Home

PGI 5 User's Guide - The Portland Group

1. display_error_number egCC only Display the error message number in any diagnostic messages that are generated 98 Chapter 3 The option may be used to determine the error number to be used when overriding the severity of a diagnostic message no_ lexceptions pgCC only Enable disable exception handling support The default is exceptions instantiation_dir lt dirname gt pgCC only Defines lt dirname gt as the instantiation directory The directory must exist This switch must appear on both the compile line and the link line The compiler will not delete objects from this directory no llalign pgCC only Do don t align long long integers on long long boundaries The default is llalign Generate a list of make dependencies and print them to stdout Compilation stops after the pre processing phase MD Generate a list of make dependencies and print them to the file lt file gt d where lt file gt is the name of the file under compilation O Command line Options 9 one_instantiation_per_object egCC only By default templates are instantiated in objects that reference them and multiple instantiations are avoided by calling pgprelnk to modify the template instantiation file an intermediate file with a ii extension This does not work well for template libraries where the ii files are not available to the archived o files As a result each file in a template library has to
2. ccceeeeseeteesteeeee 192 C calling Cisnreniseirei inisi 193 C calling Fortran eee 197 Index character case conventionsS 186 character return values 0 0 189 compatible data types eee 186 complex return values ceceeee 190 Fortran calling C o eeceeceesseeseereeeeees 191 Fortran calling C sasse 195 Fortran named COMMON s es 188 UNErSCOPES E E 186 Sb sea E EE E E 37 IPA building a program n se 39 building a program without IPA several SLEDS Fas elio o EE AEA 38 building a program without IPA single SUCP ieee PEE TA 37 building using the make utility 38 Language Options cseeceeseeseeeteeseeeees 72 Libraries BLAS iciniciratie dinni nna 171 BEDS E EE AR ee solic teeel 171 LAPACK 2 endian dee eas 171 LIBBE i icspcssctett Actes ae 171 shared object files ceeeeeseeseeeees 163 Dink 2 elaine echoes 15 Header Files sinusni ien aiin 15 Parallelization eccceseeseesteeteeseeenes 16 Listing Files 0 00 0 ceeeeeseeseeereeees 79 81 82 Loop unrolling cceceeseesseeseeseeereeeseees 24 Loops failed auto parallelization cee 34 SCALATS OAE ania ten mn 35 MINE ico sees Kiar nado Rone 34 Command line Options cscesseecees 84 OpenMP C C Pragmaas eeeeeees 129 OpenMP C C Support Routines omp_destroy_lock s es 143 omp get thread_num 0 142 omp_ get dynamic cece
3. e A cast can be used to select one out of a set of overloaded functions when taking the address of a function e Template friend declarations and definitions are permitted in class definitions and class template definitions e Type template parameters are permitted to have default arguments e Function templates may have nontype template parameters e Areference to const volatile cannot be bound to an rvalue e Qualification conversions such as conversion from T toT const const are allowed e Digraphs are recognized e Operator keywords e g and bitana etc are recognized e Static data member declarations can be used to declare member constants e wchar_t is recognized as a keyword and a distinct type e bool is recognized C Dialect Supported 259 260 RTTI runtime type identification including dynamic_cast and the typeid operator are implemented Declarations in tested conditions in if switch for and while statements are supported Array new and delete are implemented New style casts static _cast reinterpret cast and const_cast are implemented Definition of a nested class outside its enclosing class is allowed mutable is accepted on nonstatic data member declarations Namespaces are implemented including using declarations and directives Access declarations are broadened to match the corresponding using declarations Explicit instantiation of templates is implemented typename keyword is
4. byte 24 byte 28 Figure 9 2 Tail Padding in a Structure 9 2 6 Other Type Keywords in C and C The void data type is neither a scalar nor an aggregate You can use void or void as the return type of a function to indicate the function does not return a value or as a pointer to an unspecified data type respectively The const and volatile type qualifiers do not in themselves define data types but associate attributes with other types Use const to specify that an identifier is a constant and is not to be changed Use volatile to prevent optimization problems with data that can be changed from outside the program such as memory mapped I O buffers 182 Chapter 9 Chapter 10 Inter language Calling This chapter describes inter language calling conventions for C C and Fortran programs using the PGI compilers The following sections describe how to call a Fortran function or subroutine from a C or C program and how to call a C or C function from a Fortran program For information on calling assembly language programs refer to Appendix A Run time Environment 10 1 Overview of Calling Conventions This chapter includes information on the following topics e Functions and subroutines in Fortran C and C e Naming and case conversion conventions e Compatible data types e Argument passing and special return values e Arrays and Indexes e Win32 calling conventions Default Fortran calling conventions under Win32 differ fro
5. extension this option is not valid for pgcc or pgCC preprocessing Source files this option is preprocessed file i not valid for pgf77 pgf90 or pghpf S compilation Source files or preprocessed assembly language files file s c assembly Source files preprocessed unlinked object files or assembly language file o files none linking Source files preprocessed executable files files assembly language a out files object files or libraries If you specify multiple input files or do not specify an object filename the compiler uses the input filenames to derive corresponding default output filenames of the following form where filename is the input filename without its extension filename f indicates a preprocessed file if you compiled a Fortran file using the F option filename Ist indicates a listing file from the M ist option filename o indicates an object file from the c option filename s indicates an assembly language file from the S option Note Unless you specify otherwise the destination directory for any output file is the current working directory If the file exists in the destination directory the compiler overwrites it The following example demonstrates the use of output filename extensions pgf90 c proto f protol F 12 Chapter 1 This produces the output files proto o and protol o both of which are binary object files Prior to compilation the file proto1 F is pre p
6. for i 0 i lt n itt cfi j aliltj blil jl In this version the compiler does not perform vectorization for the entire source file Another use of the pragma scoping mechanism turns an option on or off locally either for a specific procedure or for a specific loop The following example shows the use of a loop scoped pragma main float a 100 100 b 100 100 c 100 100 int time maxtime n i j maxtime 10 n 100 pragma loop novector for time 0 time lt maxtime timet t for j 0 j lt n j for i 0 i lt n i cfijJ 3 alil 3 blil jl Loop level scoping does not apply to nested loops That is the pragma only applies to the following loop In this example the pragma turns off vector transformations for the top level loop If the outer loop were a timing loop this would be a practical use for a loop scoped pragma Optimization Directives and Pragmas 159 The following example shows routine pragma scope include math h funcl pragma routine novector float a 100 100 b 100 100 float c 100 100 d 100 100 intipa for i 0 i lt 100 i for j 0 j lt 100 j alil F alial j Di selia cll HI GHEIBB blig alil sl func2 float a 200 200 b 200 200 float c 200 200 d 200 200 INE iy for i 0 i lt 200 i for j 0 j lt 200 j afiJ j3 alia Abar Selil tJ cE eI DE daa lil When this source is c
7. in csh or ulimit s unlimited in bash sh or ksh 1 6 Using the PGI Compilers on Win32 On Win32 the tools that ship with the PGI compilers include a full featured UNIX like command environment After installation you should have a PGI icon on your Win32 desktop Double left click on this icon to cause an instance of the BASH command shell to appear on your screen Working within BASH is very much like working within the sh or ksh shells on a UNIX system 16 Chapter 1 but in addition BASH has a command history feature similar to csh and several other unique features Shell programming is fully supported A complete BASH User s Guide is available in the PGI online manual set Select PGI Workstation under Start gt Programs and double left click on the documentation icon to see the online manual set You must have a web browser installed on your system in order to read the online manuals The BASH shell window is pre initialized for usage of the PGI compilers so there is no need to set environment variables or modify your command path when the command window comes up In addition to the PGI compiler commands referenced above within BASH you have access to over 100 common UNIX like commands and utilities including but not limited to the following vi emacs make tar untar gzip gunzip ftp sed grep egrep fgrep awk cat cksum cp date diff du find kill Is more less mv printenv env rm rmdir touch we If
8. provides a convenient mechanism for the distribution of loop iterations across the available threads in a parallel region Items to note about clauses are e Variables declared in a PRIVATE list are treated as private to each processor participating in parallel execution of the loop meaning that a separate copy of the variable exists on each processor e Variables declared in a FIRSTPRIVATE list are PRIVATE and in addition are initialized from the original object existing before the construct e Variables declared in a LASTPRIVATE list are PRIVATE and in addition the thread that executes the sequentially last iteration updates the version of the object that existed before the construct e The REDUCTION clause is as described in Section 5 2 PARALLEL END PARALLEL 118 Chapter 5 e The SCHEDULE clause is explained in the following section e IfORDERED code blocks are contained in the dynamic extent of the DO directive the ORDERED clause must be present For more information on ORD Section 5 12 ORDERED ER ED code blocks see The DO END DO directive pair directs the compiler to distribute the iterative DO loop immediately following the SomP Do directive across the threads available to the program The Do loop is executed in parallel by the team that was started by an enclosing parallel region If the SOMP END DO directive is not specified the S OMP DO
9. w 1 0 100 w wtw y w 1 Z y w if z gt 0 goto 100 C now w is just big enough that w 1 w 1 gt 1 print w end In this case where the variables are implicitly real 4 operations are performed on the floating point stack where optimization removed unnecessary loads and stores from memory The general case of copy propagation being performed follows this pattern a x 2 0 a Instead of storing x into a then loading a to perform the addition the value of x can be left on the floating point stack and added to 2 0 Thus memory accesses in some cases can be avoided leaving answers in the extended real format If copy propagation is disabled stores of all left hand sides will be performed automatically and reloaded when needed This will have the effect of rounding any results to their declared sizes For the above program w has a value of 1 8446744E 19 when executed using default extended precision If however Kieee is set the value becomes 1 6777216E 07 single precision This difference is due to the fact that Kieee disables copy propagation so all intermediate results are stored into memory then reloaded when needed Copy propagation is only disabled for floating point operations not integer With this particular example setting the pc switch will also adjust the result The switch Kieee also has the effect of making function calls to perform all transcendental operations Although the function still
10. Function which accept a variable number of arguments Certain C C functions can only be inlined into the file that contains their definition 110 Static functions Functions which call a static function Functions which reference a static variable Chapter 4 Chapter 5 Open MP Directives for Fortran The PGF77 and PGF90 Fortran compilers support the OpenMP Fortran Application Program Interface The OpenMP shared memory parallel programming model is defined by a collection of compiler directives library routines and environment variables that can be used to specify shared memory parallelism in Fortran programs The directives include a parallel region construct for writing coarse grain SPMD programs work sharing constructs which specify that DO loop iterations should be split among the available threads of execution and synchronization constructs The data environment is controlled using clauses on the directives or with additional directives Run time library routines are provided to query the parallel runtime environment for example to determine how many threads are participating in execution of a parallel region Finally environment variables are provided to control the execution behavior of parallel programs For more information on OpenMP see http www openmp org For an introduction to how to execute programs that use multiple processors along with some pointers to example code see Section 4 Parallel Programming Using th
11. Seax sesi and edi secx and edx EFLAGS Run time Environment The stack pointer holds the limit of the current stack frame which is the address of the stack s bottom most valid word At all times the stack pointer should point to a word aligned area The frame pointer holds a base address for the current stack frame Consequently a function has registers pointing to both ends of its frame Incoming arguments reside the previous frame referenced as positive offsets from ebp while local variables reside in the current frame referenced as negative offsets from ebp A function must preserve this register value for its caller Integral and pointer return values appear in eax A function that returns a struct or union value places the address of the result in eax Otherwise this is a scratch register These Jocal registers have no specified role in the standard calling sequence Functions must preserve their values for the caller Scratch registers have no specified role in the standard calling sequence Functions do not have to preserve their values for the caller Floating point return values appear on the top of the floating point register stack there is no difference in the representation of single or double precision values in floating point registers If the function does not return a floating point value then this register must be empty Sst 7 Floating point scratch registers have no specified role in the st
12. compilers Set precision for certain calculations Selects variations for compiler steps R lt directory gt Linux only Passed to the Linker Hard code lt directory gt into the search path for shared object files Creates a relocatable object file Interpret DOUBLE PRECISION variables as REAL Interpret REAL variables as DOUBLE PRECISION Specifies the name of the driver s startup file Stops after the compiling phase and saves the assembly language code in filename s Strips the symbol table information from the object file Linux only Passed to the linker Instructs the linker to generate a shared object file Display driver s configuration parameters after startup Do not print warning messages Print execution times for the various compilation steps Command line Options 47 Description Specify the type of the target processor tp p5 for Pentium processors tp p for Pentium Pro II III processors tp p7 for Pentium 4 processors tp k8 32 for Athlon64 Opteron processors running a 32 bit operating system tp k8 64 for Athlon64 Opteron processors running a 64 bit operating system and tp px for generic 32 bit X86 code generation Undefine a preprocessor macro Initializes the symbol table with symbol which is undefined for the linker An undefined symbol triggers loading of the first member of an archive library V release_number Displays the version messages and other information v
13. pragma omp section lt C C structured block executed by processor i gt pragma omp section lt C C structured block executed by processor j gt Clauses private list shared list default shared none firstprivate list lastprivate list reduction operator list copyin list if scalar_expression nowait The omp parallel sections pragma defines a non iterative work sharing construct without the need to define an enclosing parallel region Semantics are identical to a parallel region containing only an omp sections pragma and the associated structured block Optimization MP Pragmas for C and C 139 6 11 omp ordered The OpenMP ordered pragma uses the following syntax pragma omp ordered lt C C structured block gt The ordered pragma can appear only in the dynamic extent of a for or parallel for pragma that includes the ordered clause The structured code block appearing after the ordered pragma is executed by only one thread at a time and in the order of the loop iterations This sequentializes the ordered code block while allowing parallel execution of statements outside the code block The following additional restrictions apply to the ordered pragma e The ordered code block must be a structured block It is illegal to branch into or out of the block e A given iteration of a loop with a Do directive cannot execute the same ORDERED directive more than once and cannot execute more
14. Commas separate clauses within the directives but commas are not allowed between the directive name and the first clause Clauses on directives may be repeated as needed subject to the restrictions listed in the description of each clause The compiler option mp enables recognition of the parallelization directives The use of this option also implies Mreentrant local variables are placed on the stack and optimizations that may result in non reentrant code are disabled e g Mnoframe Miomutex critical sections are generated around Fortran I O statements Many of the directives are presented in pairs and must be used in pairs In the examples given with each section the routines omp_get_ num threads and omp get _thread_num are used refer to the section below for more information They return the number of threads currently in the team executing the parallel region and the thread number within the team respectively 5 2 PARALLEL END PARALLEL The OpenMP PARALLEL END PARALLEL directive is supported using the following syntax Syntax SOMP PARALLEL Clauses lt Fortran code executed in body of parallel region gt SOMP END PARALLEL Clauses PRIVATE list SHARED list DEFAULT PRIVATE SHARED NONE FIRSTPRIVATE list REDUCTION operator intrinsic list COPYIN list IF scalar logical expression This directive pair declares a region of parallel execution It directs the
15. Default For arguments that you do not specify the defaults are as follows nobackslash noiomutex nodclchk noonetrip nodefaultunit nosave nodlines nounixlogical dollar _ noupcase This section describes the Mpgflag options that affect C C language interpretations by the PGI C and C compilers These options are only valid to the pgcc and pgcc compilation drivers Syntax Masmkeyword 72 instructs the compiler to allow the asm keyword in C source files The syntax of the asm statement is as follows asm statement Where statement is a legal assembly language statement The quote marks are required Chapter 3 Mnoasmkeyword instructs the compiler not to allow the asm keyword in C source files If you use this option and your program includes the asm keyword unresolved references will be generated Mdollar char char specifies the character to which the compiler maps the dollar sign The PGCC compiler allows the dollar sign in names ANSI C does not allow the dollar sign in names Mfcon instructs the compiler to treat floating point constants as float data types instead of double data types This option can improve the performance of single precision code Mschar specifies signed char characters The compiler treats plain char declarations as signed char Msingle do not to convert float parameters to double parameters in non prototyped functions This option can result in faster code
16. In a parallel region of code there may be a sub region of code that should execute only on the master thread Instead of ending the parallel region before this subregion and then starting it up again after this subregion the MASTER END MASTER directive pair let you conveniently designate code that executes on the master thread and is skipped by the other threads There is no implied barrier on entry to or exit from a MASTER END MASTER section of code Nested master sections are ignored Branching into or out of a master section is not supported PROGRAM MASTER_USE INTEGER A 0 1 INTEGER omp_get_thread_num A 1 OMP PARALLEL A omp_get_thread_num omp_get_thread_num SOMP MASTER PRINT YOU SHOULD ONLY SEE THIS ONCE SOMP END MASTER SOMP END PARALLEL 116 Chapter 5 PRINT A 0 A 0 A 1 A 1 END 5 5 SINGLE END SINGLE The OpenMP SINGLE END SINGLE directive uses the following syntax SOMP SINGLE Clauses lt Fortran code in body of SINGLE processor section gt SOMP END SINGLE NOWAIT Clauses PRIVATE list FIRSTPRIVATE list In a parallel region of code there may be a sub region of code that will only execute correctly on a single thread Instead of ending the peal region before this subregion and then starting it up again after this subregion the SINGL END SINGLE directive pair lets you conveniently designate co
17. Level zero optimization specifies no optimization O0 At this level the compiler generates a basic block for each Fortran statement This level is useful for the initial execution of a program Performance will almost always be slowest using this optimization level Level zero is useful for debugging since there is a direct correlation between the Fortran program text and the code generated Level one optimization specifies local optimization O The compiler performs scheduling of basic blocks as well as register allocation This optimization level is a good choice when the code is very irregular that is it contains many short statements containing IF statements and the program does not contain loops DO or DO WHILE statements For certain types of code this optimization level may perform better than level two O2 although this case rarely occurs The PGI compilers perform many different types of local optimizations including but not limited to e Algebraic identity removal e Constant folding e Common subexpression elimination e Local register optimization e Peephole optimizations e Redundant load and store elimination 22 Chapter 2 e Strength reductions Level two optimization O2 or O specifies global optimization The fast option generally will specify global optimization however the fast switch will vary from release to release depending on a reasonable selection of switches for any one particular r
18. STCC IPA requires three phases e Collection Determine which data about each function is potentially useful for interprocedural optimizations and save this data in the generated o file e Propagation Process all the o files to propagate information across function boundaries This is analogous to the link step combining information from all object files The IPA propagation step generates output files with suffix ipa one for each o file containing the newly propagated information e Inheritance When program units are recompiled STCC IPA exploits the information saved in ipa files to improve or enable a large variety of optimizations The corresponding additional steps in the STCC compilation sequence required for IPA are 1 Collect data and save in a o file this is done during the compilation step if the Mipa command line option is present 2 Propagate information and save in a ipa file this is done during the link step if the Mipa command line option is present Optimization amp Parallelization 39 3 Inherit information from a ipa file this is done during the second and subsequent compilations if the Mipa command line option and relevant ipa files are present When invoked with Mipa the STCC compilers detect whether source files have changed between the collection and inherit steps They do this by always generating the o file and comparing the new o file with any existing o file If th
19. Table 9 2 Real Data Type Ranges Data Type Binary Range Decimal Range Digits of Precision REAL 1037 to 10 1007 to 1058 on 174 Chapter 9 Table 9 3 Scalar Type Alignment Type Is Aligned on a E 1 byte boundary 2 byte boundary Sye bounds INTEGE 8 byte boundary byte bounda REAL 8 byte boundary ES COMPLE 8 byte bounda 9 1 2 FORTRAN 77 Aggregate Data Type Extensions The PGF77 compiler supports de facto standard extensions to FORTRAN 77 that allow for aggregate data types An aggregate data type consists of one or more scalar data type objects You can declare the following aggregate data types array consists of one or more elements of a single data type placed in contiguous locations from first to last structure is a structure that can contain different data types The members are allocated in the order they appear in the definition but may not occupy contiguous locations union is a single location that can contain any of a specified set of scalar or aggregate data types A union can have only one value at a time The data type of the union member to which data is assigned determines the data type of the union after that assignment Fortran C and C Data Types 175 The alignment of an array a structure or union an aggregate affects how much space the object occupies and how efficiently the processor can address members Arrays use the alignment of their members Ar
20. The compilation drivers link in several libraries by default For more information about libraries refer to Chapter 8 Libraries 1 3 2 Output Files By default an executable output file produced by one of the PGI compilers is placed in the file a out a exe on Win32 As shown in the preceding section you can use the o option to specify the output file name If you use one of the options F Fortran only P C C only S or c the compiler produces a file containing the output of the last phase that completes for each input file as specified by the option supplied The output file will be a preprocessed source file an assembly language file or an unlinked object file respectively Similarly the E option does not produce a file but displays the preprocessed source file on the standard output Using any of these options the o option is valid only if you specify a single input file If no errors occur during processing you can use the files created by these options as input to a future invocation of any of the PGI compilation drivers The following table lists the stop after options and the output files that the compilers create when you use these options Getting Started 11 Table 1 1 Stop after Options Inputs and Outputs all Stop after Input Output preprocessing Source files must have F preprocessed file to extension for Fortran standard out preprocessing Source files must have F preprocessed file
21. Use the P option to halt the compilation process after preprocessing and write the preprocessed output to the file filename i where the input file is filename c or filename cc Use the suffix option with this option to save the intermediate file in a file with the specified suffix Default The compiler produces an executable file Usage In the following example the compiler produces the preprocessed file myprog i in the current directory pgCC P myprog cc Cross references C c E Mkeepasm o S pgCC only Automatically use and or create a precompiled header file If use_pch or create_pch manual PCH mode appears on the command line following this option its effect is erased pch_dir directoryname pgCC only The directory in which to search for and or create a precompiled header file This option may be used with automatic PCH mode pch or manual PCH mode create_pch or use_pch no_ pch_messages pgCC only Enable or disable the display of a message indicating that a precompiled header file was created or used in the current compilation Command line Options 101 preinclude lt filename gt egCC only Specifies the name of a file to be included at the beginning of the compilation This option can be used to set system dependent macros and types for example prelink_objects egCC only Creates the necessary template instantiations for template librari
22. def file contains the symbols exported by the DLL Generating a def file is not required when building a DLL but can be a useful debugging tool if the DLL does not contain the symbols that you expect it to contain You can also create your own def file containing the symbols you want to export to the DLL To use your def file add it to the link line and omit output def out implib lt file gt Passed to linker Generate an import library named lt file gt for the DLL A DLL s import library is the interface used when linking an executable that depends on routines in a DLL export all symbols Passed to linker Use this flag to export all global and weak defined symbols to the DLL Even with this flag some symbols are not exported see no default excludes no default excludes Passed to linker When export all symbols is used there are still some special symbols i e D1 1Main 12 that are not exported Use no default excludes to export these symbols to the DLL To use the PGI compilers to create an executable that links to the DLL form of the runtime use the compiler flag Mdll The executable built will be smaller than one built without Mdil the PGI 166 Chapter 8 runtime DLLs however must be available when the executable is run The Mdi flag must be used when an executable is linked against a DLL built by the PGI compilers Each PGI compiler can also create DLLs for Win32 The following example
23. enddo end Example 2 3 Vector operation using SSE instructions Assume the above program is compiled as follows pgf90 fast Minfo vadd f vector op 4 Loop unrolled 4 times loop 18 Loop unrolled 4 times Following is the result if the generated executable is run and timed on a standalone AMD Opteron 1 6 Ghz system bin time a out 1 000000 771 000 3618 000 6498 00 9999 00 7 72user 0 00system 0 07 72 elapsed 99 CPU Now recompile with SSE vectorization enabled pgf90 fast Mvect sse Minfo vadd f vector op 4 Unrolling inner loop 8 times Loop unrolled 7 times completely unrolled loop 18 Generating sse code for inner loop Generated prefetch instructions for 3 loads Note the informational message indicating that the loop has been vectorized and SSE instructions have been generated The second part of the informational message notes that prefetch instructions have been generated for 3 loads to minimize latency of transfers of data from main memory Optimization amp Parallelization 29 Executing again you should see results similar to the following bin time a out 1 000000 771 000 3618 00 6498 00 9999 0 7 76user 0 00system 0 07 76elapsed 99 CPU No performance improvement is realized Why As noted above vector SSE instructions are most efficient when operating on data that is cache aligned In the compilation above there is no guarantee that the starting addresses of vector dat
24. input output of Fortran unformatted data files instructs the compiler to check for internal consistency of the X86 floating point stack in the prologue of a function and after returning from a function or subroutine call Floating point stack corruption may occur in many ways one of which is Fortran code calling floating point functions as subroutines i e with the CALL statement If the PGI_ CONTINUE environment variable is set upon execution of a program compiled with Mchkfpstk the stack will be automatically cleaned up and execution will continue There is a performance penalty associated with the stack cleanup If PGI_CONTINUE is set to verbose the stack will be automatically cleaned up and execution will continue after printing of a warning message instructs the compiler to check for pointers that are de referenced while initialized to NULL pgf 90 and pghpf only instructs the compiler to check the stack for available space in the prologue of a function and before the start of a parallel region Prints a warning message and aborts the program gracefully if stack space is insufficient Useful when many local and private variables are declared in an OpenMP program Win32 only link with the DLL versions of the runtime libraries This flag is required when linking with any DLL built by any of The Portland Chapter 3 Group Compiler Technology compilers Minfo option option instructs the compiler to produ
25. main res lt lt c lt lt endl Example 10 8 C Main Program cpmain c Calling a C Function To compile this C function and C main program use the following commands pgcc c csub c pgCC cpmain C csub o Executing the resulting a out file should produce the following output main a 8 b 2 ptr c Oxbffffb94 func a 8 b 2 ptr c bffffb94 func res 4 main res 4 10 11 Example C Calling C include lt iostream gt extern C void cpfunc int numl int num2 int res 194 Chapter 10 cout lt lt func a lt lt numl lt lt b lt lt num2 lt lt ptr c lt lt res lt lt endl res numl num2 cout lt lt func res lt lt res lt lt endl Example 10 9 Simple C Function cpfunc c with Extern C extern void cpfunc int a int b int c include lt stdio h gt main int az Dre a 8 b 2 printf main a d b d ptr c x n a b amp c cpfunc a b amp c printf main res d n c Example 10 10 C Main Program cmain c Calling a C Function To compile this C function and C main program use the following commands pgcc c cmain c pgCC cmain o cpsub C Executing the resulting a out file should produce the following output main a 2 ptrc Oxbffffb94 2 ptr c bffffb94 o func a Aa BO O ll func res main res Note that you cannot use the extern c form of declaration for an object s member functions 10 1
26. scalars within loops often need to be privatized that is each execution thread will have its own independent copy of the scalar Problems can arise if a privatized scalar is accessed outside the loop For example consider the following loop for i 1 i lt N i if fxi gt 5 0 t x i VS tt The value of t may not be computed on the last iteration of the loop Normally if a scalar is assigned within a loop and used following the loop the PGI compilers save the last value of the scalar However if the loop is parallelized and the scalar is not assigned on every iteration it may be difficult without resorting to costly critical sections to determine on what iteration t is last assigned Analysis allows the compiler to determine that a scalar is assigned on each iteration and hence that the loop is safe to parallelize if the scalar is used later For example for i 1 i lt n i df xlr O08 ef t 2 0 else t 3 0 yfi t Vat where t is assigned on every iteration of the loop However there are cases where a scalar may be privatizable but if it is used after the loop it is unsafe to parallelize Examine this loop for i 1 i lt N i if x i gt 0 0 t x il yl t vi tt 36 Chapter 2 where each use of t within the loop is reached by a definition from the same iteration Here t is privatizable but the use of t outside the loop may yield incorrect results since t
27. ssseseseeseesesseeeessreeesee 67 Command line Options sssssesseeseeseee 9 45 i ROA RE 51 e ENE A E A EE 51 fno lali giie 99 no_ alternative tokens 96 no bool ennie 97 no_ exceptions sssssesseeseeseeeesseeees ee 99 no_ pch_messages esceceereee 101 no_Jusing Std eeeeereeeee 102 T lien aie ee ein eel 95 EE O ates tain een 96 ADS R E TEN T E 96 byteswapio s sssseessseseessesersessesresseseesee 51 E a E e a Ea 52 a E E E E iene send iene 52 Cfromt 2 1 cece EN E 97 Cfromt 3 0 oo eecceseeseeeeeeeeeseeeeeeeeees 97 ereite Pehini rnana eaeties 98 AD E E E tera dea 53 Cia Cf OP i seccccsecvecetscccsecessteeiesscesienecs 98 diag remark c cee eeseeseeeteeseeteeees 98 diaG SUPPIESS eeeeeseesteeseereeeeeees 98 diaG WAIN ee eeeeseeseeeeereeeeeees 98 display_error_numbet ce 98 SOIYTUMN aiii iin eeri 54 a E E E E EASE 54 EE E T E E T 55 6 E 55 flags ernn e R E E e 55 e E 55 ARIC peie r Re as 56 Bcc EEEE E EEE EEE 56 e E E EE E E E S 56 o7 Th bsrinn ennn i ER 56 A E E TN 57 E E T P E TA E Muute inane 57 12 214 andige 58 instantiation_ dit eee eeeeeteerteereees 99 KGS ES uaketettrseesi te A 59 ocr A ators Aare Kea 60 Sis stink Head oe Sea 60 M rre bine daieeieuntea denies 99 MaN Ognin iann 79 Masmkeyword ccccescesseesteeseeseeees 72 Mbacksl shiesontin anen eas 70 268 MBOUTGS nyereni TA 80 Mbyteswapio ss sssss
28. the PGI Fortran compilers Generally optimal set of flags for the target includes at least O Munroll Mnoframe on X86 and AMD64 targets Display valid driver options Generate position independent code Equivalent to fpic Includes debugging information in the object module Allow object files generated by g77 to be linked into PGI main programs Display driver help message Adds a directory to the search path for include files Passed to the linker Treat INTEGER variables as 2 bytes Treat INTEGER variables as 4 bytes Treat INTEGER variables as 8 bytes and use 64 bits for INTEGER 8 operations Kflag Requests special compilation semantics with regard to conformance to IEEE 754 Specifies a library directory llibrary Loads a library Selects variations for code generation and optimization Displays a link map on the standard output module Save search for module files in directory lt moduledir gt only valid for the PGF90 and PGHPF compilers Chapter 3 Description Interpret and process user inserted shared memory parallel programming directives see Chapters 10 and 11 Win32 only use the Microsoft linker and include files and link against the Microsoft Visual C libraries Microsoft Visual C must be installed in order to use this switch rather than the default crtdll dll Stops after the preprocessing phase and saves the preprocessed file in filename i only valid for the PGI C C
29. 1 b3 cfront3 0 p A b3 egCC only Enable compilation of C with cfront 3 0 compatibility This causes the compiler to accept language constructs that while not part of the C language definition are accepted by the AT amp T C Language System cfront release 3 0 This option also enables acceptance of anachronisms Default The compiler does not accept cfront language constructs that are not part of the C language definition Usage In the following example the compiler accepts cfront constructs 96 Chapter 3 pgCC b3 myprog cc Cross references cfront2 1 b cfront3 0 p A no_ bool pgCC only Enable or disable recognition of bool The default value is bool pgCC only Enable compilation of C with cfront 2 1 compatibility This causes the compiler to accept language constructs that while not part of the C language definition are accepted by the AT amp T C Language System cfront release 2 1 This option also enables acceptance of anachronisms Default The compiler does not accept cfront language constructs that are not part of the C language definition Usage In the following example the compiler accepts cfront constructs pgCC cfront_2 1 myprog cc Cross references b b3 cfront3 0 p A pgCC only Enable compilation of C with cfront 3 0 compatibility This causes the compiler to accept language constructs that while not part of th
30. 143 omp_get_max_threads cee 142 omp_ get nested eeceeseeeeeseeeees 143 omp get MUM _PIOCS cee eeeeeteeeeee 142 omp get num threads ee 141 omp _in parallel 142 omp_init_lock eere 143 omp_set_dynamic ceceeeeeeeeees 143 omp_ Set_lock eee eeceeseeseeeeeeteeeees 144 omp _set_nested ccceeeeseeeeeteeeees 143 omp_set_num_threads eee 142 omp_test_lOCK sesser 144 omp_unset_lock ee eeeeeeeeeeeees 144 OpenMP Directives SVMAK Hs nenni i einir 111 OpenMP Environment Variables MBPS TR Z nnno oE 128 145 272 OMP_DYNAMIC esenee 128 144 145 OMP NESTE D neeeden eiee icsse 128 145 OMP_NUM_ THREADS 128 144 OMP SCHEDULE ssessesseesessessreeseeees 128 OpenMP Fortran Directives 111 OpenMP Fortran Support Routines omp_destroy_lock seese 127 omp get dynamic s es 127 omp_ get _max_threads cee 126 omp get _nested cccesceseesteeteeees 127 omp get MUM procs 126 omp_ get num_threads 0 eee 125 omp_ get thread num cee 126 omp _in parallel 126 omp init lock eee eeeeeseeeteeseeeees 127 omp_set_dynamic cescesesseeseees 126 omp_ Set_lock 00 ceeeeeeeseeeteesteeteeees 127 omp_set_nested cceceseeteesteeteeees 127 omp_set_num_threads cee 126 omp_test_lock essees 127 omp_unset_lock eeeeseeteeeteeeee 127 OpenM
31. 3 277libs Use the g77libs option on the link line if you are linking g77 compiled program units into a pgf90 compiled main program using the pgf 90 driver When this option is present the pgf90 driver will search the necessary g77 support libraries to resolve references specific to g77 compiled program units The g77 compiler must be installed on the system on which linking occurs in order for this option to function correctly Default The compiler does not search g77 support libraries to resolve references at link time Usage The following command line requests that g77 support libraries be searched at link time pgf90 g77libs myprog f g77_object o Used with no other options he p displays options recognized by the driver on the standard output When used in combination with one or more additional options usage information for those options is displayed to standard output Usage In the following example usage information for Minline is printed to standard output stcc help Minline Minline lib lt extlib gt lt func gt name lt func gt size lt n gt levels lt n gt Enable function inlining lib lt extlib gt Use extracted functions from extlib lt func gt Inline function func name lt func gt Inline function func except lt func gt Do not inline function func size lt n gt Inline only functions smaller than n levels lt n gt Inline n levels of functions Minline Inline all functions that were
32. 5 16 Run time Library Routines eoe A EEE E E R E ee 125 5 17 Environment Variables ci rA A nian 128 OPENMP PRAGMAS FOR C AND C 129 6 1 Parallelization Pra gimas 2 sc c cccieseccesecceceececsec dace sane n iii i 129 G 2somp Parallel ities Ae ee ties ects eee hac eet E tate Sa da sck PED Slade E T 130 O 3 0Mp CritiCal a9 A iueel os wake dl hehe adie bale oe heeled file eee cc 132 GAO MASTET esse ceeds onia E E EEO EEEE E E ES E TEER ONE E 133 6 5 OMP SIN Ge aant E a a eet A E E E E E E ee 134 66 OMP 0 ele R ERE eee eee 135 EAA DEt EE EE EEE A A A E A 137 6 8 0mp parallel for nne kun n ache a E EE e S A R 138 69 Omp se OS e ees ce RI ee RAGA E era hod on wR 138 6 10 omp parallel sections c2 c ce ccccscceee eens di ceciestecedinseesacescedencacdeestevidecstieddotiadudeedtadscduteesecess 139 6 TAvomp Ordered A E E hess ve cd ces ooh E A Leenks hes 2th hs E tothe dg had sc aesbiahentesteds 140 O21 2 Omp atomiin edocs eee oeee a a a aus ME Aiea Sed sehevere de Baca es rects 140 vi Table of Contents O13 0mp Aushe teehee ewatatele a a a a meester tues EREE 141 6 1 4 omp threadprivatens e E EE A es O ete O 141 6 15 Run time Library Routines lt cc se ccc cccscesceescdesceccecseces sti seccscesgnceedeasuccecesisdeacedactieeveei essted 141 6 16 Environment Variables sa a esses eid dane pe uh a eran A ae iia 144 OPTIMIZATION DIRECTIVES AND PRAGMAS 147 7 1 Adding Directives to Fortran ccccccccssessseessesseeescesecesecese
33. Displays the compiler assembler and linker phase invocations Passes arguments to a specific phase w Donotprint warning messages Do not print warning messages Table 3 2 C and C specific Compiler Options Option Description no_ alternative tokens pgcc only Enable disable recognition of E E alternative tokens These are tokens that make it possible to write C without the use of the amp and and characters The alternative tokens include the operator keywords e g and bitand etc and digraphs The default is no alternative tokens pgcc only Compile with cfront 2 1 compatibility This accepts constructs and a version of C that is not part of the language definition but is accepted by cfront pgcc only Compile with cfront 3 0 compatibility See b above pgCc only Enable or disable recognition of bool The default value is bool Chapter 3 Option no builtin cfront_2 1 cfront_3 0 create_pch filename dependencies see M Dependencies_to_file filename diag_error tag diag remark tag diag suppress tag diag_ warning tag display_error_number enumber no_ exceptions gnu_ extensions instantiation_dir no llalign lt 5 Command line Options Description Do don t compile with math subroutine builtin support which causes selected math library routines to be inlin
34. Inlining To invoke the function inliner use the Minline option If you do not specify an inline library the compiler performs a special prepass on all source files named on the compiler command line before it compiles any of them This pass extracts functions that meet the requirements for inlining and puts them in a temporary inline library for use by the compilation pass Several Minline options let you determine the selection criteria for functions to be inlined These selection criteria include except func Inline all eligible functions except func a function in the source text Multiple functions can be listed comma separated name func A function name which is a string matching func a function in the source text size n A size which instructs the compiler to select functions with a statement count less than or equal to n the specified size Function Inlining 105 Note the size n may not exactly equal the number of statements in a selected function the size parameter is used as a rough gauge levels n A level number which represents the number of function calling levels to be inlined The default number is one 1 Using a level greater than one indicates that function calls within inlined functions may be replaced with inlined code This allows the function inliner to automatically perform a sequence of inline and extract processes lib file ext A library file name This instructs the inliner to inline th
35. Msafe_lastval n se 67 Mno frame s 002 cc0csccedessiesecesenttesceictee 77 Msafept nenni iieii 77 EOI E setiven d sade seas ESET 77 BMSAVGs aaa ae dicta EAE EEE 71 aee e E eceeeeeeeeneesees 71 Macha rnn ai N RA 73 Mnolist s 4 cicetcceasioctetec eth eee See 82 Msecond_underscote ccceeseeeees 67 MnOMaIN 5 8c0c0sccccecescessiceteesteceseeieee 66 Msignextend 2 0 0 0 ceeeeseeseeeteeeeeeneeenes 67 Mnoonetrip 0 0 eeeeeeeeeeeeeeeeeeeeeeeenee 71 AMSIN BIE ssic sactescctiastaanstatsdeeieasiategtanvbes 73 Mnoopenmp ircen nanana s 82 MSPS avs Wee Ae ti ee 84 Mnopgdllmain ossos 82 Mstandard 00 ce ceceeeceseeeeeseeneeeeeneees 71 MDOT ransana e nine 77 MstrideO earen 68 Mnor8intrinsics oe eee cess 77 MMS VOTE e E ih econ trees 84 HMn0Save s i iaa A a E 71 Mue hat oenen A AAR 73 MioSSiMp assented Saoetieee hs 82 M n es aes oh ee ee as eset so 68 Mnosignextend s s 67 Munixlogical oo eeecsesseeseeeeeeeeeeees 71 Mnsingle osccsinniieiiriioniissni 73 sMuntoll c hiteecoiciinanwnreiinen 78 Mnostartup cne ot 68 mIMUPCASC Se eene e e i 72 Mnostdlibitin3 cncacidtin ces ea 68 CMVCCt ane ceeds Khater aet 78 Mnostride0 eee sceeeeceseeeeeeeneeeens 68 aE R AE E were seaueeniseeanavs aotees 86 Mnounixlogical seeeeeeereeeeee T2 O a duke ER A 85 Mnoumnroll assseneciiriiniriiiiiek 78 one_instantiation_per_object 100 MNOUP CASE ceecceeseeeeeeseeteeeteeeteensees 72 optk_allow_dollar_in_i
36. Parallel code activated if loop count gt 100 block distribution Generating sse code for inner loop Generated prefetch instructions for 3 loads You can see from the Minfo messages that the data initialization loops in the main program as well as the computation loop in subroutine loop have been auto parallelized Following is the result if the generated executable is run and timed on a standalone AMD Opteron 1 6 Ghz multi CPU system setenv NCPUS 2 bin time a out 1 000000 771 000 3618 000 6498 00 9999 00 o 6 o 6 8 00user 0 55system 0 04 27 elapsed 200 CPU Optimization amp Parallelization 33 Note that the user time is the aggregate of the user time spent in all executing threads in this case there are 2 The elapsed time is almost 35 less than the elapsed time required to execute the program using only 1 thread so the speed up on two processors is about 1 52 times over the single processor execution time 2 6 3 Loops That Fail to Parallelize In spite of the sophisticated analysis and transformations performed by the compiler programmers will often note loops that are seemingly parallel but are not parallelized In this subsection we ll look at some examples of common situations where parallelization does not occur Timing Loops Often loops will occur in programs that are similar to timing loops The outer loop in the example below is one such loop do 1j 41 2 do fay dy a i b 1 conti
37. Unable to open include file File is missing read protected or a maximum include depth 10 is exceeded 0018 Illegal label field The label field first five characters of the indicated line contains a non numeric character 0019 Illegally placed continuation line A continuation line does not follow an initial line or more than 99 continuation lines were specified 0020 Unrecognized compiler directive 0021 Label field of continuation line is not blank The first five characters of a continuation line must be blank 0022 Unexpected end of file missing END statement 0023 Syntax error unbalanced parentheses 0024 CHARACTER or Hollerith constant truncated to fit data type A character or Hollerith constant was converted to a data type that was not large enough to contain all of the characters in the constant This type conversion occurs when the constant is used in an arithmetic expression or is assigned to a non character variable The character or Hollerith constant is truncated on the right that is if 4 characters are needed then the first 4 are used and the remaining characters are discarded 230 Appendix B 0025 Illegal character ignored The current line contains a character possibly nonprinting which is not a legal Fortran character characters inside of character or Hollerith constants cannot cause this error As a general rule all non printing characters are treated as white space characters blanks and ta
38. a command line option affects the entire source file that is being compiled pragmas apply the effects of a particular command line option to selected functions or to selected loops in the source file 154 Chapter 7 Pragmas may also toggle an option selectively enabling and disabling the option Pragmas let you tune selected functions or loops based on your knowledge of the code The general syntax of a pragma is pragma scope pragma body The optional scope field is an indicator for the scope of the pragma some pragmas ignore the scope indicator The valid scopes are global indicates the pragma applies to the entire source file routine indicates the pragma applies to the next function loop indicates the pragma applies to the next loop but not to any loop contained within the loop body Loop scoped pragmas are only applied to for and while loops If a scope indicator is not present the default scope if any is applied Whitespace must appear after the pragma keyword and between the scope indicator and the body of the pragma Whitespace may also surround any special characters such as a comma or an equal sign Case is significant for the names of the pragmas and any variable names that appear in the body of the pragma 7 5 C C Pragma Summary Table 7 2 summarizes the supported pragmas The scope entry in the table indicates the permitted scope indicators for each pragma the letters L R and G indicate loop routin
39. a parallel region of code there may be a sub region of code that should execute only on the master thread Instead of ending the parallel region before this subregion and then starting it up again after this subregion the omp master pragma allows the user to conveniently designate code that executes on the master thread and is skipped by the other threads There is no implied barrier on entry to or exit from a master section Nested master sections are ignored Branching into or out of a master section is not supported include lt stdio h gt include lt omp h gt main int a 2 1 1 pragma omp parallel afomp get thread_num omp_get_thread_num pragma omp master printf YOU SHOULD ONLY SEE THIS ONCE n printf a 0 d a 1 d n a 0 a 1 6 5 omp single The OpenMP omp single pragma uses the following syntax pragma omp single Clauses lt C C structured block gt Clauses private list firstprivate list nowait In a parallel region of code there may be a subregion of code that will only execute correctly on a single thread Instead of ending the parallel region before this subregion and then starting it up again after this subregion the omp single pragma allows the user to conveniently designate code that executes on a single thread and is skipped by the other threads There is an implied barrier on exit from a single process section unless the optional
