Skip to content

Feature: Implement cal_force_op for sincos parallel #347

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

Closed
jieli-matrix wants to merge 4 commits into from
Closed

Feature: Implement cal_force_op for sincos parallel #347

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
9a42117
Feature: implement k continuity initialization strategy & kernel
jieli-matrix Feb 8, 2025
c605cae
implement gpu op for sincos loops
jieli-matrix May 26, 2025
fbfc91a
add cpu kernel for cal_force_loc & cal_force_ew
jieli-matrix May 27, 2025
8a6339c
fix sincos op for gpu&cpu
jieli-matrix May 30, 2025
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
9 changes: 7 additions & 2 deletions source/module_base/module_device/device.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@

#include <base/macros/macros.h>
#include <cstring>

#include <iostream>
#ifdef __MPI
#include "mpi.h"
#endif
Expand Down Expand Up @@ -166,6 +166,11 @@ int device_count = -1;
cudaGetDeviceCount(&device_count);
#elif defined(__ROCM)
hipGetDeviceCount(&device_count);
/***auto start_time = std::chrono::high_resolution_clock::now();
std::cout << "Starting hipGetDeviceCount.." << std::endl;
auto end_time = std::chrono::high_resolution_clock::now();
auto duration = std::chrono::duration_cast<std::chrono::duration<double>>(end_time - start_time);
std::cout << "hipGetDeviceCount took " << duration.count() << "seconds" << std::endl;***/
#endif
if (device_count <= 0)
{
Expand Down Expand Up @@ -711,4 +716,4 @@ void record_device_memory<base_device::DEVICE_GPU>(
#endif

} // end of namespace information
} // end of namespace base_device
} // end of namespace base_device
16 changes: 12 additions & 4 deletions source/module_esolver/esolver.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -107,13 +107,21 @@ std::string determine_type()
}
if (GlobalV::MY_RANK == 0)
{
std::cout << " RUNNING WITH DEVICE : " << device_info << " / "
/***auto start_time = std::chrono::high_resolution_clock::now();
std::cout << "Starting hipGetDeviceInfo..." << std::endl;***/
std::cout << " RUNNING WITH DEVICE : " << device_info << " / "
<< base_device::information::get_device_info(PARAM.inp.device) << std::endl;
/***auto end_time = std::chrono::high_resolution_clock::now();
auto duration = std::chrono::duration_cast<std::chrono::duration<double>>(end_time - start_time);
std::cout << "hipGetDeviceInfo took " << duration.count() << " seconds" << std::endl;***/
}

/*** auto start_time = std::chrono::high_resolution_clock::now();
std::cout << "Starting hipGetDeviceInfo..." << std::endl;***/
GlobalV::ofs_running << "\n RUNNING WITH DEVICE : " << device_info << " / "
<< base_device::information::get_device_info(PARAM.inp.device) << std::endl;

<< base_device::information::get_device_info(PARAM.inp.device) << std::endl;
/***auto end_time = std::chrono::high_resolution_clock::now();
auto duration = std::chrono::duration_cast<std::chrono::duration<double>>(end_time - start_time);
std::cout << "hipGetDeviceInfo took " << duration.count() << " seconds" << std::endl;***/
return esolver_type;
}

Expand Down
3 changes: 2 additions & 1 deletion source/module_esolver/esolver_ks_pw.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -556,7 +556,8 @@ void ESolver_KS_PW<T, Device>::hamilt2density_single(UnitCell& ucell,
hsolver::DiagoIterAssist<T, Device>::SCF_ITER,
hsolver::DiagoIterAssist<T, Device>::PW_DIAG_NMAX,
hsolver::DiagoIterAssist<T, Device>::PW_DIAG_THR,
hsolver::DiagoIterAssist<T, Device>::need_subspace);
hsolver::DiagoIterAssist<T, Device>::need_subspace,
PARAM.inp.use_k_continuity);

hsolver_pw_obj.solve(this->p_hamilt,
this->kspw_psi[0],
Expand Down
Loading
Loading