Fix: correct nvc++ compiler compatibility issues

source/source_base/kernels/math_kernel_op.cpp

@@ -81,7 +81,7 @@ struct matrixTranspose_op<T, base_device::DEVICE_CPU>
         T* temp = nullptr;
         base_device::memory::resize_memory_op<T, base_device::DEVICE_CPU>()(temp, row * col, "MTransOp");
 #ifdef _OPENMP
-#pragma omp parallel for collapse(2) schedule(static, 8192 / sizeof(T))
+#pragma omp parallel for collapse(2) schedule(static)
 #endif
         for (int j = 0; j < col; j++)
         {
@@ -91,7 +91,7 @@ struct matrixTranspose_op<T, base_device::DEVICE_CPU>
             }
         }
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 8192 / sizeof(T))
+#pragma omp parallel for schedule(static)
 #endif
         for (int i = 0; i < row * col; i++)
         {
@@ -107,7 +107,7 @@ struct matrixCopy<T, base_device::DEVICE_CPU>
     void operator()(const int& n1, const int& n2, const T* A, const int& LDA, T* B, const int& LDB)
     {
 #ifdef _OPENMP
-#pragma omp parallel for collapse(2) schedule(static, 8192 / sizeof(T))
+#pragma omp parallel for collapse(2) schedule(static)
 #endif
         for (int i = 0; i < n1; i++)
         {
@@ -130,7 +130,7 @@ struct matrix_mul_vector_op<T, base_device::DEVICE_CPU> {
                   T *c,
                   const int &ldc){
 #ifdef _OPENMP
-#pragma omp parallel for collapse(2) schedule(static, 8192 / sizeof(T))
+#pragma omp parallel for collapse(2) schedule(static)
 #endif
         for (int j = 0; j < n; j++){
             for (int i = 0; i < m; i++){

source/source_base/kernels/math_kernel_op_vec.cpp

@@ -24,7 +24,7 @@ struct vector_mul_real_op<T, base_device::DEVICE_CPU>
     void operator()(const int dim, T* result, const T* vector, const Real constant)
     {
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(Real))
+#pragma omp parallel for schedule(static)
 #endif
         for (int i = 0; i < dim; i++)
         {
@@ -42,7 +42,7 @@ struct vector_mul_vector_op<T, base_device::DEVICE_CPU>
         if (add)
         {
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(Real))
+#pragma omp parallel for schedule(static)
 #endif
             for (int i = 0; i < dim; i++)
             {
@@ -52,7 +52,7 @@ struct vector_mul_vector_op<T, base_device::DEVICE_CPU>
         else
         {
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(Real))
+#pragma omp parallel for schedule(static)
 #endif
             for (int i = 0; i < dim; i++)
             {
@@ -69,7 +69,7 @@ struct vector_div_constant_op<T, base_device::DEVICE_CPU>
     void operator()(const int& dim, T* result, const T* vector, const Real constant)
     {
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(Real))
+#pragma omp parallel for schedule(static)
 #endif
         for (int i = 0; i < dim; i++)
         {
@@ -85,7 +85,7 @@ struct vector_div_vector_op<T, base_device::DEVICE_CPU>
     void operator()(const int& dim, T* result, const T* vector1, const Real* vector2)
     {
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(Real))
+#pragma omp parallel for schedule(static)
 #endif
         for (int i = 0; i < dim; i++)
         {
@@ -121,7 +121,7 @@ struct vector_add_vector_op<T, base_device::DEVICE_CPU>
                     const Real constant2)
     {
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 8192 / sizeof(T))
+#pragma omp parallel for schedule(static)
 #endif
         for (int i = 0; i < dim; i++)
         {

source/source_base/module_device/memory_op.cpp

@@ -108,7 +108,7 @@ struct cast_memory_op<FPTYPE_out, FPTYPE_in, base_device::DEVICE_CPU, base_devic
                     const size_t size)
     {
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(FPTYPE_out))
+#pragma omp parallel for schedule(static)
 #endif
         for (int ii = 0; ii < size; ii++)
         {

source/source_base/module_external/blas_connector_vector.cpp

@@ -368,7 +368,7 @@ void vector_mul_vector(const int& dim, T* result, const T* vector1, const T* vec
 	using Real = typename GetTypeReal<T>::type;
 	if (device_type == base_device::AbacusDevice_t::CpuDevice) {
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(Real))
+#pragma omp parallel for schedule(static)
 #endif
         for (int i = 0; i < dim; i++)
         {
@@ -391,7 +391,7 @@ void vector_div_vector(const int& dim, T* result, const T* vector1, const T* vec
 	using Real = typename GetTypeReal<T>::type;
 	if (device_type == base_device::AbacusDevice_t::CpuDevice) {
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(Real))
+#pragma omp parallel for schedule(static)
 #endif
 		for (int i = 0; i < dim; i++)
         {
@@ -412,7 +412,7 @@ void vector_add_vector(const int& dim, float *result, const float *vector1, cons
 {
 	if (device_type == base_device::CpuDevice){
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 8192 / sizeof(float))
+#pragma omp parallel for schedule(static)
 #endif
         for (int i = 0; i < dim; i++)
         {
@@ -433,7 +433,7 @@ void vector_add_vector(const int& dim, double *result, const double *vector1, co
 {
 	if (device_type == base_device::CpuDevice){
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 8192 / sizeof(double))
+#pragma omp parallel for schedule(static)
 #endif
         for (int i = 0; i < dim; i++)
         {
@@ -454,7 +454,7 @@ void vector_add_vector(const int& dim, std::complex<float> *result, const std::c
 {
 	if (device_type == base_device::CpuDevice){
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 8192 / sizeof(std::complex<float>))
+#pragma omp parallel for schedule(static)
 #endif
         for (int i = 0; i < dim; i++)
         {
@@ -475,7 +475,7 @@ void vector_add_vector(const int& dim, std::complex<double> *result, const std::
 {
 	if (device_type == base_device::CpuDevice){
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 8192 / sizeof(std::complex<double>))
+#pragma omp parallel for schedule(static)
 #endif
         for (int i = 0; i < dim; i++)
         {

