PGI Server Complete for Linux, Release 7.2. Optimization, Compiler, and Other flags for use by SPEC CPU2006

pgi72_linux_flags PGI Server Complete for Linux, Release 7.2. Optimization, Compiler, and Other flags for use by SPEC CPU2006

Compilers: PGI Server Complete 7.2

Operating systems: Linux

]]>

Linux Huge Page settings

In order to take full advantage of using PGI's huge page runtime library, your system must be configured to use huge pages. It is safe to run binaries compiled with "-Msmartalloc=huge" on systems not configured to use huge pages, however, you will not benefit from the performance improvements huge pages offer. To configure your system for huge pages perform the following steps:

Create a mount point for the huge pages: "mkdir /mnt/hugepages"
The huge page file system needs to be mounted when the systems reboots. Add the following to a system boot configuration file before any services are started: "mount -t hugetlbfs nodev /mnt/hugepages"
Set vm/nr_hugepages=N in your /etc/sysctl.conf file where N is the maximum number of pages the system may allocate.
Reboot to have the changes take effect.

Note that further information about huge pages may be found in your Linux documentation file: /usr/src/linux/Documentation/vm/hugetlbpage.txt

PGI_HUGE_PAGES

The maximum number of huge pages an application is allowed to use can be set at run time via the environment variable PGI_HUGE_PAGES. If not set, then the process may use all available huge pages when compiled with "-Msmartalloc=huge" or a maximum of n pages where the value of n is set via the compile time flag "-Msmartalloc=huge:n.

Using numactl to bind processes and memory to cores

For multi-copy runs or single copy runs on systems with multiple sockets, it is advantageous to bind a process to a particular core. Otherwise, the OS may arbitrarily move your process from one core to another. This can effect performance. To help, SPEC allows the use of a "submit" command where users can specify a utility to use to bind processes. We have found the utility 'numactl' to be the best choice.

numactl runs processes with a specific NUMA scheduling or memory placement policy. The policy is set for a command and inherited by all of its children. The numactl flag "--physcpubind" specifies which core(s) to bind the process. "-l" instructs numactl to keep a process memory on the local node while "-m" specifies which node(s) to place a process memory. For full details on using numactl, please refer to your Linux documentation, 'man numactl'

Note that some versions of numactl, particularly the version found on SLES 10, we have found that the utility incorrectly interprets application arguments as it's own. For example, with the command "numactl --physcpubind=0 -l a.out -m a", numactl will interpret a.out's "-m" option as it's own "-m" option. To work around this problem, a user can put the command to be run in a shell script and then run the shell script using numactl. For example: "echo 'a.out -m a' > run.sh ; numactl --physcpubind=0 bash run.sh"

ulimit -s <n>

Sets the stack size to n kbytes, or unlimited to allow the stack size to grow without limit.

ulimit -l <n>

Sets the maximum size of memory that may be locked into physical memory.

NCPUS

Sets the maximum number of OpenMP parallel threads auto-parallelized (-Mconcur) applications may use.

]]> The PGI C compiler for Windows.

]]> pgcc.exe The PGI C++ compiler for Windows.

]]> pgcpp.exe The PGI Fortran 95 compiler for Windows.

]]> pgf95.exe The PGI C compiler for Linux.

]]> pgcc The PGI C++ compiler for Linux.

]]> pgcpp The PGI Fortran 95 compiler for Linux.

]]> pgf95 Disable warning messages.

]]> -w Don't include Fortran main program object module.

]]> -Mnomain Use C99 language features.

]]> -c9x Chooses generally optimal flags for the target platform. As of the PGI 7.0 release, the flags "-fast" and "-fastsse" are equivlent for 64-bit compilations. For 32-bit compilations "-fast" does not include "-Mscalarsse", "-Mcache_align", or "-Mvect=sse".

]]> -fast Disable C++ exception handling support.

]]> --no_exceptions Disable C++ run time type information support.

