTPR_SCBB % 00.0018 00.0024 MTPR_SIRR % 00.0019 00.0025 MFPR_SISR & 00.001A 00.0026 MFPR_TBCHK% 00.001B 00.0027 MTPR_TBIA & 00.001C 00.0028 MTPR_TBIAP% 00.001D 00.0029 MTPR_TBIS $ 00.001E 00.0030 MFPR_ESP$ 00.001F 00.0031 MTPR_ESP$ 00.0020 00.0032 MFPR_SSP I (continued on next page)1 A-20 5 I MACRO-64 Alpha AXP Architecture Quick Reference I A.8 PALcode Opcodes in Numerical Order1 I Table_A-9_(Cont.)_PALcode_Opcodes_in_Numerical_Order_____________ I Opcode(16)Opcode(10)OpenVMS______________________________________ $ 00.0021 00.0033 MTPR_SSP$ 00.0022 00.0034 MFPR_USP$ 00.0023 00.0035 MTPR_USP& 00.0024 00.0036 MTPR_TBISD& 00.0025 00.0037 MTPR_TBISI& 00.0026 00.0038 MFPR_ASTEN& 00.0027 00.0039 MFPR_ASTSR% 00.0029 00.0040 MFPR_VPTB % 00.002A 00.0041 MTPR_VPTBT( 00.002B 00.0042 MTPR_PERFMON& 00.002E 00.0043 MTPR_DATFX& 00.003F 00.0063 MFPR_WHAMI 00.0080 00.0128 BPT " 00.0081 00.0129 BUGCHK 00.0082 00.0130 CHME 00.0083 00.0131 CHMK 00.0084 00.0132 CHMS 00.0085 00.0133 CHMU 00.0086 00.0134 IMB # 00.0087 00.0135 INSQHIL # 00.0088 00.0136 INSQTIL # 00.0089 00.0137 INSQHIQ # 00.008A 00.0138 INSQTIQ # 00.008B 00.0139 INSQUEL # 00.008C 00.0140 INSQUEQa% 00.008D 00.0141 INSQUEL/D % 00.008E 00.0142 INSQUEQ/DN" 00.008F 00.0143 PROBER" 00.0090 00.0144 PROBEW! 00.0091 00.0145 RD_PSp 00.0092 00.0146 REI_# 00.0093 00.0147 REMQHIL # 00.0094 00.0148 REMQTILT# 00.0095 00.0149 REMQHIQ # 00.0096 00.0150 REMQTIQ # 00.0097 00.0151 REMQUEL # 00.0098 00.0152 REMQUEQ6 I (continued on next page) I A-21 7 MACRO-64 Alpha AXP Architecture Quick Reference . A.8 PALcode Opcodes in Numerical Order I Table_A-9_(Cont.)_PALcode_Opcodes_in_Numerical_Order_____________ I Opcode(16)Opcode(10)OpenVMS______________________________________ % 00.0099 00.0153 REMQUEL/DS% 00.009A 00.0154 REMQUEQ/D3# 00.009B 00.0155 SWASTEN $ 00.009C 00.0156 WR_PS_SW 00.009D 00.0157 RSCC$ 00.009E 00.0158 READ_UNQ% 00.009F 00.0159 WRITE_UNQ " 00.00A0 00.0160 AMOVRR" 00.00A1 00.0161 AMOVRM$ 00.00A2 00.0162 INSQHILR$ 00.00A3 00.0163 INSQTILR$ 00.00A4 00.0164 INSQHIQR$ 00.00A5 00.0165 INSQTIQR$ 00.00A6 00.0166 REMQHILR$ 00.00A7 00.0167 REMQTILR$ 00.00A8 00.0168 REMQHIQR$ 00.00A9 00.0169 REMQTIQRI 00.00AA___00.0170___GENTRAP______________________________________ A-22 U S I MACRO-64 Alpha AXP Architecture Quick Reference6I A.9 Common Architecture Instructions2 , A.9 Common Architecture Instructions F Table A-10 lists common architecture instructions. WhereF enclosed in square brackets, Ra.wq defaults to R31. This7 occurs for instructions BR, JMP, and RET. I Table_A-10_Common_Architecture_Instructions______________________M I Mnemonic________Code____Operands______________Operation__________6 E ADDF 15.080 Fa.rf,Fb.rf,Fc.wf Fc< - Fav + Fbv ADDF/C 15.000 ADDF/S 15.480 ADDF/SC 15.400 ADDF/SUC 15.500 ADDF/SU 15.580 ADDF/U 15.180 ADDF/UC 15.100E ADDG 15.0A0 Fa.rg,Fb.rg,Fc.wg Fc< - Fav + Fbv ADDG/C 15.020 ADDG/S 15.4A0 ADDG/SC 15.420 ADDG/SU 15.5A0 ADDG/SUC 15.520 ADDG/U 15.1A0 ADDG/UC 15.120G ADDL 10.00 Ra.rl,{Rb.rl Rc< - SEXT((Rav +-A /#b.ib},Rc.wq Rbv)<31:0>)i ADDL/V 10.40 E ADDQ 10.20 Ra.rq,{Rb.rq Rc< - Rav + Rbv - /#b.ib},Rc.wq_ ADDQ/V 10.60 E ADDS 16.080 Fa.rs,Fb.rs,Fc.ws Fc< - Fav + Fbv_ ADDS/C 16.000 ADDS/D 16.0C0 ADDS/M 16.040 ADDS/SU 16.580 ADDS/SUC 16.500 ADDS/SUD 16.5C0 ADDS/SUI 16.780 ADDS/SUIC 16.700 I (continued on next page)A I A-23O N 7 MACRO-64 Alpha AXP Architecture Quick ReferenceA, A.9 Common Architecture Instructions I Table_A-10_(Cont.)_Common_Architecture_Instructions______________ I Mnemonic________Code____Operands______________Operation__________ ADDS/SUID 16.7C0 ADDS/SUIM 16.740 ADDS/SUM 16.540 ADDS/U 16.180 ADDS/UC 16.100 ADDS/UD 16.1C0 ADDS/UM 16.140E ADDT 16.0A0 Fa.rt,Fb.rt,Fc.wt Fc< - Fav + Fbv ADDT/C 16.020 ADDT/D 16.0E0 ADDT/M 16.060 ADDT/SU 16.5A0 ADDT/SUC 16.520 ADDT/SUD 16.5E0 ADDT/SUI 16.7A0 ADDT/SUIC 16.720 ADDT/SUID 16.7E0 ADDT/SUIM 16.760 ADDT/SUM 16.560 ADDT/U 16.1A0 ADDT/UC 16.120 ADDT/UD 16.1E0 ADDT/UM 16.160G AND 11.00 Ra.rq,{Rb.rq Rc< - Rav AND RbvA- /#b.ib},Rc.wqA@ BEQ 39 Ra.rq,disp.al If Rav = 0E Then PC< - PC + D {4*SEXT(disp)}@ BGE 3E Ra.rq,disp.al If Rav 0E Then PC< - PC +lD {4*SEXT(disp)}@ BGT 3F Ra.rq,disp.al If Rav > 0E Then PC< - PC + D {4*SEXT(disp)}H BIC 11.08 Ra.rq,{Rb.rq Rc< - Rav AND {NOT: /#b.ib},Rc.wq Rbv} I (continued on next page)0 A-24 p i I MACRO-64 Alpha AXP Architecture Quick Reference I A.9 Common Architecture Instructionsm I Table_A-10_(Cont.)_Common_Architecture_Instructions______________e I Mnemonic________Code____Operands______________Operation__________A F BIS 11.20 Ra.rq,{Rb.rq Rc< - Rav OR Rbv- /#b.ib},Rc.wq8C BLBC 38 Ra.rq,disp.al If Rav<0> = 0 E Then PC< - PC +eD {4*SEXT(disp)}C BLBS 3C Ra.rq,disp.al If Rav<0> = 1 E Then PC< - PC + D {4*SEXT(disp)}@ BLE 3B Ra.rq,disp.al If Rav 0E Then PC< - PC + D {4*SEXT(disp)}@ BLT 3A Ra.rq,disp.al If Rav < 0E Then PC< - PC +0D {4*SEXT(disp)}A BNE 3D Ra.rq,disp.al If Rav /= 0 E Then PC< - PC + D {4*SEXT(disp)}J BR 30 [Ra.wq],disp.al Ra< - PC; PC< - PC +D {4*SEXT(disp)}J BSR 34 Ra.wq,disp.al Ra< - PC; PC< - PC +D {4*SEXT(disp)}E CALL_PAL 00 fnc.ir Trap to PALcode_E CMOVEQ 11.24 Ra.rq,{Rb.rq If Rav = 0 Thenr? /#b.ib},Rc.wq Rc< - Rbv-E CMOVGE 11.46 Ra.rq,{Rb.rq If Rav 0 Then ? /#b.ib},Rc.wq Rc< - RbvIE CMOVGT 11.66 Ra.rq,{Rb.rq If Rav > 0 Then ? /#b.ib},Rc.wq Rc< - RbveH CMOVLBC 11.16 Ra.rq,{Rb.rq If Rav<0> = 0 Then? /#b.ib},Rc.wq Rc< - RbveH CMOVLBS 11.14 Ra.rq,{Rb.rq If Rav<0> = 1 Then? /#b.ib},Rc.wq Rc< - Rbv E CMOVLE 11.64 Ra.rq,{Rb.rq If Rav 0 Then ? /#b.ib},Rc.wq Rc< - Rbv E CMOVLT 11.44 Ra.rq,{Rb.rq If Rav < 0 Thene? /#b.ib},Rc.wq Rc< - Rbv I (continued on next page) I A-25u o q 7 MACRO-64 Alpha AXP Architecture Quick Referencee, A.9 Common Architecture Instructions I Table_A-10_(Cont.)_Common_Architecture_Instructions______________o I Mnemonic________Code____Operands______________Operation__________d F CMOVNE 11.26 Ra.rq,{Rb.rq If Rav /= 0 Then? /#b.ib},Rc.wq Rc< - RbveD CMPBGE 10.0F Ra.rq,{Rb.rq Rc< - BytewiseE /#b.ib},Rc.wq compare mask of_A {Rav Rbv} G CMPEQ 10.2D Ra.rq,{Rb.rq If Rav = Rbv Then J /#b.ib},Rc.wq Rc< - 1 Else Rc< - 0G CMPGEQ 15.0A5 Fa.rg,Fb.rg,Fc.wq If Fav = Fbv Then.D Fc< - 0.5 Else? Fc< - 0.0 CMPGEQ/S 15.4A5G CMPGLE 15.0A7 Fa.rg,Fb.rg,Fc.wq If Fav Fbv Then D Fc< - 0.5 Else? Fc< - 0.0A CMPGLE/S 15.4A7G CMPGLT 15.0A6 Fa.rg,Fb.rg,Fc.wq If Fav < Fbv Then5D Fc< - 0.5 Else? Fc< - 0.0i CMPGLT/S 15.4A6G CMPLE 10.6D Ra.rq,{Rb.rq If Rav Rbv Then0J /#b.ib},Rc.wq Rc< - 1 Else Rc< - 0G CMPLT 10.4D Ra.rq,{Rb.rq If Rav < Rbv Then J /#b.ib},Rc.wq Rc< - 1 Else Rc< - 0G CMPTEQ 16.0A5 Fa.rt,Fb.rt,Fc.wq If Fav = Fbv TheneD Fc< - 2.0 Else? Fc< - 0.0p CMPTEQ/SU 16.5A5G CMPTLE 16.0A7 Fa.rt,Fb.rt,Fc.wq If Fav Fbv ThenRD Fc< - 2.0 Else? Fc< - 0.0 CMPTLE/SU 16.5A7G CMPTLT 16.0A6 Fa.rt,Fb.rt,Fc.wq If Fav < Fbv ThenrD Fc< - 2.0 Else? Fc< - 0.0v CMPTLT/SU 16.5A6 I (continued on next page) A-26 7 M I MACRO-64 Alpha AXP Architecture Quick ReferenceeI A.9 Common Architecture Instructionsr I Table_A-10_(Cont.)_Common_Architecture_Instructions______________5 I Mnemonic________Code____Operands______________Operation__________ F CMPTUN 16.0A4 Fa.rt,Fb.rt,Fc.wq If Fav UnorderedH Fbv Then Fc< - 2.0D Else Fc< - 0.0 CMPTUN/SU 16.5A4H CMPULE 10.3D Ra.rq,{Rb.rq If Rav U Rbv ThenJ /#b.ib},Rc.wq Rc< - 1 Else Rc< - 0H CMPULT 10.1D Ra.rq,{Rb.rq If Rav U< Rbv ThenJ /#b.ib},Rc.wq Rc< - 1 Else Rc< - 0F CPYS 17.020 Fa.rq,Fb.rq,Fc.wq Fc< - Fav<63> ||? Fbv<62:0>TI CPYSE 17.022 Fa.rq,Fb.rq,Fc.wq Fc< - Fav<63:52> ||I? Fbv<51:0> G CPYSN 17.021 Fa.rq,Fb.rq,Fc.wq Fc< - NOT Fav<63> B || Fbv<62:0>F CVTDG 15.09E Fb.rd,Fc.wg Fc< - D_Float toD G_Float of Fbv CVTDG/C 15.01E CVTDG/S 15.49E CVTDG/SC 15.41E CVTDG/SU 15.59E CVTDG/SUC 15.51E CVTDG/U 15.19E CVTDG/UC 15.11EF CVTGD 15.0AD Fb.rg,Fc.wd Fc< - G_Float toD D_Float of Fbv CVTGD/C 15.02D CVTGD/S 15.4AD CVTGD/SC 15.42D CVTGD/SU 15.5AD CVTGD/SUC 15.52D CVTGD/U 15.1AD CVTGD/UC 15.12DF CVTGF 15.0AC Fb.rg,Fc.wf Fc< - G_Float toD F_Float of Fbv CVTGF/C 15.02C CVTGF/S 15.4AC CVTGF/SC 15.42C I (continued on next page)l I A-27_ _ _ 7 MACRO-64 Alpha AXP Architecture Quick Reference , A.9 Common Architecture Instructions I Table_A-10_(Cont.)_Common_Architecture_Instructions______________ I Mnemonic________Code____Operands______________Operation__________N CVTGF/SU 15.5AC CVTGF/SUC 15.52C CVTGF/U 15.1AC CVTGF/UC 15.12CF CVTGQ 15.0AF Fb.rg,Fc.wq Fc< - G_Float toA quad of Fbv CVTGQ/C 15.02F CVTGQ/S 15.4AF CVTGQ/SC 15.42F CVTGQ/SV 15.5AF CVTGQ/SVC 15.52F CVTGQ/V 15.1AF CVTGQ/VC 15.12FX CVTLQ 17.010 Fb.rl,Fc.wq Fc< - SEXT(Fbv<63:62>||Fbv<58:29>)F CVTQF 15.0BC Fb.rq,Fc.wf Fc< - Quad to F_B Float of Fbv CVTQF/C 15.03CF CVTQG 15.0BE Fb.rq,Fc.wg Fc< - Quad to G_B Float of Fbv CVTQG/C 15.03EH CVTQL 17.030 Fb.rq,Fc.wl Fc< - Quad to long< of Fbv CVTQL/SV 17.530 CVTQL/V 17.130F CVTQS 16.0BC Fb.rq,Fc.ws Fc< - Quad to S_B Float of Fbv CVTQS/C 16.03C CVTQS/D 16.0FC CVTQS/M 16.07C CVTQS/SUI 16.7BC CVTQS/SUIC 16.73C CVTQS/SUID 16.7FC CVTQS/SUIM 16.77CF CVTQT 16.0BE Fb.rq,Fc.wt Fc< - Quad to T_B Float of Fbv CVTQT/C 16.03E CVTQT/D 16.0FE I (continued on next page) A-28 I MACRO-64 Alpha AXP Architecture Quick Reference I A.9 Common Architecture Instructions I Table_A-10_(Cont.)_Common_Architecture_Instructions______________ I Mnemonic________Code____Operands______________Operation__________ CVTQT/M 16.07E CVTQT/SUI 16.7BE CVTQT/SUIC 16.73E CVTQT/SUID 16.7FE CVTQT/SUIM 16.77EF CVTST 16.2AC Fb.rs,Fc.wt Fc< - S_Float toD T_Float of Fbv CVTST/S 16.6ACF CVTTQ 16.0AF Fb.rt,Fc.wq Fc< - T_Float toA quad of Fbv CVTTQ/C 16.02F CVTTQ/D 16.0EF CVTTQ/M 16.06F CVTTQ/SV 16.5AF CVTTQ/SVC 16.52F CVTTQ/SVD 16.5EF CVTTQ/SVI 16.7AF CVTTQ/SVIC 16.72F CVTTQ/SVID 16.7EF CVTTQ/SVIM 16.76F CVTTQ/SVM 16.56F CVTTQ/V 16.1AF CVTTQ/VC 16.12F CVTTQ/VD 16.1EF CVTTQ/VM 16.16FF CVTTS 16.0AC Fb.rt,Fc.ws Fc< - T_Float toD S_Float of Fbv CVTTS/C 16.02C CVTTS/D 16.0EC CVTTS/M 16.06C CVTTS/SU 16.5AC CVTTS/SUC 16.52C CVTTS/SUD 16.5EC CVTTS/SUI 16.7AC CVTTS/SUIC 16.72C CVTTS/SUID 16.7EC CVTTS/SUIM 16.76C I (continued on next page)M I A-29 0 7 MACRO-64 Alpha AXP Architecture Quick ReferenceR, A.9 Common Architecture Instructions I Table_A-10_(Cont.)_Common_Architecture_Instructions______________0 I Mnemonic________Code____Operands______________Operation___________ CVTTS/SUM 16.56C CVTTS/U 16.1AC CVTTS/UC 16.12C CVTTS/UD 16.1EC CVTTS/UM 16.16CE DIVF 15.083 Fa.rf,Fb.rf,Fc.wf Fc< - Fav / Fbvu DIVF/C 15.003 DIVF/S 15.483 DIVF/SC 15.403 DIVF/SU 15.583 DIVF/SUC 15.503 DIVF/U 15.183 DIVF/UC 15.103E DIVG 15.0A3 Fa.rg,Fb.rg,Fc.wg Fc< - Fav / Fbvm DIVG/C 15.023 DIVG/S 15.4A3 DIVG/SC 15.423 DIVG/SU 15.5A3 DIVG/SUC 15.523 DIVG/U 15.1A3 DIVG/UC 15.123E DIVS 16.083 Fa.rs,Fb.rs,Fc.ws Fc< - Fav / Fbv5 DIVS/C 16.003 DIVS/D 16.0C3 DIVS/M 16.043 DIVS/SU 16.583 DIVS/SUC 16.503 DIVS/SUD 16.5C3 DIVS/SUI 16.783 DIVS/SUIC 16.703 DIVS/SUID 16.7C3 DIVS/SUIM 16.743 DIVS/SUM 16.543 DIVS/U 16.183 DIVS/UC 16.103 DIVS/UD 16.1C3 DIVS/UM 16.143 I (continued on next page) A-30 L I MACRO-64 Alpha AXP Architecture Quick ReferenceaI A.9 Common Architecture Instructions I Table_A-10_(Cont.)_Common_Architecture_Instructions______________ I Mnemonic________Code____Operands______________Operation__________ E DIVT 16.0A3 Fa.rt,Fb.rt,Fc.wt Fc< - Fav / Fbv DIVT/C 16.023 DIVT/D 16.0E3 DIVT/M 16.063 DIVT/SU 16.5A3 DIVT/SUC 16.523 DIVT/SUD 16.5E3 DIVT/SUI 16.7A3 DIVT/SUIC 16.723 DIVT/SUID 16.7E3 DIVT/SUIM 16.763 DIVT/SUM 16.563 DIVT/U 16.1A3 DIVT/UC 16.123 DIVT/UD 16.1E3 DIVT/UM 16.163H EQV 11.48 Ra.rq,{Rb.rq Rc< - Rav XOR {NOT: /#b.ib},Rc.wq Rbv}G EXCB 18.0400 Exception barrierUH EXTBL 12.06 Ra.rq,{Rb.rq Rc< - Byte extractE /#b.ib},Rc.wq low from Rav at > Rbv<2:0>H EXTLH 12.6A Ra.rq,{Rb.rq Rc< - Long extractF /#b.ib},Rc.wq high from Rav at> Rbv<2:0>H EXTLL 12.26 Ra.rq,{Rb.rq Rc< - Long extractE /#b.ib},Rc.wq low from Rav at1> Rbv<2:0>H EXTQH 12.7A Ra.rq,{Rb.rq Rc< - Quad extractF /#b.ib},Rc.wq high from Rav at> Rbv<2:0>H EXTQL 12.36 Ra.rq,{Rb.rq Rc< - Quad extractE /#b.ib},Rc.wq low from Rav at > Rbv<2:0>H EXTWH 12.5A Ra.rq,{Rb.rq Rc< - Word extractF /#b.ib},Rc.wq high from Rav at> Rbv<2:0> I (continued on next page) I A-31, . 7 MACRO-64 Alpha AXP Architecture Quick Reference., A.9 Common Architecture Instructions I Table_A-10_(Cont.)_Common_Architecture_Instructions______________ I Mnemonic________Code____Operands______________Operation__________e H EXTWL 12.16 Ra.rq,{Rb.rq Rc< - Word extractE /#b.ib},Rc.wq low from Rav at_> Rbv<2:0>@ FBEQ 31 Fa.rq,disp.al If Fav = 0E Then PC< - PC +#D {4*SEXT(disp)}@ FBGE 36 Fa.rq,disp.al If Fav 0E Then PC< - PC + D {4*SEXT(disp)}@ FBGT 37 Fa.rq,disp.al If Fav > 0E Then PC< - PC + D {4*SEXT(disp)}@ FBLE 33 Fa.rq,disp.al If Fav 0E Then PC< - PC + D {4*SEXT(disp)}@ FBLT 32 Fa.rq,disp.al If Fav < 0E Then PC< - PC + D {4*SEXT(disp)}A FBNE 35 Fa.rq,disp.al If Fav /= 0 E Then PC< - PC + D {4*SEXT(disp)}E FCMOVEQ 17.02A Fa.rq,Fb.rq,Fc.wq If Fav = 0 Then ? Fc< - FbviE FCMOVGE 17.02D Fa.rq,Fb.rq,Fc.wq If Fav 0 Then ? Fc< - Fbv E FCMOVGT 17.02F Fa.rq,Fb.rq,Fc.wq If Fav > 0 Then ? Fc< - Fbv E FCMOVLE 17.02E Fa.rq,Fb.rq,Fc.wq If Fav 0 Then ? Fc< - Fbv E FCMOVLT 17.02C Fa.rq,Fb.rq,Fc.wq If Fav < 0 Then ? Fc< - Fbv F FCMOVNE 17.02B Fa.rq,Fb.rq,Fc.wq If Fav /= 0 Then? Fc< - Fbv E FETCH 18.8000 0(Rb.ab) Prefetch around ; (Rbv) I (continued on next page) A-32 f v I MACRO-64 Alpha AXP Architecture Quick Reference I A.9 Common Architecture Instructions I Table_A-10_(Cont.)_Common_Architecture_Instructions______________ I Mnemonic________Code____Operands______________Operation__________ E FETCH_M 18.A000 0(Rb.ab) Prefetch around C (Rbv), modify < intentG INSBL 12.0B Ra.rq,{Rb.rq Rc< - Byte insert_E /#b.iq},Rc.wq low from Rav atE> Rbv<2:0>G INSLH 12.67 Ra.rq,{Rb.rq Rc< - Long insertEF /#b.iq},Rc.wq high from Rav at> Rbv<2:0>G INSLL 12.2B Ra.rq,{Rb.rq Rc< - Long insert E /#b.iq},Rc.wq low from Rav at > Rbv<2:0>G INSQH 12.77 Ra.rq,{Rb.rq Rc< - Quad insert F /#b.iq},Rc.wq high from Rav at> Rbv<2:0>G INSQL 12.3B Ra.rq,{Rb.rq Rc< - Quad insertrE /#b.iq},Rc.wq low from Rav at > Rbv<2:0>G INSWH 12.57 Ra.rq,{Rb.rq Rc< - Word insertMF /#b.iq},Rc.wq high from Rav at> Rbv<2:0>G INSWL 12.1B Ra.rq,{Rb.rq Rc< - Word insertME /#b.iq},Rc.wq low from Rav at > Rbv<2:0>I JMP 1A.0 [Ra.wq],(Rb.ab),hint Ra< - PC; PC< - RbvRA AND {NOT 3}#I JSR 1A.1 Ra.wq,(Rb.ab),hint Ra< - PC; PC< - RbvaA AND {NOT 3} I JSR_COROUTINE 1A.3 Ra.wq,(Rb.ab),hint Ra< - PC; PC< - Rbv A AND {NOT 3} A LDA 08 Ra.wq,disp.ab(Rb.ab) Ra< - Rbv +R@ SEXT(disp)A LDAH 09 Ra.wq,disp.ab(Rb.ab) Ra< - Rbv +.F SEXT(disp*65536)C LDF 20 Fa.wf,disp.ab(Rb.ab) Fa< - ({Rbv + B SEXT(disp)}) I (continued on next page) I A-33 - 7 MACRO-64 Alpha AXP Architecture Quick Referencei, A.9 Common Architecture Instructions I Table_A-10_(Cont.)