40. a useful tool when working with shared object files and executables that reference them When applied to an executable as follows ldd myprog ldd lists all shared object files referenced in the executable along with the pathname of the directory from which they will be extracted If the pathname is not hard coded using the R option andif LD_LIBRARY_PATH is not initialized the pathname is listed as not found See the online man page for 1dd for more information on options and usage 8 3 Creating and Using Dynamic Link Libraries on Win32 Some of the PGI compiler runtime libraries are available in both static library and dynamic link library DLL form for Win32 There are several differences between these two types of libraries Both libraries are used when resolving external references when linking an executable but the process differs for each type of library When linking with a static library the code needed from the library is incorporated into the executable Once the executable has been built the library is no longer needed the executable does not rely on the static library at runtime When linking with a DLL external references are resolved using the DLL s import library not the DLL itself The code in the DLL associated with the external references does not become a part of the executable The DLL is loaded when the executable that needs it is run For the DLL to be loaded in this manner the DLL must be in your path Wind
41. algebraic identity removal constant folding common sub expression elimination pipelining redundant load and store elimination scheduling strength reduction and peephole optimizations Optimization amp Parallelization 19 Global Optimization This optimization is performed on code over all its basic blocks The optimizer performs control flow and data flow analysis for an entire program All loops including those formed by IFs and GO TOs are detected and optimized Global optimization includes constant propagation copy propagation dead store elimination global register allocation invariant code motion and induction variable elimination Loop Optimization Unrolling Vectorization and Parallelization The performance of certain classes of loops may be improved through vectorization or unrolling options Vectorization transforms loops to improve memory access performance Unrolling replicates the body of loops to reduce loop branching overhead and provide better opportunities for local optimization and scheduling of instructions Performance for loops on systems with multiple processors may also improve using the parallelization features of the compiler Inter Procedural Analysis and Optimization IPA Interprocedural analysis allows use of information across function call boundaries to perform optimizations that would otherwise be unavailable For example if the actual argument to a function is in fact a constant in the caller
42. all statements execute with the same semantics as if they were running in serial However parallel statements or constructs operating on explicitly distributed data are in fact executed in parallel The programmer must manually insert compiler directives to cause data to be distributed to the available processors See the PGHPF User s Guide and The High Performance Fortran Handbook for more details on constructing and executing data parallel programs on shared memory or distributed memory cluster systems using PGHPF Several ready made examples of data parallel HPF programs are available at the URL ftp ftp pgroup com pub HPF examples In particular the matrix multiply example matmul tar is a good example to use Unpack it within a UNIX shell window or a BASH for Win32 command window using the command o tar xvf matmul tar and follow the instructions in the supplied README file Also available at this URL is HPF implementations of several of the NAS Parallel Benchmarks 1 5 Using the PGI Compilers on Linux 1 5 1 Linux Header Files The Linux system header files contain many GNU gcc extensions Many of these extensions are supported This should allow the PGCC C and C compilers to compile most programs compilable with the GNU compilers A few header files not interoperable with previous revisions of the PGI compilers have been rewritten and are included in PGVlinux86 include These files Getting Started 15 are sigset h
43. and pointers are treated with the same copyin and copyout semantics as Fortran dummy arguments global instructs the compiler that global or external pointers and arrays do not overlap or conflict with each other and are independent local auto instructs the compiler that local pointers and arrays do not overlap or conflict with each other and are independent Command line Options 77 78 static instructs the compiler that static pointers and arrays do not overlap or conflict with each other and are independent Munroll option option invokes the loop unroller This also sets the optimization level to 2 if the level is set to less than 2 The option is one of the following Com instructs the compiler to completely unroll loops with a constant loop count less than or equal to m a supplied constant If this value is not supplied the m count is set to 4 instructs the compiler to unroll u times a loop that is not completely unrolled or has a non constant loop count If u is not supplied the unroller computes the number of times a candidate loop is unrolled Mnounroll instructs the compiler not to unroll loops Mvect option option invokes the code vectorizer where option is one of the following altcode n noaltcode aSSOC NOASSOC cachesizen Instructs the vectorizer to generate alternate scalar code for vectorized loops If altcode is specified without arguments the
44. booll character letterl integer numintl numint2 double precision numdoub1 real numfloatl integer 2 numshorl booll true letterl v numintl 11 192 Chapter 10 numint2 44 numdoubl 902 numfloatl 39 6 numshorl 299 return end Example 10 5 Fortran Subroutine forts f main char booll letterl Int numintl numint2 float numfloatl double numdoub1 short numshor1l extern void forts forts amp booll amp letter1 amp numint1 amp numint2 amp numfloatl numdoubl amp numshorl 1 printf 3s c d Sd 3 1f 0f d n booll TRUE FALSE letterl numintl numint2 numfloatl numdoubl numshorl Example 10 6 C Main Program cmain c To compile this Fortran subroutine and C program use the following commands pgcce c cmain f pgf90 Mnomain cmain o forts f Executing the resulting a out file should produce the following output TRUE v 11 44 39 6 902 299 10 10 Example C Calling C void cfunc numl num2 res int numl num2 res Inter language Calling 193 printf func a d b d ptr c x n numl num2 res res numl num2 printf func res d n res Example 10 7 Simple C Function cfunc c extern C void cfunc int n int m int p include lt iostream gt main int a b c a 8 b 2 cout lt lt main a lt lt a lt lt b lt lt b lt lt ptr c lt lt amp c lt lt endl cfunc a b amp c cout lt lt
45. compiler to create an executable in which the statements between PARALLEL and END PARALLEL are executed by multiple lightweight threads The code that lies between PARALLEL and END PARALLEL is called a parallel region 112 Chapter 5 The OpenMP parallelization directives support a fork join execution model in which a single thread executes all statements until a parallel region is encountered At the entrance to the parallel region a system dependent number of symmetric parallel threads begin executing all statements in the parallel region redundantly These threads share work by means of work sharing constructs such as parallel Do loops see below The number of threads in the team is controlled by the OMP_ NUM THREADS environment variable If OoMP_NUM_THREADS is not defined the program will execute parallel regions using only one processor Branching into or out of a parallel region is not supported All other shared memory parallelization directives must occur within the scope of a parallel region Nested PARALLEL END PARALLEL directive pairs are not supported and are ignored The END PARALLEL directive denotes the end of the parallel region and is an implicit barrier When all threads have completed execution of the parallel region a single thread resumes execution of the statements that follow Note By default there is no work distribution in a parallel region Each active
46. constant more than 4 characters In certain contexts Hollerith constants may not be more than 4 characters long Messages 237 0091 Constant expression of wrong data type 0092 Illegal use of variable length character expression A character expression used as an actual argument or in certain contexts within I O statements must not consist of a concatenation involving a passed length character variable 0093 Type conversion of expression performed An expression of some data type appears in a context that requires an expression of some other data type The compiler generates code to convert the expression into the required type 0094 Variable is of wrong data type The indicated variable is used in a context that requires a variable of some other data type 0095 Expression has wrong data type An expression of some data type appears in a context that requires an expression of some other data type 0096 Illegal complex comparison The relations LT GT GE and LE are not allowed for complex values 0097 Statement label has been defined more than once More than one statement with the indicated statement number occurs in the subprogram 0098 Divide by zero 099 Illegal use of an aggregate RECORD Aggregate record references may only appear in aggregate assignment statements unformatted I O statements and as parameters to subprograms They may not appear for example in expressions Also records with differe
47. contains fewer and simpler name spaces than there are in the C language Specifically e Overloaded function names are not allowed in the intermediate language e Classes have their own scopes in C but not in the generated intermediate language For example an entity x from inside a class must not conflict with an entity x from the file scope e External names in the object code form a completely flat name space The names of entities with external linkage must be projected onto that name space so that they do not conflict with one another A function f from a class A for example must not have the same external name as a function f from class B e Some names are not names in the conventional sense of the word they re not strings of alphanumeric characters for example operator We can see that there are two problems here 1 Generating external names that will not clash 2 Generating alphanumeric names for entities with strange names in C Name mangling solves these problems by generating external names that will not clash and alphanumeric names for entities with strange names in C It also solves the problem of generating hidden names for some behind the scenes language support in such a way that they will match up across separate compilations You will see mangled names if you view files that are translated by PGCC C and you do not use tools that demangle the C names Intermediate files that use mangled names includ
48. cross iteration data dependencies Cross iteration dependencies from reductions and expandable scalars are excluded from consideration enabling more loops to be parallelizable In general loops with calls are not parallelized due to unknown side effects Also loops with low trip counts are not parallelized since the overhead in setting up and starting a parallel loop will likely outweigh the potential benefits compiler switches let you override some of these restrictions on auto parallelization 2 6 1 Auto parallelization Sub options The parallelizer performs various operations that can be controlled by arguments to the Mconcur command line option The following sections describe these arguments that affect the operation of the vectorizer In addition these vectorizer operations can be controlled from within code using Optimization amp Parallelization 31 directives and pragmas For details on the use of directives and pragmas refer to Chapter 7 Optimization Directives and Pragmas By default Mvect without any sub options is equivalent to Mvect dist block This enables parallelization of loops with blocked iteration allocation across the available threads of execution These defaults may vary depending on the target system Altcode Option The option Mconcur altcode instructs the parallelizer to generate alternate scalar code for parallelized loops If altcode is specified without arguments the parallelizer determi
49. equivalent to bounds and Mopt is equivalent to opt Table 7 1 Fortran Directive Summary Directive Function Default altcode Do don t generate scalar code for vector altcode noaltcode regions assoc Do don t perform associative assoc noassoc transformations bounds Do don t perform array bounds checking nobounds nobounds cncall Loops are considered for parallelization nocncall nocncall even if they contain calls to user defined subroutines or functions or if their loop counts do not exceed usual thresholds concur Do don t enable auto concurrentization of concur noconcur loops depchk Do don t ignore potential data depchk nodepchk dependencies eqvchk Do don t check EQUIVALENCE s for eqvchk noeqvchk data dependencies Ignore potential data dependencies depchk Istval Do don t compute last values Istval nolstval Select optimization level N A safe_lastval Parallelize when loop contains a scalar not enabled used outside of loop invarif Do don t remove invariant if constructs invarif foinvarif from loops 148 Chapter 7 Directive Function Default Scope unroll Do don t unroll loops nounroll 1 rg nounroll vector Do don t perform vectorizations vector 1 xg novector vintr Do don t recognize vector intrinsics vintr 1 xg novintr In the case of the vector novector directive the scope is the code following the directive until the end of the routine for r scoped
50. executes the sequentially last iteration updates the version of the object that existed before the construct Optimization MP Pragmas for C and C 135 The reduction clause is as described in Section 6 2 omp parallel The schedule clause is explained below If ordered code blocks are contained in the dynamic extent of the for directive the ordered clause must be present See Section 6 omp ordered for more information on ordered code blocks The omp for pragma directs the compiler to distribute the iterative for loop immediately following across the threads available to the program The for loop is executed in parallel by the team that was started by an enclosing parallel region Branching into or out of an omp for loop is not supported and omp for pragmas may not be nested By default there is an implicit barrier after the end of the parallel loop the first thread to comp lete its portion of the work will wait until the other threads have finished their portion of work If nowait is specified the threads will not synchronize at the end of the parallel loop Other items to note about omp for loops The for loop index variable is always private and must be a signed integer omp for loops must be executed by all threads participating in the parallel region or none at all e The for loop must be a structured block and its execution must not be terminated by break e Values of the loop control expressions and the chunk expr
51. file empty Source input file does not contain any Fortran statements other than comments or compiler directives 0007 Subprogram too large to compile at this optimization level There is an internal compiler data structure overflow working storage exhausted or some other non recoverable problem related to the size of the subprogram If this error occurs at opt 2 or greater reducing the opt level to 1 may solve the problem Moving the subprogram being compiled to its own source file may eliminate the problem If this error occurs while compiling a subprogram of fewer than 2000 statements it should be reported to PGI as a possible compiler problem 0008 Error limit exceeded The compiler gives up after 50 severe errors 0009 Unable to open assembly file The user probably does not have write permission for the current working directory 0011 Unrecognized command line switch Refer to Chapter 3 Command Line Options for a list of allowed compiler switches 0012 Value required for command line switch Certain switches require a value immediately following 0013 Unrecognized value specified for command line switch 0014 Ambiguous command line switch Messages 229 Too short an abbreviation was used for one of the switches 0015 Hexadecimal or octal constant truncated to fit data type 0016 Identifier truncated to 31 chars An identifier may be at most 31 characters characters after the 31st are ignored 0017
52. file3 c stcc o a out filel o file2 o0 file3 o The stcc driver invokes the compiler and assembler as required to process each source file and for the link command stcc invokes the linker If you modify one of the source files for example filel c the executable can be rebuilt by compiling just that file and then relinking stcc c filel c S stec o a out filel o file2 o0 file3 o 2 7 3 Building a Program Without IPA Using the make Utility The program compilation and linking process can be simplified greatly using the make utility on systems where it is supported Using a makefile such as the following 38 Chapter 2 a out filel o file2 o file3 o stcc OPT o a out filel o file2 o file3 o filel o filel c stcc OPT c filel c file2 o file2 c stcc OPT c file2 c file3 o0 file3 c stcc OPT c file3 c It is possible to type a single make command make a out The make utility determines which object files are out of date with respect to their corresponding source files and invokes stcc to recompile only those source files and relink the executable If the user subsequently edits one or more source files the executable can be rebuilt with the minimum number of compilations using the same single make command 2 7 4 Building a Program Using IPA STCC interprocedural analysis and optimization are designed to alter the standard STCC and make utility command level interfaces outlined above as little as possible
53. function name A static data member name A member function declaration A template function declaration C Template Instantiation A lt int gt A lt int gt f A lt int gt i void A lt int gt char f int f int char float 205 A pragma directive in which the argument is a template class name for example A lt int gt is equivalent to repeating the pragma for each member function and static data member declared in the class When instantiating an entire class a given member function or static data member may be excluded using the pragma do_not_instantiate For example pragma instantiate A lt int gt pragma do_not_instantiate A lt int gt f The template definition of a template entity must be present in the compilation for an instantiation to occur If an instantiation is explicitly requested by use of the instantiate pragma and no template definition is available or a specific definition is provided an error is issued template lt class T gt void f1 T No body provided template lt class T gt void g1 T No body provided void fl int Specific definition void main int ay double d error specific definition pragma instantiate void f1 int error no body provided pragma instantiate void gl int f1 double and g1 double will not be instantiated because no bodies were supplied but no errors will be produced during the compilation if no bodies are supplied at
54. group can only consist of arrays and scalars that are not dummy arguments or pointer based variables Messages 239 109 Overflow in hexadecimal constant constant truncated at left A hexadecimal constant requiring more than 64 bits produces an overflow The hexadecimal constant is truncated at left e g 1234567890abcdef1 x will be 234567890abcdefl1 x 110 Overflow in octal constant constant truncated at left An octal constant requiring more than 64 bits produces an overflow The octal constant is truncated at left e g 2777777777777777777777 0 will be 777777777777777777777 0 111 Underflow of real or double precision constant 112 Overflow of real or double precision constant 113 Label is referenced but never defined 116 Illegal use of pointer based variable 117 Statement not allowed within STRUCTURE definition 118 Statement not allowed in DO IF or WHERE block 119 Redundant specification for Data type of indicated symbol is specified more than once 120 Label is defined but never referenced 121 Operation requires logical or integer data types An operation in an expression was attempted on data having a data type incompatible with the operation For example a logical expression can consist of only logical elements of type integer or logical Real data would be invalid 122 Character string truncated Character string or Hollerith constant appearing in a DATA statement or PARAMETER statement has been truncated
55. have its own local copy of each template it instantiates With the one instantiation per object implementation each template instantiation becomes an object in the instantiation directory default name Template dir Asa result each template can be linked in independently At link time pgpre nk removes all the unnecessary instantiations in the local Template dir objects and passes the remaining objects to the linker This is particularly useful in template libraries where it results in only one instantiation of any template in the library The one_instantiation_per_object flag must appear on both the compile line and the link line It will create an additional o in the instantiation directory for each instantiation If instantiation_dir is not used see below a temporary directory Template dir is created and used as the instantiation directory then deleted after linking of the executable If you use one_instantiation_per_object to create your own template libraries you must add the objects in the Template directory to the archive list See the example below Note Before using this switch for the first time all ii files should be removed Old ii files will cause unpredictable results Note that the compiler also creates ti_ files for use during instantiation optk_allow_dollar_in_id_chars pgCC only Accept dollar signs in identifiers 100 Chapter 3 Stops compilation after the preprocessing phase
56. if your program uses only float parameters However since ANSI C specifies that routines must convert float parameters to double parameters in non prototyped functions this option results in non ANSI conformant code Mnosingle instructs the compiler to convert float parameters to double parameters in non prototyped functions Muchar instructs the compiler to treat plain char declarations as unsigned char Default For arguments that you do not specify the defaults are as follows noasmkeyword nosingle dollar _ schar Usage In this example the compiler allows the asm keyword in the source file pgcc Masmkeyword myprog c In the following example the compiler maps the dollar sign to the dot character pgcc Mdollar myprog c Command line Options 73 In the following example the compiler treats floating point constants as float values pgcc Mfcon myprog c In the following example the compiler does not convert float parameters to double parameters pgcc Msingle myprog c Without Muchar or with Mschar the variable ch is a signed character char ch Signed char sch If Muchar is specified on the command line pgcc Muchar myprog c char ch above is equivalent to unsigned char ch Optimization Controls Mcache_align Align unconstrained objects of length greater than or equal to 16 bytes on cache line boundaries An unconstrained object is a data object that is not a
57. implemented explicit is accepted to declare Non converting constructors The scope of a variable declared in a for init statement ofa loop is the scope of the loop not the surrounding scope Member templates are implemented The new specialization syntax using template lt gt is implemented Cy qualifiers are retained on rvalues in particular on function return values The distinction between trivial and nontrivial constructors has been implemented as has the distinction between PODs and non PODs with trivial constructors The linkage specification is treated as part of the function type affecting function overloading and implicit conversions extern inline functions are supported and the default linkage for inline functions is external A typedef name may be used in an explicit destructor call Placement delete is implemented An array allocated via a placement new can be deallocated via delete Covariant return types on overriding virtual functions are supported enum types are considered to be non integral types Appendix C Partial specialization of class templates is implemented Partial ordering of function templates is implemented Function declarations that match a function template are regarded as independent functions not as guiding declarations that are instances of the template It is possible to overload operators using functions that take enum types and no class types Explicit specification o
58. is assumed to end with the enclosed Do loop DO END DO directive pairs may not be nested Branching into or out of a SOMP DO loop is not supported By default there is an implicit barrier after the end of the parallel loop the first thread to complete its portion of the work will wait until the other threads have finished their portion of work If NOWAIT is specified the threads will not synchronize at the end of the parallel loop Other items to note about SOMP DO loops e The DO loop index variable is always private e SOMP DO loops must be executed by all threads participating in the parallel region or none at all e The END DO directive is optional but if it is present it must appear immediately after the end of the enclosed DO loop PROGRAM DO_USE REAL A 1000 B 1000 DO I 1 1000 B I FLOAT I ENDDO SOMP PARALLEL SOMP DO DO I 1 1000 A I SQRT B I ENDDO SOMP END PARALLEL END OpenMP Directives for Fortran 119 The SCHE DULE clause specifies how iterations of the Do loop are divided up between processors Given a SCHEDULE type chunk clause type can be STATIC DYNAMIC GUIDED or RUNTIME These are defined as follows When SCHEDULE STATIC chunk is specified iterations are allocated in contiguous blocks of size chunk The blocks of iterations are statically assigned to threads in a round robin fashion in order of the thread ID
59. library is created or updated using one of the PGI compilers the last change date of the library directory is updated This allows a library to be listed as a dependence in a makefile and ensures that the necessary compilations will be performed when a library is changed 4 2 2 Updating Inline Libraries Makefiles If you use inline libraries you need to be certain that they remain up to date with the source files they are inlined into One way to assure that inline libraries are updated is to include them in your makefiles The makefile fragment shown in Example 4 1 assumes that the file utils f contains a number of small functions that are used in the files parser f and alloc f The makefile also maintains the inline library utils il Note that the makefile updates the library whenever you change utils f or one of the include files it uses In turn the makefile compiles parser f and alloc f whenever you update the library SRC mydir FC pgf90 FFLAGS 02 main o SRC main f SRC global h FC FFLAGS c SRC main f utils o SRC utils f SRC global h SRC utils h FC FFLAGS c SRC utils f utils il SRC utils f SRC global h SRC utils h S FC FFLAGS Mextract 15 o utils il parser o SRC parser f SRC global h utils il FC FFLAGS SRC parser f alloc o SRC alloc f SRC global h utils il FC FFLAGS SRC alloc f myprog main o utils o parser o alloc o inline uti
60. link time a linker error will be produced A member function name e g A lt int gt f can only be used as a pragma argument if it refers to a single user defined member function i e not an overloaded function Compiler generated functions are not considered so a name may refer to a user defined constructor even if a compiler generated copy constructor of the same name exists Overloaded member functions can be instantiated by providing the complete member function declaration as in pragma instantiate char A lt int gt f int char 206 Chapter 11 The argument to an instantiation pragma may not be a compiler generated function an inline function or a pure virtual function 11 3 Automatic template instantiation The goal of an automatic instantiation mode is to provide painless instantiation You should be able to compile source files to object code then link them and run the resulting program and never have to worry about how the necessary instantiations get done In practice this is hard for a compiler to do Our approach requires that for each instantiation required there is some normal top level explicitly compiled source file that contains both the definition of the template entity and of any types required for the particular instantiation Isn t this always the case No Suppose that file A contains a definition of class x and a reference to Stack lt X gt push and that file B contains the definition for
61. masked the processor recovers from the exceptions and continues If a floating point exception occurs and its corresponding mask bit is off or unmasked execution terminates with an arithmetic exception C s SIGFPE signal Ktrap fp is equivalent to Ktrap inv divz ovf Default The default is Knoieee Specifies a directory to search for libraries Use L to add directories to the search path for library files Multiple Z options are valid However the position of multiple Z options is important relative to options supplied Syntax Ldirectory Where directory is the name of the library directory Default Search the standard library directory Usage In the following example the library directory is ib and the linker links in the standard libraries required by PGF90 from lib pgf90 L lib myprog f In the following example the library directory lib is searched for the library file ibx a and both the directories lib and libz are searched for liby a pgf90 L lib 1x L libz ly myprog f Loads a library The linker searches lt 1ibrary gt in addition to the standard libraries Libraries specified with are searched in order of appearance and before the standard libraries Syntax llibrary 60 Chapter 3 Where library is the name of the library to search The compiler prepends the characters ib to the library name and adds the a extension following the library name Usag
62. member of an aggregate structure or common block This option does not affect the alignment of allocatable or automatic arrays Note To effect cache line alignment of stack based local variables the main program or function must be compiled with Mcache_align Mconcur option option Instructs the compiler to enable auto concurrentization of loops If Mconcur is specified multiple processors will be used to execute loops that the compiler determines to be parallelizable Where option is one of the following altcode n Instructs the parallelizer to generate alternate scalar code for parallelized loops If a tcode is specified without arguments the parallelizer determines an appropriate cutoff length and generates 74 Chapter 3 noaltcode dist block dist cyclic cncall noassoc scalar code to be executed whenever the loop count is less than or equal to that length If altcode n is specified the scalar altcode is executed whenever the loop count is less than or equal ton If noaltcode is specified the parallelized version of the loop is always executed regardless of the loop count Parallelize with block distribution this is the default Contiguous blocks of iterations of a parallelizable loop are assigned to the available processors Parallelize with cyclic distribution The outermost parallelizable loop in any loop nest is parallelized If a parallelized loop is innermost its iterations are allocate
63. messages Command line Options 81 Mkeepasm Mlist Mnolist Mmakedll Mnoopenmp Mnosgimp Mnopegdllmain severe instructs the compiler to display severe and fatal error messages warn instructs the compiler to display warning severe and fatal error messages inform instructs the compiler to display all error messages inform warn severe and fatal instructs the compiler to keep the assembly file as compilation continues Normally the assembler deletes this file when it is finished The assembly file has the same filename as the source file but with a s extension instructs the compiler to create a listing file The listing file is filename I st where the name of the source file is filename f the compiler does not create a listing file This is the default Win32 only generate a dynamic link library DLL when used in combination with the mp option causes the compiler to ignore OpenMP parallelization directives or pragmas but still process SGI style parallelization directives or pragmas when used in combination with the mp option causes the compiler to ignore SGI style parallelization directives or pragmas but still process OpenMP parallelization directives or pragmas Win32 only do not link the module containing the default D Main into the DLL This flag applies to building DLLs with the PGF 90 and PGHPF compilers If you want to replace the default D Main routine with a
64. nowait clause is specified Nested single process sections are ignored Branching into or out of a single process section is not supported The private and firstprivate clauses are as described in Section 6 2 omp parallel 134 Chapter 6 6 6 omp for The OpenMP omp for pragma uses the following syntax pragma omp for Clauses lt C C for loop to be executed in parallel gt Clauses private list firstprivate list lastprivate list reduction operator list schedule kind chunk ordered nowait The real purpose of supporting parallel execution is the distribution of work across the available threads You can explicitly manage work distribution with constructs such as if omp_get_ thread num 0 else if omp_get_thread_num 1 However these constructs are not in the form of pragmas The omp for pragma provides a convenient mechanism for the distribution of loop iterations across the available threads in a parallel region The following variables can be used e Variables declared in a private list are treated as private to each processor participating in parallel execution of the loop meaning that a separate copy of the variable exists on each processor e Variables declared in a firstprivate list are private and in addition are initialized from the original object existing before the construct e Variables declared in a lastprivate list are private and in addition the thread that
65. numbers The chunk must be a scalar integer expression If chunk is not specified a default chunk size is chosen equal to number_of iterations omp_num_threads 1 omp_num_threads When SCHEDULE DYNAMIC chunk is specified iterations are allocated in contiguous blocks of size chunk As each thread finishes a piece of the iteration space it dynamically obtains the next set of iterations The chunk must be a scalar integer expression If no chunk is specified a default chunk size is chosen equal to 1 When SCHEDULE GUIDED chunk is specified the chunk size is reduced in an exponentially decreasing manner with each dispatched piece of the iteration space Chunk specifies the minimum number of iterations to dispatch each time except when there are less than chunk iterations remaining to be processed at which point all remaining iterations are assigned If no chunk is specified a default chunk size is chosen equal to 1 When SCHEDULE RUNTIME is specified the decision regarding iteration scheduling is deferred until runtime The schedule type and chunk size can be chosen at runtime by setting the OMP_ SCHEDULE environment variable If this environment variable is not set the resulting schedule is equivalent to SCHEDULE STATIC 5 7 BARRIER The OpenMP BARRIER directive uses the following syntax SOMP BARRIER There may be occasions in a parallel
66. on data conversion Messages 253 233 too many constants to initialize group item 234 invalid edit descriptor An invalid edit descriptor has been found in a format statement 235 edit descriptor does not match item type Data types specified by I O list item and corresponding edit descriptor conflict 236 formatted record longer than 2000 characters 237 quad precision type unsupported 238 tab value out of range A tab value of less than one has been specified 239 entity name is not member of group 242 illegal operation on direct access file 243 format parentheses nesting depth too great 244 syntax error entity name expected 245 syntax error within group definition 246 infinite format scan for edit descriptor 248 illegal subscript or substring specification 249 error in format illegal E F G or D descriptor 250 error in format number missing after or t 251 illegal character in format string 252 operation attempted after end of file 254 Appendix B 253 attempt to read non existent record direct access 254 illegal repeat count in format Messages 255 Appendix C C Dialect Supported The PGCC C compiler accepts the C language as defined by The Annotated C Reference Manual ARM by Ellis and Stroustrup Addison Wesley 1990 including templates exceptions and support for the anachronisms described in section 18 of the ARM This is the same language defined by the language reference for ATT s cfr
67. on the context will treat the name as a variable or an external function 0036 Illegal implicit character range First character must alphabetically precede second 0037 Contradictory data type specified for The indicated identifier appears in more than one type specification statement and different data types are specified for it 0038 Symbol has not been explicitly declared The indicated identifier must be declared in a type statement this is required when the IMPLICIT NONE statement occurs in the subprogram 0039 Symbol appears illegally in a SAVE statement An identifier appearing in a SAVE statement must be a local variable or array 0040 Illegal common variable Indicated identifier is a dummy variable is already in a common block or has previously been defined to be something other than a variable or array 0041 Illegal use of dummy argument This error can occur in several situations It can occur if dummy arguments were specified on a PROGRAM statement It can also occur if a dummy argument name occurs ina DATA COMMON SAVE or EQUIVALENCE statement A program statement must have an empty argument list 0042 is a duplicate dummy argument 0043 Illegal attempt to redefine 232 Appendix B An attempt was made to define a symbol in a manner inconsistent with an earlier definition of the same symbol This can happen for a number of reasons The message attempts to indicate the situation that occurr
68. option it passes the option to the linker This chapter uses the following notation item Square brackets indicate that the enclosed item is optional item item Braces indicate that you must select one and only one of the enclosed items A vertical bar separates the choices Horizontal ellipses indicate that zero or more instances of the preceding item are valid Note Some options do not allow a space between the option and its argument or within an argument This fact is noted in the syntax section of the respective option Table 3 1 Generic PGI Compiler Options Option Cc c cyglibs Command line Options Description Display invocation information Show but do not execute the driver commands same as dryrun Swap bytes from big endian to little endian or vice versa on input output of unformatted data Perform array bounds checking Stops after the assembly phase and saves the object code in filename o Win32 only link against the Cygnus libraries and use the Cygnus include files You must have the full Cygwin32 environment installed in order to use this switch Defines a preprocessor macro Show but do not execute driver commands 45 46 Description E Stops after the preprocessing phase and displays the preprocessed file on the standard output Stops after the preprocessing phase and saves the preprocessed file in filename f this option is only valid for
69. out of bounds n return v i In this case we would like to use the specialized member function element C also dictates that unreferenced template functions should not be compiled So for all of these reasons the compiler cannot know what instantiations are required or in which modules to put them until the whole program is linked You should have an idea where these templates should be expanded First we discuss two methods you can use to tell the compiler where to put template instantiations Then we discuss an automatic instantiation scheme 11 1 Command Line control of template instantiation Normally when a file is compiled no template entities are instantiated except those assigned to the file by automatic instantiation see below The overall instantiation mode can however be changed by a command line option We tnone Do not automatically create instantiations of any template entities This is the default It is also the usually appropriate mode when automatic instantiation is done 204 Chapter 11 We tused We tall We tlocal Instantiate those template entities that were used in the compilation This will include all static data members for which there are template definitions Instantiate all template entities declared or referenced in the compilation unit For each fully instantiated template class all of its member functions and static data members will be instantiated whether or not they
70. physical processors can produce inefficient execution Executing a program on multiple processors in an environment where some of the processors are being time shared with another executing job can also result in inefficient execution As with the vectorizer the Mconcur option can speed up code if it contains well behaved countable loops and or computationally intensive nested loops that operate on arrays However it is possible that some codes will show a decrease in performance on multi processor systems when 32 Chapter 2 compiled with Mconcur due to parallelization overheads memory bandwidth limitations in the target system false sharing of cache lines or other architectural or code generation factors For this reason it is recommended that you check carefully whether particular program units or loops show improved performance when compiled using this option If the compiler is not able to successfully auto parallelize your application you should refer to Chapter 5 Open MP Directives for Fortran or Chapter 6 OpenMP Pragmas for C and C to see if insertion of explicit parallelization directives or pragmas and use of the mp compiler option enables the application to run in parallel 2 6 2 Auto parallelization Example Assume the program from example 2 3 is compiled as follows pgf90 fast Mvect sse Mconcur Minfo vadd f vector op 4 Parallel code generated block distribution Unrolling inner loop 8 times loop 18
71. precompiled header file of the specified name as part of the current compilation no_ using std pgcc only Enable disable implicit use of the std namespace when standard header files are included X pgcc only Generate cross reference information and place output in specified file Chapter 3 Option Description gcc only Allow in names pgcc only Enable exception handling suffix see P pgCC only Use with E F or P to save intermediate file in a file with the specified suffix 3 1 Generic PGI Compiler Options Use the option to display the invocations of the compiler assembler and linker These invocations are command lines created by the driver from your command line input and the default values Default The compiler does not display individual phase invocations Usage The following command line requests verbose invocation information pgf90 prog f Cross reference Minfo V v Use the option to display the invocations of the compiler assembler and linker but do not execute them These invocations are command lines created by the compilation driver from the PGIRC files and the command line options Default The compiler does not display individual phase invocations Usage The following command line requests verbose invocation information pgf90 myprog f Cross reference Minfo V dryrun Command line Options 51 Use the
72. quoted string exceeded the maximum allowed currently 2048 248 Appendix B 244 Syntax error in define formal not identifier A formal parameter that was not an identifier was used in a macro definition 245 Syntax error in define missing blank after name or arglist There was no space or tab between a macro name or argument list and the macro s definition 246 Syntax error in if A syntax error was found while parsing the expression following a if or elif directive 247 Syntax error in include The include directive was not correctly formed 248 Syntax error in line A line directive was not correctly formed 249 Syntax error in module A module directive was not correctly formed 250 Syntax error in undef A undef directive was not correctly formed 251 Token after ifdef must be identifier The ifdef directive was not followed by an identifier 252 Token after ifndef must be identifier The ifndef directive was not followed by an identifier 253 Too many actual parameters to The number of actual arguments to the indicated macro exceeded the maximum allowed currently 31 254 Too many formal parameters to The number of formal arguments to the indicated macro exceeded the maximum allowed currently 31 Messages 249 255 Too much pushback The preprocessor ran out of space while processing a macro expansion The macro may be recursive 256 Undefined directive The identif
73. recognized but currently always returns FALSE subroutine omp_init_lock integer_var initializes a lock associated with the variable integer_var for use in subsequent calls to lock routines This initial state of integer_var is unlocked It is illegal to make a call to this routine if integer_var is already associated with a lock subroutine omp_ destroy lock integer_var disassociates a lock associated with the variable integer_var subroutine omp_set_lock integer_var causes the calling thread to wait until the specified lock is available The thread gains ownership of the lock when it is available It is illegal to make a call to this routine if integer_var has not been associated with a lock subroutine omp_unset_lock integer var causes the calling thread to release ownership of the lock associated with integer var It is illegal to make a call to this routine if integer_var has not been associated with a lock logical function omp_test_lock integer_ var causes the calling thread to try to gain ownership of the lock associated with integer_var The function returns TRUE if the thread gains ownership of the lock and FALSE otherwise It is illegal to make a call to this routine if integer_var has not been associated with a lock OpenMP Directives for Fortran 127 5 17 Environment Variables OMP_NUM_THREADS specifies the number of threads to use during execution of parallel regions The default value for thi
74. region when it is necessary that all threads complete work to that point before any thread is allowed to continue The BARRIER directive synchronizes all threads at such a point in a program Multiple barrier points are allowed within a parallel region The BARRIER directive must either be executed by all threads executing the parallel region or by none of them 120 Chapter 5 5 8 DOACROSS The CSDOACROSS directive is not part of the OpenMP standard but is supported for compatibility with programs parallelized using legacy SGI style directives Syntax CSDOACROSS Clauses lt Fortran DO loop to be executed in parallel gt Clauses PRIVATE LOCAL list SHARED SHARE list MP_SCHEDTYPE SIMPLE INTERLEAVE CHUNK lt integer _expression gt IF logical_expression The CSDOACROSS directive has the effect of a combined parallel region and parallel Do loop applied to the loop immediately following the directive It is very similar to the OpenMP PARALLEL DO directive but provides for backward compatibility with codes parallelized for SGI systems prior to the OpenMP standardization effort The CSDOACROSS directive must not appear within a parallel region It is a shorthand notation that tells the compiler to parallelize the loop to which it applies even though that loop is not contained within a parallel region While this syntax is more convenient it should be noted that if multiple successive Do loops ar
75. same storage area for SHARED data The DEFAULT clause lets you specify the default attribute for variables in the lexical extent of the parallel region EFAULT Spec iw Individual clauses specifying PRIVATE SHARED etc status override the declared ifying DEFAULT NONE declares that there is no implicit default and in this case each variable in the parallel region must be explicitly listed with an attribute of PRIVATE SHARED FIRSTPRIVATE LASTPRIVATE or REDUCTION Variables that appear in the list of a FIRSTPRIVATE clause are subject to the same semantics as PRIVATE variables but in addition are initialized from the original object existing prior to entering the parallel region Variables that appear in the list of a REDUCTION clause must be SHARED A private copy of each variable in ist is created for each thread as if the PRIVATE clause had been specified Each private copy is initialized according to the operator as specified in table 5 1 Table 5 1 Initialization of REDUCTION Variables Operator Intrinsic Initialization At the end of th Smallest Representable Number Largest Representable Number All bits on e parallel region a reduction is performed on the instances of variables appearing in list using operator or intrinsic as specified in the REDUCTION clause The initial value of each REDUCTION variable is include
76. source file that is being compiled directives and pragmas let you do the following e Apply or disable the effects of a particular command line option to selected subprograms or to selected loops in the source file for example an optimization e Globally override command line options e Tune selected routines or loops based on your knowledge or on information obtained through profiling Chapter 7 Optimization Directives and Pragmas provides details on how to add directives and pragmas to your source files 2 10 Execution Timing and Instruction Counting As this chapter shows once you have a program that compiles executes and gives correct results you may optimize your code for execution efficiency Selecting the correct optimization level requires some thought and may require that you compare several optimization levels before arriving at the best solution To compare optimization levels you need to measure the execution time for your program There are several approaches you can take for timing execution You can use shell commands that provide execution time statistics you can include system calls in your code that provides timing information or you can profile sections of code In general any of these approaches will work however there are several important timing considerations to keep in mind e Execution should take at least five seconds the choice of five seconds is somewhat arbitrary the interval should be statistic
77. text Syntactic information includes information such as syntax errors Semantic includes information includes such as unreachable code You can specify that the compiler displays error messages at a certain level with the Minform option The compiler messages refer to a severity level a message number and the line number where the error occurs The compiler can also display internal error messages on standard errors If your compilation produces any internal errors contact your Portland Group technical support representative If you use the listing file option M ist the compiler places diagnostic messages after the source lines in the listing file in the following format PGFTN etype enum message filename line Where etype is a character signifying the severity level enum is the error number message is the error message filename is the source filename line is the line number where the compiler detected an error Messages 227 B 2 Phase Invocation Messages You can display compiler assembler and linker phase invocations by using the v command line option For further information about this option see Chapter 3 Command Line Options B 3 Compiler Error Messages This section presents the error messages generated by the compiler The compiler displays error messages in the program listing and on standard output it can also display internal error messages on standard error If a compilation produces any internal err