]]> --no_rtti Generate zero-overhead C++ exception handlers.

]]> Inline functions declared with the inline keyword.

]]> -Mautoinline Disable inlining of functions declared with the inline keyword.

]]> -Mnoautoinline Align "unconstrained" data objects of size greater than or equal to 16 bytes on cache-line boundaries. An "unconstrained" object is a variable or array that is not a member of an aggregate structure or common block, is not allocatable, and is not an automatic array. On by default on 64-bit Linux systems.

]]> -Mcache_align Align doubles on double alignment boundaries

]]> -Mdalign Do not align doubles on double alignment boundaries

]]> -Mnodalign Set SSE to flush-to-zero mode; if a floating-point underflow occurs, the value is set to zero.

]]> -Mflushz Treat denormalized numbers as zero. Included with "-fast" on Intel based systems. For AMD based systems, "-Mdaz" is not included by default with "-fast".

]]> -Mdaz Generate code to set up a stack frame.

]]> -Mframe 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.

]]> -Mnoframe CPU2006 flags file rule used to split an optimization flag containing sub-options into multiple flag descriptions. Please refer to the flag file rule of the various sub-options for the actual flag description. Instructs the compiler to use relaxed precision in the calculation of floating-point reciprocal square root (1/sqrt). Can result in improved performance at the expense of numerical accuracy.

]]> -Mfprelaxed=rsqrt Instructs the compiler to use relaxed precision in the calculation of floating-point square root. Can result in improved performance at the expense of numerical accuracy.

]]> -Mfprelaxed=sqrt Instructs the compiler to use relaxed precision in the calculation of floating-point division. Can result in improved performance at the expense of numerical accuracy.

]]> -Mfprelaxed=div Instructs the compiler to allow floating-point expression reordering, including factoring. Can result in improved performance at the expense of numerical accuracy.

]]> -Mfprelaxed=order Instructs the compiler to use relaxed precision in the calculation of some intrinsic functions. Can result in improved performance at the expense of numerical accuracy. The default on an AMD system is "-Mfprelaxed=sqrt,rsqrt,order". The default on an Intel system is "-Mfprelaxed=rsqrt,sqrt,div,order"

]]> -Mfprelaxed CPU2006 flags file rule used to split an optimization flag containing sub-options into multiple flag descriptions. Please refer to the flag file rule of the various sub-options for the actual flag description. Instructs the compiler to use low-precision approximation in the calculation of reciprocal square root (1/sqrt). Can result in improved performance at the expense of numerical accuracy.

]]> -Mfpapprox=rsqrt Instructs the compiler to use low-precision approximation in the calculation of square root. Can result in improved performance at the expense of numerical accuracy.

]]> -Mfpapprox=sqrt Instructs the compiler to use low-precision approximation in the calculation of divides. Can result in improved performance at the expense of numerical accuracy.

]]> -Mfpapprox=div Instructs the compiler to perform low-precision approximation in the calculation of floating-point division, square-root, and reciprocal square root. Can result in improved performance at the expense of numerical accuracy.

]]> -Mfpapprox -Mpre -Mpre CPU2006 flags file rule used to split an optimization flag containing sub-options into multiple flag descriptions. Please refer to the flag file rule of the various sub-options for the actual flag description. Set the fetch-ahead distance for prefetch instructions to $1 cache lines

]]> -Mprefetch=d:m Set maximum number of prefetch instructions to generate for a given loop to $1.

]]> -Mprefetch=n:p Use the prefetchnta instruction.

]]> -Mprefetch_nta Use the prefetch instruction.

]]> -Mprefetch=plain Use the prefetcht0 instruction.

]]> -Mprefetch=t0 Use the AMD-specific prefetchw instruction.

]]> -Mprefetch=w Enable generation of prefetch instructions on processors where they are supported.

]]> -Mprefetch Disable generation of prefetch instructions.

