src/gromacs/mdlib/gmx_omp_nthreads.h

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2012,2013,2014,2015,2016 by the GROMACS development team.
   5  * Copyright (c) 2017,2018,2019,2020, by the GROMACS development team, led by
   6  * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
   7  * and including many others, as listed in the AUTHORS file in the
   8  * top-level source directory and at http://www.gromacs.org.
   9  *
  10  * GROMACS is free software; you can redistribute it and/or
  11  * modify it under the terms of the GNU Lesser General Public License
  12  * as published by the Free Software Foundation; either version 2.1
  13  * of the License, or (at your option) any later version.
  14  *
  15  * GROMACS is distributed in the hope that it will be useful,
  16  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  17  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  18  * Lesser General Public License for more details.
  19  *
  20  * You should have received a copy of the GNU Lesser General Public
  21  * License along with GROMACS; if not, see
  22  * http://www.gnu.org/licenses, or write to the Free Software Foundation,
  23  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
  24  *
  25  * If you want to redistribute modifications to GROMACS, please
  26  * consider that scientific software is very special. Version
  27  * control is crucial - bugs must be traceable. We will be happy to
  28  * consider code for inclusion in the official distribution, but
  29  * derived work must not be called official GROMACS. Details are found
  30  * in the README & COPYING files - if they are missing, get the
  31  * official version at http://www.gromacs.org.
  32  *
  33  * To help us fund GROMACS development, we humbly ask that you cite
  34  * the research papers on the package. Check out http://www.gromacs.org.
  35  */
  36
  37 #ifndef GMX_OMP_NTHREADS_H
  38 #define GMX_OMP_NTHREADS_H
  39
  40 #include <stdio.h>
  41
  42 #include "gromacs/utility/basedefinitions.h"
  43
  44 struct t_commrec;
  45
  46 namespace gmx
  47 {
  48 class MDLogger;
  49 }
  50
  51 /** Enum values corresponding to multithreaded algorithmic modules. */
  52 typedef enum module_nth
  53 {
  54     /* Default is meant to be used in OMP regions outside the named
  55      * algorithmic modules listed below. */
  56     emntDefault,
  57     emntDomdec,
  58     emntPairsearch,
  59     emntNonbonded,
  60     emntBonded,
  61     emntPME,
  62     emntUpdate,
  63     emntVSITE,
  64     emntLINCS,
  65     emntSETTLE,
  66     emntNR
  67 } module_nth_t;
  68
  69 /*! \brief
  70  * Initializes the per-module thread count.
  71  *
  72  * It is compatible with tMPI, thread-safety is ensured (for the features
  73  * available with tMPI).
  74  * This function should caled only once during the initialization of mdrun. */
  75 void gmx_omp_nthreads_init(const gmx::MDLogger& fplog,
  76                            t_commrec*           cr,
  77                            int                  nthreads_hw_avail,
  78                            int                  numRanksOnThisNode,
  79                            int                  omp_nthreads_req,
  80                            int                  omp_nthreads_pme_req,
  81                            gmx_bool             bCurrNodePMEOnly);
  82
  83 /*! \brief
  84  * Returns the number of threads to be used in the given module \p mod. */
  85 int gmx_omp_nthreads_get(int mod);
  86
  87 /*! \brief
  88  * Returns the number of threads to be used in the given module \p mod for simple rvec operations.
  89  *
  90  * When the, potentially, parallel task only consists of a loop of clear_rvec
  91  * or rvec_inc for nrvec elements, the OpenMP overhead might be higher than
  92  * the reduction in computional cost due to parallelization. This routine
  93  * returns 1 when the overhead is expected to be higher than the gain.
  94  */
  95 static inline int gmx_omp_nthreads_get_simple_rvec_task(int mod, int nrvec)
  96 {
  97     /* There can be a relatively large overhead to an OpenMP parallel for loop.
  98      * This overhead increases, slowly, with the numbe of threads used.
  99      * The computational gain goes as 1/#threads. The two effects combined
 100      * lead to a cross-over point for a (non-)parallel loop at loop count
 101      * that is not strongly dependent on the thread count.
 102      * Note that a (non-)parallel loop can have benefit later in the code
 103      * due to generating more cache hits, depending on how the next lask
 104      * that accesses the same data is (not) parallelized over threads.
 105      *
 106      * A value of 2000 is the switch-over point for Haswell without
 107      * hyper-threading. With hyper-threading it is about a factor 1.5 higher.
 108      */
 109     const int nrvec_omp = 2000;
 110
 111     if (nrvec < nrvec_omp)
 112     {
 113         return 1;
 114     }
 115     else
 116     {
 117         return gmx_omp_nthreads_get(mod);
 118     }
 119 }
 120
 121 /*! \brief Sets the number of threads to be used in module.
 122  *
 123  * Intended for use in testing. */
 124 void gmx_omp_nthreads_set(int mod, int nthreads);
 125
 126 /*! \brief
 127  * Read the OMP_NUM_THREADS env. var. and check against the value set on the
 128  * command line. */
 129 void gmx_omp_nthreads_read_env(const gmx::MDLogger& mdlog, int* nthreads_omp);
 130
 131 #endif