Skip to content

Fix SSF issues on large grids #197

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
vmitq wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Fix SSF issues on large grids #197

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions src/xc_integrator/local_work_driver/device/cuda/cuda_aos_scheme1_weights.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@ namespace GauXC {
void cuda_aos_scheme1_weights_wrapper( int32_t npts, int32_t natoms,
const double* points_x, const double* points_y, const double* points_z,
const double* RAB, int32_t ldRAB, const double* coords,
double* dist, int32_t lddist, const int32_t* iparent,
double* dist, size_t lddist, const int32_t* iparent,
const double* dist_nearest, double* weights, cudaStream_t stream ) {

constexpr auto weight_unroll =
Expand Down Expand Up @@ -64,7 +64,7 @@ void cuda_aos_scheme1_weight_1st_deriv_wrapper(
int32_t npts, int32_t natoms,
const double* points_x, const double* points_y, const double* points_z,
const double* RAB, int32_t ldRAB, const double* coords,
double* dist, int32_t lddist, const int32_t* iparent,
double* dist, size_t lddist, const int32_t* iparent,
const double* dist_nearest, const double* w_times_f,
double* exc_grad_w, cudaStream_t stream ){

Expand Down
4 changes: 2 additions & 2 deletions src/xc_integrator/local_work_driver/device/cuda/cuda_aos_scheme1_weights.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,14 +14,14 @@ namespace GauXC {
void cuda_aos_scheme1_weights_wrapper( int32_t npts, int32_t natoms,
const double* points_x, const double* points_y, const double* points_z,
const double* RAB, int32_t ldRAB, const double* coords,
double* dist, int32_t lddist, const int32_t* iparent,
double* dist, size_t lddist, const int32_t* iparent,
const double* dist_nearest, double* weights, cudaStream_t stream );

void cuda_aos_scheme1_weight_1st_deriv_wrapper(
int32_t npts, int32_t natoms,
const double* points_x, const double* points_y, const double* points_z,
const double* RAB, int32_t ldRAB, const double* coords,
double* dist, int32_t lddist, const int32_t* iparent,
double* dist, size_t lddist, const int32_t* iparent,
const double* dist_nearest, const double* w_times_f,
double* exc_grad_w, cudaStream_t stream );

Expand Down
8 changes: 4 additions & 4 deletions src/xc_integrator/local_work_driver/device/cuda/kernels/cuda_ssf_1d.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,7 +28,7 @@ __global__ void modify_weights_ssf_kernel_1d(
int32_t ldRAB,
const double* coords,
const double* dist_scratch,
int32_t lddist,
size_t lddist,
const int32_t* iparent_device,
const double* dist_nearest_device,
double* weights_device
Expand Down Expand Up @@ -131,7 +131,7 @@ __global__ void modify_weights_ssf_kernel_1d(
}

void partition_weights_ssf_1d( int32_t npts, int32_t natoms, const double* RAB,
int32_t ldRAB, const double* coords, const double* dist, int32_t lddist,
int32_t ldRAB, const double* coords, const double* dist, size_t lddist,
const int32_t* iparent, const double* dist_nearest, double* weights,
cudaStream_t stream ) {

Expand All @@ -153,7 +153,7 @@ __global__ void eval_weight_1st_deriv_contracted_ssf_kernel_1d(
const double* points_y,
const double* points_z,
const double* dist_scratch,
int32_t lddist,
size_t lddist,
const int32_t* iparent_device,
const double* dist_nearest_device,
const double* __restrict__ w_times_f_device,
Expand Down Expand Up @@ -352,7 +352,7 @@ __global__ void eval_weight_1st_deriv_contracted_ssf_kernel_1d(
void eval_weight_1st_deriv_contracted_ssf_1d( int32_t npts, int32_t natoms, const double* RAB,
int32_t ldRAB, const double* coords,
const double* points_x, const double* points_y, const double* points_z,
const double* dist, int32_t lddist,
const double* dist, size_t lddist,
const int32_t* iparent, const double* dist_nearest, const double* w_times_f,
double* exc_grad_w, cudaStream_t stream){

Expand Down
4 changes: 2 additions & 2 deletions src/xc_integrator/local_work_driver/device/cuda/kernels/cuda_ssf_1d.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,14 +13,14 @@
namespace GauXC {

void partition_weights_ssf_1d( int32_t npts, int32_t natoms, const double* RAB,
int32_t ldRAB, const double* coords, const double* dist, int32_t lddist,
int32_t ldRAB, const double* coords, const double* dist, size_t lddist,
const int32_t* iparent, const double* dist_nearest, double* weights,
cudaStream_t stream);

void eval_weight_1st_deriv_contracted_ssf_1d( int32_t npts, int32_t natoms, const double* RAB,
int32_t ldRAB, const double* coords,
const double* points_x, const double* points_y, const double* points_z,
const double* dist, int32_t lddist,
const double* dist, size_t lddist,
const int32_t* iparent, const double* dist_nearest, const double* w_times_f,
double* exc_grad_w, cudaStream_t stream);

Expand Down
Loading