moebiusband/MD-Bench
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Initial refactoring of CUDA code

Browse Source 
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
This commit is contained in:
Rafael Ravedutti 2022-08-12 04:19:38 +02:00
parent 959ff65126
commit 065b596074
9 changed files with 127 additions and 144 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
config.mk
View File

@ -38,6 +38,10 @@ XTC_OUTPUT ?= false
# Check if cj is local when decreasing reaction force # Check if cj is local when decreasing reaction force
HALF_NEIGHBOR_LISTS_CHECK_CJ ?= false HALF_NEIGHBOR_LISTS_CHECK_CJ ?= false
# Configurations for CUDA
# Use CUDA host memory to optimize transfers
USE_CUDA_HOST_MEMORY ?= false
#Feature options #Feature options
OPTIONS = -DALIGNMENT=64 OPTIONS = -DALIGNMENT=64
#OPTIONS += More options #OPTIONS += More options

49
lammps/allocate.c
View File

@ -49,36 +49,61 @@ void *allocate(int alignment, size_t bytesize) {
return ptr; return ptr;
} }
void *reallocate(void* ptr, int alignment, size_t newBytesize, size_t oldBytesize) { void *reallocate(void* ptr, int alignment, size_t new_bytesize, size_t old_bytesize) {
void *newarray = allocate(alignment, newBytesize); void *newarray = allocate(alignment, new_bytesize);
if(ptr != NULL) { if(ptr != NULL) {
memcpy(newarray, ptr, oldBytesize); memcpy(newarray, ptr, old_bytesize);
free(ptr); free(ptr);
} }
return newarray; return newarray;
} }
#ifndef CUDA_TARGET #ifndef CUDA_TARGET
void *allocate_gpu(int alignment, size_t bytesize) { return NULL; } void *allocateGPU(size_t bytesize) { return NULL; }
void *reallocate_gpu(void *ptr, int alignment, size_t newBytesize, size_t oldBytesize) { return NULL; } void *reallocateGPU(void *ptr, size_t new_bytesize) { return NULL; }
void memcpyToGPU(void *d_ptr, void *h_ptr, size_t bytesize) {}
void memcpyFromGPU(void *h_ptr, void *d_ptr, size_t bytesize) {}
void memsetGPU(void *d_ptr, int value, size_t bytesize) {}
#else #else
#include <cuda_runtime.h> #include <cuda_runtime.h>
#include <cuda_atom.h> #include <cuda_atom.h>
void *allocate_gpu(int alignment, size_t bytesize) { void *allocateGPU(size_t bytesize) {
void *ptr; void *ptr;
checkCUDAError("allocate", cudaMallocHost((void **) &ptr, bytesize)); #ifdef CUDA_HOST_MEMORY
cuda_assert("allocateGPU", cudaMallocHost((void **) &ptr, bytesize));
#else
cuda_assert("allocateGPU", cudaMalloc((void **) &ptr, bytesize));
#endif
return ptr; return ptr;
} }
// Data is not preserved // Data is not preserved
void *reallocate_gpu(void *ptr, int alignment, size_t newBytesize, size_t oldBytesize) { void *reallocateGPU(void *ptr, size_t new_bytesize) {
void *newarray = allocate_gpu(alignment, newBytesize);
if(ptr != NULL) { if(ptr != NULL) {
#ifdef CUDA_HOST_MEMORY
cudaFreeHost(ptr); cudaFreeHost(ptr);
#else
cudaFree(ptr);
#endif
} }
return newarray; return allocateGPU(new_bytesize);
}
void memcpyToGPU(void *d_ptr, void *h_ptr, size_t bytesize) {
#ifndef CUDA_HOST_MEMORY
cuda_assert("memcpyToGPU", cudaMemcpy(d_ptr, h_ptr, bytesize, cudaMemcpyHostToDevice));
#endif
}
void memcpyFromGPU(void *h_ptr, void *d_ptr, size_t bytesize) {
#ifndef CUDA_HOST_MEMORY
cuda_assert("memcpyFromGPU", cudaMemcpy(h_ptr, d_ptr, bytesize, cudaMemcpyDeviceToHost));
#endif
}
void memsetGPU(void *d_ptr, int value, size_t bytesize) {
cuda_assert("memsetGPU", cudaMemset(d_ptr, value, bytesize));
} }
#endif #endif