78. than one ORDERED directive 6 12 omp atomic The omp atomic pragma uses the following syntax pragma omp atomic lt C C expression statement gt The omp atomic pragma is semantically equivalent to subjecting the following single C C expression statement to an omp critical pragma The expression statement must be of one of the following forms e x lt binary_operator gt expr e xXt e x xy e y where x is a scalar variable of intrinsic type expr is a scalar expression that does not reference x lt binary_operator gt is not overloaded and is one of amp r lt lt or gt gt 140 Chapter 6 6 13 omp flush The omp flush pragma uses the following syntax pragma omp flush list The omp flush pragma ensures that all processor visible data items or only those specified in list when it s present are written back to memory at the point at which the directive appears 6 14 omp threadprivate The omp threadprivate pragma uses the following syntax pragma omp threadprivate list Where list is a list of variables to be made private to each thread but global within the thread This pragma must appear in the declarations section of a program unit after the declaration of any variables listed On entry to a parallel region data in a threadprivate variable is undefined unless copyin is specified on the omp parallel pragma When a variable appears in an omp threadprivate p
79. the array the location where the error occurred the source file and the line number in the source and information about the subscript which is out of bounds its value its upper bound and its dimension Cross reference Mbounds Mnobounds Stops after the assembling phase Use the c option to halt the compilation process after the 52 Chapter 3 assembling phase and write the object code to the file filename o where the input file is filename f Default The compiler produces an executable file does not use the c option Usage In this example the compiler produces the object file myprog o in the current directory pgf90 c myprog f Cross reference E Mkeepasm o and S cyglibs Win32 only link against the Cygnus libraries and use the Cygnus include files You must have the full Cygwin32 environment installed in order to use this switch Default The compiler does not link against the Cygnus libraries Defines a preprocessor macro Use the D option to create a macro with a given value The value must be either an integer or a character string You can use the D option more than once on a compiler command line The number of active macro definitions is limited only by available memory You can use macros with conditional compilation to select source text during preprocessing A macro defined in the compiler invocation remains in effect for each module on the command line unless you r
80. the first step of building DLLs with mutual imports These errors will be resolved in the next step where each DLL is built and linked against the import library previously created for the other DLL pgcc object3 o0 Mmakedll o obj3 dll L lobj4 pgcc object4 0 Mmakedll o obj4 dll L lobj3 a a Step 2 Compile the main program and link against the import libraries for obj 3 d11 and obj4 dll o pgcc Mdll o prog2 prog2 c L lobj3 lobj4 Step 3 Execute prog2 to ensure that the DLLs were created properly 170 Chapter 8 prog2 func_3a calling a routine in obj4 dll func_4b func_4a calling a routine in obj3 d11 func_3b 8 5 Using LIB3F on Win32 Previous releases of the PGI Fortran compilers on Win32 required special compiler options or include statements for program units using LIB3F routines LIB3F is now supported on Win32 by default without the need for specific steps taken by the programmer See the PGF77 Workstation Reference Manual for further information on the contents of LIB3F 8 6 LAPACK the BLAS and FFTs Pre compiled versions of the public domain LAPACK and BLAS libraries are included with the PGI compilers on UNIX and Win32 systems in the files SPGI lt target gt lib lapack a and PGI lt target gt lib blas a respectively where lt target gt is replaced with the appropriate target name linux86 linux86 64 or nt86 To use these libraries simply link them in using the option when linki
81. to fit the declared size of the corresponding identifier 240 Appendix B 123 Hollerith length specification too big reduced The length specifier field of a Hollerith constant specified more characters than were present in the character field of the Hollerith constant The length specifier was reduced to agree with the number of characters present 124 Relational expression mixes character with numeric data A relational expression is used to compare two arithmetic expressions or two character expressions A character expression cannot be compared to an arithmetic expression 125 Dummy procedure not declared EXTERNAL A dummy argument which is not declared in an EXTERNAL statement is used as the subprogram name in a CALL statement or is called as a function and is therefore assumed to be a dummy procedure This message can result from a failure to declare a dummy array 126 Name is not an intrinsic function 127 Optimization level for changed to opt 1 128 Integer constant truncated to fit data type An integer constant will be truncated when assigned to data types smaller than 32 bits such as a BYTE 129 Floating point overflow Check constants and constant expressions 130 Floating point underflow Check constants and constant expressions 131 Integer overflow Check floating point expressions cast to integer 132 Floating pt invalid oprnd Check constants and constant expressions 133 Divide by 0 0 Check constants and cons
82. vect_index t vect_index_t size public vect vect_index t x size x v new int x T amp operator vect_index_t T amp element vect_index t i E7 template lt class T gt T amp vect lt T gt element vect_index_t i return v i This program fragment shows the template class vect being used use the template vect lt int gt x 80 vect lt double gt d 20 void foo void int j x element 5 double f d element 6 C Template Instantiation 203 The previous program requires two instantiations of template class vect one where T is int and one where T is double It would seem the compiler could just generate these instantiations but unfortunately things are not that simple If the template declaration of class vect was in an include file and another module included it and used it other instantiations of this template might be needed Another module might also require an instantiation where T is int In this case we would like only one instantiation of template class vect where T is int C also allows specialization of a template entity This is a type specific version to be used in place of the version that would have been generated from the template In the above example someone could write a specialization for type int int amp vect lt int gt element vect_index_t i check bounds for int vectors if i gt size extern void error char error vect index
83. with internal unformatted formatted and list directed I O Also I O lists must not be present 213 is not a NAMELIST group name 214 Input item is not a variable reference 215 Assumed sized array name cannot be used as an I O item or specifier An assumed sized array was used as an item to be read or written or as an I O specifier i e FMT array name In these contexts the size of the array must be known 216 STRUCTURE UNION cannot be used as an I O item 217 ENCODE DECODE buffer must be a variable array or array element 221 elif after else 246 Appendix B A preprocessor elif directive was found after a else directive only endif is allowed in this context 222 else after els A preprocessor else directive was found after a else directive only endif is allowed in this context 223 if directives too deeply nested Preprocessor if directive nesting exceeded the maximum allowed currently 10 224 Actual parameters too long for The total length of the parameters in a macro call to the indicated macro exceeded the maximum allowed currently 2048 225 Argument mismatch for The number of arguments supplied in the call to the indicated macro did not agree with the number of parameters in the macro s definition 226 Can t find include file The indicated include file could not be opened 227 Definition too long for The length of the macro definition of the indicated macro exceeded the maximum all
84. you are familiar with program development in a UNIX environment editing compiling and executing programs within BASH will be very comfortable If you have not previously used such an environment you should take time to familiarize yourself with either the vi or emacs editors and with makefiles The emacs editor has an extensive online tutorial which you can start by bringing up emacs and selecting the appropriate option under the pull down help menu In addition there is a full online EMACS User s Guide included as part of the PGI compilers online manual set referenced above You can get a thorough introduction to the construction and use of makefiles in the online Makefile User s Guide A simple example makefile is included in the Linpack100 benchmark example referenced above Getting Started 17 Chapter 2 Optimization amp Parallelization Source code that is readable maintainable and produces correct results is not always organized for efficient execution Normally the first step in the program development process involves producing code that executes and produces the correct results This first step usually involves compiling without much worry about optimization After code is compiled and debugged code optimization and parallelization become an issue Invoking one of the PGI compiler commands with certain options instructs the compiler to generate optimized code Optimization is not always performed since it increases compilation ti
85. 0 13 Example C Calling Fortran Example 10 13 shows a Fortran subroutine called by the C main program shown in Example 10 14 Notice that each call uses the amp operator to pass by reference Also notice that the call to the Fortran subroutine uses all lower case and a trailing _ subroutine forts booll letterl numintl amp numint2 numfloatl numdoubl numshor1 logical 1 booll character letterl integer numintl numint2 double precision numdoub1 real numfloatl integer 2 numshorl booll true letterl v numintl 11 numint2 44 numdoubl 902 numfloatl 39 6 numshorl 299 return end Example 10 13 Fortran Subroutine forts include lt iostream gt extern C extern void forts char char int int float double short main char booll letterl int numintl numint2 float numfloat1 double numdoub1 short numshor1l Inter language Calling 197 forts amp booll amp letterl amp numintl amp numint2 amp numfloatl cou cout lt lt cout lt lt cout lt lt cout lt lt cout lt lt cout lt lt amp numdoubl amp numshorl1 t lt lt booll booll cout lt lt TRUE cout letterl N lt lt numintl lt lt numint2 lt lt numfloatl lt lt numdoubl lt lt numshorl lt lt lt lt FALSE cout lt lt endl letterl lt lt endl numintl lt lt endl nu
86. 158 Preprocessor CD Dist E ER SAE 11 Forta R a Eis 11 Run time Environment cece 217 Shared object files 0 0 00 ceeceeeesesteeseeeees 163 Timing CXECULON sateen Ane eaa 42 SYSTEM CLOCK cisions 42 Tools PGDB Gini Re ae es 4 PG PROB REE ER 4 Vectorization cccceccccceesseceesseeeeeeees 25 78 prefetch instructions ceceeeeseeeee 79 SSE instructions ceceeceseseeeseeneeeeeees 79 Win32 Calling Conventions C 198 201 Defaults 20 ike otsene 198 200 STDGCALL wisn caunksicess 198 200 symbol name construction 200 UNIX style eee AEAEE 198 201 273 274 Index
87. 1991 1991 e International Language Standard ISO Standard 1539 199 E e Fortran 90 Handbook Intertext McGraw Hill New York NY 1992 e High Performance Fortran Language Specification Revision 1 0 Rice University Houston Texas 1993 http www crpc rice edu HPFF e High Performance Fortran Language Specification Revision 2 0 Rice University Houston Texas 1997 http www crpc rice edu HPFF e OpenMP Fortran Application Program Interface Version 1 1 November 1999 http www openmp org e OpenMP C and C Application Program Interface Version 1 0 October 1998 http www openmp org e Programming in VAX Fortran Version 4 0 Digital Equipment Corporation September 1984 e IBM VS Fortran IBM Corporation Rev GC26 4119 e Military Standard Fortran DOD Supplement to American National Standard Programming Language Fortran ANSI x 3 1978 MIL STD 1753 November 9 1978 e American National Standard Programming Language C ANSI X3 159 1989 Organization This manual is divided into the following chapters and appendices Chapter 1 Getting Started provides an introduction to the PGI compilers and describes their use and overall features Chapter 2 Optimization amp Parallelization describes standard optimization techniques that with little effort allow users to significantly improve the performance of programs Chapter 3 Command Line Options provides a detailed description of each command line opti
88. 2 Example Fortran Calling C The Fortran main program shown in Example 6 11 calls the C function shown in Example 6 12 Notice that each argument is defined as a pointer in the C function since Fortran passes by reference Also notice that the C function name uses all lower case and a trailing _ Inter language Calling 195 196 logical 1 booll character letter1 integer 4 numintl numint2 real numfloatl double precision numdoubl integer 2 numshorl external cfunc call cpfunc booll letterl numintl amp numint2 numfloatl numdoubl numshor1 Write s7 L2 A2 255 Lop F621 6 1 ES amp booll letterl numintl numint2 numfloatl amp numdoubl numshorl end Example 10 11 Fortran Main Program fmain f calling a C function define TRUE Oxff define FALSE 0 non extern void cpfunc_ booll int numintl float numfloatl double numdoubl1 extern char letterl numint2 short numshortl int len _letterl booll TRUE letterl v numintl 11 numint2 44 numfloatl 39 6 numdoubl 39 2 numshortl 981 Example 10 12 C function cpfunc c Chapter 10 Assuming the Fortran program is in a file fmain f and the C function is in a file cpfunc C create an executable using the following command lines pgCC c cpfunc C pgf90 cpfunc o fmain f Executing the a out file should produce the following output T v 11 44 39 6 39 2 981 1
89. 4 processor based systems running a 64 bit operating system because the application binary interface ABI for those systems is still evolving See http www x86 64 org abi pdf for the latest version of this ABI 9 1 Fortran Data Types 9 1 1 Fortran Scalars A scalar data type holds a single value such as the integer value 42 or the real value 112 6 Table 9 1 lists scalar data types their size format and range Table 9 2 shows the range and approximate precision for Fortran real data types Table 9 3 shows the alignment for different scalar data types The alignments apply to all scalars whether they are independent or contained in an array a structure or a union Table 9 1 Representation of Fortran Data Types Fortran Format Range Data Type wrecer 2 s complement integer SS 2 s complement integer 32768 to 32767 same as INTEGER same as INTEGER lh Fortran C and C Data Types 173 Fortran Format Data Type 2 s complement 128 to 127 ae es E COMPLEX See DOUBLE PRECISION Sete __ DOUBLE ae Sameasabove Sequence of n bytes SSS 1 Approximate value The logical constants TRUE and FALSE are all ones and all zeroes respectively Internally the value of a logical variable is true if the least significant bit is one and false otherwise When the option Munixlogical is set a logical variable with a non zero value is true and with a zero value is false
90. 77 pgf90 and pghpf only no recursive allocate do not allocate local variables on the Code stack this allows recursion SAVEd data Generation initialized or namelist members are always allocated statically regardless of the setting of this switch pg 77 pg 90 and pghpf only no reentrant specifies whether the compiler avoids Code optimizations that can prevent code from being Generation reentrant no ref_externals do don t force references to names appearing Code in EXTERNAL statements pg 77 pgf90 and Generation pghpf only safeptr instructs the compiler to override data Optimization dependencies between pointers and arrays pgcc and pgCC only safe_lastval In the case where a scalar is used after a loop Code but is not defined on every iteration of the loop Generation the compiler does not by default parallelize the loop However this option tells the compiler it safe to parallelize the loop For a given loop the last value computed for all scalars make it safe to parallelize the loop no save determines whether the compiler assumes that Fortran all local variables are subject to the SAVE Language statement pgf77 pgf90 and pghpf only schar specifies signed char for characters pgcc C C and pgCc only also see uchar Language no second_underscore do don t add the second underscore to the Code name of a Fortran global if its name already Generation contains an underscore pgf77 pgf
91. 86 and AMD64 targets and includes additional options on some targets For C programs add Minline levels 10 no_exceptions pgCC fast Minline levels 10 no exceptions prog cc By experimenting with individual compiler options on a file by file basis further significant performance gains can sometimes be realized However individual optimizations can sometimes cause slowdowns depending on coding style and must be used carefully to ensure performance improvements result In addition to fast the optimization flags most likely to further improve performance are O3 Mvect sse on targets where SSE and or SSE2 are supported Mipa Mconcur on targets with multiple processors and Minline You can see a brief specification of any option by invoking one of the PGI compilers with that option in combination with help without specifying any input files For example pgf90 help Mvect Mvect assoc noassoc altcode lt n gt noaltcode cachesize lt c gt levels lt n gt prefetch recog norecog smallvect lt n gt sse nosse transform notransform idiom noidiom Control automatic vector pipelining The remainder of this chapter describes the O options the loop unroller option Munroll the vectorizer option Mvect the auto parallelization option Mconcur and the inter procedural analysis optimization Mipa Usually you should be able to get very near optimal compiled performance using some combinat
92. 90 and pghpf only no signextend specifies whether the compiler extends the sign Code bit if it is set Generation no single converts float parameters to double C C parameter characters pgcc and pgCC only Language Command line Options 65 pgflag Description Category standard causes the compiler to flag source code that Fortran does not conform to the ANSI standard pgf 77 Language pgf90 and pghpf only nostride0 the compiler generates does not generate Code alternate code for a loop that contains an Generation induction variable whose increment may be zero pg 77 pgf90 and pghpf only uchar specifies unsigned char for characters pgcec C C and pgCc only also see schar Language unix uses UNIX calling and naming conventions for Code Fortran subprograms pg 77 pgf 90 and Generation pghpf for Win32 only no unixlogical determines whether logical TRUE and Fortran FALSE are determined by non zero TRUE Language and zero FALSE values for unixlogical With nounixlogical the default 1 values are TRUE and 0 values are FALSE pgf77 pgf90 and pghpf only no unroll controls loop unrolling no upcase determines whether the compiler allows Fortran uppercase letters in identifiers pg 77 pgf90 Language and pghpf only Mpegflag Code Generation Controls Syntax Mnomain instructs the compiler of to include the object file that calls the Fortran main program as part of the
93. CALL The C or STDCALL conventions are used if an appropriate compiler directive is placed in a program unit containing the call The C and STDCALL conventions are typically used to call routines coded in C or assembly language that depend on these conventions e UNIX The UNIX convention is used in any Fortran program unit compiled using the Munix compilation flag Table 10 3 outlines each of these calling conventions Table 10 3 Calling Conventions Supported by the PGI Fortran Compilers Convention Default STDCALL C UNIX Leading underscore Yes Yes Yes Argument byte count Yes Yes No No added Arguments passed by Yes reference Character argument byte After each End of counts passed char argument argument list Character strings truncated to first character and passed by value Except arrays which are always passed by reference even in the STDCALL and C conventions Inter language Calling 199 Note While it is compatible with the Fortran implementations of Microsoft and several other vendors the C calling convention supported by the PGI Fortran compilers for Win32 is not strictly compatible with the C calling convention used by most C C compilers In particular symbol names produced by PGI Fortran compilers using the C convention are all lower case The standard C convention is to preserve mixed case symbol names You can cause any of the PGI Fortran compilers to preserve mixed case symbol names using the Mupcase
94. CCC ANSI and K amp R C PGCC C ANSI C with cfront features PGDBG Source code debugger PGPROF Performance profiler In general the designation PGF90 is used to refer to The Portland Group Compiler Technology s Fortran 90 compiler and pgf90 is used to refer to the command that invokes the compiler A similar convention is used for each of the PGI compilers and tools Preface For simplicity examples of command line invocation of the compilers generally reference the pg 90 command and most source code examples are written in Fortran Usage of the PGF 77 compiler whose features are a subset of PGF90 is similar Usage of PGHPF PGCC C and PGCC ANSI C is consistent with PGF90 and PGF77 but there are command line options and features of these compilers that do not apply to PGF90 and PGF77 and vice versa There are a wide variety of X86 compatible processors in use All are supported by the PGI compilers and tools Most of these processors are forward compatible but not backward compatible That means code compiled to target a given processor will not necessarily execute correctly on a previous generation processor The most important processor types along with a list of the features utilized by the PGI compilers that distinguish them from a compatibility standpoint are listed in table P 2 Table P 2 Processor Options Prefetch SSE1 SSE2 32 bit 64 bit AMD Athlon x AMD Athlon XP MP X X X AMD Ath
95. CTIONS END SECTIONS The OpenMP SECTIONS END SECTIONS directive pair uses the following syntax Syntax SOMP SECTIONS Clauses SOMP SECTION lt Fortran code block executed by processor i gt SOMP SECTION lt Fortran code block executed by processor j gt SOMP END SECTIONS NOWAIT Clauses PRIVATE list FIRSTPRIVATE list 122 Chapter 5 LASTPRIVATE list REDUCTION operator intrinsic list The SECTIONS END SECTIONS directives define a non iterative work sharing construct within a parallel region Each section is executed by a single processor If there are more processors than sections some processors will have no work and will jump to the implied barrier at the end of the construct If there are more sections than processors one or more processors will execute more than one section A SECTION directive may only appear within the lexical extent of the enclosing SECTIONS END SECTIONS directives In addition the code within the SECTIONS END SECTIONS directives must be a structured block and the code in each SECTION must be a structured block The available clauses are as defined in Section 5 2 PARALLEL END PARALLEL and Section 5 6 DO END DO 5 11 PARALLEL SECTIONS The OpenMP PARALLEL SECTIONS END SECTIONS directive pair uses the following syntax Syntax SOMP PARALLEL SECTIONS CLAUSE
96. Controls Syntax Mnostartup instructs the linker not to link in the standard startup routine that contains the entry point _ start for the program Note If you use the Mnostartup option and do not supply an entry point the linker issues the following error message Warning cannot find entry symbol _ start Mnostdlib instructs the linker not to link in the standard libraries libpgfinrtl a libm a libe a and libpge a in the library directory lib within the standard directory You can link in your own library with the option or specify a library directory with the L option Default For arguments that you do not specify the default environment option depends on your configuration Cross reference D I L l U 68 Chapter 3 Mpeflag Inlining Controls This section describes the Mpegflag options that control function inlining Syntax Mextract option option Extracts functions from the file indicated on the command line and creates or appends to the specified extract directory option can be any of name func size number lib dirname instructs the extractor to extract function func from the file instructs the extractor to extract functions with number or fewer statements from the file Use directory dirname as the extract directory required in order to save and re use inline libraries If you specify both name and size the compiler extracts functions that ma
97. Mode ccccceesesseessceseceseceeceseceseeeecseeeseeeseseneensees 262 C 5 cfront 2 1 Compatibility Moden inican e a a 263 C 6 cfront 2 1 3 0 Compatibility Mode ssssseessseseesseseesesseeesssesresesersessesresseseeseesersesseenesseseeses 265 Table of Contents INDEX 267 LIST OF TABLES Table P 1 PGI Compilers and Commands ceeccesccescceseceseescecseeeneeeseeeneeeesacenseceseseeenreneeeeensees 4 Table P 2 Processor Options ccscccssesssessseesceeeceecesecsecaecssecaeecaeesaecaeecaeeeseseseeeseseeeeeressrenseenseserens 5 Table 1 1 Stop after Options Inputs and Outputs cccecccceesseesceesceesceseeeseeeseeeceecsaeesseesaeeneeenes 12 Table 2 1 Optimization and O g and M lt opt gt Options cccesceeseesceeseeeseeeeeeeeceseceneeseeeseeenes 41 Table 3 1 Generic PGI Compiler Options 0 ccccececeseesceeseeeseeeeeeeeceeeeeeeeeeeenseenseeaeesaeeaecaeeenes 45 Table 3 2 C and C specific Compiler Options 0 ccccescceseesseeseeeseeeseeeeceeeeeensecssecnseeeeeneeenes 48 Table 3 3 M Options Summary sonne eai eh ite es eee else 62 Table 3 4 Optimization and O g Mvect and Mconcur Options escesseeeeeceesecneeeeeneees 85 Table 5 1 Initialization of REDUCTION Variables 0 0 eececcsseecceeceeeeceeeeceeeeceaeeaeeseeneeetenee 114 Table 6 1 Initialization of reduction Variables 0 ecssssccsesesceseeeceseeeeeseceeeeecnaeeeceaecateneeneeatea
98. NIZATION CHARACTER LEN 30 EMAIL END TYPE ATTENDEE In order to declare a variable of type ATTENDEE and access the contents of such a variable code such as the following would be used TYPE ATTENDEE ATTLIST 100 ATTLIST 1 NAME JOHN DOB 9 2 C and C Data Types 9 2 1 C and C Scalars Table 9 4 lists C and C scalar data types their size and format The alignment of a scalar data type is equal to its size Table 9 5 shows scalar alignments that apply to individual scalars and to scalars that are elements of an array or members of a structure or union Wide characters are supported character constants prefixed with an L The size of each wide character is 4 bytes Table 9 4 C C Scalar Data Types Data Type Size bytes Format Range signed char 1 two s complement 128 to 127 integer signed s 2 two s complement 32768 to 32767 integer signed int 4 two s complement 23 to 231 1 signed long int integer signed long long two s complement 263 to 263 1 int integer Fortran C and C Data Types 177 Data Type Size bytes Format Range unsigned long long ordinal 0 to 264 1 int float IEEE single precision floating point double IEEE double precision floating point long double IEEE double precision floating point bit fiela 1 to 32 bits ordinal unsigned value bit fiela 1 to 32 bits two s complement signed value integer enum two s comple
99. P Pragmas SYMUAX ieo r EE N 129 Optimization 2 0 eeeceseeeseeeseeteeeeeeteeeees 147 C C pragma ceceeeeseeteeeees 41 154 C C pragmas SCOPE ceseeeeeteees 158 cache tiling e nenta eae 78 default optimization levels 41 Index Fortran directives ccccccceeeeee 41 147 Fortran directives Scope ceceeeees 153 function lining eee 20 105 global optimization 0 0 0 eee 20 23 line bTaneS 4 8 cea seeee ets 106 Inter Procedural Analysis 20 37 TRA Saeed arie tiene E 20 local optimization 19 loop optimization s s s 20 loop unrolling e seseeeseeeeeeeeeeee 20 24 78 T O AE E E A Rac ae 85 OD sek ei ae AA AEO E ee 22 ROL EE E A EN AAA E eases 22 I EA E E E E 22 BOB E E E E we etter es 22 Olevelernioi a A atin eee 22 parallelization 20 31 POE S 3 040 Sse A E PERRE aT EERS 77 VECCOFIZATION 0 0 eeeceeceeeceesseeeeeeeeees 20 25 Parallelization 0 0 eeeeeeeceseeeeetecneeeeeneeeres 31 auto parallelization ce eeeeeeeeeeeneees 31 failed auto parallelization 34 81 Mconcur auto parallelization 74 NCPUS environment variable 32 safe lastvalcscncacicisew danced 36 user directe dd ice iei 84 Index Parallelization Directives 0 111 Parallelization Pragmas cecee 129 Pragmas GC E REEE ATT 9 Omp barrier a 65 cece KARA ete 138 optimization rr 155 SCOPE aE NE E A RNE A
100. PA linker has not yet propagated IPA information back to the compiler If the object files are removed and the program re linked IPA information in the ipa files is propagated to enable interprocedural optimizations in each generated object file rm 0 a out make a out OPT Mipa rm During the link phase the IPA linker detects that no IPA information has changed Finally the executable linker is invoked to build the executable properly optimized using IPA information Assume file c is edited and the program is rebuilt There are several possible sources of error One would be to rebuild without IPA enabled make a out Changing file c may result in changes to values of arguments passed to routines in the other source files resulting in stale file2 0 and file3 o object files Assume the program is rebuilt with IPA enabled 40 Chapter 2 make a out OPT Mipa rm This will cause file c to be recompiled and will invoke the IPA linker Even though ile c has changed the IPA information may or may not have changed If not the IPA linker succeeds and the executable linker generates a correct executable file If the IPA information has changed say for file2 c the IPA linker will issue a message and because the m sub option was specified the object file2 0 will be removed Any stale objects can then be rebuilt with one more make command make a out OPT Mipa rm Note that only two make commands ar
101. PGI User s Guide Parallel Fortran C and C for Scientists and Engineers The Portland Group Compiler Technology STMicroelectronics 9150 SW Pioneer Court Suite H Wilsonville OR 97070 http www pgroup com While every precaution has been taken in the preparation of this document The Portland Group Compiler Technology Microelectronics makes no warranty for the use of its products and assumes no responsibility for any errors that may appear or for damages resulting from the use of the information contained herein The Portland Group Compiler Technology Microelectronics retains the right to make changes to this information at any time without notice The software described in this document is distributed under license from The Portland Group Compiler Technology STMicroelectronics and may be used or copied only in accordance with the terms of the license agreement No part of this document may be reproduced or transmitted in any form or by any means for any purpose other than the purchaser s personal use without the express written permission of The Portland Group Compiler Technology STMicroelectronics Many of the designations used by manufacturers and sellers to distinguish their products are claimed as trademarks Where those designations appear in this manual The Portland Group Compiler Technology Microelectronics was aware of a trademark claim The designations have been printed in caps or initial caps Thanks is g
102. PGIS pgf90 hello f Linking PGIS By default the executable output is placed in the file a out a exe on Win32 Use the o option to specify an output file name To place the executable output in the file Aello PGIS pgf90 o hello hello f Linking PGIS To execute the resulting program simply type the filename at the command prompt and press Return PGI hello hello PGIS 1 2 1 Command line Syntax The command line syntax using pgf90 as an example is pgf90 options path filename 8 Chapter 1 Where options path filename is One or more command line options Case is significant for options and their arguments The compiler drivers recognize characters preceded by a hyphen as command line options For example the Mlist option specifies that the compiler creates a listing file in the text of this manual we show command line options using a dash instead of a hyphen for example Mlist In addition the pgcc command recognizes a group of characters preceded by a plus sign as command line options The order of options and the filename is not fixed That is you can place options before and after the filename argument on the command line However the placement of some options is significant for example the option If two or more options contradict each other the last one in the command line takes precedence is the pathname to the directory containing the file named by filena
103. R EQV or NEQV 5 14 FLUSH The OpenMP FLUSH directive uses the following syntax SOMP FLUSH list The FLUSH directive ensures that all processor visible data items or only those specified in ist when it s present are written back to memory at the point at which the directive appears 5 15 THREADPRIVATE The OpenMP THREADPRIVATE directive uses the following syntax SOMP THREADPRIVATE common_block1 common_block2 Where common_blockn is the name of a common block to be made private to each thread but global within the thread This directive must appear in the declarations section of a program unit after the declaration of any common blocks listed On entry to a parallel region data in a THREADPRIVATE common block is undefined unless COPYIN is specified on the PARALLEL directive When a common block that is initialized using DATA statements appears in a THREADPRIVATE directive each thread s copy is initialized once prior to its first use The following restrictions apply to the THREADPRIVATE directive e The THREADPRIVATE directive must appear after every declaration of a thread private common block e Only named common blocks can be made thread private e It is illegal for a THREADPRIVATE common block or its constituent variables to appear in any clause other than a COPYIN clause 5 16 Run time Library Routines User callable functions are ava
104. S SOMP SECTION lt Fortran code block executed by processor i gt SOMP SECTION lt Fortran code block executed by processor j gt SOMP END SECTIONS NOWAIT Clauses PRIVATE list SHARED list DEFAULT PRIVATE SHARED NONE FIRSTPRIVATE list LASTPRIVATE list REDUCTION operator intrinsic list COPYIN list IF scalar _logical_expression The PARALLEL SECTIONS END SECTIONS directives define a non iterative work sharing construct without the need to define an enclosing parallel region Each section is executed by a single processor If there are more processors than sections some processors will have no work OpenMP Directives for Fortran 123 and will jump to the implied barrier at the end of the construct If there are more sections than processors one or more processors will execute more than one section A SECTION directive may only appear within the lexical extent of the enclosing PARALLEL SECTIONS END SECTIONS directives In addition the code within the PARALLEL SECTIONS D SECTIONS directives must be a structured block and the code in each SECTION must be a structured block The available clauses are as defined in Section 5 2 PARALLEL END PARALLEL and Section 5 6 DO END DO 5 12 ORDERED The OpenMP ORDERED directive is supported using the following syntax SOMP ORDERED lt Fortran
105. The caller must then remove the value from the fpu stack even if it doesn t use the value Failure of either side to meet its obligations leads to undefined program behavior The standard calling sequence does not include any method to detect such failures nor to detect return value type mismatches Therefore the user must declare all functions properly There is no difference in the representation of single double or extended precision values in floating point registers e Functions that return no value also called procedures or void functions put no particular value in any register e A call instruction pushes the address of the next instruction the return address onto the stack The return instruction pops the address off the stack and effectively continues execution at the next instruction after the call instruction A function that returns a scalar or no value must preserve the caller s registers as described above Additionally the called function must remove the return address from the stack leaving the stack pointer esp with the value it had before the call instruction was executed Functions Returning Structures or Unions If a function returns a structure of union then the caller provides space for the return value and places its address on the stack as argument word zero In effect this address becomes a hidden first argument Having the caller supply the return objects spaces allows re entrancy 220 Appendix A A fun
106. a computed on using SSE instructions are aligned to cache line boundaries To ensure alignment of local arrays and common blocks the Mcache_align switch can be used Using this switch in combination with those used previously results in the following pgf90 fast Mvect ss Mcache align Minfo vadd f vector op 4 Unrolling inner loop 8 times Loop unrolled 7 times completely unrolled loop 18 Generating sse code for inner loop Generated prefetch instructions for 3 loads Note that the same informational messages are emitted Executing this version of the code you should see results similar to the following o bin time a out 1 000000 771 000 3618 00 6498 00 9999 0 6 52user 0 00system 0 06 52elapsed 100 CPU The result is a speed up of 15 over the equivalent scalar i e non SSE version of the program Speed up realized by a given loop or program can vary widely based on a number of factors e Performance improvement using vector SSE or SSE2 instructions is most effective when the vectors of data are resident in the data cache e If data is aligned properly performance will be better in general than when using vector SSE operations on unaligned data e Ifthe compiler can guarantee that data is aligned properly even more efficient sequences of SSE instructions can be generated 30 Chapter 2 e SSE2 vector instructions can operate on 4 single precision elements concurrently but only 2 double precision el
107. abled or disabled independently and implicit inclusion is still useful when automatic instantiation is not done The implicit inclusion mode can be turned on or off using the implicit_include and no_implicit_include command line options Implicit inclusions are only performed during the normal compilation of a file i e not when doing only preprocessing A common means of investigating certain kinds of problems is to produce a preprocessed source file that can be inspected When using implicit inclusion it is sometimes desirable for the preprocessed source file to include any implicitly included files This may be done using the no_preproc_only command line option This causes the preprocessed output to be generated as part of a normal compilation When implicit inclusion is being used the implicitly included files will appear as part of the preprocessed output in the precise location at which they were included in the compilation C Template Instantiation 209 11 5 Template Libraries Template libraries contain template definitions Special care must be taken when building template libraries that reference other template libraries Since template instantiation occurs at link time and libraries are archived rather than linked the user must invoke a special pre link step to instantiate templates that are directly referenced in the library code The command line flags one_instantiation_per_object template_dir and prelink_object
108. achronism by cfront 2 1 e A type qualifier is allowed but ignored on the declaration of a constructor or destructor For example class A A const No error in cfront 2 1 mode C 6 cfront 2 1 3 0 Compatibility Mode The following extensions are accepted in both cfront 2 1 and cfront 3 0 compatibility mode i e these are features or problems that exist in both cfront 2 1 and 3 0 e Type qualifiers on the this parameter may to be dropped in contexts such as this example struct A void f const void A fp amp A f This is actually a safe operation A pointer to a const function may be put into a pointer to non const because a call using the pointer is permitted to modify the object and the function pointed to will actually not modify the object The opposite assignment would not be safe e Conversion operators specifying conversion to void are allowed C Dialect Supported 265 e A nonstandard friend declaration may introduce a new type A friend declaration that omits the elaborated type specifier is allowed in default mode but in cfront mode the declaration is also allowed to introduce a new type name struct A friend B e The third operator of the operator is a conditional expression instead of an assignment expression as it is in the current X3J16 WG21 Working Paper e A reference to a pointer type may be initialized from a pointer value without use of a temporary even when the refe
109. acted library of functions can have other functions inlined into the library Using both options enables you to obtain more than one level of inlining In this situation if you do not specify a library with the Minline option the inline process consists of two extract passes The first pass is a hidden pass implied by the Minline option during which the compiler extracts functions and places them into a temporary library The second pass uses the results of the first pass but puts its results into the library that you specify with the o option 4 2 1 Working with Inline Libraries An inline library is implemented as a directory with each inline function in the library stored as a file using an encoded form of the inlinable function A special file named TOC in the inline library directory serves as a table of contents for the inline library This is a printable ASCII file which can be examined to find out information about the library contents such as names and sizes of functions the source file from which they were extracted the version number of the extractor which created the entry etc Libraries and their elements can be manipulated using ordinary system commands e Inline libraries can be copied or renamed e Elements of libraries can be deleted or copied from one library to another Function Inlining 107 e The s command can be used to determine the last change date of a library entry Dependencies in Makefiles When a
110. adjusts threshold to which n unrolling applies v is a constant a loop to which n unrolling applies is unrolled v times The directives unroll and nounrol1 only apply if Munroll is selected on the command line vector novector The directive novector is used to disable vectorization vector and novector only apply if Mvect has been selected on the command line vintr novintr The directive novintr directs the vectorizer to disable recognition of vector intrinsics The Mvect transform option always disables vector intrinsic recognition The directive norecog takes precedence over vintr The directive vintr affects the compiler only when Mvect is specified 7 3 Scope of Directives and Command Line options This section presents several examples showing the effect of directives and the scope of directives Remember that during compilation the effect of a directive may be to either turn an option on or turn an option off Directives apply to the section of code following the directive corresponding to the specified scope that is the following loop the following routine or the rest of the program Consider the following code integer maxtime time parameter n 1000 maxtime 10 double precision a n n b n n c n n do time 1 maxtime do i il n do j 1 c i j a i j b i j enddo enddo enddo end When compiled with Mvect both interior loops are interchanged with the outer loop pgf90 Mvect
111. age but operations use 64 bits instead of 32 bits 58 Chapter 3 Requests that the compiler provide special compilation semantics Syntax Kflag Where flag is one of the following ieee noieee PIC pic Perform floating point operations in strict conformance with the IEEE 754 standard Some optimizations are disabled and on some systems a more accurate math library is linked if Kieee is used during the link step Use the fastest available means to perform floating point operations link in faster non IEEE libraries if available and disable underflow traps Linux only Generate position independent code Equivalent to pic Provided for compatibility with other compilers Linux only Generate position independent code Equivalent to fpic Provided for compatibility with other compilers trap option option Command line Options Controls the behavior of the processor when floating point exceptions occur Possible options include Jp align ignored inv denorm divz ovf unf inexact Ktrap is only processed by the compilers when compiling main functions programs The options inv denorm divz ovf unf and inexact correspond to the processor s exception mask bits invalid operation denormalized operand divide by zero overflow underflow and precision respectively Normally the processor s exception mask bits are 59 on floating point exceptions are
112. al only enough space is set aside for a pointer to the actual storage which appears later Fortran C and C Data Types 179 Inthe case of a non virtual direct base class enough storage is set aside for its own non virtual base classes its virtual base class pointers its own fields and its virtual function information but no space is allocated for its virtual base classes e Next storage for the class s own fields e Next storage for virtual function information typically a pointer to a virtual function table e Finally storage for its virtual base classes with space enough in each case for its own non virtual base classes virtual base class pointers fields and virtual function information 9 2 4 Aggregate Alignment The alignment of an array a structure or union an aggregate affects how much space the object occupies and how efficiently the processor can address members Arrays use the alignment of their members Arrays align according to the alignment of the array elements For example an array of short data type aligns on a 2 byte boundary structures and unions align according to the most restrictive alignment of the enclosing members For example the union un1 below aligns on a 4 byte boundary since the alignment of c the most restrictive element is four union unl short a 2 bytes char b 1 byte int c 4 bytes l Structure alignment can result in unused space called paddin
113. allel region is the same as a single parallel region By default the value returned by this function is equal to the value of the environment variable OMP_NUM_ THREADS or to the value set by the last previous call to the omp_set_num threads function defined below include lt omp h gt void omp_set_num_threads int num_threads sets the number of threads to use for the next parallel region This function can only be called from a serial region of code If it is called from within a parallel region or within a function that is called from within a parallel region the results are undefined This function has precedence over the OMP_NUM_THREADS environment variable include lt omp h gt int omp_get thread _num void returns the thread number within the team The thread number lies between 0 and omp_get num threads 1 When called from a serial region this function returns 0 A nested parallel region is the same as a single parallel region include lt omp h gt int omp_get_max threads void returns the maximum value that can be returned by calls to omp_get_num_threads If omp set num threads is used to change the number of processors subsequent calls to omp_get_max threads will return the new value This function returns the maximum value whether executing from a parallel or serial region of code include lt omp h gt int omp_get_num procs void returns the number of processors that are available to the
114. ally significant If the program does not execute for five seconds increase the iteration count of some internal loops or try to place a loop around the main body of the program to extend execution time 42 Chapter 2 e Timing should eliminate or reduce the amount of system level activities such as program loading and I O and task switching e Use one of the 3F timing routines if available or a similar call available on your system or use the SECNDS pre declared function in PGF77 or PGF90 or the SYSTEM CLOCK intrinsic in PGF 90 or PGHPF Example 2 4 shows a fragment that indicates how to use SYSTEM_CLOCK effectively within either an HPF or F90 program unit integer nprocs hz clock0 clockl real time integer allocatable t hpf S distribute t cyclic if defined HPF allocate t number _of processors elif defined _OPENMP allocate t OMP_GET_NUM THREADS else allocate t 1 endif call system_clock count_rate hz call system_clock count clock0 lt do work gt call system_clock count clock1 t clockl clock0O time real sum t real hz size t Example 2 4 Using sYSTEM_CLOCK Optimization amp Parallelization 43 Chapter 3 Command Line Options This chapter describes the syntax and operation of each compiler option The options are arranged in alphabetical order On a command line options need to be preceded by a hyphen If the compiler does not recognize an
115. andard calling sequence These registers must be empty before entry and upon exit from a function The flags register contains the system flags such as the direction flag and the carry flag The direction flag must be set to the forward i e zero direction before entry and upon exit from a function Other user flags have no specified role in the standard calling sequence and are not reserved 219 Floating Point Control Word The control word contains the floating point flags such as the rounding mode and exception masking This register is initialized at process initialization time and its value must be preserved Signals can interrupt processes Functions called during signal handling have no unusual restriction on their use of registers Moreover if a signal handling function returns the process resumes its original execution path with registers restored to their original values Thus programs and compilers may freely use all registers without danger of signal handlers changing their values A 3 Functions Returning Scalars or No Value Functions can perform as follows e A function that returns an integral or pointer value places its result in register eax e A function that returns a long long integer value places its result in the registers 3edx and eax The most significant word is placed in edx and the least significant word is placed in Seax e A floating point return value appears on the top of the fpu register stack
116. anged Chapter 3 norm don t remove o files even if IPA information has changed safe assume missing functions are safe vestigial remove uncalled vestigial functions Mnoi4 pg 77 and pgf 90 only the compiler treats INTEGER variables as INTEGER 2 Mnoframe Eliminates operations that set up a true stack frame pointer for every function With this option enabled you cannot perform a traceback on the generated code and you cannot access local variables Mrs pgf77 pgf90 and pghpf only the compiler promotes REAL variables and constants to DOUBLE PRECISION spectively DOUBLE PRECISION elements are 8 bytes in length variables and constants re Mnors pg 77 pgf90 and pghpf only the compiler does not promote REAL variables and constants to DOUBLE PRECISION REAL variables will be single precision 4 bytes in length Mr8intrinsics pg 77 and pgf 90 only the compiler treats the intrinsics CMPLX and MnorSintrinsics REAL as DCMPLX and DBLE respectively pg 77 and pgf 90 only the compiler does not promote the intrinsics CMPLX and RI Msafeptr option option EAL to DCMPLX and DBLI E respectively pgcc and pgCC only instructs the C C compiler to override data dependencies between pointers of a given storage class Possible values of option include arg instructs the compiler that arrays
117. asm byteorder h stddef h and asm posix_types h Also PGI s version of stdarg h should support changes in newer versions of Linux If you are using the PGCC C or C compilers please make sure that the supplied versions of these include files are found before the system versions This will happen by default unless you explicitly add a option that references one of the system include directories 1 5 2 Running Parallel Programs on Linux The PGI compilers use the LinuxThreads package by Xavier Leroy for SMP parallelism on Linux Because of potential incompatibilities with installed versions of this package PGI now ships a custom version of the library in PG inux86 lib ibpgthread a This version of the library will be included by default before any system version of the library This library is used only when Mconcur or mp is specified on the command line There have been problems with some parallel programs on Linux systems when the per thread stack size is set to the default 2MB If you have unexplained failures please try setting the environment variable MPSTKZ to a larger value such as 8MB This can be accomplished with the command setenv MPSTKZ 8M in csh or with MPSTKZ 8M export MPSTKZ in bash sh or ksh If your program is still failing you may be encountering the hard 8 MB limit on main process stack sizes in Linux You can work around the problem by issuing the command o limit stacksize unlimited
