Loading src/trapping/cfrfme.cpp +6 −4 Original line number Diff line number Diff line Loading @@ -667,7 +667,9 @@ void offload_loop( int nkvs = nkv * nkv; const dcomplex cc0 = 0.0 + I * 0.0; const dcomplex uim = 0.0 + I * 1.0; #pragma omp target teams distribute parallel for #pragma omp target teams distribute parallel for \ map(to: jlmf, jlml, nkv, nkvs, nrvc, nlmmt, nxv, nyv, nzv) \ map(to: delks, frsh) map(to: cc0, uim) for (int j80 = jlmf - 1; j80 < jlml; j80++) { dcomplex *vec_w = global_vec_w + nkvs * (j80 - jlmf + 1); #pragma omp parallel for simd Loading Loading @@ -700,9 +702,9 @@ void offload_loop( double vkx = (jx55 == 0) ? vkv[nkv - 1] : vkv[jx55]; double vkz = vec_vkzm[(jx55 * nkv) + jy60]; double sign = (jx55 == 0) ? -1.0 : 1.0; double rpart = cos(sign * vkx * x + vky * y + vkz * z); double ipart = sin(vkx * x + vky * y + vkz * z); dcomplex phas = rpart + I * ipart; double rpart = cos(vkx * x + vky * y + vkz * z); double ipart = sin(sign * vkx * x + vky * y + vkz * z); dcomplex phas = rpart + uim * ipart; // dcomplex phas = cexp(uim * (sign * vkx * x + vky * y + vkz * z)); dcomplex sumx = vec_w[w_index] * phas; double factor1 = ((jx55 == 0) || (jx55 == (nkv - 1))) ? 0.5 : 1.0; Loading Loading
src/trapping/cfrfme.cpp +6 −4 Original line number Diff line number Diff line Loading @@ -667,7 +667,9 @@ void offload_loop( int nkvs = nkv * nkv; const dcomplex cc0 = 0.0 + I * 0.0; const dcomplex uim = 0.0 + I * 1.0; #pragma omp target teams distribute parallel for #pragma omp target teams distribute parallel for \ map(to: jlmf, jlml, nkv, nkvs, nrvc, nlmmt, nxv, nyv, nzv) \ map(to: delks, frsh) map(to: cc0, uim) for (int j80 = jlmf - 1; j80 < jlml; j80++) { dcomplex *vec_w = global_vec_w + nkvs * (j80 - jlmf + 1); #pragma omp parallel for simd Loading Loading @@ -700,9 +702,9 @@ void offload_loop( double vkx = (jx55 == 0) ? vkv[nkv - 1] : vkv[jx55]; double vkz = vec_vkzm[(jx55 * nkv) + jy60]; double sign = (jx55 == 0) ? -1.0 : 1.0; double rpart = cos(sign * vkx * x + vky * y + vkz * z); double ipart = sin(vkx * x + vky * y + vkz * z); dcomplex phas = rpart + I * ipart; double rpart = cos(vkx * x + vky * y + vkz * z); double ipart = sin(sign * vkx * x + vky * y + vkz * z); dcomplex phas = rpart + uim * ipart; // dcomplex phas = cexp(uim * (sign * vkx * x + vky * y + vkz * z)); dcomplex sumx = vec_w[w_index] * phas; double factor1 = ((jx55 == 0) || (jx55 == (nkv - 1))) ? 0.5 : 1.0; Loading