_Common_Architecture_Instructions______________( I Mnemonic________Code____Operands______________Operation___________ C LDG 21 Fa.wg,disp.ab(Rb.ab) Fa< - ({Rbv +NB SEXT(disp)})H LDL 28 Ra.wq,disp.ab(Rb.ab) Ra< - SEXT(({Rbv +I SEXT(disp)})<31:0>)SH LDL_L 2A Ra.wq,disp.ab(Rb.ab) Ra< - SEXT(({Rbv +I SEXT(disp)})<31:0>) C LDQ 29 Ra.wq,disp.ab(Rb.ab) Ra< - ({Rbv + B SEXT(disp)})C LDQ_L 2B Ra.wq,disp.ab(Rb.ab) Ra< - ({Rbv +rB SEXT(disp)})D LDQ_U 0B Ra.wq,disp.ab(Rb.ab) Ra< - ({{Rbv +I SEXT(disp)} AND NOT 9 7})rC LDS 22 Fa.ws,disp.ab(Rb.ab) Fa< - ({Rbv + B SEXT(disp)})C LDT 23 Fa.wt,disp.ab(Rb.ab) Fa< - ({Rbv + B SEXT(disp)})D MB 18.4000 Memory barrier@ MF_FPCR 17.025 Fa.rq,Fa.rq,Fa.wq Fa< - FPCRE MSKBL 12.02 Ra.rq,{Rb.rq Rc< - Byte maskGE /#b.iq},Rc.wq low from Rav at > Rbv<2:0>E MSKLH 12.62 Ra.rq,{Rb.rq Rc< - Long mask/F /#b.iq},Rc.wq high from Rav at> Rbv<2:0>E MSKLL 12.22 Ra.rq,{Rb.rq Rc< - Long mask_E /#b.iq},Rc.wq low from Rav at > Rbv<2:0>E MSKQH 12.72 Ra.rq,{Rb.rq Rc< - Quad mask F /#b.iq},Rc.wq high from Rav at> Rbv<2:0>E MSKQL 12.32 Ra.rq,{Rb.rq Rc< - Quad mask.E /#b.iq},Rc.wq low from Rav ato> Rbv<2:0>E MSKWH 12.52 Ra.rq,{Rb.rq Rc< - Word maskGF /#b.iq},Rc.wq high from Rav at> Rbv<2:0> I (continued on next page) A-34 I MACRO-64 Alpha AXP Architecture Quick Reference I A.9 Common Architecture Instructions/ I Table_A-10_(Cont.)_Common_Architecture_Instructions______________F I Mnemonic________Code____Operands______________Operation__________| E MSKWL 12.12 Ra.rq,{Rb.rq Rc< - Word mask E /#b.iq},Rc.wq low from Rav at > Rbv<2:0>@ MT_FPCR 17.024 Fa.rq,Fa.rq,Fa.wq FPCR< - FaE MULF 15.082 Fa.rf,Fb.rf,Fc.wf Fc< - Fav * Fbv MULF/C 15.002 MULF/S 15.482 MULF/SC 15.402 MULF/SU 15.582 MULF/SUC 15.502 MULF/U 15.182 MULF/UC 15.102E MULG 15.0A2 Fa.rg,Fb.rg,Fc.wg Fc< - Fav * Fbv MULG/C 15.022 MULG/S 15.4A2 MULG/SC 15.422 MULG/SU 15.5A2 MULG/SUC 15.522 MULG/U 15.1A2 MULG/UC 15.122G MULL 13.00 Ra.rl,{Rb.rl Rc< - SEXT((Rav * A /#b.ib},Rc.wq Rbv)<31:0>) MULL/V 13.40 E MULQ 13.20 Ra.rq,{Rb.rq Rc< - Rav * Rbv - /#b.ib},Rc.wq MULQ/V 13.60xE MULS 16.082 Fa.rs,Fb.rs,Fc.ws Fc< - Fav * FbvR MULS/C 16.002 MULS/D 16.0C2 MULS/M 16.042 MULS/SU 16.582 MULS/SUC 16.502 MULS/SUD 16.5C2 MULS/SUI 16.782 MULS/SUIC 16.702 MULS/SUID 16.7C2 MULS/SUIM 16.742 I (continued on next page) I A-35r c 7 MACRO-64 Alpha AXP Architecture Quick Reference , A.9 Common Architecture Instructions I Table_A-10_(Cont.)_Common_Architecture_Instructions______________ I Mnemonic________Code____Operands______________Operation__________ MULS/SUM 16.542 MULS/U 16.182 MULS/UC 16.102 MULS/UD 16.1C2 MULS/UM 16.142E MULT 16.0A2 Fa.rt,Fb.rt,Fc.wt Fc< - Fav * Fbv MULT/C 16.022 MULT/D 16.0E2 MULT/M 16.062 MULT/SU 16.5A2 MULT/SUC 16.522 MULT/SUD 16.5E2 MULT/SUI 16.7A2 MULT/SUIC 16.722 MULT/SUID 16.7E2 MULT/SUIM 16.762 MULT/SUM 16.562 MULT/U 16.1A2 MULT/UC 16.122 MULT/UD 16.1E2 MULT/UM 16.162G ORNOT 11.28 Ra.rq,{Rb.rq Rc< - Rav OR {NOT : /#b.ib},Rc.wq Rbv}F RC 18.E000 Ra.wq Ra< - intr_flag;D intr_flag< - 0I RET 1A.2 [Ra.wq],(Rb.ab),hint Ra< - PC; PC< - Rbv A AND {NOT 3}AI RPCC 18.C000 Ra.wq Ra< - Process cyclet= countersF RS 18.F000 Ra.wq Ra< - intr_flag;D intr_flag< - 1N S4ADDL 10.02 Ra.rl,{Rb.rl Rc< - SEXT(((SLL(Rav,2))C /#b.ib},Rc.wq + Rbv)<31:0>)5H S4ADDQ 10.22 Ra.rq,{Rb.rq Rc< - SLL(Rav,2) +9 /#b.ib},Rc.wq Rbv N S4SUBL 10.0B Ra.rl,{Rb.rl Rc< - SEXT(((SLL(Rav,2))C /#b.ib},Rc.wq - Rbv)<31:0>) I (continued on next page) A-36 I S I MACRO-64 Alpha AXP Architecture Quick Reference I A.9 Common Architecture InstructionsG I Table_A-10_(Cont.)_Common_Architecture_Instructions______________ I Mnemonic________Code____Operands______________Operation__________ H S4SUBQ 10.2B Ra.rq,{Rb.rq Rc< - SLL(Rav,2) -9 /#b.ib},Rc.wq RbvIN S8ADDL 10.12 Ra.rl,{Rb.rl Rc< - SEXT(((SLL(Rav,3))C /#b.ib},Rc.wq + Rbv)<31:0>) H S8ADDQ 10.32 Ra.rq,{Rb.rq Rc< - SLL(Rav,3) +9 /#b.ib},Rc.wq Rbv N S8SUBL 10.1B Ra.rl,{Rb.rl Rc< - SEXT(((SLL(Rav,3))C /#b.ib},Rc.wq - Rbv)<31:0>)AH S8SUBQ 10.3B Ra.rq,{Rb.rq Rc< - SLL(Rav,3) -9 /#b.ib},Rc.wq RbvtM SLL 12.39 Ra.rq,{Rb.rq Rc< - SLL(Rav,Rvb<5:0>)o- /#b.ib},Rc.wq M SRA 12.3C Ra.rb,{Rb.rq Rc< - SRA(Rav,Rvb<5:0>) - /#b.ib},Rc.wq M SRL 12.34 Ra.rq,{Rb.rq Rc< - SRL(Rav,Rvb<5:0>) - /#b.ib},Rc.wq = STF 24 Fa.rf,disp.ab(Rb.ab) ({Rbv + I SEXT(disp)})< - Fav = STG 25 Fa.rg,disp.ab(Rb.ab) ({Rbv + I SEXT(disp)})< - FavT] STL 2C Ra.rl,disp.ab(Rb.ab) ({Rbv + SEXT(disp)})<31:0>< - Rav<31:0> ] STL_C 2E Ra.ml,disp.ab(Rb.ab) ({Rbv + SEXT(disp)})<31:0>< - Rav<31:0> = STQ 2D Ra.rq,disp.ab(Rb.ab) ({Rbv +0I SEXT(disp)})< - Rav = STQ_C 2F Ra.mq,disp.ab(Rb.ab) ({Rbv + I SEXT(disp)})< - Rava> STQ_U 0F Ra.rq,disp.ab(Rb.ab) ({{Rbv +E SEXT(disp)} AND D NOT 7})< - Rav= STS 26 Fa.rs,disp.ab(Rb.ab) ({Rbv + I SEXT(disp)})< - Fav = STT 27 Fa.rt,disp.ab(Rb.ab) ({Rbv + I SEXT(disp)})< - Fav E SUBF 15.081 Fa.rf,Fb.rf,Fc.wf Fc< - Fav - Fbv SUBF/C 15.001 SUBF/S 15.481 SUBF/SC 15.401 I (continued on next page)m I A-37 X 7 MACRO-64 Alpha AXP Architecture Quick Reference , A.9 Common Architecture Instructions I Table_A-10_(Cont.)_Common_Architecture_Instructions______________< I Mnemonic________Code____Operands______________Operation__________a SUBF/SU 15.581 SUBF/SUC 15.501 SUBF/U 15.181 SUBF/UC 15.101E SUBG 15.0A1 Fa.rg,Fb.rg,Fc.wg Fc< - Fav - Fbvs SUBG/C 15.021 SUBG/S 15.4A1 SUBG/SC 15.421 SUBG/SU 15.5A1 SUBG/SUC 15.521 SUBG/U 15.1A1 SUBG/UC 15.121G SUBL 10.09 Ra.rl,{Rb.rl Rc< - SEXT((Rav -mA /#b.ib},Rc.wq Rbv)<31:0>)< SUBL/V 10.491E SUBQ 10.29 Ra.rq,{Rb.rq Rc< - Rav - Rbv - /#b.ib},Rc.wq SUBQ/V 10.69 E SUBS 16.081 Fa.rs,Fb.rs,Fc.ws Fc< - Fav - Fbvp SUBS/C 16.001 SUBS/D 16.0C1 SUBS/M 16.041 SUBS/SU 16.581 SUBS/SUC 16.501 SUBS/SUD 16.5C1 SUBS/SUI 16.781 SUBS/SUIC 16.701 SUBS/SUID 16.7C1 SUBS/SUIM 16.741 SUBS/SUM 16.541 SUBS/U 16.181 SUBS/UC 16.101 SUBS/UD 16.1C1 SUBS/UM 16.141E SUBT 16.0A1 Fa.rt,Fb.rt,Fc.wt Fc< - Fav - Fbv SUBT/C 16.021 SUBT/D 16.0E1 I (continued on next page) A-38 I MACRO-64 Alpha AXP Architecture Quick ReferenceEI A.9 Common Architecture Instructions I Table_A-10_(Cont.)_Common_Architecture_Instructions______________ I Mnemonic________Code____Operands______________Operation__________r SUBT/M 16.061 SUBT/SU 16.5A1 SUBT/SUC 16.521 SUBT/SUD 16.5E1 SUBT/SUI 16.7A1 SUBT/SUIC 16.721 SUBT/SUID 16.7E1 SUBT/SUIM 16.761 SUBT/SUM 16.561 SUBT/U 16.1A1 SUBT/UC 16.121 SUBT/UD 16.1E1 SUBT/UM 16.161G TRAPB 18.0000 Drain any pending ; traps,C UMULH 13.30 Ra.rq,{Rb.rq Rc< - {Rav *U B /#b.ib},Rc.wq Rbv}<127:64>B WMB 18.4400 Write memory= barrier G XOR 11.40 Ra.rq,{Rb.rq Rc< - Rav XOR Rbv - /#b.ib},Rc.wq I ZAP 12.30 Ra.rq,{Rb.rq Rc< - Zap from Rav,-H /#b.iq},Rc.wq byte mask Rbv<7:0>D ZAPNOT 12.31 Ra.rq,{Rb.rq Rc< - Zap fromH /#b.iq},Rc.wq Rav, byte mask NOTI ______________________________________________Rbv<7:0>____________ I A-39 I B I _________________________________________________________________ H Programming with MACRO-64 I This appendix contains information to help you program morebI effectively using the MACRO-64 assembly language within thesI framework of the OpenVMS Calling Standard. It discusses thet$ following information: - o The OpenVMS Calling Standard 5 o Accessing Memory with Base Registers ) o Types of Data Structures " o Types of Routines o Example Program $ o Self Addressability 5 o Optimization and Automatic Alignment E For more information on the AXP environment, you should 9 become familiar with the following manuals:Q 4 o Alpha Architecture Reference Manual ) o OpenVMS Calling Standard, ' o Alpha AXP architecture $ B.1 OpenVMS Calling Standard C All OpenVMS languages supplied by Digital support the I OpenVMS Calling Standard. With most compiled AXP languages,oE the OpenVMS Calling Standard is invisible to you as the>C programmer. However, when programming with the MACRO-sD 64 assembly language, you are responsible for definingC the appropriate data structures and using appropriate>G instruction sequences in order to implement routines thatb7 comply with the OpenVMS Calling Standard. I B-1 b ! Programming with MACRO-64 $ B.1 OpenVMS Calling Standard E Refer to Chapter 6 for a description of several library C macros that are supplied with the MACRO-64 Assembler.D B An important role of the OpenVMS Calling Standard isF support for debugging, exception handling, and tracebackG routines. These routines are aided by the OpenVMS Calling G Standard data structures and conventions to determine the0F current routine and the source of its call. In addition,E these structures and conventions enable the debugger to E interpret other information, such as data stored on the $ stack or in registers. G Again, when programming in MACRO-64 assembly language, itcF is important that you as the programmer correctly defineG the data structures and adhere to the conventions defined . by the OpenVMS Calling Standard. 0 B.2 Accessing Memory with Base Registers I The Alpha AXP architecture is a base-register architecture. F Memory locations can only be accessed relative to a base: address stored in one of the base registers. H The Alpha AXP architecture has 31 general, base registers,C R0 through R30. R31 is a special base register, which G always reads as zero. Any attempt to store a value in R31 is ignored.( F Since memory-reference instructions use a 16-bit, signedF displacement relative to the base address, elevated-modeC programs may access the low and high 32K bytes of the D virtual address space relative to R31. See Figure B-1. F However, these areas of virtual memory are typically not/ accessible to user-mode programs. ( Base Register Architecture B Base-register architectures use different methods toF obtain an initial base address for a routine. Typically,B base-register architectures load the current programB counter (PC) into a base register. Thus, data can beF accessed relative to a location in the code. This methodI is inconsistent with the high performance objectives of the ? Alpha AXP architecture for the following reasons: B-2c H Programming with MACRO-64I B.2 Accessing Memory with Base Registers/ G o It requires placing data areas adjacent to code areas, H which results in inefficient use of the instruction and data caches. C o The memory locations near the boundary of the data E area and code would likely be duplicated in both the G instruction cache and the data cache, thus diminishingc= the effectiveness of the caching algorithms. Procedure Value F The OpenVMS Calling Standard simplifies obtaining a baseE address without requiring the data and code areas to be G adjacent. When a program or routine needs to call another C routine, it must load R27 with a procedure value. The5D procedure value is a special base address which is theF address of the procedure descriptor of the routine it is calling. " Procedure Descriptor G The procedure value for a given routine is the address of F the procedure descriptor for that routine. Section B.3.1@ describes the procedure descriptor in more detail. F The procedure descriptor is a data block that resides inF a special data area owned by the routine. This data area) called the linkage section. D Unlike VAX VMS external routine names, which referenceC its entry point, AXP external routine names reference G the procedure descriptor which contains the entry point'sC address. A Since the called routine has the address of its ownrE procedure descriptor in R27, it can access not only itsUD own procedure descriptor, but more importantly, it canD use that address as a base address and access any dataF that is within 32K bytes on either side of its procedure descriptor. Linkage Section B The special data area where the procedure descriptorH resides is called the linkage section. The linkage sectionA is actually a program section (psect) that contains E the routine's procedure descriptor. The linkage section . performs the following function: I B-3 r R ! Programming with MACRO-64 0 B.2 Accessing Memory with Base Registers G o Allows other routines to link to the routine using the & procedure descriptor. E o Uses it to link to other areas of memory in order to ( receive and store data. E o Stores the addresses of other routines, allowing the 0 routine to call those routines. E Figure B-2 illustrates how you access memory using base addresses: H 1 The called routine can access memory in its own linkageB section relative to R27 since R27 contains a baseF address within its linkage section, specifically, the5 address of its procedure descriptor.h A 2 Since memory-reference instructions use a 16-bit_G displacement relative to the base address, the linkage6C section can be up to 64K bytes in size with direct ( access relative to R27. I In addition, the linkage section psect typically has thecH NOWRT attribute and contains only address constants andH possibly other constants in order to facilitate placingF the linkage section in a shared section in a sharable image. D 3 If the called routine needs to access memory beyondF its own linkage section, it can store additional baseF addresses in its linkage section, load them into baseG registers as needed, and thereby access any legitimateR$ location in memory. F 4 To access read/write data, a separate read/write dataE section is defined in a different psect with the WRT G attribute and the address of the read/write section is / stored in the linkage section.b E As shown in Figure B-2, the special base address passed ? in R27 provides the called routine with a seed of F addressability from which it can directly access its ownI linkage section and from which it can indirectly access all G of memory. Each routine in a call chain receives its seed 0 of addressability from its caller. B-4v ) H Programming with MACRO-64I B.2 Accessing Memory with Base Registers % MAIN Program Call Chain F The main routine in a program works in the same way. TheB operating system passes it a base address within itsD own linkage section in R27 just like any other callingC standard-conforming routine. The operating system can F establish its own seed of addressability in a variety ofG ways. The important programming consideration is that theaH main routine is given its requisite seed of addressabilityC via the base address passed in R27 when the operating E system calls the main routine. The main routine must in)G turn pass a seed of addressability in R27 to any routines8; that it calls, and so on down the call chain.R E Figure B-3 illustrates program start up and the program> call chain. F The call chain shown in Figure B-3 illustrates a typicalG program's call chain originating with the transfer to the,I program's entry point (MAIN) from the operating system. The#I MAIN routine calls routine A, which in turn calls B. A fullRF prologue and epilogue are required for routines MAIN andF A since they both make standard calls. The full prologueI allocates the fixed stack area and saves the return address H (R26), procedure value (R27), frame pointer (R29), and anyE preserved registers that are to be used by the routine. H Finally, the full prologue establishes the routine's frameE as the current frame. If MAIN or A need to access theiraG respective linkage sections after making a standard call, H they would save a preserved register in the prologue, moveG R27 to that register before using R27 in a standard call, G use that register to access the linkage section after theaF call, and restore that register in the epilogue. Since BI does not make any standard calls, its prologue and epilogue / can be significantly streamlined.