118. ault use the steps that follow to create object files with position independent code and shared object files that are to include them The following steps describe how to create and use a shared object file Step 1 To create an object file with position independent code compile it with the appropriate PGI compiler using the pic option the PIC Kpic and KPIC options are supported for compatibility with other systems you may have used and are equivalent to fpic For example use the following command to create an object file with position independent code using pgf 90 o pgf90 c fpic tobeshared f Step 2 To produce a shared object file use the appropriate PGI compiler to invoke the linker supplied with your system It is customary to name such files using a so filename extension On Linux this is done by passing the shared option to the linker o pgf90 shared o tobeshared so tobeshared o Note that compilation and generation of the shared object can be performed in one step using both the fpic option and the appropriate option for generation of a shared object file Step 3 To use a shared object file compile and link the program which will reference functions or subroutines in the shared object file using the appropriate PGI compiler and listing the shared object on the link line o pgf90 o myprog myprof f tobeshared so Step 4 You now have an executable myprog which does not include any code from fun
119. ber instructs the inliner to perform number levels of inlining The default number is 1 If you specify both func and number the compiler inlines functions that match the function name or have number or fewer statements For examples of inlining functions see Chapter 4 Function Inlining Usage In the following example the compiler extracts functions that have 500 or fewer statements from the source file myprog f and saves them in the file extract il pgf90 Mextract 500 oextract il myprog f In the following example the compiler inlines functions with fewer than approximately 100 statements in the source file myprog f and writes the executable code in the default output file a out pgf90 Minline size 100 myprog f Cross reference o Fortran Language Controls This section describes the Mpgflag options that affect Fortran language interpretations by the PGI Fortran compilers These options are only valid to the pg 77 pg 90 and pghpf compilation drivers Syntax Mbackslash the compiler treats the backslash as a normal character and not as an escape character in quoted strings Mnobackslash the compiler recognizes a backslash as an escape character in quoted strings in accordance with standard C usage Mdclchk the compiler requires that all program variables be declared Mnodclchk the compiler does not require that all program variables be declared 70 Chapter 3 Mdefaultunit Mnodefaul
120. bs no error message is generated when this occurs If for some reason a non printing character is not treated as a white space character its hex representation is printed in the form dd where each d is a hex digit 0026 Unmatched quote 0027 Illegal integer constant Integer constant is too large for 32 bit word 0028 Illegal real or double precision constant 0029 Illegal hexadecimal constant A hexadecimal constant consists of digits 0 9 and letters A F or a f Any other character in a hexadecimal constant is illegal 0030 Illegal octal constant An octal constant consists of digits 0 7 Any other digit or character in an octal constant is illegal 0031 Illegal data type length specifier for The data type length specifier e g 4 in INTEGER 4 is not a constant expression that is a member of the set of allowed values for this particular data type 0032 Data type length specifier not allowed for The data type length specifier e g 4 in INTEGER 4 is not allowed in the given syntax e g DIMENSION A 10 4 0033 Illegal use of constant A constant was used in an illegal context such as on the left side of an assignment statement or as the target of a data initialization statement Messages 231 0034 Syntax error at or near 0035 Predefined intrinsic loses intrinsic property An intrinsic name was used in a manner inconsistent with the language definition for that intrinsic The compiler based
121. byteswapio option to swap the byte order of data in unformatted Fortran data files on input output When this option is used the order of bytes is swapped in both the data and record control words the latter occurs in unformatted sequential files Specifically this option can be used to convert big endian format data files produced by most RISC workstations and high end servers to the little endian format used on X86 or AMD64 systems on the fly during file reads writes This option assumes that the record layouts of unformatted sequential access and direct access files are the same on the systems Also the assumption is that the IEEE representation is used for floating point numbers In particular the format of unformatted data files produced by PGI Fortran compilers is identical to the format used on Sun and SGI workstations that allows you to read and write unformatted Fortran data files produced on those platforms from a program compiled for an X86 or AMD64 platform using the byteswapio option Default The compiler does not byte swap data on input output Usage The following command line requests byte swapping are performed on input output pgf90 byteswapio myprog f Perform array bounds checking If an array bounds violation occurs when a program is executed an error message describing where the error occurred is printed and the program terminates Usage pgf90 C myprog f The text of the error message includes the name of
122. ce information on standard error where option is one of the following all instructs the compiler to produce all available Minfo information inline instructs the compiler to display information about extracted or inlined functions This option is not useful without either the Mextract or Minline option ipa instructs the compiler to display information about interprocedural optimizations loop instructs the compiler to display information about loops such as information on vectorization opt instructs the compiler to display information about optimization time instructs the compiler to display compilation statistics unroll instructs the compiler to display information about loop unrolling Mneginfo option option instructs the compiler to produce information on standard error where option is one of the following concur instructs the compiler to produce all available information on why loops are not automatically parallelized In particular if a loop is not parallelized due to potential data dependence the variable s that cause the potential dependence will be listed in the Mneginfo messages loop instructs the compiler to produce information on why memory hierarchy optimizations on loops are not performed Minform level instructs the compiler to display error messages at the specified and higher levels where level is one of the following fatal instructs the compiler to display fatal error
123. cesecaecaecseecaeeeaeeeeeeneeeneeserenerensees 147 7 2 Fortran Directive Summa ry ccceccecccesscessescessceesceecesecesecesecssecseecaeeeaeeeaeeeseeeeeeeeneeeneeaees 148 7 3 Scope of Directives and Command Line options cccceseseecceseeereeeeeeeeeeeeeeseeeteenteensees 153 TA Adding Pragmas toG and C rrur cai eet teeccdewe ssc E E tiene ERE ena 154 TS C C t Pragma Summary ocisnionincscreiinsisii iioii e ie a E iaa ie iina 155 7 6 Scope of C C Pragmas and Command Line OptionS ssssesssssesessseeessesersresereessreressese 158 LIBRARIES 163 8 1 Using builtin Math Functions in C C cecccecccsseesseessesseeesceescesecesecaecaecaeecseeeseeeseeneeeereens 163 8 2 Creating and Using Shared Object Files on UNIX 00 eccecccescceseeseceseeeseceeeceeeseesseeeneeneeees 163 8 3 Creating and Using Dynamic Link Libraries on Win32 0 ceeceeeceeseeeteeeceeeeeeeenseeeeeneeees 165 8 5 Using LIBSF on Win e seven ticacba totes A N tai ee ees ieee 171 8 6 LAPACK the BLAS and FFTS orione i a EAE EEE TENOR 171 8 7 The C Standard Template Library cceccececseescesseeesceeeceecesecesecsaeceeeceeeseeeeeeneeensees 171 FORTRAN C AND C DATA TYPES 173 SA UR OSa ta a Data BAA OSEE AE sts A E ak cd caste A ER EER EE AE E ATE 173 Dele Fortran ScalarS eeens enaar an eann E a a a ac dienes asletereuboveens EAEAN EEEE Eeoae 173 9 1 2 FORTRAN 77 Aggregate Data Type Extensions sesessesrseeeiersessrerersessssrsrrersrsesses 175 9 1 3 F
124. ch serves to make the class name distinct from all user names Local class names are encoded as L followed by a number which has no special meaning it s just an identifying number assigned to the class followed by _ followed by the mangled name of the class this is not in the ARM and cfront encodes local class names slightly differently void f class A L1__ 1A This form is used when encoding the local class name as a type It s not necessary to mangle the name of the local class itself unless it s also a nested class C Name Mangling 215 12 2 4 Template Class Name Mangling Template classes have mangled names that encode the arguments of the template template lt class Tl class T2 gt class abc abc lt int int gt x abc pt__3_ ii This describes two template arguments of type int with the total length of template argument list string including the underscore and a fixed string indicates parameterized type as well the name of the class template 216 Chapter 12 Appendix A Run time Environment This appendix describes the programming model supported for compiler code generation including register conventions and common object file formats for X86 and AMD64 processor based systems running a 32 bit operating system It specifically does not address these conventions for AMD64 processor based systems running a 64 bit operating system See the application binary interface ABI for AMD64 processor based systems
125. cies exist the compiler by default assumes that there is a data dependence that in turn may inhibit certain optimizations or vectorizations nodepchk directs the compiler to ignore unknown data dependencies eqvchk noeqvchk When examining data dependencies noeqvchk directs the compiler to ignore any dependencies between variables appearing in EQUIVALENCE statements invarif noinvarif There is no command line option corresponding to this directive Normally the compiler removes certain invariant if constructs from within a loop and places them outside of the loop The directive noinvarif directs the compiler to not move such constructs The directive invarif toggles a previous noinvarif ivdep The ivdep directive is equivalent to the directive nodepchk opt The syntax of this directive is 150 Chapter 7 cpgi lt scope gt opt lt level gt where the optional lt scope gt is r or g and lt leve1 gt is an integer constant representing the optimization level to be used when compiling a subprogram routine scope or all subprograms in a file global scope The opt directive overrides the value specified by the command line option On Istval nolstval There is no command line option corresponding to this directive The compiler determines whether the last values for loop iteration control variables and promoted scalars need to be computed In certain cases the compiler must assume that the last values of these variable
126. code block executed by processor gt SOMP END ORDERED The ORDERED directive can appear only in the dynamic extent of a DO or PARALLEL DO directive that includes the ORDERED clause The code block between the ORDERED END ORDERED directives is executed by only one thread at a time and in the order of the loop iterations This sequentializes the ordered code block while allowing parallel execution of statements outside the code block The following additional restrictions apply to the ORDERED directive e The ORDERED code block must be a structured block It is illegal to branch into or out of the block e A given iteration of a loop with a Do directive cannot execute the same ORDERED directive more than once and cannot execute more than one ORDERED directive 5 13 ATOMIC The OpenMP ATOMIC directive uses following syntax SOMP ATOMIC The ATOMIC directive is semantically equivalent to enclosing the following single statement in a CRITICAL END CRITICAL directive pair The statement must be of one of the following forms e xX x operator expr e x expr operator x e x intrinsic x expr 124 Chapter 5 e x intrinsic expr x where x is a scalar variable of intrinsic type expr is a scalar expression that does not reference x intrinsic is one of MAX MIN IAND IOR or IEOR and operator is one of AND O
127. cription of how to build and execute data parallel HPF programs Note the Mconcur option is valid with the PGHPF compiler and can be used to create hybrid shared distributed memory parallel programs In this manual the first two types of parallel programs are collectively referred to as SMP parallel programs The third type is referred to as a data parallel program or simply as a HPF program 1 4 1 Running SMP Parallel Programs When you execute an SMP parallel program by default it will use only 1 processor If you wish to run on more than one processor set the NCPUS environment variable to the desired number of processors subject to a maximum of 4 for PGI s workstation class products You can set this environment variable by issuing the following command setenv NCPUS lt number gt Getting Started 13 on Linux under csh or with NCPUS lt number gt export NCPUS in sh or ksh or within a BASH for Win32 command window Note If you set NCPUS to a number larger than the number of physical processors your program will execute very slowly A ready made example of an auto parallelizable benchmark is available at the URL ftp ftp pgroup com pub SMP linpack tar Unpack it within a UNIX shell window or a BASH for Win32 command window using the command o tar xvf linpack tar and follow the instructions in the supplied README file In addition to the NCPUS environment variable directive based parallel pro
128. ction that returns a structure or union also sets eax to the value of the original address of the caller s area before it returns Thus when the caller receives control again the address of the returned object resides in register eax and can be used to access the object Both the calling and the called functions must cooperate to pass the return value successfully e The calling function must supply space for the return value and pass its address in the stack frame e The called function must use the address from the frame and copy the return value to the object so supplied e The called function must remove this address from the stack before returning Failure of either side to meet its obligation leads to undefined program behavior The standard function calling sequence does not include any method to detect such failures nor to detect structure and union type mismatches Therefore you must declare the function properly Table A 3 illustrates the stack contents when the function receives control after the call instruction and when the calling function again receives control after the ret instruction Table A 3 Stack Contents for Functions Returning struct union Position After Call After Return Position sesp argument wordn argumentwordn _4n 4 esp Pe eam _ argument word argument word 0 eeap sesp valueaddress undefmea io tees retumaddress The following sections of this appendix
129. ctions or subroutines in tobeshared so but which can be executed and dynamically linked to that code By default when the program is linked to produce myprog no assumptions are made on the location of tobeshared so In order for myprog to execute correctly you must initialize the environment variable LD_LIBRARY_PATH to include the directory containing tobeshared so IfLD_LIBRARY_PATH is already initialized it is important not to overwrite its contents Assuming you have placed tobeshared so ina directory home myusername bin you can initialize LD_LIBRARY_PATH to include that directory and preserve its existing contents as follows setenv LD LIBRARY PATH SLD LIBRARY PATH home myusername bin If you know that tobeshared so will always reside in a specific directory you can create the executable myprog ina form that assumes this using the R link time option For example you can link as follows o pgf90 o myprog myprof f tobeshared so R home myusername bin 164 Chapter 8 Note that there is no space between R and the directory name As with the L option no space can be present If the R option is used it is not necessary to initialize LD LIBRARY PATH In the example above the dynamic linker will always look in home myusername bin to resolve references to tobeshared so By default ifthe LD_LIBRARY_PATH environment variable is not set the linker will only search usr 1ib for shared objects The UNIX command 1dd is
130. custom D Main use this flag and add the object containing the custom DilMain to the link line The latest version of the default D Main used by PGF 90 and PGHPF is included in the Release Notes for each release the PGF90 and PGHPF specific code in this routine must be incorporated into the custom version of Di Main to ensure the appropriate function of your DLL Default For arguments that you do not specify the default miscellaneous options are as follows inform warn nolist nobounds 82 Chapter 3 Usage In the following example the compiler includes Fortran source code with the assembly code pgf90 Manno S myprog f In the following example the compiler displays information about inlined functions with fewer than approximately 20 source lines in the source file myprog f pgf90 Minfo inline Minline 20 myprog f In the following example the assembler does not delete the assembly file myprog s after the assembly pass pgf90 Mkeepasm myprog f In the following example the compiler creates the listing file myprog Ist pgf90 Mlist myprog f In the following example array bounds checking is enabled pgf90 Mbounds myprog f Cross reference m S V v module lt moduledir gt Use the module option to specify a particular directory in which generated intermediate mod files should be placed If the nodule lt moduledir gt option is present and USE statements are present in a comp
131. d AthlonXP compatible processors p7 generate 32 bit code for Pentium 4 and compatible processors See Table P 2 for a concise list of the features of these processors that distinguish them as separate targets when using the PGI compilers and tools Syntax tp k8 32 k8 64 p5 po p7 px Usage In the following example pgf90 sets the target architecture to Pentium 4 pgf90 tp p7 myprog f Default The default style of code generation is auto selected depending on the type of processor on which compilation is performed Undefines a preprocessor macro Use the U option or the undef preprocessor directive to undefine macros Syntax Usymbol Where symbol is a symbolic name Usage The following examples undefine the macro test pgf90 Utest myprog F pgf90 Dtest Utest myprog F Cross reference D Mnostdde Command line Options 93 V release_number Displays additional information including version messages If a release_number is appended the compiler driver will attempt to compile using the specified release instead of the default release There can be no space between V and release_number The specified release must be co installed with the default release and must have a release number greater than or equal to 4 1 the first release for which this functionality is supported Usage The following command line shows the output using the V option o pgf90 V myprog f The follo
132. d in the reduction operation Ifthe operator intrinsic portion of the REDUCTION clause is omitted the default reduction operator is addition 114 Chapter 5 The COPYIN clause applies only to THREADPRIVATE common blocks In the presence of the COPYIN clause data from the master thread s copy of the common block is copied to the threadprivate copies upon entry to the parallel region In the presence of an IF clause the parallel region will be executed in parallel only if the corresponding scalar_logical_expression evaluates to TRUE Otherwise the code within the region will be executed by a single processor regardless of the value of the environment variable OMP_NUM THREADS 5 3 CRITICAL END CRITICAL The OpenMP END CRITICAL directive uses the following syntax SOMP CRITICAL name lt Fortran code executed in body of critical section gt SOMP END CRITICAL name Within a parallel region you may have code that will not execute properly when multiple threads act upon the same sub region of code This is often due to a shared variable that is written and then read again The CRITICAL END CRITICAL directive pair defines a subsection of code within a parallel region referred to as a critical section which will be executed one thread at a time The optional name argument identifies the critical section The first thread to arrive at a critical section will be the first to execut
133. d to processors cyclically For example if there are 3 processors executing a loop processor 0 performs iterations 0 3 6 etc processor performs iterations 1 4 7 etc and processor 2 performs iterations 2 5 8 etc Calls in parallel loops are safe to parallelize Loops containing calls are candidates for parallelization Also no minimum loop count threshold must be satisfied before parallelization will occur and last values of scalars are assumed to be safe Disables parallelization of loops with reductions When linking the Mconcur switch must be specified or unresolved references will result The NCPUS environment variable controls how many processors are used to execute parallelized loops Note this option applies only on shared memory multi processor systems Mcray option option pgf 77 and pgf90 only Force Cray Fortran CF77 compatibility with respect to the listed options Possible values of option include pointer for purposes of optimization it is assumed that pointer based variables do not overlay the storage of any other variable Mdepchk instructs the compiler to assume unresolved data dependencies actually conflict Command line Options 75 76 Mnodepchk Mi4 instructs the compiler to assume potential data dependencies do not conflict However if data dependencies exist this option can produce incorrect code pgf77 and pgf90 only the compiler tr
134. d_chars 100 MODUS ses 2 ATE E wees 83 SP EE eB A EAE E 101 Index 269 SOOM AA EE AEE 101 SPCh dif vised eeaesceeiiedea vacua sieves 101 preinclude oo eeeeseeseeeeeeeeeeteeeees 102 prelink Objects ecceseeeeseeeteeeees 102 OS ath anki nein au aie tee 89 e EE dee hee enka heed 90 sf4 and 8 A 90 SE Cie astscses det cite scatcusesussseeee petivausieituagatess 90 Dc E E E ne henna tate 91 Shared ievsesscecvsestavessvesdcsvteerecscteceneeeseens 91 SHOW icine ete naa tee 91 E EEE A ETE 92 E ELE E E A EE E E 103 Aee a n ao a 92 MPi ooi e a i a test 92 a ser E A E E eeienePeaer ee 93 SUSE PCH Moves entitle E 102 Va a a spent 94 OV E E A E E denevteetaneens 94 AEE E OA 95 OWS ATE E E E E E EE 94 Command line Syntax s sseseesseeeeeseeeesseeee Compilation driver ccceceeseesseeseeeseeeeees Compilers Invoke at command level eee PGCC ANSI res seiselen tae ieee tgs PGCG CHT ted costes wetestneeeser etna eens 270 PGE LF ia TEE watiwneu wads 4 POP YO eosin ese tecteeten eed aii E 4 POPI eee atic ee tii wah 4 COD T AE E eta dteesbeg neath 11 Data Types ninne e canteens 173 Dithelds E E E E 181 C C aggregate alignment 180 C C scalar data types 0 cece 177 CICE SU ea a a Sattte 179 CLE a css E A O 182 C class and object layout 179 CTH classes E 179 DEC Structures cece cece centres 175 DEC Uni0nS 3
135. de that executes on a single thread and is skipped by the other threads There is an implied barrier on exit from a SINGLE END SINGLE section of code unless the optional NOWAIT clause is specified Nested single process sections are ignored Branching into or out of a single process section is not supported PROGRAM SINGLE USE INTEGER A 0 1 INTEGER omp_get_thread_num OMP PARALLEL A omp_get_thread_num omp_get_thread_num OMP SINGLE PRINT YOU SHOULD ONLY SEE THIS ONCE OMP END SINGLE SOMP END PARALLEL PRINT A O A 0 A 1 A 1 END The PRIVATE and FIRSTPRIVATE clauses are as described in Section 5 2 PARALLEL END PARALLEL OpenMP Directives for Fortran 117 5 6 DO END DO The OpenMP DO END DO directive uses the following syntax Syntax SOMP DO Clauses lt Fortran DO loop to be executed in parallel gt SOMP END DO NOWAIT Clauses PRIVATE list FIRSTPRIVATE list LASTPRIVATE list REDUCTION operator intrinsic list SCHEDULE type chunk ORDERED The real purpose of supporting parallel execution is the distribution of work across the available threads You can explicitly manage work distribution with constructs such as IF omp_get_thread_num EQ 0 THEN ELSE IF omp_get_thread_num EQ 1 THEN ENDIF However these constructs are not in the form of directives The DO END DO directive pair
136. der to use this switch This switch can be used to link Visual C compiled program units into PGI main programs on Win32 Default The compiler uses the PGI supplied linker and include files and links against PGI supplied libraries Cross reference msvcrt msvcrt Win32 only Use the msvcrt option to instruct the compiler to use Microsoft s msvcrt d11 at runtime rather than the default crtd11 d11 These files contain the Microsoft C runtime library and the default mingw32 C runtime library respectively It is recommended that you use the msvcrt option in combination with the ms libs option Default The compiler uses crtd11 d11 at runtime Cross reference mslibs 84 Chapter 3 Invokes code optimization at the specified level Syntax O level Where evel is one of the following 0 creates a basic block for each statement Neither scheduling nor global optimization is done To specify this level supply a 0 zero argument to the O option schedules within basic blocks and performs some register allocations but does no global optimization performs all level 1 optimizations and also performs global scalar optimizations such as induction variable elimination and loop invariant movement level three specifies aggressive global optimization This level performs all level one and level two optimizations and enables more aggressive hoisting and scalar replacement optimizations that may or may n
137. describe where arguments appear on the stack The examples are written as if the function prologue described above had been used A 4 Integral and Pointer Arguments As mentioned a function receives all its arguments through the stack the last argument is pushed first In the standard calling sequence the first argument is at offset 8 Sebp the second argument is at offset 12 ebp etc Functions pass all integer valued arguments as words expanding or padding signed or unsigned bytes and halfwords as needed Run time Environment 221 Table A 4 Integral and Pointer Arguments Argument Stack Address A 5 Floating Point Arguments The stack also holds floating point arguments single precision values use one word and double precision use two The example below uses only double precision arguments Table A 5 Floating point Arguments Call Argument Stack Address hi l 414 1 20998610 word 0 1 414 8 Sebp po wor 2 2 414 12_ Sebp Po word 0 2 998010 20 ebp po d woran 2 998010 24 hebpi a o be Oo o ion o o A 6 Structure and Union Arguments As described in the data representation section structures and unions can have byte halfword or word alignment depending on the constituents An argument s size is increased if necessary to make it a multiple of words This may require tail padding depending on the size of the argument To ensure that data in the stack is prope
138. directives as opposed to the entire routine or until the end of the file for g scoped directives as opposed to the entire file altcode noaltcode Instructs the parallelizer to generate alternate scalar code for parallelized loops If altcode is specified the parallelizer determines an appropriate cutoff length and generates scalar code to be executed whenever the loop count is less than or equal to that length The noaltcode directive disables these transformations This directive affects the compiler only when Mconcur is enabled on the command line altcode n concur This directive sets the loop count threshold for parallelization of non reduction loops to n Without this directive the compiler assumes a default of 100 Under this directive innermost loops without reductions are executed in parallel only if their iteration counts exceed n altcode n concurreduction This directive sets the loop count threshold for parallelization of reduction loops to n Without this directive the compiler assumes a default of 200 Under this directive innermost loops with reductions are executed in parallel only if their iteration counts exceed n noaltcode This directive sets the loop count thresholds for parallelization of all innermost loops to 0 assoc noassoc This directive toggles the effects of the Mvect noassoc command line option an Optimization M control By default when scalar reductions are present the vect
139. dirvectl f Directives alter this behavior either globally or on a routine or loop by loop basis To assure that vectorization is not applied use the novector directive with global scope Optimization Directives and Pragmas 153 cpgi g novector integer maxtime time parameter n 1000 maxtime 10 double precision a n n b n n c n n do time 1 maxtime do i 1 n do 7 Tyn eia al AoT enddo enddo enddo end In this version the compiler disables vectorization for the entire source file Another use of the directive scoping mechanism turns an option on or off locally either for a specific procedure or for a specific loop integer maxtime time parameter n 1000 maxtime 10 double precision a n n b n n c n n cpgi l novector do time 1 maxtime do i l n do j c i j a i j b i j enddo enddo enddo end Le Loop level scoping does not apply to nested loops That is the directive only applies to the following loop In this example the directive turns off vector transformations for the top level loop If the outer loop were a timing loop this would be a practical use for a loop scoped directive 7 4 Adding Pragmas to C and C Pragmas may be supplied in a C C source file to provide information to the compiler Like directives in Fortran pragmas alter the effects of certain command line options or default behavior of the compiler many pragmas have a corresponding command line option While
140. e If variable names occur after safe each name is the name of a pointer argument in the current function The named argument is considered to be safe Note that if just one variable name is specified the surrounding parentheses may be omitted There is no command line option corresponding to this pragma safeptr nosafeptr The pragma safeptr directs the compiler to treat pointer variables of the indicated storage class as safe The pragma nosafeptr directs the compiler to treat pointer variables of the indicated storage class as unsafe This pragma alters the effects of the Msafeptr command line option Optimization Directives and Pragmas 157 The syntax of this pragma is pragma scope value where value is no safeptr arg local auto global staticlall Note that the values local and auto are equivalent For example in a file containing multiple functions the command line option Msafeptr might be helpful for one function but can t be used because another function in the file would produce incorrect results In such a file the safeptr pragma used with routine scope could improve performance and produce correct results single nosingle The pragma single directs the compiler not to convert float parameters to double in non prototyped functions This can result in faster code if the program uses only float parameters Note Since ANSI C specifies that routines must convert float parameters to double in non
141. e In the following example if the standard library directory is ib the linker loads the library ib libmylib a in addition to the standard libraries pgf90 myprog f lmylib Mpegflag Selects options for code generation The options are divided into the following categories e Code generation e Environment e Inlining e Fortran Language Controls e C C Language Controls e Optimization e Miscellaneous Command line Options 61 Table 3 3 lists and briefly describes the options alphabetically and includes a field showing the category Table 3 3 M Options Summary pgflag Description Category annotate the assembly code with source code no asmkeyword specifies whether the compiler allows the asm C C keyword in C C source files pgec and pgCc Language only no backslash determines how the backslash character is Fortran treated in quoted strings pg 77 pgf90 and Language pghpf only no bounds specifies whether array bounds checking is Miscellaneous enabled or disabled no builtin Do don t compile with math subroutine builtin Optimization support which causes selected math library routines to be inlined pgcec and pgCC only byteswapio Swap byte order big endian to little endian or Miscellaneous vice versa during I O of Fortran unformatted data cache_align where possible align data objects of size greater Optimization than or equal to 16 bytes on cache line boundaries chkfpstk check fo
142. e and global scope respectively The default scope is surrounded by parentheses The in the scope field indicates that the scope is the code following the pragma until the end of the routine for R scoped pragmas as opposed to the entire routine or until the end of the file for G scoped pragmas as opposed to the entire file Many of the pragmas can be preceded by no The default entry in the table indicates the default of the pragma N A appears if a default does not apply The name of any pragma may be prefixed with m for example Mnoassoc is equivalent to noassoc and Mvintr is equivalent to vintr The section following the table provides brief descriptions of the pragmas that are unique to C C Pragmas that have a corresponding directive in Fortran are described in Section 7 2 Optimization Directives and Pragmas 155 Table 7 2 C C Pragma Summary Function Default Scope altcode Do don t generate scalar code for vector altcode L RG noaltcode regions assoc Do don t perform associative a9SQC L RG noassoc transformations bounds Do don t perform array bounds checking nobounds R G nobounds concur Do don t enable auto concurrentization of concur L RG noconcur loops depchk Do don t ignore potential data depchk L RG nodepchk dependencies fcon Do don t assume unsuffixed real constants nofcon R G nofcon are single precision invarif Do don t remove invariant if constructs invarif L RG noinvarif from l
143. e declarations would be CHARACTER ARRAYA 8 4 and INTEGER ARRAYB 4 8 0047 More than seven dimensions specified for array 0048 Illegal use of in declaration of array An asterisk may be used only as the upper bound of the last dimension Messages 233 0049 Illegal use of in non subroutine subprogram The alternate return specifier is legal only in the subroutine statement Programs functions and block data are not allowed to have alternate return specifiers 0050 Adjustable or assumed size array is not a dummy argument o 0051 Unrecognized built in function The allowable built in functions are VAL REF LOC and FILL One was encountered that did not match one of these allowed forms 0052 Illegal argument to VAL or LOC 0053 SREF or VAL not legal in this context The built in functions REF and VAL can only be used as actual parameters in procedure calls 0054 Implicit character used in a previous implicit statement An implicit character has been given an implied data type more than once The implied data type for the implicit character is changed anyway 0055 Multiple implicit none statements The IMPLICIT NONE statement can occur only once in a subprogram 0056 Implicit type declaration The Mdclchk switch and an implicit declaration following an IMPLICIT NONE statement will produce a warning message for IMPLICIT statements 0057 Illegal equivalence of dummy variable Dummy arg
144. e pgf90 prints the execution times for the various compilation steps pgf90 time myprog f Cross reference Set the target architecture By default the PGI compilers produce code specifically targeted to the type of processor on which the compilation is performed In particular the default is to use all supported instructions wherever possible when compiling on a given system As a result executables created on a given system may not be useable on previous generation systems for example executables created on a Pentium 4 may fail to execute on a Pentium III or Pentium II Processor specific optimizations can be specified or limited explicitly by using the p option In this way it is possible to create executables that are useable on previous generation systems With the exception of k8 64 any of these sub options are valid on any X86 or AMD64 processor based system The k8 64 sub option is valid only on AMD64 processor based systems running a 64 bit operating system Following is a list of possible sub options to tp and the processors they are intended to target k8 32 generate 32 bit code for AMD AthlonXP and compatible processors k8 64 generate 64 bit code for AMD Athlon64 AMD Opteron and compatible processors 92 Chapter 3 px generate 32 bit code that is useable on any X86 processor based system ps generate 32 bit code for Pentium and Athlon compatible processors p generate 32 bit code for Pentium Pro II II an
145. e C language definition are accepted by the AT amp T C Language System cfront release 3 0 This option also enables acceptance of anachronisms Default The compiler does not accept cfront language constructs that are not part of the C language definition Usage In the following example the compiler accepts cfront constructs pgCC cfront_3 0 myprog cc Cross references cfront2 1 b b3 p A Command line Options 97 create_pch filename egCC only If other conditions are satisfied create a precompiled header file with the specified name If pch automatic PCH mode appears on the command line following this option its effect is erased diag_ suppress tag egCC only Override the normal error severity of the specified diagnostic messages The message s may be specified using a mnemonic error fag or using an error number diag_ remark tag egCC only Override the normal error severity of the specified diagnostic messages The message s may be specified using a mnemonic error fag or using an error number diag_ warning tag egCC only Override the normal error severity of the specified diagnostic messages The message s may be specified using a mnemonic error fag or using an error number diag error tag egCC only Override the normal error severity of the specified diagnostic messages The message s may be specified using a mnemonic error fag or using an error number
146. e PGI Compilers The file ftp ftp pgroup com pub SMP fftpde tar gz contains a more advanced self guided tutorial on how to parallelize the NAS FT fast Fourier transform benchmark using OpenMP directives You can retrieve it using a web browser and unpack it using the following commands within a UNIX shell window or a BASH for Win32 command window gunzip fftpde tar gz tar xvf fftpde tar ae ol Follow the instructions in the README file to work through the tutorial 5 1 Parallelization Directives Parallelization directives are comments in a program that are interpreted by the PGI Fortran compilers when the option mp is specified on the command line The form of a parallelization directive is sentinel directive _name clauses With the exception of the SGI compatible DOACROSS directive the sentinel must be OMP CSOMP or OMP must start in column 1 one and must appear as a single word without embedded white space The sentinel marking a DOACROSS directive is C Standard Fortran syntax restrictions line length case insensitivity etc apply to the directive line Initial directive lines must have a space or zero in column six and continuation directive lines must have a character OpenMP Directives for Fortran 111 other than space or zero in column six Continuation lines for CSDOACROSS directives are specified using the CS sentinel The order in which clauses appear in the parallelization directives is not significant
147. e amounts of thread local storage in the form of private variables or local variables in functions or subroutines called within parallel regions The value should be an integer lt n gt concatenated with M or m to specify stack sizes of n megabytes For example setenv MPSTKZ 8M Optimization MP Pragmas for C and C 145 Chapter 7 Optimization Directives and Pragmas Directives are Fortran comments that the user may supply in a Fortran source file to provide information to the compiler Directives alter the effects of certain command line options or default behavior of the compiler While a command line option affects the entire source file that is being compiled directives apply or disable the effects of a command line option to selected subprograms or to selected loops in the source file for example an optimization Use directives to tune selected routines or loops 7 1 Adding Directives to Fortran Directives may have any of the following forms cpgisg directive cpgiSr directive cpgi 1 directive cpgi directive The c must be in column 1 Either or is allowed in place of c The scope indicator occurs after the this indicator controls the scope of the directive Some directives ignore the scope indicator The valid scopes as shown above are g global indicates the directive applies to the end of the source file t routine indicates the directive applies to the next subprogram 1 loop indicates the directive appli
148. e ever needed since the IPA propagation step is comprehensive 2 8 Default Optimization Levels Table 2 1 shows the interaction between the O g and M lt opf gt options In the table level can be 0 1 2 or 3 and lt opt gt can be vect unroll or ipa The default optimization level is dependent upon these command line options Table 2 1 Optimization and O g and M lt opt gt Options Optimize Debug M lt opt gt Optimization Level Option Option Option none none none none none M lt opt gt none O 03 noneor g none Optimization amp Parallelization 41 Unoptimized code compiled using the option O0 can be significantly slower than code generated at other optimization levels The M lt opt gt option where lt opt gt is vect concur unroll or ipa sets the optimization level to level 2 if no O options are supplied The fast option sets the optimization level to a target dependent optimization level if no O options are supplied 2 9 Local Optimization Using Directives and Pragmas Command line options let you specify optimizations for an entire source file Directives supplied within a Fortran source file and pragmas supplied within a C or C source file provide information to the compiler and alter the effects of certain command line options or default behavior of the compiler many directives have a corresponding command line option While a command line option affects the entire
149. e functions within the library file file ext Create the library file using the Mextract option If no inline library is specified functions are extracted from a temporary library created during an extract prepass If you specify both a function name and a size n the compiler inlines functions that match the function name or have n or fewer statements If a keyword name lib or size is omitted then a name with a period is assumed to be an inline library a number is assumed to be a size and a name without a period is assumed to be a function name In the following example the compiler inlines functions with fewer than approximately 100 statements in the source file myprog f and writes the executable code in the default output file a out pgf90 Minline size 100 myprog f Refer to Chapter 3 Command Line Options for more information on the Minline options 4 1 1 Using an Inline Library If you specify one or more inline libraries on the command line with the Minline option the compiler does not perform an initial extract pass The compiler selects functions to inline from the specified inline library If you also specify a size or function name all functions in the inline library meeting the selection criteria are selected for inline expansion at points in the source text where they are called If you do not specify a function name or a size limitation for the Minline option the compiler inlines every function in the
150. e of prefetch instructions Default For arguments that you do not specify the default optimization control options are as follows depchk nors i4 nor intrinsics If you do not supply an option to Mvect the compiler uses defaults that are dependent upon the target system Usage In this example the compiler invokes the vectorizer with idiom recognition for Pentium III SSE instructions enabled pgf90 Mvect sse Mcache align myprog f Cross reference g O Mpeflag Miscellaneous Controls Syntax Manno annotate the generated assembly code with source code Command line Options 79 80 Mbounds Mnobounds Mbyteswapio Mchkfpstk Mchkptr Mchkstk Madll enables array bounds checking If an array is an assumed size array the bounds checking only applies to the lower bound If an array bounds violation occurs during execution an error message describing the error is printed and the program terminates The text of the error message includes the name of the array the location where the error occurred the source file and the line number in the source and information about the out of bounds subscript its value its lower and upper bounds and its dimension For example PGFTN F Subscript out of range for array a a f 2 subscript 3 lower bound 1 upper bound 2 dimension 2 disables array bounds checking swap byte order from big endian to little endian or vice versa upon
151. e subject to the SAVE statement the compiler flags non ANSI conforming source code directs the compiler to treat logical values as true if the value is non zero and false if the value is zero UNIX F77 convention When Munixlogical is enabled a logical value or test that is non zero is TRUE and a value or test that is zero is FALSE In addition the value of a logical expression is guaranteed to be one 1 when the result is TRUE directs the compiler to use the VMS convention for logical values for true 71 Mupcase Mnoupcase and false Even values are true and odd values are false the compiler allows uppercase letters in identifiers With Mupcase the identifiers X and x are different and keywords must be in lower case This selection affects the linking process if you compile and link the same source code using Mupcase on one occasion and Mnoupcase on another you may get two different executables depending on whether the source contains uppercase letters The standard libraries are compiled using the default Mnoupcase the compiler converts all identifiers to lower case This selection affects the linking process If you compile and link the same source code using Mupcase on one occasion and Mnoupcase on another you may get two different executables depending on whether the source contains uppercase letters The standard libraries are compiled using Mnoupcase
152. e the assembly and object files created by the pgcc command and the C like file that can be viewed as output from pgcc using the i command line option The name mangling algorithm for the PGCC C compiler is the same as that for cfront and also matches the description in Section 7 2 Function Name Encoding of The Annotated C Reference Manual except for some minor details Refer to the ARM for a complete description of name mangling C Name Mangling 213 12 1 Types of Mangling The following entity names are mangled e Function names including non member function names are mangled to deal with overloading Names of functions with extern C linkage are not mangled e Mangled function names have the function name followed by ___ followed by F followed by the mangled description of the types of the parameters of the function Ifthe function is a member function the mangled form of the class name precedes the F If the member function is static an S also precedes the F int f float _ Ff class A int float __1AFf static int g float g__1ASFE e Special and operator function names like constructors and operator The encoding is similar to that for normal functions but a coded name is used instead of the routine name class A int operator float __pl__1Aff A float Ct IAFF int operator A float __pl__FI1Af e Static data member names The mangled form is the member name followed by _ fo
153. e the code within the section The second thread to arrive will not begin execution of statements in the critical section until the first thread has exited the critical section Likewise each of the remaining threads will wait its turn to execute the statements in the critical section Critical sections cannot be nested and any such specifications are ignored Branching into or out of a critical section is illegal If a name argument appears on a CRITICAL directive the same name must appear on the END CRITICAL directive PROGRAM CRITICAL USE REAL A 100 100 MX LMX INTEGER I J MX 1 0 LMX 1 0 CALL RANDOM_SEED CALL RANDOM NUMBER A SOMP PARALLEL PRIVATE I FIRSTPRIVATE LMX SOMP DO DO J 1 100 OpenMP Directives for Fortran 115 DO I 1 100 LMX MAX A I J LMX ENDDO ENDDO SOMP CRITICAL MX MAX MX LMX SOMP END CRITICAL SOMP END PARALLEL PRINT MAX VALUE OF A IS MX END Note that this program could also be implemented without the critical region by declaring Mx as a reduction variable and performing the MAX calculation in the loop using Mx directly rather than using LMX See Sections 5 2 PARALLEL END PARALLEL and 5 6 DO END DO for more information on how to use the REDUCTION clause on a parallel Do loop 5 4 MASTER END MASTER The OpenMP END MASTER directive uses the following syntax SOMP MASTER lt Fortran code in body of MASTER section gt SOMP END MASTER
154. e to be parallelized it is more efficient to define a single enclosing parallel region and parallelize each loop using the OpenMP Do directive A variable declared PRIVATE or LOCAL to a CSDOACROSS loop is treated the same as a private variable in a parallel region or DO see above A variable declared SHARED or SHARE to a CSDOACROSS loop is shared among the threads meaning that only 1 copy of the variable exists to be used and or modified by all of the threads This is equivalent to the default status of a variable that is not listed as PRIVATE in a parallel region or Do this same default status is used in CSDOACROSS loops as well 5 9 PARALLEL DO The OpenMP PARALLEL Do directive uses the following syntax OpenMP Directives for Fortran 121 Syntax SOMP PARALLEL DO CLAUSES lt Fortran DO loop to be executed in parallel gt SOMP END PARALLEL DO Clauses PRIVATE list SHARED list DEFAULT PRIVATE SHARED NONE FIRSTPRIVATE list LASTPRIVATE list REDUCTION operator COPYIN list IF scalar logical expression SCHEDULE type chunk ORDERED intrinsic list The semantics of the PARALLEL DO directive are identical to those of a parallel region containing only a single parallel Do loop and directive Note that the END PARALLEL DO directive is optional The available clauses are as defined in Section 5 2 PARALLEL END PARALLEL and Section 5 6 DO END DO 5 10 SE
155. e to use the extern c keyword to declare the C function This keeps the C compiler from mangling the name of the function e Youcanuse the cplusplus macro to allow a program or header file to work for both C and C For example the following defines in the header file stdio h allow this file to work for both C and C ifndef _STDIO H define _STDIO H ifdef cplusplus extern C endif _ cplusplus Functions and data types defined ifdef cplusplus endif _ cplusplus endif C member functions cannot be declared extern as their names will always be mangled Therefore C member functions cannot be called from C or Fortran 10 3 Functions and Subroutines Fortran C and C define functions and subroutines differently For a Fortran program calling a C or C function observe the following return value convention e When the C or C function returns a value call it from Fortran as a function and otherwise call it as a subroutine For a C C program calling a Fortran function the call should return a similar type Table 6 1 lists compatible types If the call is to a Fortran subroutine or a Fortran CHARACTER function call it from C C as a function that returns void The exception to this convention is when a Fortran subroutine has alternate returns call such a subroutine from C C as a function returning int whose value is the value of the integer expression specified in the alterna
156. eats INTEGER variables as INT EG ER 4 Mipa lt option gt lt option gt Pass options to the interprocedural analyzer Note Mipa align arg collect const noconst fast globals inline inlimit n localarg noerror ptr pure rm implies O2 and the minimum optimization level that can be specified in combination with Mipa is O2 For example if you specify Mipa O1 on the command line the optimization level will automatically be elevated to O2 by the compiler driver The choices of option are recognize when targets of a pointer dummy are aligned remove arguments replaced by const ptr perform interprocedural analysis and data collection only do not perform any interprocedural code generation optimizations perform interprocedural constant propagation const is default when compiling with Mipa don t perform interprocedural constant propagation equivalent to const ptr vestigial globals localarg optimize references to global variables perform automatic function inlining limit inlining growth to n equivalent to arg plus externalization of local pointer targets normally Mipa will return an error status if there are changes in the IPA information Mipa noerror modifies this so that an error status is never returned perform pointer target disambiguation pure function detection remove o files even if IPA information has ch
157. ectorizer to perform associativity conversions that can change the results of a computation due to roundoff error Mvect noassoc disables this option For example a typical optimization is to change one arithmetic operation to another arithmetic operation that is mathematically correct but can be computationally different and generate faster code This option is provided to enable or disable this transformation since roundoff error for such associativity conversions may produce unacceptable results Cachesize Option The option Mvect cachesize n instructs the vectorizer to tile nested loop operations assuming a data cache size of n bytes By default the vectorizer attempts to tile nested loop operations such as matrix multiply using multi dimensional strip mining techniques to maximize re use of items in the data cache SSE Option The option Mvect sse instructs the vectorizer to automatically generate SSE SSE2 streaming SIMD extensions and prefetch instructions when vectorizable loops are encountered SSE 26 Chapter 2 instructions first introduced on Pentium III and AthlonXP processors operate on 32 bit floating point data and hence apply only to vectorizable loops that operate on 32 bit floating point data SSE2 instructions first introduced on Pentium 4 Xeon and Opteron processors operate on 64 bit floating point data Prefetch instructions first introduced on Pentium HI and AthlonXP processors can be used to improve t
158. ecute parallel regions using only one processor Branching into or out of a parallel region is not supported All other shared memory parallelization pragmas must occur within the scope of a parallel region Nested omp parallel pragmas are not supported and are ignored There is an implicit barrier at 130 Chapter 6 the end of a parallel region When all threads have completed execution of the parallel region a single thread resumes execution of the statements that follow It should be emphasized that by default there is no work distribution in a parallel region Each active thread executes the entire region redundantly until it encounters a directive that specifies work distribution For work distribution see the omp for pragma include lt stdio h gt include lt omp h gt main int a 2 1 1 pragma omp parallel afomp_get_thread_num omp_get_thread_num printf a 0 d a 1 d a 0 a 1 The variables specified ina private list are private to each thread in a team In effect the compiler creates a separate copy of each of these variables for each thread in the team When an assignment to a private variable occurs each thread assigns to its local copy of the variable When operations involving a private variable occur each thread performs the operations using its local copy of the variable Other important points to note about private variables are the following e Variables declared private i
159. ed Intrinsic An attempt was made to redefine an intrinsic function A symbol that represents an intrinsic function may be redefined if that symbol has not been previously verified to be an intrinsic function For example the intrinsic SIN can be defined to be an integer array If a symbol is verified to be an intrinsic function via the INTRINSIC statement or via an intrinsic function reference then it must be referred to as an intrinsic function for the remainder of the program unit Symbol An attempt was made to redefine a symbol that was previously defined An example of this is to declare a symbol to be a PARAMETER that was previously declared to be a subprogram argument 0044 Multiple declaration for symbol An attempt was made to declare a symbol as an entry point when that symbol was previously declared as an entry point or another data type 0045 Data type of entry point disagrees with function The current function has entry points with data types inconsistent with the data type of the current function For example the function returns type character and an entry point returns type complex 0046 Data type length specifier in wrong position A character data type specifier has a different position for the length specifier from the other data types Suppose we want to declare arrays ARRAYA and ARRAYB to have 8 elements each having an element length of 4 bytes The difference is that ARRAYA is character and ARRAYB is integer Th
160. ed This means that if function A calls function B and B calls C and both B and C are inlinable then the version of B which is inlined into A will have had C inlined into it pgf90 dhry f Minline size 10 levels 2 4 5 Restrictions on Inlining The following Fortran subprograms cannot be extracted e Main or BLOCK DATA programs e Subprograms containing alternate return assigned GO TO DATA SAVE or EQUIVALENCE statements Function Inlining 109 Subprograms containing FORMAT statements Subprograms containing multiple entries A Fortran subprogram is not inlined if any of the following applies It is referenced in a statement function A common block mismatch exists i e the caller must contain all common blocks specified in the callee and elements of the common blocks must agree in name order and type except that the caller s common block can have additional members appended to the end of the common block An argument mismatch exists i e the number and type size of actual and formal parameters must be equal A name clash exists e g a call to subroutine xyz in the extracted subprogram and a variable named xyz in the caller The following types of C and C functions cannot be inlined Functions whose return type is a struct data type or functions which have a struct argument Functions containing switch statements Functions which reference a static variable whose definition is nested within the function
161. ed The default is builtin pgcc only Enable compilation of C with compatibility with cfront version 2 1 pgCC only Enable compilation of C with mpatibility with cfront version 3 0 pgCc only Create a precompiled header file with the name filename pgCc only Print makefile dependencies to stdout pgCc only Print makefile dependencies to file filename pgCc only Override the normal error severity of the specified diagnostic messages pgCC only Override the normal error severity of the specified diagnostic messages Q pgCc only Override the normal error severity of the specified diagnostic messages pgCc only Override the normal error severity of the specified diagnostic messages pgCc only Display the error message number in any diagnostic messages that are generated pgCc only Set the C front end error limit to the specified number pgCc only Disable enable exception handling support The default is exceptions pgcc only Allow GNU extensions like include next which are required to compile Linux system header files pgcc only If one_instantiation_per_object is used define dirname as the instantiation directory pgcc only Do don t align long long integers on integer boundaries The default is I lalign Generate make dependence lists Generate make dependence lists 49 50 Option Description MD filename pgCc only Generate make dependence l
162. eeeeeeneeeeenaeenaeenaes 220 A 4 Integral and Pointer Argument cccecccesseesccesecesecesecaeecaeeseecaeecaeeeaeceseeeeeeeeesereneeneenaees 221 A 5 Floating Point Arguments ccccceescessceseceseceseenecseecaeeesecaeecaeecaeeeneseeeseesseneensenseenaeenaes 222 A 6 Structure and Union Argument cccccecccesscesscesceeeeeseceseceaeceeceecaeeeaeeeaeseeeserenseseeeneenaees 222 MESSAGES 227 Bil Dia gnOstic Messages etic ede esos coche feces eel cass EEE A dean dk Eka esheseeh see tees 227 B 2 Phase Invocation Messages ss s 5 sees exseveevesveeeeteba vesveanteee a eines acted 228 B 3 Compiler Error Messages eeina E toe ihes ei eden ds ase esata eee ats 228 B3 1 Message Format ices ciczcctesscsaeccvis ceetedad trceess ves teenecevas teen dad E REER 228 B32 Messages Last AE T A dace des at ttc antabec taestaed te epee bea ets 228 B 4 Runtime Error Messages narenn e ee e E eee nie em Erene 251 BAcT MeSsage Format eccsci ccs ccoveces hast scece ent aioi i e AE E cwearees E 251 B 4 2 Message Listiccccncicrcccstececesciestcescusssancdeqitceieedaciucsusesancvach a e a a e i 251 C DIALECT SUPPORTED 257 C FAnachronisms Acceptednneianconiienn a a aaa nce eee 257 C 2 New Language Features Accepted ssesssessseeeesseseesresetsessteresstsesesessestesteseesessrenesseseesee 258 C 3 The following language features are not accepted ce ecceesceeseeeseeeseceeeceeeseeeeeeeeeeerenseens 261 C 4 Extensions Accepted in Normal C
163. eeeeeseeses 37 2 7 1 Building a Program Without IPA Single Step cc eecesceseceteeseceneceeceseeneeeeeeseeees 37 2 7 2 Building a Program Without IPA Several Steps 0 ccccesceesceseceeeceseceecseeeseeeeeeneeees 38 iv Table of Contents 2 7 3 Building a Program Without IPA Using the make Utility eeceseeeeeseeneeeecneeeeenee 38 2 7 4 Building a Program Using IPA cceccsseesseescesseeeseeseceseeeeeeseceseenaeceaeceaecaeeeaeeeseenseeneeees 39 2 8 Default Optimization Levels iccc cc cc ccdscccecesge vet ceectccteescedetsecceccnsedee dus secdvcevseteadeasuectcessecetceseest 41 2 9 Local Optimization Using Directives and Pragmas 0 0 0 0 cscsessssecseeeeceeeeeceseeeeeseeaeeeeeneeeres 42 2 10 Execution Timing and Instruction Counting ccccccecseescessceeeceseceseceaeceecseeeseeeneeeeeeereees 42 COMMAND LINE OPTIONS 45 3 1 Generic PGI Compiler Options ccecceecceeseesseseceseceesceseenseceseceaecaecaecsecsaecnaecaecaeesseeeneeses 51 3 2 C and C specific Compiler Options ccccceesceescesseesecesecesecseecseeeseeeneeeeecaeeeseenseenseeereeas 95 FUNCTION INLINING 105 4 1 Invoking Function nlining 000 0 eeeeccescceseceeceseceeecseecneeeseeeeeeceseseeeeeeeeenseeneeeeaeenseenaes 105 4 12 Usine an Inline Library ic8 a cve ite eae Stick eon eR ea AS deka are A E 106 4 2 Creating an Inline Library iseci ean ia i EE E a e i 107 4 2 1 Working with Inline Libraries eseeseseesseeeeseeeesseeeesse
164. elease The O or O2 level performs all level one local optimizations as well as global optimizations Control flow analysis is applied and global registers are allocated for all functions and subroutines Loop regions are given special consideration This optimization level is a good choice when the program contains loops the loops are short and the structure of the code is regular The PGI compilers perform many different types of global optimizations including but not limited to e Branch to branch elimination e Constant propagation e Copy propagation e Dead store elimination e Global register allocation e Invariant code motion e Induction variable elimination You select the optimization level on the command line For example level two optimization results in global optimization as shown below pgf90 02 prog f Specifying O on the command line without a level designation is equivalent to O2 The default optimization level changes depending on which options you select on the command line For example when you select the g debugging option the default optimization level is set to level zero O0 However you can override this default by placing Olevel option after g on the command line if you need to debug optimized code Refer to section 2 8 Default Optimization Levels for a description of the default levels As noted above the fast option includes O2 on all X86 and AMD64 targets If you wish to
165. ements As a result the efficiency of loops that operate on single precision data can be higher By careful coding in combination with the Mvect sse and Mcache_align switches it is possible to get substantial speed ups on programs which operate on floating point vectors However in some cases codes which operate on unaligned or strided data will result in performance degradations when compiling with Mvect sse For this reason PGI recommends that you always measure the performance of codes with and without Mvect sse and Mcache_align rather than using these switches as a default for optimization Note Compiling with Mvect sse can result in numerical differences from the generated executable Certain vectorizable operations for example dot products are sensitive to order of operations and the associative transformations necessary to enable vectorization or parallelization 2 6 Auto Parallelization Mconcur With the Mconcur option the compiler scans code searching for loops that are candidates for auto parallelization Mconcur must be used at both compile time and link time When the parallelizer finds opportunities for auto parallelization it parallelizes loops and you are informed of the line or loop being parallelized if the Vinfo option is present on the compile line See Chapter 3 Command Line Options for a complete specification of Mconcur A loop is considered parallelizable if doesn t contain any
166. emove the macro with an undef preprocessor directive or with the U option The compiler processes all of the U options in a command line after processing the D options Syntax Dname value Where name is the symbolic name and value is either an integer value or a character string Default If you define a macro name without specifying a value the preprocessor assigns the string to the macro name Usage In the following example the macro PATHLENGTH has the value 256 until a subsequent compilation If the D option is not used PATHLENGTH s value is set to 128 Command line Options 53 pgf90 DPATHLENGTH 256 myprog F Where the source text is ifndef PATHLENGTH define PATHLENGTH 128 fendif SUBROUTINE SUB CHARACTER PATHLENGTH path END Cross reference U dryrun Use the dryrun option to display the invocations of the compiler assembler and linker but do not execute them These invocations are command lines created by the compilation driver from the PGIRC file and the command line supplied with dryrun Default The compiler does not display individual phase invocations Usage The following command line requests verbose invocation information pgf90 dryrun myprog f Cross reference Minfo V Stops after the preprocessing phase Use the E option to halt the compilation process after the preprocessing phase and display the preprocessed output on the standard out
167. end the compiler accepts 132 column source code Fortran otherwise it accepts 72 column code pgf77 Language pgf90 and pghpf only invokes the function extractor instructs the compiler to treat floating point C C constants as float data types pgcec and pgCc Language only noi4 determines how the compiler treats INTEGER Optimization variables pg 77 pgf 90 and pghpf only info prints informational messages regarding Miscellaneous optimization and code generation to standard output as compilation proceeds inform specifies the minimum level of error severity Miscellaneous that the compiler displays inline invokes the function inline the function inliner Inlining Se E inter procedural analysis and optimization no iomutex determines whether critical sections are Fortran generated around Fortran I O calls pgf 77 Language pgf90 and pghpf only instructs the compiler to keep the assembly file Command line Options 63 Description Category no list specifies whether the compiler creates a listing Miscellaneous file makedll Generate a dynamic link library DLL Win32 Miscellaneous only neginfo instructs the compiler to produce information on Miscellaneous why certain optimizations are not performed noframe eliminates operations that set up a true stack Optimization frame pointer for functions nomain when the link step is called don t include the Code object file that calls
168. ength Parameter Lists Parameter passing in registers can handle a variable number of parameters The C language uses a special method to access variable count parameters The stdarg h and varargs h files define several functions to access these parameters A C routine with variable parameters must use the va_start macro to set up a data structure before the parameters can be used The va_arg macro must be used to access the successive parameters C Parameter Conversion In C for a called prototyped function the parameter type in the called function must match the argument type in the calling function If the called function is not prototyped the calling convention uses the types of the arguments but promotes char or short to int and unsigned char or unsigned short to unsigned int and promotes float to double unless you use the Msingle option For more information on the Msingle option refer to Chapter 3 Command Line Options If the called function is prototyped the unused bits of a register containing a char or Run time Environment 223 short parameter are undefined and the called function must extend the sign of the unused bits when needed Calling Assembly Language Programs 224 File testmain c main long 1_ paral 0x3f 800000 float f_para2 1 0 double d_para3 0 5 float f return extern float sum 3 long paral float para2 double para3 f return sum_3 1_ paral f _para2 d_para3 pr
169. enseenseeseenaecaeceeeaeeeseeaes 222 Table A 6 Structure and Union Arguments 0 cccccesccseesseeseeesceeeeeeeeseceseceaecneceseeeaeenaeenaeeneeenes 223 LIST OF FIGURES Figure 9 1 Internal Padding in a Structure eee eceseeccesecseeeecneeeecsseeecseceeesecnessecsaeeeeeaeeaeeseenees 181 Figure 9 2 Tail Padding in a Structure eecsseecsseeeceseeeeesecseesecneeseceaeeeeeaecaessecneseseesesereeaees 182 xii Table of Contents Preface This guide describes how to use The Portland Group Compiler Technology PGI Fortran C and C compilers and program development tools In particular these include the PGF77 PGF 90 PGHPF PGCC C and PGCC ANSI C compilers the PGPROF profiler and the PGDBG debugger These compilers and tools work in conjunction with a 32 bit X86 Intel PII PHI P4 Xeon AMD Athlon AthlonXP or 64 bit AMD64 AMD Opteron Athlon64 assembler and linker You can use the PGI compilers and tools to compile debug optimize and profile serial uni processor and parallel multi processor applications for X86 or AMD64 processor based systems This PGI User s Guide provides operating instructions for the command level compilation environment and general information about PGI s implementation of the Fortran C and C languages This guide does not teach the Fortran C or C programming languages Audience Description This guide is intended for scientists and engineers using the PGI compilers To use these compilers y
170. er unit 208 NEW specified for file that already exists 209 OLD specified for file that does not exist 210 dynamic memory allocation failed Memory allocation operations occur only in conjunction with namelist I O The most probable cause of fixed buffer overflow is exceeding the maximum number of simultaneously open file units 211 invalid file name 212 invalid unit number A file unit number less than or equal to zero has been specified 252 Appendix B 215 formatted unformatted file conflict Formatted unformatted file operation conflict 217 attempt to read past end of file 219 attempt to read write past end of record For direct access the record to be read written exceeds the specified record length 220 write after last internal record 221 syntax error in format string A runtime encoded format contains a lexical or syntax error 222 unbalanced parentheses in format string 223 illegal P or T edit descriptor value missing 224 illegal Hollerith or character string in format An unknown token type has been found in a format encoded at run time 225 lexical error unknown token type 226 unrecognized edit descriptor letter in format An unexpected Fortran edit descriptor FED was found in a runtime format item 228 end of file reached without finding group 229 end of file reached while processing group 230 scale factor out of range 128 to 127 Fortran P edit descriptor scale factor not within range of 128 to 127 231 error
171. ere are differences a message is generated and the inherit step is disabled This prevents erroneous information from being inherited by another object file A potential difficulty with IPA is that editing a single source file may result in recompilation of many other source files as well since any IPA information propagated to those files is then stale However the compiler and make utility can t detect whether the IPA information is stale until the IPA link step which is too late to effect the necessary recompilations incorporating the new IPA information To enable a simple and effective workaround to this difficulty the Vipa command line option to the STCC compilers includes possibly many sub options of the form Mipa option1 option2 One of the sub options is VMipa rm which causes all object files with stale corresponding ipo files to be removed and prevents the executable link step This allows for minimal recompilations and a correct link to occur with just one more invocation of the make command For example given the makefile above assume the make utility is invoked as follows make a out OPT Mipa rm The source files will be compiled o files will be generated the IPA linker will be invoked and the executable linker will then be invoked to generate a correct executable In this first step the compiler collects all the IPA information However it has not yet used it for any optimizations because the I
172. es 132 Table 7 1 Fortran Directive Summary oein e E E E E AE SEEE 148 Table 7 2 C C Pragma SumMMaty enren ranni ee aE E E E E E aaa 156 Table 9 1 Representation of Fortran Data Types c c cecccessessscesecesecesecseeeseeeaeeeseeseeeeeeeeeesrensees 173 Table 9 2 Real Data Type Ranges sssssssseseressesssssersesseeresstsresreserstsstentssesessestesteseeseeseereseeseese 174 Table 9 3 Scalar Type Alignment sssseseesesseeseeesssessseerssstsrrssesetsteseesessteessestesteseesesseeresseseesss 175 Table 9 4 C C Scalar Data Ty pesi e508 6h occ sei keaton Aa 177 Table9 5 Scalar Alignment seriei iiini ie ses dveyes reeves E E 178 Table 10 1 Fortran and C C Data Type Compatibility 2 0 00 ccceccccecsceeseeeseeeeeeeseeereeteeeseenaees 187 Table 10 2 Fortran and C C Representation of the COMPLEX Type ccscecsseeseeseeeeeeteenees 187 Table 10 3 Calling Conventions Supported by the PGI Fortran Compilers ecseeeeeeereeee 199 Table A T Register Allocation asck seco aA ions hee A ATR Bee see ee aes 217 Table A 2 Standard Stack Frame occ iein ey e O ENR ENO iE 218 Table A 3 Stack Contents for Functions Returning struct union sssesssssesesseseseesersreseereseeseesee 221 Table A 4 Integral and Pointer Arguments 0 ccccessceesceseceeceseeeeecaeecaeeeneeeeeeeeeeeeeneeeneeeerenseees 222 Table of Contents xi Table A 5 Floating point Argument c cccccescceseceseceseeeecseeeseeeeeeecese
173. es that reference other template libraries In previous releases for example libraries that reference templates in the Rogue Wave STL would generate undefined template references Now when the user builds the library object files with the one_instantiation_per_object flag and pre links the object files with the command pgCC one_instantiation_per object prelink objects o the pre linker will instantiate the templates required by the library use_pch filename egCC only Use a precompiled header file of the specified name as part of the current compilation If pch automatic PCH mode appears on the command line following this option its effect is erased no_ using_std egCC only Enable or disable implicit use of the std namespace when standard header files are included Default The default is using_std Usage The following command line disables implicit use of the std namespace pgCC no_using_ std hello cc 102 Chapter 3 pgCC only Control instantiation of template functions Syntax t arg where arg is one of the following all Instantiates all functions whether or not they are used local Instantiates only the functions that are used in this compilation and forces those functions to be local to this compilation Note This may cause multiple copies of local static variables If this occurs the program may not execute correctly none Instantiates no functions this is the defau
174. es to the next loop but not to any loop contained within the loop body Loop scoped directives are only applied to Do loops blank indicates that the default scope for the directive is applied The body of the directive may immediately follow the scope indicator Alternatively any number of blanks may precede the name of the directive Any names in the body of the directive including the directive name may not contain embedded blanks Blanks may surround any special characters such as a comma or an equal sign The directive name including the directive prefix may contain upper or lower case letters case is not significant Case is significant for any variable names that appear in the body of the directive Optimization Directives and Pragmas 147 if the command line option Mupcase is selected For compatibility with other vendors directives the prefix cpgi may be substituted with cdir or cvds 7 2 Fortran Directive Summary Table 7 1 summarizes the supported Fortran directives The scope entry indicates the allowed scope indicators for each directive the default scope is surrounded by parentheses The system field indicates the target system type for which the pragma applies Many of the directives can be preceded by no The default entry in the table indicates the default of the directive n a appears if a default does not apply The name of a directive may also be prefixed with V for example the directive Mbounds is
175. esessteressesttsreseestssteesseeesresteseestesessrereseene 3 CONVEMHIONS essan aa a a a r eee aden ey 4 Related Publications ipp E E anes a 6 GETTING STARTED 7 Te OVERVIEW yee ann E ek Ss A A A ee ee 7 1 2 Invoking the Command level PGI Compilers ccceccecscesseeceesceesceeeceecesecesecsaeenseceeeeeenes 8 1 2 1 Command line Syntax sssrini iieii iiie a e i EEA E ea 8 1 2 2 Command line Options sssesessesseesessreeessestsreseesessteressesttsteseestssteesststeseseestsseentssreressee 9 1 2 3 Fortran Directives and C C Pragimas cccecccsceesseeeeeseeeeceeeceeecesecnseceseceaeeaecneeeneeenes 9 13 Filename Conventions sess sence es sch a a suas tueeweancte dass A AEE rE EARE 10 M3 input Bales fo ecececcs E Reece E N aes Rete E 10 Table of Contents iii TS e2 Output EE Se ctr testectewtteet ahesuhtvteccnee aa hated ei aetl cw octte mate 11 1 4 Parallel Programming Using the PGI Compilers 00 cc ceesescceseeseeeecneeseeseceeeceaeeecaeeneeaeens 13 1 4 1 Running SMP Parallel Programs ccccceesseescessceeseeeeceseceseeesecsaecseecseeeaeceeceeeseeeneeees 13 1 4 2 Running Data Parallel HPF Programs ccccccesscesscesseeseceseceeecseeeaeeeeeeseeeeeeeeteneeeeseenaes 14 1 5 Using the PGI Compilers on Linux cceccccccesecssecsseeeeeseeeeeeeseeseceecensecaecaecasecseecaecseeeneeenes 15 t54 Linux Header Files ss cc chcteons dee au ROR hea aes 15 1 5 2 Running Parallel Programs on Linux ccceceesse
176. esseescessceeeceeceeceseeeceseenseceaeenseeneeeneeenes 16 1 6 Using the PGI Compilers on Win32 ccceeceecseesceeseeesceecesecsecssecseecaeecaeeeaeeseeeeeesesesrenseenaes 16 OPTIMIZATION amp PARALLELIZATION 19 2 1 Overview of OptimmiZations cccciccc ce cee cc ecsevsc cee siei E Ea r e EISi iE i 19 2 2 Getting Started with Optimmizations cceeccesecseesseeseeeseeeeceeeceseceseceseesecaecaecseeceeeeeeneeses 20 2 3 Local and Global Optimization O cccesccesecsseesseeseeeeeeeeeeseceeceseceseeeseenseenaeeeeceeeeeeneeses 22 2 4 Loop Unrollitig Munro ere a ere a eaaa AEI EE iniata ts 24 2 5 Vector zation MYyect aeaee a a aa a a tea apache a ere Seeks 25 2 5 1 Vectorization Sub options ccccecccsseessessecsseeseeeeeeeeeeeecesecnsecesecaecaecaaecseecaeeeseeeaeeneeenes 25 2 5 2 Vectorization Example Using SSE SSE2 Instructions eeeeeeceeeeeeeceeeeeeeseeeeeeeeeees 27 2 6 Auto Parallelization Mconcur eccecccesscceseesseesseeseeeseeeseesecesecesecnseensecesecnaecnaeeseeneeseeeneeses 31 2 6 1 Auto parallelization Sub Options ccceceesseesseeseeeseeeseeeeceseceseensecnaecseecseecaeeeaeeeneenneenes 31 2 6 2 Auto parallelization Example cccccsceesseessessceeeceeeceseceeeesecaeceaecseecaeecseeeaeeseeeeeeneeaes 33 2 6 3 Loops That Fail to Parallelize 22 0 5 ncchs katie Hien Gb ais Ie cde oe 34 2 7 Inter Procedural Analysis and Optimization Mipa cccccccesseesseessceseceseceeenseeseecne
177. esseeseesessreesseserseeees 80 Mceache align ceceseesesteeseeeseeseeees 74 Mech fpstk 0 0 ceeceeeeseeeseeseeereeeseeneeenes 80 Mehk ptt tiiic ciecacconidtenancane iets 80 Mehki nanas detinecies 80 MeoncUr anatina enn daai 74 Meta ives cinersteeccuseceeteee innin e 75 MD AMEE E E E E 99 MEER eene ora aeaa 70 Mdefaultunit seercsninmncnuani 71 Milepehk vs s 2scesiecsceieceuseesleciceeseseeaes 76 Midlines iosscicctessssntabe ceesesastntebsetianeute dens 71 Mil acticin Ra Ret desea tetera tse 80 1197 4 0 1 1 eee nennsnie n 71 73 Mextend canaine 71 MCX ACh arenadan aen 69 Micon 5 4 basen kan led eccte feed sation 73 Misi ccstsdeesee tt tpeetsdtet aattab teense Tabs 76 MIN e ec teehee een en 81 Maintorm rye tsi EORNA 81 Mainline c chc gids AEE 69 Monte oaae r Auk 71 Mipas serranon kien tones 76 Mkeepastitic cucceci hitiee aceesisnieca eases 82 MiStsss ess chests setts certian Aietiesaeauecienett 82 Mimakedll a aA 82 Mneginfo sn eonna 81 Index Mnoasmkeywotd sssr 73 SMonetri pssst ee rekpestetie dee ostiess 71 Mnobackslash cesceeseeseeeeeseeeees 70 MP tethers ose ieee ee 84 Mnobounds eccecseeseeseeeeeeeeeeteeesees 80 lt M pre fle is vicina decceseeteeitacescvensenciesstbecds 61 Mnodclehk ue eceeeeeseeseereeeeeeeeeeees 71 Mpo e 66 Mnodefaultunit nogrann 71 SMES oonan o AEA 77 Mridd pchksvenncn ake 76 MYrBintrinsics i asenin e 77 Mnod lines 00 0 ceeeeseeseeeseeseeeteeeteeeeees 71
178. essions must be the same for 136 all threads executing the loop include lt stdio h gt include lt math h gt main float a 1000 b 1000 int i for i 0 i lt 1000 i b i i pragma omp parallel pragma omp for for i 0 i lt 1000 i a i sqrt b i Chapter 6 The schedule clause specifies how iterations of the for loop are divided up between processors Given a schedule kind chunk clause kind can be static dynamic guided or runtime These are defined as follows When schedule static chunk is specified iterations are allocated in contiguous blocks of size chunk The blocks of iterations are statically assigned to threads in a round robin fashion in order of the thread ID numbers The chunk must be a scalar integer expression If chunk is not specified a default chunk size is chosen equal to number_of iterations omp_num_threads 1 omp_num_threads When schedule dynamic chunk is specified iterations are allocated in contiguous blocks of size chunk As each thread finishes a piece of the iteration space it dynamically obtains the next set of iterations The chunk must be a scalar integer expression If no chunk is specified a default chunk size is chosen equal to 1 When schedule guided chunk is specified the chunk size is reduced in an exponentially decreasing manner with each dispatched piece of the iteration space Chunk specifies the minimum number of iterations to d
179. extracted Cross reference show V flags Command line Options 57 Adds a directory to the search path for files that are included using the INCLUDE statement or the preprocessor directive include Use the option to add a directory to the list of where to search for the included files The compiler searches the directory specified by the option before the default directories Syntax Idirectory Where directory is the name of the directory added to the standard search path for include files Usage The Fortran INCLUDE statement directs the compiler to begin reading from another file The compiler uses two rules to locate the file 1 Ifthe file name specified in the INCLUDE statement includes a path name the compiler begins reading from the file it specifies 2 Ifno path name is provided in the INCLUDE statement the compiler searches in order any directories specified using the option in the order specified the directory containing the source file the current directory For example the compiler applies rule 1 to the following statements INCLUDE INCLUDE pob include filel absolute path name filel relative path name and rule 2 to this statement INCLUDE filel Cross reference Mnostdinc i2 i4 and i8 Treat INTEGER variables as either two four or eight bytes INTEGER 8 values not only occupy 8 bytes of stor
180. f code within a parallel region referred to as a critical section which will be executed one thread at a time The first thread to arrive at a critical section will be the first to execute the code within the section The second thread to arrive will not begin execution of statements in the critical section until the first thread has exited the critical section Likewise each of the remaining threads will wait its turn to execute the statements in the critical section An optional name may be used to identify the critical region Names used to identify critical regions have external linkage and are in a name space separate from the name spaces used by labels tags members and ordinary identifiers Critical sections cannot be nested and any such specifications are ignored Branching into or out of a critical section is illegal include lt stdlib h gt main int a 100 100 mx 1 lmx 1 i j for j 0 3 lt 100 j for i 0 i lt 100 i a i j 1 int 10 0 rand RAND MAX 1 0 pragma omp parallel private i firstprivate lmx pragma omp for for j 0 3 lt 100 j for i 0 i lt 100 i lmx lmx gt a i j Imx afi jl pragma omp critical mx mx gt lmx mx lmx printf max value of a is d n mx 6 4 omp master The OpenMP omp master pragma uses the following syntax pragma omp master lt C C structured block gt Optimization MP Pragmas for C and C 133 In
181. f function template arguments is supported Unnamed template parameters are supported The new lookup rules for member references of the form x A PB and p gt A B are supported The notation template and gt template etc is supported C 3 The following language features are not accepted The following features not in the ARM but in the X3J16 WG21 Working Paper are accepted enum types cannot contain values larger than can be contained in an int reinterpret_cast does not allow casting a pointer to a member of one class to a pointer to a member of another class if the classes are unrelated Two phase name binding in templates as described in temp res and temp dep of the Working Paper is not implemented In a reference of the form f gt g with g a static member function f is not evaluated This is as required by the ARM The WP however requires that f be evaluated Class name injection is not implemented Putting a try catch around the initializers and body of a constructor is not implemented Template parameters are not implemented Koenig lookup of function names on all calls is not implemented Finding friend functions of the argument class types on name lookup on the function name in calls is not implemented String literals do not have const type Universal character set escapes e g uabcd are not implemented C Dialect Supported 261 C 4 Extensions Accepted in Normal C Mode The
182. following extensions are accepted in all modes except when strict ANSI violations are diagnosed as errors see the A option e A friend declaration for a class may omit the class keyword class A friend B Should be friend class B e Constants of scalar type may be defined within classes class A const int size 10 int a size e Inthe declaration of a class member a qualified name may be used struct Af int A f Should be int e The preprocessing symbol c_plusp1lus is defined in addition to the standard __ cplusplus e An assignment operator declared in a derived class with a parameter type matching one of its base classes is treated as a default assignment operator that is such a declaration blocks the implicit generation of a copy assignment operator This is cfront behavior that is known to be relied upon in at least one widely used library Here s an example struct A struct B public A B amp operator A amp By default as well as in cfront compatibility mode there will be no implicit declaration of B operator const B amp whereas in strict ANSI mode B operator A amp is nota copy assignment operator and B operator const B amp is implicitly declared e Implicit type conversion between a pointer to an extern cC function and a pointer to an extern C function is permitted Here s an example extern C void s
183. function has a frame on the run time stack This stack grows downward from high addresses Table A 2 shows the stack frame organization Table A 2 Standard Stack Frame Position Contents Frame ant8_ Sebp 8 sebp argument word 4 sebp retumaddress 4 ebp nbytesoflocal n ebp variablesandtemps Several key points concerning the stack frame e The stack is kept double word aligned for best performance e Argument words are pushed onto the stack in reverse order i e the rightmost argument in a C call syntax has the highest address A dummy word may be pushed ahead of the rightmost argument in order to preserve doubleword alignment All incoming arguments appear on the stack residing in the stack frame of the caller e An argument s size is increased if necessary to make it a multiple of words This may require tail padding depending on the size of the argument 218 Appendix A All registers on an IA system are global and thus visible to both a calling and a called function Registers ebp ebx edi esi and esp belong to the calling function Therefore a called function must preserve these registers values for its caller Remaining registers belong to the called function If a calling function wants to preserve such a register value across a function call it must save a value in its local stack frame Some registers have assigned roles in the standard calling sequence Sesp Sebp
184. g cd double d com 10 6 Argument Passing and Return Values In Fortran arguments are passed by reference i e the address of the argument is passed rather than the argument itself In C C arguments are passed by value except for strings and arrays which are passed by reference Due to the flexibility provided in C C you can work around these differences Solving the parameter passing differences generally involves intelligent use of the amp and operators in argument passing when C C calls Fortran and in argument declarations when Fortran calls C C For strings declared in Fortran as type CHARACTER an argument representing the length of the string is passed to a calling function On UNIX systems or when using the UNIX calling 188 Chapter 10 convention on Win32 Munix the compiler places the length argument s at the end of the parameter list following the other formal arguments The length argument is passed by value not by reference 10 6 1 Passing by Value VAL When passing parameters from a Fortran subprogram to a C C function it is possible to pass by value using the vaL function If you enclose a Fortran parameter with svaAL the parameter is passed by value For example the following call passes the integer I and the logical BVAR by value INTEGER 1 I LOGICAL 1 BVAR CALL CVALUE VAL I sVAL BVAR 10 6 2 Character Return Values Section 70 3 Functions and Subroutines describes t
185. g Padding between members of a structure is called internal padding Padding between the last member and the end of the space occupied by the structure is called tail padding Figure 5 1 illustrates structure alignment Consider the following structure struct strcl char a occupies byte 0 sa A short b occupies bytes 2 and 3 char c occupies byte 4 X int d occupies bytes 8 through 11 180 Chapter 9 byte 0 byte 4 byte 8 Figure 9 1 Internal Padding in a Structure Figure 5 2 below shows how tail padding is applied to a structure aligned on a doubleword boundary SELUCU StERGZ f int m1 4 occupies bytes 0 through 15 double m2 occupies bytes 16 through 23 short m3 occupies bytes 24 and 25 eof St 9 2 5 Bit field Alignment Bit fields have the same size and alignment rules as other aggregates with several additions to these rules Bit fields are allocated from right to left A bit field must entirely reside in a storage unit appropriate for its type Bit fields never cross unit boundaries Bit fields may share a storage unit with other structure union members including members that are not bit fields Unnamed bit field s types do not affect the alignment of a structure or union Items of signed unsigned long long type may not appear in field declarations Fortran C and C Data Types 181 byte 0 byte 4 byte 8 byte 12 byte 16 byte 20
186. grams built using the OpenMP features of PGF77 and PGF90 recognize the OpenMP standard environment variable OMP_NUM_THREADS Initialization and usage are identical to that for NCPUS A ready made example of an OpenMP parallel program is available at the URL ftp ftp pgroup com pub SMP matmul tar Unpack it within a UNIX shell window or a BASH for Win32 command window using the command o tar xvf matmul tar and follow the instructions in the supplied README file In addition to showing the OpenMP capabilities of PGF77 and PGF90 this example also further illustrates auto parallelization and provides a brief glimpse of the capabilities of the PGHPF data parallel compiler on SMP systems 1 4 2 Running Data Parallel HPF Programs When you execute a HPF program by default it will use only one processor If you wish to run on more than one processor use the pghpf np runtime option For example to compile and run the hello f example defined above on one processor you would issue the following commands pghpf o hello hello f Linking 14 Chapter 1 hello hello xe To execute it on two processors you would issue the following commands hello pghpf np 2 hello xe Note If you specify a number larger than the number of physical processors your program will execute very slowly Note that you still only see a single hello printed to your screen This is because HPF is a single threaded model meaning that
187. h the compilers 2 7 1 Building a Program Without IPA Single Step Using the STCC command level C compiler driver three for example source files can be compiled and linked into a single executable with one command o stcc o a out filel c file2 c file3 c In actuality the stcc driver executes several steps to produce the assembly code and object files corresponding to each source file and subsequently link the object files together into a single executable file Thus the command above is equivalent to the following commands performed individually stcc S o filel s filel c asstl o filel o filel s stcc S o file2 s file2 c asstl o file2 o file2 s monn nn Optimization amp Parallelization 37 S stcc S o file3 s file3 c S asstl o file3 0 file3 s stcc o a out filel o file2 o file3 0 lstc If any of the three source files is edited the executable can be rebuilt using the same command line stcc o a out filel c file2 c file3 c This always works as intended but has the side effect of recompiling all of the source files For applications that consist of a large number of source files this can be time consuming and inefficient 2 7 2 Building a Program Without IPA Several Steps It is also possible to type discrete stcc commands to compile each source file into a corresponding object file and one to link the resulting object files into an executable stcc c filel c stcc c file2 c stee c
188. he Munix compilation flag is used meaning you cannot mix UNIX style argument passing and C calling conventions within the same file Syntax for the directive is as follows MSSATTRIBUTES C work Where work is the name of the subroutine to be called using C conventions More than one subroutine may be listed separated by commas See above for a complete description of the implementation of the C calling convention Note The directive prefix DEC is also supported but requires a space between the prefix and the directive keyword ATTRIBUTES The must begin the prefix when compiling using Fortran 90 freeform format The characters C or can be used in place of in either form of the prefix when compiling used fixed form F77 style format The directives are completely case insensitive 10 14 6 Using the UNIX Calling Convention Using the UNIX calling convention is straightforward Any program unit compiled using Vunix compilation flag will use the UNIX convention 202 Chapter 10 Chapter 11 C Template Instantiation A template defines a family of types or functions For example the following code fragment shows a template declaration of a class vect This template declaration can be used to declare vector objects By supplying different types for the parameter T different template class definitions will be instantiated or generated define a template template lt class T gt class vectf private Te as typedef int
189. he compiler may not be able to detect on which iteration of the parallelized loop t is last assigned The compiler detects the above cases Where a scalar is used after the loop but is not defined on every iteration of the loop parallelization will not occur When the programmer knows that the scalar is assigned on the last iteration of the loop the programmer may use a directive or pragma to let the compiler know the loop is safe to parallelize The Fortran directive which tells the compiler that for a given loop the last value computed for all scalars make it safe to parallelize the loop is cpgi l1 safe lastval In addition a command line option Msafe_lastval provides this information for all loops within the routines being compiled essentially providing global scope 2 7 Inter Procedural Analysis and Optimization Mipa The PGI Fortran C and C compilers use a two phase interprocedural analysis IPA process that results in minimal changes to makefiles and the standard edit build run application development cycle This process requires that compilation or more generally the make command be performed twice in order to take advantage of IPA Other than adding the necessary compile and link time command line options no other changes are required to the program development cycle For reference and background the process of building a program without IPA is described below followed by the minor modifications required to use IPA wit
190. he general rules for return values for C C and Fortran inter language calling There is a special return value to consider When a Fortran function returns a character two arguments need to be added at the beginning of the C C calling function s argument list e the address of the return character or characters e the length of the return character Example 10 1 illustrates the extra parameters tmp and 10 supplied by the caller CHARACTER FUNCTION CHF Cl I CHARACTER C1 INTEGER I END extern void chf char tmp 10 char cl 9 int i chf tmp 10 cl amp i 9 Example 10 1 Character Return Parameters Inter language Calling 189 If the Fortran function is declared to return a character value of constant length for example CHARACTER 4 FUNCTION CHF the second extra parameter representing the length must still be supplied but is not used Note The value of the character function is not automatically NULL terminated 10 6 3 Complex Return Values Fortran complex functions return their values in multiple floating point registers consequently you cannot directly call a Fortran complex function from C C It is possible for a C C function to pass a pointer to a memory area to a function which calls the comPLEx function and stores the value returned by the complex function Example 10 2 illustrates COMPLEX return values extern void inter cf typedef struct float real imag cplx cplx c
191. he performance of vectorizable loops that operate on either 32 bit or 64 bit floating point data See table P 2 for a concise list of processors that support SSE SSE2 and prefetch instructions Note Programs units compiled with Mvect sse will not execute on Pentium Pentium Pro Pentium II or first generation AMD Athlon processors They will only execute correctly on Pentium III Pentium 4 Xeon AthlonXP Athlon64 and Opteron systems running an SSE enabled operating system Prefetch Option The option Mvect prefetch instructs the vectorizer to automatically generate prefetch instructions when vectorizable loops are encountered even in cases where SSE or SSE2 instructions are not generated Usually explicit prefetching is not necessary on Pentium 4 Xeon and Opteron because these processors support hardware prefetching nonetheless it sometimes it can be worthwhile to experiment with explicit prefetching Note Program units compiled with Mvect prefetch will not execute correctly on Pentium Pentium Pro or Pentium II processors They will execute correctly only on Pentium III Pentium 4 Xeon AthlonXP or Opteron systems In addition to these sub options to Mvect several other sub options are supported See the description of Mvect in Chapter 7 Command line Options for a detailed description of all available sub options 2 5 2 Vectorization Example Using SSE SSE2 Instructions One of the most important vectorization opt
192. he valid options Cross reference v Command line Options 55 l mha go mi o ie Linux only Generate position independent code suitable for inclusion in shared object dynamically linked library files Cross reference shared G R Hh E A Linux only Equivalent to fpic Provided for compatibility with other compilers Cross reference fpic shared G R Passed to the linker Instructs the linker to produce a shared object dynamically linked library file Cross reference fpic shared R The g option instructs the compiler to include symbolic debugging information in the object module Debuggers such as PGDBG require symbolic debugging information in the object module to display and manipulate program variables and source code Note that including symbolic debugging information increases the size of the object module If you specify the g option on the command line the compiler sets the optimization level to O0 zero unless you specify the O option For more information on the interaction between the g and O options see the O entry Symbolic debugging may give confusing results if an optimization level other than zero is selected Default The compiler does not put debugging information into the object module Usage In the following example the object file a out contains symbolic debugging information pgf90 g myprog f 56 Chapter
193. hereof struct A A int A operator A amp A operator const A amp main A b 1 b A 1 A 2 Allowed as anachronism C 2 New Language Features Accepted The following features not in the ARM but in the X3J16 WG21 Working paper are accepted e The dependent statement of an if while do while or for is considered to be a scope and the restriction on having such a dependent statement be a declaration is removed 258 Appendix C e The expression tested in an if while do while or for as the first operand of a operator or as an operand of the amp amp or operators may have a pointer to member type or a class type that can be converted to a pointer to member type in addition to the scalar cases permitted by the ARM e Qualified names are allowed in elaborated type specifiers e Use of a global scope qualifier in member references of the form x A B and p gt A B e The precedence of the third operand of the operator is changed e If control reaches the end of the main routine and main has an integral return type it is treated as ifa return 0 statement were executed e Pointers to arrays with unknown bounds as parameter types are diagnosed as errors e A functional notation cast of the form A can be used even if A is a class without a nontrivial constructor The temporary created gets the same default initialization to zero as a static object of the class type
194. hreads to use during execution of parallel regions The default value for this variable is 1 For historical reasons the environment variable NCPUS is supported with the same functionality In the event that both OMP_NUM_THREADS and NCPUS are defined the value of OMP_NUM_THREADS takes precedence Note OMP_NUM_THREADS threads will be used to execute the program regardless of the number of physical processors available in the system As a result you can run programs using more threads than physical processors and they will execute correctly However performance of programs executed in this manner can be unpredictable and oftentimes will be inefficient OMP_ SCHEDULE specifies the type of iteration scheduling to use for omp for and omp parallel for loops that include the schedule runtime clause The default value for this variable is static Ifthe optional chunk size is not set a chunk size of 1 is assumed except in the case of a static schedule Fora static schedule the default is as defined in Section 6 6 omp for 144 Chapter 6 Examples of the use of OMP_ SCHEDULE are as follows setenv OMP_SCHEDULE static 5 setenv OMP_SCHEDULE guided 8 setenv OMP_SCHEDULE dynamic OMP_ DYNAMIC currently has no effect OMP_NESTED currently has no effect MPSTKzZ increase the size of the stacks used by threads executing in parallel regions For use with programs that utilize larg
