First run but segfault at the moment after a few seconds

This commit is contained in:
Maximilian Gaul 2021-11-11 15:23:46 +01:00
parent 280f595b7f
commit 1a54314c8b

View File

@ -20,6 +20,7 @@
* with MD-Bench. If not, see <https://www.gnu.org/licenses/>.
* =======================================================================================
*/
#include <math.h>
#include <stdio.h>
#include <stdlib.h>
#include <stddef.h>
@ -37,7 +38,7 @@ extern "C" {
// cuda kernel
__global__ void calc_force(
Atom *atom,
Atom a,
MD_FLOAT xtmp, MD_FLOAT ytmp, MD_FLOAT ztmp,
MD_FLOAT *fix, MD_FLOAT *fiy, MD_FLOAT *fiz,
int i, int numneighs, int *neighs) {
@ -48,6 +49,8 @@ __global__ void calc_force(
return;
}
Atom *atom = &a;
int j = neighs[k];
MD_FLOAT delx = xtmp - atom_x(j);
MD_FLOAT dely = ytmp - atom_y(j);
@ -109,30 +112,29 @@ double computeForce(
const int type_i = atom->type[i];
#endif
Atom *c_atom;
cudaMalloc((void**)&c_atom, sizeof(Atom));
cudaMemcpy(c_atom, atom, sizeof(Atom), cudaMemcpyHostToDevice);
Atom c_atom;
memcpy(&c_atom, atom, sizeof(Atom));
cudaMalloc((void**)&c_atom->x, sizeof(MD_FLOAT) * atom->Nmax * 3);
cudaMemcpy(c_atom->x, atom->x, sizeof(MD_FLOAT) * atom->Nmax * 3, cudaMemcpyHostToDevice);
cudaMalloc((void**)&(&c_atom)->x, sizeof(MD_FLOAT) * atom->Nmax * 3);
cudaMemcpy(c_atom.x, atom->x, sizeof(MD_FLOAT) * atom->Nmax * 3, cudaMemcpyHostToDevice);
cudaMalloc((void**)&c_atom->y, sizeof(MD_FLOAT) * atom->Nmax * 3);
cudaMemcpy(c_atom->y, atom->y, sizeof(MD_FLOAT) * atom->Nmax * 3, cudaMemcpyHostToDevice);
cudaMalloc((void**)&(&c_atom)->y, sizeof(MD_FLOAT) * atom->Nmax * 3);
cudaMemcpy(c_atom.y, atom->y, sizeof(MD_FLOAT) * atom->Nmax * 3, cudaMemcpyHostToDevice);
cudaMalloc((void**)&c_atom->z, sizeof(MD_FLOAT) * atom->Nmax * 3);
cudaMemcpy(c_atom->z, atom->z, sizeof(MD_FLOAT) * atom->Nmax * 3, cudaMemcpyHostToDevice);
cudaMalloc((void**)&(&c_atom)->z, sizeof(MD_FLOAT) * atom->Nmax * 3);
cudaMemcpy(c_atom.z, atom->z, sizeof(MD_FLOAT) * atom->Nmax * 3, cudaMemcpyHostToDevice);
cudaMalloc((void**)&c_atom->type, sizeof(int) * atom->Nmax);
cudaMemcpy(c_atom->type, atom->type, sizeof(int) * atom->Nmax, cudaMemcpyHostToDevice);
cudaMalloc((void**)&(&c_atom)->type, sizeof(int) * atom->Nmax);
cudaMemcpy(c_atom.type, atom->type, sizeof(int) * atom->Nmax, cudaMemcpyHostToDevice);
cudaMalloc((void**)&c_atom->epsilon, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes);
cudaMemcpy(c_atom->epsilon, atom->epsilon, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes, cudaMemcpyHostToDevice);
cudaMalloc((void**)&(&c_atom)->epsilon, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes);
cudaMemcpy(c_atom.epsilon, atom->epsilon, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes, cudaMemcpyHostToDevice);
cudaMalloc((void**)&c_atom->sigma6, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes);