source/source_base/module_mixing/broyden_mixing.cpp

@@ -32,7 +32,7 @@ void Broyden_Mixing::tem_push_data(Mixing_Data& mdata,
     const size_t length = mdata.length;
     std::vector<FPTYPE> F_tmp(length);
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(FPTYPE))
+#pragma omp parallel for schedule(static)
 #endif
     for (int i = 0; i < length; ++i)
     {
@@ -72,7 +72,7 @@ void Broyden_Mixing::tem_push_data(Mixing_Data& mdata,
         dF = malloc(sizeof(FPTYPE) * length * mixing_ndim);
         FP_dF = static_cast<FPTYPE*>(dF);
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(FPTYPE))
+#pragma omp parallel for schedule(static)
 #endif
         for (int i = 0; i < length; ++i)
         {
@@ -84,7 +84,7 @@ void Broyden_Mixing::tem_push_data(Mixing_Data& mdata,
         this->ndim_cal_dF = std::min(this->ndim_cal_dF + 1, this->mixing_ndim);
         start_dF = (this->start_dF + 1) % this->mixing_ndim;
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(FPTYPE))
+#pragma omp parallel for schedule(static)
 #endif
         for (int i = 0; i < length; ++i)
         {
@@ -192,7 +192,7 @@ void Broyden_Mixing::tem_cal_coef(const Mixing_Data& mdata, std::function<double
 
     FPTYPE* dFnext = FP_dF + dFindex_move(1) * length;
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(FPTYPE))
+#pragma omp parallel for schedule(static)
 #endif
     for (int i = 0; i < length; ++i)
     {

source/source_base/module_mixing/mixing_data.h

@@ -52,7 +52,7 @@ class Mixing_Data
         ++this->ndim_history;
         FPTYPE* FP_startdata = static_cast<FPTYPE*>(this->data) + this->start * this->length;
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096/sizeof(FPTYPE))
+#pragma omp parallel for schedule(static)
 #endif
         for (std::size_t i = 0; i < length; ++i)
         {

source/source_base/module_mixing/plain_mixing.cpp

@@ -30,7 +30,7 @@ void Plain_Mixing::tem_push_data(Mixing_Data& mdata,
     const size_t length = mdata.length;
     std::vector<FPTYPE> F_tmp(length);
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(FPTYPE))
+#pragma omp parallel for schedule(static)
 #endif
     for (int i = 0; i < length; ++i)
     {
@@ -68,7 +68,7 @@ void Plain_Mixing::simple_mix(FPTYPE* data_new,
     if (screen == nullptr)
     {
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(FPTYPE))
+#pragma omp parallel for schedule(static)
 #endif
         for (int ig = 0; ig < length; ig++)
         {
@@ -79,15 +79,15 @@ void Plain_Mixing::simple_mix(FPTYPE* data_new,
     {
         std::vector<FPTYPE> F_tmp(length);
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(FPTYPE))
+#pragma omp parallel for schedule(static)
 #endif
         for (int i = 0; i < length; ++i)
         {
             F_tmp[i] = data_out[i] - data_in[i];
         }
         screen(F_tmp.data());
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(FPTYPE))
+#pragma omp parallel for schedule(static)
 #endif
         for (int i = 0; i < length; ++i)
         {

source/source_base/module_mixing/pulay_mixing.cpp

@@ -32,7 +32,7 @@ void Pulay_Mixing::tem_push_data(Mixing_Data& mdata,
     std::vector<FPTYPE> F_tmp(length);
 
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(FPTYPE))
+#pragma omp parallel for schedule(static)
 #endif
     for (std::size_t i = 0; i < length; ++i)
     {
@@ -67,7 +67,7 @@ void Pulay_Mixing::tem_push_data(Mixing_Data& mdata,
         F = malloc(sizeof(FPTYPE) * length * mixing_ndim);
         FP_F = static_cast<FPTYPE*>(F);
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(FPTYPE))
+#pragma omp parallel for schedule(static)
 #endif
         for (std::size_t i = 0; i < length; ++i)
         {
@@ -79,7 +79,7 @@ void Pulay_Mixing::tem_push_data(Mixing_Data& mdata,
         start_F = (this->start_F + 1) % this->mixing_ndim;
         FPTYPE* FP_startF = FP_F + start_F * length;
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096 / sizeof(FPTYPE))
+#pragma omp parallel for schedule(static)
 #endif
         for (std::size_t i = 0; i < length; ++i)
         {

source/source_basis/module_pw/pw_gatherscatter.h

@@ -102,7 +102,7 @@ void PW_Basis::gathers_scatterp(std::complex<T>* in, std::complex<T>* out) const
     if(this->poolnproc == 1) //In this case nrxx=fftnx*fftny*nz, nst = nstot, 
     {
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096/sizeof(T))
+#pragma omp parallel for schedule(static)
 #endif
         for(int i = 0; i < this->nrxx; ++i)
         {
@@ -160,7 +160,7 @@ void PW_Basis::gathers_scatterp(std::complex<T>* in, std::complex<T>* out) const
 	}
 
 #ifdef _OPENMP
-#pragma omp parallel for schedule(static, 4096/sizeof(T))
+#pragma omp parallel for schedule(static)
 #endif
     for(int i = 0; i < this->nrxx; ++i)
     {