Neko
0.9.99
A portable framework for high-order spectral element flow simulations
|
Go to the source code of this file.
Functions | |
template<typename T > | |
__global__ void | cmult_kernel (T *__restrict__ a, const T c, const int n) |
template<typename T > | |
__global__ void | masked_red_copy_kernel (T *__restrict__ a, T *__restrict__ b, int *__restrict__ mask, const int n, const int m) |
template<typename T > | |
__global__ void | masked_copy_kernel (T *__restrict__ a, T *__restrict__ b, int *__restrict__ mask, const int n, const int m) |
template<typename T > | |
__global__ void | cfill_mask_kernel (T *__restrict__ a, const T c, const int size, int *__restrict__ mask, const int mask_size) |
template<typename T > | |
__global__ void | cmult2_kernel (T *__restrict__ a, T *__restrict__ b, const T c, const int n) |
template<typename T > | |
__global__ void | cadd_kernel (T *__restrict__ a, const T c, const int n) |
template<typename T > | |
__global__ void | cadd2_kernel (T *__restrict__ a, T *__restrict__ b, const T c, const int n) |
template<typename T > | |
__global__ void | cfill_kernel (T *__restrict__ a, const T c, const int n) |
template<typename T > | |
__global__ void | add2_kernel (T *__restrict__ a, const T *__restrict__ b, const int n) |
template<typename T > | |
__global__ void | add3_kernel (T *__restrict__ a, const T *__restrict__ b, const T *__restrict__ c, const int n) |
template<typename T > | |
__global__ void | add4_kernel (T *__restrict__ a, const T *__restrict__ b, const T *__restrict__ c, const T *__restrict__ d, const int n) |
template<typename T > | |
__global__ void | add2s1_kernel (T *__restrict__ a, const T *__restrict__ b, const T c1, const int n) |
template<typename T > | |
__global__ void | add2s2_many_kernel (T *__restrict__ x, const T **p, const T *alpha, const int p_cur, const int n) |
template<typename T > | |
__global__ void | add2s2_kernel (T *__restrict__ a, const T *__restrict__ b, const T c1, const int n) |
template<typename T > | |
__global__ void | addsqr2s2_kernel (T *__restrict__ a, const T *__restrict__ b, const T c1, const int n) |
template<typename T > | |
__global__ void | add3s2_kernel (T *__restrict__ a, const T *__restrict__ b, const T *__restrict__ c, const T c1, const T c2, const int n) |
template<typename T > | |
__global__ void | invcol1_kernel (T *__restrict__ a, const int n) |
template<typename T > | |
__global__ void | invcol2_kernel (T *__restrict__ a, const T *__restrict__ b, const int n) |
template<typename T > | |
__global__ void | col2_kernel (T *__restrict__ a, const T *__restrict__ b, const int n) |
template<typename T > | |
__global__ void | col3_kernel (T *__restrict__ a, const T *__restrict__ b, const T *__restrict__ c, const int n) |
template<typename T > | |
__global__ void | subcol3_kernel (T *__restrict__ a, const T *__restrict__ b, const T *__restrict__ c, const int n) |
template<typename T > | |
__global__ void | sub2_kernel (T *__restrict__ a, const T *__restrict__ b, const int n) |
template<typename T > | |
__global__ void | sub3_kernel (T *__restrict__ a, const T *__restrict__ b, const T *__restrict__ c, const int n) |
template<typename T > | |
__global__ void | addcol3_kernel (T *__restrict__ a, const T *__restrict__ b, const T *__restrict__ c, const int n) |
template<typename T > | |
__global__ void | addcol4_kernel (T *__restrict__ a, const T *__restrict__ b, const T *__restrict__ c, const T *__restrict__ d, const int n) |
template<typename T > | |
__global__ void | vdot3_kernel (T *__restrict__ dot, const T *__restrict__ u1, const T *__restrict__ u2, const T *__restrict__ u3, const T *__restrict__ v1, const T *__restrict__ v2, const T *__restrict__ v3, const int n) |
template<typename T > | |
