KAP Fortran for OpenVMS Alpha
User Guide

April 1997

This document provides information about how to run and use KAP Fortran on OpenVMS Alpha systems.

Revision /Update Information This is a new document.

Operating System and Version: OpenVMS Alpha, Versions 7.1

Software Version: KAP Fortran, Version 3.1

Digital Equipment Corporation
Maynard, Massachusetts
_____________________________________________________________________ First Printing, December 1992 Revised, April 1997

Possession, use, or copying of the software described in this publication is authorized only pursuant to a valid written license from Digital or an authorized sublicensor.

Digital Equipment Corporation makes no representations that the use of its products in the manner described in this publication will not infringe on existing or future patent rights, nor do the descriptions contained in this publication imply the granting of licenses to make, use, or sell equipment or software in accordance with the description. © Digital Equipment Corporation 1996, 1997. All rights reserved. © Kuck & Associates, Inc. 1996, 1997. All rights reserved.

The following are trademarks of Digital Equipment Corporation: AlphaGeneration, DECthreads, Digital, VAX DOCUMENT, and the DIGITAL logo.

KAP is a trademark of Kuck & Associates, Inc.

All other trademarks and registered trademarks are the property of their respective holders.

This document is available on CD-ROM.

This document was produced using SDMLWEB.

Contents

Preface

1 Overview

2 How to Run KAP

2.1 General KAP Information

2.2 Installing KAP

2.3 Compiling a Program Using the kfort Driver

2.4 Kfort Qualifiers

2.4.1 /c, (/c)

2.4.2 /verbose, /verbose, (/noverbose)

2.5 KAP Command Qualifiers Determined by Compiler Qualifiers

2.6 Compiling a Program Using kapf

2.7 Using KAP Syntax

2.8 Using File Naming Conventions

2.9 Introducing the Five Minute KAP Guide

2.9.1 Optimizing Small Programs with KAP

2.9.2 Optimizing Large Programs with KAP

2.10 Improving and Customizing KAP Performance

2.11 General Optimization Tips

2.12 Using Additional Performance Improvement Techniques

2.13 Correcting KAP Problems

3 KAP Parallel Processing

3.1 Automatic Parallelization Using the kfort Driver

3.1.1 Parallel Processing Options

3.2 Directed Parallelization Using the kfort Driver and PCF Directives

3.2.1 PCF Directive Syntax and Lexical Rules

3.2.2 PARALLEL REGION Directive

3.2.3 PARALLEL DO Directive

3.2.4 DO Loop Example with PCF Directives

3.2.5 Program Example with PCF Directives

3.2.6 CRITICAL SECTION Directive

3.2.7 ONE PROCESSOR SECTION Direcitve

3.2.8 Comparison of KAP PCF and Cray Autotasking Directives

3.3 Combined Automatic and Directed Parallelization Using the kfort Driver

3.4 Compiling a Program for Parallel Execution Using kapf

3.5 Running a Parallel Program

3.6 Parallel Run-Time Support Library Routines

3.7 Parallel Programming Tips

4 Command Qualifiers

4.1 General Optimization Qualifiers

4.1.1 /interchange, /nointerchange, (/interchange)

4.1.2 /namepartitioning, /namepart, /nnamepart, (/nonamepartitioning)

4.1.3 /optimize, /o, (/optimize=5)

4.1.4 /roundoff, /r, (/roundoff=3)

4.1.5 /scalaropt, /so, (/scalaropt=3)

4.1.6 /skip, /sk, /nsk, (/noskip)

4.1.7 /tune, /tune, (/tune=<architecture>)

4.2 Parallel Processing Qualifiers

4.2.1 /concurrentize, /conc, /noconc, (/noconcurrentize)

4.2.2 /minconcurrent, /mc, (/minconcurrent=1000)

4.2.3 /parallelio, /nopio, /pio, (/noparallelio)

4.2.4 /scheduling=<list>, /schd=<list>, (/scheduling=e)

4.3 Fortran Dialect Qualifiers

4.3.1 /assume, /a, (/assume=cel), /noassume, /na

4.3.2 /case, (/nocase), /ncase

4.3.3 /datasave, /ds, (/datasave), /nodatasave, /nds

4.3.4 /dlines, /dl, (/nodlines), /ndl

4.3.5 /escape, /noescape, (/escape)

4.3.6 /integer, /int, (/integer=4)

4.3.7 /intlog, (/intlog)

4.3.8 /kind, (/kind), (/kind=4)

4.3.9 /logical, /log, (/logical=4)

4.3.10 /natural, /nat, /nonatural

4.3.11 /onetrip, 1, (/noonetrip), /n1

4.3.12 /real, /rl, (/real=4)

4.3.13 /recursion, /rc, (/norecursion), /nrc

4.3.14 /save, /sv, (/save=manual_adjust)

4.3.15 /scan, (/scan=72)

4.3.16 /syntax, /sy, (off)

4.3.17 /type, /ty, (/notype), /nty

4.4 Inlining and Interprocedural Analysis Qualifiers

4.4.1 /inline, /inl, (off) /noinline, /ninl, /ipa, /ipa, (off), /noipa, /nipa

4.4.2 /inline_and_copy, /inlc, (off)

4.4.3 /inline_create, /incr, (off), /ipa_create, /ipacr, (off)

4.4.4 /inline_depth, /ind, (/inline_depth=2), /ipa_depth, /ipad, (/ipa_depth=2)

4.4.5 /inline_from_files, /inff, (current source file)

4.4.6 /inline_from_libraries, /infl, (off)

4.4.7 /ipa_from_files, /ipaff, (current source file)

4.4.8 /ipa_from_libraries, /ipafl, (off)

4.4.9 /inline_looplevel, /inll, (/inline_looplevel=2), /ipa_looplevel, /ipall, (/ipa_looplevel=2)

4.4.10 /inline_manual, /inm, (off), /ipa_manual, /ipam, (off)

4.4.11 /inline_optimize, (/inline_optimize=0), /ipa_optimize, (/ipa_optimize=0)

4.5 Advanced Optimization Control

4.5.1 /aggressive, /ag, (/noaggressive), /nag

4.5.2 /arclimit, /arclm, (/arclimit=5000)

4.5.3 /cacheline, /chl, (/cacheline=32,32)

4.5.4 /cache_prefetch_line_count, /cplc, (/cplc=0)

4.5.5 /cachesize, /chs, (/cachesize=8,0)

4.5.6 /dpregisters, /dpr, (/dpregisters=32)

4.5.7 /each_invariant_if_growth, /eiifg, (/eiifg=20)

4.5.8 /fpregisters, /fpr, (/fpregisters=32)

4.5.9 /fuse, /nfuse, (/nofuse)

4.5.10 /fuselevel, /fuselevel=1, (/fuselevel=0)

4.5.11 /heaplimit, /heap, (/heaplimit=100)

4.5.12 /hoist_loop_invariants, /hli, (/hoist_loop_invariants=1)

4.5.13 /interleave, /intl, (/interleave)

4.5.14 /library_calls, /lc, (off)

4.5.15 /limit, /lm, (/limit=10)

4.5.16 /machine, /ma, /noma, (machine=s)

4.5.17 /max_invariant_if_growth, /miifg, (/miifg=500)

4.5.18 /routine, /rt, /nrt, (/noroutine)

4.5.19 /setassociativity, /sasc, (/setassociativity=1)

4.5.20 /srlcd, /nsrlcd, (/nosrlcd)

4.5.21 /unroll, /ur, (unroll=4), /unroll2, /ur2, (/unroll2=160), /unroll3, /ur3, (/unroll3=20)

4.6 Directive Recognition Qualifiers

4.6.1 /directives, /dr, (/directives=akpv), /nodirectives, /ndr

4.6.2 /ignoreoptions, /ig, (/noignoreoptions), /nig

4.7 Input-Output Qualifiers

4.7.1 /cmp, (<file>.cmp.f), (<file>.cmp.f), /nocmp, /ncmp

4.7.2 /fortran, /f, (<file>.cmp.f), (<file>.cmp.f), /nofortran, /nf

4.7.3 /include, /inc, (/noinclude), ninc

4.7.4 /list, /l, (<file>.out), /nolist, /nl

4.8 Listing Qualifiers

4.8.1 /cmpoptions, /cp, /cmpoptions=n, (/nocmpoptions), /ncp

4.8.2 /lines, /ln, (/lines=55)

4.8.3 /listingwidth, /lw, (/listingwidth=132)

4.8.4 /listoptions, /lo, (/listoptions=o)

4.8.5 /suppress, /su, (off)

