<<

KMP_AFFINITY
   OpenMP* Environment Variables
   Thread Affinity Interface (Linux* and Windows*)
    modifier
    offset
    permute
    type
KMP_ALL_THREADS
KMP_BLOCKTIME
KMP_LIBRARY
   OpenMP* Environment Variables
   OpenMP* Support Libraries
KMP_MONITOR_STACKSIZE
KMP_STACKSIZE
KMP_VERSION



LASTPRIVATE
    summary of data scope attribute clauses
    using
libraries
   Improving I/O Performance
   Parallel Region Directives
   Timing your Application
    libintrins.lib
    OpenMP* run-time routines
       Intel Extension Routines/Functions
       OpenMP* Run-time Library Routines
library functions
library routines
    Intel extension
    OpenMP* run-time routines
    to control number of threads
linking options
linking tools
    xild
       Creating a Library from IPO Objects
       Creating a Multi-file IPO Executable
       Interprocedural Optimization (IPO) Overview
       IPO-Related Performance Issues
    xilibtool
    xilink
       Creating a Multi-file IPO Executable
       Interprocedural Optimization (IPO) Overview
       IPO-Related Performance Issues
linking with IPO
little-endian-to-big-endian conversion
locating samples
lock routines
LOOP COUNT
    and loop distribution
loop interchange
loop unrolling
    limitations of
    support for
    using the HLO optimizer
       HLO Overview
       Optimizer Report Generation
loops
   Loop Constructs
   Loop Count and Loop Distribution
    anti dependency
    collapsing
    constructs
    count
       Loop Count and Loop Distribution
       Loop Unrolling Support
    dependencies
    distribution
       HLO Overview
       Loop Count and Loop Distribution
    flow dependency
    independence
    interchange
       Applying Optimization Strategies
       Coding Guidelines for Intel(R) Architectures
       HLO Overview
       Understanding Runtime Performance
    manual transformation
    output dependency
    parallelization
       Loop Parallelization and Vectorization
       Parallelism Overview
       Programming with Auto-parallelization
    reductions
    transformations
       Efficient Compilation
       HLO Overview
       Understanding Runtime Performance
    unrolling
       Loop Unrolling
       Loop Unrolling Support
    using for arrays
    vectorization



maintainability
manual transformations
master thread
    copying data in
memory
    access
    allocation
    dependency
    layout
memory aliasing
memory file system
misaligned data
mixing vectorizable types in a loop
MMX(TM)
mock object files
multidimensional arrays
    using effectively
multithreaded programs
   Auto-parallelization Overview
   Coding Guidelines for Intel Architectures
   Parallelism Overview
multithreading
   OpenMP* Support Libraries
   Programming with Auto-parallelization



natural alignment
naturally aligned
    data
       Alignment Options
       Efficient Compilation
    records
    storage
non-unit memory access
NOPREFETCH
    using
NOSWP
    using
NOVECTOR directive



obj files
OMP directives
   Examples of OpenMP* Usage
   Parallelism Overview
   Programming with OpenMP*
OMP_DYNAMIC
OMP_NESTED
OMP_NUM_THREADS
OMP_SCHEDULE
OpenMP*
    advanced issues
    cluster
       Cluster OpenMP* Support (Linux*)
       Thread Affinity Interface (Linux* and Windows*)
    debugging
    directives
       Combined Parallel and Worksharing Constructs
       Parallel Region Directives
       Synchronization Constructs
       THREADPRIVATE Directive
       Worksharing Construct Directives
    environment variables
       OpenMP* Environment Variables
       Thread Affinity Interface (Linux* and Windows*)
    KMP_AFFINITY
       OpenMP* Environment Variables
       Thread Affinity Interface (Linux* and Windows*)
    KMP_ALL_THREADS
    KMP_BLOCKTIME
    KMP_LIBRARY
    KMP_MONITOR_STACKSIZE
    KMP_STACKSIZE
    KMP_VERSION
    omp.h
    OMP_DYNAMIC
    OMP_NESTED
    OMP_NUM_THREADS
    OMP_SCHEDULE
    parallel processing thread model
    performance
    reports
    run-time library routines
    support libraries
OpenMP* Fortran directives
    clauses for
    examples of
    features of
    for synchronization
    for worksharing
    Intel extensions for
    programming using
optimal records to improve performance
optimization
    analyzing applications
    application-specific
    hardware-related
    library-related
    methodology
    options
       Enabling Automatic Optimizations
       Optimizing Different Application Types Overview
        restricting
        setting
    OS-related
    reports
       Optimization Support Features Overview
       Optimizer Report Generation
       Pipelining for IA-64 Architecture
    resources
    restricting
    strategies
    system-related
optimization support
optimizations
   Enabling Automatic Optimizations
   Optimizing Applications Overview
   Restricting Optimizations
    compilation process
    default level of
    for specific processors
    helper thread
    high-level language
    options for IA-32 architectures
    options for IA-64 architecture
    overview of
       Automatic Optimizations Overview
       Profile-guided Optimizations Overview
    parallelization
    PGO methodology
    profile-guided
    SSP
    support features for
optimizer report generation
optimizing
    applications
    helping the compiler
    overview
    technical applications
optimizing performance
options for efficient compilation
options for IA-32 architectures
options used for IPO
OptReport support
ORDERED
    example of
    in DO directives
    overview of OpenMP* directives and clauses
overflow
overriding
    call to a runtime library routine
    loop unrolling
    software pipelining
    the threads number
overriding vectorization
overview
    of data scope attribute clauses
    of optimizing compilation
    of optimizing for specific processors
    of parallelism



