URL https://opencores.org/ocsvn/openrisc/openrisc/trunk
Subversion Repositories openrisc

[/] [openrisc/] [trunk/] [gnu-stable/] [gcc-4.5.1/] [libstdc++-v3/] [doc/] [xml/] [manual/] [parallel_mode.xml] - Blame information for rev 826

Details | Compare with Previous | View Log



 "http://www.oasis-open.org/docbook/xml/4.5/docbookx.dtd"
[ ]>
 


 

  
    
      C++
    
    
      library
    
    
      parallel
    
  

 
Parallel Mode
 
 The libstdc++ parallel mode is an experimental parallel
implementation of many algorithms the C++ Standard Library.

 

Several of the standard algorithms, for instance
std::sort, are made parallel using OpenMP
annotations. These parallel mode constructs and can be invoked by
explicit source declaration or by compiling existing sources with a
specific compiler flag.

 
 

  Intro
 
The following library components in the include
numeric are included in the parallel mode:

  std::accumulate
  std::adjacent_difference
  std::inner_product
  std::partial_sum

 
The following library components in the include
algorithm are included in the parallel mode:

  std::adjacent_find
  std::count
  std::count_if
  std::equal
  std::find
  std::find_if
  std::find_first_of
  std::for_each
  std::generate
  std::generate_n
  std::lexicographical_compare
  std::mismatch
  std::search
  std::search_n
  std::transform
  std::replace
  std::replace_if
  std::max_element
  std::merge
  std::min_element
  std::nth_element
  std::partial_sort
  std::partition
  std::random_shuffle
  std::set_union
  std::set_intersection
  std::set_symmetric_difference
  std::set_difference
  std::sort
  std::stable_sort
  std::unique_copy

 

 

  Semantics
 
 The parallel mode STL algorithms are currently not exception-safe,
i.e. user-defined functors must not throw exceptions.
Also, the order of execution is not guaranteed for some functions, of course.
Therefore, user-defined functors should not have any concurrent side effects.

 
 Since the current GCC OpenMP implementation does not support
OpenMP parallel regions in concurrent threads,
it is not possible to call parallel STL algorithm in
concurrent threads, either.
It might work with other compilers, though.
 

 

  Using
 

  Prerequisite Compiler Flags
 

  Any use of parallel functionality requires additional compiler
  and runtime support, in particular support for OpenMP. Adding this support is
  not difficult: just compile your application with the compiler
  flag -fopenmp. This will link
  in libgomp, the GNU
  OpenMP implementation,
  whose presence is mandatory.

 

In addition, hardware that supports atomic operations and a compiler
  capable of producing atomic operations is mandatory: GCC defaults to no
  support for atomic operations on some common hardware
  architectures. Activating atomic operations may require explicit
  compiler flags on some targets (like sparc and x86), such
  as -march=i686,
  -march=native or -mcpu=v9. See
  the GCC manual for more information.

 

 

  Using Parallel Mode
 

  To use the libstdc++ parallel mode, compile your application with
  the prerequisite flags as detailed above, and in addition
  add -D_GLIBCXX_PARALLEL. This will convert all
  use of the standard (sequential) algorithms to the appropriate parallel
  equivalents. Please note that this doesn't necessarily mean that
  everything will end up being executed in a parallel manner, but
  rather that the heuristics and settings coded into the parallel
  versions will be used to determine if all, some, or no algorithms
  will be executed using parallel variants.

 
Note that the _GLIBCXX_PARALLEL define may change the
  sizes and behavior of standard class templates such as
  std::search, and therefore one can only link code
  compiled with parallel mode and code compiled without parallel mode
  if no instantiation of a container is passed between the two
  translation units. Parallel mode functionality has distinct linkage,
  and cannot be confused with normal mode symbols.


 

  Using Specific Parallel Components
 
When it is not feasible to recompile your entire application, or
  only specific algorithms need to be parallel-aware, individual
  parallel algorithms can be made available explicitly. These
  parallel algorithms are functionally equivalent to the standard
  drop-in algorithms used in parallel mode, but they are available in
  a separate namespace as GNU extensions and may be used in programs
  compiled with either release mode or with parallel mode.

 
 
An example of using a parallel version
of std::sort, but no other parallel algorithms, is:

 

#include <vector>
#include <parallel/algorithm>
 
int main()
{
  std::vector<int> v(100);
 
  // ...
 
  // Explicitly force a call to parallel sort.
  __gnu_parallel::sort(v.begin(), v.end());
  return 0;
}

 

Then compile this code with the prerequisite compiler flags
(-fopenmp and any necessary architecture-specific
flags for atomic operations.)

