65 use math,
only:
rzero,
rone,
copy,
cmult,
cadd,
cfill,
invcol1,
vdot3,
add2, &
80 use,
intrinsic :: iso_c_binding, only: c_ptr
99 integer,
intent(in),
optional :: n
100 type(
field_t),
intent(inout) :: a
112 call rzero(a%x, size)
118 integer,
intent(in),
optional :: n
119 type(
field_t),
intent(inout) :: a
137 integer,
intent(in),
optional :: n
139 type(
field_t),
intent(inout) :: a
151 call copy(a%x, b%x, size)
157 integer,
intent(in),
optional :: n
158 type(
field_t),
intent(inout) :: a
159 real(kind=
rp),
intent(in) :: c
171 call cmult(a%x, c, size)
177 integer,
intent(in),
optional :: n
178 type(
field_t),
intent(inout) :: a
179 real(kind=
rp),
intent(in) :: s
191 call cadd(a%x, s, size)
197 integer,
intent(in),
optional :: n
198 type(
field_t),
intent(inout) :: a
199 real(kind=
rp),
intent(in) :: c
211 call cfill(a%x, c, size)
217 integer,
intent(in),
optional :: n
218 type(
field_t),
intent(inout) :: a
237 integer,
intent(in),
optional :: n
238 type(
field_t),
intent(inout) :: a
252 call invcol3(a%x, b%x, c%x, size)
260 integer,
intent(in),
optional :: n
261 type(
field_t),
intent(in) :: u1, u2, u3
262 type(
field_t),
intent(in) :: v1, v2, v3
263 type(
field_t),
intent(out) :: dot
274 u1%x_d, u2%x_d, u3%x_d, &
275 v1%x_d, v2%x_d, v3%x_d, &
288 integer,
intent(in),
optional :: n
289 type(
field_t),
intent(inout) :: a
302 call add2(a%x, b%x, size)
309 integer,
intent(in),
optional :: n
310 type(
field_t),
intent(inout) :: a
311 type(
field_t),
intent(in) :: b, c
323 call add3(a%x, b%x, c%x, size)
330 integer,
intent(in),
optional :: n
331 type(
field_t),
intent(inout) :: a
332 type(
field_t),
intent(in) :: b, c, d
344 call add4(a%x, b%x, c%x, d%x, size)
351 integer,
intent(in),
optional :: n
352 type(
field_t),
intent(inout) :: a
353 type(
field_t),
intent(inout) :: b
365 call sub2(a%x, b%x, size)
372 integer,
intent(in),
optional :: n
373 type(
field_t),
intent(inout) :: a
387 call sub3(a%x, b%x, c%x, size)
396 integer,
intent(in),
optional :: n
397 type(
field_t),
intent(inout) :: a
398 type(
field_t),
intent(inout) :: b
399 real(kind=
rp),
intent(in) :: c1
411 call add2s1(a%x, b%x, c1, size)
419 integer,
intent(in),
optional :: n
420 type(
field_t),
intent(inout) :: a
421 type(
field_t),
intent(inout) :: b
422 real(kind=
rp),
intent(in) :: c1
434 call add2s2(a%x, b%x, c1, size)
441 integer,
intent(in),
optional :: n
442 type(
field_t),
intent(inout) :: a
444 real(kind=
rp),
intent(in) :: c1
463 integer,
intent(in),
optional :: n
464 type(
field_t),
intent(inout) :: a
466 real(kind=
rp),
intent(in) :: c
478 call cmult2(a%x, b%x, c, size)
485 integer,
intent(in),
optional :: n
486 type(
field_t),
intent(inout) :: a
507 integer,
intent(in),
optional :: n
508 type(
field_t),
intent(inout) :: a
521 call col2(a%x, b%x, size)
528 integer,
intent(in),
optional :: n
529 type(
field_t),
intent(inout) :: a
543 call col3(a%x, b%x, c%x, size)
550 integer,
intent(in),
optional :: n
551 type(
field_t),
intent(inout) :: a
565 call subcol3(a%x, b%x, c%x, size)
572 integer,
intent(in),
optional :: n
573 type(
field_t),
intent(inout) :: a
576 real(kind=
rp),
intent(in) :: c1, c2
588 call add3s2(a%x, b%x, c%x, c1, c2, size)
595 integer,
intent(in),
optional :: n
596 type(
field_t),
intent(inout) :: a
610 call addcol3(a%x, b%x, c%x, size)
617 integer,
intent(in),
optional :: n
618 type(
field_t),
