Remove extra semicolons causing clang warnings.

[gromacs.git] / src / gromacs / mdlib / nbnxn_pairlist.h

/*
 * This file is part of the GROMACS molecular simulation package.
 *
 * Copyright (c) 2012,2013,2014,2015,2016,2017,2018,2019, by the GROMACS development team, led by
 * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
 * and including many others, as listed in the AUTHORS file in the
 * top-level source directory and at http://www.gromacs.org.
 *
 * GROMACS is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public License
 * as published by the Free Software Foundation; either version 2.1
 * of the License, or (at your option) any later version.
 *
 * GROMACS is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with GROMACS; if not, see
 * http://www.gnu.org/licenses, or write to the Free Software Foundation,
 * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
 *
 * If you want to redistribute modifications to GROMACS, please
 * consider that scientific software is very special. Version
 * control is crucial - bugs must be traceable. We will be happy to
 * consider code for inclusion in the official distribution, but
 * derived work must not be called official GROMACS. Details are found
 * in the README & COPYING files - if they are missing, get the
 * official version at http://www.gromacs.org.
 *
 * To help us fund GROMACS development, we humbly ask that you cite
 * the research papers on the package. Check out http://www.gromacs.org.
 */

#ifndef _nbnxn_pairlist_h
#define _nbnxn_pairlist_h

#include "config.h"

#include <cstddef>

#include "gromacs/gpu_utils/hostallocator.h"
#include "gromacs/math/vectypes.h"
#include "gromacs/mdlib/nbnxn_consts.h"
#include "gromacs/mdtypes/nblist.h"
#include "gromacs/utility/basedefinitions.h"
#include "gromacs/utility/bitmask.h"
#include "gromacs/utility/defaultinitializationallocator.h"
#include "gromacs/utility/real.h"

struct NbnxnPairlistCpuWork;
struct NbnxnPairlistGpuWork;
struct tMPI_Atomic;

/* Convenience type for vector with aligned memory */
template<typename T>
using AlignedVector = std::vector < T, gmx::AlignedAllocator < T>>;

/* Convenience type for vector that avoids initialization at resize() */
template<typename T>
using FastVector = std::vector < T, gmx::DefaultInitializationAllocator < T>>;

/*! \cond INTERNAL */

/*! \brief The setup for generating and pruning the nbnxn pair list.
 *
 * Without dynamic pruning rlistOuter=rlistInner.
 */
struct NbnxnListParameters
{
    /*! \brief Constructor producing a struct with dynamic pruning disabled
     */
    NbnxnListParameters(real rlist) :
        useDynamicPruning(false),
        nstlistPrune(-1),
        rlistOuter(rlist),
        rlistInner(rlist),
        numRollingParts(1)
    {
    }

    bool useDynamicPruning; //!< Are we using dynamic pair-list pruning
    int  nstlistPrune;      //!< Pair-list dynamic pruning interval
    real rlistOuter;        //!< Cut-off of the larger, outer pair-list
    real rlistInner;        //!< Cut-off of the smaller, inner pair-list
    int  numRollingParts;   //!< The number parts to divide the pair-list into for rolling pruning, a value of 1 gives no rolling pruning
};

/*! \endcond */

/* With CPU kernels the i-cluster size is always 4 atoms. */
static constexpr int c_nbnxnCpuIClusterSize = 4;

/* With GPU kernels the i and j cluster size is 8 atoms for CUDA and can be set at compile time for OpenCL */
#if GMX_GPU == GMX_GPU_OPENCL
static constexpr int c_nbnxnGpuClusterSize = GMX_OPENCL_NB_CLUSTER_SIZE;
#else
static constexpr int c_nbnxnGpuClusterSize = 8;
#endif

/* The number of clusters in a pair-search cell, used for GPU */
static constexpr int c_gpuNumClusterPerCellZ = 2;
static constexpr int c_gpuNumClusterPerCellY = 2;
static constexpr int c_gpuNumClusterPerCellX = 2;
static constexpr int c_gpuNumClusterPerCell  = c_gpuNumClusterPerCellZ*c_gpuNumClusterPerCellY*c_gpuNumClusterPerCellX;