195. ier following a was not a directive name 257 EOF in include directive End of file was encountered while processing a include directive 258 Unmatched elif A elif directive was encountered with no preceding if or elif directive 259 Unmatched else A else directive was encountered with no preceding if or elif directive 260 Unmatched endif A endif directive was encountered with no preceding if ifdef or ifndef 261 Include files nested too deeply The nesting depth of include directives exceeded the maximum currently 20 262 Unterminated macro definition for A newline was encountered in the formal parameter list for the indicated macro 263 Unterminated string or character constant A newline with no preceding backslash was found in a quoted string 264 Possible nested comment The characters were found within a comment 268 Cannot inline subprogram common block mismatch 269 Cannot inline subprogram argument type mismatch This message may be classified as severe if compilation has gone too far to undo the inlining process 250 Appendix B 270 Missing exlib option 271 Can t inline wrong number of arguments 272 Argument of inlined function not used 273 Inline library not specified on command line inlib switch 274 Unable to access file TOC 275 Unable to open file while extracting or inlining 276 Assignment to constant actual parameter in inlined subprogram 277 Inlining of fu
196. iguration file 90 Chapter 3 Default The driver uses the configuration file pgirc Usage In the following example the file pgf90rctest relative to usr pgi linux86 bin the value of DRIVER is the driver configuration file pgf90 rc pgf90rctest myprog f Cross reference show Stops compilation after the compiling phase and writes the assembly language output to the file filename s where the input file is filename f Default The compiler produces an executable file Usage In this example pgf90 produces the file myprog s in the current directory pgf90 S myprog f Cross reference c E F Mkeepasm o shared Valid only on Linux and is passed to the linker Instructs the linker to produce a shared object dynamically linked library file Cross reference fpic G R show Produce driver help information describing the current driver configuration Usage In the following example the driver displays configuration information to the standard output after processing the driver configuration file pgf90 show myprog f Cross reference V v help rc Command line Options 91 silent Do not print warning messages Usage In the following example the driver does not display warning messages pgf90 silent myprog f Cross reference v V w time Print execution times for various compilation steps Usage In the following exampl
197. ilable to the Fortran programmer to query and alter the parallel execution environment integer omp_get_num_ threads OpenMP Directives for Fortran 125 returns the number of threads in the team executing the parallel region from which it is called When called from a serial region this function returns 1 A nested parallel region is the same as a single parallel region By default the value returned by this function is equal to the value of the environment variable OMP_NUM_ THREADS or to the value set by the last previous call to the omp_set_num_ threads subroutine defined below subroutine omp_set_num_threads scalar_ integer exp sets the number of threads to use for the next parallel region This subroutine can only be called from a serial region of code If it is called from within a parallel region or within a subroutine or function that is called from within a parallel region the results are undefined This subroutine has precedence over the OMP_NUM_THREADS environment variable integer omp_get_thread_num returns the thread number within the team The thread number lies between 0 and omp_get_num_ threads 1 When called from a serial region this function returns 0 A nested parallel region is the same as a single parallel region integer function omp_get_max threads returns the maximum value that can be returned by calls to omp_get_num_threads If omp set num threads is used to change the number of proces
198. iled program unit lt moduledir gt will search for mod intermediate files prior to the search in the default local directory Default The compiler places mod files in the current working directory and searches only in the current working directory for pre compiled intermediate mod files Usage The following command line requests that any intermediate module file produced during compilation of myprog f be placed in the directory mymods in particular the file mymods myprog mod will be used pgf90 module mymods myprog f Command line Options 83 Use the mp option to instruct the compiler to interpret user inserted OpenMP shared memory parallel programming directives and generate an executable file which will utilize multiple processors in a shared memory parallel system See Chapter 5 Open MP Directives for Fortran and Chapter 6 OpenMP Pragmas for C and C for a detailed description of this programming model and the associated directives and pragmas Default The compiler ignores user inserted shared memory parallel programming directives and pragmas Usage The following command line requests processing of any shared memory directives present in myprog f pgf90 mp myprog f Cross reference Mconcur and Mvect Win32 only Use the ms ibs option to instruct the compiler to use the Microsoft linker and include files and link against the Microsoft Visual C libraries Microsoft Visual C must be installed in or
199. ing to the operator as specified in table 11 1 Table 6 1 Initialization of reduction Variables Operator Initialization e At the end of the parallel region a reduction is performed on the instances of variables appearing in list using operator as specified in the reduction clause The initial value of each reduction variable is included in the reduction operation If the operator portion of the reduction clause is omitted the default reduction operator is addition e The copyin clause applies only to threadprivate variables In the presence of the copyin clause data from the master thread s copy of the threadprivate variable is copied to the thread private copies upon entry to the parallel region e In the presence of an if clause the parallel region will be executed in parallel only if the corresponding scalar_expression evaluates to a non zero value Otherwise the code within the region will be executed by a single processor regardless of the value of the environment variable OMP_NUM_ THREADS 6 3 omp critical The OpenMP omp critical pragma uses the following syntax pragma omp critical name lt C C structured block gt 132 Chapter 6 Within a parallel region there may exist subregions of code that will not execute properly when executed by multiple threads simultaneously This is often due to a shared variable that is written and then read again The omp critical pragma defines a subsection o
200. inline library that matches a function in the source text In the following example the compiler inlines the function proc from the inline library ib i and writes the executable code in the default output file a out 106 Chapter 4 pgf90 Minline name proc lib lib il myprog f The following command line is equivalent to the line above the only difference in this example is that the name and lib inline keywords are not used The keywords are provided so you can avoid name conflicts if you use an inline library name that does not contain a period Otherwise without the keywords a period lets the compiler know that the file on the command line is an inline library pgf90 Minline proc lib il myprog f 4 2 Creating an Inline Library You can create or update an inline library using the Mextract command line option If you do not specify a selection criteria along with the Mextract option the compiler attempts to extract all subprograms When you use the Mextract option only the extract phase is performed the compile and link phases are not performed The output of an extract pass is a library of functions available for inlining It is placed in the inline library file specified on the command line with the o filename specification If the library file exists new information is appended to it If the file does not exist it is created You can use the Minline option with the Mextract option In this case the extr
201. intf Parameter one type long 08x n 1 paral printf Parameter two type float f n f para2 printf Parameter three type double g n d_para3 printf The sum after conversion f n f return File sum_3 s Computes paral para2 para3 PGC Rel 1 1 opt 1 align 4 long EN1 sum_3 0xc8000000 text align 16 globl sum sum 3 EN1 Appendix A subl 12 esp movl Sebp 8 Sesp leal 8 Sesp sebp fildl 8 Sebp fadds 12 ebp faddl 16 Sebp fsts 12 Sebp fstps 4 Sebp flds 4 ebp movl Sebp sesp popl Sebp ret type sum_3 function sum_3 sum_3 Example A 1 C Program Calling an Assembly language Routine Run time Environment 225 Appendix B Messages This appendix describes the various messages that the compiler produces These messages include the sign on message and diagnostic messages for remarks warnings and errors The compiler always displays any error messages along with the erroneous source line on the screen If you specify the Mlist option the compiler places any error messages in the listing file You can also use the v option to display more information about the compiler assembler and linker invocations and about the host system For more information on the Miist and v options refer to Chapter 3 Command Line Options B 1 Diagnostic Messages Diagnostic messages provide syntactic and semantic information about your source
202. invoked with the Mvect option the vectorizer scans code searching for loops that are candidates for high level transformations such as loop distribution loop interchange cache tiling and idiom recognition replacement of a recognizable code sequence such as a reduction loop with optimized code sequences or function calls When the vectorizer finds vectorization opportunities it internally rearranges or replaces sections of loops the vectorizer changes the code generated your source code s loops are not altered In addition to performing these loop transformations the vectorizer produces extensive data dependence information for use by other phases of compilation and detects opportunities to use vector Streaming SIMD Extensions SSE SSE2 instructions on processors where these are supported The Mvect option can speed up code which contains well behaved countable loops which operate on large REAL REAL 4 REAL 8 INTEGER 4 COMPLEX Or COMPLEX DOUBLE arrays in Fortran and their C C counterparts However it is possible that some codes will show a decrease in performance when compiled with Mvect due to the generation of conditionally executed code segments inability to determine data alignment and other code generation factors For this reason it is recommended that you check carefully whether particular program units or loops show improved performance when compiled with this option enabled 2 5 1 Vectorization Sub options The
203. ion of these switches The following overview will help if you are just getting started with one of the PGI compilers or wish to experiment with individual optimizations Complete specifications of each of these options are listed in Chapter 3 Command Line Options The chapters that follow provide more detailed information on other Mpgflag options that control specific optimizations including function inlining Explicit parallelization through the use of OpenMP directives or pragmas is invoked using the mp option described in detail in Chapter 5 Open MP Directives for Fortran and Chapter 6 OpenMP Pragmas for C and C Optimization amp Parallelization 21 2 3 Local and Global Optimization 0 Using the PGI compiler commands with the Oleve option you can specify any of the following optimization levels the capital O is for Optimize 00 level zero specifies no optimization A basic block is generated for each Fortran statement O1 level one specifies local optimization Scheduling of basic blocks is performed Register allocation is performed O2 level two specifies global optimization This level performs all level one local optimization as well as level two global optimization O3 level three specifies aggressive global optimization This level performs all level one and level two optimizations and enables more aggressive hoisting and scalar replacement optimizations that may or may not be profitable
204. ions is Mvect sse This section contains an example of the use and potential effects of Mvect sse When the compiler switch Mvect sse is used the vectorizer in the PGI Workstation compilers automatically uses SSE and SSE2 instructions where possible when targeting processors where these are supported This capability is supported by all of the PGI Fortran C and C compilers See table P 2 for a complete specification of which X86 and AMD64 processors support SSE and SSE2 instructions Using Mvect sse performance improvements of up to two times over equivalent scalar code sequences are possible Optimization amp Parallelization 27 In the program in example 2 3 the vectorizer recognizes the vector operation in subroutine loop when the compiler switch Mvect sse is used This example shows the compilation informational messages and runtime results using the SSE instructions on an AMD Opteron processor based system along with issues that affect SSE performance First note that the arrays in Example 2 3 are single precision and that the vector operation is done using a unit stride loop Thus this loop can potentially be vectorized using SSE instructions on any processor that supports SSE or SSE2 instructions SSE operations can be used to operate on pairs of single precision floating point numbers and do not apply to double precision floating point numbers SSE2 instructions can be used to operate on quads of single precision f
205. ispatch each time except when there are less than chunk iterations remaining to be processed at which point all remaining iterations are assigned If no chunk is specified a default chunk size is chosen equal to 1 When schedule runtime is specified the decision regarding iteration scheduling is deferred until runtime The schedule type and chunk size can be chosen at runtime by setting the OMP_ SCHEDULE environment variable If this environment variable is not set the resulting schedule is equivalent to schedule static 6 7 omp barrier The OpenMP omp barrier pragma uses the following syntax pragma omp barrier There may be occasions in a parallel region when it is necessary that all threads complete work to that point before any thread is allowed to continue The omp barrier pragma synchronizes all threads at such a point in a program Multiple barrier points are allowed within a parallel region The omp barrier pragma must either be executed by all threads executing the parallel region or by none of them Optimization MP Pragmas for C and C 137 6 8 omp parallel for The omp parallel for pragma uses the following syntax pragma omp parallel for clauses lt C C for loop to be executed in parallel gt Clauses private list shared list default shared none firstprivate list lastprivate list reduction operator list copyin list if scalar_expression ordered schedule kind chun
206. ists and print them to file filename one_instantiation_per_object pgCC only Put out each template instantiation function or static data member in a separate object file The primary object file contains everything else in the compilation Allows users of libraries to pull only the instantiations that are needed Necessary for template libraries that rely on other template libraries pgcc only Accept dollar signs in identifiers optk_allow_dollar_in_id_chars pch pgCc only Automatically use and or create a precompiled header file pch_dir directoryname pgcc only The directory dirname in which to search for and or create a precompiled header file no_ pch_messages pgCc only Enable disable the display of a message indicating that a precompiled header file was created or used pgCC only Disallow all anachronistic constructs Stops after the preprocessing phase and saves the preprocessed file in filename i preinclude lt filename gt pgcc only Specify file to be included at the beginning of compilation to set system dependent macros types etc prelink_objects pgcc only If one_instantiation_per_object is used create template instantiations for a set of objects that are about to become a template library Required for template libraries that reference other template libraries lee Control instantiation of template functions use_pch filename pgcc only Use a
207. it may be possible to propagate that constant into the callee and perform optimizations that are not valid if the dummy argument is treated as a variable Function Inlining This optimization allows a call to a function to be replaced by a copy of the body of that function This optimization will sometimes speed up execution by eliminating the function call and return overhead Function inlining may also create opportunities for other types of optimization Function inlining is not always beneficial When used improperly it may increase code size and generate less efficient code 2 2 Getting Started with Optimizations Your first concern should be getting your program to execute and produce correct results To get your program running start by compiling and linking without optimization Use the optimization level O0 or select g that performs minimal optimization At this level you will be able to debug your program easily and isolate any coding errors exposed during porting to X86 or AMD64 platforms If you want to get started quickly with optimization a good option to use with any of the PGI compilers is fast For example pgf90 fast prog f 20 Chapter 2 For all of the PGI Fortran C and C compilers this option will generally produce code that is well optimized without the possibility of significant slowdowns due to pathological cases The fast option is equivalent to specifying at least 02 Munroll Mnoframe on all X
208. ities for instruction grouping or filling instruction delays found within the loop Examples 3 3 and 3 4 show the effect of code unrolling on a segment that computes a dot product REAL 4 A 100 B 100 Z INTEGER I DO I 1 100 Z 2Z2 A i B i END DO END Example 2 1 Dot Product Code REAL 4 A 100 B 100 Z INTEGER I DO I 1 100 2 Z 2 A i B i Z Z A itl B itl END DO END Example 2 2 Unrolled Dot Product Code 24 Chapter 2 Using the Minfo or Minfo loop option the compiler informs you when a loop is being unrolled For example a message indicating the line number and the number of times the code is unrolled similar to the following will display when a loop is unrolled dot 5 Loop unrolled 5 times Using the c lt m gt and n lt m gt sub options to Munroll or using Mnounroll you can control whether and how loops are unrolled on a file by file basis Using directives or pragmas as specified in Chapter 7 Optimization Directives and Pragmas you can precisely control whether and how a given loop is unrolled See Chapter 3 Command Line Options for a detailed description of the Munroll option 2 5 Vectorization Mvect If your program contains many loops try the Mvect option to see if it is helpful If you select Minfo loop and your code contains loops that can be vectorized the compiler reports relevant information on the optimizations applied When a PGI compiler command is
209. iven to the Parallel Tools Consortium and in particular to the High Performance Debugging Forum for their efforts PGF90 PGC Cluster Development Kit CDK and The Portland Group are trademarks and PGI PGHPF PGF77 PGCC PGPROF and PGDBG are registered trademarks of STMicroelectronics Inc Other brands and names are the property of their respective owners The use of STLport a C Library is licensed separately and license distribution and copyright notice can be found in the Release Notes for a given release of the PGI compilers and tools PGI User s Guide Copyright 1998 2000 The Portland Group Inc Copyright 2000 2003 STMicroelectronics Inc All rights reserved Printed in the United States of America Part Number 2030 990 888 0603 First Printing Release 1 7 June 1998 Second Printing Release 3 0 January 1999 Third Printing Release 3 1 September 1999 Fourth Printing Release 3 2 September 2000 Fifth Printing Release 4 0 May 2002 Sixth Printing Release 5 0 June 2003 Technical support trs pgroup com Sales sales pgroup com http www pgroup com Table of Contents TABLE OF CONTENTS IH PREFACE 1 Audience Description ii cnn d cians ea ee 1 Compatibility and Conformance to Standards 00 cccccesccsseesceeseeesceesceeeeeeeeseeesecesecaeeaeceeeaeeenes 1 Organization eee see eiei iieii ieoor a EEEE EE EA E EEEE ES E EE 2 Hardware and Software Constraints sssesessseseeessseeeese
210. k The semantics of the omp parallel for pragma are identical to those of a parallel region containing only a single parallel for loop and pragma The available clauses are as defined in Sections 6 2 omp parallel and 6 6 omp for 6 9 omp sections The omp sections pragma uses the following syntax pragma omp sections Clauses pragma omp section lt C C structured block executed by processor i gt pragma omp section lt C C structured block executed by processor j gt Clauses private list firstprivate list lastprivate list 138 Chapter 6 reduction operator list nowait The omp sections pragma defines a non iterative work sharing construct within a parallel region Each section is executed by a single thread If there are more threads than sections some threads will have no work and will jump to the implied barrier at the end of the construct If there are more sections than threads one or more threads will execute more than one section An omp section pragma may only appear within the lexical extent of the enclosing omp sections pragma In addition the code within the omp sections pragma must be a structured block and the code in each omp section must be a structured block The available clauses are as defined in Sections 6 2 omp parallel and 6 6 omp for 6 10 omp parallel sections The omp parallel sections pragma uses the following syntax pragma omp parallel sections clauses
211. l aH oh eapalge inter cf amp cl amp i SUBROUTINE INTER CF C I COMPLEX C COMPLEX CF COMPLEX FUNCTION CF T Example 10 2 COMPLEX Return Values 10 7 Array Indices C C arrays and Fortran arrays use different default initial array index values By default C C arrays start at 0 and Fortran arrays start at 1 If you adjust your array comparisons so that a Fortran second element is compared to a C C first element and adjust similarly for other elements you should not have problems working with this difference If this is not satisfactory you can declare your Fortran arrays to start at zero 190 Chapter 10 Another difference between Fortran and C C arrays is the storage method used Fortran uses column major order and C C use row major order For one dimensional arrays this poses no problems For two dimensional arrays where there are an equal number of rows and columns row and column indexes can simply be reversed For arrays other than single dimensional arrays and square two dimensional arrays inter language function mixing is not recommended 10 8 Example Fortran Calling C Example 10 4 shows a C function that is called by the Fortran main program shown in Example 10 3 Notice that each argument is defined as a pointer since Fortran passes by reference Also notice that the C function name uses all lower case and a trailing logical 1 character intege
212. led within parallel regions The value should be an integer lt n gt concatenated with M or m to specify stack sizes of n megabytes For example setenv MPSTKZ 8M 128 Chapter 5 Chapter 6 OpenMP Pragmas for C and C The PGCC ANSI C and C compilers support the OpenMP C C Application Program Interface The OpenMP shared memory parallel programming model is defined by a collection of compiler directives or pragmas library routines and environment variables that can be used to specify shared memory parallelism in Fortran C and C programs The OpenMP C C pragmas include a parallel region construct for writing coarse grain SPMD programs work sharing constructs which specify that C C for loop iterations should be split among the available threads of execution and synchronization constructs The data environment is controlled using clauses on the pragmas or with additional pragmas Run time library functions are provided to query the parallel runtime environment for example to determine how many threads are participating in execution of a parallel region Finally environment variables are provided to control the execution behavior of parallel programs For more information on OpenMP and a complete copy of the OpenMP C C API specification see http www openmp org 6 1 Parallelization Pragmas Parallelization pragmas are pragma statements in a C or C program that are interpreted by the PGCC C and C compilers when the opti
213. les given with each section the functions omp_get_num_threads and omp get thread num are used refer to Section 6 15 Run time Library Routines They return the number of threads currently in the team executing the parallel region and the thread number within the team respectively 6 2 omp parallel The OpenMP omp parallel pragma uses the following syntax Syntax pragma omp parallel clauses lt C C structured block gt Clauses private list shared list default shared none firstprivate list reduction operator list copyin list if scalar_expression This pragma declares a region of parallel execution It directs the compiler to create an executable in which the statements within the following C C structured block are executed by multiple lightweight threads The code that lies within the structured block is called a parallel region The OpenMP parallelization pragmas support a fork join execution model in which a single thread executes all statements until a parallel region is encountered At the entrance to the parallel region a system dependent number of symmetric parallel threads begin executing all statements in the parallel region redundantly These threads share work by means of work sharing constructs such as parallel for loops see the next example The number of threads in the team is controlled by the oOMP_NUM_THREADS environment variable If OoMP_NUM_THREADS is not defined the program will ex
214. link step This option is useful for linking programs in which the main program is written in C C and one or more subroutines are written in Fortran pg 77 pgf90 and pghpf only Mprof option option Set profile options option can be any of the following func perform PGI style function level profiling 66 Chapter 3 line Mrecursive Mnorecursive Mref externals perform PGI style line level profiling instructs the compiler to allow Fortran subprograms to be called recursively Fortran subprograms may not be called recursively force references to names appearing in EXTERNAL statements pgf77 pgf90 and pghpf only Mnoref_externals Mreentrant Mnoreentrant do not force references to names appearing in EXTERNAL statements pgf77 pgf 90 and pghpf only instructs the compiler to avoid optimizations that can prevent code from being reentrant instructs the compiler not to avoid optimizations that can prevent code from being reentrant Msecond_underscore instructs the compiler to add a second underscore to the name of a Fortran global symbol if its name already contains an underscore This option is useful for maintaining compatibility with object code compiled using g77 which uses this convention by default og 77 pgf90 and pghpf only Mnosecond_underscore Msignextend Mnosignextend Msafe_lastval Command line Options instructs the compiler
215. llowed by the mangled form of the class name class A static int i i__1A e Names of variables generated for virtual function tables These have names like vtblmangled class name or vtblmangled base class namemangled class name e Names of variables generated to contain runtime type information These have names like Ttype encoding and TIDtype encoding 214 Chapter 12 12 2 Mangling Summary This section lists some of the C entities that are mangled and provides some details on the mangling algorithm For more details refer to The Annotated C Reference Manual 12 2 1 Type Name Mangling Using PGCC C each type has a corresponding mangled encoding For example a class type is represented as the class name preceded by the number of characters in the class name as in 5abcde for abcde Simple types are encoded as lower case letters as in i for int or f for float Type modifiers and declarators are encoded as upper case letters preceding the types they modify as in U for unsigned or P for pointer 12 2 2 Nested Class Name Mangling Nested class types are encoded as a Q followed by a digit indicating the depth of nesting followed by a _ followed by the mangled form names of the class types in the fully qualified name of the class from outermost to innermost class A class B Q2_1A1B 12 2 3 Local Class Name Mangling The name of the nested class itself is mangled to the form described above with a prefix __ whi
216. loating point numbers or on pairs of double precision floating point numbers Loops vectorized using SSE or SSE2 instructions operate much more efficiently when processing vectors that are aligned to a cache line boundary You can cause unconstrained data objects of size 16 bytes or greater to be cache aligned by compiling with the Mcache_align switch An unconstrained data object is a data object that is not a common block member and not a member of an aggregate data structure Note In order for stack based local variables to be properly aligned the main program or function must be compiled with Mcache_align The Mcache_align switch has no effect on the alignment of Fortran allocatable or automatic arrays If you have arrays that are constrained for example vectors that are members of Fortran common blocks you must specifically pad your data structures to ensure proper cache alignment Mcache_align causes only the beginning address of each common block to be cache aligned The following examples show results of compiling the example code with and without Mcache_align program vector op parameter N 9999 real 4 x n y n z n w n do i 1 n y i i zZ i 2 i w i 4 i enddo do j 1 200000 call loop x y z w 1 0e0 n enddo print x 1 x 771 x 3618 x 6498 x 9999 end 28 Chapter 2 subroutine loop a b c d s n integer i n real 4 a n b n c n d n s do i 1 n a i b i c i s d i
217. lon64 X X X X X AMD Opteron X X X X X Intel Celeron X Intel Pentium H X Intel Pentium HI x x x Intel Pentium 4 x x x X Intel Xeon X X X X In this manual the convention is to use X86 to specify the group of processors in Table P 2 that are listed 32 bit but not 64 bit The convention is to use AMD64 to specify the group of processors that are listed as both 32 bit and 64 bit X86 processor based systems can run only under 32 bit operating systems AMD64 processor based systems can run either 32 bit or 64 bit operating systems and can execute all 32 bit X86 binaries in either case AMD64 processors have additional registers and 64 bit addressing capabilities that are utilized by the PGI compilers and tools when operating under a 64 bit operating system The prefetch SSE1 and SSE2 processor features further distinguish the various processors Where such distinctions are important with respect to a given compiler option or feature it is explicitly noted in this manual Preface S Related Publications The following documents contain additional information related to the X86 and AMD64 architectures and the compilers and tools available from The Portland Group Compiler Technology PGF77 Reference Manual describes the FORTRAN 77 statements data types input output format specifiers and additional reference material PGHPF Reference Manual describes the HPF statements data types inp
218. ls il c inline utils il c FC o myprog main o utils o parser o alloc o Example 4 1 Sample Makefile 108 Chapter 4 4 3 Error Detection during Inlining To request inlining information from the compiler when you invoke the inliner specify the Minfo inline option For example pgf90 Minline mylib il Minfo inline myext f 4 4 Examples Assume the program dhry consists of a single source file dhry f The following command line builds an executable file for dhry in which proc7 is inlined wherever it is called pgf90 dhry f Minline proc7 The following command lines build an executable file for dhry in which proc7 plus any functions of approximately 10 or fewer statements are inlined one level only Note that the specified functions are inlined only if they are previously placed in the inline library temp i1 during the extract phase pgf90 dhry f Mextract o temp il pgf90 dhry f Minline 10 Proc7 temp il Assume the program fibo f contains a single function fibo that calls itself recursively The following command line creates the file fibo o in which fibo is inlined into itself pgf90 fibo f c Mrecursive Minline fibo Because this version of fibo recurses only half as deeply it executes noticeably faster Using the same source file dhry f the following example builds an executable for dhry in which all functions of roughly ten or fewer statements are inlined Two levels of inlining are perform
219. lso applies to substring and function references 0076 Subscripts specified for non array variable 0077 Subscripts omitted from array 0078 Wrong number of subscripts specified for 0079 Keyword form of intrinsic argument illegal infor 236 Appendix B 0080 Subscript for array is out of bounds 0081 Matrix vector illegal as subprogram argument A matrix vector reference cannot be used as a subprogram argument 0082 Illegal substring expression for variable Substring expressions must be of type integer and if constant must be greater than zero 0083 Vector expression used where scalar expression required A vector expression was used in an illegal context For example iscalar iarray where a scalar is assigned the value of an array Also character and record references are not vectorizable 0084 Illegal use of symbol This message is used for many different errors 0085 Incorrect number of arguments to statement function 0086 Dummy argument to statement function must be a variable 0087 Non constant expression where constant expression required 0088 Recursive subroutine or function call of A function may not call itself 0089 Illegal use of symbol with character length Symbols of type CHARACTER must be dummy variables and must not be used as statement function dummy parameters and statement function names Also a dummy variable of type CHARACTER cannot be used as a function 0090 Hollerith
220. lt used Instantiates only the functions that are used in this compilation Usage In the following example all templates are instantiated pgCC tall myprog cc Command line Options 103 Chapter 4 Function Inlining Function inlining replaces a call to a function or a subroutine with the body of the function or subroutine This can speed up execution by eliminating parameter passing and function subroutine call and return overhead It also allows the compiler to optimize the function with the rest of the code Note that using function inlining indiscriminately can result in much larger code size and no increase in execution speed The PGI compilers provide two categories of inlining e Automatic inlining During the compilation process a hidden pass precedes the compilation pass This hidden pass extracts functions that are candidates for inlining The inlining of functions occurs as the source files are compiled e Inline libraries You create inline libraries for example using the pg 90 command and the Mextract and o options There is no hidden extract pass but you must ensure that any files that depend on the inline library use the latest version of the inline library There are important restrictions on inlining Inlining only applies to certain types of functions Refer to Section 4 5 Restrictions on Inlining at the end of this chapter for more details on function inlining limitations 4 1 Invoking Function
221. m those used under UNIX operating systems Win32 programs compiled using the Vunix Fortran command line option use the UNIX convention rather than the default Win32 convention Sections 6 1 through 6 13 describe how to perform inter language calling using the UNIX convention All information in those sections pertaining to compatibility of arguments applies to Win32 as well See Section 70 14 Win32 Calling Conventions for details on the symbol name and argument passing conventions used on Win3 2 10 2 Inter language Calling Considerations In general when argument data types and function return values agree you can call a C or C function from Fortran and likewise you can call a Fortran function from C or C You may need to develop special procedures in cases where data types for arguments do not agree For example the Fortran COMPLEX type does not have a matching type in C it is still possible to provide inter language calls but there are no general calling conventions for such cases In this instance you need to develop a special procedure Follow these guidelines Inter language Calling 183 e Note that ifa C function contains objects with constructors and destructors calling such a function from either C or Fortran will not be possible unless the initialization in the main program is performed from a C program where constructors and destructors are properly initialized e In general you can call a C function from C witho
222. me If you do not specify path for a filename the compiler uses the current directory You must specify path separately for each filename not in the current directory Note that within the BASH for Win32 shell UNIX directory naming conventions are assumed In particular is used to separate directory names in the directory hierarchy is the name of a source file assembly language file object file or library to be processed by the compilation system You can specify more than one path filename 1 2 2 Command line Options The command line options control various aspects of the compilation process For a complete alphabetical listing and a description of all the command line options refer to Chapter 3 Command Line Options 1 2 3 Fortran Directives and C C Pragmas Fortran directives or C C pragmas inserted in program source code allow you to alter the effects of certain command line options and control various aspects of the compilation process for a specific routine or a specific program loop For a complete alphabetical listing and a description Getting Started of all the Fortran directives and C C pragmas refer to Chapter 5 Open MP Directives for Fortran Chapter 6 OpenMP Pragmas for C and C and Chapter7 Optimization Directives and Pragmas 1 3 Filename Conventions The PGI compilers use the filenames that you specify on the command line to find and to create input and output files This section describes
223. me and may make debugging difficult However optimization produces more efficient code that usually runs significantly faster than code that is not optimized The compilers optimize code according to the specified optimization level Using the O Mvect and Mconcur options you specify the optimization levels In addition several Mpgflag switches control specific types of optimization and parallelization This chapter describes the optimization options and describes how to choose optimization options to use with the PGI compilers Chapter 4 Function Inlining describes how to use the function inlining options 2 1 Overview of Optimization In general optimization involves using transformations and replacements that generate more efficient code This is done by the compiler and involves replacements that are independent of the particular target processor s architecture as well as replacements that take advantage of the X86 or AMD64 architecture instruction set and registers For the discussion in this and the following chapters optimization is divided into the following categories Local Optimization This optimization performs on a block by block basis within a program s basic blocks A basic block is a sequence of statements in which the flow of control enters at the beginning and leaves at the end without the possibility of branching except at the end The PGI compilers perform many types of local optimization including
224. ment integer 1 Approximate value 10 37 to 1038 10307 to 10308 1 10 307 to 10308 1 0 to 25 2 1 where size is the number of bits in the bit field 28i2e 1 to 28ize 1_1 where size is the number of bits in the bit field 2 Bit fields occupy as many bits as you assign them up to 4 bytes and their length need not be a multiple of 8 bits 1 byte Table 9 5 Scalar Alignment Data Type Alignment is aligned on a 1 byte boundary is aligned on a 2 byte boundary is aligned on a 4 byte boundary signed or unsigned 178 Chapter 9 9 2 2 Cand C Aggregate Data Types An aggregate data type consists of one or more scalar data type objects You can declare the following aggregate data types array consists of one or more elements of a single data type placed in contiguous locations from first to last class C only is a class that defines an object and its member functions The object can contain fundamental data types or other aggregates including other classes The class members are allocated in the order they appear in the definition but may not occupy contiguous locations struct is a structure that can contain different data types The members are allocated in the order they appear in the definition but may not occupy contiguous locations When a struct is defined with member functions its alignment issues are the same as those for a class union is a single location that can co
225. mint2 lt lt endl numfloatl lt lt endl numdoubl lt lt endl numshorl lt lt endl Example 10 14 C main program cpmain C To compile this Fortran subroutine and C program use the following command lines pgf90 c forts f pgCC forts o cpmain C lpgf90 lpgf90 rpml lpgf902 lpgf90rtl lpgftnrtl Executing this C main should produce the following output booll letterl numintl numint2 numfloatl numdoub1 numshorl TRUE v 11 44 39 6 902 299 Note that you must explicitly link in the PGF90 runtime support libraries when linking pgf 90 compiled program units into C or C main programs When linking pg 77 compiled program units into C or C main programs you need only link in pgfinrtl 10 14 Win32 Calling Conventions Aside from name mangling considerations in C the calling convention i e the symbol name to which the subroutine or function name is mapped and the means by which arguments are passed for C C is identical between most compilers on Win32 and UNIX variants However Fortran calling conventions vary widely between Win32 and UNIX and between UNIX implementations 198 Chapter 10 10 14 1 Win32 Fortran Calling Conventions Four styles of calling conventions are supported using the PGI Fortran compilers for Win32 Default C STDCALL and UNIX e Default Default is the method used in the absence of compilation flags or directives to alter the default e Cor STD
226. mization Directives and Pragmas 161 Chapter 8 Libraries This chapter discusses issues related to PGlI supplied compiler libraries It also addresses the use of C C builtin functions in place of the corresponding ibc routines creation of dynamically linked libraries also known as shared objects or shared libraries on UNIX systems and math libraries 8 1 Using builtin Math Functions in C C The name of the math header file is math h Include the math header file in all of your source files that use a math library routine as in the following example which calculates the inverse cosine of n 3 include lt math h gt define PI 3 1415926535 main double x y x PI 3 0 y acos x Including math h will cause PGCC C and C to use builtin functions which are much more efficient than library calls In particular the following intrinsics calls will be processed using builtins if you include math h atan2 cos fabs exp atan sin log pow tan sqrt log10 8 2 Creating and Using Shared Object Files on UNIX All of the PGI Fortran C and C compilers support creation of shared object files Unlike statically linked object and library files shared object files link and resolve references with an executable at runtime via a dynamic linker supplied with your operating system The PGI compilers must generate position independent code to support creation of shared objects by the Libraries 163 linker This is not the def
227. n a 4 byte word The following is an example of the pseudo code for the above call using STDCALL conventions call _work 20 Sval E Sval a addr b sval n Chapter 10 Note that in this case there are still 20 bytes in the argument list However rather than 5 4 byte quantities as in the Default convention there are 3 4 byte quantities and 1 8 byte quantity the double precision value of a e C The symbol name for the subroutine is constructed by pre pending an underscore and converting to all lower case Character strings are truncated to the first character in the string which is passed by value as the first byte in a 4 byte word The following is an example of the pseudo code for the above call using C conventions call work sval E Sval a addr b val n e UNIX The symbol name for the subroutine is constructed by pre pending an underscore converting to all lower case and appending an underscore Byte counts for character strings appear in sequence following the last argument in the argument list The following is an example of the pseudo code for the above call using UNIX conventions call _work_ addr ERR saddr a addr b addr n 3 10 14 3 Using the Default Calling Convention Using the Default calling convention is straightforward Use the default convention if no directives are inserted to modify calling conventions and if the Munix compilation flag is not used See the previ
228. n a parallel region are undefined upon entry to the parallel region If the first use of a private variable within the parallel region is in a right hand side expression the results of the expression will be undefined i e this is probably a coding error e Likewise variables declared private in a parallel region are undefined when serial execution resumes at the end of the parallel region The variables specified ina shared list are shared between all threads in a team meaning that all threads access the same storage area for shared data The default clause allows the user to specify the default attribute for variables in the lexical extent of the parallel region Individual clauses specifying private shared etc status override the declared default Specifying default none declares that there is no implicit default and in this case each variable in the parallel region must be explicitly listed with an attribute of private shared firstprivate or reduction Variables that appear in the list of a firstprivate clause are subject to the same semantics as private variables but in addition are initialized from the original object existing prior to entering the parallel region Variables that appear in the list of a reduction clause must be Optimization MP Pragmas for C and C 131 shared A private copy of each variable in list is created for each thread as if the private clause had been specified Each private copy is initialized accord
229. n for the compilation based on the supplied sourcetype parameter Valid sourcetypes are 1 c s and o These indicate respectively stop after preprocessing compiling assembling or linking Usage The following examples show the different Q options pgf90 Qproduce s hello f pgf90 Qoption ld s hello f Command line Options 89 pgf90 Qpath home test hello f pgf90 Qdir home comp new hello f Cross reference p R lt directory gt Valid only on Linux and is passed to the linker Instructs the linker to hard code the pathname lt directory gt into the search path for generated shared object dynamically linked library files Note that there cannot be a space between R and lt directory gt Cross reference fpic shared G r4 and r8 Interpret DOUBLE PRECISION variables as REAL r4 or REAL variables as DOUBLE PRECISION r8 Usage pgf90 r4 myprog f Cross reference i2 i4 8 E Specifies the name of the driver startup configuration file If the file or pathname supplied is not a full pathname the path for the configuration file loaded is relative to the DRIVER path the path of the currently executing driver If a full pathname is supplied that file is used for the driver configuration file Syntax rc path filename Where path is either a relative pathname relative to the value of SDRIVER or a full pathname beginning with Filename is the driver conf
230. nction may result in recursion 278 Can t inline too large 280 Syntax error in directive 281 Directive ignored B 4 Runtime Error Messages This section presents the error messages generated by the runtime system The runtime system displays error messages on standard output B 4 1 Message Format The messages are numbered but have no severity indicators because they all terminate program execution B 4 2 Message List Here are the runtime error messages 201 illegal value for specifier An improper specifier value has been passed to an I O runtime routine Example within an OPEN statement form unknown Messages 251 202 conflicting specifiers Conflicting specifiers have been passed to an I O runtime routine Example within an OPEN statement form unformatted blank null 203 record length must be specified A recl specifier required for an I O runtime routine has not been passed Example within an OPEN statement access direct has been passed but the record length has not been specified recl specifier 204 illegal use of a readonly file Self explanatory Check file and directory modes for readonly status 205 SCRATCH and SAVE KEEP both specified In an OPEN statement a file disposition conflict has occurred Example within an OPEN statement status scratch and dispose keep have been passed 206 attempt to open a named file as SCRATCH 207 file is already connected to anoth
231. nd when a member of a class or one of its base classes should actually be found This bug is not emulated in cfront compatibility mode e A parameter of type const void is allowed on operator delete it is treated as equivalent to void e A period may be used for qualification where should be used Only may be used as a global qualifier Except for the global qualifier the two kinds of qualifier operators may not be mixed in a given name i e you may say A B C orA B C butnotA B C or A B C A period may not be used in a vacuous destructor reference nor in a qualifier that follows a template reference such as A lt T gt B e cfront 2 1 does not correctly look up names in friend functions that are inside class definitions In this example function f should refer to the functions and variables e g 1 and a1 from the class declaration Instead the global definitions are used int al int el void f1 class A int al void 1 friend void f int il al cfront uses global al fl cfront uses global f1 j Only the innermost class scope is incorrectly skipped by cfront as illustrated in the following example 264 Appendix C int al int bl struct A static int al class B static int bl friend void f int il al cfront uses A al int j1 bl cfront uses global bl e operator may be declared as a nonmember function This is flagged as an an
232. nes an appropriate cutoff length and generates scalar code to be executed whenever the loop count is less than or equal to that length If altcode n is specified the scalar altcode is executed whenever the loop count is less than or equal to n If noaltcode is specified no alternate scalar code is generated Dist Option The option Mconcur dist block cyclic option specifies whether to assign loop iterations to the available threads in blocks or in a cyclic round robin fashion Block distribution is the default If cyclic is specified iterations are allocated to processors cyclically That is processor 0 performs iterations 0 3 6 etc processor performs iterations 1 4 7 etc and processor 2 performs iterations 2 5 8 etc Cncall Option The option Mconcur cncall specifies that it is safe to parallelize loops that contain subroutine or function calls By default such loops are excluded from consideration for auto parallelization Also no minimum loop count threshold must be satisfied before parallelization will occur and last values of scalars are assumed to be safe The environment variable Ncpus is checked at runtime for a parallel program If NCPUS is set to 1 a parallel program runs serially but will use the parallel routines generated during compilation If ncrus is set to a value greater than 1 the specified number of processors will be used to execute the program Setting NCPUS to a value exceeding the number of
