math.cu File Reference

#include "math_kernel.h"
#include <device/device_config.h>
#include <device/cuda/check.h>
#include <stdio.h>
#include <stdlib.h>
#include <math/bcknd/device/device_mpi_reduce.h>
#include <math/bcknd/device/device_mpi_op.h>

Go to the source code of this file.

Functions
void	cuda_copy (void *a, void *b, int *n)
void	cuda_masked_copy (void *a, void *b, void *mask, int *n, int *m)
void	cuda_masked_red_copy (void *a, void *b, void *mask, int *n, int *m)
void	cuda_masked_atomic_reduction (void *a, void *b, void *mask, int *n, int *m)
void	cuda_cfill_mask (void *a, real *c, int *size, int *mask, int *mask_size)
void	cuda_rzero (void *a, int *n)
void	cuda_cmult (void *a, real *c, int *n)
void	cuda_cmult2 (void *a, void *b, real *c, int *n)
void	cuda_cadd (void *a, real *c, int *n)
void	cuda_cadd2 (void *a, void *b, real *c, int *n)
void	cuda_cfill (void *a, real *c, int *n)
void	cuda_add2 (void *a, void *b, int *n)
void	cuda_add3 (void *a, void *b, void *c, int *n)
void	cuda_add4 (void *a, void *b, void *c, void *d, int *n)
void	cuda_add2s1 (void *a, void *b, real *c1, int *n)
void	cuda_add2s2 (void *a, void *b, real *c1, int *n)
void	cuda_add2s2_many (void *x, void **p, void *alpha, int *j, int *n)
void	cuda_addsqr2s2 (void *a, void *b, real *c1, int *n)
void	cuda_add3s2 (void *a, void *b, void *c, real *c1, real *c2, int *n)
void	cuda_invcol1 (void *a, int *n)
void	cuda_invcol2 (void *a, void *b, int *n)
void	cuda_col2 (void *a, void *b, int *n)
void	cuda_col3 (void *a, void *b, void *c, int *n)
void	cuda_subcol3 (void *a, void *b, void *c, int *n)
void	cuda_sub2 (void *a, void *b, int *n)
void	cuda_sub3 (void *a, void *b, void *c, int *n)
void	cuda_addcol3 (void *a, void *b, void *c, int *n)
void	cuda_addcol4 (void *a, void *b, void *c, void *d, int *n)
void	cuda_vdot3 (void *dot, void *u1, void *u2, void *u3, void *v1, void *v2, void *v3, int *n)
void	cuda_vcross (void *u1, void *u2, void *u3, void *v1, void *v2, void *v3, void *w1, void *w2, void *w3, int *n)
real	cuda_vlsc3 (void *u, void *v, void *w, int *n)
real	cuda_glsc3 (void *a, void *b, void *c, int *n)
void	cuda_glsc3_many (real *h, void *w, void *v, void *mult, int *j, int *n)
real	cuda_glsc2 (void *a, void *b, int *n)
real	cuda_glsum (void *a, int *n)
void	cuda_absval (void *a, int *n)
void	cuda_pwmax_vec2 (void *a, void *b, int *n)
void	cuda_pwmax_vec3 (void *a, void *b, void *c, int *n)
void	cuda_pwmax_sca2 (void *a, real *c, int *n)
void	cuda_pwmax_sca3 (void *a, void *b, real *c, int *n)
void	cuda_pwmin_vec2 (void *a, void *b, int *n)
void	cuda_pwmin_vec3 (void *a, void *b, void *c, int *n)
void	cuda_pwmin_sca2 (void *a, real *c, int *n)
void	cuda_pwmin_sca3 (void *a, void *b, real *c, int *n)

Variables
int	red_s = 0
real *	bufred = NULL
void *	bufred_d = NULL

Function Documentation

cuda_absval()

void cuda_absval	(	void *	a,
		int *	n
	)

Fortran wrapper absval Take the abs value of a vector of length n

Definition at line 826 of file math.cu.

Here is the call graph for this function:

cuda_add2()

void cuda_add2	(	void *	a,
		void *	b,
		int *	n
	)

Fortran wrapper for add2 Vector addition $a = a + b$

Definition at line 207 of file math.cu.