8
lammps/atom.c
View File

@ -71,14 +71,6 @@ void createAtom(Atom *atom, Parameter *param) {
atom->Natoms = 4 * param->nx * param->ny * param->nz; atom->Natoms = 4 * param->nx * param->ny * param->nz;
atom->Nlocal = 0; atom->Nlocal = 0;
atom->ntypes = param->ntypes; atom->ntypes = param->ntypes;
#ifdef CUDA_TARGET
checkCUDAError( "atom->epsilon cudaMallocHost", cudaMallocHost((void**)&(atom->epsilon), atom->ntypes * atom->ntypes * sizeof(MD_FLOAT)) ); // atom->epsilon = allocate(ALIGNMENT, atom->ntypes * atom->ntypes * sizeof(MD_FLOAT));
checkCUDAError( "atom->sigma6 cudaMallocHost", cudaMallocHost((void**)&(atom->sigma6), atom->ntypes * atom->ntypes * sizeof(MD_FLOAT)) ); // atom->sigma6 = allocate(ALIGNMENT, atom->ntypes * atom->ntypes * sizeof(MD_FLOAT));
checkCUDAError( "atom->cutforcesq cudaMallocHost", cudaMallocHost((void**)&(atom->cutforcesq), atom->ntypes * atom->ntypes * sizeof(MD_FLOAT)) ); // atom->cutforcesq = allocate(ALIGNMENT, atom->ntypes * atom->ntypes * sizeof(MD_FLOAT));
checkCUDAError( "atom->cutneighsq cudaMallocHost", cudaMallocHost((void**)&(atom->cutneighsq), atom->ntypes * atom->ntypes * sizeof(MD_FLOAT)) ); // atom->cutneighsq = allocate(ALIGNMENT, atom->ntypes * atom->ntypes * sizeof(MD_FLOAT));
#endif
atom->epsilon = allocate(ALIGNMENT, atom->ntypes * atom->ntypes * sizeof(MD_FLOAT)); atom->epsilon = allocate(ALIGNMENT, atom->ntypes * atom->ntypes * sizeof(MD_FLOAT));
atom->sigma6 = allocate(ALIGNMENT, atom->ntypes * atom->ntypes * sizeof(MD_FLOAT)); atom->sigma6 = allocate(ALIGNMENT, atom->ntypes * atom->ntypes * sizeof(MD_FLOAT));
atom->cutforcesq = allocate(ALIGNMENT, atom->ntypes * atom->ntypes * sizeof(MD_FLOAT)); atom->cutforcesq = allocate(ALIGNMENT, atom->ntypes * atom->ntypes * sizeof(MD_FLOAT));

43
lammps/cuda/atom.cu
View File

@ -37,38 +37,29 @@ void initCuda(Atom *atom, Neighbor *neighbor, Atom *c_atom, Neighbor *c_neighbor
c_atom->Nghost = atom->Nghost; c_atom->Nghost = atom->Nghost;
c_atom->Nmax = atom->Nmax; c_atom->Nmax = atom->Nmax;
c_atom->ntypes = atom->ntypes; c_atom->ntypes = atom->ntypes;
c_atom->border_map = NULL; c_atom->border_map = NULL;
const int Nlocal = atom->Nlocal; c_atom->x = (MD_FLOAT *) allocateGPU(sizeof(MD_FLOAT) * atom->Nmax * 3);
c_atom->vx = (MD_FLOAT *) allocateGPU(sizeof(MD_FLOAT) * atom->Nmax * 3);
c_atom->fx = (MD_FLOAT *) allocateGPU(sizeof(MD_FLOAT) * atom->Nmax * 3);
c_atom->epsilon = (MD_FLOAT *) allocateGPU(sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes);
c_atom->sigma6 = (MD_FLOAT *) allocateGPU(sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes);
c_atom->cutforcesq = (MD_FLOAT *) allocateGPU(sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes);
c_atom->type = (int *) allocateGPU(sizeof(int) * atom->Nmax * 3);
c_neighbor->neighbors = (int *) allocateGPU(sizeof(int) * atom->Nmax * neighbor->maxneighs);
c_neighbor->numneigh = (int *) allocateGPU(sizeof(int) * atom->Nmax);
checkCUDAError( "c_atom->x malloc", cudaMalloc((void**)&(c_atom->x), sizeof(MD_FLOAT) * atom->Nmax * 3) ); memcpyToGPU(c_atom->x, atom->x, sizeof(MD_FLOAT) * atom->Nmax * 3);
checkCUDAError( "c_atom->x memcpy", cudaMemcpy(c_atom->x, atom->x, sizeof(MD_FLOAT) * atom->Nmax * 3, cudaMemcpyHostToDevice) ); memcpyToGPU(c_atom->vx, atom->vx, sizeof(MD_FLOAT) * atom->Nmax * 3);
memcpyToGPU(c_atom->sigma6, atom->sigma6, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes);
checkCUDAError( "c_atom->fx malloc", cudaMalloc((void**)&(c_atom->fx), sizeof(MD_FLOAT) * Nlocal * 3) ); memcpyToGPU(c_atom->epsilon, atom->epsilon, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes);
memcpyToGPU(c_atom->cutforcesq, atom->cutforcesq, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes);
checkCUDAError( "c_atom->vx malloc", cudaMalloc((void**)&(c_atom->vx), sizeof(MD_FLOAT) * Nlocal * 3) ); memcpyToGPU(c_atom->type, atom->type, sizeof(int) * atom->Nmax);
checkCUDAError( "c_atom->vx memcpy", cudaMemcpy(c_atom->vx, atom->vx, sizeof(MD_FLOAT) * Nlocal * 3, cudaMemcpyHostToDevice) );
checkCUDAError( "c_atom->type malloc", cudaMalloc((void**)&(c_atom->type), sizeof(int) * atom->Nmax) );
checkCUDAError( "c_atom->epsilon malloc", cudaMalloc((void**)&(c_atom->epsilon), sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes) );
checkCUDAError( "c_atom->sigma6 malloc", cudaMalloc((void**)&(c_atom->sigma6), sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes) );
checkCUDAError( "c_atom->cutforcesq malloc", cudaMalloc((void**)&(c_atom->cutforcesq), sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes) );
checkCUDAError( "c_neighbor->neighbors malloc", cudaMalloc((void**)&c_neighbor->neighbors, sizeof(int) * Nlocal * neighbor->maxneighs) );
checkCUDAError( "c_neighbor->numneigh malloc", cudaMalloc((void**)&c_neighbor->numneigh, sizeof(int) * Nlocal) );
checkCUDAError( "c_atom->type memcpy", cudaMemcpy(c_atom->type, atom->type, sizeof(int) * atom->Nmax, cudaMemcpyHostToDevice) );
checkCUDAError( "c_atom->sigma6 memcpy", cudaMemcpy(c_atom->sigma6, atom->sigma6, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes, cudaMemcpyHostToDevice) );
checkCUDAError( "c_atom->epsilon memcpy", cudaMemcpy(c_atom->epsilon, atom->epsilon, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes, cudaMemcpyHostToDevice) );
checkCUDAError( "c_atom->cutforcesq memcpy", cudaMemcpy(c_atom->cutforcesq, atom->cutforcesq, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes, cudaMemcpyHostToDevice) );
} }
void checkCUDAError(const char *msg, cudaError_t err) { void cuda_assert(const char *label, cudaError_t err) {
if (err != cudaSuccess) { if (err != cudaSuccess) {
//print a human readable error message printf("[CUDA Error]: %s: %s\r\n", label, cudaGetErrorString(err));
printf("[CUDA ERROR %s]: %s\r\n", msg, cudaGetErrorString(err));
exit(-1); exit(-1);
} }
} }

