MPI-AMRVAC 3.1
The MPI - Adaptive Mesh Refinement - Versatile Advection Code (development version)
Loading...
Searching...
No Matches
mod_comm_lib.t
Go to the documentation of this file.
1module mod_comm_lib
2
3 implicit none
4 private
5
6
7 public :: comm_start
8 public :: comm_finalize
9 public :: init_comm_types
10 public :: mpistop
11
12contains
13
14
15 !> Initialize the MPI environment
16 subroutine comm_start
17 use mod_global_parameters
18
19 integer(kind=MPI_ADDRESS_KIND) :: lb
20 integer(kind=MPI_ADDRESS_KIND) :: sizes
21
22 ! Initialize MPI
23 call mpi_init(ierrmpi)
24
25 ! Each process stores its rank, which ranges from 0 to N-1, where N is the
26 ! number of processes.
27 call mpi_comm_rank(mpi_comm_world,mype,ierrmpi)
28
29 ! Store the number of processes
30 call mpi_comm_size(mpi_comm_world,npe,ierrmpi)
31
32 ! Use the default communicator, which contains all the processes
33 icomm = mpi_comm_world
34
35 ! Get size of double/integer
36 call mpi_type_get_extent(mpi_real,lb,sizes,ierrmpi)
37 if (sizes /= size_real) call mpistop("Incompatible real size")
38 call mpi_type_get_extent(mpi_double_precision,lb,sizes,ierrmpi)
39 if (sizes /= size_double) call mpistop("Incompatible double size")
40 call mpi_type_get_extent(mpi_integer,lb,sizes,ierrmpi)
41 if (sizes /= size_int) call mpistop("Incompatible integer size")
42 call mpi_type_get_extent(mpi_logical,lb,sizes,ierrmpi)
43 if (sizes /= size_logical) call mpistop("Incompatible logical size")
44
45 end subroutine comm_start
46
47 !> Finalize (or shutdown) the MPI environment
48 subroutine comm_finalize
49 use mod_global_parameters
50
51 call mpi_barrier(mpi_comm_world,ierrmpi)
52 call mpi_finalize(ierrmpi)
53
54 end subroutine comm_finalize
55
56 !> Create and store the MPI types that will be used for parallel communication
57 subroutine init_comm_types
58 use mod_global_parameters
59
60 integer, dimension(ndim+1) :: sizes, subsizes, start
61 integer :: i^d, ic^d, nx^d, nxco^d, nxg^d, idir
62
63 nx^d=ixmhi^d-ixmlo^d+1;
64 nxg^d=ixghi^d-ixglo^d+1;
65 nxco^d=nx^d/2;
66
67 ^d&sizes(^d)=ixghi^d;
68 sizes(ndim+1)=nw
69 ^d&subsizes(^d)=nxg^d;
70 subsizes(ndim+1)=nw
71 ^d&start(^d)=ixglo^d-1;
72 start(ndim+1)=0
73 call mpi_type_create_subarray(ndim+1,sizes,subsizes,start, &
74 mpi_order_fortran,mpi_double_precision, &
75 type_block,ierrmpi)
76 call mpi_type_commit(type_block,ierrmpi)
77 size_block={nxg^d*}*nw*size_double
78
79 ^d&sizes(^d)=ixghi^d/2+nghostcells;
80 sizes(ndim+1)=nw
81 ^d&subsizes(^d)=nxco^d;
82 subsizes(ndim+1)=nw
83 ^d&start(^d)=ixmlo^d-1;
84 start(ndim+1)=0
85 call mpi_type_create_subarray(ndim+1,sizes,subsizes,start, &
86 mpi_order_fortran,mpi_double_precision, &
87 type_coarse_block,ierrmpi)
88 call mpi_type_commit(type_coarse_block,ierrmpi)
89
90 if(stagger_grid) then
91 ^d&sizes(^d)=ixghi^d+1;
92 sizes(ndim+1)=nws
93 ^d&subsizes(^d)=nx^d+1;
94 subsizes(ndim+1)=nws
95 ^d&start(^d)=ixmlo^d-1;
96 start(ndim+1)=0
97 call mpi_type_create_subarray(ndim+1,sizes,subsizes,start, &
98 mpi_order_fortran,mpi_double_precision, &
99 type_block_io_stg,ierrmpi)
100 call mpi_type_commit(type_block_io_stg,ierrmpi)
101 size_block_io_stg={(nx^d+1)*}*nws*size_double
102
103 ^d&sizes(^d)=ixghi^d/2+nghostcells+1;
104 sizes(ndim+1)=nws
105 {do ic^db=1,2\}