Here is the call graph for this function:

cuda_add2s1()

void cuda_add2s1	(	void *	a,
		void *	b,
		real *	c1,
		int *	n
	)

Fortran wrapper for add2s1 Vector addition with scalar multiplication $a = c_1 a + b$ (multiplication on first argument)

Definition at line 252 of file math.cu.

Here is the call graph for this function:

cuda_add2s2()

void cuda_add2s2	(	void *	a,
		void *	b,
		real *	c1,
		int *	n
	)

Fortran wrapper for add2s2 Vector addition with scalar multiplication $a = a + c_1 b$ (multiplication on second argument)

Definition at line 268 of file math.cu.

Here is the call graph for this function:

cuda_add2s2_many()

void cuda_add2s2_many	(	void *	x,
		void **	p,
		void *	alpha,
		int *	j,
		int *	n
	)

Fortran wrapper for add2s2 Vector addition with scalar multiplication $x = x + c_1 p1 + c_2p2 + ... + c_jpj$ (multiplication on second argument)

Definition at line 285 of file math.cu.

Here is the call graph for this function:

cuda_add3()

void cuda_add3	(	void *	a,
		void *	b,
		void *	c,
		int *	n
	)

Fortran wrapper for add3 Vector addition $a = b + c$

Definition at line 222 of file math.cu.

Here is the call graph for this function:

cuda_add3s2()

void cuda_add3s2	(	void *	a,
		void *	b,
		void *	c,
		real *	c1,
		real *	c2,
		int *	n
	)

Fortran wrapper for add3s2 Vector addition with scalar multiplication $a = c_1 b + c_2 c$ (multiplication on second argument)

Definition at line 319 of file math.cu.

Here is the call graph for this function:

cuda_add4()

void cuda_add4	(	void *	a,
		void *	b,
		void *	c,
		void *	d,
		int *	n
	)

Fortran wrapper for add4 Vector addition $a = b + c + d$

Definition at line 237 of file math.cu.

Here is the call graph for this function:

cuda_addcol3()

void cuda_addcol3	(	void *	a,
		void *	b,
		void *	c,
		int *	n
	)

Fortran wrapper for addcol3 $a = a + b * c$

Definition at line 435 of file math.cu.

Here is the call graph for this function:

cuda_addcol4()

void cuda_addcol4	(	void *	a,
		void *	b,
		void *	c,
		void *	d,
		int *	n
	)

Fortran wrapper for addcol4 $a = a + b * c * d$

Definition at line 450 of file math.cu.

Here is the call graph for this function:

cuda_addsqr2s2()

void cuda_addsqr2s2	(	void *	a,
		void *	b,
		real *	c1,
		int *	n
	)

Fortran wrapper for addsqr2s2 Vector addition with scalar multiplication $a = a + c_1 (b * b)$ (multiplication on second argument)

Definition at line 302 of file math.cu.

Here is the call graph for this function:

cuda_cadd()

void cuda_cadd	(	void *	a,
		real *	c,
		int *	n
	)

Fortran wrapper for cadd Add a scalar to vector $a_i = a_i + c$

Definition at line 161 of file math.cu.

Here is the call graph for this function:

cuda_cadd2()

void cuda_cadd2	(	void *	a,
		void *	b,
		real *	c,
		int *	n
	)

Fortran wrapper for cadd2 Add a scalar to vector $a_i = b_i + c$

Definition at line 176 of file math.cu.

Here is the call graph for this function:

cuda_cfill()

void cuda_cfill	(	void *	a,
		real *	c,
		int *	n
	)

Fortran wrapper for cfill Set all elements to a constant c $a = c$

Definition at line 190 of file math.cu.

Here is the call graph for this function:

cuda_cfill_mask()

void cuda_cfill_mask	(	void *	a,
		real *	c,
		int *	size,
		int *	mask,
		int *	mask_size
	)

Fortran wrapper for cfill_mask Fill a scalar to vector $a_i = s, for i \in mask$

Definition at line 112 of file math.cu.

Here is the call graph for this function:

cuda_cmult()

void cuda_cmult	(	void *	a,
		real *	c,
		int *	n
	)

Fortran wrapper for cmult Multiplication by constant c $a = c \cdot a$

Definition at line 133 of file math.cu.

Here is the call graph for this function:

cuda_cmult2()

void cuda_cmult2	(	void *	a,
		void *	b,
		real *	c,
		int *	n
	)

Fortran wrapper for cmult2 Multiplication by constant c $a = c \cdot b$

Definition at line 147 of file math.cu.

Here is the call graph for this function:

cuda_col2()

void cuda_col2	(	void *	a,
		void *	b,
		int *	n
	)

Fortran wrapper for col2 Vector multiplication with 2 vectors $a = a \cdot b$

Definition at line 363 of file math.cu.

Here is the call graph for this function:

cuda_col3()

void cuda_col3	(	void *	a,
		void *	b,
		void *	c,
		int *	n
	)

Fortran wrapper for col3 Vector multiplication with 3 vectors $a = b \cdot c$

Definition at line 377 of file math.cu.

Here is the call graph for this function:

cuda_copy()

void cuda_copy	(	void *	a,
		void *	b,
		int *	n
	)

Fortran wrapper for copy Copy a vector $a = b$

Definition at line 59 of file math.cu.

Here is the call graph for this function:

cuda_glsc2()

real cuda_glsc2	(	void *	a,
		void *	b,
		int *	n
	)

Fortran wrapper glsc2 Weighted inner product $a^T b c$

Definition at line 685 of file math.cu.

Here is the call graph for this function:

cuda_glsc3()

real cuda_glsc3	(	void *	a,
		void *	b,
		void *	c,
		int *	n
	)

Fortran wrapper glsc3 Weighted inner product $a^T b c$

Definition at line 546 of file math.cu.

Here is the call graph for this function:

cuda_glsc3_many()

void cuda_glsc3_many	(	real *	h,
		void *	w,
		void *	v,
		void *	mult,
		int *	j,
		int *	n
	)

Fortran wrapper for doing an reduction to an array Weighted inner product $w^T v(n,1:j) c$

Definition at line 614 of file math.cu.

Here is the call graph for this function:

cuda_glsum()

real cuda_glsum	(	void *	a,
		int *	n
	)

Fortran wrapper glsum Sum a vector of length n

Definition at line 756 of file math.cu.

Here is the call graph for this function:

cuda_invcol1()

void cuda_invcol1	(	void *	a,
		int *	n
	)

Fortran wrapper for invcol1 Invert a vector $a = 1 / a$

Definition at line 335 of file math.cu.

Here is the call graph for this function:

cuda_invcol2()

void cuda_invcol2	(	void *	a,
		void *	b,
		int *	n
	)

Fortran wrapper for invcol2 Vector division $a = a / b$

Definition at line 349 of file math.cu.

Here is the call graph for this function:

cuda_masked_atomic_reduction()

void cuda_masked_atomic_reduction	(	void *	a,
		void *	b,
		void *	mask,
		int *	n,
		int *	m
	)

Fortran wrapper for masked atomic reduction update a vector $a += b(mask)$ where mask is not unique

Definition at line 96 of file math.cu.

Here is the call graph for this function:

cuda_masked_copy()

void cuda_masked_copy	(	void *	a,
		void *	b,
		void *	mask,
		int *	n,
		int *	m
	)

Fortran wrapper for masked copy Copy a vector $a(mask) = b(mask)$

Definition at line 68 of file math.cu.

Here is the call graph for this function:

cuda_masked_red_copy()

void cuda_masked_red_copy	(	void *	a,
		void *	b,
		void *	mask,
		int *	n,
		int *	m
	)

Fortran wrapper for masked copy Copy a vector $a(mask) = b(mask)$

Definition at line 82 of file math.cu.

Here is the call graph for this function:

cuda_pwmax_sca2()

void cuda_pwmax_sca2	(	void *	a,
		real *	c,
		int *	n
	)

Fortran wrapper for pwmax_sca2

Compute the maximum of vector and scalar $a = \max(a, c)$

Definition at line 875 of file math.cu.

Here is the call graph for this function:

cuda_pwmax_sca3()

void cuda_pwmax_sca3	(	void *	a,
		void *	b,
		real *	c,
		int *	n
	)

Fortran wrapper for pwmax_sca3

Compute the maximum of vector and scalar $a = \max(b, c)$

Definition at line 890 of file math.cu.

Here is the call graph for this function:

cuda_pwmax_vec2()

void cuda_pwmax_vec2	(	void *	a,
		void *	b,
		int *	n
	)

Fortran wrapper for pwmax_vec2

Compute the maximum of two vectors $a = \max(a, b)$

Definition at line 845 of file math.cu.

Here is the call graph for this function:

cuda_pwmax_vec3()

void cuda_pwmax_vec3	(	void *	a,
		void *	b,
		void *	c,
		int *	n
	)

Fortran wrapper for pwmax_vec3

Compute the maximum of two vectors $a = \max(b, c)$

Definition at line 860 of file math.cu.

Here is the call graph for this function:

cuda_pwmin_sca2()

void cuda_pwmin_sca2	(	void *	a,
		real *	c,
		int *	n
	)

Fortran wrapper for pwmin_sca2

Compute the minimum of vector and scalar $a = \min(a, c)$

Definition at line 935 of file math.cu.

Here is the call graph for this function:

cuda_pwmin_sca3()

void cuda_pwmin_sca3	(	void *	a,
		void *	b,
		real *	c,
		int *	n
	)

Fortran wrapper for pwmin_sca3

Compute the minimum of vector and scalar $a = \min(b, c)$

Definition at line 950 of file math.cu.

Here is the call graph for this function:

cuda_pwmin_vec2()

void cuda_pwmin_vec2	(	void *	a,
		void *	b,
		int *	n
	)

Fortran wrapper for pwmin_vec2

Compute the minimum of two vectors $a = \min(a, b)$

Definition at line 905 of file math.cu.

Here is the call graph for this function:

cuda_pwmin_vec3()

void cuda_pwmin_vec3	(	void *	a,
		void *	b,
		void *	c,
		int *	n
	)

Fortran wrapper for pwmin_vec3

Compute the minimum of two vectors $a = \min(b, c)$

Definition at line 920 of file math.cu.

Here is the call graph for this function:

cuda_rzero()

void cuda_rzero	(	void *	a,
		int *	n
	)

Fortran wrapper for rzero Zero a real vector

Definition at line 125 of file math.cu.

Here is the call graph for this function:

cuda_sub2()

void cuda_sub2	(	void *	a,
		void *	b,
		int *	n
	)

Fortran wrapper for sub2 Vector subtraction $a = a - b$

Definition at line 406 of file math.cu.

Here is the call graph for this function:

cuda_sub3()

void cuda_sub3	(	void *	a,
		void *	b,
		void *	c,
		int *	n
	)

Fortran wrapper for sub3 Vector subtraction $a = b - c$

Definition at line 420 of file math.cu.

Here is the call graph for this function:

cuda_subcol3()

void cuda_subcol3	(	void *	a,
		void *	b,
		void *	c,
		int *	n
	)

Fortran wrapper for subcol3 Vector multiplication with 3 vectors $a = a - b \cdot c$

Definition at line 391 of file math.cu.

Here is the call graph for this function:

cuda_vcross()

void cuda_vcross	(	void *	u1,
		void *	u2,
		void *	u3,
		void *	v1,
		void *	v2,
		void *	v3,
		void *	w1,
		void *	w2,
		void *	w3,
		int *	n
	)

Fortran wrapper for vcross $u = v \times w$

Definition at line 483 of file math.cu.

Here is the call graph for this function:

cuda_vdot3()

void cuda_vdot3	(	void *	dot,
		void *	u1,
		void *	u2,
		void *	u3,
		void *	v1,
		void *	v2,
		void *	v3,
		int *	n
	)

Fortran wrapper for vdot3 $dot = u \cdot v$

Definition at line 465 of file math.cu.

Here is the call graph for this function:

cuda_vlsc3()

real cuda_vlsc3	(	void *	u,
		void *	v,
		void *	w,
		int *	n
	)

Fortran wrapper vlsc3 Compute multiplication sum $dot = u \cdot v \cdot w$

Definition at line 512 of file math.cu.

Here is the call graph for this function:

Variable Documentation

bufred

real* bufred = NULL

Definition at line 505 of file math.cu.

bufred_d

void* bufred_d = NULL

Definition at line 506 of file math.cu.

red_s

int red_s = 0

Definition at line 504 of file math.cu.