 
 The following table provides the names and headers of all the
  parallel algorithms that can be used in a similar manner:

 

 

 

 

  Design
  
  

  Interface Basics
 

All parallel algorithms are intended to have signatures that are
equivalent to the ISO C++ algorithms replaced. For instance, the
std::adjacent_find function is declared as:


namespace std
{
  template<typename _FIter>
    _FIter
    adjacent_find(_FIter, _FIter);
}

 

Which means that there should be something equivalent for the parallel
version. Indeed, this is the case:

 

namespace std
{
  namespace __parallel
  {
    template<typename _FIter>
      _FIter
      adjacent_find(_FIter, _FIter);
 
    ...
  }
}

 
But.... why the ellipses?

 
 The ellipses in the example above represent additional overloads
required for the parallel version of the function. These additional
overloads are used to dispatch calls from the ISO C++ function
signature to the appropriate parallel function (or sequential
function, if no parallel functions are deemed worthy), based on either
compile-time or run-time conditions.

 
 The available signature options are specific for the different
algorithms/algorithm classes.
 
 The general view of overloads for the parallel algorithms look like this:


   ISO C++ signature
   ISO C++ signature + sequential_tag argument
   ISO C++ signature + algorithm-specific tag type
    (several signatures)

 
 Please note that the implementation may use additional functions
(designated with the _switch suffix) to dispatch from the
ISO C++ signature to the correct parallel version. Also, some of the
algorithms do not have support for run-time conditions, so the last
overload is therefore missing.

 
 

 

  Configuration and Tuning
 
 

  Setting up the OpenMP Environment
 

Several aspects of the overall runtime environment can be manipulated
by standard OpenMP function calls.

 

To specify the number of threads to be used for the algorithms globally,
use the function omp_set_num_threads. An example:

 

#include <stdlib.h>
#include <omp.h>
 
int main()
{
  // Explicitly set number of threads.
  const int threads_wanted = 20;
  omp_set_dynamic(false);
  omp_set_num_threads(threads_wanted);
 
  // Call parallel mode algorithms.
 
  return 0;
}

 

 Some algorithms allow the number of threads being set for a particular call,
 by augmenting the algorithm variant.
 See the next section for further information.

 

Other parts of the runtime environment able to be manipulated include
nested parallelism (omp_set_nested), schedule kind
(omp_set_schedule), and others. See the OpenMP
documentation for more information.

 

 

  Compile Time Switches
 

To force an algorithm to execute sequentially, even though parallelism
is switched on in general via the macro _GLIBCXX_PARALLEL,
add __gnu_parallel::sequential_tag() to the end
of the algorithm's argument list.

 

Like so:

 

std::sort(v.begin(), v.end(), __gnu_parallel::sequential_tag());

 

Some parallel algorithm variants can be excluded from compilation by
preprocessor defines. See the doxygen documentation on
compiletime_settings.h and features.h for details.

 

For some algorithms, the desired variant can be chosen at compile-time by
appending a tag object. The available options are specific to the particular
algorithm (class).

 

For the "embarrassingly parallel" algorithms, there is only one "tag object
type", the enum _Parallelism.
It takes one of the following values,
__gnu_parallel::parallel_tag,
__gnu_parallel::balanced_tag,
__gnu_parallel::unbalanced_tag,
__gnu_parallel::omp_loop_tag,
__gnu_parallel::omp_loop_static_tag.
This means that the actual parallelization strategy is chosen at run-time.
(Choosing the variants at compile-time will come soon.)

 

For the following algorithms in general, we have
__gnu_parallel::parallel_tag and
__gnu_parallel::default_parallel_tag, in addition to
__gnu_parallel::sequential_tag.
__gnu_parallel::default_parallel_tag chooses the default
algorithm at compiletime, as does omitting the tag.
__gnu_parallel::parallel_tag postpones the decision to runtime
(see next section).
For all tags, the number of threads desired for this call can optionally be
passed to the respective tag's constructor.

 

The multiway_merge algorithm comes with the additional choices,
__gnu_parallel::exact_tag and
__gnu_parallel::sampling_tag.
Exact and sampling are the two available splitting strategies.

 

For the sort and stable_sort algorithms, there are
several additional choices, namely
__gnu_parallel::multiway_mergesort_tag,
__gnu_parallel::multiway_mergesort_exact_tag,
__gnu_parallel::multiway_mergesort_sampling_tag,
__gnu_parallel::quicksort_tag, and
__gnu_parallel::balanced_quicksort_tag.
Multiway mergesort comes with the two splitting strategies for multi-way
merging. The quicksort options cannot be used for stable_sort.

 

 

  Run Time Settings and Defaults
 

The default parallelization strategy, the choice of specific algorithm
strategy, the minimum threshold limits for individual parallel
algorithms, and aspects of the underlying hardware can be specified as
desired via manipulation
of __gnu_parallel::_Settings member data.

 

First off, the choice of parallelization strategy: serial, parallel,
or heuristically deduced. This corresponds
to __gnu_parallel::_Settings::algorithm_strategy and is a
value of enum __gnu_parallel::_AlgorithmStrategy
type. Choices
include: heuristic, force_sequential,
and force_parallel. The default is heuristic.