intent(inout) :: a
633 call addcol4(a%x, b%x, c%x, d%x, size)
639 integer,
intent(in),
optional :: n
653 sum =
glsum(a%x, size)
659 integer,
intent(in),
optional :: n
660 type(
field_t),
intent(in) :: a, b
661 real(kind=
rp) :: norm
673 norm =
glsc2(a%x, b%x, size)
679 integer,
intent(in),
optional :: n
680 type(
field_t),
intent(in) :: a, b, c
681 real(kind=
rp) :: norm
693 norm =
glsc3(a%x, b%x, c%x, size)
699 integer,
intent(in),
optional :: n
700 type(
field_t),
intent(in) :: a, b
701 real(kind=
rp) :: norm
727 integer,
intent(in) :: n, n_mask
728 real(kind=
rp),
dimension(n_mask),
intent(inout) :: a
730 integer,
dimension(0:n_mask) ::
mask
731 type(c_ptr) :: mask_d, a_d
752 integer,
intent(in) :: n, n_mask
753 real(kind=
rp),
dimension(n_mask),
intent(in) :: b
754 type(
field_t),
intent(inout) :: a
755 integer,
dimension(0:n_mask) ::
mask
756 type(c_ptr) :: mask_d, b_d
771 integer,
intent(in),
optional :: n
772 type(
field_t),
intent(inout) :: a
785 call pwmax2(a%x, b%x, size)
793 integer,
intent(in),
optional :: n
794 type(
field_t),
intent(inout) :: a
808 call pwmax3(a%x, b%x, c%x, size)
815 integer,
intent(in),
optional :: n
816 type(
field_t),
intent(inout) :: a
817 real(kind=
rp),
intent(in) :: b
837 integer,
intent(in),
optional :: n
838 type(
field_t),
intent(inout) :: a
840 real(kind=
rp),
intent(in) :: c
852 call cpwmax3(a%x, b%x, c, size)
860 integer,
intent(in),
optional :: n
861 type(
field_t),
intent(inout) :: a
874 call pwmin2(a%x, b%x, size)
882 integer,
intent(in),
optional :: n
883 type(
field_t),
intent(inout) :: a
897 call pwmin3(a%x, b%x, c%x, size)
904 integer,
intent(in),
optional :: n
905 type(
field_t),
intent(inout) :: a
906 real(kind=
rp),
intent(in) :: b
926 integer,
intent(in),
optional :: n
927 type(
field_t),
intent(inout) :: a
929 real(kind=
rp),
intent(in) :: c
941 call cpwmin3(a%x, b%x, c, size)
Return the device pointer for an associated Fortran array.
subroutine, public device_pwmin2(a_d, b_d, n, strm)
Compute the point-wise minimum of two vectors .
subroutine, public device_add2s1(a_d, b_d, c1, n, strm)
subroutine, public device_sub3(a_d, b_d, c_d, n, strm)
Vector subtraction .
subroutine, public device_masked_scatter_copy_0(a_d, b_d, mask_d, n, n_mask, strm)
Scatter a masked vector .
subroutine, public device_add2s2(a_d, b_d, c1, n, strm)
Vector addition with scalar multiplication (multiplication on first argument)
subroutine, public device_add2(a_d, b_d, n, strm)
Vector addition .
subroutine, public device_addcol3(a_d, b_d, c_d, n, strm)
Returns .
real(kind=rp) function, public device_glsum(a_d, n, strm)
Sum a vector of length n.
subroutine, public device_pwmax3(a_d, b_d, c_d, n, strm)
Compute the point-wise maximum of two vectors .
subroutine, public device_invcol1(a_d, n, strm)
Invert a vector .
subroutine, public device_add3s2(a_d, b_d, c_d, c1, c2, n, strm)
Returns .
subroutine, public device_cpwmax3(a_d, b_d, c, n, strm)
Compute the point-wise maximum of a vector and a scalar .
subroutine, public device_rzero(a_d, n, strm)
Zero a real vector.
subroutine, public device_rone(a_d, n, strm)
Set all elements to one.
subroutine, public device_cmult(a_d, c, n, strm)
Multiplication by constant c .
subroutine, public device_vdot3(dot_d, u1_d, u2_d, u3_d, v1_d, v2_d, v3_d, n, strm)
Compute a dot product (3-d version) assuming vector components etc.
real(kind=rp) function, public device_glsubnorm(a_d, b_d, n, strm)
Returns the norm of the difference of two vectors .
subroutine, public device_sub2(a_d, b_d, n, strm)
Vector substraction .
subroutine, public device_copy(a_d, b_d, n, strm)
Copy a vector .
subroutine, public device_invcol3(a_d, b_d, c_d, n, strm)
Vector division .
subroutine, public device_pwmin3(a_d, b_d, c_d, n, strm)
Compute the point-wise minimum of two vectors .
subroutine, public device_col2(a_d, b_d, n, strm)
Vector multiplication .
subroutine, public device_add4(a_d, b_d, c_d, d_d, n, strm)
subroutine, public device_cpwmax2(a_d, c, n, strm)
Compute the point-wise maximum of a vector and a scalar .
subroutine, public device_subcol3(a_d, b_d, c_d, n, strm)
Returns .
subroutine, public device_masked_gather_copy_0(a_d, b_d, mask_d, n, n_mask, strm)
Gather a masked vector .
subroutine, public device_invcol2(a_d, b_d, n, strm)
Vector division .
subroutine, public device_addsqr2s2(a_d, b_d, c1, n, strm)
Returns .
real(kind=rp) function, public device_glsc3(a_d, b_d, c_d, n, strm)
Weighted inner product .
subroutine, public device_cpwmin3(a_d, b_d, c, n, strm)
Compute the point-wise minimum of a vector and a scalar .
real(kind=rp) function, public device_glsc2(a_d, b_d, n, strm)
Weighted inner product .
subroutine, public device_cmult2(a_d, b_d, c, n, strm)
Multiplication by constant c .
subroutine, public device_col3(a_d, b_d, c_d, n, strm)
Vector multiplication with 3 vectors .
subroutine, public device_addcol4(a_d, b_d, c_d, d_d, n, strm)
Returns .
subroutine, public device_cfill(a_d, c, n, strm)
Set all elements to a constant c .
subroutine, public device_cpwmin2(a_d, c, n, strm)
Compute the point-wise minimum of a vector and a scalar .
subroutine, public device_add3(a_d, b_d, c_d, n, strm)
Vector addition .
subroutine, public device_pwmax2(a_d, b_d, n, strm)
Compute the point-wise maximum of two vectors .
Device abstraction, common interface for various accelerators.
subroutine, public field_cpwmax2(a, b, n)
Point-wise max operation for field and constant .
subroutine, public field_add2s1(a, b, c1, n)
Vector addition with scalar multiplication (multiplication on first argument)
subroutine, public field_cpwmin3(a, b, c, n)
Point-wise min operation for field and constant .
subroutine, public field_invcol2(a, b, n)
Vector division .
subroutine field_add4(a, b, c, d, n)
Vector addition .
subroutine, public field_cadd(a, s, n)
Add a scalar to vector .
subroutine, public field_invcol1(a, n)
Invert a vector .
subroutine, public field_col2(a, b, n)
Vector multiplication .
subroutine, public field_sub2(a, b, n)
Vector substraction .
subroutine, public field_sub3(a, b, c, n)
Vector subtraction .
subroutine, public field_cmult2(a, b, c, n)
Multiplication by constant c .
real(kind=rp) function, public field_glsc2(a, b, n)
subroutine, public field_cfill(a, c, n)
Set all elements to a constant c .
subroutine, public field_add2s2(a, b, c1, n)
Vector addition with scalar multiplication (multiplication on second argument)
real(kind=rp) function, public field_glsum(a, n)
subroutine, public field_rzero(a, n)
Zero a real vector.
subroutine, public field_addcol3(a, b, c, n)
Returns .
real(kind=rp) function, public field_glsubnorm(a, b, n)
subroutine, public field_rone(a, n)
Set all elements to one.
subroutine, public field_invcol3(a, b, c, n)
Invert a vector .
subroutine, public field_add3s2(a, b, c, c1, c2, n)
Returns .
subroutine, public field_add2(a, b, n)
Vector addition .
subroutine, public field_cpwmin2(a, b, n)
Point-wise min operation for field and constant .
subroutine, public field_pwmax2(a, b, n)
Point-wise max operation .
subroutine, public field_pwmin3(a, b, c, n)
Point-wise min operation .
subroutine, public field_col3(a, b, c, n)
Vector multiplication with 3 vectors .
subroutine, public field_copy(a, b, n)
Copy a vector .
subroutine, public field_subcol3(a, b, c, n)
Returns .
subroutine, public field_add3(a, b, c, n)
Vector addition .
subroutine, public field_pwmax3(a, b, c, n)
Point-wise max operation .
subroutine, public field_cpwmax3(a, b, c, n)
Point-wise max operation for field and constant .
subroutine, public field_cmult(a, c, n)
Multiplication by constant c .
real(kind=rp) function, public field_glsc3(a, b, c, n)
subroutine, public field_addcol4(a, b, c, d, n)
Returns .
subroutine, public field_addsqr2s2(a, b, c1, n)
Returns .
subroutine, public field_vdot3(dot, u1, u2, u3, v1, v2, v3, n)
Compute a dot product (3-d version) assuming vector components etc.
subroutine, public field_pwmin2(a, b, n)
Point-wise min operation .
subroutine, public field_masked_scatter_copy_0(a, b, mask, n, n_mask)
Gather a contigous array into a field .
subroutine, public field_masked_gather_copy_0(a, b, mask, n, n_mask)
Gather a field to reduced contigous array .
Object for handling masks in Neko.
subroutine, public cmult(a, c, n)
Multiplication by constant c .
subroutine, public cmult2(a, b, c, n)
Multiplication by constant c .
subroutine, public invcol2(a, b, n)
Vector division .
real(kind=rp) function, public glsc3(a, b, c, n)
Weighted inner product .
subroutine, public cadd(a, s, n)
Add a scalar to vector .
subroutine, public addsqr2s2(a, b, c1, n)
Returns .
subroutine, public add2s1(a, b, c1, n)
Vector addition with scalar multiplication (multiplication on first argument)
subroutine, public cpwmin2(a, b, n)
Point-wise minimum of scalar and vector .
real(kind=rp) function, public glsc2(a, b, n)
Weighted inner product .
subroutine, public masked_scatter_copy_0(a, b, mask, n, n_mask)
Scatter a contigous vector to masked positions in a target array .
subroutine, public subcol3(a, b, c, n)
Returns .
subroutine, public rone(a, n)
Set all elements to one.
subroutine, public cpwmin3(a, b, c, n)
Point-wise minimum of scalar and vector .
subroutine, public pwmax3(a, b, c, n)
Point-wise maximum of two vectors .
subroutine, public add3(a, b, c, n)
Vector addition .
real(kind=rp) function, public glsum(a, n)
Sum a vector of length n.
subroutine, public sub3(a, b, c, n)
Vector subtraction .
subroutine, public addcol4(a, b, c, d, n)
Returns .
subroutine, public add2(a, b, n)
Vector addition .
subroutine, public cfill(a, c, n)
Set all elements to a constant c .
subroutine, public invcol3(a, b, c, n)
Invert a vector .
subroutine, public add3s2(a, b, c, c1, c2, n)
Returns .
subroutine, public pwmax2(a, b, n)
Point-wise maximum of two vectors .
subroutine, public pwmin2(a, b, n)
Point-wise minimum of two vectors .
subroutine, public masked_gather_copy_0(a, b, mask, n, n_mask)
Gather a masked vector to reduced contigous vector .
subroutine, public addcol3(a, b, c, n)
Returns .
subroutine, public invcol1(a, n)
Invert a vector .
subroutine, public cpwmax3(a, b, c, n)
Point-wise maximum of scalar and vector .
subroutine, public col2(a, b, n)
Vector multiplication .
subroutine, public copy(a, b, n)
Copy a vector .
subroutine, public add4(a, b, c, d, n)
Vector addition .
subroutine, public col3(a, b, c, n)
Vector multiplication with 3 vectors .
subroutine, public vdot3(dot, u1, u2, u3, v1, v2, v3, n)
Compute a dot product (3-d version) assuming vector components etc.
subroutine, public rzero(a, n)
Zero a real vector.
real(kind=rp) function, public glsubnorm(a, b, n)
Returns the norm of the difference of two vectors .
subroutine, public cpwmax2(a, b, n)
Point-wise maximum of scalar and vector .
subroutine, public sub2(a, b, n)
Vector substraction .
subroutine, public pwmin3(a, b, c, n)
Point-wise minimum of two vectors .
subroutine, public add2s2(a, b, c1, n)
Vector addition with scalar multiplication (multiplication on second argument)
integer, parameter neko_bcknd_device
integer, parameter, public rp
Global precision used in computations.