/* In CUDA the number of threads in a warp is 32 and we have cluster pairs
 * of 8*8=64 atoms, so it's convenient to store data for cluster pair halves.
 */
static constexpr int c_nbnxnGpuClusterpairSplit = 2;

/* The fixed size of the exclusion mask array for a half cluster pair */
static constexpr int c_nbnxnGpuExclSize = c_nbnxnGpuClusterSize*c_nbnxnGpuClusterSize/c_nbnxnGpuClusterpairSplit;

/* A buffer data structure of 64 bytes
 * to be placed at the beginning and end of structs
 * to avoid cache invalidation of the real contents
 * of the struct by writes to neighboring memory.
 */
typedef struct {
    int dummy[16];
} gmx_cache_protect_t;

/* Abstract type for pair searching data */
typedef struct nbnxn_search * nbnxn_search_t;

/* Function that should return a pointer *ptr to memory
 * of size nbytes.
 * Error handling should be done within this function.
 */
typedef void nbnxn_alloc_t (void **ptr, size_t nbytes);

/* Function that should free the memory pointed to by *ptr.
 * NULL should not be passed to this function.
 */
typedef void nbnxn_free_t (void *ptr);

/* This is the actual cluster-pair list j-entry.
 * cj is the j-cluster.
 * The interaction bits in excl are indexed i-major, j-minor.
 * The cj entries are sorted such that ones with exclusions come first.
 * This means that once a full mask (=NBNXN_INTERACTION_MASK_ALL)
 * is found, all subsequent j-entries in the i-entry also have full masks.
 */
struct nbnxn_cj_t
{
    int          cj;    /* The j-cluster                    */
    unsigned int excl;  /* The exclusion (interaction) bits */
};

/* In nbnxn_ci_t the integer shift contains the shift in the lower 7 bits.
 * The upper bits contain information for non-bonded kernel optimization.
 * Simply calculating LJ and Coulomb for all pairs in a cluster pair is fine.
 * But three flags can be used to skip interactions, currently only for subc=0
 * !(shift & NBNXN_CI_DO_LJ(subc))   => we can skip LJ for all pairs
 * shift & NBNXN_CI_HALF_LJ(subc)    => we can skip LJ for the second half of i
 * !(shift & NBNXN_CI_DO_COUL(subc)) => we can skip Coulomb for all pairs
 */
#define NBNXN_CI_SHIFT          127
#define NBNXN_CI_DO_LJ(subc)    (1<<(7+3*(subc)))
#define NBNXN_CI_HALF_LJ(subc)  (1<<(8+3*(subc)))
#define NBNXN_CI_DO_COUL(subc)  (1<<(9+3*(subc)))

/* Cluster-pair Interaction masks
 * Bit i*j-cluster-size + j tells if atom i and j interact.
 */
// TODO: Rename according to convention when moving into Nbnxn namespace
/* All interaction mask is the same for all kernels */
constexpr unsigned int NBNXN_INTERACTION_MASK_ALL       = 0xffffffffU;
/* 4x4 kernel diagonal mask */
constexpr unsigned int NBNXN_INTERACTION_MASK_DIAG      = 0x08ceU;
/* 4x2 kernel diagonal masks */
constexpr unsigned int NBNXN_INTERACTION_MASK_DIAG_J2_0 = 0x0002U;
constexpr unsigned int NBNXN_INTERACTION_MASK_DIAG_J2_1 = 0x002fU;
/* 4x8 kernel diagonal masks */
constexpr unsigned int NBNXN_INTERACTION_MASK_DIAG_J8_0 = 0xf0f8fcfeU;
constexpr unsigned int NBNXN_INTERACTION_MASK_DIAG_J8_1 = 0x0080c0e0U;

/* Simple pair-list i-unit */
struct nbnxn_ci_t
{
    int ci;             /* i-cluster             */
    int shift;          /* Shift vector index plus possible flags, see above */
    int cj_ind_start;   /* Start index into cj   */
    int cj_ind_end;     /* End index into cj     */
};