33
lammps/cuda/force.cu
View File

@ -33,6 +33,7 @@
extern "C" { extern "C" {
#include <allocate.h>
#include <atom.h> #include <atom.h>
#include <cuda_atom.h> #include <cuda_atom.h>
#include <allocate.h> #include <allocate.h>
@ -123,33 +124,31 @@ __global__ void kernel_final_integrate(MD_FLOAT dtforce, int Nlocal, Atom a) {
extern "C" { extern "C" {
void finalIntegrate_cuda(bool doReneighbour, Parameter *param, Atom *atom, Atom *c_atom) { void finalIntegrate_cuda(bool reneigh, Parameter *param, Atom *atom, Atom *c_atom) {
const int Nlocal = atom->Nlocal; const int Nlocal = atom->Nlocal;
const int num_threads_per_block = get_num_threads(); const int num_threads_per_block = get_num_threads();
const int num_blocks = ceil((float)Nlocal / (float)num_threads_per_block); const int num_blocks = ceil((float)Nlocal / (float)num_threads_per_block);
kernel_final_integrate <<< num_blocks, num_threads_per_block >>> (param->dtforce, Nlocal, *c_atom); kernel_final_integrate <<< num_blocks, num_threads_per_block >>> (param->dtforce, Nlocal, *c_atom);
cuda_assert("kernel_final_integrate", cudaPeekAtLastError());
cuda_assert("kernel_final_integrate", cudaDeviceSynchronize());
checkCUDAError( "PeekAtLastError FinalIntegrate", cudaPeekAtLastError() ); if(reneigh) {
checkCUDAError( "DeviceSync FinalIntegrate", cudaDeviceSynchronize() ); memcpyFromGPU(atom->vx, c_atom->vx, sizeof(MD_FLOAT) * atom->Nlocal * 3);
if(doReneighbour) {
checkCUDAError( "FinalIntegrate: velocity memcpy", cudaMemcpy(atom->vx, c_atom->vx, sizeof(MD_FLOAT) * atom->Nlocal * 3, cudaMemcpyDeviceToHost) );
} }
} }
void initialIntegrate_cuda(bool doReneighbour, Parameter *param, Atom *atom, Atom *c_atom) { void initialIntegrate_cuda(bool reneigh, Parameter *param, Atom *atom, Atom *c_atom) {
const int Nlocal = atom->Nlocal; const int Nlocal = atom->Nlocal;
const int num_threads_per_block = get_num_threads(); const int num_threads_per_block = get_num_threads();
const int num_blocks = ceil((float)Nlocal / (float)num_threads_per_block); const int num_blocks = ceil((float)Nlocal / (float)num_threads_per_block);
kernel_initial_integrate <<< num_blocks, num_threads_per_block >>> (param->dtforce, param->dt, Nlocal, *c_atom); kernel_initial_integrate <<< num_blocks, num_threads_per_block >>> (param->dtforce, param->dt, Nlocal, *c_atom);
cuda_assert("kernel_initial_integrate", cudaPeekAtLastError());
cuda_assert("kernel_initial_integrate", cudaDeviceSynchronize());
checkCUDAError( "PeekAtLastError InitialIntegrate", cudaPeekAtLastError() ); if(reneigh) {
checkCUDAError( "DeviceSync InitialIntegrate", cudaDeviceSynchronize() ); memcpyFromGPU(atom->vx, c_atom->vx, sizeof(MD_FLOAT) * atom->Nlocal * 3);
if(doReneighbour) {
checkCUDAError( "InitialIntegrate: velocity memcpy", cudaMemcpy(atom->vx, c_atom->vx, sizeof(MD_FLOAT) * atom->Nlocal * 3, cudaMemcpyDeviceToHost) );
} }
} }
@ -176,26 +175,20 @@ double computeForceLJFullNeigh_cuda(Parameter *param, Atom *atom, Neighbor *neig
// HINT: Run with cuda-memcheck ./MDBench-NVCC in case of error // HINT: Run with cuda-memcheck ./MDBench-NVCC in case of error
// checkCUDAError( "c_atom->fx memset", cudaMemset(c_atom->fx, 0, sizeof(MD_FLOAT) * Nlocal * 3) ); // checkCUDAError( "c_atom->fx memset", cudaMemset(c_atom->fx, 0, sizeof(MD_FLOAT) * Nlocal * 3) );
cudaProfilerStart(); cudaProfilerStart();
const int num_blocks = ceil((float)Nlocal / (float)num_threads_per_block); const int num_blocks = ceil((float)Nlocal / (float)num_threads_per_block);
double S = getTimeStamp(); double S = getTimeStamp();
LIKWID_MARKER_START("force"); LIKWID_MARKER_START("force");
calc_force <<< num_blocks, num_threads_per_block >>> (*c_atom, cutforcesq, sigma6, epsilon, Nlocal, neighbor->maxneighs, c_neighbor->neighbors, c_neighbor->numneigh); calc_force <<< num_blocks, num_threads_per_block >>> (*c_atom, cutforcesq, sigma6, epsilon, Nlocal, neighbor->maxneighs, c_neighbor->neighbors, c_neighbor->numneigh);
cuda_assert("calc_force", cudaPeekAtLastError());
checkCUDAError( "PeekAtLastError ComputeForce", cudaPeekAtLastError() ); cuda_assert("calc_force", cudaDeviceSynchronize());
checkCUDAError( "DeviceSync ComputeForce", cudaDeviceSynchronize() );
cudaProfilerStop(); cudaProfilerStop();
LIKWID_MARKER_STOP("force"); LIKWID_MARKER_STOP("force");
double E = getTimeStamp(); double E = getTimeStamp();
return E-S; return E-S;
} }

59
lammps/cuda/neighbor.cu
View File

@ -199,30 +199,25 @@ void binatoms_cuda(Atom *c_atom, Binning *c_binning, int *c_resize_needed, Neigh
while(resize > 0) { while(resize > 0) {
resize = 0; resize = 0;
checkCUDAError("binatoms_cuda c_binning->bincount memset", cudaMemset(c_binning->bincount, 0, c_binning->mbins * sizeof(int))); memsetGPU(c_binning->bincount, 0, c_binning->mbins * sizeof(int));
checkCUDAError("binatoms_cuda c_resize_needed memset", cudaMemset(c_resize_needed, 0, sizeof(int)) ); memsetGPU(c_resize_needed, 0, sizeof(int));
/*binatoms_kernel(Atom a, int* bincount, int* bins, int c_binning->atoms_per_bin, Neighbor_params np, int *resize_needed) */
binatoms_kernel<<<num_blocks, threads_per_block>>>(*c_atom, c_binning->bincount, c_binning->bins, c_binning->atoms_per_bin, *np, c_resize_needed); binatoms_kernel<<<num_blocks, threads_per_block>>>(*c_atom, c_binning->bincount, c_binning->bins, c_binning->atoms_per_bin, *np, c_resize_needed);
cuda_assert("binatoms", cudaPeekAtLastError());
cuda_assert("binatoms", cudaDeviceSynchronize());
checkCUDAError( "PeekAtLastError binatoms kernel", cudaPeekAtLastError() ); memcpyFromGPU(&resize, c_resize_needed, sizeof(int));
checkCUDAError( "DeviceSync binatoms kernel", cudaDeviceSynchronize() );
checkCUDAError("binatoms_cuda c_resize_needed memcpy back", cudaMemcpy(&resize, c_resize_needed, sizeof(int), cudaMemcpyDeviceToHost) );
if(resize) { if(resize) {
cudaFree(c_binning->bins);
c_binning->atoms_per_bin *= 2; c_binning->atoms_per_bin *= 2;
checkCUDAError("binatoms_cuda c_binning->bins resize malloc", cudaMalloc(&c_binning->bins, c_binning->mbins * c_binning->atoms_per_bin * sizeof(int)) ); c_binning->bins = (int *) reallocateGPU(c_binning->bins, c_binning->mbins * c_binning->atoms_per_bin * sizeof(int));
} }
} }
atoms_per_bin = c_binning->atoms_per_bin; atoms_per_bin = c_binning->atoms_per_bin;
const int sortBlocks = ceil((float) mbins / (float) threads_per_block); const int sortBlocks = ceil((float) mbins / (float) threads_per_block);
/*void sort_bin_contents_kernel(int* bincount, int* bins, int mbins, int atoms_per_bin)*/
sort_bin_contents_kernel<<<sortBlocks, threads_per_block>>>(c_binning->bincount, c_binning->bins, c_binning->mbins, c_binning->atoms_per_bin); sort_bin_contents_kernel<<<sortBlocks, threads_per_block>>>(c_binning->bincount, c_binning->bins, c_binning->mbins, c_binning->atoms_per_bin);
checkCUDAError( "PeekAtLastError sort_bin_contents kernel", cudaPeekAtLastError() ); cuda_assert("sort_bin", cudaPeekAtLastError());
checkCUDAError( "DeviceSync sort_bin_contents kernel", cudaDeviceSynchronize() ); cuda_assert("sort_bin", cudaDeviceSynchronize());
} }
void buildNeighbor_cuda(Atom *atom, Neighbor *neighbor, Atom *c_atom, Neighbor *c_neighbor) { void buildNeighbor_cuda(Atom *atom, Neighbor *neighbor, Atom *c_atom, Neighbor *c_neighbor) {
@ -231,18 +226,18 @@ void buildNeighbor_cuda(Atom *atom, Neighbor *neighbor, Atom *c_atom, Neighbor *
c_neighbor->maxneighs = neighbor->maxneighs; c_neighbor->maxneighs = neighbor->maxneighs;
cudaProfilerStart(); cudaProfilerStart();
/* upload stencil */
// TODO move all of this initialization into its own method // TODO move all of this initialization into its own method
if(c_stencil == NULL) { if(c_stencil == NULL) {
checkCUDAError( "buildNeighbor c_n_stencil malloc", cudaMalloc((void**)&c_stencil, nstencil * sizeof(int)) ); c_stencil = (int *) allocateGPU(nstencil * sizeof(int));
checkCUDAError( "buildNeighbor c_n_stencil memcpy", cudaMemcpy(c_stencil, stencil, nstencil * sizeof(int), cudaMemcpyHostToDevice )); memcpyToGPU(c_stencil, stencil, nstencil * sizeof(int));
} }
if(c_binning.mbins == 0) { if(c_binning.mbins == 0) {
c_binning.mbins = mbins; c_binning.mbins = mbins;
c_binning.atoms_per_bin = atoms_per_bin; c_binning.atoms_per_bin = atoms_per_bin;
checkCUDAError( "buildNeighbor c_binning->bincount malloc", cudaMalloc((void**)&(c_binning.bincount), c_binning.mbins * sizeof(int)) ); c_binning.bincount = (int *) allocateGPU(c_binning.mbins * sizeof(int));
checkCUDAError( "buidlNeighbor c_binning->bins malloc", cudaMalloc((void**)&(c_binning.bins), c_binning.mbins * c_binning.atoms_per_bin * sizeof(int)) ); c_binning.bins = (int *) allocateGPU(c_binning.mbins * c_binning.atoms_per_bin * sizeof(int));