4.9 C*$* options

5 Directives

5.1 Directive Usage and Syntax

5.2 KAP Directives

5.3 General Optimization Directives

5.3.1 C*$* arclimit (0-5000)

5.3.2 C*$* beginblock <directive block> C*$* endblock

5.3.3 C*$* each_invariant_if_growth (0-100)

5.3.4 C*$* limit (> 0)

5.3.5 C*$* max_invariant_if_growth (0-1000)

5.3.6 C*$* optimize (0-5)

5.3.7 C*$* roundoff (0-3)

5.3.8 C*$* scalar optimize (0-3 )

5.3.9 C*$* unroll( <#it>[<weight>])

5.4 Parallel Processing Directives

5.4.1 C*$* [no]concurrentize

5.4.2 C*$* minconcurrent (0-999999)

5.5 Inlining and IPA

5.5.1 C*$* [no]inline [here|routine|global] [(name [,name...])]

5.5.2 C*$* [no]ipa [here|routine|global] [(name [,name...])]

5.6 Assertions

5.6.1 C*$* [no]assertions

5.7 Memory Management

5.7.1 C*$* padding (var-list)

5.7.2 C*$* storage order (var-list)

6 Assertions

6.1 KAP Assertions

6.2 Descriptions

6.2.1 C*$* assert [no]argument aliasing

6.2.2 C*$* assert [no]bounds violations

6.2.3 C*$* assert [no]equivalence hazard

6.2.4 C*$* assert [no]last value needed

6.2.5 C*$* assert permutation

6.2.6 C*$* assert no recurrence

6.2.7 C*$* assert relation ( <name> .XX. <variable/constant>)

6.2.8 C*$* assert no sync

6.2.9 C*$* assert [no] temporaries for constant arguments

6.3 Parallel Processing Assertions

6.3.1 C*$* assert concurrent call

6.3.2 C*$* assert do (concurrent)

6.3.3 C*$* assert do (concurrent call)

6.3.4 C*$* assert do (serial)

6.3.5 C*$* assert do prefer (concurrent)

6.3.6 C*$* assert do prefer (serial)

7 Inlining and IPA

7.1 Inlining and IPA Command Qualifiers

7.1.1 inline_from/ipa_from qualifiers

7.1.2 Library Creation

7.1.3 Naming Specific Routines

7.1.4 DO Loop Level

7.1.5 Recursive Inlining

7.1.6 Manual Control

7.2 Inlining and IPA Directives

7.3 Listing File Support

7.3.1 /Listoptions=c

7.4 Inlining/IPA Examples

7.4.1 Inlining Example - Same Source File

7.4.2 Inlining Example with a Library

7.4.3 IPA Example

7.4.4 Recursive Inlining Examples

7.4.5 Manual Inlining Example

7.4.6 Notes on Inlining and IPA

7.5 Conditions Inhibiting Inlining/IPA

8 Transformations

8.1 Memory Management

8.1.1 Command Qualifiers

8.1.2 Memory Management Tactics

8.2 Serial Optimizations

8.2.1 Dead-Code Elimination

8.2.2 Induction Variable Recognition

8.2.3 Global Forward Substitution

8.2.4 Loop Peeling

8.2.5 Lifetime Analysis

8.2.6 Invariant-IF Restructuring

8.2.7 Reciprocal Substitution

8.3 Scalar (Dusty-Deck) IF Transformations

8.3.1 IF to Block IF

8.3.2 IF to DO Loop

8.3.3 Semantic IF Merging

8.3.4 Zero-Trip IF Removal

8.4 Loop Unrolling

8.5 Loop Rerolling

9 KAP Listing File

9.1 Listing Qualifiers

9.1.1 Original Program Listing (O)

9.1.2 Calling Tree (C)

9.1.3 KAP Switches (K)

9.1.4 Loop Table (L)

9.1.5 Name (N)

9.1.6 Compilation Performance Statistics (P)

9.1.7 Summary Table (S)

9.1.8 Transformed Program Listing (T)

9.2 Listing Information

9.2.1 Line Numbers

9.2.2 DO Loop Markings

9.2.3 INCLUDE File Markings

9.2.4 Footnotes

9.2.5 Syntax Error/Warning Messages

9.2.6 Questions Generated by KAP

9.2.7 Action Summary

9.3 Loop Table Messages

9.4 KAP Listing Messages

A Data Dependence Analysis

A.1 Data Dependence Definitions

A.2 Varieties of Data Dependence

A.3 Input and Output Sets

A.4 Data Dependence Relations

A.5 Data Dependence Direction Vectors

A.6 Loop-Carried Dependence

A.7 Data Dependence Examples

B Listing File Messages

B.1 Classes of Messages

B.2 Messages

B.2.1 Data Dependence (DD)

B.2.2 Error (E)

B.2.3 Extension (EX)

B.2.4 Inlining/IPA (INL)

B.2.5 Informational (INF)

B.2.6 Inserted (I)

B.2.7 Loop Reordering (LR)

B.2.8 Warning (MIS)

B.2.9 Option Error (OW)

B.2.10 Not Optimized (NO)

B.2.11 Output Translation (OT)

B.2.12 Output Trans Fails (OTF)

B.2.13 Program Too Large (NO)

B.2.14 Question (Q)

B.2.15 Scalar Optimization (SO)

B.2.16 Standardized (STD)

B.2.17 Translator Error (TE)

B.2.18 Vector Enhanced (VE)

B.2.19 Warning (W)

C KAP and Incorrect Programs

Index

Tables

2-1 User Actions for Specific Goals

3-1 Comparison of KAP PCF and Cray Autotasking Directives

4-1 Command-Line Qualifiers

5-1 KAP Directives

6-1 KAP Assertions


(no previous page) | Next Page | Contents | Index |
Command-Line Qualifiers