/* Grouped pair-list i-unit */
typedef struct {
    /* Returns the number of j-cluster groups in this entry */
    int numJClusterGroups() const
    {
        return cj4_ind_end - cj4_ind_start;
    }

    int sci;            /* i-super-cluster       */
    int shift;          /* Shift vector index plus possible flags */
    int cj4_ind_start;  /* Start index into cj4  */
    int cj4_ind_end;    /* End index into cj4    */
} nbnxn_sci_t;

/* Interaction data for a j-group for one warp */
struct nbnxn_im_ei_t
{
    // The i-cluster interactions mask for 1 warp
    unsigned int imask    = 0U;
    // Index into the exclusion array for 1 warp, default index 0 which means no exclusions
    int          excl_ind = 0;
};

typedef struct {
    int           cj[c_nbnxnGpuJgroupSize];         /* The 4 j-clusters */
    nbnxn_im_ei_t imei[c_nbnxnGpuClusterpairSplit]; /* The i-cluster mask data       for 2 warps   */
} nbnxn_cj4_t;

/* Struct for storing the atom-pair interaction bits for a cluster pair in a GPU pairlist */
struct nbnxn_excl_t
{
    /* Constructor, sets no exclusions, so all atom pairs interacting */
    nbnxn_excl_t()
    {
        for (unsigned int &pairEntry : pair)
        {
            pairEntry = NBNXN_INTERACTION_MASK_ALL;
        }
    }

    /* Topology exclusion interaction bits per warp */
    unsigned int pair[c_nbnxnGpuExclSize];
};

/* Cluster pairlist type for use on CPUs */
struct NbnxnPairlistCpu
{
    gmx_cache_protect_t     cp0;

    int                     na_ci;       /* The number of atoms per i-cluster        */
    int                     na_cj;       /* The number of atoms per j-cluster        */
    real                    rlist;       /* The radius for constructing the list     */
    FastVector<nbnxn_ci_t>  ci;          /* The i-cluster list                       */
    FastVector<nbnxn_ci_t>  ciOuter;     /* The outer, unpruned i-cluster list       */

    FastVector<nbnxn_cj_t>  cj;          /* The j-cluster list, size ncj             */
    FastVector<nbnxn_cj_t>  cjOuter;     /* The outer, unpruned j-cluster list       */
    int                     ncjInUse;    /* The number of j-clusters that are used by ci entries in this list, will be <= cj.size() */

    int                     nci_tot;     /* The total number of i clusters           */

    NbnxnPairlistCpuWork   *work;

    gmx_cache_protect_t     cp1;
};

/* Cluster pairlist type, with extra hierarchies, for on the GPU
 *
 * NOTE: for better performance when combining lists over threads,
 *       all vectors should use default initialization. But when
 *       changing this, excl should be intialized when adding entries.
 */
struct NbnxnPairlistGpu
{
    /* Constructor
     *
     * \param[in] pinningPolicy  Sets the pinning policy for all buffers used on the GPU
     */
    NbnxnPairlistGpu(gmx::PinningPolicy pinningPolicy);

    gmx_cache_protect_t            cp0;

    int                            na_ci; /* The number of atoms per i-cluster        */
    int                            na_cj; /* The number of atoms per j-cluster        */
    int                            na_sc; /* The number of atoms per super cluster    */
    real                           rlist; /* The radius for constructing the list     */
    // The i-super-cluster list, indexes into cj4;
    gmx::HostVector<nbnxn_sci_t>   sci;
    // The list of 4*j-cluster groups
    gmx::HostVector<nbnxn_cj4_t>   cj4;
    // Atom interaction bits (non-exclusions)
    gmx::HostVector<nbnxn_excl_t>  excl;
    // The total number of i-clusters
    int                            nci_tot;