} }
Neighbor_params np { Neighbor_params np {
@ -264,13 +259,13 @@ void buildNeighbor_cuda(Atom *atom, Neighbor *neighbor, Atom *c_atom, Neighbor *
}; };
if(c_resize_needed == NULL) { if(c_resize_needed == NULL) {
checkCUDAError("buildNeighbor c_resize_needed malloc", cudaMalloc((void**)&c_resize_needed, sizeof(int)) ); c_resize_needed = (int *) allocateGPU(sizeof(int));
} }
/* bin local & ghost atoms */ /* bin local & ghost atoms */
binatoms_cuda(c_atom, &c_binning, c_resize_needed, &np, num_threads_per_block); binatoms_cuda(c_atom, &c_binning, c_resize_needed, &np, num_threads_per_block);
if(c_new_maxneighs == NULL) { if(c_new_maxneighs == NULL) {
checkCUDAError("c_new_maxneighs malloc", cudaMalloc((void**)&c_new_maxneighs, sizeof(int) )); c_new_maxneighs = (int *) allocateGPU(sizeof(int));
} }
int resize = 1; int resize = 1;
@ -278,35 +273,26 @@ void buildNeighbor_cuda(Atom *atom, Neighbor *neighbor, Atom *c_atom, Neighbor *
/* extend c_neighbor arrays if necessary */ /* extend c_neighbor arrays if necessary */
if(nall > nmax) { if(nall > nmax) {
nmax = nall; nmax = nall;
if(c_neighbor->numneigh) cudaFree(c_neighbor->numneigh); c_neighbor->neighbors = (int *) reallocateGPU(c_neighbor->neighbors, nmax * c_neighbor->maxneighs * sizeof(int));
if(c_neighbor->neighbors) cudaFree(c_neighbor->neighbors); c_neighbor->numneigh = (int *) reallocateGPU(c_neighbor->numneigh, nmax * sizeof(int));
checkCUDAError( "buildNeighbor c_numneigh malloc", cudaMalloc((void**)&(c_neighbor->numneigh), nmax * sizeof(int)) );
checkCUDAError( "buildNeighbor c_neighbors malloc", cudaMalloc((void**)&(c_neighbor->neighbors), nmax * c_neighbor->maxneighs * sizeof(int)) );
} }
/* loop over each atom, storing neighbors */ /* loop over each atom, storing neighbors */
while(resize) { while(resize) {
resize = 0; resize = 0;
memsetGPU(c_new_maxneighs, 0, sizeof(int));
checkCUDAError("c_new_maxneighs memset", cudaMemset(c_new_maxneighs, 0, sizeof(int) ));
// TODO call compute_neigborhood kernel here
const int num_blocks = ceil((float)atom->Nlocal / (float)num_threads_per_block); const int num_blocks = ceil((float)atom->Nlocal / (float)num_threads_per_block);
/*compute_neighborhood(Atom a, Neighbor neigh, Neighbor_params np, int nstencil, int* stencil,
int* bins, int atoms_per_bin, int *bincount, int *new_maxneighs)
* */
compute_neighborhood<<<num_blocks, num_threads_per_block>>>(*c_atom, *c_neighbor, compute_neighborhood<<<num_blocks, num_threads_per_block>>>(*c_atom, *c_neighbor,
np, nstencil, c_stencil, np, nstencil, c_stencil,
c_binning.bins, c_binning.atoms_per_bin, c_binning.bincount, c_binning.bins, c_binning.atoms_per_bin, c_binning.bincount,
c_new_maxneighs, c_new_maxneighs,
cutneighsq); cutneighsq);
checkCUDAError( "PeekAtLastError ComputeNeighbor", cudaPeekAtLastError() ); cuda_assert("compute_neighborhood", cudaPeekAtLastError());
checkCUDAError( "DeviceSync ComputeNeighbor", cudaDeviceSynchronize() ); cuda_assert("compute_neighborhood", cudaDeviceSynchronize());
// TODO copy the value of c_new_maxneighs back to host and check if it has been modified
int new_maxneighs; int new_maxneighs;
checkCUDAError("c_new_maxneighs memcpy back", cudaMemcpy(&new_maxneighs, c_new_maxneighs, sizeof(int), cudaMemcpyDeviceToHost)); memcpyFromGPU(&new_maxneighs, c_new_maxneighs, sizeof(int));
if (new_maxneighs > c_neighbor->maxneighs){ if (new_maxneighs > c_neighbor->maxneighs){
resize = 1; resize = 1;
} }
@ -315,8 +301,7 @@ void buildNeighbor_cuda(Atom *atom, Neighbor *neighbor, Atom *c_atom, Neighbor *
printf("RESIZE %d\n", c_neighbor->maxneighs); printf("RESIZE %d\n", c_neighbor->maxneighs);
c_neighbor->maxneighs = new_maxneighs * 1.2; c_neighbor->maxneighs = new_maxneighs * 1.2;
printf("NEW SIZE %d\n", c_neighbor->maxneighs); printf("NEW SIZE %d\n", c_neighbor->maxneighs);
cudaFree(c_neighbor->neighbors); c_neighbor->neighbors = (int *) reallocateGPU(c_neighbor->neighbors, c_atom->Nmax * c_neighbor->maxneighs * sizeof(int));
checkCUDAError("c_neighbor->neighbors resize malloc", cudaMalloc((void**)(&c_neighbor->neighbors), c_atom->Nmax * c_neighbor->maxneighs * sizeof(int)));
} }
} }