{ ? Note that all three routines access data in theirdF respective linkage sections relative to the base addressF that is in R27 on entry. The caller refers to the calleeF by the address of the callee's procedure descriptor. TheG procedure-descriptor address is also called the procedure F value (PV), and it is this address that is passed to theF callee in R27. To make a call, the caller must store theE address of the callee's procedure descriptor in its own D linkage section. The caller can then load R27 from itsD own linkage section prior to the call. The caller then I B-5U C ! Programming with MACRO-6440 B.2 Accessing Memory with Base Registers E goes on to obtain the address of the callee's code from G the callee's procedure descriptor. With the callee's code G address (CA) in R26 and the callee's procedure-descriptor D address in R27, the caller can transfer control to theH callee with a jump-to-subroutine (JSR) instruction throughI R26, storing the return address in R26. A more optimal call F sequence using a linkage pair is shown in Section B.3.3. B.3 Data Structures1 F This section describes the OpenVMS Calling Standard dataH structures, and how to use them in practical applications. D There are three data structures defined by the OpenVMS@ Calling Standard that you should be familiar with: % o Procedure descriptor o Signature block o Linkage pair " B.3.1 Procedure Descriptor @ The procedure descriptor contains information thatB describes the routine. More importantly, it containsF the address of the code for the routine. It includes the$ following information: % o Stack usage (if any)0 , o Register save/restore masks # o Exception handlers @ Usually, a routine's name references the routine'sB procedure descriptor, rather than the address of itsF code. For example, a global name for an external routineC represents the address where the procedure descriptorUF resides, not the address of the routine's code. ProgramsH which call the routine obtain the address of its code from? its procedure descriptor or using a linkage pair. G The .PROCEDURE_DESCRIPTOR and .LOCAL_PROCEDURE_DESCRIPTORnC assembler directives define the name of the procedure A descriptor, and indicates that the block of storage > that follows is the procedure descriptor for the: specified routine. .PROCEDURE_DESCRIPTOR andA .LOCAL_PROCEDURE_DESCRIPTOR also associate the code_ B-6_ _ H Programming with MACRO-64I B.3 Data Structures6 F addresss of the routine's entry point with the procedureD name. The association between the code address and theG procedure name allows the assembler and linker to process F the .CODE_ADDRESS and .LINKAGE_PAIR directives. For more> information on linkage pairs, see Section B.3.3. @ You need to declare the correct amount of storage,9 and specify the correct initial values. The C .PROCEDURE_DESCRIPTOR and .LOCAL_PROCEDURE_DESCRIPTOR D directives do not create the storage for the procedureE descriptor. Instead, they mark the storage that follows4G as a procedure descriptor, resulting in a special object- G module record. Use the $ROUTINE and $PROCEDURE_DESCRIPTOR E macros to both mark the procedure descriptor and define G the storage for the procedure descriptor. For information D on these macros, see Chapter 6. For information on theB procedure descriptor format, see the OpenVMS Calling Standard._ B.3.2 Signature Block E The signature block describes the parameters, parameter3D passing methods, and return value used by the routine.C Using the signature block is optional. If it is used,_B the procedure descriptor references it. $ROUTINE andD $PROCEDURE_DESCRIPTOR macros can by used to define theC signature block. For information on these macros, see Chapter 6. B.3.3 Linkage Pair H You can make calls to routines more efficient by using theH .LINKAGE_PAIR and .LOCAL_LINKAGE directives. .LINKAGE_PAIRH and .LOCAL_LINKAGE store both the address of the specifiedE routine's code and procedure descriptor (a total of two quad-words). G For example, the assembler provides the .CODE_ADDRESS and H .LOCAL_CODE_ADDRESSS directives to obtain the address of aG routine's code (a quadword). To obtain the address of thefI routine's code and procedure descriptor, you can use either F the .CODE_ADDRESS or .LOCAL_CODE_ADDRESSS in conjunctionF with the .ADDRESS directive. The following example showsC how you would use these directives to obtain the code @ address and and procedure descriptor of routine X: I B-7u > ! Programming with MACRO-64 B.3 Data Structures S .CODE_ADDRESS X ; Store address of X's code (entry point)pU .ADDRESS X ; Store address of X's procedure descriptorn ? .LINKAGE_PAIR performs the same function as usingpB .CODE_ADDRESS and .ADDRESS together, as shown in the following example: ! .LINKAGE_PAIR Xi - Using Linkage Pairs in Routinesd G Another example of how you can use linkage pairs is shown F in Example B-1 and Example B-2. Example B-1 shows a code8 sequence that does not use a linkage pair: 7 Example B-1 Routine Without Linkage Pairsh ) .enable local_blocks + .psect $LINKAGE,noexen< LS: .procedure_descriptor MY_PROC, MY_CODE? ; MY_PROC procedure descriptor details...tV X_ADDR: .address X ; Store address of X's proc desc & .psect $CODE,exe- MY_CODE: ; R27 -> LS upon entry 7 ; Prologue omitted for clarity...gN LDQ R27,X_ADDR-LS(R27) ; R27 -> proc desc for XL LDQ R26,8(R27) ; R26 -> code for X 1L JSR R26,(R26) ; CALL X 27 ; Epilogue omitted for clarity... F While the previous code sequence works correctly, it has? two immediate stalls in the instruction sequence: G 1 The first stall occurs when the second LDQ instruction E must wait for the preceding load to R27 to complete. F 2 The second stall occurs when the JSR instruction must@ wait for the preceding load to R26 to complete. F Since routine calls occur frequently, these stalls wouldG significantly increase the amount of time for the routine to execute. B-8a m a H Programming with MACRO-64I B.3 Data Structurest H The following example is similar to Example B-1, except it4 uses a linkage pair to call routine X. 4 Example B-2 Routine With Linkage Pairs ) .enable local_blockd + .psect $LINKAGE,noexea< LS: .procedure_descriptor MY_PROC, MY_CODE= ;MYPROC procedure descriptor details... Q X_LP: .LINKAGE_PAIR X ; Store address of X's codeiR ; followed by address of X'sA ; proc desc & .psect $CODE,exe- MY_CODE: ; R27 -> LS upon entry 7 ; Prologue omitted for clarity...rI LDQ R26,<X_LP+0>-LS(R27) ; R26 -> code for XeN LDQ R27,<X_LP+8>-LS(R27) ; R27 -> proc desc for X> JSR R26,(R26) ; CALL X7 ; Epilogue omitted for clarity...a D Although the same number of instructions are required,F the two immediate stalls in the instruction sequence are eliminated.a C You can also use the $LINKAGE_PAIR macro or the $CALLhI macro. For more information on these macros, see Chapter 6.sI Note that the $CALL macro generates an instruction sequence 3 similar to that shown in Example B-2. B.4 Types of Routines2 A The OpenVMS Calling Standard defines four differenteF types of routines with four different kinds of procedure descriptors. C The format varies slightly for each kind of procedure 9 descriptor. The four types of routines are:d o Null framec o Register frameS 1 o Stack frame (fixed and variable)e o Bound frame I B-9 p ! Programming with MACRO-64s B.4 Types of Routinesd D A null frame routine is a light-weight routine similarF in concept to a JSB-linkage routine on VAX. A null-frameB routine executes in the context of its caller and isI therefore restricted in the operations that it can perform. F As shown in Table B-1, a null-frame routine requires the? least overhead and offers the least capabilities.c @ A register frame routine creates its own frame and@ therefore executes in its own context. A register-C frame routine is an intermediate-weight routine, with F intermediate capabilities and overhead. A register-frameH routine maintains the context of its callers in registers. A A stack frame routine has two categories: fixed andl variable.t F o With the fixed frame stack, the stack frame size doesF not vary. A fixed stack frame routine creates its ownI frame and stores the context of its caller on the stack.RD A fixed stack frame routine is also an intermediateB weight routine with intermediate capabilities andB overhead. The only significant capability lackingH with a fixed stack frame routine is the ability to make standard calls. D o With the variable frame stack, the stack frame sizeE may vary. A variable stack frame routine is the full F function, heavyweight type of routine. It creates itsH own stack frame and stores the context of its caller onG the stack. The variable stackframe routine is the only7> type of routine that can make standard calls. A A bound frame routine is generally used in compiledeF languages such as Ada and Pascal that require a dynamic,G up level link to data in an outer routine. Most assembly- I language programming does not involve bound frame routines.c " B.4.1 Routine Capabilities H The types of operations that a routine may perform and theG amount of required overhead in the routine's prologue andtH epilogue instruction sequences are different for each type of routine.t B-10 h H Programming with MACRO-64I B.4 Types of Routinesa G The capabilities and overhead requirements of null frame,iI register frame, fixed stack frame, and variable stack framen3 routines are summarized in Table B-1:c I Table_B-1_Frame_Attributes_______________________________________a 6 Variable+ FixedaF RegisterRegisterFixed VariableC Null Stack StackBI ______________________Frame___Frame___Frame___Frame___Frame______l @ Executes in Frame of Yes No No No No Caller C Where Caller's - RegisterRegisterstack Stacki Context is Kept A Can Allocate Stack - Yes Yes Yes Yes in Proloques A Must Allocate Stack - No No Yes Yes in Prologue A Can Allocate Stack RestrictNo Yes No Yes in Routine Bodyh A Can Have an No Yes Yes Yes Yes Exception Handler A Can Save/Restore RestrictNo No Yes Yesp Registersr A Can Make Standard RestrictNo No No YesyI Calls____________________________________________________________s G For more information on routines, see the OpenVMS Callingt Standard. 8 B.4.2 Entry Prologue and Exit Epilogue Sequences F In general, a standard sequence of instructions are usedD when a routine first starts and when it completes. TheC following code example shows a typical entry prologue D sequence for a variable stack-frame procedure. SeveralF elements of the prologue may be omitted for less complex routines.t I B-11 ! Programming with MACRO-64s B.4 Types of Routinesa L ENTRY: LDA SP, -32(SP) ; Allocate fixed stack area 1D STQ R27, (SP) ; Save procedure valueC STQ R26, 8(SP) ; Save return addressrU STQ R2, 16(SP) ; Save any preserved regs that are used ; STQ FP, 24(SP) ; Save old FPeG MOV SP, FP ; Establish current frame,O $END_PROLOGUE ; Marks the end of the proloque 2i C 1 The ENTRY label marks the code entry point for thef routine.e G 2 The $END_PROLOGUE macro is used to mark the end of thea proloque. F The following code example shows a typical exit epilogueD sequence for a variable stack frame procedure. SeveralE elements of the epilogue may be omitted for routines of less complexity. T $BEGIN_EPILOGUE ; Mark the beginning of the epilogue 1O MOV FP, SP ; Release the variable stack areadD LDQ R28, 8(FP) ; Fetch return addressK LDQ R2, 16(FP) ; Restore preserved registerspJ LDQ FP, 24(FP) ; Reestablish caller's frameL LDA SP, 32(SP) ; Release the fixed stack area@ RET R28 ; Return to callerN $END_EPILOGUE ; Mark the end of the epilogue 1 G 1 In this example, the $BEGIN_EPILOGUE and $END_EPILOGUE H macros are used to begin and end the epilogue sequence. I The $ROUTINE macro optionally generates a standard prologueeF instruction sequence at the beginning of the routine. InH addition, you can also use the $RETURN macro to generate aG standard epilogue sequence. For more information on theseC$ macros, see Chapter 6. B.5 Example Program E The following example shows a simple routine, ROUTINE1, & that does the following: = o Takes a single, longword parameter as input. B o Calls another function, ROUTINE2, passing its own6 parameter as a parameter to ROUTINE2. B-12 i ( H Programming with MACRO-64I B.5 Example Programi H o Adds the return value from ROUTINE2 and the value of anH external longword, I, storing the result in a register. H o Returns the result of the addition to its caller in R0. G $ROUTINE ROUTINE1, KIND=STACK, SAVED_REGS=<R2,FP>a& $LINKAGE_SECTIONM I_ADDR: .address I ; Linkage to external I # $CODE_SECTION.Y ; The prologue instruction sequence S ; is generated by $ROUTINE by ? ; default $ ; Routine bodyM .BASE R27,$LS ; Inform assembler that N ; R27 -> linkage section? LDQ R0, I_ADDR ; R0 -> Ir? LDL R2, (R0) ; R2 <- IaU MOV 1, R25 ; AI <- 1 (R16 already containseF ; the argument)X $CALL ROUTINE2 ; Call ROUTINE2, R27->linkage sectL ADDL R2, R0, R0 ; retval <- retval + IV $RETURN ; $RETURN generates the epilogueA ; sequence.r+ $END_ROUTINE ROUTINE1 , B.6 Establishing Self-Addressability H A MACRO-64 routine can establish addressability to its ownI linkage section without the help of its caller (the calling B routine). In this case, the caller does not pass theG procedure value in R27, as in a standard call. Therefore, H the routine must establish its own base address within its linkage section. E To accomplish self-addressability, the code and linkage I sections must be adjacent to each other. This requires them I to reside in the same psect. Because the assembler does not H allow placement of data in a EXE/NOMIX psect, instructionsE must be placed in an EXE/MIX psect. When debugging, the F debugger views execution to have transferred into a dataF area, because the debugger views all MIX psects as data. I B-13o e a ! Programming with MACRO-64d, B.6 Establishing Self-Addressability F ________________________ NOTE ________________________ : The debugger can correctly display decodedB instructions in a MIX psect, but it cannot achieveC source line /program-counter correlation and sourcet line scrolling.r F ______________________________________________________ I Example B-3 represents the CLEAR_X_ARRAY routine, which has & no procedure descriptor. 