cudaMemcpy(c_atom->sigma6, atom->sigma6, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes, cudaMemcpyHostToDevice);
cudaMalloc((void**)&(&c_atom)->sigma6, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes);
cudaMemcpy(c_atom.sigma6, atom->sigma6, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes, cudaMemcpyHostToDevice);
cudaMalloc((void**)&c_atom->cutforcesq, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes);
cudaMemcpy(c_atom->cutforcesq, atom->cutforcesq, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes, cudaMemcpyHostToDevice);
cudaMalloc((void**)&(&c_atom)->cutforcesq, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes);
cudaMemcpy(c_atom.cutforcesq, atom->cutforcesq, sizeof(MD_FLOAT) * atom->ntypes * atom->ntypes, cudaMemcpyHostToDevice);
int *c_neighs;
cudaMalloc((void**)&c_neighs, sizeof(int) * numneighs);
@ -144,8 +146,8 @@ double computeForce(
cudaMalloc((void**)&c_fiz, sizeof(MD_FLOAT) * numneighs);
const int num_blocks = 64;
const int num_threads_per_block = numneighs / num_blocks;
printf("numneighs: %d => num-blocks: %d, num_threads => %d\r\n", numneighs, num_blocks, num_threads_per_block);
const int num_threads_per_block = ceil((float)numneighs / (float)num_blocks);
// printf("numneighs: %d => num-blocks: %d, num_threads_per_block => %d\r\n", numneighs, num_blocks, num_threads_per_block);
// launch cuda kernel
calc_force <<< num_blocks, num_threads_per_block >>> (c_atom, xtmp, ytmp, ztmp, c_fix, c_fiy, c_fiz, i, numneighs, c_neighs);
@ -156,9 +158,9 @@ double computeForce(
d_fix = (MD_FLOAT*)malloc(sizeof(MD_FLOAT) * numneighs);
d_fiy = (MD_FLOAT*)malloc(sizeof(MD_FLOAT) * numneighs);
d_fiz = (MD_FLOAT*)malloc(sizeof(MD_FLOAT) * numneighs);
cudaMemcpy((void**)d_fix, c_fix, sizeof(MD_FLOAT) * numneighs, cudaMemcpyDeviceToHost);
cudaMemcpy((void**)d_fiy, c_fiy, sizeof(MD_FLOAT) * numneighs, cudaMemcpyDeviceToHost);
cudaMemcpy((void**)d_fiz, c_fiz, sizeof(MD_FLOAT) * numneighs, cudaMemcpyDeviceToHost);
cudaMemcpy((void**)&d_fix, c_fix, sizeof(MD_FLOAT) * numneighs, cudaMemcpyDeviceToHost);
cudaMemcpy((void**)&d_fiy, c_fiy, sizeof(MD_FLOAT) * numneighs, cudaMemcpyDeviceToHost);
cudaMemcpy((void**)&d_fiz, c_fiz, sizeof(MD_FLOAT) * numneighs, cudaMemcpyDeviceToHost);
for(int k = 0; k < numneighs; k++) {
fx[i] += d_fix[k];
@ -166,14 +168,16 @@ double computeForce(
fz[i] += d_fiz[k];
}
cudaFree(c_fix); cudaFree(c_fiy); cudaFree(c_fiz);
cudaFree(c_atom); cudaFree(c_neighs);
cudaFree(c_fix); cudaFree(c_fiy); cudaFree(c_fiz); cudaFree(c_neighs);
cudaFree(c_atom.x); cudaFree(c_atom.y); cudaFree(c_atom.z); cudaFree(c_atom.type);
cudaFree(c_atom.epsilon); cudaFree(c_atom.sigma6); cudaFree(c_atom.cutforcesq);
free(d_fix); free(d_fiy); free(d_fiz);
}
LIKWID_MARKER_STOP("force");
double E = getTimeStamp();
return E-S;
return 0;
}
}