49
lammps/cuda/pbc.cu
View File

@ -94,30 +94,25 @@ void updatePbc_cuda(Atom *atom, Atom *c_atom, Parameter *param, bool doReneighbo
if (atom->Nmax > c_atom->Nmax){ // the number of ghost atoms has increased -> more space is needed if (atom->Nmax > c_atom->Nmax){ // the number of ghost atoms has increased -> more space is needed
c_atom->Nmax = atom->Nmax; c_atom->Nmax = atom->Nmax;
if(c_atom->x != NULL){ cudaFree(c_atom->x); } c_atom->x = (MD_FLOAT *) reallocateGPU(c_atom->x, sizeof(MD_FLOAT) * atom->Nmax * 3);
if(c_atom->type != NULL){ cudaFree(c_atom->type); } c_atom->type = (int *) reallocateGPU(c_atom->type, sizeof(int) * atom->Nmax);
checkCUDAError( "updatePbc c_atom->x malloc", cudaMalloc((void**)&(c_atom->x), sizeof(MD_FLOAT) * atom->Nmax * 3) );
checkCUDAError( "updatePbc c_atom->type malloc", cudaMalloc((void**)&(c_atom->type), sizeof(int) * atom->Nmax) );
} }
// TODO if the sort is reactivated the atom->vx needs to be copied to GPU as well
checkCUDAError( "updatePbc c_atom->x memcpy", cudaMemcpy(c_atom->x, atom->x, sizeof(MD_FLOAT) * atom->Nmax * 3, cudaMemcpyHostToDevice) ); memcpyToGPU(c_atom->x, atom->x, sizeof(MD_FLOAT) * atom->Nmax * 3);
checkCUDAError( "updatePbc c_atom->type memcpy", cudaMemcpy(c_atom->type, atom->type, sizeof(int) * atom->Nmax, cudaMemcpyHostToDevice) ); memcpyToGPU(c_atom->type, atom->type, sizeof(int) * atom->Nmax);
if(c_NmaxGhost < NmaxGhost) { if(c_NmaxGhost < NmaxGhost) {
c_NmaxGhost = NmaxGhost; c_NmaxGhost = NmaxGhost;
if(c_PBCx != NULL){ cudaFree(c_PBCx); } c_PBCx = (int *) reallocateGPU(c_PBCx, NmaxGhost * sizeof(int));
if(c_PBCy != NULL){ cudaFree(c_PBCy); } c_PBCy = (int *) reallocateGPU(c_PBCy, NmaxGhost * sizeof(int));
if(c_PBCz != NULL){ cudaFree(c_PBCz); } c_PBCz = (int *) reallocateGPU(c_PBCz, NmaxGhost * sizeof(int));
if(c_atom->border_map != NULL){ cudaFree(c_atom->border_map); } c_atom->border_map = (int *) reallocateGPU(c_atom->border_map, NmaxGhost * sizeof(int));
checkCUDAError( "updatePbc c_PBCx malloc", cudaMalloc((void**)&c_PBCx, NmaxGhost * sizeof(int)) );
checkCUDAError( "updatePbc c_PBCy malloc", cudaMalloc((void**)&c_PBCy, NmaxGhost * sizeof(int)) );
checkCUDAError( "updatePbc c_PBCz malloc", cudaMalloc((void**)&c_PBCz, NmaxGhost * sizeof(int)) );
checkCUDAError( "updatePbc c_atom->border_map malloc", cudaMalloc((void**)&(c_atom->border_map), NmaxGhost * sizeof(int)) );
} }
checkCUDAError( "updatePbc c_PBCx memcpy", cudaMemcpy(c_PBCx, PBCx, NmaxGhost * sizeof(int), cudaMemcpyHostToDevice) );
checkCUDAError( "updatePbc c_PBCy memcpy", cudaMemcpy(c_PBCy, PBCy, NmaxGhost * sizeof(int), cudaMemcpyHostToDevice) ); memcpyToGPU(c_PBCx, PBCx, NmaxGhost * sizeof(int));