packed structures
PARALLEL
    using SHARED clause in
parallel construct
PARALLEL DO
    and synchronization constructs
    example of
    SCHEDULE clause
    summary of OpenMP* directives and clauses
    using COPYIN clause in
    using DEFAULT clause in
    using FIRSTPRIVATE clause in
    using LASTPRIVATE clause in
    using PRIVATE clause in
    using REDUCTION clause in
    using SHARED clause in
parallel invocations with makefile
PARALLEL OpenMP* directive
   COPYIN Clause
   DEFAULT Clause
   OpenMP Directives and Clauses Summary
   PRIVATE, FIRSTPRIVATE, and LASTPRIVATE Clauses
   REDUCTION Clause
   SHARED Clause
   Synchronization Constructs
parallel processing
    thread model
parallel programming
   Optimizing Applications Overview
   Parallelism Overview
parallel regions
    directive defining
    directives affecting
    library routine affecting
PARALLEL SECTIONS
    and synchronization constructs
    example of
    summary of OpenMP* directives
    using COPYIN clause in
    using DEFAULT clause in
    using FIRSTPRIVATE clause in
    using LASTPRIVATE clause in
    using PRIVATE clause in
    using REDUCTION clause in
    using SHARED clause in
PARALLEL WORKSHARE
    using
parallelism
   Auto-parallelization Overview
   OpenMP* Run-time Library Routines
   Parallelism: an Overview
parallelization
   Auto-parallelization Overview
   Loop Parallelization and Vectorization
   Parallelization Overview
   Programming with Auto-parallelization
    diagnostic
passing
    array arguments efficiently
performance analyzer
performance issues with IPO
PGO
PGO API
    _PGOPTI_Prof_Dump_And_Reset
    _PGOPTI_Prof_Reset
    _PGOPTI_Set_Interval_Prof_Dump
    enable
PGO tools
    code coverage tool
    helper threads
    profmerge
    proforder
    profrun
    software-based precomputation
    test prioritization tool
pgopti.dpi file
   Basic PGO Options
   PGO Environment Variables
pgopti.spi file
pgouser.h header file
pipelining
    affect of LOOP COUNT on
    for IA-64 architecture based applications
pointer aliasing
preempting functions
PREFETCH
    options used for
    using
prefetches of data
    optimizations for
preparing code
prioritizing application tests
PRIVATE
    in DEFAULT clause
    in the DO directive
    relationship to REDUCTION clause
    summary of data scope attribute clauses
       Data Scope Attribute Clauses Overview
       OpenMP Directives and Clauses Summary
    used in PARALLEL directive
    using
processor
    achieving optimum performance for
processors
    targeting IA-32 architecture using options
    targeting Itanium(R) processors using options
PROF_DIR environment variable
PROF_DUMP_INTERVAL environment variable
PROF_NO_CLOBBER environment variable
profile-guided optimization
    API support
    dumping profile information
    environment variables
    example of
    interval profile dumping
    options
    overview
    phases
    resetting dynamic profile counters
    resetting profile information
    support
    usage model
profile-optimized code
   Basic PGO Options
   PGO API Support Overview
profile data
    dumping
       Dumping Profile Information
       Interval Profile Dumping
    resetting dynamic counters for
profiling
    generating information
profmerge
profrun
    .hpi file
    .tb5 file
    requirements
    SSP
program loops
pseudo code
    parallel processing model


>>