 
 

Next, the sub-choices for algorithm variant, if not fixed at compile-time.
Specific algorithms like find or sort
can be implemented in multiple ways: when this is the case,
a __gnu_parallel::_Settings member exists to
pick the default strategy. For
example, __gnu_parallel::_Settings::sort_algorithm can
have any values of
enum __gnu_parallel::_SortAlgorithm: MWMS, QS,
or QS_BALANCED.

 

Likewise for setting the minimal threshold for algorithm
parallelization.  Parallelism always incurs some overhead. Thus, it is
not helpful to parallelize operations on very small sets of
data. Because of this, measures are taken to avoid parallelizing below
a certain, pre-determined threshold. For each algorithm, a minimum
problem size is encoded as a variable in the
active __gnu_parallel::_Settings object.  This
threshold variable follows the following naming scheme:
__gnu_parallel::_Settings::[algorithm]_minimal_n.  So,
for fill, the threshold variable
is __gnu_parallel::_Settings::fill_minimal_n,

 

Finally, hardware details like L1/L2 cache size can be hardwired
via __gnu_parallel::_Settings::L1_cache_size and friends.

 


 

All these configuration variables can be changed by the user, if
desired.
There exists one global instance of the class _Settings,
i. e. it is a singleton. It can be read and written by calling
__gnu_parallel::_Settings::get and
__gnu_parallel::_Settings::set, respectively.
Please note that the first call return a const object, so direct manipulation
is forbidden.
See 
  settings.h
for complete details.

 

A small example of tuning the default:

 

#include <parallel/algorithm>
#include <parallel/settings.h>
 
int main()
{
  __gnu_parallel::_Settings s;
  s.algorithm_strategy = __gnu_parallel::force_parallel;
  __gnu_parallel::_Settings::set(s);
 
  // Do work... all algorithms will be parallelized, always.
 
  return 0;
}

 

 

 

  Implementation Namespaces
 
 One namespace contain versions of code that are always
explicitly sequential:
__gnu_serial.

 
 Two namespaces contain the parallel mode:
std::__parallel and __gnu_parallel.

 
 Parallel implementations of standard components, including
template helpers to select parallelism, are defined in namespace
std::__parallel. For instance, std::transform from algorithm has a parallel counterpart in
std::__parallel::transform from parallel/algorithm. In addition, these parallel
implementations are injected into namespace
__gnu_parallel with using declarations.

 
 Support and general infrastructure is in namespace
__gnu_parallel.

 
 More information, and an organized index of types and functions
related to the parallel mode on a per-namespace basis, can be found in
the generated source documentation.

 

 

 

  Testing
 
  
    Both the normal conformance and regression tests and the
    supplemental performance tests work.
  
 
  
    To run the conformance and regression tests with the parallel mode
    active,
  
 
  
  make check-parallel
  
 
  
    The log and summary files for conformance testing are in the
    testsuite/parallel directory.
  
 
  
    To run the performance tests with the parallel mode active,
  
 
  
  make check-performance-parallel
  
 
  
    The result file for performance testing are in the
    testsuite directory, in the file
    libstdc++_performance.sum. In addition, the
    policy-based containers have their own visualizations, which have
    additional software dependencies than the usual bare-boned text
    file, and can be generated by using the make
    doc-performance rule in the testsuite's Makefile.


 

Bibliography
 
  
    </code></pre></td>
      </tr>
      <tr valign="middle">
         <td>840</td>
         <td></td>
         <td></td>
         <td class="code"><pre><code>      Parallelization of Bulk Operations for STL Dictionaries</code></pre></td>
      </tr>
      <tr valign="middle">
         <td>841</td>
         <td></td>
         <td></td>
         <td class="code"><pre><code>    
 
    
      Johannes
      Singler
    
    
      Leonor
      Frias
    
 
    
      2007
      
    
 
    
      
        Workshop on Highly Parallel Processing on a Chip (HPPC) 2007. (LNCS)
      
    
  
 
  
    </code></pre></td>
      </tr>
      <tr valign="middle">
         <td>866</td>
         <td></td>
         <td></td>
         <td class="code"><pre><code>      The Multi-Core Standard Template Library</code></pre></td>
      </tr>
      <tr valign="middle">
         <td>867</td>
         <td></td>
         <td></td>
         <td class="code"><pre><code>    
 
    
      Johannes
      Singler
    
    
      Peter
      Sanders
    
    
      Felix
      Putze
    
 
    
      2007
      
    
 
    
      
         Euro-Par 2007: Parallel Processing. (LNCS 4641)
      
    
  
 

 

Browse

Tools

Subversion Repositories openrisc

[/] [openrisc/] [trunk/] [gnu-stable/] [gcc-4.5.1/] [libstdc++-v3/] [doc/] [xml/] [manual/] [parallel_mode.xml] - Blame information for rev 826