checkCUDAError( "updatePbc c_PBCz memcpy", cudaMemcpy(c_PBCz, PBCz, NmaxGhost * sizeof(int), cudaMemcpyHostToDevice) ); memcpyToGPU(c_PBCy, PBCy, NmaxGhost * sizeof(int));
checkCUDAError( "updatePbc c_atom->border_map memcpy", cudaMemcpy(c_atom->border_map, atom->border_map, NmaxGhost * sizeof(int), cudaMemcpyHostToDevice) ); memcpyToGPU(c_PBCz, PBCz, NmaxGhost * sizeof(int));
memcpyToGPU(c_atom->border_map, atom->border_map, NmaxGhost * sizeof(int));
} }
MD_FLOAT xprd = param->xprd; MD_FLOAT xprd = param->xprd;
@ -125,13 +120,9 @@ void updatePbc_cuda(Atom *atom, Atom *c_atom, Parameter *param, bool doReneighbo
MD_FLOAT zprd = param->zprd; MD_FLOAT zprd = param->zprd;
const int num_blocks = ceil((float)atom->Nghost / (float)num_threads_per_block); const int num_blocks = ceil((float)atom->Nghost / (float)num_threads_per_block);
/*__global__ void computePbcUpdate(Atom a, int* PBCx, int* PBCy, int* PBCz,
* MD_FLOAT xprd, MD_FLOAT yprd, MD_FLOAT zprd)
* */
computePbcUpdate<<<num_blocks, num_threads_per_block>>>(*c_atom, c_PBCx, c_PBCy, c_PBCz, xprd, yprd, zprd); computePbcUpdate<<<num_blocks, num_threads_per_block>>>(*c_atom, c_PBCx, c_PBCy, c_PBCz, xprd, yprd, zprd);
checkCUDAError( "PeekAtLastError UpdatePbc", cudaPeekAtLastError() ); cuda_assert("computePbcUpdate", cudaPeekAtLastError());
checkCUDAError( "DeviceSync UpdatePbc", cudaDeviceSynchronize() ); cuda_assert("computePbcUpdate", cudaDeviceSynchronize());
} }
void updateAtomsPbc_cuda(Atom* atom, Atom *c_atom, Parameter *param) { void updateAtomsPbc_cuda(Atom* atom, Atom *c_atom, Parameter *param) {
@ -141,10 +132,8 @@ void updateAtomsPbc_cuda(Atom* atom, Atom *c_atom, Parameter *param){
MD_FLOAT zprd = param->zprd; MD_FLOAT zprd = param->zprd;
const int num_blocks = ceil((float)atom->Nlocal / (float)num_threads_per_block); const int num_blocks = ceil((float)atom->Nlocal / (float)num_threads_per_block);
/*void computeAtomsPbcUpdate(Atom a, MD_FLOAT xprd, MD_FLOAT yprd, MD_FLOAT zprd)*/
computeAtomsPbcUpdate<<<num_blocks, num_threads_per_block>>>(*c_atom, xprd, yprd, zprd); computeAtomsPbcUpdate<<<num_blocks, num_threads_per_block>>>(*c_atom, xprd, yprd, zprd);
cuda_assert("computeAtomsPbcUpdate", cudaPeekAtLastError());
checkCUDAError( "PeekAtLastError UpdateAtomsPbc", cudaPeekAtLastError() ); cuda_assert("computeAtomsPbcUpdate", cudaDeviceSynchronize());
checkCUDAError( "DeviceSync UpdateAtomsPbc", cudaDeviceSynchronize() ); memcpyFromGPU(atom->x, c_atom->x, sizeof(MD_FLOAT) * atom->Nlocal * 3);
checkCUDAError( "updateAtomsPbc position memcpy back", cudaMemcpy(atom->x, c_atom->x, sizeof(MD_FLOAT) * atom->Nlocal * 3, cudaMemcpyDeviceToHost) );
} }