]]> -Mnoprefetch Use SSE/SSE2 instructions to perform scalar floating-point arithmetic on targets where these instructions are supported.

]]> -Mscalarsse Do not use SSE/SSE2 instructions to perform scalar floating-point arithmetic; use x87 operations instead.

]]> -Mnoscalarsse Instructs the compiler to extend the sign bit that is set as a result of an object's conversion from one data type to an object of a larger signed data type.

]]> -Msignextend Aligns or does not align innermost loops on 32 byte boundaries with -tp barcelona. Small loops on barcelona systems may run fast if aligned on 32-byte boundaries; however, in practice, most assemblers do not yet implement efficient padding causing some programs to run more slowly with this as default. Use -Mloop32 on systems with an assembler tuned for barcleona. The default is -Mnoloop32.

]]> -Mloop32 Treat individual array element references as candidates for possible loop-carried redundancy elimination. The default is to eliminate only redundant expressions involving two or more operands.

]]> -Mlre_array Allow expression re-association; specifying this sub-option can increase opportunities for loop-carried redundancy elimination.

]]> -Mlre=assoc Disable expression re-association.

]]> -Mlre=noassoc Enables loop-carried redundancy elimination, an optimization that can reduce the number of arithmetic operations and memory references in loops.

]]> -Mlre Disable loop-carried redundancy elimination.

]]> -Mnolre Instructs the compiler not to perform idiom recognition or introduce calls to hand-optimized vector functions.

]]> -Mnovintr Generate profile-feedback instrumentation (PFI); this includes extra code to collect run-time statistics and dump them to a trace file for use in a subsequent compilation. PFI gathers information about a program's execution and data values but does not gather information from hardware performance counters. PFI does gather data for optimizations which are unique to profile-feedback optimization.

The indirect sub-option enables collection of indirect function call targets, which can be used for indirect function call inlining.

]]> -Mpfi=indirect Enable profile-feedback optimizations including indirect function call inlining. This option requires a pgfi.out file generated from a binary built with -Mpfi=indirect.

]]> -Mpfo=indirect Generate profile-feedback instrumentation (PFI); this includes extra code to collect run-time statistics and dump them to a trace file for use in a subsequent compilation. PFI gathers information about a program's execution and data values but does not gather information from hardware performance counters. PFI does gather data for optimizations which are unique to profile-feedback optimization.

]]> -Mpfi Enable profile-feedback optimizations.

]]> -Mpfo CPU2006 flags file rule used to split an optimization flag containing sub-options into multiple flag descriptions. Please refer to the flag file rule of the various sub-options for the actual flag description. Interprocedural Analysis option: Recognize when targets of pointer dummy are aligned.

]]> -Mipa=align Interprocedural Analysis option: Disable recognizition when targets of pointer dummy are aligned.

]]> -Mipa=noalign Interprocedural Analysis option: Remove arguments replaced by -Mipa=ptr,const

]]> -Mipa=arg Interprocedural Analysis option: Do not remove arguments replaced by -Mipa=ptr,const

]]> -Mipa=noarg Interprocedural Analysis option: Generate call graph information for pgicg tool.

]]> -Mipa=cg Interprocedural Analysis option: Do not generate call graph information for pgicg tool.

]]> -Mipa=nocg Interprocedural Analysis option: Enable interprocedural constant propagation.

]]> -Mipa=const Interprocedural Analysis option: Disable interprocedural constant propagation.

]]> -Mipa=noconst Interprocedural Analysis option: Used with -Mipa=inline to specify functions which should not be inlined.

]]> -Mipa=except:func Instructs the compiler to perform interprocedural analysis. Equivalant to -Mipa=align,arg,const,f90ptr,shape,globals,libc,localarg,ptr,pure.

]]> -Mipa=fast Interprocedural Analysis option: Force all objects to recompile regardless whether IPA information has changed.

]]> -Mipa=force Interprocedural Analysis option: Optimize references to global values.

]]> -Mipa=globals Interprocedural Analysis option: Do not optimize references to global values.