5 Example B-3 Establishing a Base Address < .psect CLEAR_X_ARRAY_CODE,exe,mix,quad X_ARRAY_ADDR:d& .address X_ARRAY X_ARRAY_SIZE:D7 .long <X_ARRAY_END - X_ARRAY> / 8 CLEAR_X_ARRAY::oA BSR R1,10$ ; R1 -> 10$kF 10$: LDQ R22,X_ARRAY_ADDR-10$(R1); R22 -> X_ARRAYH LDL R23,X_ARRAY_SIZE-10$(R1); R23 <- REMAININGJ 20$: CLR R31,(R22) ; [X_ARRAY_PTR] <- 0R LDA R22,8(R22) ; X_ARRAY_PTR<-X_ARRAY_PTR+8N LDA R23,-1(R23) ; REMAINING<-REMAINING-1O BGT R23,20$ ; Branch if REMAINING > 0h> RET (R26) ; Return .end E Other routines may call the CLEAR_X_ARRAY routine using B only its code address and a nonstandard linkage. TheB CLEAR_X_ARRAY routine executes in the context of itsF caller, it does not create its own frame. It establishesH a base address in its code section and because its linkageI section is adjacent to its code section, it can access data % in its linkage section.M B-14 c e H Programming with MACRO-64I B.7 Optimization and Automatic AlignmentX 0 B.7 Optimization and Automatic Alignment < MACRO-64 provides the following optimizations: ) o Automatic data alignment ) o Automatic code alignment o Scheduling o PeepholingR C All optimizations and automatic alignment are off, byiE default. When invoking MACRO-64, the default qualifiers / are /NOOPTIMIZE and /NOALIGNMENT.t H You can control optimization and automatic alignment using5 the following qualfiers and directives: 8 o The /OPTIMIZE and /ALIGNMENT qualifiers 4 o The .ENABLE and .DISABLE directives ; o The .BEGIN_EXACT and .END_EXACT directives F The .ENABLE and .DISABLE directives control optimizationD and automatic code alignment for an assembly unit; youE cannot use these directives to control optimization andcE automatic code alignment for a range of statements. Use H .BEGIN_EXACT and .END_EXACT to suppress optimization for a" range of statements. & B.7.1 Automatic Data Alignment D Access to naturally aligned data is faster than access@ to unaligned data. MACRO-64 does not align data byD default. Instead, it places data exactly as specified,F with no padding. With automatic data alignment selected,D data directives are automatically aligned on a naturalC boundary. Pad bytes are added as necessary to achievehD natural alignment. Labels that occur with the data are2 automatically aligned with the data. I B-15e o ! Programming with MACRO-64o0 B.7 Optimization and Automatic Alignment * B.7.1.1 Controlling Data Alignment E You can control when you want to have data alignment in.G your program. You do this by using the .ENABLE ALIGN_DATA E and .DISABLE ALIGN_DATA directives. Note that automaticnF data alignment is disabled by default. Example B-4 shows7 how to enable and disable data alignment. ? Example B-4 Enabling and Disabling Data Alignmento # .PSECT D1,DATA,NONEXE C .DISABLE ALIGN_DATA ; This is the default state 1N2 .BYTE 1 ; offset = 02 .QUAD 2 ; offset = 1" .PSECT D2,DATA,NOEXEE .ENABLE ALIGN_DATA ; Turn on automatic alignment 2 2 .BYTE 1 ; offset = 0Y .QUAD 2 ; offset = 8 (pad bytes are added at offsets 1 - 7)r G 1 The .DISABLE_ALIGN_DATA directive causes the assemblereG to allocate data at the current location with padding,cG regardless of whether it is within a natural boundary. F 2 The .ENABLE_ALIGN_DATA directive causes the assemblerB to align data objects on natural rather than byte boundaries. 7 B.7.1.2 Directives for Automatic Data Alignment D The following MACRO-64 data storage directives provide; optional, automatic alignment. See Table B-2.a I Table_B-2_Directives_Using_Automatic_Alignment_____________q I Directive_____________Alignment_Boundary___________________y , .WORD word (2) , .SIGNED_WORD word (2) 0 .LONG longword (4) 0 .ADDRESS quadword (8) 1 .OCTA octaword (16)d 0 .QUAD quadword (8) I (continued on next page)m B-16 S H Programming with MACRO-64I B.7 Optimization and Automatic Alignment I Table_B-2_(Cont.)_Directives_Using_Automatic_Alignment_____ I Directive_____________Alignment_Boundary___________________ 0 .CODE_ADDRESS quadword (8) 0 .LOCAL_CODE_ADDRESS quadword (8) 1 .LINKAGE_PAIR octaword (16)t 1 .LOCAL_LINKAGE_PAIR octaword (16) 0 .PROCEDURE_ quadword (8) DESCRIPTOR 0 .LOCAL_PROCEDURE_ quadword (8) DESCRIPTOR 0 .F_FLOATING (.FLOAT) longword (4) 0 .D_FLOATING quadword (8) (.DOUBLE)t 0 .G_FLOATING quadword (8) 0 .S_FLOATING longword (4) 0 .T_FLOATING quadword (8) 0 .BLKA quadword (8) 0 .BLKD quadword (8) 0 .BLKF longword (4) 0 .BLKG quadword (8) 0 .BLKL longword (4) 1 .BLKO octaword (16)h 0 .BLKQ quadword (8) 0 .BLKS longword (4) 0 .BLKT quadword (8) , .BLKW word (2) 0 .ASCID quadword (8) I .INSTRUCTION__________longword_(4)_________________________ H Specify a .PSECT directive alignment attribute that alignsF the psect on a boundary that is at least as stringent as3 the automatically aligned data items.t I B-17 t ! Programming with MACRO-64e0 B.7 Optimization and Automatic Alignment , B.7.2 Automatic Code Label Alignment I Automatic code label alignment improves I-cache utilizationcC and multi-instruction issue. This optimization aligns H certain code labels to a quadword boundary by padding withE NOP and FNOP instructions. Use the /ENVIRONMENT=NOFLOATr6 qualifier to suppress FNOP instructions. F Code labels that are the target of backward branches areB aligned. Also, labels that are the target of forwardE branches that are not also the target of a fall throughc are aligned. I Use the /ALIGNMENT=CODE qualifier or the .ENABLE ALIGN_CODE A directive to select automatic code label alignment. % B.7.3 Scheduling Optimizationa H Scheduling reorders instructions to reduce pipeline stallsD and increases the chances for multi-instruction issue.C It only reorders instructions; it does not change the E instructions or the registers used in the instructions. H In addition, instructions are not scheduled across labels.C Therefore, the structure of your program is retained.e F ________________________ Note ________________________ F If you use the same register for two or more logically? distinct actions, the scheduler schedules these D actions serially in the sequence you specify. If youA use different registers for two or more logicallysE distinct actions, the scheduler schedules the actions D in parallel, providing faster execution performance. F ______________________________________________________ A Use the /OPTIMIZE=SCHEDULE qualifier or the .ENABLEhC SCHEDULE directive to select scheduling optimization. # B.7.4 Peephole Optimization H Peephole optimization restructures your program for fasterH execution. To do so, it reduces the number of instructionsD executed and replaces long-executing instructions withE instructions that require less execution time. Although G this optimization may increase the number of instructionsiD in memory, the number actually executed, as you followE the code paths, decreases or remains the same. Peephole B-18 T H Programming with MACRO-64I B.7 Optimization and Automatic Alignmentl H optimization matches short code patterns and replaces themH with more optimal sequences. It also eliminates dead code.G (Note that dead code is often created by the code changeseI made by peephole optimization. This dead code is eliminatedeE by successive iterations where peephole optimization isaG selected.) Peephole optimization eliminates some branchesrH by inlining or replicating short sequences from the branch target.u A Use the /OPTIMIZE=PEEPHOLE qualifier or the .ENABLE A PEEPHOLE directive to select peephole optimization.m , B.7.5 Using MACRO-64 for Performance C High-order language compilers do optimizations calledlA loop unrolling and inlining. MACRO-64 allows you to A perform these optimizations manually through coding H techniques, although it does not offer specific qualifiers4 or directives for these optimizations. H Loop unrolling reduces the number of branches that must beI executed. You can unroll loops using the .REPEAT directive..@ The following example shows manual loop unrolling: 7 10$: ; Copy loopPH .repeat 10 ; Unroll the loop, factor=10G LDA R18, -1(R18) ; Decrement quad-word count2K BLT R18, 20$ ; Branch if quad-word count < 0P> LDQ R0, (R17) ; Load a quad-wordA STQ R0, (R16) ; Store the quad-word J LDA R16, 8(R16) ; Increment the target addressJ LDA R17, 8(R17) ; Increment the source address BR 10$ 20$: .endr C Loop unrolling complimented with MACRO-64 shcheduling C and peephole optimizations can significantly increaseeG performance. For better loop scheduling, you can make the < counter decrement and the loop exit separable. I B-19 ! Programming with MACRO-64.0 B.7 Optimization and Automatic Alignment 7 10$: ; Copy looptH .repeat 10 ; Unroll the loop, factor=10G LDA R18, -1(R18) ; Decrement quad-word countg> LDQ R0, (R17) ; Load a quad-wordK BLT R18, 20$ ; Branch if quad-word count < 0 A STQ R0, (R16) ; Store the quad-word J LDA R16, 8(R16) ; Increment the target addressJ LDA R17, 8(R17) ; Increment the source address BR 10$ 20$: .endrr F High oder language compilers often inline a routine callG if the called routine is small and is defined in the sameAG module from which it is called. To inline a routine call,gE the compiler replaces the routine call with the body of G the called routine. You can use macros to manually inline 3 short, local routines with MACRO-64.. * B.7.6 Viewing Optimization Results C You can view the results of optimization by comparing D code from before and after optimization. Use the /LISTD and /MACHINE_CODE command line qualifiers described inG Section 1.2.1. Using these qualifiers provides you with a0* binary machine code listing. B-20 I C I _________________________________________________________________ I Using LSE with MACRO-64 F This appendix explains how to use the Language-SensitiveA Editor (LSE) with the MACRO-64 language. For LSE toaD function correctly, LSE must be installed prior to the! MACRO-64 assembler. C.1 Invoking LSE B To use LSE with the MACRO-64 language, you must nameF your program source file with a .M64 file extension. ForI example, to edit a file named TEST.M64, enter the followingn command: $ LSE TEST.M64 F Using the .M64 file extension invokes the LSE editor and> places you in the MACRO-64 language environment. F If you choose to use a different file extension, you canE still use LSE in the MACRO-64 language environment. FornI example, to edit a file named TEST.ASM, enter the followings command: 2 $ LSE/LANGUAGE=MACRO64 TEST.ASM C.2 Running Diagnostics F You can run diagnostics in LSE to debug programs withoutC leaving the LSE editor. For more information, see the D Guide to Language-Sensitive Editor for VMS Systems and) the MACRO-64 release notes. H When running diagnostics in an LSE editing session, MACRO-F 64 displays error messages of different severity levels.E For more information on error messages, see Appendix E. I C-1s I I DuI __________________________________________________________________ I Differences from VAX MACROt F MACRO-64 is a new product that provides a macro assemblyD language for the native Alpha AXP instruction set. TheG MACRO-64 instruction set is completely different from theAH VAX MACRO instruction set. Although many of the directivesG are the same between VAX MACRO and MACRO-64, it would not I be practical to share instruction sequences between the two ! assembly languages. H While MACRO-64 is compatible with VAX MACRO, even allowingB shared source modules between VAX MACRO and MACRO-64G using common assembler directives and through conditional I assembly, several differences exist. This appendix contains A two sections describing the differences in assemblyG& directives and features: H o Section D.1 describes new features in MACRO-64 that are* not present in VAX MACRO. I o Section D.2 describes features in VAX MACRO that are noto% present in MACRO-64.d * D.1 Assembler Features in MACRO-64 E This section describes MACRO-64 features not present in VAX MACRO. o .BASE directive B The .BASE directive is provided for implicit base" register support. F o .PROCEDURE_DESCRIPTOR and .LOCAL_PROCEDURE_DESCRIPTOR directives I D-1w R " Differences from VAX MACRO* D.1 Assembler Features in MACRO-64 D These directives mark the specified identifier as aE procedure descriptor. .PROCEDURE_DESCRIPTOR declarestA an identifier as a global label that defines the ? address of the procedure's description (a data C address) and associate the procedure's entry point A (a code address) with the name of the procedure. I .LOCAL_PROCEDURE_DESCRIPTOR performs a similar function,oF defining a local label rather than a global label. NoD storage is allocated. For more information on these+ directives, see Chapter 5.N A o .LINKAGE_PAIR and .LOCAL_LINKAGE_PAIR directives C These directives cause a linkage pair to be storediE at the current psect and current location counter. ArH linkage pair consists of a code address and the addressE of the specified identifier. For more information one1 these directives, see Chapter 5.d A o .CODE_ADDRESS AND .LOCAL_CODE_ADDRESS directivesa C These directives cause the code address(es) of theeD specified identifier(s) to be placed at the currentF psect and current location counter. See Chapter 5 for" more information. D-2i a a I Differences from VAX MACROnI D.1 Assembler Features in MACRO-64h o .ELSE directive ? MACRO-64 introduces a new conditional-assemblyeE directive, .ELSE. It can be used in conjunction with F the .IF and .ENDC directives (common to VAX MACRO andD MACRO-64) to specify a sequence of statements to beD assembled when the condition in the controlling .IFI directive is false. While the .ELSE directive is similariG to VAX MACRO's .IF_FALSE directive, it differs in that I you may specify at most one .ELSE directive within a .IFbG /.ENDC block. .IF_FALSE is also provided for VAX MACROaC compatibility. See Chapter 5 for more information. # o .INCLUDE directiveA I MACRO-64 provides a new .INCLUDE directive that causes a I specified file to be textually included in the assembly. < The format of this directive is as follows: H .INCLUDE "file_specification" ["Directory_search_list"] H The effect of the .INCLUDE directive is similar to thatH of the #include directive in C or the REQUIRE statement in BLISS. 