    NbnxnPairlistGpuWork          *work;

    gmx_cache_protect_t            cp1;
};

typedef struct {
    int                nnbl;                  /* number of lists */
    NbnxnPairlistCpu **nbl;                   /* lists for CPU */
    NbnxnPairlistCpu **nbl_work;              /* work space for rebalancing lists */
    NbnxnPairlistGpu **nblGpu;                /* lists for GPU */
    gmx_bool           bCombined;             /* TRUE if lists get combined into one (the 1st) */
    gmx_bool           bSimple;               /* TRUE if the list of of type "simple"
                                                 (na_sc=na_s, no super-clusters used) */
    int                natpair_ljq;           /* Total number of atom pairs for LJ+Q kernel */
    int                natpair_lj;            /* Total number of atom pairs for LJ kernel   */
    int                natpair_q;             /* Total number of atom pairs for Q kernel    */
    t_nblist         **nbl_fep;               /* List of free-energy atom pair interactions */
    int64_t            outerListCreationStep; /* Step at which the outer list was created */
} nbnxn_pairlist_set_t;

enum {
    nbatXYZ, nbatXYZQ, nbatX4, nbatX8
};

// Struct that holds force and energy output buffers
struct nbnxn_atomdata_output_t
{
    /* Constructor
     *
     * \param[in] nb_kernel_type          Type of non-bonded kernel
     * \param[in] numEnergyGroups         The number of energy groups
     * \param[in] simdEnergyBufferStride  Stride for entries in the energy buffers for SIMD kernels
     * \param[in] pinningPolicy           Sets the pinning policy for all buffers used on the GPU
     */
    nbnxn_atomdata_output_t(int                nb_kernel_type,
                            int                numEnergyGroups,
                            int                simdEnergyBUfferStride,
                            gmx::PinningPolicy pinningPolicy);

    gmx::HostVector<real> f;      // f, size natoms*fstride
    gmx::HostVector<real> fshift; // Shift force array, size SHIFTS*DIM
    gmx::HostVector<real> Vvdw;   // Temporary Van der Waals group energy storage
    gmx::HostVector<real> Vc;     // Temporary Coulomb group energy storage
    AlignedVector<real>   VSvdw;  // Temporary SIMD Van der Waals group energy storage
    AlignedVector<real>   VSc;    // Temporary SIMD Coulomb group energy storage
};

/* Block size in atoms for the non-bonded thread force-buffer reduction,
 * should be a multiple of all cell and x86 SIMD sizes (i.e. 2, 4 and 8).
 * Should be small to reduce the reduction and zeroing cost,
 * but too small will result in overhead.
 * Currently the block size is NBNXN_BUFFERFLAG_SIZE*3*sizeof(real)=192 bytes.
 */
#if GMX_DOUBLE
#define NBNXN_BUFFERFLAG_SIZE   8
#else
#define NBNXN_BUFFERFLAG_SIZE  16
#endif

/* We store the reduction flags as gmx_bitmask_t.
 * This limits the number of flags to BITMASK_SIZE.
 */
#define NBNXN_BUFFERFLAG_MAX_THREADS  (BITMASK_SIZE)

/* Flags for telling if threads write to force output buffers */
typedef struct {
    int               nflag;       /* The number of flag blocks                         */
    gmx_bitmask_t    *flag;        /* Bit i is set when thread i writes to a cell-block */
    int               flag_nalloc; /* Allocation size of cxy_flag                       */
} nbnxn_buffer_flags_t;

/* LJ combination rules: geometric, Lorentz-Berthelot, none */
enum {
    ljcrGEOM, ljcrLB, ljcrNONE, ljcrNR
};

/* Struct that stores atom related data for the nbnxn module
 *
 * Note: performance would improve slightly when all std::vector containers
 *       in this struct would not initialize during resize().
 */
struct nbnxn_atomdata_t
{   //NOLINT(clang-analyzer-optin.performance.Padding)
    struct Params
    {
        /* Constructor
         *
         * \param[in] pinningPolicy  Sets the pinning policy for all data that might be transfered to a GPU
         */
        Params(gmx::PinningPolicy pinningPolicy);

        // The number of different atom types
        int                   numTypes;
        // Lennard-Jone 6*C6 and 12*C12 parameters, size numTypes*2*2
        gmx::HostVector<real> nbfp;
        // Combination rule, see enum defined above
        int                   comb_rule;
        // LJ parameters per atom type, size numTypes*2
        gmx::HostVector<real> nbfp_comb;
        // As nbfp, but with a stride for the present SIMD architecture
        AlignedVector<real>   nbfp_aligned;
        // Atom types per atom
        gmx::HostVector<int>  type;
        // LJ parameters per atom for fast SIMD loading
        gmx::HostVector<real> lj_comb;
        // Charges per atom, not set with format nbatXYZQ
        gmx::HostVector<real> q;
        // The number of energy groups
        int                   nenergrp;
        // 2log(nenergrp)
        int                   neg_2log;
        // The energy groups, one int entry per cluster, only set when needed
        gmx::HostVector<int>  energrp;
    };

    // Diagonal and topology exclusion helper data for all SIMD kernels
    struct SimdMasks
    {
        SimdMasks();

        // Helper data for setting up diagonal exclusion masks in the SIMD 4xN kernels
        AlignedVector<real>     diagonal_4xn_j_minus_i;
        // Helper data for setting up diaginal exclusion masks in the SIMD 2xNN kernels
        AlignedVector<real>     diagonal_2xnn_j_minus_i;
        // Filters for topology exclusion masks for the SIMD kernels
        AlignedVector<uint32_t> exclusion_filter;
        // Filters for topology exclusion masks for double SIMD kernels without SIMD int32 logical support
        AlignedVector<uint64_t> exclusion_filter64;
        // Array of masks needed for exclusions
        AlignedVector<real>     interaction_array;
    };

    /* Constructor
     *
     * \param[in] pinningPolicy  Sets the pinning policy for all data that might be transfered to a GPU
     */
    nbnxn_atomdata_t(gmx::PinningPolicy pinningPolicy);

    /* Returns a const reference to the parameters */
    const Params &params() const
    {
        return params_;
    }

    /* Returns a non-const reference to the parameters */
    Params &paramsDeprecated()
    {
        return params_;
    }

    /* Returns the current total number of atoms stored */
    int numAtoms() const
    {
        return numAtoms_;
    }

    /* Return the coordinate buffer, and q with xFormat==nbatXYZQ */
    gmx::ArrayRef<const real> x() const
    {
        return x_;
    }

    /* Return the coordinate buffer, and q with xFormat==nbatXYZQ */
    gmx::ArrayRef<real> x()
    {
        return x_;
    }

    /* Resizes the coordinate buffer and sets the number of atoms */
    void resizeCoordinateBuffer(int numAtoms);

    /* Resizes the force buffers for the current number of atoms */
    void resizeForceBuffers();

    private:
        // The LJ and charge parameters
        Params                     params_;
        // The total number of atoms currently stored
        int                        numAtoms_;
    public:
        int                        natoms_local; /* Number of local atoms                           */
        int                        XFormat;      /* The format of x (and q), enum                      */
        int                        FFormat;      /* The format of f, enum                              */
        gmx_bool                   bDynamicBox;  /* Do we need to update shift_vec every step?    */
        gmx::HostVector<gmx::RVec> shift_vec;    /* Shift vectors, copied from t_forcerec              */
        int                        xstride;      /* stride for a coordinate in x (usually 3 or 4)      */
        int                        fstride;      /* stride for a coordinate in f (usually 3 or 4)      */
    private:
        gmx::HostVector<real>      x_;           /* x and possibly q, size natoms*xstride              */

    public:
        // Masks for handling exclusions in the SIMD kernels
        const SimdMasks          simdMasks;

        /* Output data */
        std::vector<nbnxn_atomdata_output_t> out; /* Output data structures, 1 per thread */

        /* Reduction related data */
        gmx_bool                 bUseBufferFlags;     /* Use the flags or operate on all atoms     */
        nbnxn_buffer_flags_t     buffer_flags;        /* Flags for buffer zeroing+reduc.  */
        gmx_bool                 bUseTreeReduce;      /* Use tree for force reduction */
        tMPI_Atomic             *syncStep;            /* Synchronization step for tree reduce */
};

#endif