]]> -Mipa=noglobals Interprocedural Analysis option: Automatically determine which functions to inline, limit to $1 levels where $1 is a supplied constant value. If no value is suppiled, then the default value of 2 is used. IPA-based function inlining is performed from leaf routines upward.

]]> n -Mipa=inline:4 Interprocedural Analysis option: Automatically determine which functions to inline, limit to 2 levels (default). IPA-based function inlining is performed from leaf routines upward.

]]> -Mipa=inline Interprocedural Analysis option: Automatically determine which functions to inline, independent of information gathered from profile guided feedback (-Mpfi), limit to $1 levels where $1 is a supplied constant value. If no value is suppiled, then the default value of 2 is used. IPA-based function inlining is performed from leaf routines upward.

]]> n -Mipa=inlinenopfo:4 Interprocedural Analysis option: Automatically determine which functions to inline, independent of information gathered from profile guided feedback (-Mpfi), limit to 2 levels (default). IPA-based function inlining is performed from leaf routines upward.

]]> -Mipa=inlinenopfo Interprocedural Analysis option: Inline static functions which are outside the scope of the current file.

]]> -Mipa=staticfunc Interprocedural Analysis option: Allow inlining of routines from libraries.

]]> -Mipa=libinline Interprocedural Analysis option: Do not inline routines from libraries.

]]> -Mipa=nolibinline Interprocedural Analysis option: Specifies the number of concurent IPA second pass compliation proccess that may be performed. This option speeds-up the compilation time on multi-core systems but does not perform any optimizations.

]]> -Mipa=libc Interprocedural Analysis option: Used to optimize calls to certain functions in the system standard C library, libc.

]]> -Mipa=libc Interprocedural Analysis option: Allow recompiling and optimization of routines from libraries using IPA information.

]]> -Mipa=libopt Interprocedural Analysis option: Don't optimize routines in libraries.

]]> -Mipa=nolibopt Interprocedural Analysis option: -Mipa=arg plus externalizes local pointer targets.

]]> -Mipa=localarg Interprocedural Analysis option: -Mipa=arg plus externalizes local pointer targets.

]]> -Mipa=localarg Interprocedural Analysis option: Do not externalize local pointer targets.

]]> -Mipa=nolocalarg Interprocedural Analysis option: Enable pointer disambiguation across procedure calls.

]]> -Mipa=ptr Interprocedural Analysis option: Disable pointer disambiguation.

]]> -Mipa=noptr Interprocedural Analysis option: Fortran 90/95 Pointer disambiguation across calls.

]]> -Mipa=f90ptr Interprocedural Analysis option: Disable Fortran 90/95 pointer disambiguation

]]> -Mipa=nof90ptr Interprocedural Analysis option: Pure function detection.

]]> -Mipa=pure Interprocedural Analysis option: Disable pure function detection.

]]> -Mipa=nopure Interprocedural Analysis option: Allows inlining in Fortran even when array shapes do not match.

]]> -Mipa=reshape Interprocedural Analysis option: Perform Fortran 90 array shape propagation.

]]> -Mipa=shape Interprocedural Analysis option: Disable Fortran 90 array shape propagation.

]]> -Mipa=noshape Interprocedural Analysis option: Remove functions that are never called.

]]> -Mipa=vestigial Interprocedural Analysis option: Do not remove functions that are never called.

]]> -Mipa=novestigial Enable Interprocedural Analysis.

]]> -Mipa CPU2006 flags file rule used to split an optimization flag containing sub-options into multiple flag descriptions. Please refer to the flag file rule of the various sub-options for the actual flag description. Instructs the parallelizer to generate alternate serial code for parallelized loops. Without arguments, the parallelizer determines an appropriate cutoff length and generates serial code to be executed whenever the loop count is less than or equal to that length.

]]> -Mconcur=altcode Instructs the parallelizer to generate alternate serial code for parallelized loops. With arguments, the serial altcode is executed whenever the loop count is less than or equal to $1.