9 o Optional case sensitivity in identifierst H You can specify the /NAMES=AS_IS command-line qualifierH to cause the assembler to distinguish between uppercaseI and lowercase letters in identifiers and external symbollB names. The default is /NAMES=UPPERCASE; uppercaseE and lowercase letters are not distinguished, as with ? VAX MACRO. See Chapter 1 for more information.e C o Lexical operator support for 10 lexical operators. G MACRO-64 provides 10 lexical operators, three of which C are compatible with VAX MACRO's three macro stringyI operators. MACRO-64 lexical operators can be used in anyOI context, and are not restricted to macros. See Chapter 3 & for more information. ' o Lexical string symbols 4 See Chapter 3 for more information. - o Lexical escape operator (%%) 4 See Chapter 3 for more information. > o Lexical substitution operator (%lexsym_name%) I D-3n u a " Differences from VAX MACRO* D.1 Assembler Features in MACRO-64 4 See Chapter 3 for more information. : o More liberal use of value-of operator (\) F MACRO-64 as well as VAX MACRO recognizes the value-ofB operator (\) in macro invocations and in default-F value strings for parameters in .MACRO directives. InH addition, MACRO-64 recognizes the value-of operator (\)+ in the following contexts:R G - With either or both of the two arguments in the .IFO; DIFFERENT and .IF IDENTICAL directives. H - With the argument to the .IF BLANK and .IF NOT_BLANK directives. C - With the second argument to the .NCHR and .IRPC directives. H - With the second and subsequent arguments to the .IRP directive. F - With arguments to lexical operators. See Chapter 4) for more information. D-4 d I Differences from VAX MACROKI D.1 Assembler Features in MACRO-64( o MACRO64$ symbol H MACRO-64 implicitly defines a new symbol, MACRO64$, forD use in conditional assembly. This symbol is defined@ in all uppercase and all lowercase letters whenD /NAMES=AS_IS is specified on the command line. This1 symbol might be used as follows: + .MACRO ADDRESS32 EXPR_LISTt) .IF DEFINED MACRO64$ $ .LONG EXPR_LIST .IF_FALSE' .ADDRESS EXPR_LIST .ENDC .ENDM ADDRESS32" ADDRESS32 <A,B,C> ) o Automatic data alignment H .ENABLE_ALIGN_DATA causes data to be naturally aligned. See Chapter 5.e o Optimizations > The assembler optionally performs a number of@ optimizations. See Chapter 2 (command line) and% Chapter 5 (.ENABLE).e o $OPDEF macro B $OPDEF is supplied in place of VAX MACRO's .OPDEF directive. 1 o OpenVMS Calling Standard Macros:I 5 $ROUTINE $SEND_PROLOGUEe 7 $CALL $LINKAGE_SECTIONa 4 $CODE_SECTION $DATA_SECTION 6 $LINKAGE_PAIR $BEGIN_EPILOGUE 4 $RETURN $END_EPILOGUE $PROCEDURE_ DESCRIPTOR ' o .INSTRUCTION directiven H For more information on the .INSTRUCTION directive, see Chapter 5.r 7 o .BEGIN_EXACT and .END_EXACT directives I D-5 f " Differences from VAX MACRO* D.1 Assembler Features in MACRO-64 I For more information on these directives, see Chapter 5.n 6 D.2 VAX MACRO Features Not Present in MACRO-64 F This section contains features in VAX MACRO that are not" present in MACRO-64. & o EXE/NOEXE restriction D MACRO-64 allows only instructions to be placed in aF psect with either the EXE or MIX attributes, or both.G Data can be placed in a psect with either the NOEXE orlG MIX attributes, or both. VAX MACRO allows instructions @ and or data to be arbitrarily placed in psects. C o Setting the current location in an EXE/NOMIX psect I MACRO-64 does not permit the current location counter to I be changed in a psect with the EXE and NOMIX attributes. : VAX MACRO does not have this restriction. D-6o w I Differences from VAX MACRO-I D.2 VAX MACRO Features Not Present in MACRO-64g & o Radix unary operators G VAX MACRO allows unary radix operators on expressions.iF MACRO-64 allows unary radix operators only to specifyI the radix of numeric constants. The octal constant ^O123tE is valid with both VAX MACRO and MACRO-64. The octal F expression ^O(10 + 10) is valid with VAX MACRO; it is) not valid with MACRO-64.a + o The Register mask operatori H The VAX MACRO ^M operator is not supported on MACRO-64.F Because register masks are not needed by any MACRO-64F instructions or directives, there is no need for this operator. - o .ALIGN does not support PAGER H Because the Alpha architecture has variable page sizes,E MACRO-64 does not support PAGE as an argument to thel" .ALIGN directive. ? o No count specifiers on data storage directivesu E MACRO-64 does not allow a repeat count expression oncG any data storage directive such as .BYTE or .WORD. TheeG %REPEAT lexical function can be used for this purpose.l / o No support for .CROSS/.NOCROSSh F There is no support for these directives in MACRO-64. & o No support for .DEBUG D There is no support for this directive in MACRO-64.F However, even though this directive is not supported,E MACRO-64 provides symbolic information in the object 0 module for use by the debugger. & o No .DEFAULT directive > This directive in not applicable on MACRO-64. 5 o .ENABLE/.DISABLE directive arguments F The MACRO-64 assembler does not support the followingA arguments to the .ENABLE or .DISABLE directives: 7 - ABSOLUTE-Not applicable on MACRO-64h - DEBUGe - SUPPRESSIONt I D-7m " Differences from VAX MACRO6 D.2 VAX MACRO Features Not Present in MACRO-64 - TRACEBACK - TRUNCATION + o .END directive differencesi I VAX MACRO requires the .END directive. MACRO-64 does not G require this directive. MACRO-64 requires the optionaleC identifier that follows the .END directive to be am: previously declared procedure descriptor. / o .ENTRY directive not supported F Instead, a $ROUTINE macro that accomplishes a similar" function is used. % o No .GLOBAL directive8 B There is no support for this directive. VAX MACROI provided this directive for MACRO-11 compatibility only. D-8o g I Differences from VAX MACROsI D.2 VAX MACRO Features Not Present in MACRO-64a ) o No .H_FLOATING directive I Since this data type is not a native MACRO-64 type, thisa, directive is not supported. " o Changes to .IDENT F The argument to the .IDENT directive must be a quoted literal.e # o No .LINK directive 8 There is no support for this directive. # o No .MASK directiveD > This directive is not applicable to MACRO-64. $ o No .NTYPE directive G This directive is not supported by MACRO-64. The %TYPE > lexical operator provides a similar function. $ o No .OPDEF directive I There is no support for this directive. The $OPDEF macro_- provides a similar function. 0 o Differences in .PSECT directive H The following .PSECT arguments are not supported: PAGE, LIB, and USR. $ o No .REFn directives : There is no support for these directives. 2 o Differences in .RESTORE directive F On successive .RESTOREs for the same psect, VAX MACROC restores the current location counter value as the F last location counter value saved for the psect beingI restored. MACRO-64 restores the current location counter B value as the location counter value saved for the* matching .SAVE directive. ' o No .TRANSFER directivef 8 There is no support for this directive. 1 o Differences with .ENDM and .ENDRr I D-9 l n " Differences from VAX MACRO6 D.2 VAX MACRO Features Not Present in MACRO-64 F VAX MACRO considers the .ENDM and .ENDR directives toH be synonyms of each other. MACRO-64 considers .ENDM andE .ENDR to be distinct. You must end macro definitionsSF with .ENDM and you must end repeat ranges with .ENDR. D-10 s I EEI _________________________________________________________________ I Error Messages H This appendix describes the messages produced by the Alpha! MACRO-64 assembler. A The description of each message gives the severity,RC followed by additional explanatory text and suggested action. @ ADDTRUNC, Storing an address expression into a storageC allocation less than the size of an address will result in data truncation.A E Informational: The assembler has stored a value too largeCB for the allocated space. Data truncation has occurred. / User Action: Allocate more storage. A ALIGNFILLIGN, The optional .ALIGN fill pattern is beingwD ignored. It is only valid for psects that do not possess) the EXE and NOMIX attributes.s C Warning: The optional fill pattern specified is ignored E since it is only valid for psects that do not possess the % EXE and NOMIX attributes.n G User Action: Omit the fill pattern or specify the MIX psect attribute. E ALIGNFILLTRUNC, The value specified for the .ALIGN optionalRJ fill pattern should be an integer in the range of 0 . . . 255.H Data truncation will occur with the currently specified fill/ pattern in a byte storage location.s D Warning: The value specified as the fill pattern for theE .ALIGN directive should be in the range 0 . . . 255. DataiE truncation occurs when the specified value is out of this range. F User Action: Specify a smaller value for the fill pattern. I E-1 - Error Messages H ALIGNLABELIGN, The ALIGN_LABEL option has been replaced by the ALIGN_CODE option. B Error: The ALIGN_LABEL option has been replaced by the ALIGN_CODE option. 8 User Action: Use the recommended new option. > ALIGNTOBIG, Specified alignment too large for PSECT. G Error: The requested alignment is too large for the current psect. C User Action: Check the psect attributes and insure thatRC the psect alignment is greater than or equal to what isf requested. H ASCIITRUNC, ASCII constant contains too many characters; value truncated. ? Error: Your ASCII constant contains more than eightKG characters with the ^A or ^a radix specifier. The assemblerr) deletes the extra characters. B User Action: Check your source code. Use eight or less characters.e 5 BADALIGN, Alignment specifier out of range. B Error: The alignment specifier used with the .PSECT or- .ALIGN directive is out of range. E User Action: Consult the documentation description of the ' .PSECT or .ALIGN directive.o 4 BADENDARG, Bad argument to .END directive. A Error: The optional argument to the .END directive ise invalid. E User Action: The argument, if specified, must reference a C procedure descriptor within the module. Specify a valid ; procedure descriptor name or omit the argument. = BADINSARG, Argument N invalid for this instruction.h ? Error: The argument number shown is invalid for the instruction. B User Action: Check the argument and required format as+ specified in the documentation.. E-2 m I Error Messagest 3 BADLIB, Error opening library file XXXXX.c H Error: The assembler encountered an error when attempting to, open the indicated library file. D User Action: Check the file format and file protections. 8 BADMACPARAMNAME, Illegal macro parameter name. A Error: The indicated macro parameter name is illegal. H User Action: Examine your source code and consult Chapter 1. + BADMACRONAME, Illegal macro name.C 7 Error: The indicated macro name is illegal. F User Action: Check your source code and consult Chapter 2. 8 BADOPERAND, Invalid operand type for operator. ? Error: The resolved operand type is invalid for the_ specified operator.f B User Action: Consult the documentation descriptions of1 operators, operands, and expressions.t 0 BADPARAMSTR, Illegal parameter string. H Error: The string specified as a macro parameter is invalid. ? User Action: Check your source code and consult theh documentation. F BASEFAIL, Argument N invalid. The assembler failed to find aF base register specified with a previous .BASE directive to> form a register expression of the form offset(Rn). ? Error: The assembler failed to find a base registeroE specified with a previous .BASE directive to form a valid07 register expression of the form offset(Rn).s B User Action: Check the use of .BASE directives and the+ instruction in the source code. G BEGEXPSC, .BEGIN_EXACT is not valid in a psect with the NOEXE ! and NOMIX attributes.t H Error: A .BEGIN_EXACT directive is not valid in a psect with+ the NOEXE and NOMIX attributes. 0 User Action: Check your source code. I E-3t 3 r Error Messages F BYTEALIGNIGN, The BYTE_ALIGN option has been replaced by the ALIGN_DATA option. H Error: The BYTE_ALIGN option has been replaced by the ALIGN_ DATA option. 8 User Action: Use the recommended new option. 6 CONPSECTATTR, Contradictory PSECT attribute. H Error: A previously specified psect attribute conflicts with( the flagged psect attribute. E User Action: Consult the documentation description of thef2 .PSECT directive and psect attributes. C CONTEOF, End of file encountered after line continuation. A Error: The assembler encountered end of file when the ? previous line encountered specified a continuation.F 0 User Action: Check your source code. F DATAALIGNTOBIG, Data requires alignment too large for PSECT. F Error: The alignment required for a specified data item is$ too large for the psect. G User Action: Check the psect attributes and insure that thenD psect alignment is greater than or equal to the required( alignment of the data items. G DATANOTINNOEXE, Data declarations must be in a psect with the $ MIX, or NOEXE attribute. H Error: A data declaration, such as a data storage directive,> has been specified in a psect with incorrect psect attributes. B User Action: Make sure the psect has the MIX, or NOEXE attribute set. F DIRNOTINNOEXE, Directive must be in a psect with the MIX, or NOEXE attribute. F Error: The directive specified must appear in a psect with+ the MIX or NOEXE attribute set. > User Action: Check to insure that a psect with the6 appropriate attributes has been specified. E-4 I Error Messages F DISPTOOLGE, Branch offset is too large for this instruction. = Error: The specified offset is too large for thism instruction. D User Action: Check the range of the specified target. ItC must be in the inclusive range -1048576 . . . +1048575. 6 DUPLEXTERN, EXTERN item is multiply defined. ? Error: The item declared as externally defined withaC the .EXTERNAL attribute also has a conflicting multiple 1 definition within this assembly unit.a E User Action: Check the definitions of the specified item. 5 DUPLGLOBAL, Duplicate global name detected.O @ Warning: A duplicate global has been detected by the assembler. D User Action: Check all references in your source code to this name. : DUPMACPARAMNAME, Duplicate macro parameter name. > Error: The indicated macro parameter name has been duplicated. 0 User Action: Check your source code. I ENDEXPSC, .END_EXACT is not valid in a psect with the NOEXE andr NOMIX attributes._ F Error: A .END_EXACT directive is not valid in a psect with+ the NOEXE and NOMIX attributes. 0 User Action: Check your source code. ' EOLEXP, End of line expected. H Error: The assembler expected no more input from the current line.e 0 User Action: Check your source code. H ESCAPE, Illegal escape sequence in string literal; expected \, ", x, or X. H Error: The specified escape sequence specified in the string literal is illegal. 0 User Action: Check your source code. I E-5E o Error Messages J EXP32BITTRUNC, Expected an integer in the range 0 . . . (2^32)-1I for unsigned expression OR -(2^31) . . . +(2^31)-1 for signed 3 expression. Data truncation to 32 bits.o A Warning: Integer found was not in the expected range.h G User Action: Check your source code. The literal must be inMF the range 0 . . . (2^32)-1 for an unsigned expression OR -I (2^31) . . . +(2^31)-1 for signed expression. Data truncation " to 32 bits will occur. I EXP32BITTYPE, Expected an integer in the range 0 . . . (2^32)-1OI for unsigned expression OR -(2^31) . . . +(2^31)-1 for signedC expression.r B Error: Expected an unsigned integer value in the rangeE 0 . . . (2^32)-1 or a signed integer value in the range -t# (2^31) . . . +(2^31)-1. 