106 do idir=1,ndim
107 ^d&subsizes(^d)=nxco^d+kr(ic^d,1)*kr(idir,^d);
108 subsizes(ndim+1)=1
109 ^d&start(^d)=ixmlo^d-kr(ic^d,1)*kr(idir,^d);
110 start(ndim+1)=idir-1
111
112 call mpi_type_create_subarray(ndim+1,sizes,subsizes,start, &
113 mpi_order_fortran,mpi_double_precision, &
114 type_coarse_block_stg(idir,ic^d),ierrmpi)
115 call mpi_type_commit(type_coarse_block_stg(idir,ic^d),ierrmpi)
116 end do
117 {end do\}
118
119 ^d&sizes(^d)=ixghi^d+1;
120 sizes(ndim+1)=nws
121 {do ic^db=1,2\}
122 do idir=1,^nd
123 ^d&subsizes(^d)=nxco^d+kr(ic^d,1)*kr(idir,^d);
124 subsizes(ndim+1)=1
125 ^d&start(^d)=ixmlo^d-kr(ic^d,1)*kr(idir,^d)+(ic^d-1)*nxco^d;
126 start(ndim+1)=idir-1
127 call mpi_type_create_subarray(ndim+1,sizes,subsizes,start, &
128 mpi_order_fortran,mpi_double_precision, &
129 type_sub_block_stg(idir,ic^d),ierrmpi)
130 call mpi_type_commit(type_sub_block_stg(idir,ic^d),ierrmpi)
131 end do
132 {end do\}
133 end if
134
135 ^d&sizes(^d)=ixghi^d;
136 sizes(ndim+1)=nw
137 {do ic^db=1,2\}
138 ^d&subsizes(^d)=nxco^d;
139 subsizes(ndim+1)=nw
140 ^d&start(^d)=ixmlo^d-1+(ic^d-1)*nxco^d;
141 start(ndim+1)=0
142 call mpi_type_create_subarray(ndim+1,sizes,subsizes,start, &
143 mpi_order_fortran,mpi_double_precision, &
144 type_sub_block(ic^d),ierrmpi)
145 call mpi_type_commit(type_sub_block(ic^d),ierrmpi)
146 {end do\}
147
148 ^d&sizes(^d)=ixghi^d;
149 sizes(ndim+1)=nw
150 ^d&subsizes(^d)=nx^d;
151 subsizes(ndim+1)=nw
152 ^d&start(^d)=ixmlo^d-1;
153 start(ndim+1)=0
154 call mpi_type_create_subarray(ndim+1,sizes,subsizes,start, &
155 mpi_order_fortran,mpi_double_precision, &
156 type_block_io,ierrmpi)
157 call mpi_type_commit(type_block_io,ierrmpi)
158 size_block_io={nx^d*}*nw*size_double
159
160 ^d&sizes(^d)=ixmhi^d-ixmlo^d+1;
161 sizes(ndim+1)=^nd
162 ^d&subsizes(^d)=sizes(^d);
163 subsizes(ndim+1)=^nd
164 ^d&start(^d)=0;
165 start(ndim+1)=0
166 call mpi_type_create_subarray(ndim+1,sizes,subsizes,start, &
167 mpi_order_fortran,mpi_double_precision, &
168 type_block_xcc_io,ierrmpi)
169 call mpi_type_commit(type_block_xcc_io,ierrmpi)
170
171 ^d&sizes(^d)=ixmhi^d-ixmlo^d+2;
172 sizes(ndim+1)=^nd
173 ^d&subsizes(^d)=sizes(^d);
174 subsizes(ndim+1)=^nd
175 ^d&start(^d)=0;
176 start(ndim+1)=0
177 call mpi_type_create_subarray(ndim+1,sizes,subsizes,start, &
178 mpi_order_fortran,mpi_double_precision, &
179 type_block_xc_io,ierrmpi)
180 call mpi_type_commit(type_block_xc_io,ierrmpi)
181
182 ^d&sizes(^d)=ixmhi^d-ixmlo^d+1;
183 sizes(ndim+1)=nw+nwauxio
184 ^d&subsizes(^d)=sizes(^d);
185 subsizes(ndim+1)=nw+nwauxio
186 ^d&start(^d)=0;
187 start(ndim+1)=0
188 call mpi_type_create_subarray(ndim+1,sizes,subsizes,start, &
189 mpi_order_fortran,mpi_double_precision, &
190 type_block_wcc_io,ierrmpi)
191 call mpi_type_commit(type_block_wcc_io,ierrmpi)
192
193 ^d&sizes(^d)=ixmhi^d-ixmlo^d+2;
194 sizes(ndim+1)=nw+nwauxio
195 ^d&subsizes(^d)=sizes(^d);
196 subsizes(ndim+1)=nw+nwauxio
197 ^d&start(^d)=0;
198 start(ndim+1)=0
199 call mpi_type_create_subarray(ndim+1,sizes,subsizes,start, &
200 mpi_order_fortran,mpi_double_precision, &
201 type_block_wc_io,ierrmpi)
202 call mpi_type_commit(type_block_wc_io,ierrmpi)
203
204 end subroutine init_comm_types
205
206 !> Exit MPI-AMRVAC with an error message
207 subroutine mpistop(message)
208 use mod_global_parameters