]]> -Mconcur=altcode:n Always execute the parallelized version of a loop regardless of the loop count.

]]> -Mconcur=noaltcode Disables parallelization of loops with reductions.

]]> -Mconcur=noassoc Assume loops containing calls are safe to parallelize and allows loops containing calls to be 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.

]]> -Mconcur=cncall Do not assume loops containing calls are safe to parallelize.

]]> -Mconcur=nocncall Parallelize with block distribution. Contiguous blocks of iterations of a parallelizable loop are assigned to the available processors.

]]> -Mconcur=dist:bloc Parallelize with cyclic distribution. The outermost parallelizable loop in any loop nest is parallelized. If a parallelized loop is innermost, its iterations are allocated to processors cyclically. For example, if there are 3 processors executing a loop, processor 0 performs iterations 0, 3, 6, etc.; processor 1 performs iterations 1, 4, 7, etc.; and processor 2 performs iterations 2, 5, 8, etc.

]]> -Mconcur=dist:cyclic Enable parallelization of innermost loops.

]]> -Mconcur=innermost Disable parallelization of innermost loops.

]]> -Mconcur=noinnermost 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.

]]> -Mconcur CPU2006 flags file rule used to split an optimization flag containing sub-options into multiple flag descriptions. Please refer to the flag file rule of the various sub-options for the actual flag description. Instructs the inliner to inline the functions within the library filename.ext.

]]> -Minline=lib:filename.ext Instructs the inliner to inline all eligible functions except $1, a function in the source text. Multiple functions can be listed, comma-separated.

]]> -Minline=except:func Instructs the inliner to inline function func.

]]> -Minline=name:func Allows inlining in Fortran even when array shapes do not match.

]]> -Minline=reshape Instructs the inliner to inline functions with $1 or fewer statements where $1 is a supplied constant value.

]]> n -Minline=size:n Instructs the inliner to perform $1 levels of inlining where $1 is a supplied constant value. If no value is suppiled, then the default value of 2 is used.

]]> n -Minline=levels:4 Instructs the inliner to perform 1 level of inlining.

]]> -Minline Disable constant propagation from assertions derived from equality conditionals.

]]> -Mnopropcond CPU2006 flags file rule used to split an optimization flag containing sub-options into multiple flag descriptions. Please refer to the flag file rule of the various sub-options for the actual flag description. Link with the huge page runtime library and allocate a maximum of $1 huge pages where $1 is a supplied constant value. If no constant value is supplied, then the maximum number of huge pages the application can use is limited by the number of huge pages the operating system has available or the value of the environment variable PGI_HUGE_PAGES. Note that setting PGI_HUGE_PAGES will override the value of $1.

]]> n -Msmartalloc=huge:150 Link with the huge page runtime library. The maximum number of huge pages the application can use is limited by the number of huge pages the operating system has available or the value of the environment variable PGI_HUGE_PAGES.

]]> -Msmartalloc=huge Link with the huge page runtime library. Use huge pages for an executable's .BSS section.

]]> -Msmartalloc\=hugebss Adds a call to the routine "mallopt" in the main routine. This option can have a dramatic impact on the performance of programs that dynamically allocate memory, especially for those which have a few large mallocs. To be effective, this switch must be specified when compiling the file containing the Fortran, C, or C++ main routine.

]]> -Msmartalloc Link with PGI's Alloc library which replaces the system's Malloc, Calloc, Realloc, and Free functions with PGI versions. Programs using -Msmartalloc must be compiled and linked with "-Bdynamic".

]]> -Msmartalloc CPU2006 flags file rule used to split an optimization flag containing sub-options into multiple flag descriptions. Please refer to the flag file rule of the various sub-options for the actual flag description. Assume all pointers and arrays are independent and safe for aggressive optimizations, and in particular that no pointers or arrays overlap of conflict with each other.

]]> -Msafeptr=all Instructs the compiler that arrays and pointers are treated with the same copyin and copyout semantics as Fortran dummy arguments.

]]> -Msafeptr=arg Instructs the compiler that local pointers and arrays do not overlap or conflict with each other and are independent.

]]> -Msafeptr=auto Instructs the compiler that local pointers and arrays do not overlap or conflict with each other and are independent.

]]> -Msafeptr=local Instructs the compiler that static pointers and arrays do not overlap or conflict with each other and are independent.

]]> -Msafeptr=static Instructs the compiler that global or external pointers and arrays do not overlap or conflict with each other and are independent.

]]> -Msafeptr=global Instructs the C/C++ compiler to override data dependencies between pointers of a given storage class.

]]> -Msafeptr CPU2006 flags file rule used to split an optimization flag containing sub-options into multiple flag descriptions. Please refer to the flag file rule of the various sub-options for the actual flag description. Instructs the compiler to completely unroll loops with a constant loop count of less than or equal to $1 where $1 is a supplied constant value. If no constant value is given, then a default of 4 is used. A value of 1 inhibits the complete unrolling of loops with constant loop counts.

]]> n -Munroll=c:1 "-Munroll=n:n" instructs the compiler to unroll loops $1 times where $1 is a supplied constant value. If no constant value is given, then a default of 4 is used.

]]> n -Munroll=n:n "-Munroll=m:n" instructs the compiler to unroll loops with multiple blocks $1 times where $1 is a supplied constant value. If no constant value is given, then a default of 4 is used.

]]> n -Munroll=m:n Instructs the compiler to unroll loops with multiple blocks using the default value of 4 times

]]> -Munroll=m Invokes the loop unroller.

]]> -Munroll Disable loop unrolling.

]]> -Mnounroll Don't check dependence relations for vector or parallel code.

]]> Allow parallelization of loops with conditional scalar assignments.

]]> Generate code to check for zero loop increments.

]]> Enable an optional post-pass instruction scheduling.

]]> -Msmart Disable an optional post-pass instruction scheduling.

]]> -Mnosmart CPU2006 flags file rule used to split an optimization flag containing sub-options into multiple flag descriptions. Please refer to the flag file rule of the various sub-options for the actual flag description. Disable automatic vector pipelining.

]]> -Mnovect Instructs the vectorizer to generate alternate code for vectorized loops when appropriate. For each vectorized loop the compiler decides whether to generate altcode and what type or types to generate, which may be any or all of:

Altcode without iteration peeling
Altcode with non-temporal stores and other data cache optimizations
Altcode base on array alignments calculated dynamically at runtime.

The compiler also determines suitable loop count and array alignment conditions for executing the altcode.

]]> -Mvect=altcode Disables alternate code generation for vectorized loops.

]]> -Mvect=noaltcode Instructs the vectorizer to enable certain associativity conversions that can change the results of a computations due to roundoff error. A typical optimization is to change an arithmetic operation to an arithmetic opteration that is mathmatically correct, but can be computationally different, due to round-off error.

]]> -Mvect=assoc Instructs the vectorizer to disable associativity conversions.

]]> -Mvect=noassoc Instructs the vectorizer, when performing cache tiling optimizations, to assume a cache size of $1. The default size is processor dependent.

]]> n -Mvect=cachesize:n Instructs the vectorizer to enable loop fusion.

]]> -Mvect=fuse Instructs the vectorizer to disable vectorization of indirect array references.

]]> -Mvect=nogather Instructs the vectorizer to enable idiom recognition.

]]> -Mvect=idiom Instructs the vectorizer to disable idiom recognition.

]]> -Mvect=noidiom Generate vector loops for all loops where possible regardless of the number of statements in the loop. This overrides a heuristic in the vectorizer that ordinarily prevents vectorization of loops with a number of statements that exceed a certain threshold.

]]> -Mvect=nosizelimit Instructs the vectorizer to generate partial vectorization.

]]> -Mvect=partial Instructs the vectorizer to generate prefetch instructions.

]]> -Mvect=prefetch Instructs the vectorizer to search for vectorizable loops and, where possible, make use of SSE, SSE2, and prefetch instructions.

]]> -Mvect=sse Instructs the driver to disable the -Mvect=sse option which is part of the "-fast" option.

]]> -Mvect=nosse Enable automatic vector pipelining.

]]> -Mvect Disables -Ktrap=fp.

]]> -Mnofptrap -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 on (floating-point exceptions are 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.

]]> -Ktrap=fp Enable long branches.

]]> -Mlongbranch Link with the AMD Core Math Library. Available from www.amd.com

]]> -lacml 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. When used strictly as a linker flag, the PGI OpenMP runtime will be linked and users can use the environment variables MP_BIND and MP_BLIST to bind a serial program to a CPU.

]]> -mp The align sub-option to -mp forces loop iterations to be allocated to OpenMP processes using an algorithm that maximizes alignment of vector sub-sections in loops that are both parallelized and vectorized for SSE. This can improve performance in program units that include many such loops. It can result in load-balancing problems that significantly decrease performance in program units with relatively short loops that contain a large amount of work in each iteration.

]]> -mp=align The numa suboption to -mp uses libnuma on systems where it is available.

]]> -mp=numa The nonuma suboption to -mp tells the driver to not link with libnuma.

]]> -mp=nonuma (For use only on 64-bit Linux targets) Generate code for the medium memory model in the linux86-64 execution environment. The default small memory model of the linux86-64 environment limits the combined area for a user's object or executable to 1GB, with the Linux kernel managing usage of the second 1GB of address for system routines, shared libraries, stacks, etc. Programs are started at a fixed address, and the program can use a single instruction to make most memory references. The medium memory model allows for larger than 2GB data areas, or .bss sections. Program units compiled using either -mcmodel=medium or -fpic require additional instructions to reference memory. The effect on performance is a function of the data-use of the application. The -mcmodel=medium switch must be used at both compile time and link time to create 64-bit executables. Program units compiled for the default small memory model can be linked into medium memory model executables as long as they are compiled -fpic, or position-independent.

]]> -mcmodel=medium Enable support for 64-bit indexing and single static data objects larger than 2GB in size. This option is default in the presence of -mcmodel=medium. Can be used separately together with the default small memory model for certain 64-bit applications that manage their own memory space.

]]> -Mlarge_arrays Enable dead store elimination.

]]> Enable optimizations using ANSI C type-based pointer disambiguation.

]]> Set the optimization level to -O2

]]> -O A basic block is generated for each C statement. No scheduling is done between statements. No global optimizations are performed.

]]> -O0 Level-one optimization specifies local optimization (-O1). 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:

Algebraic identity removal
Constant folding
Common subexpression elimination
Local register optimization
Peephole optimizations
Redundant load and store elimination
Strength reductions

]]> -O1 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 release. 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:

Branch to branch elimination
Constant propagation
Copy propagation
Dead store elimination
Global register allocation
Invariant code motion
Induction variable elimination

]]> -O2 All level 1 and 2 optimizations are performed. In addition, this level enables more aggressive code hoisting and scalar replacement optimizations that may or may not be profitable.

]]> -O3 Performs all level 1, 2, and 3 optimizations and enables hoisting of guarded invariant floating point expressions.

]]> -O4 Create a Unified Binary using multiple targets. Specify the type of the target processor as AMD64 Processor 32-bit mode.

]]> -tp k8-32 Specify the type of the target processor as AMD64 Processor 64-bit mode.

]]> -tp k8-64 Specify the type of the target processor as AMD64 Barcelona Processor 64-bit mode.

]]> -tp barcelona-64 Specify the type of the target processor as AMD64 Barcelona Processor 32-bit mode.

]]> -tp barcelona-32 Specify the type of the target processor as AMD64 Barcelona Processor 32-bit mode.