0 User Action: Check your source code. H EXPBINEXPTERM, Found XXXXX when expecting a binary operator or" expression terminator. H Error: The assembler expected a binary operator, such as theE plus sign (+) for binary addition, or an item to end the 6 expression, such as the right bracket (>). H User Action: Check the flagged item in the source statement. A EXPFPREG, Argument N invalid. Expected a floating pointp register. H Error: The instruction argument cited is invalid. A floating' point register is expected. C User Action: Check your source code and the instruction documentation. E EXPGENREG, Argument N invalid. Expected a general register. G Error: The instruction argument cited is invalid. A generalB! register is expected. C User Action: Check your source code and the instruction documentation. E-6 I Error Messages ? EXPIDPROC, Argument N invalid. Expected an identifiere+ representing a procedure value. H Error: The argument cited is invalid. The assembler expectedA a user identifier which represents a procedure value.o C User Action: Check your source code and the instruction documentation. ? EXPINTPAL, Expected integer expression or PAL opcode. ; Error: Integer expession or PAL opcode missing. H User Action: Replace the flagged item with an integer or PAL opcode. E EXPLAB, Argument N invalid. Expected a label defined in the same psect. H Error: The cited argument is invalid. The assembler expected@ a label definition to occur in the same psect as its reference. C User Action: Check your source code and the instructionR documentation. D EXPLITVAL, Argument N invalid. Expected an integer literal5 value in the inclusive range 0 . . . 255.- A Error: The instruction argument cited is invalid. TherF assembler expected an integer literal in the range 0 . . . 255. C User Action: Check your source code and the instruction documentation. 4 EXPMACRONAME, Expected a valid macro name. C Error: A valid macro name was expected in this context. B User Action: Check your source code to insure that theC item flagged is a user identifier, opcode, or non-macroc directive. F EXPPALOPLIT, Argument N invalid. Expected an integer literal: value in the inclusive range 0 . . . 67108863. A Error: The instruction argument cited is invalid. TheX2 assembler expected an integer literal. C User Action: Check your source code and the instructiona documentation. I E-7 C g Error Messages D EXPREGOFF, Argument N invalid. Expected a general register. expression of the form offset(Rn). H Error: The cited argument is invalid. The assembler expectedH a general register expression of the form integer_offset(Rn) for this argument. B User Action: Check the source code and the instruction documentation. G EXPRESEXP, Argument N invalid. Expected an expression with no > forward references resolvable to psect +/- offset. H Error: The argument cited is invalid. The assembler expected5 an expression with no forward references.E C User Action: Check your source code and the instruction documentation. ? EXPSTACKOVER, Internal SEM expression stack overflow. 2 Fatal: An internal error has occurred. E User Action: Gather as much information as possible about H the circumstances under which the error occurred, and submit an SPR.o = EXPTOOCMPLX, Expression is too complex to evaluate.t E Error: The expression is too complex for the assembler toe evaluate.t E User Action: Try grouping the expression components usingfB angle brackets (< >). The most complex expression formA handled by the assembler resolves to the form: <psectfH /symbol +/- offset> OPERATOR <psect/symbol +/- offset> whereF OPERATOR is one of: +, -, *, /, @, \, &, or !. Consult theC documentation for further descriptions of the assemblerA& evaluation of expressions. D EXPZEROFF, Argument N invalid. Expected a general register1 expression of the form 0(Rn) or (Rn).A H Error: The cited argument is invalid. The assembler expected< a general register expression of the form 0(Rn). C User Action: Check your source code and the instruction documentation. E-8 I Error Messages D FOUNDEXP, Found XXXXX when expecting one of the following: XXXXX. C Error: An unexpected item was found in a location wherer( something else was expected. ? User Action: Check the unexpected item found in theRC source statement. Examine those items cited as expectede4 as alternatives for the unexpected item. $ GENERROR, Generated ERROR: A Error: This was generated using the .ERROR directive. 2 User Action: Examine your source code. $ GENPRINT, Generated PRINT: E Informational: This statement was generated with a .PRINTs directive. 2 User Action: Examine your source code. % GENWARN, Generated WARNING:: E Warning: This was generated using the .WARNING directive.i 2 User Action: Examine your source code. H HEXSTR, Illegal hexadecimal escape sequence in string literal. H Error: The specified hexadecimal escape sequence is invalid. F User Action: Check your source code and the documentation. F IDFOUND, Found identifier in the opcode field when expectingF one of the following: opcode, directive, macro invocation,! or symbol definition.c B Error: The identifier cited was unexpected. Rather, anE opcode, directive, macro invocation, or symbol definitiono was expected.c 0 User Action: Check your source code. = IDTOOLONG, Identifier is longer than 31 characters.p H Error: The identifier exceeds the 31 character maximum size. F User Action: Check your source code and rename or truncate the identifier. I E-9 Error Messages + ILLASCII, Illegal ASCII constant. E Error: The assembler found an illegal ASCII constant witho) the 6A or ^a radix specifier.G 0 User Action: Check your source code. * ILLBIN, Illegal binary constant. F Error: The assembler found an illegal binary constant with) the ^B or ^b radix specifier.g 0 User Action: Check your source code. + ILLDEC, Illegal decimal constant.n F Error: The assembler found an illegal binary constant with) the ^D or ^d radix specifier. 0 User Action: Check your source code. @ ILLEXPON, Illegal exponent in floating point constant. H Error: The specified exponent of the floating point constant is illegal. F User Action: Check your source code and the documentation. 4 ILLFLOAT, Illegal floating point constant. D Error: The specified floating point constant is illegal. F User Action: Check your source code and the documentation. / ILLHEX, Illegal hexadecimal constant.e F Error: The assembler found an illegal binary constant with) the ^X or ^x radix specifier.e 0 User Action: Check your source code. ( ILLIFOP, Illegal .IF operator. ? Error: An illegal operator was encountered as a .IFo operator.D E User Action: Check your source code and the documentationm" for the .IF directive. 7 ILLINCL, Illegal .INCLUDE file specification.f A Error: An illegal .INCLUDE directive was encountered.t H User Action: Check the use of the .INCLUDE directive by your source code. E-10 n i I Error Messagesn ) ILLOCT, Illegal octal constant.n F Error: The assembler found an illegal binary constant with) the ^O or ^o radix specifier.M 0 User Action: Check your source code. = ILLOPERANDMIX, Illegal operand mixing for operator.a C Error: The resolved operand types are invalid when used.1 together with the specified operator.a B User Action: Consult the documentation descriptions of1 operators, operands, and expressions.o @ ILLPROCRET, Illegal procedure return; linkage registerD (argument 1) must be R31 when software hint (argument 3) is 1. , Error: Illegal procedure return. G User Action: Check the instruction arguments. When argumenteE 3, software hint, is 1, the first argument specifying the ) linkage register must be R31. H ILLRADIX, Illegal radix specifier in numeric constant; specify A, B, C, D, O, or X. B Error: The assembler found an illegal radix specifier. G User Action: Check your source code and use one of A, B, C,n D, O, or X.r G INCLDEPTH, .INCLUDE nest depth exceeds N - check for circularo .INCLUDE. C Error: The maximum level of include file depth has been exceeded. A User Action: Check your source code for circular file inclusion. - INCLOPEN, .INCLUDE file open error. 5 Error: Included file could not be opened. ? User Action: Check the file attributes, etc. of the $ specified .INCLUDE file. I E-11v a t Error Messages H INSNOTINPSC, Instructions must be in a MIX, NOEXE; MIX,EXE; or NOMIX,EXE PSECT. D Error: An instruction has been specified in a psect with' incorrect psect attributes. F User Action: Make sure the psect has MIX, or EXE and NOMIX attributes set. C INTERNAL, Internal assembler error. Please submit an SPR. 2 Fatal: An internal error has occurred. E User Action: Gather as much information as possible aboutoH the circumstances under which the error occurred, and submit an SPR.A @ INTERR, Internal processing error in the SYN facility. 2 Fatal: An internal error has occurred. E User Action: Gather as much information as possible abouttH the circumstances under which the error occurred, and submit an SPR.n D INVALIGNFILL, Invalid optional fill pattern specified with .ALIGN directive.d 9 Error: An invalid fill pattern was specified. @ User Action: Check the second argument to the .ALIGN@ directive which is the alignment fill specifier. TheG argument must resolve to an integer. Check your source coder) and the .ALIGN documentation. = INVBASEEXP, Invalid expression for .BASE directive. C Error: The expression is not valid for .BASE directive.e E User Action: The expression specified for a base register > with the .BASE directive should contain no forwardH references and resolve to one of the following at this pointH in assembly: psect +/- offset, external symbol reference +/-H offset, integer, label +/- offset where the label is definedE in a psect with the EXE and NOMIX attributes. Consult thetB documentation for further description of the assembler& evaluation of expressions. E-12 E D I Error Messages I INVBASEREG, Invalid base register specified. Base register musth% be one of R0 through R30. 3 Error: Invalid base register specified.e F User Action: Specify a base register as a general registerJ from the range R0 . . . R30. R31 cannot be specified as a base? register and is implicitly defined as .BASE R31, 0.r C INVBRTGT, Invalid branch target. Branch target label musttD be defined in same psect as the branch instruction which! references the label. D Error: The specified label referenced as the target of aC branch instruction must be defined in the same psect in # which it is referenced.a G User Action: Consult the documentation for more informationc on labels. > INVCA, Invalid code address specified with procedureB descriptor. Code address must be a non-temporary labelF defined in a psect with the EXE or MIX attribute after its+ use with .PROCEDURE_DESCRIPTOR. G Error: The code address specified as the second argument tou; the .PROCEDURE_DESCRIPTOR directive is invalid.R G User Action: The code address must be a non-temporary labeldE defined in a psect with the EXE or NOMIX attribute. Check your source code.c @ INVEXP, Found XXXXX when expecting a valid expression. F Error: One of the following was expected by the assembler:C integer, floating point constant, identifier, register, : period (.), left bracket (<), unary operator. F User Action: Check the unexpected item found in the source statement. ? INVEXPRFORDIR, Invalid expression type for directive. E Error: The assembler resolved value for the expression ina+ the cited directive is invalid. > User Action: Consult the documentation for further= explanation of the directive arguments and types.l I E-13 Error Messages < INVEXPRFORSYM, Invalid expression type for symbol. H Error: The assembler resolved value for the expression which? is assigned to a local or global symbol is invalid.e F User Action: Expressions assigned to a symbol must containD no forward references and resolve to an integer or psectD /label +/- offset. Consult the documentation for furtherC descriptions of how the assembler determines symbol and expression values. H INVFPCONST, Invalid floating point value detected. Check value/ range for floating point data type.r E Error: An invalid floating point value has been detected.. D User Action: Check the specified range for the directive type.e I INVINSQUAL, Instruction qualifier list specified is invalid forr the opcode. B Error: The instruction qualifier(s) specified with the" opcode is/are invalid. F User Action: Consult the documentation for a complete list8 of opcodes and valid instruction qualifiers. G INVLCA, Invalid or undefined code address found for procedure descriptor.n F Error: An invalid or undefined code address corresponds to/ the specified procedure descriptor.u F User Action: Check your source code for the specified code address. F INVLISTOPT, Invalid option specified with the .LIST or .SHOW directive. F Error: An invalid option has been specified with the .LIST or .SHOW directive. F User Action: Consult the documentation for valid directive options. 9 INVLPD, Invalid procedure descriptor specified.v F Error: An invalid procedure descriptor has been specified.D There was no definition of a procedure descriptor by the specified name. 0 User Action: Check your source code. E-14 r M I Error Messages B INVNLISTOPT, Invalid option specified with the .NLIST or .NOSHOW directive. G Error: An invalid option has been specified with the .NLISTv" or .NO_SHOW directive. F User Action: Consult the documentation for valid directive options. G INVOFF, Attempting to specify data intialization with currentpF psect offset that is outside the range of 0 to 2147483647. G Error: The current psect offset is invalid for specifying a, data initialization. D User Action: Check your source code and the value of the! current psect offset.e I INVREPCOUNT, The integer value of the .REPEAT expression is notiH within the inclusive range of 0 . . . 65535. A zero value is assumed. G Warning: The value of the .REPEAT expression must be in theEJ inclusive range of 0 . . . 65535. Therefore, a zero expression value is assumed.e H User Action: Specify a repetition count between 0 and 65535. C INVSAVEOPT, Invalid option specified with the .SAVE_PSECT directive. G Error: An invalid option has been specified with the .SAVE_E PSECT directive. F User Action: Consult the documentation for valid directive options. 5 INVTEMPLAB, Invalid use of temporary label.e E Error: A temporary label reference is not allowed in this context. F User Action: Consult the documentation on use of temporary- labels for a further description.. B INVTERM, Found N when expecting a valid expression term. G Error: An unexpected item was found where an expression was F expected. It expected one of the following: floating pointI number, integer, register, decimal point (.), identifier, or left bracket (<). A User Action: Check the item flagged by the assembler.. I E-15 Error Messages 3 LABELNOTDEF, Undefined label encountered. 4 Error: The specified label is undefined. F User Action: Consult the documentation for descriptions of the valid labels. 