Neko  0.9.99
A portable framework for high-order spectral element flow simulations
Macros | Functions
coef.cu File Reference
#include <stdio.h>
#include "coef_kernel.h"
#include <device/device_config.h>
#include <device/cuda/check.h>
Include dependency graph for coef.cu:

Go to the source code of this file.

Macros

#define GEO_CASE(LX)
 
#define DXYZDRST_CASE(LX)
 

Functions

void cuda_coef_generate_geo (void *G11, void *G12, void *G13, void *G22, void *G23, void *G33, void *drdx, void *drdy, void *drdz, void *dsdx, void *dsdy, void *dsdz, void *dtdx, void *dtdy, void *dtdz, void *jacinv, void *w3, int *nel, int *lx, int *gdim)
 
void cuda_coef_generate_dxyzdrst (void *drdx, void *drdy, void *drdz, void *dsdx, void *dsdy, void *dsdz, void *dtdx, void *dtdy, void *dtdz, void *dxdr, void *dydr, void *dzdr, void *dxds, void *dyds, void *dzds, void *dxdt, void *dydt, void *dzdt, void *dx, void *dy, void *dz, void *x, void *y, void *z, void *jacinv, void *jac, int *lx, int *nel)
 

Macro Definition Documentation

◆ DXYZDRST_CASE

#define DXYZDRST_CASE (   LX)
Value:
case LX: \
coef_generate_dxyz_kernel<real, LX, 1024> \
<<<nblcks_dxyz, nthrds, 0, stream>>> \
((real *) dxdr, (real *) dydr, (real *) dzdr, \
(real *) dxds, (real *) dyds, (real *) dzds, \
(real *) dxdt, (real *) dydt, (real *) dzdt, \
(real *) dx, (real *) dy, (real *) dz, \
(real *) x, (real *) y, (real *) z); \
CUDA_CHECK(cudaGetLastError()); \
break
dx
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dx
Definition: ax_helm_full_kernel.h:49
dz
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dz
Definition: ax_helm_full_kernel.h:51
dy
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dy
Definition: ax_helm_full_kernel.h:50
x
__global__ void const T *__restrict__ x
Definition: cdtp_kernel.h:106
real
double real
Definition: device_config.h:12

◆ GEO_CASE

#define GEO_CASE (   LX)
Value:
case LX: \
coef_generate_geo_kernel<real, LX, 1024> \
<<<nblcks, nthrds, 0, stream>>> \
((real *) G11, (real *) G12, (real *) G13, \
(real *) G22, (real *) G23, (real *) G33, \
(real *) drdx, (real *) drdy, (real *) drdz, \
(real *) dsdx, (real *) dsdy, (real *) dsdz, \
(real *) dtdx, (real *) dtdy, (real *) dtdz, \
(real *) jacinv, (real *) w3, *gdim); \
CUDA_CHECK(cudaGetLastError()); \
break
drdy
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ drdy
Definition: ax_helm_full_kernel.h:54
drdz
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ drdz
Definition: ax_helm_full_kernel.h:55
dsdz
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dsdz
Definition: ax_helm_full_kernel.h:58
dsdy
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dsdy
Definition: ax_helm_full_kernel.h:57
dtdy
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dtdy
Definition: ax_helm_full_kernel.h:60
drdx
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ drdx
Definition: ax_helm_full_kernel.h:53
dtdz
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dtdz
Definition: ax_helm_full_kernel.h:61
dsdx
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dsdx
Definition: ax_helm_full_kernel.h:56
dtdx
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dtdx
Definition: ax_helm_full_kernel.h:59
jacinv
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ jacinv
Definition: ax_helm_full_kernel.h:62
w3
__global__ void const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ w3
Definition: cdtp_kernel.h:113

Function Documentation

◆ cuda_coef_generate_dxyzdrst()

void cuda_coef_generate_dxyzdrst ( void *  drdx,
void *  drdy,
void *  drdz,
void *  dsdx,
void *  dsdy,
void *  dsdz,
void *  dtdx,
void *  dtdy,
void *  dtdz,
void *  dxdr,
void *  dydr,
void *  dzdr,
void *  dxds,
void *  dyds,
void *  dzds,
void *  dxdt,
void *  dydt,
void *  dzdt,
void *  dx,
void *  dy,
void *  dz,
void *  x,
void *  y,
void *  z,
void *  jacinv,
void *  jac,
int *  lx,
int *  nel 
)

Fortran wrapper for generating geometric factors

Definition at line 97 of file coef.cu.

Here is the caller graph for this function:

◆ cuda_coef_generate_geo()

void cuda_coef_generate_geo ( void *  G11,
void *  G12,
void *  G13,
void *  G22,
void *  G23,
void *  G33,
void *  drdx,
void *  drdy,
void *  drdz,
void *  dsdx,
void *  dsdy,
void *  dsdz,
void *  dtdx,
void *  dtdy,
void *  dtdz,
void *  jacinv,
void *  w3,
int *  nel,
int *  lx,
int *  gdim 
)

Fortran wrapper for generating geometric factors

Definition at line 45 of file coef.cu.

Here is the caller graph for this function: