Neko
0.9.0
A portable framework for high-order spectral element flow simulations
|
#include <math/bcknd/device/cuda/math_kernel.h>
Go to the source code of this file.
Functions | |
template<typename T > | |
__global__ void | fusedcg_update_p_kernel (T *__restrict__ p, const T *__restrict__ z, const T *__restrict__ po, const T beta, const int n) |
template<typename T > | |
__global__ void | fusedcg_update_x_kernel (T *__restrict__ x, const T **p, const T *__restrict__ alpha, const int p_cur, const int n) |
template<typename T > | |
__global__ void | fusedcg_part2_kernel (T *__restrict__ a, const T *__restrict__ b, const T *__restrict__ c, const T alpha, T *buf_h, const int n) |
__global__ void fusedcg_part2_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const T *__restrict__ | c, | ||
const T | alpha, | ||
T * | buf_h, | ||
const int | n | ||
) |
Device kernel for fusedcg_part2
Definition at line 85 of file fusedcg_kernel.h.
__global__ void fusedcg_update_p_kernel | ( | T *__restrict__ | p, |
const T *__restrict__ | z, | ||
const T *__restrict__ | po, | ||
const T | beta, | ||
const int | n | ||
) |
Kernel for update of p
Definition at line 43 of file fusedcg_kernel.h.
__global__ void fusedcg_update_x_kernel | ( | T *__restrict__ | x, |
const T ** | p, | ||
const T *__restrict__ | alpha, | ||
const int | p_cur, | ||
const int | n | ||
) |
Kernel for update of x
Definition at line 62 of file fusedcg_kernel.h.