233. ng your main program o pgf90 myprog f lblas llapack Highly optimized assembly coded versions of the BLAS and certain FFT routines may be available for your platform In some cases these are shipped with the PGI compilers See the current release notes for the PGI compilers you are using to determine if these optimized libraries exist where they can be downloaded if necessary and how to incorporate them into your installation as the default 8 7 The C Standard Template Library The PGCC C compiler includes a bundled copy of the STLPort Standard C Library See the online Standard C Library tutorial and reference manual at http www stlport com for further details and licensing The version of the Rogue Wave Standard Template Library included with older versions of PGCC C Version 3 3 and older requires that the header files be encrypted so you won t be able to read those Rogue Wave files Libraries 171 Chapter 9 Fortran C and C Data Types This chapter describes the scalar and aggregate data types recognized by the PGI Fortran C and C compilers the format and alignment of each type in memory and the range of values each type can take on X86 or AMD64 processor based systems running a 32 bit operating system For more information on X86 specific data representation refer to the System V Application Binary Interface Processor Supplement listed in the Preface This chapter specifically does not address AMD6
234. nitialized from an argument that requires a non class standard conversion the conversion counts as a user defined conversion This is an outright bug which unfortunately happens to be exploited in the NIH class libraries When a builtin operator is considered alongside overloaded operators in overload resolution the match of an operand of a builtin type against the builtin type required by the builtin operator is considered a standard conversion in all cases e g even when the type is exactly right without conversion A reference to a non const type may be initialized from a value that is a const qualified version of the same type but only if the value is the result of selecting a member from a const Class object or a pointer to such an object A cast to an array type is allowed it is treated like a cast to a pointer to the array element type A warning is issued C Dialect Supported 263 e When an array is selected from a class the type qualifiers on the class object if any are not preserved in the selected array In the normal mode any type qualifiers on the object are preserved in the element type of the resultant array e An identifier in a function is allowed to have the same name as a parameter of the function A warning is issued e An expression of type void may be supplied on the return statement in a function with a void return type A warning is issued e cfront has a bug that causes a global identifier to be fou
235. not to add a second underscore to the name of a Fortran global symbol if its name already contains an underscore pgf77 pgf90 and pghpf only instructs the compiler to extend the sign bit that is set as a result of converting an object of one data type to an object of a larger signed data type instructs the compiler not to extend the sign bit that is set as the result of converting an object of one data type to an object of a larger data type In the case where a scalar is used after a loop but is not defined on every iteration of the loop the compiler does not by default parallelize the loop However this option tells the compiler it s safe to parallelize the loop For a given loop the last value computed for all scalars make it safe to 67 parallelize the loop Mstrided instructs the compiler to inhibit certain optimizations and to allow for stride 0 array references This option may degrade performance and should only be used if zero stride induction variables are possible MnostrideO instructs the compiler to perform certain optimizations and to disallow for stride 0 array references Munix use UNIX symbol and parameter passing conventions for Fortran subprograms pgf77 pg 90 and pghpf for Win32 only Default For arguments that you do not specify the default code generation controls are as follows norecursive nostrideO noreentrant signextend nosecond_underscore noref_externals Mpegflag Environment
236. nt structure types may not be assigned to one another 238 Appendix B 100 Expression cannot be promoted to a vector An expression was used that required a scalar quantity to be promoted to a vector illegally For example the assignment of a character constant string to a character array Records too cannot be promoted to vectors 101 Vector operation not allowed on Record and character typed entities may only be referenced as scalar quantities 102 Arithmetic if expression has wrong data type The parenthetical expression of an arithmetic if statement must be an integer real or double precision scalar expression 103 Type conversion of subscript expression for The data type of a subscript expression must be integer If it is not it is converted 104 Illegal control structure This message is issued for a number of errors involving IF THEN statements and DO loops If the line number specified is the last line END statement of the subprogram the error is probably an unterminated DO loop or IF THEN statement 105 Unmatched ELSEIF ELSE or ENDIF statement An ELSEIF ELSE or ENDIF statement cannot be matched with a preceding IF THEN statement 106 DO index variable must be a scalar variable The DO index variable cannot be an array name a subscripted variable a PARAMETER name a function name a structure name etc 107 Illegal assigned goto variable 108 Illegal variable in NAMELIST group A NAMELIST
237. ntain any of a specified set of scalar or aggregate data types A union can have only one value at a time The data type of the union member to which data is assigned determines the data type of the union after that assignment 9 2 3 Class and Object Data Layout Class and structure objects with no virtual entities and with no base classes that is just direct data field members are laid out in the same manner as C structures The following section describes the alignment and size of these C like structures C classes and structures as a special case of a class are more difficult to describe Their alignment and size is determined by compiler generated fields in addition to user specified fields The following paragraphs describe how storage is laid out for more general classes The user is warned that the alignment and size of a class or structure is dependent on the existence and placement of direct and virtual base classes and of virtual function information The information that follows is for informational purposes only reflects the current implementation and is subject to change Do not make assumptions about the layout of complex classes or structures All classes are laid out in the same general way using the following pattern in the sequence indicated e First storage for all of the direct base classes which implicitly includes storage for non virtual indirect base classes as well When the direct base class is also virtu
238. nue n i c i The outer loop above is not parallelized because the compiler detects a cross iteration dependence in the assignment to a i Suppose the outer loop were parallelized Then both processors would simultaneously attempt to make assignments into a 1 n Now in general the values computed by each processor for a 1 n will differ so that simultaneous assignment into a 1 n will produce values different from sequential execution of the loops In this example values computed for a 1 n don t depend on j so that simultaneous assignment by both processors will not yield incorrect results However it is beyond the scope of the compilers dependence analysis to determine that values computed in one iteration of a loop don t differ from values computed in another iteration So the worst case is assumed and different iterations of the outer loop are assumed to compute different values for a 1 n Is this assumption too pessimistic If j doesn t occur anywhere within a loop the loop exists only to cause some delay most probably to improve timing resolution And it s not usually valid to parallelize timing loops to do so would distort the timing information for the inner loops 34 Chapter 2 Scalars Quite often scalars will inhibit parallelization of non innermost loops There are two separate cases that present problems In the first case scalars appear to be expandable but appear in non innermost loops as in the f
239. ock for use in subsequent calls to lock routines This initial state of Jock is unlocked It is illegal to make a call to this routine if Jock is already associated with a lock include lt omp h gt void omp_destroy lock omp_lock t lock void omp_destroy nest_lock omp_nest_lock_t lock disassociates a lock associated with the variable Jock include lt omp h gt void omp_set_lock omp_lock t lock void omp_set_nest_lock omp_nest_lock_t lock Optimization MP Pragmas for C and C 143 causes the calling thread to wait until the specified lock is available The thread gains ownership of the lock when it is available It is illegal to make a call to this routine if Jock has not been associated with a lock include lt omp h gt void omp_unset_lock omp_lock_t lock void omp_unset_nest_lock omp_nest_lock_t lock causes the calling thread to release ownership of the lock associated with Jock It is illegal to make a call to this routine if lock has not been associated with a lock include lt omp h gt int omp_test_lock omp_lock_t lock int omp_test_nest_lock omp_nest_lock_t lock causes the calling thread to try to gain ownership of the lock associated with ock The function returns non zero if the thread gains ownership of the lock and zero otherwise It is illegal to make a call to this routine if Jock has not been associated with a lock 6 16 Environment Variables OMP NUM THREADS specifies the number of t
240. ollowing example dol j 1 n x b j do 1 i 1 n a i j x c i j 1 continue There are a number of technical problems to be resolved in order to recognize expandable scalars in non innermost loops Until this generalization occurs scalars like x above will inhibit parallelization of loops in which they are assigned In the following example scalar k is not expandable and it is not an accumulator for a reduction do3i 1 n dol j 1 n 1 a j i b k x kei 2 if i gt n 2 k n i n 2 3 continue If the outer loop is parallelized conflicting values will be stored into k by the various processors The variable k cannot be made local to each processor because the value of k must remain coherent among the processors It is possible the loop could be parallelized if all assignments to k are placed in critical sections However it is not clear where critical sections should be introduced because in general the value for k could depend on another scalar or on k itself and code to obtain the value of other scalars must reside in the same critical section In the example above the assignment to k within a conditional at label 2 prevents k from being recognized as an induction variable If the conditional statement at label 2 is removed k would be an induction variable whose value varies linearly with j and the loop could be parallelized Optimization amp Parallelization 35 Scalar Last Values During parallelization
241. ompiled using the Mvect command line option func2 is vectorized but funcl is not vectorized In the following example the global novintr pragma turns off vectorization for the entire file 160 include math h funcl pragma global novector float a 100 100 b 100 100 float c 100 100 d 100 100 inte yp WF for i 0 1 lt 100 i for j3 0 3 lt 100 j alili afliltj3 bfailt3 clillil Chapter 7 efi j clil g bfal 3 dlillsl func2 float a 200 200 b 200 200 float c 200 200 da 200 200 int i j for i1 0 1 lt 200 it for j 0 43 lt 200 j aLI afil 3 bfil 3 cliltjl Cd gh Chil Ig BENEI sd LF oO Special Scope Rules Special rules apply for a pragma with loop routine and global scope When the pragma is placed within a routine it applies to the routine from its point in the routine to the end of the routine The same rule applies for one of these pragmas with global scope However there are several pragmas for which only routine and global scope applies and which affect code immediately following the pragma e bounds and fcon The bounds and fcon pragmas behave in a similar manner to pragmas with loop scope That is they apply to the code following the pragma e opt and safe When the opt and safe pragmas are placed within a routine they apply to the entire routine as if they had been placed at the beginning of the routine Opti
242. on Chapter 4 Function Inlining describes how to use function inlining and shows how to create an inline library Chapter 5 Open MP Directives for Fortran provides a description of the OpenMP 2 Preface Chapter 6 Chapter 7 Chapter 8 Chapter 9 Chapter 10 Chapter 11 Chapter 12 Appendix A Appendix B Appendix C Fortran parallelization directives and shows examples of their use OpenMP Pragmas for C and C provides a description of the OpenMP C and C parallelization pragmas and shows examples of their use Optimization Directives and Pragmas provides a description of each Fortran optimization directive and C C optimization pragma and shows examples of their use Libraries discusses PGI support libraries and shared object files Fortran C and C Data Types describes the data types that are supported by the PGI Fortran C and C compilers Inter language Calling provides examples showing how to place C Language calls in a Fortran program and Fortran Language calls in a C program C Template Instantiation describes the options available for template instantiation C Name Mangling describes the name mangling facility and explains the transformations of names of entities to names that include information on aspects of the entity s type and a fully qualified name Run time Environment describes the assembly language calling conventions and examples of assembly language calls Message
243. on mp is specified on the command line The form of a parallelization pragma is pragma omp pragmas name clauses The pragmas follow the conventions of the C and C standards White space can appear before and after the Preprocessing tokens following the pragma omp are subject to macro replacement The order in which clauses appear in the parallelization pragmas is not significant Spaces separate clauses within the pragmas Clauses on pragmas may be repeated as needed subject to the restrictions listed in the description of each clause For the purposes of the OpenMP pragmas a C C structured block is defined to be a statement or compound statement a sequence of statements beginning with and ending with that has a single entry and a single exit No statement or compound statement is a C C structured block if there is a jump into or out of that statement The compiler option mp enables recognition of the parallelization pragmas The use of this option also implies Mreentrant local variables are placed on the stack and optimizations that may result in non reentrant code are disabled e g Vnoframe Optimization MP Pragmas for C and C 129 Also note that calls to I O library functions are system dependent and are not necessarily guaranteed to be thread safe I O library calls within parallel regions should be protected by critical regions see below to ensure they function correctly on all systems In the examp
244. on from the same iteration Here t is privatizable but the use of t outside the loop may yield incorrect results since the compiler may not be able to detect on which iteration of the parallelized loop t is assigned last The compiler detects the above cases Where a scalar is used after the loop but is not defined on every iteration of the loop parallelization will not occur If you know that the scalar is assigned on the last iteration of the loop making it safe to parallelize the loop a pragma is available to let the compiler know the loop is safe to parallelize Use the following C pragma to tell the compiler that for a given loop the last value computed for all scalars make it safe to parallelize the loop cpgi l safe _lastval In addition a command line option Msafe_lastval provides this information for all loops within the routines being compiled essentially providing global scope unroll nounroll The directive nounro11 is used to disable loop unrolling and unro11 to enable unrolling The directive takes arguments c and n A c specifies that c complete unrolling should be turned on or off An n specifies that n count unrolling should be turned on or off In addition the following arguments may be added to the unroll directive unroll crv This sets the threshold to which c unrolling applies v is a constant a loop whose constant loop count is lt v is completely unrolled unroll n v 152 Chapter 7 This
245. ont version 3 0 1 with the addition of exceptions PGCC C optionally accepts a number of features erroneously accepted by cfront version 2 1 Using the b option PGCC C accepts these features which may never have been legal C but have found their way into some user s code Command line options provide full support of many C variants including strict standard conformance PGCC C provides command line options that enable the user to specify whether anachronisms and or cfront 2 1 compatibility features should be accepted Refer to Section C 4 for details on features that are not part of the ARM but are part of the ANSI C working draft X3J16 WG21 C 1 Anachronisms Accepted The following anachronisms are accepted when anachronisms are enabled when the p option is not used e overload is allowed in function declarations It is accepted and ignored e Definitions are not required for static data members that can be initialized using default initialization This anachronism does not apply to static data members of template classes they must always be defined e The number of elements in an array may be specified in an array delete operation The value is ignored e A single operator and operator function can be used to overload both prefix and postfix operations e The base class name may be omitted in a base class initializer if there is only one immediate base class e Assignment to this in constructors and de
246. onvention cccccccscceseessessceesceesceeeeeceaeesaeesseceenseenaes C TEMPLATE INSTANTIATION 11 1 Command Line control of template instantiation eeeecseeeeceeeeeceeeeeceseceeesecseeeeeneens 11 2 Pragma control of template instantiation eee eeeeceseeeceeceeeeceeeceeeeeaecatenecaeeeteneees 11 3 Automatic template instantiation ccccesceseceecseecsseeseeeseeseceeeeseenseenseeeseseceeceeenseenaes 11 4 Implicit inclusion 11 5 Template Libraries ic 0 5 cAceccceie cesses icles see ctacs ce cut e EE aa C NAME MANGLING 12T Types of Mans liti gc select oi E E Gaul eaten cae eee 12 2 Mangling Summary cenene i R E E E E ia 12 2 1 Type Name Mangling sssseesssssesseeeessrserssesersrseeressesessesersrsseesessteesseseeseseesseseesreseese 12 2 2 Nested Class Name Mangling s ssesesesesserseeseesesseeressesersreseenessreressesreseeseesesseeressese 12 2 3 Local Class Name Mangling ccceccesceseesseesceceeeseeeseeseeeeenseeeseenseceaeenaeeaecaeeneenaes 12 2 4 Template Class Name Mangling ccccccesecssecsceeeeseeeseeeeeeseecesecnseenseceaeeeeneesaeenaes RUN TIME ENVIRONMENT A A Programming Model sic feces bce a ten eg ice ie Ree ee as Table of Contents 203 204 205 207 209 210 213 214 215 215 215 215 216 217 217 A 2 Function Calling Sequence serina aa a a eiii 217 A 3 Functions Returning Scalars or No Value cccesecesessecssececeeeecseeeeeesee
247. oops lstval Do don t compute last values lstval L RG nolstval t Select optimization level R G Do don t treat pointer arguments as safe R G fe_lastval Parallelize when loop contains a scalar used not enabled L outside of loop safeptr Do don t ignore potential data dependencies nosafeptr L R G nosafeptr to pointers single Do don t convert float parameters to nosingle R G nosingle double 156 Chapter 7 Function Default Scope unroll Do don t unroll loops nounroll L RG nounroll vector Do don t perform vectorizations vector L RG novector vintr Do don t recognize vector intrinsics L RG novintr fcon nofcon This pragma alters the effects of the Mfcon command line option a M Language control The pragma instructs the compiler to treat non suffixed floating point constants as float rather than double By default all non suffixed floating point constants are treated as double safe nosafe By default the compiler assumes that all pointer arguments are unsafe That is the storage located by the pointer can be accessed by other pointers The forms of the safe pragma are pragma scope no safe pragma safe variable variable where scope is either global or routine When the pragma safe is not followed by a variable name or name list all pointer arguments appearing in a routine if scope is routine or all routines if scope is global will be treated as saf
248. option but be aware that this could have other ramifications on your program 10 14 2 Symbol Name Construction and Calling Example This section presents an example of the rules outlined in table 10 3 In the pseudocode used below saddr refers to the address of a data item while val refers to the value of that data item Subroutine and function names are converted into symbol names according to the rules outlined in table 10 3 Consider the following subroutine call call work ERR a b n where a is a double precision scalar b is a real vector of size n and n is an integer 200 Default The symbol name for the subroutine is constructed by pre pending an underscore converting to all upper case and appending an sign followed by an integer indicating the total number of bytes occupied by the argument list Byte counts for character arguments appear immediately following the corresponding argument in the argument list The following is an example of the pseudo code for the above call using Default conventions call WORK 20 saddr ERR 3 addr a addr b saddr n STDCALL The symbol name for the subroutine is constructed by pre pending an underscore converting to all lower case and appending an sign followed by an integer indicating the total number of bytes occupied by the argument list Character strings are truncated to the first character in the string which is passed by value as the first byte i
249. or reference use of structure field reference nonstandard form of constant amp alternate return mixed non character and CHARACTER elements in COMMON mixed non character and CHARACTER EQUIVALENCE type elements numeric and or character types in COMMON Mixed numeric and or character type EQUIVALENCE Possible use of before definition in Invalid qualifier or qualifier value in OPTIONS statement An illegal qualifier was found or a value was specified for a qualifier that does not expect a value In either case the qualifier for which the error occurred is indicated in the error message 198 19 9 in ALLOCATE DEALLOCATE Unaligned memory reference A memory reference occurred whose address does not meet its data alignment requirement 200 201 202 Messages Missing UNIT FILE specifier Illegal I O specifier Repeated I O specifier 245 203 FORMAT statement has no label 204 Syntax error unbalanced angle brackets 205 Illegal specification of scale factor The integer following or has been omitted or P does not follow the integer value 206 Repeat count is zero 207 Integer constant expected in edit descriptor 208 Period expected in edit descriptor 209 Illegal edit descriptor 210 Exponent width not used in the Ew dEe or Gw dEe edit descriptors 211 Internal I O not allowed in this I O statement 212 Illegal NAMELIST I O Namelist I O cannot be performed
250. ord or member reference was found in a context that is not supported For example the use of structures records or members within a data statement is disallowed 151 Empty STRUCTURE UNION or MAP STRUCTURE ENDSTRUCTURE UNION ENDUNION or MAP ENDMAP declaration contains no members 158 Alternate return not specified in SUBROUTINE or ENTRY An alternate return can only be used if alternate return specifiers appeared in the SUBROUTINE or ENTRY statements 159 Alternate return illegal in FUNCTION subprogram An alternate return cannot be used in a FUNCTION 160 ENDSTRUCTURE ENDUNION or ENDMAP does not match top 162 Not equal test of loop control variable and replaced with lt or gt test 163 Cannot data initialize member of ALLOCATABLE COMMON 164 Overlapping data initializations of An attempt was made to data initialize a variable or array element already initialized 165 appeared more than once as a subprogram A subprogram name appeared more than once in the source file The message is applicable only when an assembly file is the output of the compiler Messages 243 166 cannot be a common block and a subprogram A name appeared as a common block name and a subprogram name The message is applicable only when an assembly file is the output of the compiler 167 Inconsistent size of common block A common block occurs in more than one subprogram of a source file and its size is not identical The maximum
251. orizer may change the order of operations so that it can generate better code e g dot product Such transformations change the result of the Optimization Directives and Pragmas 149 computation due to roundoff error The noassoc directive disables these transformations This directive affects the compiler only when Mvect is enabled on the command line bounds nobounds This directive alters the effects of the Mbounds command line option This directive enables the checking of array bounds when subscripted array references are performed By default array bounds checking is not performed cncall nocncall Loops within the specified scope are considered for parallelization even if they contain calls to user defined subroutines or functions or if their loop counts do not exceed the usual thresholds A nocncall directive cancels the effect of a previous cncall concur noconcur This directive alters the effects of the Mconcur command line option The directive instructs the auto parallelizer to enable auto concurrentization of loops If concur is specified multiple processors will be used to execute loops which the auto parallelizer determines to be parallelizable The noconcur directive disables these transformations This directive affects the compiler only when Mconcur is enabled on the command line depchk nodepchk This directive alters the effects of the Mdepchk command line option When potential data dependen
252. ors contact your PGI representative B 3 1 Message Format Each message is numbered Each message also lists the line and column number where the error occurs A dollar sign in a message represents information that is specific to each occurrence of the message B 3 2 Message List 0000 Internal compiler error This message indicates an error in the compiler not a user error although a user can cause an internal error The severity may vary if it is informative or warning the compiler probably generated correct object code but there is no way to be sure Regardless of the severity or cause internal errors should be reported to PGI 0001 Source input file name not specified On the command line the source file name should be specified either before all the switches or after them 0002 Unable to open source input file The source file name misspelled the file not in current working directory or the file is read protected 0003 Unable to open listing file The user probably does not have write permission for the current working directory 228 Appendix B 0004 Unable to open object file The user probably does not have write permission for the current working directory 0005 Unable to open temporary file The compiler uses directory usr tmp or tmp in which to create temporary files If neither of these directories is available on the node on which the compiler is being used this error will occur 0006 Input
253. ortran 90 Aggregate Data Types Derived Types c ccscccsseesseeseeeseeeeeseeseeeereeeens 176 Table of Contents vii 92 Cand C2 Data Ty pes te 2ictesrestdece a ieee a a a E raa aeesthaaetenee st 177 9 21 Grand Grr Sealainn sgoes ccs sees stave E EEEE RA ENE 177 9 2 2 Cand C Aggregate Data Types mrc eia ie e aE KE AA EENE RAEE ENEN 179 9 2 3 Class and Object Data Lay ubenieore ii n E E A 179 02 A Aperegate Alignment a E P thoy PA ET E 180 0 27 Bit feld APANE eons iri ir A E E TENO AT EAR 181 9 2 6 Other Type Keywords in C and C iriiieracdeniiraniiidor Eea AE 182 INTER LANGUAGE CALLING 183 10 1 Overview of Calling Conventions c cccceccceeseessceseceecescesecesecesecseecaeecaeseseeeeeeeeeeerenrens 183 10 2 Inter language Calling Considerations cccceeccesscesscesecesecesecseecseeeseeeaeeneeeeeseneeneeenseees 183 10 3 Functions and Subroutines 0 0 ce e S AAA R Ae 185 10 4 Upper and Lower Case Conventions Underscores ccccesccssessteesseereeeseeeeeeeseeeseneeeerees 186 10 5 Compatible Data Types c38 8s cid Behn ae dea nai ae 186 10 5 1 Fortran Named Common BLOCKS escasenco ieni E 188 10 6 Argument Passing and Return Values ccccecssesssessceescesecesceeceseceaecsaeeseecaeeenseeeeeneeeneeeas 188 10 6 1 Passing by Valle SVAT coiecccecttecceictetscteeiecc oes i nerds se dense Seater i 189 106 2 Character Retumm Values lt i ated cenin ities tera nate telet eines te beiesee Ge
254. ot be profitable Default Table 3 4 shows the interaction between the O option g option and Mvect options Command line Options Table 3 4 Optimization and O g Mvect and Mconcur Options Optimize Debug M Optimization Option Option Option Level none none none none none Mvect none e ne O noneor g none 85 Optimize Debug M Optimization Option Option Option Level Olevel lt 2 Olevel lt 2 Unoptimized code compiled using the option O0 can be significantly slower than code generated at other optimization levels Like the Mvect option the Munroll option sets the optimization level to level 2 if no O or g options are supplied For more information on optimization see Chapter 2 Optimization amp Parallelization Usage In the following example since no optimization level is specified and a O option is specified the compiler sets the optimization to level 2 pgf90 O myprog f Cross reference g Mpgflag Names the executable file Use the o option to specify the filename of the compiler object file The final output is the result of linking Syntax o filename Where filename is the name of the file for the compilation output The filename must not have a f extension Default The compiler creates executable filenames as needed If you do not specify the o option the default filename is the linker output file a out Usage In the foll
255. ou should be aware of the role of high level languages e g Fortran C C and assembly language in the software development process and should have some level of understanding of programming The PGI compilers are available on a variety of 32 bit X86 or AMD64 hardware platforms and operating systems You need to be familiar with the basic commands available on your system Finally your system needs to be running a properly installed and configured version of the compilers For information on installing PGI compilers and tools refer to the installation instructions Compatibility and Conformance to Standards This guide describes how to use The Portland Group Compiler Technology PGI Fortran C and C compilers and program development tools In particular these include the PGF 77 PGF 90 PGHPF PGCC C and PGCC ANSI C compilers the PGPROF profiler and the PGDBG debugger These compilers and tools work in conjunction with a 32 bit X86 hereafter just X86 or 64 bit AMD64 hereafter just AMD64 assembler and linker You can use the PGI compilers and tools to compile debug optimize and profile serial and parallel applications for X86 Intel Pentium II III 4 AMD Athlon and AMD64 AMD Opteron Athlon64 processor based systems For further information refer to the following e American National Standard Programming Language FORTRAN ANSI X3 1978 1978 Preface 1 e American National Standard Programming Language FORTRAN ANSI X3
256. ous section for a complete description of the Default convention 10 14 4 Using the STDCALL Calling Convention Using the STDCALL calling convention requires the insertion of a compiler directive into the declarations section of any Fortran program unit which calls the STDCALL program unit This directive has no effect when the Vunix compilation flag is used meaning you cannot mix UNIX style argument passing and STDCALL calling conventions within the same file Syntax for the directive is as follows MSSATTRIBUTES STDCALL work Where work is the name of the subroutine to be called using STDCALL conventions More than one subroutine may be listed separated by commas See Section 0 14 2 Symbol Name Construction and Calling Example for a complete description of the implementation of STDCALL Inter language Calling 201 Note The directive prefix DECS is also supported but requires a space between the prefix and the directive keyword ATTRIBUTES The must begin the prefix when compiling using Fortran 90 freeform format The characters C or can be used in place of in either form of the prefix when compiling used fixed form F77 style format The directives are completely case insensitive 10 14 5 Using the C Calling Convention Using the C calling convention requires the insertion of a compiler directive into the declarations section of any Fortran program unit which calls the C program unit This directive has no effect when t
257. override this with O3 while maintaining all other elements of fast simply compile as follows pgf90 fast 03 prog f Optimization amp Parallelization 23 2 4 Loop Unrolling Munroll This optimization unrolls loops executing multiple instances of the loop during each iteration This reduces branch overhead and can improve execution speed by creating better opportunities for instruction scheduling A loop with a constant count may be completely unrolled or partially unrolled A loop with a non constant count may also be unrolled A candidate loop must be an innermost loop containing one to four blocks of code The following shows the use of the Munroll option pgf90 Munroll prog f The Munroll option is included as part of fast on all X86 and AMD64 targets The loop unroller expands the contents of a loop and reduces the number of times a loop is executed Branching overhead is reduced when a loop is unrolled two or more times since each iteration of the unrolled loop corresponds to two or more iterations of the original loop the number of branch instructions executed is proportionately reduced When a loop is unrolled completely the loop s branch overhead is eliminated altogether Loop unrolling may be beneficial for the instruction scheduler When a loop is completely unrolled or unrolled two or more times opportunities for improved scheduling may be presented The code generator can take advantage of more possibil
258. owed currently 2048 228 EOF in comment The end of a file was encountered while processing a comment 229 EOF in macro call to The end of a file was encountered while processing a call to the indicated macro 230 EOF in string The end of a file was encountered while processing a quoted string 231 Formal parameters too long for Messages 247 The total length of the parameters in the definition of the indicated macro exceeded the maximum allowed currently 2048 232 Identifier too long The length of an identifier exceeded the maximum allowed currently 2048 234 Illegal directive name The sequence of characters following a directive prefix was not an identifier 235 Illegal macro name A macro name was not an identifier 236 Illegal number The indicated number contained a syntax error 237 Line too long The input source line length exceeded the maximum allowed currently 2048 238 Missing endif End of file was encountered before a required endif directive was found 239 Missing argument list for A call of the indicated macro had no argument list 240 Number too long The length of a number exceeded the maximum allowed currently 2048 241 Redefinition of symbol The indicated macro name was redefined 242 Redundant definition for symbol A definition for the indicated macro name was found that was the same as a previous definition 243 String too long The length of a
259. owing example the executable file is myprog instead of the default a out pgf90 myprog f o myprog Cross reference c E F S 86 Chapter 3 Syntax pc 32 64 80 The X86 architecture implements a floating point stack using 8 80 bit registers Each register uses bits 0 63 as the significand bits 64 78 for the exponent and bit 79 is the sign bit This 80 bit real format is the default format called the extended format When values are loaded into the floating point stack they are automatically converted into extended real format The precision of the floating point stack can be controlled however by setting the precision control bits bits 8 and 9 of the floating control word appropriately In this way you can explicitly set the precision to standard IEEE double precision using 64 bits or to single precision using 32 bits The default precision is system dependent To alter the precision in a given program unit the main program must be compiled with the same pc option The command line option pc val lets the programmer set the compiler s precision preference Valid values for val are 32 single precision 64 double precision 80 extended precision Extended Precision Option Operations performed exclusively on the floating point stack using extended precision without storing into or loading from memory can cause problems with accumulated values within the extra 16 bits of extended precision values Thi
260. ows directory or Windows systems directory Static libraries and DLLs also handle global data differently If two static libraries contain global data with the same name and both libraries are linked to the executable the global data item in the libraries will be resolved to the same memory location If this situation occurs with two DLLs however the global data items in each DLL are resolved to separate memory locations The PGI runtime DLLs can be used to create both executables and other DLLs The following switches apply Mdil Link with the DLL version of the runtime libraries This flag is required when Libraries 165 linking with any DLL built by the PGI compilers Mmakedll Generate a dynamic link library or DLL Mnopgdllmain Do not link the module containing the default D Main into the DLL This flag applies to building DLLs with the PGF90 and PGHPF compilers If you want to replace the default D Main routine with a custom Di Main use this flag and add the object containing the custom D Main to the link line The latest version of the default D Main is included in the Release Notes the code in this routine specific to PGF90 and PGHPF must be incorporated into the custom version of DilMain to ensure the appropriate function of your DLL o lt file gt Passed to the linker Name the DLL lt file gt output def lt file gt Passed to linker Generate a def named lt file gt for the DLL The
261. produces the X86 machine instruction for computation unless in C the Mnobuiltin switch is set arguments are passed on the stack which results in a memory store and load Finally Kieee also disables reciprocal division for constant divisors That is for a b with unknown a and constant b the expression is usually converted at compile time to a 1 b thus 88 Chapter 3 turning an expensive divide into a relatively fast scalar multiplication However numerical discrepancies can occur when this optimization is used Understanding and correctly using the pc Mnobuiltin and Kieee switches should enable you to produce the desired and expected precision for calculations which utilize floating point operations Usage pgf90 pc 64 myprog c Selects variations for compilation There are four uses for the O option Syntax Qdirdirectory The first variety using the dir keyword lets you supply a directory parameter that indicates the directory where the compiler driver is located Qopt ionprog opt The second variety using the option keyword lets you supply the option opt to the program prog The prog parameter can be one of pgftn as Or ld Qpat hpathname The third Q variety using the path keyword lets you supply an additional pathname to the search path for the compiler s required o files Qproducesourcetype The fourth O variety using the produce keyword lets you choose a stop after locatio
262. prog is necessary Just recreate obj 12 d11 and the new obj 12 d11 will be loaded at runtime Example 2 Build two DLLs when each DLL is dependent on the other and use them to build the main program In the following source files obj ect3 c makes calls to routines defined in object4 c and vice versa This situation of mutual imports requires two steps to build each DLL object3 c extern void func_4b void void func _3a void printf func 3a calling a routine in obj4 dll1l n func_4b void func _3b void printf func _3b n object4 c extern void func_3b void void func 4a void printf func _4a calling a routine in obj3 dll1l n func_3b void func _4b void printf func _4b n Libraries 169 prog2 c extern void func_3a void extern void func _4a void int main func_3a func_4a Step 1 To make 0b 3 d11 and obj 4 d11 first compile the source and create an import library for each DLL that will be built pgcc c object3 c pgcc object3 0 Mmakedll o obj3 d1l out implib obj3 1lib Creating library file obj3 lib object3 o0 text 0x24 object3 c undefined reference to func _4b Q Q pgcc c object4 c pgcc object4 0 Mmakedll o obj4 dl1l out implib obj4 1lib Creating library file obj4 lib object4 o text 0x24 object4 c undefined reference to func_3b Q Q The undefined reference errors are to be expected in
263. program include lt omp h gt int omp_in parallel void returns non zero if called from within a parallel region and zero if called outside of a parallel region When called from within a parallel region that is serialized for example in the presence of an if clause evaluating to zero the function will return zero 142 Chapter 6 include lt omp h gt void omp_set_dynamic int dynamic_threads is designed to allow automatic dynamic adjustment of the number of threads used for execution of parallel regions This function is recognized but currently has no effect include lt omp h gt int omp_ get dynamic void is designed to allow the user to query whether automatic dynamic adjustment of the number of threads used for execution of parallel regions is enabled This function is recognized but currently always returns zero include lt omp h gt void omp_set_nested int nested is designed to allow enabling disabling of nested parallel regions This function is recognized but currently has no effect include lt omp h gt int omp_get_nested void is designed to allow the user to query whether dynamic adjustment of the number of threads available for execution of parallel regions is enabled This function is recognized but currently always returns zero include lt omp h gt void omp_init_lock omp_lock_t lock void omp_init_nest_lock omp_nest_lock_t lock initializes a lock associated with the variable l
264. prototyped functions this pragma results in non ANSI conforming code 7 6 Scope of C C Pragmas and Command Line Options This section presents several examples showing the effect of pragmas and the use of the pragma scope indicators Note during compilation a pragma either turns an option on or turns an option off Pragmas apply to the section of code corresponding to the specified scope that is the entire file the following loop or the following or current routine For pragmas that have only routine and global scope there are two rules for determining the scope of a pragma We cover these special scope rules at the end of this section In all cases pragmas override a corresponding command line option Consider the program main float a 100 100 b 100 100 c 100 100 int time maxtime n i Jj maxtime 10 n 100 for time 0 time lt maxtime timet for j 0 j lt n j for i 0 i lt n i 158 Chapter 7 c i j a i j b i j When this is compiled using the Mvect command line option both interior loops are interchanged with the outer loop Pragmas alter this behavior either globally or on a routine or loop by loop basis To assure that vectorization is not applied use the novector pragma with global scope main pragma global novector float a 100 100 b 100 100 c 100 100 int time maxtime n i j maxtime 10 n 100 for time 0 time lt maxtime timet t for j 0 j lt n j
265. put Default The compiler produces an executable file Usage In the following example the compiler displays the preprocessed myprog f on the standard output pgf90 E myprog f Cross reference See the options C c Mkeepasm o F S 54 Chapter 3 l Stops compilation after the preprocessing phase Use the F option to halt the compilation process after preprocessing and write the preprocessed output to the file filename f where the input file is filename F Default The compiler produces an executable file Usage In the following example the compiler produces the preprocessed file myprog fin the current directory pgf90 F myprog F Cross reference c E Mkeepasm o S fast A generally optimal set of options is chosen depending on the target system Always includes the options O2 Munroll and Mnoframe on all X86 and AMD64 platforms In addition the appropriate tp option is automatically included to enable generation of code optimized for the type of system on which compilation is performed Note Auto selection of the appropriate tp option means that programs built using the fast option on a given system is not necessarily backward compatible with older systems Cross reference O Munroll Mnoframe tp flags Displays driver options on the standard output Use this option with v to list options that are recognized and ignored as well as t
266. put files with unrecognized extensions or no extension are also passed to the linker Files with a F Capital F suffix are first preprocessed by the Fortran compilers and the output is passed to the compilation phase The Fortran preprocessor functions similar to cpp for C C programs but is built in to the Fortran compilers rather than implemented through an invocation of cpp This ensures consistency in the pre processing step regardless of the type or revision of operating system under which you re compiling Any input files not needed for a particular phase of processing are not processed For example if on the command line you use an assembly language file filename s and the S option to stop before the assembly phase the compiler takes no action on the assembly language file Processing stops after compilation and the assembler does not run in this case compilation must have been completed in a previous pass which created the s file Refer to the following section Output Files for a description of the S option In addition to specifying primary input files on the command line code within other files can be compiled as part of include files using the INCLUDE statement in a Fortran source file or the preprocessor include directive in Fortran source files that use a F extension or C and C source files When linking a program module with a library the linker extracts only those library modules that the program needs
267. r 4 real double precision integer 2 external cfunc call cfunc booll amp numfloatl numdoubl write L2 booll letterl numdoubl numshorl end booll letterl numintl numint2 numfloatl numdoub1 numshorl letterl numintl numint2 numshor1 Loyo Ebr Lg Erla LO e numintl numint2 numfloatl Example 10 3 Fortran Main Program fmain f define TRUE Oxff define FALSE 0 void cfunc_ booll letterl numdoubl numshorl char booll letterl Lars numintl numint2 float numfloatl double numdoubl short numshorl Int len_letter1 Inter language Calling numintl numint2 numfloat1 len_letter1 191 booll TRUE letterl v numintl 11 numint2 44 numfloatl 39 6 numdoubl 39 2 numshorl 981 Example 10 4 C function cfunc_ Compile and execute the program fmain f with the call to cfunc_ using the following command lines pgcc c cfunc c pgf90 cfunc o fmain f Executing the a out file should produce the following output T v 11 44 39 6 39 2 981 10 9 Example C Calling Fortran Example 10 6 shows a C main program that calls the Fortran subroutine shown in Example 10 5 Notice that each call uses the amp operator to pass by reference Also notice that the call to the Fortran subroutine uses all lower case and a trailing _ subroutine forts booll letterl numintl amp numint2 numfloatl numdoubl numshorl1 logical 1
268. r internal consistency of the X86 FP Miscellaneous stack in the prologue of a function and after returning from a function or subroutine call chkptr check for NULL pointers pg 90 and pghpf Miscellaneous only chkstk check the stack for available space upon entry to Miscellaneous and before the start of a parallel region Useful when many private variables are declared concur enable auto concurrentization of loops Multiple Optimization processors will be used to execute parallelizable loops only valid on shared memory multi CPU systems cray Force Cray Fortran CF77 compatibility Optimization pg 77 pgf90 and pghpf only 62 Chapter 3 Description Category determines whether all program variables must Fortran be declared pg 77 pgf 90 and pghpf only Language no defaultunit determines how the asterisk character is Fortran treated in relation to standard input and standard Language output regardless of the status of I O units 5 and 6 pgf77 pgf90 and pghpf only no depchk checks for potential data dependencies no dlines determines whether the compiler treats lines Fortran containing the letter D in column one as Language executable statements pg 77 pgf90 and pghpf only dil Link with the DLL version of the runtime Miscellaneous libraries Win32 only dollar specifies the character to which the compiler Fortran maps the dollar sign code pgf 77 pgf90 and Language pghpf only ext
269. ragma each thread s copy is initialized once at an unspecified point prior to its first use as the master copy would be initialized in a serial execution of the program The following restrictions apply to the omp _threadprivate pragma e The omp threadprivate pragma must appear after the declaration of every threadprivate variable included in list e It is illegal for an omp threadprivate variable to appear in any clause other than a copyin schedule or if clause e Ifa variable is specified in an omp threadprivate pragma in one translation unit it must be specified in an omp threadprivate pragma in every translation unit in which it appears e The address of an omp _threadprivate variable is not an address constant e Anomp threadprivate variable must not have an incomplete type or a reference type 6 15 Run time Library Routines User callable functions are available to the OpenMP C C programmer to query and alter the parallel execution environment Any program unit that invokes these functions should include the statement include lt omp h gt The omp h include file contains definitions for each of the C C library routines and two required type definitions Optimization MP Pragmas for C and C 141 include lt omp h gt int omp_get_num_ threads void returns the number of threads in the team executing the parallel region from which it is called When called from a serial region this function returns 1 A nested par
270. ray types align according to the alignment of the array elements For example an array of INTEGER 2 data aligns on a 2 byte boundary Structures Unions align according to the alignment of the most restricted data type of the structure or union In the next example the union aligns on a 4 byte boundary since the alignment of c the most restrictive element is four STRUCTURE ASTR UNION AP INTEGER 2 A 2 bytes END MAP AP BYTE B 1 byte END MAP AP INTEGER 4 C 4 bytes END MAP END UNION END STRUCTURE Structure alignment can result in unused space called padding between members of the structure internal padding between the last member and the end of the space occupied by the structure The padding at the end of the structure is called tail padding The offset of a structure member from the beginning of the structure is a multiple of the member s alignment For example since an INTEGER 2 aligns on a 2 byte boundary the offset of an INTEGER 2 member from the beginning of a structure is a multiple of two bytes 9 1 3 Fortran 90 Aggregate Data Types Derived Types The Fortran 90 standard added formal support for aggregate data types The TYPE statement begins a derived type data specification or declares variables of a specified user defined type For example the following would define a derived type ATTENDEE TYPE ATTENDEE CHARACTER LEN 30 NAME 176 Chapter 9 CHARACTER LEN 30 ORGA
271. rence pointer type has additional type qualifiers above those present in the pointer value For example int p const int amp r p No temporary used e A reference may be initialized with a null 266 Appendix C Index Auto parallelization cee ceeeeeeneeeeeeee 31 Basic DIOCK eee eeeeeecrseseeecneeeeeeeneees 19 Bounds checking cccesceeseeseeeteetees 80 C C Builtin Functions cece 163 C C Math Header File 163 C C Parallelization Pragmas Omp AtOMIC eH 05 cc ceiscceeessteiesseeeeseeees 140 Op US heee as 141 omp parallel for ssseseseeseeeesseeeeseeeee 138 omp parallel sections ceceee 139 OMP Section Sine oii ie 138 omp threadprivate sesseeseeeesseeeeseeeee 141 C Name Mangling s eseseeeeseeeseeeeee 213 C Standard Template Library 171 C Template Instantiation 0 0 0 0 203 AULOMALIC cect za Heese lies 207 command line control 0 0 0 eee 204 implicit inclusion ec eeeeeeeeeeeees 209 pragma control eeeseeseereeereeeeees 205 C Template Libraries 0 eee 210 Cache Tiling failed cache tiling eeeeeeeeeeeeeees 81 Command line options MMOTLeECUISIVE 00 eect s 67 Index Mnoreentr nt sinisini 67 Mnoref externals 0 0 0 0 eeeeeesteereees 67 Mnosecond underscore ccce 67 MTOCUTSIVE arae tei e ATA 67 lt Mreentranit msanii eein 67 Mref_externals