209
210 character(len=*), intent(in) :: message !< The error message
211 integer :: ierrcode
212
213 write(*, *) "ERROR for processor", mype, ":"
214 write(*, *) trim(message)
215
216 call mpi_abort(icomm, ierrcode, ierrmpi)
217
218 end subroutine mpistop
219
220end module mod_comm_lib
mod_comm_lib
Definition mod_comm_lib.t:1
mod_comm_lib::init_comm_types
subroutine, public init_comm_types
Create and store the MPI types that will be used for parallel communication.
Definition mod_comm_lib.t:58
mod_comm_lib::comm_start
subroutine, public comm_start
Initialize the MPI environment.
Definition mod_comm_lib.t:17
mod_comm_lib::comm_finalize
subroutine, public comm_finalize
Finalize (or shutdown) the MPI environment.
Definition mod_comm_lib.t:49
mod_comm_lib::mpistop
subroutine, public mpistop(message)
Exit MPI-AMRVAC with an error message.
Definition mod_comm_lib.t:208
mod_global_parameters
This module contains definitions of global parameters and variables and some generic functions/subrou...
Definition mod_global_parameters.t:4
mod_global_parameters::type_coarse_block
integer type_coarse_block
MPI type for block coarsened by 2, and for its children blocks.
Definition mod_global_parameters.t:236
mod_global_parameters::ixghi
integer ixghi
Upper index of grid block arrays.
Definition mod_global_parameters.t:281
mod_global_parameters::kr
integer, dimension(3, 3) kr
Kronecker delta tensor.
Definition mod_global_parameters.t:426
mod_global_parameters::ndim
integer, parameter ndim
Number of spatial dimensions for grid variables.
Definition mod_global_parameters.t:217
mod_global_parameters::stagger_grid
logical stagger_grid
True for using stagger grid.
Definition mod_global_parameters.t:620
mod_global_parameters::type_coarse_block_stg
integer, dimension(^nd, 2^d &) type_coarse_block_stg
MPI type for staggered block coarsened by 2, and for its children blocks.
Definition mod_global_parameters.t:238
mod_global_parameters::icomm
integer icomm
The MPI communicator.
Definition mod_global_parameters.t:226
mod_global_parameters::mype
integer mype
The rank of the current MPI task.
Definition mod_global_parameters.t:223
mod_global_parameters::d
double precision, dimension(:), allocatable, parameter d
Definition mod_global_parameters.t:23
mod_global_parameters::ierrmpi
integer ierrmpi
A global MPI error return code.
Definition mod_global_parameters.t:229
mod_global_parameters::size_block
integer size_block
Definition mod_global_parameters.t:234
mod_global_parameters::npe
integer npe
The number of MPI tasks.
Definition mod_global_parameters.t:220
mod_global_parameters::type_block
integer type_block
MPI type for block including ghost cells and its size.
Definition mod_global_parameters.t:234
mod_global_parameters::nghostcells
integer nghostcells
Number of ghost cells surrounding a grid.
Definition mod_global_parameters.t:290
mod_global_parameters::type_block_io_stg
integer type_block_io_stg
MPI type for IO of staggered variables.
Definition mod_global_parameters.t:242
mod_global_parameters::size_block_io_stg
integer size_block_io_stg
Definition mod_global_parameters.t:242
mod_global_parameters::ixglo
integer, parameter ixglo
Lower index of grid block arrays (always 1)
Definition mod_global_parameters.t:278
Generated by doxygen 1.9.8