<<

READ
real object files
    compiling with
RECL
    specifier for OPEN
record buffers
    efficient use of
record structures
    and alignment
REDUCTION
   Data Scope Attribute Clauses Overview
   OpenMP* Directives and Clauses Summary
   Parallel Region Directives
   REDUCTION Clause
   Worksharing Construct Directives
    variables
report generation
report software pipelining (SWP)
resetting
    dynamic profile counters
    profile information
routines
    Intel extension
    OpenMP* run-time
    timing
run-time checks
    processor-specific
run-time performance
    improving
    slowing down



sample of timing
scalar clean-up iterations
scalar replacement
scalars
    allocation of
SCHEDULE
   Specifying Schedule Type and Chunk Size
   Worksharing Construct Directives
    DYNAMIC
    GUIDED
    RUNTIME
    STATIC
serial execution
shared scalars
shared scoping
shared variables
significand
single-precision real
   Coding Guidelines for IntelŪ Architectures
   Improving Run-time Efficiency
SMP systems
software pipelining
   Parallelism Overview
   SWP Reports
    affect of LOOP COUNT on
    for ItaniumŪ-based applications
    optimization
    reports
source code
specialized code
   Automatic Processor-specific Optimization (IA-32 Only)
   Coding Guidelines for Intel Architectures
   Parallelism Overview
specifiers
    -Qoption compiler option
    /Qoption compiler option
specifying symbol visibility
SSE
    optimizing
SSE2
SSP
    profun
    using with -prof-gen
    using with -prof-use
    using with -ssp
    using with /Qprof-gen
    using with /Qprof-use
    using with /Qssp
stacks
   Optimizing Compilation Process Overview
   Stacks: Automatic Allocation and Checking
statement functions
statements
    in the loop body
strategies for optimization
Stream_LF records
Streaming SIMD Extensions
   Coding Guidelines for IntelŪ Architectures
   Key Programming Guidelines for Vectorization
   Strip-mining and Cleanup
stripmining
structures
    record
subroutines in the OpenMP* run-time library
support
    for loop unrolling
    for OpenMP*
    for optimization
    for prefetching
    for vectorization
    parallel run-time
SWP
    SWP reports
symbol visibility
    specifying
symbol visibility on Linux*
synchronization
   Auto-parallelization Overview
   Parallelism Overview
    constructs



targeting
    IA-32 processors
    ItaniumŪ processors
targeting optimizations
targeting processors
    IA-32
    ItaniumŪ
technical applications
test-prioritization tool
    examples
    options
    requirements
    usage
testing applications
thread pooling
threads
    changing the number of
    parallel processing model for
    thread-level parallelism
    thread sleep time
threshold control for auto-parallelization
timeout
timing
    OpenMP* routines for
tool options
    code-coverage tool
    profmerge
    proforder
    profrun
    test-prioritization tool
tools
    code-coverage tool
    strategies
    test-prioritization tool
    tuning
transformations
    loop
    reordering
tuning
types of loop vectorized



UBC buffers
unaligned data
unbuffered WRITEs
underflow
   Coding Guidelines for IntelŪ Architectures
   Floating-point Options for Multiple Architectures
   Stacks: Automatic Allocation and Checking
unformatted files
unvectorizable copy
usage rules
   Creating Multifile IPO Executable
   Programming with OpenMP*
user functions
using
    advanced PGO
    auto-parallelization
       Auto-parallelization Overview
       Parallelism: an Overview
    dynamic libraries
    EQUIVALENCE statements
    floating-point conversions
    formatted or unformatted files
    implied-DO loops
    intrinsics
    memory
    noniterative worksharing SECTIONS
    OpenMP*
    profile-guided optimization
    slow arithmetic operators
    timing for an application
    unbuffered WRITEs
    worksharing
utilities
    profmerge
    proforder



variables
    automatic
    length of
    loop assigns for
    PGO environment
    private scoping for
    renaming
vector copy
vector dependencies
vectorization
    examples
    options
    options for
    overview
    patterns
       Data Dependency
       Vectorization Support
    programming guidelines
       Key Programming Guidelines for Vectorization
       Vectorization Overview (IA-32 Only)
    support for
vectorizing
    loops
       Loop Constructs
       Profile-guided Optimizations Overview
VOLATILE
VTune(TM) Analyzer
   Profrun Utility
   Using Intel Performance Analysis Tools
    profrun



worker thread
WORKSHARE
worksharing
   Auto-parallelization Overview
   Combined Parallel and Worksharing Constructs
   OpenMP* Directives and Clauses Summary
   Parallel Region Directives
   Parallelism Overview
   Worksharing Construct Directives
    directives
       Combined Parallel and Worksharing Constructs
       Worksharing Construct Directives



zero denormal values