272. rly aligned the stack pointer should always point to a double word boundary Structure and union arguments are pushed onto the stack in the same manner as integral arguments described above This provides call by value semantics letting the called function modify its arguments without affecting the calling functions object 222 Appendix A Table A 6 Structure and Union Arguments Argument Stack Address IZ a eS orans e ee Implementing a Stack In general compilers and programmers must maintain a software stack Register esp is the stack pointer Register esp is set by the operating system for the application when the program is started The stack must be a grow down stack A separate frame pointer enables calls to routines that change the stack pointer to allocate space on the stack at run time e g alloca Some languages can also return values from a routine allocated on stack space below the original top of stack pointer Such a routine prevents the calling function from using sesp relative addressing to get at values on the stack If the compiler does not call routines that leave esp in an altered state when they return a frame pointer is not needed and is not used if the compiler option Mnoframe is specified Although not required the stack should be kept aligned on 8 byte boundaries Each routine should use stack space in multiples of 8 bytes PGI s compilers allocate stack space in multiples of 8 bytes Variable L
273. rocessed because it has a F filename extension 1 4 Parallel Programming Using the PGI Compilers The PGI compilers support three styles of parallel programming e Automatic shared memory parallel programs compiled using the Mconcur option to pg 77 pgf90 pgcc or pgcC parallel programs of this variety can be run on shared memory parallel SMP systems such as dual processor workstations e User directed shared memory parallel programs compiled using the mp option to pgf77 pgf90 pgcc or pgccC parallel programs of this variety can be run on SMP systems Carefully coded user directed parallel programs using OpenMP directives can often achieve significant speed ups on large numbers of processors on SMP server systems Chapter 5 Open MP Directives for Fortran and Chapter 6 OpenMP Pragmas for C and C contain complete descriptions of user directed parallel programming e Data parallel shared or distribute memory parallel programs compiled using the PGHPF High Performance Fortran compiler parallel programs of this variety can be run on SMP workstations or servers distributed memory clusters of workstations or clusters of SMP workstations or servers Coding a data parallel version of an application can be more work than using OpenMP directives but has the advantage that the resulting executable is usable on all types of parallel systems regardless of whether shared memory is available See the PGHPF User s Guide for a complete des
274. running 64 bit Linux at http www x86 64 org abi pdf A 1 Programming Model This section defines compiler and assembly language conventions for the use of certain aspects of the processor These standards must be followed to guarantee that compilers applications programs and operating systems written by different people and organizations will work together The conventions supported by the PGCC ANSI C compiler implement the application binary interface ABI as defined in the System V Application Binary Interface Intel Processor Supplement and the System V Application Binary Interface listed in the Related Publications section in the Preface A 2 Function Calling Sequence This section describes the standard function calling sequence including the stack frame register usage and parameter passing Register Assignment Table A 1 defines the standard for register allocation The 32 bit X86 Architecture X86 provides a number of registers All the integer registers and all the floating point registers are global to all procedures in a running program Table A 1 Register Allocation Type Name Purpose General integer return value dividend register for divide operations Coo ae Run time Environment 217 Type Name Purpose Po tesp stack pointer S floating point stack top return value ptt floating point next to stack top ee ee eee pst floating point stack bottom In addition to the registers each
275. s 54ers cake eile 175 F90 derived types ceccescceseesseereees 176 Fortra sno e R EE 173 internal padding cceeeeeseeseeeteeee 180 tail padding eeeeceesseeseeeeeeeeeeeees 180 Directives Fortrade ee tes coveencevedivees 9 optimization eeeeeeeseeseeseeseeeseeeees 147 Parallelization 00 0 eseeeeeceeeeeeeeeeeeees 111 SCOPE tein Anata ences 153 Environment Variables MPS TAZ S ll tegcesses iets 16 128 145 OMP DYNAMIC eee 128 144 145 OMP NESTED eeeeeeteeeeeeeeeees 128 145 OMP_NUM_ THREADS 128 144 Index OME SCHEDULE porrer 128 Filename Conventions 10 CXteNSiONS Acie erann h 10 Input Fil s sc outs cccsdenieisachvaies 10 Output PES r reana eiar 11 Floating point stack 00 eee eeeeeeeeeeeneeeees 87 Fortran Parallelization Directives ATOMI Orii seceetesceneoncureedistenctes 124 140 CRITICAL END CRITICAL 115 DOACROS Senini i eiie a 121 FEUSH ege a e R 125 PARALLEL DO uu eects 121 PARALLEL SECTIONS ee 123 SECTIONS END SECTIONS 122 THREADPRIVATE cece 125 Function inlining inlining and makefiles eee eee 108 Function Inlining inlining examples ccseeseseeteees 109 inlining restrictions eee eens 109 Inter language Calling eee 183 PVA Dennen niei ies 189 arguments and return values 188 array indices 0 eeeseeeseeseeeeeseeereeeneeees 190 C calling CH se eon hid 194 C calling Fortran
276. s are needed and therefore computes their last values The directive nolstval directs the compiler not to compute the last values for those cases safe_lastval During parallelization scalars within loops need to be privatized Problems are possible if a scalar is accessed outside the loop For example do i 1 N LE EAI gt 50 t x i enddo v t creates a problem since the value of t may not be computed on the last iteration of the loop Ifa scalar assigned within a loop is used outside the loop we normally save the last value of the scalar Essentially the value of the scalar on the last iteration is saved in this case when i N If the loop is parallelized and the scalar is not assigned on every iteration it may be difficult to determine on what iteration t is last assigned without resorting to costly critical sections Analysis allows the compiler to determine if a scalar is assigned on every iteration thus the loop is safe to parallelize if the scalar is used later An example loop is do i 1 N x 1 TECE O80 t 2 0 else VO He ibe Optimization Directives and Pragmas 151 where t is assigned on every iteration of the loop However there are cases where a scalar may be privatizable If it is used after the loop it is unsafe to parallelize Examine this loop do i l x N i gt 0 0 ati if i Va DE ENE endif enddo vet where each use of t within the loop is reached by a definiti
277. s can lead to answers when rounded that do not match expected results For example if the argument to sin is the result of previous calculations performed on the floating point stack then an 80 bit value used instead of a 64 bit value can result in slight discrepancies Results can even change sign due to the sin curve being too close to an x intercept value when evaluated To maintain consistency in this case you can assure that the compiler generates code that calls a function According to the X86 ABI a function call must push its arguments on the stack in this way memory is guaranteed to be accessed even if the argument is an actual constant Thus even if the called function simply performs the inline expansion using the function call as a wrapper to sin has the effect of trimming the argument precision down to the expected size Using the Vnobuiltin option on the command line for C accomplishes this task by resolving all math routines in the library ibm performing a function call of necessity The According to Intel documentation this only affects the operations of add subtract multiply divide and square root Command line Options 87 other method of generating a function call for math routines but one that may still produce the inline instructions is by using the Kieee switch A second example illustrates the precision control problem using a section of code to determine machine precision program find precision
278. s facilitate this pre link step The option one_instantiation_per_object is used on each c file to put each template instantiation in its own object so that the linker may access it independently The option template_dir is used only if the user wants direct control over the template directory name The option prelink_objects is used in combination with one_instantiation_per_object on the list of archivable objects to generate the instantiations For example a makefile that looks as follows CC pgCC CCFLAGS 02 filel o filel cc CC CCLAGS c filel cc file2 o file2 cc CC S CCLAGS c file2 cc libX a filel o file2 o ar cr libX a filel o file2 o might be modified to look as follows 210 CC pgCC CCFLAGS 02 one instantiation per object filel o filel cc CC CCLAGS c filel cc file2 o file2 cc S CC S CCLAGS c file2 cc libX a filel o file2 o This implies one_instantiation per object prelink_objects Chapter 11 CC S CCFLAGS prelink objects filel o file2 o Template dir o contains the templates instantiated above ar cr libX a filel o file2 o Template dir o C Template Instantiation 211 Chapter 12 C Name Mangling Name mangling transforms the names of entities so that the names include information on aspects of the entity s type and fully qualified name This is necessary since the intermediate language into which a program is translated
279. s of template entities and for the added information about entities that could be instantiated 3 Ifpgprelnk finds a reference to a template entity for which there is no definition anywhere in the set of object files it looks for a file that indicates that it could instantiate that template entity When it finds such a file it assigns the instantiation to it The set of instantiations assigned to a given file say abc C is recorded in an associated ii file for example abc ii 4 The pgpreink then executes the compiler again to recompile each file for which the ii file was changed 5 When the compiler compiles a file it reads the 1i file for that file and obeys the instantiation requests therein It produces a new object file containing the requested template entities and all the other things that were already in the object file 6 pgprelnk repeats steps 3 through 5 until there are no more instantiations to be adjusted 7 The object files are linked together Once the program has been linked correctly the ii files contain a complete set of instantiation assignments From then on whenever source files are recompiled the compiler will consult the ii files and do the indicated instantiations as it does the normal compilations That means that except in cases where the set of required instantiations changes the pgprelnk step from then on will find that all the necessary instantiations are present in the object files and no in
280. s outline how to use Mmakedll to do so Example 1 Build a DLL out of two source files object1 f and object2 f and use it to build the main source file prog1 f Step 1 object1 f subroutine subf1 integer n n 1 print n n return end object2 f function funf2 real funf2 funf2 2 0 return end progl f program test external subfl real funf2 val integer n call subf1 n val funf2 write val stop end Libraries 167 Step 2 Create the DLL 0b3j12 d11 and its import library 06312 1ib using the following series of commands pgf90 c objectl f object2 f pgf90 objectl o object2 0 Mmakedll o obj12 dll out implib obj12 1lib o a Step 3 Compile the main program pgf90 Mdll o progl progl f L lobj12 The Mdil switch causes the compiler to link against the PGI runtime DLLs instead of the PGI runtime static libraries The Mdl switch is required when linking against any PGI compiled DLL such as obj312 d11 The switch is used to specify that obj312 1ib the DLL s import library will be used to resolve the calls to subf1 and funf2 in prog1 f 168 Chapter 8 Step 4 Ensure that obj 12 d11 is in your path then run the executable prog to determine if the DLL was successfully created and linked progl n 1 val 2 000000 FORTRAN STOP Should you wish to change obj 12 d11 without changing the subroutine or function interfaces no rebuilding of
281. s provides a list of compiler error messages C Dialect Supported lists more details of the version of the C language that PGCC C supports Hardware and Software Constraints This guide describes versions of the PGI compilers that produce assembly code for X86 and AMD64 processor based systems Details concerning environment specific values and defaults and system specific features or limitations are presented in the release notes sent with the PGI compilers Preface Conventions This User s Guide uses the following conventions italic Constant Width item item2 item 3 filename FORTRAN is used for commands filenames directories arguments options and for emphasis is used in examples and for language statements in the text including assembly language statements square brackets indicate optional items In this case item is optional braces indicate that a selection is required In this case you must select either item2 or item3 ellipsis indicate a repetition Zero or more of the preceding item may occur In this example multiple filenames are allowed Fortran language statements are shown in the text of this guide using upper case characters and a reduced point size The following table lists the PGI compilers and tools and their corresponding commands Table P 1 PGI Compilers and Commands Compiler Language Command PGF77 FORTRAN 77 pgf77 PGHPF High Performance Fortran PG
282. s variable is 1 For historical reasons the environment variable NCPUS is supported with the same functionality In the event that both omp_NUM_THREADS and NCPUS are defined the value of omPp_NUM_THREADS takes precedence Note OMP_NUM_THREADS threads will be used to execute the program regardless of the number of physical processors available in the system As a result you can run programs using more threads than physical processors and they will execute correctly However performance of programs executed in this manner can be unpredictable and oftentimes will be inefficient OMP_SCHEDULE specifies the type of iteration scheduling to use for Do and PARALLEL DO loops which include the SCHEDULE RUNTIME clause The default value for this variable is static If the optional chunk size is not set a chunk size of 1 is assumed except in the case of a STATIC schedule For a static schedule the default is as defined in Section 5 6 DO END DO Examples of the use of oMP_ SCHEDULE are as follows setenv OMP_SCHEDULE STATIC 5 setenv OMP_ SCHEDULE GUIDED 8 setenv OMP_ SCHEDULE DYNAMIC OMP_ DYNAMIC currently has no effect OMP_ NESTED currently has no effect MPSTKzZ increase the size of the stacks used by threads executing in parallel regions It is for use with programs that utilize large amounts of thread local storage in the form of private variables or local variables in functions or subroutines cal
283. sesreserstsseeressesesesersesseenesseseesse 107 4 2 2 Updating Inline Libraries Makefile 0 0 ccccescceseesseesceeeceeeeeeeceeeececeecneeneeneenseenaes 108 4 3 Error Detection during Inlining ccceececcesecsseeseeseeesceeeceeeceseeeeeeeecnecneenseenaeensecaeeaeeenes 109 A A EXAM PleSt 5 ciecscetest docs E O E gtd ndtenc decide A E ech Wace eee 109 4 5 Restrictions on Imlining 2c 3 25 ccccccecee sec seceseedeacsted seventesccecetdsccesssgeddedscsedescesseuvecesdeeeeeitesedeescee 109 OPEN MP DIRECTIVES FOR FORTRAN 111 5 1 Parallelization Directives ss 2 csccscecocesccvccsdescadeacs de ccececca ce vsnc evcd e i E eii ii 111 5 2 PARALLEL END PARALLEL w eeececccceccessesssseeesceseescaeeseesecaeeseesecaecsesseeeseaeesesaeenenaes 112 5 3 CRITICAL END CRITICAL aiaia aE Aa SEAE E aa 115 SASMASTER END MASTER ovadonar onekana aa a a 116 5 5 SINGLE s END SINGLE reiii e A E E RAEE A NS 117 Table of Contents v 5 6 DO wv END DO vis ieisoseetecssperndseossd a a a a a sh cbeyeesdeostd suentuecantettvsto 118 5 7 BARRIER seiess Gis sires e a E T EE E EE A A heeds 120 5 8 DOACROSS inon iana an EEEO RE EREEREER EE bts 121 5 9 N OAN A D S D DI O E T E ETE 121 51O SECTIONS s END SECTIONS hree ee a a E E akg Anes ah 122 SILPARAELELE SECTIONS S enai r E E E A T E a a 123 5 12 ORDERED nig o n ieri TE AE AEREE E KE N E R 124 MA OME a a e aa A AAA 124 DAA BLUSH BEEE RTE A AE EE TA 125 II THREADPRIVYA TE rieton e r a A ATE 125
284. she 189 10 6 3 Complex Return Valles 2i h0 4 en teeta hhh kA ek oh en Oe 190 10 7 Array Indices sccccicecciscesiecde tine ects ca cians Cdescets ia E E stadia cd aa a ots 190 10 8 Example Fortran Calling C cccccecccsscesseessessceesceeeeeecneecesecaecsaecseecaeeeaeseneeeeeeeeeenreeerens 191 10 9 Example C Calling Fortran ccccccsssssssesssesseesesscesssessecsoesnaeseneconsconsensesnsesnsensesssesseees 192 10 10 Example lt G 4 5 Calling 2c cccccccseek tei ce o ee 193 10 11 Example C Calling C roisean aai a i a 194 10 12 Example Fortran Calling C eseseseesseeeessesesssesersesseerssstsressesersteseesessesresseseseseeseesee 195 viii Table of Contents 10 13 Example C Calling Fortran cccccesccesecssecseecseeeseeeeeeseceeeesecnseeneenseesaeeneecseeeneeenes 10214 Win32 Calling Convento oi icce cccecesecc ects cbcceeteendeciy e e a 10 14 1 Win32 Fortran Calling Conventions ccceccecsseeeessceeeceeecesceecesecesecesecnsecaeeaeeenes 10 14 2 Symbol Name Construction and Calling Example cccccccccceeseeeseseeeeeeeeseeeseenees 10 14 3 Using the Default Calling Convention cccccccsceesseeceeseeesceeeeeseceeeeenseeeeenseenaes 10 14 4 Using the STDCALL Calling Convention cccccccsecsceesceeseeeeeeeeceseesseenseeneeeeeenes 10 14 5 Using the C Calling Convention 0 ccecccesecssecsseesseseeeeseeeeceeseeseenseceaeenaeensecseenaeenaes 10 14 6 Using the UNIX Calling C
285. sion are integer and Messages 235 0070 Incorrect sequence of statements The statement order is incorrect For instance an IMPLICIT NONE statement must precede a specification statement which in turn must precede an executable statement 0071 Executable statements not allowed in block data 0072 Assignment operation illegal to The destination of an assignment operation must be a variable an array reference or a vector reference The assignment operation may be by way of an assignment statement a data statement or the index variable of an implied DO loop The compiler has determined that the identifier used as the destination is not a storage location The error message attempts to indicate the type of entity used Entry point An assignment to an entry point that was not a function procedure was attempted External procedure An assignment to an external procedure or a Fortran intrinsic name was attempted an external procedure if the identifier is the name of an entry point that is not a function 0073 Intrinsic or predeclared cannot be passed as an argument 0074 Illegal number or type of arguments to The indicated symbol is an intrinsic or generic function or a predeclared subroutine or function requiring a certain number of arguments of a fixed data type 0075 Subscript substring or argument illegal in this context for This can happen if you try to doubly index an array such as ra 2 3 This a
286. size is chosen The message is applicable only when an assembly file is the output of the compiler 168 Incompatible size of common block A common block occurs in more than one subprogram of a source file and is initialized in one subprogram Its initialized size was found to be less than its size in the other subprogram s The message is applicable only when an assembly file is the output of the compiler 169 Multiple data initializations of common block A common block is initialized in more than one subprogram of a source file Only the first set of initializations apply The message is applicable only when an assembly file is the output of the compiler 170 F77 extension Use of a nonstandard feature A description of the feature is provided 171 F77 extension nonstandard statement type 172 F77 extension numeric initialization of CHARACTER A CHARACTER variable or array element was initialized with a numeric value 173 F77 extension nonstandard use of data type length specifier 174 F77 extension type declaration contains data initialization 175 F77 extension IMPLICIT range contains nonalpha characters 244 Appendix B 176 190 197 F77 F77 F77 F77 F77 F77 F77 F77 ex ex ex ex ex ex ex ex Mixed tension tension tension tension tension tension tension tension nonstandard operator nonstandard use of keyword argument matrix vect
287. sors subsequent calls to omp get max threads will return the new value This function returns the maximum value whether executing from a parallel or serial region of code integer function omp_get_num procs returns the number of processors that are available to the program logical function omp_in parallel returns TRUE if called from within a parallel region and FALSE if called outside of a parallel region When called from within a parallel region that is serialized for example in the presence of an IF clause evaluating FALSE the function will return FALSE subroutine omp_set_dynamic scalar_logical_exp is designed to allow automatic dynamic adjustment of the number of threads used for execution of parallel regions This function is recognized but currently has no effect logical function omp_get_dynamic 126 Chapter 5 is designed to allow the user to query whether automatic dynamic adjustment of the number of threads used for execution of parallel regions is enabled This function is recognized but currently always returns FALSE subroutine omp_set_nested scalar_logical_exp is designed to allow enabling disabling of nested parallel regions This function is recognized but currently has no effect logical function omp_get_nested is designed to allow the user to query whether dynamic adjustment of the number of threads available for execution of parallel regions is enabled This function is
288. stantiation assignment adjustments need be done That s true even if the entire program is recompiled If the programmer provides a specialization of a template entity somewhere in the program the specialization will be seen as a definition by the pgpre nk step Since that definition satisfies whatever references there might be to that entity the pgpre nk program will see no need to request an instantiation of the entity If the programmer adds a specialization to a program that has previously been compiled the pgpre nk program will notice that too and remove the assignment of the instantiation from the proper ii file The ii files should not in general require any manual intervention One exception e Ifa definition is changed in such a way that some instantiation no longer compiles it gets errors and at the same time a specialization is added in another file and the first file is being recompiled before the specialization file and is getting errors e In this case the ii file for the file getting the errors must be deleted manually to allow the pgpreink to regenerate it 208 Chapter 11 If the pgpre nk changes an instantiation assignment it will issue a message like C pgprelnk f 10A_ pt__2 iFv assigned to file test o C pgprelnk executing usr pgi bin pgCC c test c The name in the message is the mangled name of the entity The automatic instantiation scheme can coexist with partial explicit control of instantiation by
289. stem 1 1 Overview In general using a PGI compiler involves three steps 1 Produce a program in a file containing a f extension or another appropriate extension see Section 7 3 1 Input Files This may be a program that you have written or a program that you are modifying 2 Compile the program using the appropriate compiler command 3 Execute debug or profile the executable file on your system The PGI compilers allow many variations on these general program development steps These variations include the following e Stop the compilation after preprocessing compiling or assembling to save and examine intermediate results e Provide options to the driver that control compiler optimization or that specifies various features or limitations e Include as input intermediate files such as preprocessor output compiler output or assembler output Getting Started 1 2 Invoking the Command level PGI Compilers To translate and link a Fortran C or C language program the pgf77 pgf90 pghpf pgcc and pgcc commands do the following e Preprocess the source text file e Check the syntax of the source text e Generate an assembly language file e Pass control to the subsequent assembly and linking steps For example if you enter the following simple Fortran program in the file hello f print hello end You can compile it from a UNIX shell prompt or BASH for Win32 prompt using the default pgf90 driver options
290. structors is allowed This is allowed only if anachronisms are enabled and the assignment to this configuration parameter is enabled C Dialect Supported 257 e A bound function pointer a pointer to a member function for a given object can be cast to a pointer to a function e A nested class name may be used as a non nested class name provided no other class of that name has been declared This anachronism is not applied to template classes e A reference to a non const type may be initialized from a value of a different type A temporary is created it is initialized from the converted initial value and the reference is set to the temporary e A reference to a non const class type may be initialized from an rvalue of the class type or a derived class thereof No additional temporary is used e A function with old style parameter declarations is allowed and may participate in function overloading as though it was prototyped Default argument promotion is not applied to parameter types of such functions when the check for compatibility is done so that the following declares the overloading of two functions named f int f int int f x char x return x It will be noted that in C this code is legal but has a different meaning a tentative declaration of f is followed by its definition e When nonconst_ref anachronism is enabled a reference to a nonconst class can be bound to a class rvalue of the same type or a derived type t
291. tant expressions 135 Missing STRUCTURE name field STRUCTURE name field is required on the outermost structure Messages 241 136 Field namelist not allowed The field namelist field of the STRUCTURE statement is disallowed on the outermost structure 137 Field namelist is required in nested structures 138 Multiply defined STRUCTURE member name A member name was used more than once within a structure 139 Structure in RECORD statement not defined A RECORD statement contains a reference toa STRUCTURE that has not yet been defined 140 Variable is not a RECORD 141 RECORD required on left of 142 is not a member of this RECORD 144 NEED ERROR MESSAGE This is used as a temporary message for compiler development 145 FILL only valid within STRUCTURE block The FILL special name was used outside of aSTRUCTURE multiline statement It is only valid when used within a STRUCTURE multiline statement even though it is ignored 146 Expression must be character type 147 Character expression not allowed in this context 148 Non record where aggregate record reference required An aggregate reference to a record was expected during statement compilation but another data type was found instead 149 Record where arithmetic value required 242 Appendix B An aggregate record reference was encountered when an arithmetic expression was expected 150 Structure Record or member not allowed in this context A structure rec
292. tch func or that have number or fewer statements For examples of extracting functions see Chapter 4 Function Inlining Minline func filename ext number levels number This passes options to the function inliner where except func name func filename ext Command line Options instructs the inliner to inline all eligible functions except func a function in the source text Multiple functions can be listed comma separated instructs the inliner to inline the function func The func name should be a non numeric string that does not contain a period You can also use a name prefix followed by the function name If name is specified what follows is always the name of a function instructs the inliner to inline the functions within the library file filename ext The compiler assumes that a filename ext option containing a period is a library file Create the library file using the Mextract option You can also use a lib prefix followed by the library name If lib is specified no period is necessary in the library name Functions from the specified library are inlined If no library is specified functions are extracted from a temporary 69 library created during an extract prepass Number instructs the inliner to inline functions with number or fewer statements You can also use a size prefix followed by a number If size is specified what follows is always taken as a number levels num
293. te RETURN statement Inter language Calling 185 10 4 Upper and Lower Case Conventions Underscores By default on UNIX systems all Fortran symbol names are converted to lower case C and C are case sensitive so upper case function names stay upper case When you use inter language calling you can either name your C C functions with lower case names or invoke the Fortran compiler command with the option Mupcase in which case it will not convert symbol names to lower case When programs are compiled using one of the PGI Fortran compilers on UNIX systems an underscore is appended to Fortran global names names of functions subroutines and common blocks This mechanism distinguishes Fortran name space from C C name space Use these naming conventions e Ifyou call a C C function from Fortran you should rename the C C function by appending an underscore or use CSPRAGMA C in the Fortran program refer to Chapter 7 Optimization Directives and Pragmas for details on CSPRAGMA C e If you call a Fortran function from C C you should append an underscore to the Fortran function name in the calling program 10 5 Compatible Data Types Table 10 1 shows compatible data types between Fortran and C C Table 10 2 shows how the Fortran COMPLEX type may be represented in C C If you can make your function subroutine parameters and return values match types you should be able to use inter language calling An exception is that
294. the programmer through the use of pragmas or command line specification of the instantiation mode 11 4 Implicit inclusion When implicit inclusion is enabled the front end assumes that if it needs a definition to instantiate a template entity declared in a h file it can implicitly include the corresponding c file to get the source code for the definition For example if a template entity asc is declared in file xyz h and an instantiation of ABC f is required in a compilation but no definition of ABC F appears in the source code processed by the compilation the compiler will look to see if a file xyz C exists and if so it will process it as if it were included at the end of the main source file To find the template definition file for a given template entity the front end needs to know the full path name of the file in which the template was declared and whether the file was included using the system include syntax e g include lt file h gt This information is not available for preprocessed source containing 1ine directives Consequently the front end will not attempt implicit inclusion for source code containing 1ine directives The set of definition file suffixes cc and c gt cc 39 ec 99 6c tried is c c cpp CPP cxx CXX Implicit inclusion works well alongside automatic instantiation but the two are independent They can be en
295. the Fortran main program Generation pg 77 pgf90 and pghpf only noopenmp when used in combination with the mp Miscellaneous option causes the compiler to ignore OpenMP parallelization directives or pragmas but still process SGI style parallelization directives or pragmas nopgdllmain Do not link the module containing the default Miscellaneous DilMain into the DLL Win32 only nosgimp when used in combination with the snp Miscellaneous option causes the compiler to ignore SGI style parallelization directives or pragmas but still process OpenMP directives or pragmas nostartup do not link in the standard startup routine Environment pg 77 pgf90 and pghpf only nostddef instructs the compiler to not recognize the Environment standard preprocessor macros nostdinc instructs the compiler to not search the standard Environment location for include files nostdlib instructs the linker to not link in the standard Environment libraries noonetrip determines whether each DO loop executes at Language least once pgf 77 pgf90 and pghpf only prof set profile options function level and line level Code profiling are supported Generation determines whether the compiler promotes Optimization REAL variables and constants to DOUBLE PRECISION pgf 77 pg 90 and pghpf only 64 Chapter 3 Description Category nor8intrinsics determines how the compiler treats the intrinsics Optimization CMPLX and REAL pgf
296. the input and output filename conventions for the phases of the compilation process 1 3 1 Input Files You can specify assembly language files preprocessed source files Fortran C C source files object files and libraries as inputs on the command line The compilation driver determines the type of each input file by examining the filename extensions The drivers use the following conventions 10 filename filename F filename f90 filename hpf filename c filename i filename C filename cc filename s filename o filename a filename so indicates a Fortran source file indicates a Fortran source file that can contain macros and preprocessor directives to be preprocessed indicates a Fortran 90 source file that is in freeform format indicates a HPF source file indicates a C source file that can contain macros and preprocessor directives to be preprocessed indicates a pre processed C or C source file indicates a C source file that can contain macros and preprocessor directives to be preprocessed indicates a C source file that can contain macros and preprocessor directives to be preprocessed indicates an assembly language file indicates an object file indicates a library of object files UNIX systems only indicates a library of shared object files Chapter 1 The driver passes files with o so and a extensions to the linker and s files to the assembler In
297. the member function push There would be no file containing both the definition of push and the definition of x Source File Method 1 Each nh file that declares a template entity also contains either the definition of the entity or includes another file containing the definition 2 Implicit inclusion when the compiler sees a template declaration in a h file and discovers a need to instantiate that entity it is given permission to look for an associated definition file having the same base name and a different suffix and it implicitly includes that file at the end of the compilation This method allows most programs written using the cfront convention to be compiled See Section 11 4 Implicit inclusion 3 The ad hoc approach the programmer makes sure that the files that define template entities also have the definitions of all the available types and adds code or pragmas in those files to request instantiation of the entities there Automatic instantiation Method The automatic instantiation method works as follows 1 The first time the source files of a program are compiled no template entities are instantiated However the generated object files contain information about things that could have been instantiated in each compilation C Template Instantiation 207 2 When the object files are linked together a special pre linker program called pgprelnk is run It examines the object files looking for references and definition
298. thread executes the entire region redundantly until it encounters a directive that specifies work distribution For work distribution see the Do PARALLEL DO or DOACROSS directives PROGRAM WHICH_PROCESSOR_AM I INTEGER A 0 1 INTEGER omp_get_thread_num A 0 1 A 1 1 SOMP PARALLEL A omp_get_thread_num omp_get_thread_num SOMP END PARALLEL PRINT A 0 A 0 A 1 A 1 END The variables specified in a PRIVATE list are private to each thread in a team In effect the compiler creates a separate copy of each of these variables for each thread in the team When an assignment to a private variable occurs each thread assigns to its local copy of the variable When operations involving a private variable occur each thread performs the operations using its local copy of the variable Important points about private variables are e Variables declared private in a parallel region are undefined upon entry to the parallel region If the first use of a private variable within the parallel region is in a right hand side expression the results of the expression will be undefined i e this is probably a coding error OpenMP Directives for Fortran 113 e Likewise variables declared private in a parallel region are undefined when serial execution resumes at the end of the parallel region The variables specified in a SHARED list are shared between all threads in a team meaning that all threads access the
299. tunit Mdlines Mnodlines Mdollar char Mextend Miomutex Mnoiomutex Monetrip Mnoonetrip Msave Mnosave Mstandard Munixlogical Mnounixlogical Command line Options the compiler treats as a synonym for standard input for reading and standard output for writing the compiler treats as a synonym for unit 5 on input and unit 6 on output the compiler treats lines containing D in column 1 as executable statements ignoring the D the compiler does not treat lines containing D in column 1 as executable statements does not ignore the D char specifies the character to which the compiler maps the dollar sign The compiler allows the dollar sign in names with Mextend the compiler accepts 132 column source code otherwise it accepts 72 column code the compiler generates critical section calls around Fortran I O statements the compiler does not generate critical section calls around Fortran I O statements the compiler forces each DO loop to execute at least once the compiler does not force each DO loop to execute at least once This option is useful for programs written for earlier versions of Fortran i the compiler assumes that all local variables are subject to the SAVE statement Note that this may allow older Fortran programs to run but it can greatly reduce performance the compiler does not assume that all local variables ar
300. type has extern C linkage void pf pf points to an extern C function amp f error unless implicit conv is allowed 262 Appendix C This extension is allowed in environments where C and C functions share the same calling conventions though it is pointless unless DEFAULT _C_AND CPP FUNTION TYPES ARE DISTINCT is TRUE When DEFAULT _IMPL CONV _ BETWEEN _ C AND CPP FUNCTION PTRS_ ALLOWED is set it is enabled by default it can also be enabled in cfront compatibility mode or with command line option implicit_extern_c_type_conversion It is disabled in strict ANSI mode C 5 cfront 2 1 Compatibility Mode The following extensions are accepted in cfront 2 1 compatibility mode in addition to the extensions listed in the following 2 1 3 0 section i e these are things that were corrected in the 3 0 release of cfront The dependent statement of an if while do while or for is not considered to define a scope The dependent statement may not be a declaration ny objects constructed within the dependent statement are destroyed at exit from the dependent statement Implicit conversion from integral types to enumeration types is allowed A non const member function may be called for a const object A warning is issued A const void value may be implicitly converted toa void value e g when passed as an argument When in determining the level of argument match for overloading a reference parameter is i
301. uments may not appear in EQUIVALENCE statements 0058 Equivalenced variables and not in same common block A common block variable must not be equivalenced with a variable in another common block 234 Appendix B 0059 Conflicting equivalence between and The indicated equivalence implies a storage layout inconsistent with other equivalences 0060 Illegal equivalence of structure variable STRUCTURE and UNION variables may not appear in EQUIVALENCE statements 0061 Equivalence of and extends common block backwards 0062 Equivalence forces to be unaligned EQUIVALENCE statements have defined an address for the variable that has an alignment not optimal for variables of its data type This can occur when INTEGER and CHARACTER data are equivalenced for instance 0063 Gap in common block before 0064 Illegal use of in DATA statement implied DO loop The indicated variable is referenced where it is not an active implied DO index variable 0065 Repeat factor less than or equal to zero 0066 Too few data constants in initialization statement 0067 Too many data constants in initialization statement 0068 Numeric initializer for CHARACTER out of range 0 through 255 A CHARACTER 1 variable or character array element can be initialized to an integer octal or hexadecimal constant if that constant is in the range 0 through 255 0069 Illegal implied DO expression The only operations allowed within an implied DO expres
302. ut output format specifiers and additional reference material System V Application Binary Interface Processor Supplement by AT amp T UNIX System Laboratories Inc Prentice Hall Inc System V Application Binary Interface X86 64 Architecture Processor Supplement http www x86 64 org abi pdf FORTRAN 90 HANDBOOK Complete ANSI ISO Reference McGraw Hill 1992 Programming in VAX Fortran Version 4 0 Digital Equipment Corporation September 1984 IBM VS Fortran IBM Corporation Rev GC26 4119 The C Programming Language by Kernighan and Ritchie Prentice Hall C A Reference Manual by Samuel P Harbison and Guy L Steele Jr Prentice Hall 1987 The Annotated C Reference Manual by Margaret Ellis and Bjarne Stroustrup AT amp T Bell Laboratories Inc Addison Wesley Publishing Co 1990 Preface Chapter 1 Getting Started This chapter describes how to use the PGI compilers The command used to invoke a compiler for example the pg 90 command is called a compilation driver The compilation driver controls the following phases of compilation preprocessing compiling assembling and linking Once a file is compiled and an executable file is produced you can execute debug or profile the program on your system Executables produced by the PGI compilers are unconstrained meaning they can be executed on any compatible X86 or AMD64 processor based system regardless of whether the PGI compilers are installed on that sy
303. ut problems as long as you use the extern C keyword to declare the C function in the C program This prevents name mangling for the C function name If you want to call a C function from C likewise you have to use the extern c keyword to declare the C function This keeps the C compiler from mangling the name of the function e Youcanuse the cplusplus macro to allow a program or header file to work for both C and C For example the following defines in the header file stdio h allow this file to work for both C and C ifndef STDIO_H define STDIO_H ifdef cplusplus extern C endif cplusplus Functions and data types defined ifdef cplusplus endif cplusplus fendif C member functions cannot be declared extern as their names will always be mangled Therefore C member functions cannot be called from C or Fortran 184 Note that if a C function contains objects with constructors and destructors calling such a function from either C or Fortran will not be possible unless the initialization in the main program is performed from a C program where constructors and destructors are properly initialized In general you can call a C function from C without problems as long as you use the extern C keyword to declare the C function in the C program This prevents name Chapter 10 mangling for the C function name If you want to call a C function from C likewise you hav
304. vectorizer determines an appropriate cutoff length and generates scalar code to be executed whenever the loop count is less than or equal to that length If altcode n is specified the scalar altcode is executed whenever the loop count is less than or equal ton If noaltcode is specified the vectorized version of the loop is always executed regardless of the loop count Instructs the vectorizer to enable certain associativity conversions that can change the results of a computation due to roundoff error A typical optimization is to change an arithmetic operation to an arithmetic operation that is mathematically correct but can be computationally different due to round off error Instructs the vectorizer to disable associativity conversions Instructs the vectorizer when performing cache tiling Chapter 3 optimizations to assume a cache size of n The default is n 262144 smallvect n Instructs the vectorizer to assume that the maximum vector length is less than or equal to n The vectorizer uses this information to eliminate generation of the stripmine loop for vectorized loops wherever possible Ifthe size n is omitted the default is 100 Note No space is allowed on either side of the colon sse Instructs the vectorizer to search for vectorizable loops and where possible make use of SSE SSE2 and prefetch instructions prefetch Instructs the vectorizer to search for vectorizable loops and where possible make us
305. vectorizer performs various operations that can be controlled by arguments to the Mvect command line option The following sections describe these arguments that affect the operation of the vectorizer In addition these vectorizer operations can be controlled from within code using Optimization amp Parallelization 25 directives and pragmas For details on the use of directives and pragmas refer to Chapter 7 Optimization Directives and Pragmas The vectorizer performs the following operations e Inner loop and outer loop distribution e Loop interchange e Memory hierarchy cache tiling optimizations e Generation of SSE SSE2 and prefetch instructions on processors where these are supported By default Mvect without any sub options is equivalent to Mvect assoc cachesize 262144 This enables the options for nested loop transformation and various other vectorizer options These defaults may vary depending on the target system The vectorizer performs high level loop transformations on countable loops A loop is countable if the number of iterations is set only before loop execution and cannot be modified during loop execution These transformations which include loop distribution loop splitting loop interchange and cache tiling allow the resulting loop to be optimized more completely and often result in more effective use of machine resources such as registers Assoc Option The option Mvect assoc instructs the v
306. were used Nonmember template functions will be instantiated even if the only reference was a declaration Similar to tused except that the functions are given internal linkage This is intended to provide a very simple mechanism for those getting started with templates The compiler will instantiate the functions that are used in each compilation unit as local functions and the program will link and run correctly barring problems due to multiple copies of local static variables However one may end up with many copies of the instantiated functions tlocal cannot be used in conjunction with automatic template instantiation 11 2 Pragma control of template instantiation Instantiation pragmas control the instantiation of specific template entities or sets of template entities There are three instantiation pragmas The instantiate pragma causes a specified entity to be instantiated The do_not_instantiate pragma suppresses the instantiation of a specified entity It is typically used to suppress the instantiation of an entity for which a specific definition will be supplied The can_instantiate pragma indicates that a specified entity can be instantiated in the current compilation but need not be it is used in conjunction with automatic instantiation to indicate potential sites for instantiation if the template entity turns out to be required The argument to the instantiation pragma may be A template class name A member
307. wing command line causes PGF90 to compile using the 4 7 release instead of the default o stcc V4 1 myprog c Cross reference Minfo v Use the v option to display the invocations of the compiler assembler and linker These invocations are command lines created by the compilation driver from the files and the W options you specify on the compiler command line Default The compiler does not display individual phase invocations Cross reference Minfo V Passes arguments to a specific phase Use the W option to specify options for the assembler compiler or linker Note A given PGI compiler command invokes the compiler driver which parses the command line and generates the appropriate commands for the compiler assembler and linker Syntax W 0 a 1 option option 94 Chapter 3 Where 0 the number zero specifies the compiler a specifies the assembler l lowercase letter 1 specifies the linker option is a string that is passed to and interpreted by the compiler assembler or linker Options separated by commas are passed as separate command line arguments Note You cannot have a space between the W and the single letter pass identifier between the identifier and the comma or between the comma and the option Usage In the following example the linker loads the text segment at address 0xf c00000 and the data segment at address 0xffe00000 pgf90 W1 k t 0xff c00000 d 0
308. xffe00000 myprog f Do not print warning messages 3 2 C and C specific Compiler Options The following options are specific to PGCC C and or C pgCC only Using this option the PGCC C compiler accepts code conforming to the proposed ANSI C standard It issues errors for non conforming code Default By default the compiler accepts code conforming to the standard C Annotated Reference Manual Usage The following command line requests ANSI conforming C Command line Options 95 pgCC A hello cc Cross references b and p no_l lalternative_tokens egCC only Enable or disable recognition of alternative tokens These are tokens that make it possible to write C without the use of the amp and characters The alternative tokens include the operator keywords e g and bitand etc and digraphs The default behavior is no_alternative_tokens l pgCC only Enable compilation of C with cfront 2 1 compatibility This causes the compiler to accept language constructs that while not part of the C language definition are accepted by the AT amp T C Language System cfront release 2 1 This option also enables acceptance of anachronisms Default The compiler does not accept cfront language constructs that are not part of the C language definition Usage In the following example the compiler accepts cfront constructs pgCC b myprog cc Cross references cfront2
309. you cannot directly call a COMPLEX function from C C Refer to Section 0 6 3 Complex Return Values for details on how to call a COMPLEX function indirectly 186 Chapter 10 Table 10 1 Fortran and C C Data Type Compatibility Type lower case character x character n x real x real 4 real 8 double intege C C Type char x n Fortran C X bytes 4 incegerse x icra toss Je Diosin Odes e Diosin eara h Diosin orea e Drosina aime iosicare hong toss e Table 10 2 Fortran and C C Representation of the COMPLEX Type Fortran Type lower case C C Type Size bytes complex x struct float 1 13 x ree eae double complex x struct 16 double dr di x Inter language Calling struct i Loak Tikai 187 10 5 1 Fortran Named Common Blocks A named Fortran common block can be represented in C C by a structure whose members correspond to the members of the common block The name of the structure in C C must have the added underscore For example the Fortran common block INTEGER I COMPLEX C DOUBLE COMPLEX CD DOUBLE PRECISION D COMMON COM i c cd d is represented in C with the following equivalent extern struct int iz struct float real imag c struct double real imag cd double d com_ and in C with the following equivalent extern C struct inte aiz struct float real imag c struct double real ima

PGI 5 User's Guide - The Portland Group

Contents

Download Pdf Manuals

Related Search

Related Contents