]]> -tp barcelona Specify the type of the target processor as Intel Penryn Processor 64-bit mode.

]]> -tp penryn-64 Specify the type of the target processor as Intel Penryn Processor 32-bit mode.

]]> -tp penryn-32 Specify the type of the target processor as Intel Penryn Processor 32-bit mode.

]]> -tp penryn Specify the type of the target processor as Intel P7 Architecture with EM64t, 64-bit mode.

]]> -tp p7-64 Specify the type of the target processor as Intel P7 Architecture (Pentium 4, Xeon, Centrino).

]]> -tp p7 Specify the type of the target processor as Intel Core 2 EM64T or compatible architecture using 64-bit mode.

]]> -tp core2-64 Specify the type of the target processor as Intel Core 2 or compatible architecture using 32-bit mode.

]]> -tp core2 Use the unified AMD/Intel 64-bit mode.

]]> -tp x64 Experimental flags.

]]> Link with static libraries.

]]> Staticily link with the PGI runtime libraries. System libraries may still be dynamically linked.

]]> Link with dynamic libraries.

]]> Link with Dynamic Link Libraries (DLL). Note that -Bdynamic must also be used during compilation as well as linking. Implies -D_DLL.

]]> -Bdynamic Disable runtime stack checking and set the stack's reservere size to $1 bytes and commit size to $2 bytes at link time.

]]> Pass the flag "-force:multiple" to the linker to create an output file whether or not the linker finds more than one definition for a symbol. This flag is required when linking with either PGI's Alloc library (-Msmartalloc) or Microquill's Smartheap library. Both libraries replace the system's Malloc, Calloc, Realloc, and Free functions.

]]> -Wl,-force:multiple The NCPUS environment variable can be used to set the number of processes or threads used in parallel regions. The default is to use only one process or thread (serial mode). If both OMP_NUM_THREADS and NCPUS are set, the value of OMP_NUM_THREADS takes precedence. Warning: setting NCPUS to a value larger than the number of physical processors or cores in your system can cause parallel programs to run very slowly.

]]> The MP_BIND environment variable can be set to yes or y to bind processes or threads executing in a parallel region to physical processors, or to no or n to disable such binding. The default is to not bind processes to processors. This is an execution time environment variable interpreted by the PGI runtime support libraries. It does not affect the behavior of the PGI compilers in any way. Note: the MP_BIND environment variable is not supported on all platforms.

]]> In addition to the MP_BIND variable, it is possible to define the thread-CPU relationship. For example, setting MP_BLIST=3,2,1,0 maps CPUs 3, 2, 1 and 0 to threads 0, 1, 2 and 3 respectively.

]]> The maximum number of huge pages an application is allowed to use can be set at run time via the environment variable PGI_HUGE_PAGES. If not set, then the process may use all available huge pages when compiled with "-Msmartalloc=huge" or a maximum of n pages where the value of n is set via the compile time flag "-Msmartalloc=huge:n."

]]> Specifies a directory to search for libraries. Use -L to add directories to the search path for library files. Multiple -L options are valid. However, the position of multiple -L options is important relative to -l options supplied.

]]> -L/path/to/libs Link using MicroQuill's SmartHeap 8 (32-bit) library for Linux. Description from Microquill:

SmartHeap is a fast (3X-100X faster than compiler-supplied libraries), portable (Windows, Linux, Solaris, HP-UX, IBM-AIX, Dec OSF Tru64, SGI Irix), reliable, ANSI-compliant malloc/operator new library. SmartHeap supports multiple memory pools, includes a fixed-size allocator, and is thread-safe. SmartHeap also includes comprehensive memory debugging APIs to detect leakage, overwrites, double-frees, wild pointers, out of memory, references to previously freed memory, and other memory errors.

]]> Link using MicroQuill's SmartHeap 8 (32-bit) library for Windows. Description from Microquill:

]]>