__global__ void | vcross_kernel (T *__restrict__ u1, T *__restrict__ u2, T *__restrict__ u3, const T *__restrict__ v1, const T *__restrict__ v2, const T *__restrict__ v3, const T *__restrict__ w1, const T *__restrict__ w2, const T *__restrict__ w3, const int n) |
template<typename T > | |
__inline__ __device__ T | reduce_warp (T val) |
template<typename T > | |
__global__ void | reduce_kernel (T *bufred, const int n) |
template<typename T > | |
__global__ void | glsc3_reduce_kernel (T *bufred, const int n, const int j) |
template<typename T > | |
__global__ void | glsc3_kernel (const T *a, const T *b, const T *c, T *buf_h, const int n) |
template<typename T > | |
__global__ void | glsc3_many_kernel (const T *a, const T **b, const T *c, T *buf_h, const int j, const int n) |
template<typename T > | |
__global__ void | glsc2_kernel (const T *a, const T *b, T *buf_h, const int n) |
template<typename T > | |
__global__ void | glsum_kernel (const T *a, T *buf_h, const int n) |
template<typename T > | |
__global__ void | absval_kernel (T *__restrict__ a, const int n) |
template<typename T > | |
__global__ void | pwmax_vec2_kernel (T *__restrict__ a, const T *__restrict__ b, const int n) |
template<typename T > | |
__global__ void | pwmax_vec3_kernel (T *__restrict__ a, const T *__restrict__ b, const T *__restrict__ c, const int n) |
template<typename T > | |
__global__ void | pwmax_sca2_kernel (T *__restrict__ a, const T c, const int n) |
template<typename T > | |
__global__ void | pwmax_sca3_kernel (T *__restrict__ a, const T *__restrict b, const T c, const int n) |
template<typename T > | |
__global__ void | pwmin_vec2_kernel (T *__restrict__ a, const T *__restrict__ b, const int n) |
template<typename T > | |
__global__ void | pwmin_vec3_kernel (T *__restrict__ a, const T *__restrict__ b, const T *__restrict__ c, const int n) |
template<typename T > | |
__global__ void | pwmin_sca2_kernel (T *__restrict__ a, const T c, const int n) |
template<typename T > | |
__global__ void | pwmin_sca3_kernel (T *__restrict__ a, const T *__restrict b, const T c, const int n) |
__global__ void absval_kernel | ( | T *__restrict__ | a, |
const int | n | ||
) |
Device kernel for abs_value
Definition at line 747 of file math_kernel.h.
__global__ void add2_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const int | n | ||
) |
Device kernel for add2
Definition at line 176 of file math_kernel.h.
__global__ void add2s1_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const T | c1, | ||
const int | n | ||
) |
Device kernel for add2s1
Definition at line 227 of file math_kernel.h.
__global__ void add2s2_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const T | c1, | ||
const int | n | ||
) |
Device kernel for add2s2
Definition at line 267 of file math_kernel.h.
__global__ void add2s2_many_kernel | ( | T *__restrict__ | x, |
const T ** | p, | ||
const T * | alpha, | ||
const int | p_cur, | ||
const int | n | ||
) |
Device kernel for add2s2 many
Definition at line 244 of file math_kernel.h.
__global__ void add3_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const T *__restrict__ | c, | ||
const int | n | ||
) |
Device kernel for add3
Definition at line 192 of file math_kernel.h.
__global__ void add3s2_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const T *__restrict__ | c, | ||
const T | c1, | ||
const T | c2, | ||
const int | n | ||
) |
Device kernel for add3s2
Definition at line 301 of file math_kernel.h.
__global__ void add4_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const T *__restrict__ | c, | ||
const T *__restrict__ | d, | ||
const int | n | ||
) |
Device kernel for add4
Definition at line 209 of file math_kernel.h.
__global__ void addcol3_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const T *__restrict__ | c, | ||
const int | n | ||
) |
Device kernel for addcol3
Definition at line 435 of file math_kernel.h.
__global__ void addcol4_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const T *__restrict__ | c, | ||
const T *__restrict__ | d, | ||
const int | n | ||
) |
Device kernel for addcol4
Definition at line 453 of file math_kernel.h.
__global__ void addsqr2s2_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const T | c1, | ||
const int | n | ||
) |
Device kernel for addsqr2s2
Definition at line 284 of file math_kernel.h.
__global__ void cadd2_kernel | ( | T *__restrict__ | a, |
T *__restrict__ | b, | ||
const T | c, | ||
const int | n | ||
) |
Device kernel for cadd2
Definition at line 143 of file math_kernel.h.
__global__ void cadd_kernel | ( | T *__restrict__ | a, |
const T | c, | ||
const int | n | ||
) |
Device kernel for cadd
Definition at line 127 of file math_kernel.h.
__global__ void cfill_kernel | ( | T *__restrict__ | a, |
const T | c, | ||
const int | n | ||
) |
Device kernel for cfill
Definition at line 160 of file math_kernel.h.
__global__ void cfill_mask_kernel | ( | T *__restrict__ | a, |
const T | c, | ||
const int | size, | ||
int *__restrict__ | mask, | ||
const int | mask_size | ||
) |
Device kernel for cfill_mask
Definition at line 94 of file math_kernel.h.
__global__ void cmult2_kernel | ( | T *__restrict__ | a, |
T *__restrict__ | b, | ||
const T | c, | ||
const int | n | ||
) |
Device kernel for cmult2
Definition at line 110 of file math_kernel.h.
__global__ void cmult_kernel | ( | T *__restrict__ | a, |
const T | c, | ||
const int | n | ||
) |
Device kernel for cmult
Definition at line 41 of file math_kernel.h.
__global__ void col2_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const int | n | ||
) |
Device kernel for col2
Definition at line 352 of file math_kernel.h.
__global__ void col3_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const T *__restrict__ | c, | ||
const int | n | ||
) |
Device kernel for col3
Definition at line 368 of file math_kernel.h.
__global__ void glsc2_kernel | ( | const T * | a, |
const T * | b, | ||
T * | buf_h, | ||
const int | n | ||
) |
Device kernel for glsc2
Definition at line 677 of file math_kernel.h.
__global__ void glsc3_kernel | ( | const T * | a, |
const T * | b, | ||
const T * | c, | ||
T * | buf_h, | ||
const int | n | ||
) |
Device kernel for glsc3
Definition at line 601 of file math_kernel.h.
__global__ void glsc3_many_kernel | ( | const T * | a, |
const T ** | b, | ||
const T * | c, | ||
T * | buf_h, | ||
const int | j, | ||
const int | n | ||
) |
Device kernel for glsc3 many
Definition at line 636 of file math_kernel.h.
__global__ void glsc3_reduce_kernel | ( | T * | bufred, |
const int | n, | ||
const int | j | ||
) |
Reduction kernel for glsc3
Definition at line 566 of file math_kernel.h.
__global__ void glsum_kernel | ( | const T * | a, |
T * | buf_h, | ||
const int | n | ||
) |
Device kernel for glsum
Definition at line 712 of file math_kernel.h.
__global__ void invcol1_kernel | ( | T *__restrict__ | a, |
const int | n | ||
) |
Device kernel for invcol1
Definition at line 320 of file math_kernel.h.
__global__ void invcol2_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const int | n | ||
) |
Device kernel for invcol2
Definition at line 336 of file math_kernel.h.
__global__ void masked_copy_kernel | ( | T *__restrict__ | a, |
T *__restrict__ | b, | ||
int *__restrict__ | mask, | ||
const int | n, | ||
const int | m | ||
) |
Device kernel for masked copy
Definition at line 76 of file math_kernel.h.
__global__ void masked_red_copy_kernel | ( | T *__restrict__ | a, |
T *__restrict__ | b, | ||
int *__restrict__ | mask, | ||
const int | n, | ||
const int | m | ||
) |
Device kernel for masked red copy
Definition at line 58 of file math_kernel.h.
__global__ void pwmax_sca2_kernel | ( | T *__restrict__ | a, |
const T | c, | ||
const int | n | ||
) |
Device kernel for point-wise max of vector and scalar a = max(a, c)
Definition at line 795 of file math_kernel.h.
__global__ void pwmax_sca3_kernel | ( | T *__restrict__ | a, |
const T *__restrict | b, | ||
const T | c, | ||
const int | n | ||
) |
Device kernel for point-wise max of vector and scalar a = max(b, c)
Definition at line 808 of file math_kernel.h.
__global__ void pwmax_vec2_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const int | n | ||
) |
Device kernel for point-wise max of two vectors a = max(a, b)
Definition at line 767 of file math_kernel.h.
__global__ void pwmax_vec3_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const T *__restrict__ | c, | ||
const int | n | ||
) |
Device kernel for point-wise max of two vectors a = max(b, c)
Definition at line 780 of file math_kernel.h.
__global__ void pwmin_sca2_kernel | ( | T *__restrict__ | a, |
const T | c, | ||
const int | n | ||
) |
Device kernel for point-wise min of vector and scalar a = min(a, c)
Definition at line 851 of file math_kernel.h.
__global__ void pwmin_sca3_kernel | ( | T *__restrict__ | a, |
const T *__restrict | b, | ||
const T | c, | ||
const int | n | ||
) |
Device kernel for point-wise min of vector and scalar a = min(b, c)
Definition at line 864 of file math_kernel.h.
__global__ void pwmin_vec2_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const int | n | ||
) |
Device kernel for point-wise min of two vectors a = min(a, b)
Definition at line 823 of file math_kernel.h.
__global__ void pwmin_vec3_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const T *__restrict__ | c, | ||
const int | n | ||
) |
Device kernel for point-wise min of two vectors a = min(b, c)
Definition at line 836 of file math_kernel.h.
__global__ void reduce_kernel | ( | T * | bufred, |
const int | n | ||
) |
Vector reduction kernel
Definition at line 534 of file math_kernel.h.
__inline__ __device__ T reduce_warp | ( | T | val | ) |
Warp shuffle reduction
Definition at line 521 of file math_kernel.h.
__global__ void sub2_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const int | n | ||
) |
Device kernel for sub2
Definition at line 402 of file math_kernel.h.
__global__ void sub3_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const T *__restrict__ | c, | ||
const int | n | ||
) |
Device kernel for sub3
Definition at line 418 of file math_kernel.h.
__global__ void subcol3_kernel | ( | T *__restrict__ | a, |
const T *__restrict__ | b, | ||
const T *__restrict__ | c, | ||
const int | n | ||
) |
Device kernel for subcol3
Definition at line 385 of file math_kernel.h.
__global__ void vcross_kernel | ( | T *__restrict__ | u1, |
T *__restrict__ | u2, | ||
T *__restrict__ | u3, | ||
const T *__restrict__ | v1, | ||
const T *__restrict__ | v2, | ||
const T *__restrict__ | v3, | ||
const T *__restrict__ | w1, | ||
const T *__restrict__ | w2, | ||
const T *__restrict__ | w3, | ||
const int | n | ||
) |
Device kernel for vcross
Definition at line 494 of file math_kernel.h.
__global__ void vdot3_kernel | ( | T *__restrict__ | dot, |
const T *__restrict__ | u1, | ||
const T *__restrict__ | u2, | ||
const T *__restrict__ | u3, | ||
const T *__restrict__ | v1, | ||
const T *__restrict__ | v2, | ||
const T *__restrict__ | v3, | ||
const int | n | ||
) |
Device kernel for vdot3
Definition at line 472 of file math_kernel.h.