7
lammps/includes/cuda_atom.h
View File

@ -6,5 +6,10 @@
#ifndef __CUDA_ATOM_H_ #ifndef __CUDA_ATOM_H_
#define __CUDA_ATOM_H_ #define __CUDA_ATOM_H_
extern void initCuda(Atom*, Neighbor*, Atom*, Neighbor*); extern void initCuda(Atom*, Neighbor*, Atom*, Neighbor*);
extern void checkCUDAError(const char *msg, cudaError_t err); extern void cuda_assert(const char *msg, cudaError_t err);
extern void *allocateGPU(size_t bytesize);
extern void *reallocateGPU(void *ptr, size_t new_bytesize);
extern void memcpyToGPU(void *d_ptr, void *h_ptr, size_t bytesize);
extern void memcpyFromGPU(void *h_ptr, void *d_ptr, size_t bytesize);
extern void memsetGPU(void *d_ptr, int value, size_t bytesize);
#endif #endif

3
lammps/main.c
View File

@ -261,9 +261,8 @@ int main(int argc, char** argv) {
if(!((n + 1) % param.nstat) && (n+1) < param.ntimes) { if(!((n + 1) % param.nstat) && (n+1) < param.ntimes) {
#ifdef CUDA_TARGET #ifdef CUDA_TARGET
checkCUDAError("computeThermo atom->x memcpy back", cudaMemcpy(atom.x, c_atom.x, atom.Nmax * sizeof(MD_FLOAT) * 3, cudaMemcpyDeviceToHost)); memcpyFromGPU(atom.x, c_atom.x, atom.Nmax * sizeof(MD_FLOAT) * 3);
#endif #endif
computeThermo(n + 1, &param, &atom); computeThermo(n + 1, &param, &atom);
} }