5 LABELREDECL, Illegal redefinition of label.a C Error: The label is being illegally defined in multiples) places in this assembly unit. C User Action: Check all references to this label in your source code. ; LABNOTINPSECT, Label must be declared in a PSECT.e E Error: A temporary, local, or global label declaration isC9 being made and no psect has been established. B User Action: Make sure the a .PSECT directive has beenC issued prior to the appearance of the label declarationm! in the source stream.r 5 LEXOPEDITSPEC, Unrecognized edit specifier:P G Error: The edit specifier for the %EDIT lexical operator ise unrecognized.e ? User Action: Check your source code and consult theT/ documentation on lexical operators. G LEXOPENDM, Illegal modification of .ENDM directive keyword by lexical operation. ? Error: While your macro definition contains a .ENDMs? directive that ends the macro definition, the .ENDM-D directive is modified by a lexical operator such that itD can no longer be recognized as a .ENDM directive keyword% after lexical processing. D User Action: Change the statement to avoid modifying theE .ENDM directive keyword with lexical operator processing.h E-16 I Error Messageso G LEXOPENDR, Illegal modification of .ENDR directive keyword by lexical operation. E Error: While your repeat range contains a .ENDR directive G that ends the repeat block, the .ENDR directive is modified ? by a lexical operator such that it can no longer benA recognized as a .ENDR directive keyword after lexical processing.C D User Action: Change the statement to avoid modifying theE .ENDR directive keyword with lexical operator processing.h G LEXOPSYNTAX, Illegal lexical operator syntax (missing left orr4 right parenthesis, missing comma, etc.). E Error: The indicated lexical operator has a syntax error., G User Action: Check the source code to insure correct syntaxe) and placement of parenthesis. C LEXSYM, XXXXX is already a lexical string symbol name; it , cannot also be a numeric symbol. G Error: You cannot have a lexical and numeric symbol by this: same name. E User Action: Check your source code and remove either thee1 lexical or numeric symbol definition.l H LIBMOD_BADFORMAT, Library module XXXXX contains illegal syntaxE (missing .MACRO or label preceding .MACRO, missing or note9 matching .ENDM, or other macro syntax error).i 2 Error: Illegal syntax was encountered. 7 User Action: Check the syntax of the macro. 6 LIBMOD_EMPTY, Library module XXXXX is empty. G Warning: The assembler encountered an empty library module. 4 User Action: Replace the library module. E LIBMOD_EXTRA, Library module XXXXX contains extraneous text , after .ENDM; extra text ignored. G Warning: The assembler encountered extraneouse text after aaF .ENDM directive in a library module. This text is ignored. 4 User Action: Correct the library module. I E-17s t Error Messages ; LIBMOD_NOT_FOUND, Library module XXXXX not found.a > Error: The indicated library module was not found. D User Action: Check the spelling of the macro library and module names.e G LOCCTRNOTDATA, Location counter cannot be set in an EXE,NOMIX PSECT. E Error: The location counter cannot be modified in a psect . with the EXE and NOMIX attributes. D User Action: If you need to modify the location counter,F specify the MIX psect attribute. Consult the documentationB for a complete description of the MIX psect attribute. E MACCASEMATCH, Library macro name is spelled using differentg= alphabetic case than in .MCALL directive or macro invocation. F Error: There is an alphabetic case difference between thatE specified in the macro library and the alphabetic case ofC the macro name. A User Action: Check the case of the macro name in your H source code and the case of the macro in the specified macro library. C MACEXPNEST, Macro expansion exceeds maximum nesting depth + (macro recursion not detected). E Error: The macro is not recursive but exceeds the maximum & allowable expansion depth. < User Action: Check your source code for possible restructuring. C MACPARAMGENDEF, You may specify a generated label defaulto; value, or a default string value, but not both. G Error: A default string value AND a generated label default @ value have both been specified, when only one may be specified. 2 User Action: Examine your source code. E-18 s t I Error Messages E MACPARAMSYNTAX, Illegal macro parameter syntax: Found XXXXX ( when expecting one of XXXXX. 5 Error: Macro parameter syntax is invalid.: G User Action: Try replacing the unexpected argument with onet- of those items cited as expected.A E MACRECURSE, Recursive macro exceeds maximum macro expansiono nesting depth. A Error: The macro is recursive and exceeds the maximum $ expansion nesting depth. H User Action: Check your source code for a missing basis step# in the recursive macro.s I MACZERO, Cannot evaluate expression. A zero expression value isf assumed. H Informational: The assembler cannot evaluate this expressionA due to errors encountered. Therefore, a zero value iss assumed. E User Action: Check the expression for forward or externala references. < MAXIF, Maximum nesting of .IF directives exceeded. G Error: The maximum depth nesting of .IF directives has beene exceeded.a < User Action: Check your source code for possible restructuring. / MISSENDC, Missing .ENDC directive(s).A C Warning: The assembler did not find a terminating .ENDC." conditional directive. 0 User Action: Check your source code. / MISSINGENDM, Missing .ENDM directive.e E Error: No terinating .ENDM directive was found to match a. .MACRO directive. 0 User Action: Check your source code. / MISSINGENDR, Missing .ENDR directive.S H Error: The assembler expected to encounter a .ENDR directive1 to terminate a .IRP or .REPEAT block. 0 User Action: Check your source code. I E-19M E Error Messages E MISSQUOTE, Missing closing double-quote character in stringn literal. F Error: The closing double-quote is missing from the string literal. D User Action: Check your source code and insert a closing. double-quote for a string literal. E MODCODLOCCTR, Restoring the location counter in the current B context causes an illegal modification of the location, counter for an EXE, NOMIX psect. E Error: The location counter cannot be modified in a psectt. with the EXE and NOMIX attributes. D User Action: If you need to modify the location counter,F specify the MIX psect attribute. Consult the documentationB for a complete description of the MIX psect attribute. > NOBEGEX, Unmatched .END_EXACT directive encountered. D Error: A .END_EXACT directive was encountered prior to a# .BEGIN_EXACT directive.i 0 User Action: Check your source code. < NOCA, No code address specified as argument 2 with" .PROCEDURE_DESCRIPTOR. G Error: No code address was specified as the second argument 3 to the .PROCEDURE_DESCRIPTOR directive. G User Action: The code address must be a non-temporary label E defined in a psect with the EXE or NOMIX attribute. Check your source code.b H NOQUAL, Instruction qualifiers are not valid with this opcode. G Error: No instruction qualifiers may be specified with this opcode.s F User Action: Consult the documentation for a complete list8 of opcodes and valid instruction qualifiers. D NOTAQUAL, An item specified in the qualifier list is not a valid qualifier. B Error: The instruction qualifier(s) specified with the" opcode is/are invalid. F User Action: Consult the documentation for a complete list8 of opcodes and valid instruction qualifiers. E-20 e I Error Messages > NOTENOUGHARGS, Not enough arguments for instruction. C Error: The instruction needs one or more arguments thanb specified. H User Action: Check the argument numbers and required formats. as specified in the documentation. D NOTINMACRO, This statement must only occur within a macro. C Error: The statement specified is only allowed within a macro. E User Action: Consult the documentation description of thee statement specified. 7 NOTINSEM, Missing functionality in assembler.e 6 Error: Missing functionality in assembler. . User Action: Please submit an SPR. H NUMSYM, XXXXX is already a numeric symbol name; it cannot also' be a lexical string symbol.e G Error: You cannot have a lexical and numeric symbol by thist same name. E User Action: Check your source code and remove either then1 lexical or numeric symbol definition. 7 OPTIGN, VAX MACRO-32 option(s) being ignored.b F Informational: A VAX MACRO-32 option has been detected and is ignored. E User Action: Remove the VAX MACRO options from your Alphau MACRO-64 program.r B OVERLAP, Overlapping initializers detected at offset NN.D Multiple initial values are specified for the same PSECT offset.i C Error: You are trying to statically initialize the samer$ location more than once. 0 User Action: Check your source code. I E-21 Error Messages ? PSECTALIGNCON, PSECT alignment conflicts with earlier declaration. C Error: A previously specified psect alignment attribute 7 conflicts with the flagged psect attribute.d E User Action: Check all declarations of the psect. ConsultaE the documentation description of the .PSECT directive andt psect attributes.. > PSECTATTRCON, PSECT attribute conflicts with earlier declaration. H Error: A previously specified psect attribute conflicts with( the flagged psect attribute. E User Action: Check all declarations of the psect. ConsultrE the documentation description of the .PSECT directive and psect attributes.d F REDUNDELSE, You cannot specify more than one .ELSE directive& within a single .IF block. E Error: More than one .ELSE directive has been encounteredq& within a single .IF block. 0 User Action: Check your source code. 6 RESTOREWOSAVE, PSECT .RESTORE without .SAVE. B Error: A .RESTORE_PSECT directive was issued without a9 previous corresponding .SAVE_PSECT directive. H User Action: Check uses of .SAVE_PSECT and .RESTORE_PSECT in your source code. A SAVESTACKOVER, Internal SEM PSECT .SAVE stack overflow.k 2 Fatal: An internal error has occurred. E User Action: Gather as much information as possible about H the circumstances under which the error occurred, and submit an SPR.h - SRCREAD, Error reading source file. E Error: The assembler encountered an error in reading your source file. E User Action: Check file specifications, protections, etc.: E-22 r e I Error Messagesd 7 SYMBOLREDECL, Illegal redefinition of symbol.s > Error: The symbol is already defined as a label orH explicitly declared as externally defined with the .EXTERNAL directive. 5 User Action: Check all uses of this item. : TOOMANYARGS, Too many arguments for instruction. F Error: The instruction contains one or more arguments than necessary. H User Action: Check the argument numbers and required formats. as specified in the documentation. G TOOMANYMACARG, More arguments specified than defined for thise macro. G Error: More arguments have been specified on the macro call,3 than were specified for its definition.r H User Action: Check the macro definition and point of call in your source code.C 6 TOOMANYMACPARAMS, Too many macro parameters. ; Error: Too many macro parameters are specified.r 0 User Action: Check your source code. - TRUNCDATA, Data truncation warning.i E Warning: The specified data value is out of range for thex< specified directive. Data truncation will occur. 1 User Action: Specify a smaller value.. @ UNDCA, Undefined code address specified with procedure descriptor.v G Error: The code address specified as the second argument toE= the .PROCEDURE_DESCRIPTOR directive is undefined.n G User Action: The code address must be a non-temporary label E defined in a psect with the EXE or NOMIX attribute. Checkt your source code. I E-23e c n Error Messages A UNDEFSYM, Undefined symbol or label. Assuming .EXTERNALN definition.x D Warning: The referenced label or symbol does not have anF explicit definition and an external definition is assumed. C User Action: Use the .EXTERNAL directive to declare thet symbol. 0 UNEXPELSE, Unexpected .ELSE directive. A Error: An unexpected .ELSE directive was encountered. E User Action: Check the use of the .ELSE directive in your C source code to insure proper positioning with a .IF ande .ENDC directive. 0 UNEXPENDC, Unexpected .ENDC directive. F Error: The assembler never encountered a terminating .ENDC; for a macro conditional directive, such as .IF.t 0 User Action: Check your source code. 0 UNEXPENDM, Unexpected .ENDM directive. @ Error: The assembler encountered an unexpected .ENDM directive. C User Action: Check your source code for matching .MACRO /.ENDM pairs. 0 UNEXPENDR, Unexpected .ENDR directive. A Error: An unexpected .ENDR directive was encountered.b D User Action: Check your source code for matching .REPEAT2 /.ENDR and .IRP/.ENDR directive pairs. : UNEXPIFF, Unexpected .IF_FALSE (.IFF) directive. D Error: The assembler encountered an unexpected .IF_FALSE directive. C User Action: Check your source code to insure that this 0 directive occurs within a .IF block. 9 UNEXPIFT, Unexpected .IF_TRUE (.IFT) directive. C Error: The assembler encountered an unexpected .IF_TRUEb directive. C User Action: Check your source code to insure that this 0 directive occurs within a .IF block. E-24 D I Error Messages A UNEXPIFTF, Unexpected .IF_TRUE_FALSE (.IFTF) directive.e D Error: The assembler encountered an unexpected .IF_TRUE_ FALSE directive. C User Action: Check your source code to insure that this 0 directive occurs within a .IF block. 2 UNEXPMEXIT, Unexpected .MEXIT directive. A Error: The assembler encountered an unexpected .MEXITo directive. 0 User Action: Check your source code. 0 UNKDIR, Unknown directive XXXXX found. 1 Error: An internal error has occured.f E User Action: Gather as much information as possible about G the circumstances under which the error occurred and please submit an SPR. D UNKENDISOPTION, Unknown .ENABLE/.DISABLE option specified. ? Error: The option specified for .ENABLE/.DISABLE is incorrect. D User Action: Check the option specified with the .ENABLE /.DISABLE directive. / UNKNOWNATTR, Unknown PSECT attribute. E Error: The specified psect attribute is not recognized bye the assembler. E User Action: Consult the documentation description of thel .PSECT directive. H UNTERMEX, N Unterminated .BEGIN_EXACT directive(s) detected in psect XXXXX. D Error: Unmatched .BEGIN_EXACT directive(s) occur for the indicated psect. 0 User Action: Check your source code. G VAXDIR, VAX MACRO-32 directive(s) or option(s) being ignored. 1 Continuing processing with next line.l F Informational: A VAX MACRO-32 directive or option has been% specified and is ignored.o H User Action: Remove the VAX MACRO directives from your Alpha MACRO-64 program.L I E-25m r c Error Messages G VMACELSE, You cannot specify .ELSE in the same .IF block with_G either .IF_FALSE (.IFF), .IF_TRUE (.IFT), or .IF_TRUE_FALSE (.IFTF). H Error: A .ELSE directive was encountered within the same .IFC block as a .IF_FALSE, .IF_TRUE, or .IF_FALSE directive.g E User Action: Check your source code and remove either the 3 .ELSE directive or the .IF_x directive. F WRONGMACID, Macro name in .ENDM does not match corresponding .MACRO.o G Error: The macro name specified as the optional argument toaF the .ENDM directive does not match the name specified with/ the corresponding .MACRO directive. H User Action: Check your souce code for matching .MACRO/.ENDM directive pairs. @ WRONGPL, The code address specified in the .PROCEDURE_D DESCRIPTOR directive must occur BEFORE its definition as$ a local or global label. G Error: The code address specified as the second argument to E the .PROCEDURE_DESCRIPTOR directive must occur BEFORE its G definition as a non-temporary label defined in a psect with ' the EXE or NOMIX attribute.i 0 User Action: Check your source code. E-26 w F _________________________________________________________________ F Index E A ASCII string storage directivec2 _______________________________ (cont'd)D Absolute expression, 2-23 string descriptor (.ASCID),0 Accessing memory, B-2 5-14B Address zero-terminated (.ASCIZ),0 base register (.BASE), 5-18 5-17> local code storage directive .ASCIZ directive, 5-17E (.LOCAL_CODE_ADDRESS), Assembly termination directive 7 5-83 (.END), 5-44 B storage directive (.CODE_ Assignment statement, 1-2,. ADDRESS), 5-28 2-30E .ADDRESS directive, 5-9, B-16 Automatic alignment directivest6 Addressing See Table B-2" controlling alignment, B-16 to linkage section, B-13rF Address storage directive B______________________________ B (.ADDRESS), 5-9 .BASE directive, 5-18, 6-9: .ALIGN directive, 5-10 Base register, B-2@ AND operator, 2-29 defining address, B-14> Argument Base register directive8 actual, 4-2 (.BASE), 5-18C determining length, 5-99 $BEGIN_EPILOGUE macro, 6-6, . formal, 4-2 6-13D in a macro, 4-2 .BEGIN_EXACT directive, 5-22= number of, 5-97 Binary operator, 2-28 = Arithmetic shift operator, .BLKA directive, B-17v= 2-29 .BLKD directive, B-17e= .ASCIC directive, 5-13 .BLKF directive, B-17e= .ASCID directive, 5-14, B-17 .BLKG directive, B-17u= .ASCII directive, 5-16 .BLKL directive, B-17r= ASCII string storage directive .BLKO directive, B-17x counted (.ASCIC), 5-13e string (.ASCII), 5-16 F Index-1 a D .BLKQ directive, B-17 $DATA_SECTION macro, 6-23? .BLKS directive, B-17 Date and time lexical 7 .BLKT directive, B-17 %TIME, 3-21E3 .BLKW directive, B-17 Delimiter @ .BLKx directive, 5-24 string argument, 4-5F Block storage allocation Direct assignment statement,6 directive (.BLKx), 5-24 1-2, 2-30? .BYTE directive, 5-27 Directives, 1-2, 5-1 = Byte storage directive (.BYTE) as operators, 2-4TD , 5-27 for auto alignment, B-16H general assembler, 1-2, 5-1,1 C______________________________ 5-4 @ $CALL macro, 6-8, 6-9, 6-10, macro, 1-2, 5-1, 5-7E 6-15 Disable assembler functionsrG instruction sequence, 6-11 directive (.DISABLE), 5-31dB Character set .DOUBLE directive, 5-29? in source statement, 2-5 Double-precision IEEE E Character string directive (.T_FLOATING),,2 determining length, 5-99 5-132@ $CODE$ symbol, 6-52, 6-53 $DP symbol, 6-4, 6-52D .CODE_ADDRESS directive, 5-28, .DSABL, DISABLE directive,1 B-17 5-31 @ $CODE_SECTION macro, 6-22 $DS symbol, 6-4, 6-52G Colon (:) .D_FLOATING directive, 5-29,h1 in label field, 2-3 B-17 " Conditional assembly blockI directive E______________________________t > .ELSE, 5-32 %EDIT lexical, 3-12D .ENDC, 5-46 Element extraction lexical: .IF, 5-57 %ELEMENT, 3-14A Created temporary label, 4-10 %ELEMENT lexical, 3-14U@ range, 2-16 .ELSE directive, 5-32I $CS symbol, 6-4, 6-51 .ENABL, ENABLE directive, 5-34eE Current location counter, 2-31 Enable assembler functions, 1 5-34r@ D______________________________ .ENDC directive, 5-46 B $DATA$ symbol, 6-52, 6-53 End conditional assemblyC Data Storage directive (.END), 5-46i? ALIGN_DATA, 5-34 .END directive, 5-44dH Data structures, B-6 End macro definition directive: linkage pair, B-7 (.ENDM), 5-47@ procedure descriptor, B-6 .ENDM directive, 5-47 signature block, B-7 Index-2e i t G .ENDR directive, 5-48 Floating-point constants (.D_a< $END_EPILOGUE macro, 6-6, 6-24 FLOATING), 5-29? .END_EXACT directive, 5-49 Floating-point number D $END_PROLOGUE macro, 6-5, 6-26 arithmetic, 5-125, 5-1328 $END_ROUTINE macro, 6-27 format, 2-10= Epilogue sequence, 6-41, B-5, .F_FLOATING, 5-53m= B-10, B-11 .G_FLOATING, 5-55CD beginning, 6-13 in source statement, 2-99 .ERROR directive, 5-50 storage, 5-29e? Error messages, E-1 storing, 5-53, 5-55a> .EVEN directive, 5-51 .S_FLOATING, 5-125> Exact Instruction block .T_FLOATING, 5-132@ beginning, 5-22 Floating-point storage6 ending, 5-49 directive= Exact Instruction block .D_FLOATING, 5-29o= directive (.BEGIN_EXACT), .F_FLOATING, 5-53a= 5-22 .G_FLOATING, 5-55s? Exclusive OR operator, 2-29 Formal argument, 4-2dG Expression, 2-22 .F_FLOATING, FLOAT directive,o1 absolute, 2-23 5-53 F evaluation of, 2-22 .F_FLOATING directive, B-17 external, 2-23I global, 2-22 G______________________________eB relocatable, 2-22, 2-32 Global expression, 2-22< .EXTERNAL, EXTRN directive, Global label, 2-3E 5-52 Global symbol, 2-15, 5-136f: External expression, 2-23 defining, 5-34G External symbol, 5-136 .G_FLOATING directive, 5-55,y1 attribute directive B-17 (.EXTERNAL), 5-52+ defining, 5-34, 5-52 I I %EXTRACT lexical, 3-15 _______________________________ A .IDENT directive, 5-56NB F______________________________ Identification directive; Fields (.IDENT), 5-56 > comment, 2-2, 2-5 .IF directive, 5-57? label, 2-2, 2-3 .IFx directive, 5-63eD operand, 2-2, 2-4 .IF_FALSE directive, 5-63C operator, 2-2, 2-4 .IF_TRUE directive, 5-63OI Floating-point arithmetic .IF_TRUE_FALSE directive, 5-63 ? directive .IIF directive, 5-67i .S_FLOATING, 5-125 .T_FLOATING, 5-132 I Index-3u t o A Immediate conditional assembly Lexical operators, 3-1 6 block directive (.IIF), list, 3-11; 5-67 processing, 3-1 = .INCLUDE directive, 5-69 syntax rules, 3-1 ? Inclusive OR operator, 2-29 Lexical string symbol:G Indefinite repeat argument with quoted string literal,u1 directive (.IRP), 5-72 3-4eF Indefinite repeat character Lexical string symbols, 3-4C directive (.IRPC), 5-75 .LIBRARY directive, 5-78aD Instruction directive, 5-70 $LINK$ symbol, 6-52, 6-53I .INSTRUCTION directive, 5-70, .LINKAGE_PAIR directive, 5-80,A1 B-17 B-17AD Instructions $LINKAGE_PAIR macro, 6-28G as operators, 2-4 $LINKAGE_SECTION macro, 6-30SF Integer Linking directive (.LINKAGE_8 in source statement, 2-9 PAIR), 5-80@ %INTEGER lexical, 3-16 .LIST directive, 5-82C Invoking assembler, 1-3 Listing control directive 8 .IRPC directive, 5-75 .IDENT, 5-567 .IRP directive, 5-72 .LIST, 5-82i9 .NLIST, 5-101y: K______________________________ .NOSHOW, 5-102C Keyword argument, 4-3 Listing directive (.SHOW, ; NOSHOW), 5-126TD L Listing table of contents,2 _______________________________ 5-131? Label Local address storagegD created temporary, 4-10 directive (.LOCAL_CODE_; global, 2-3 ADDRESS), 5-83f; local, 2-3 Local label, 2-3 9 temporary, 2-3 saving, 5-124,; user-defined temporary, Local label block 8 2-15, 4-10 ending, 5-34: Labeling directive starting, 5-34A .LOCAL_DESCRIPTOR, 5-85 Local linking directivemH .PROCEDURE_DESCRIPTOR, 5-109 (.LOCAL_LINKAGE_PAIR), 5-84= Length Determining lexical Local symbol, 2-14iH %LENGTH, 3-17 .LOCAL_CODE_ADDRESS directive,7 %LENGTH lexical, 3-17 5-83, B-17cH %%-Lexical escape operator, .LOCAL_LINKAGE_PAIR directive,7 3-5 5-84, B-17 E Lexical escape operator (%%), .LOCAL_PROCEDURE_DESCRIPTORsB 3-5 directive, 5-85, B-17 Index-4 B %LOCATE lexical, 3-18 Macro arguments (cont'd)D Location control directive SCRATCH_REGS, 6-11, 6-18> .ALIGN, 5-10 SET_ARG_INFO, 6-17A .BLKx, 5-24 SIGNATURE_BLOCK, 6-19g: Location counter SIZE, 6-4, 6-7D current, 2-31 STACK_RETURN_VALUE, 6-17B Location counter alignment STANDARD_PROLOGUE, 6-57 directive (.ODD), 5-105 string, 4-5 5 Location counter control TIE, 6-18rB directive (.EVEN), 5-51 USES_VAX_ARGLIST, 6-19> Logical AND operator Macro call directive= See AND operator (.LIBRARY), 5-92 ; Logical exclusive OR operator Macro calls, 4-1r= See Exclusive OR operator as operators, 2-4v8 nested, 5-89E Logical inclusive OR operator number of arguments, 5-97 H See Inclusive OR operator MACRO command qualifiers, 1-3@ .LONG directive, 5-87, B-16 Macro definition, 4-1> Longword storage directive default value, 4-38 (.LONG), 5-87 ending, 5-47= $LS symbol, 6-4, 6-10, 6-52 labeling in, 4-10 D Macro definition directive; M______________________________ (.MACRO), 5-88eB MACRO-64 Assembler for OpenVMS Macro deletion directive= AXP Systems (.MDELETE), 5-93tA introduction, 1-1 .MACRO directive, 5-88tH Macro arguments, 4-2 Macro exit directive (.MEXIT),1 actual, 4-2 5-94o? ARGS, 6-8, 6-16 Macro expansion, 4-1t9 ARGS parameters, 6-16 printing, 4-1e= concatenated, 4-9 terminating, 5-94mA delimited, 4-5, 4-8 Macro include directive = formal, 4-2 (.INCLUDE), 5-69tA FUNC_RETURN, 6-19 Macro library directive-= keyword, 4-3 (.LIBRARY), 5-78 ; KIND, 6-5, 6-6 Macro name, 2-13p6 LOCAL, 6-10, 6-16 Macros, 4-1A LS, 6-16 calling routines, 6-8 G NAMES, 6-12 $CALL instruction sequence,i1 NONSTANDARD, 6-20 6-8iC positional, 4-3 defining a routine, 6-6 @ Rls, 6-9, 6-15 ending routines, 6-2E RSA_END, 6-4 for calling routines, 6-2eE RSA_OFFSET, 6-4 in epilogue sections, 6-6 SAVED_REGS, 6-6s I Index-5 u G Macros (cont'd) Number of arguments directiveu: in prologue sections, 6-5, (.NARG), 5-97H 6-6 Number of characters directive: in source code, 6-9 (.NCHR), 5-99F MACRO-64 supplied, 6-1 Numeric complement operator,1 MACRO-64 supplied library, 2-27P8 6-1 Numeric symbolG making multiple calls, 6-11 with lexical string symbol, 1 nested, 4-6, 5-89 3-4n? passing numeric value to, Numeric symbols, 3-4 4-9_I recursive, 5-89 O______________________________o7 redefining, 5-89 Object module = supplied, 6-12 to 6-53 identifying, 5-56 9 switching psects, 6-3, 6-6 naming, 5-133 8 syntax rules, 6-12 title, 5-133G within routines, 6-2 Obtaining information lexicalt7 with same name as opcode, %TYPE, 3-22rG 5-89 .OCTA directive, 5-103, B-16rD .MCALL directive, 5-92 Octaword storage directive; .MDELETE directive, 5-93 (.OCTA), 5-103 @ Memory .ODD directive, 5-105: accessing, B-2 One's complement? Message display directive of expression, 2-27.0 .ERROR, 5-50 Opcode< .PRINT, 5-108 redefining, 5-89I Messages with the same name as a macroy5 errors, E-1 , 5-89S= Message warning display $OPDEF macro, 6-31a7 directive (.WARN), 5-135 Operand, 2-4 8 .MEXIT directive, 5-94 Operator, 2-4 5 AND, 2-29tB N______________________________ arithmetic shift, 2-29 8 .NARG directive, 5-97 binary, 2-28< .NCHR directive, 5-99 complement, 2-27> .NLIST directive, 5-101 exclusive OR, 2-29> .NOSHOW directive, 5-102 inclusive OR, 2-29D Number numeric complement, 2-27? See also Integer, Floating- radix control, 2-26 7 point number, and Packed unary, 2-25B> decimal string Operator field, 2-4 " in source statement, 2-8 Index-6u i I P______________________________ Q______________________________6 G .PACKED directive, 5-106 .QUAD directive, 5-118, B-16_D .PACKED macro, 6-37 Quadword storage directive; Packed storage directive (.QUAD), 5-118 ? (.PACKED), 5-106 Quoted string literal_G .PAGE directive, 5-107 with lexical string symbol, 1 Page ejection directive 3-4m (.PAGE), 5-107 I Period (.) R______________________________ G current location counter, Radix control operator, 2-26 B 2-31 Range extraction lexical: Permanent symbol, 2-11, 2-13 %EXTRACT, 3-15D Positional argument, 4-3 Register name, 2-11, 2-14G .PRINT directive, 5-108 Relocatable expression, 2-22 6 .PROCEDURE_DESCRIPTOR Repeat blockG directive, 5-109, B-17 argument substitution, 5-72TH $PROCEDURE_DESCRIPTOR macro, character substitution, 5-75H 6-40 terminating repetition, 5-94@ Program section Repeat block directive= attributes, 5-111 (.REPEAT), 5-119BC code, 6-2 .REPEAT directive, 5-119 @ data, 6-2 %REPEAT lexical, 3-19D defining, 5-111 Repeat range end directive: directive (.ENDR), 5-48A .PSECT, 5-111 .REPT directive, 5-119 B .RESTORE, 5-122 .RESTORE_PSECT, .RESTORE= .SAVE, 5-124 directive, 5-122 C linkage, 6-2 $RETURN macro, 6-6, 6-41 D name, 5-116 $ROUTINE macro, 6-5, 6-42D restoring context of, 5-122 qualifiers, 6-42 to 6-518 saving context of, 5-124 Routines, B-99 saving local label, 5-124 See Table B-1e= Prologues bound frame, B-10eG entry and exit, B-11 defining type by macro, 6-5 < Prologue sequence, B-5, B-10, null frame, B-10= B-11 performance, B-11 @ ending, 6-26 register frame, B-10= .PSECT directive, 5-111 stack frame, B-10 E $RSA_END symbol, 6-4, 6-522H $RSA_OFFSET symbol, 6-4, 6-5,1 6-52t I Index-7 e : Strings (cont'd)? S______________________________ extracting elementsi ? .SAVE_PSECT, .SAVE directive, (%ELEMENT), 3-14sH 5-124 extracting range (%EXTRACT),2 Shift operator, 2-29 3-15G .SHOW, .NOSHOW directive, locating lexical (%LOCATE),u2 5-126 3-18I Signed byte data directive obtaining information lexicalD< (.SIGNED_BYTE), 5-129 (%TYPE), 3-22C Signed word storage directive obtaining value lexical > (.SIGNED_WORD), 5-130 (%STRING), 3-20H .SIGNED_BYTE directive, 5-129 repeating lexical (%REPEAT),2 .SIGNED_WORD directive, 5-130, 3-19G B-16 Subconditional assembly block56 Single-precision IEEE directive; directive .IF_FALSE, 5-63F: .S_FLOATING, 5-125 .IF_TRUE, 5-63@ $SIZE symbol, 6-4, 6-5, 6-52 .IF_TRUE_FALSE, 5-63E $STACK_ARG_SIZE symbol, 6-4 .SUBTITLE, SBTTL directive, 2 Statements 5-131B character set, 2-5 Subtitle listing controlC comment, 2-5 directive (.SUBTITLE),42 continuation of, 2-2 5-1317 field, 2-1 Symbol, 2-11 A format, 2-1 defined by $CALL, 6-4 D label, 2-3 defined by $ROUTINE, 6-4F operand, 2-4 determining value of, 2-13A operator, 2-4 external, 5-52, 5-136i? types, 1-1 global, 2-15, 5-136oB String argument, 4-5 in operand field, 2-14C String editing lexical in operator field, 2-13d7 %EDIT, 3-12 local, 2-14 < %STRING lexical, 3-20 macro name, 2-13A String locating lexical permanent, 2-11, 2-13 E %LOCATE, 3-18 register name, 2-11, 2-14 ; String repeating lexical undefined, 5-34 D %REPEAT, 3-19 user-defined, 2-12, 2-14D Strings Symbol attribute directive; date/time lexical (%LEX), (.WEAK), 5-1366H 3-21 .S_FLOATING directive, 5-125,1 determining length (%LENGTH), B-176 3-17! editing lexical (%EDIT),- 3-12 Index-8, 5 G User-defined temporary label,r1 T______________________________ 2-15r7 Tab stops range, 2-16 " in source statement, 2-2I Temporary label, 2-3 V______________________________T B user-defined, 2-15 Value conversion lexical: Term in MACRO statement, 2-22 %INTEGER, 3-16A %TIME lexical, 3-21 Value obtaining lexical 9 .TITLE directive, 5-133 %STRING, 3-2050 Title listing control ValuesG directive (.TITLE), 5-133 converting (%INTEGER), 3-16n %TYPE lexical, 3-22 I .T_FLOATING directive, 5-132, W______________________________-A B-17 .WARN directive, 5-135 A U .WEAK directive, 5-136gG _______________________________ .WORD directive, 5-138, B-16 H Unary operator, 2-25 Word storage directive (.WORD)4 User-defined symbol, 2-12, , 5-138 2-14 I Index-9e
