Skip to content

GitLab

Commit 3a0fb34f authored by Andreas Marek's avatar Andreas Marek
Browse files

C autotune test programs

parent 997cee12
Hide whitespace changes
Inline Side-by-side
Showing with 308 additions and 10 deletions
generate_automake_test_programs.py
View file @ 3a0fb34f
......@@ -173,7 +173,7 @@ for lang, m, g, q, t, p, d, s, lay in product(sorted(language_flag.keys()),
print("endif\n" * endifs)
for p, d in product(sorted(prec_flag.keys()), sorted(domain_flag.keys())):
for lang, p, d in product(sorted(language_flag.keys()), sorted(prec_flag.keys()), sorted(domain_flag.keys())):
endifs = 0
if (p == "single"):
if (d == "real"):
......@@ -184,12 +184,20 @@ for p, d in product(sorted(prec_flag.keys()), sorted(domain_flag.keys())):
raise Exception("Oh no!")
endifs += 1
name = "test_autotune_{0}_{1}".format(d, p)
name = "test_autotune{langsuffix}_{d}_{p}".format(langsuffix=language_flag[lang], d=d, p=p)
print("check_SCRIPTS += " + name + ".sh")
print("noinst_PROGRAMS += " + name)
print(name + "_SOURCES = test/Fortran/test_autotune.F90")
print(name + "_LDADD = $(test_program_ldadd)")
print(name + "_FCFLAGS = $(test_program_fcflags) \\")
if lang == "Fortran":
print(name + "_SOURCES = test/Fortran/test_autotune.F90")
print(name + "_LDADD = $(test_program_ldadd)")
print(name + "_FCFLAGS = $(test_program_fcflags) \\")
elif lang == "C":
print(name + "_SOURCES = test/C/test_autotune.c")
print(name + "_LDADD = $(test_program_ldadd) $(FCLIBS)")
print(name + "_CFLAGS = $(test_program_fcflags) \\")
print(" " + " \\\n ".join([
domain_flag[d],
prec_flag[p]]))
......
legacy_test_programs.am
View file @ 3a0fb34f
noinst_PROGRAMS += \
autotune_c_version@SUFFIX@ \
legacy_real_1stage@SUFFIX@ \
legacy_complex_1stage@SUFFIX@ \
legacy_real_2stage@SUFFIX@ \
......@@ -73,10 +72,6 @@ noinst_PROGRAMS += \
endif
endif
autotune_c_version@SUFFIX@_SOURCES = test/C/autotune_c_version.c
autotune_c_version@SUFFIX@_LDADD = $(test_program_ldadd) $(FCLIBS)
autotune_c_version@SUFFIX@_FCFLAGS = $(test_program_fcflags)
legacy_real_1stage_c_version@SUFFIX@_SOURCES = test/C/elpa1/legacy_interface/legacy_real_1stage_c_version.c
legacy_real_1stage_c_version@SUFFIX@_LDADD = $(test_program_ldadd) $(FCLIBS)
legacy_real_1stage_c_version@SUFFIX@_FCFLAGS = $(test_program_fcflags)
......
test/C/autotune_c_version.c → test/C/test_autotune.c
View file @ 3a0fb34f
/* This file is part of ELPA. */
/* */
/* The ELPA library was originally created by the ELPA consortium, */
/* consisting of the following organizations: */
/* */
/* - Max Planck Computing and Data Facility (MPCDF), formerly known as */
/* Rechenzentrum Garching der Max-Planck-Gesellschaft (RZG), */
/* - Bergische Universität Wuppertal, Lehrstuhl für angewandte */
/* Informatik, */
/* - Technische Universität München, Lehrstuhl für Informatik mit */
/* Schwerpunkt Wissenschaftliches Rechnen , */
/* - Fritz-Haber-Institut, Berlin, Abt. Theorie, */
/* - Max-Plack-Institut für Mathematik in den Naturwissenschaften, */
/* Leipzig, Abt. Komplexe Strukutren in Biologie und Kognition, */
/* and */
/* - IBM Deutschland GmbH */
/* */
/* */
/* More information can be found here: */
/* http://elpa.mpcdf.mpg.de/ */
/* */
/* ELPA is free software: you can redistribute it and/or modify */
/* it under the terms of the version 3 of the license of the */
/* GNU Lesser General Public License as published by the Free */
/* Software Foundation. */
/* */
/* ELPA is distributed in the hope that it will be useful, */
/* but WITHOUT ANY WARRANTY; without even the implied warranty of */
/* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the */
/* GNU Lesser General Public License for more details. */
/* */
/* You should have received a copy of the GNU Lesser General Public License */
/* along with ELPA. If not, see <http://www.gnu.org/licenses/> */
/* */
/* ELPA reflects a substantial effort on the part of the original */
/* ELPA consortium, and we ask you to respect the spirit of the */
/* license that we chose: i.e., please contribute any changes you */
/* may have back to the original ELPA library distribution, and keep */
/* any derivatives of ELPA under the same license that we chose for */
/* the original distribution, the GNU Lesser General Public License. */
/* */
/* */
#include "config-f90.h"
/* This file is part of ELPA.
The ELPA library was originally created by the ELPA consortium,
consisting of the following organizations:
- Max Planck Computing and Data Facility (MPCDF), formerly known as
Rechenzentrum Garching der Max-Planck-Gesellschaft (RZG),
- Bergische Universität Wuppertal, Lehrstuhl für angewandte
Informatik,
- Technische Universität München, Lehrstuhl für Informatik mit
Schwerpunkt Wissenschaftliches Rechnen ,
- Fritz-Haber-Institut, Berlin, Abt. Theorie,
- Max-Plack-Institut für Mathematik in den Naturwissenschaften,
Leipzig, Abt. Komplexe Strukutren in Biologie und Kognition,
and
- IBM Deutschland GmbH
More information can be found here:
http://elpa.mpcdf.mpg.de/
ELPA is free software: you can redistribute it and/or modify
it under the terms of the version 3 of the license of the
GNU Lesser General Public License as published by the Free
Software Foundation.
ELPA is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License
along with ELPA. If not, see <http://www.gnu.org/licenses/>
ELPA reflects a substantial effort on the part of the original
ELPA consortium, and we ask you to respect the spirit of the
license that we chose: i.e., please contribute any changes you
may have back to the original ELPA library distribution, and keep
any derivatives of ELPA under the same license that we chose for
the original distribution, the GNU Lesser General Public License.
*/
#include "config.h"
#include <string.h>
#include <stdio.h>
#include <stdlib.h>
#ifdef WITH_MPI
......@@ -53,143 +53,120 @@
#include <elpa/elpa.h>
#include <assert.h>
#include <string.h>
#include <test/shared/generated.h>
#include "test/shared/generated.h"
#define DOUBLE_PRECISION_REAL 1
#if !(defined(TEST_REAL) ^ defined(TEST_COMPLEX))
//#error "define exactly one of TEST_REAL or TEST_COMPLEX"
#endif
#define assert_elpa_ok(x) assert(x == ELPA_OK)
#if !(defined(TEST_SINGLE) ^ defined(TEST_DOUBLE))
//#error "define exactly one of TEST_SINGLE or TEST_DOUBLE"
#endif
int main(int argc, char** argv) {
int myid;
int nprocs;
#ifndef WITH_MPI
int MPI_COMM_WORLD;
#if !(defined(TEST_SOLVER_1STAGE) ^ defined(TEST_SOLVER_2STAGE))
//#error "define exactly one of TEST_SOLVER_1STAGE or TEST_SOLVER_2STAGE"
#endif
int na, nev, nblk;
int status;
#ifdef TEST_SINGLE
# define EV_TYPE float
# ifdef TEST_REAL
# define MATRIX_TYPE float
# else
# define MATRIX_TYPE complex float
# endif
#else
# define EV_TYPE double
# ifdef TEST_REAL
# define MATRIX_TYPE double
# else
# define MATRIX_TYPE complex double
# endif
#endif
int np_cols, np_rows, np_colsStart;
#define assert_elpa_ok(x) assert(x == ELPA_OK)
int my_blacs_ctxt, my_prow, my_pcol;
int mpierr;
int main(int argc, char** argv) {
/* matrix dimensions */
int na, nev, nblk;
int my_mpi_comm_world;
int mpi_comm_rows, mpi_comm_cols;
/* mpi */
int myid, nprocs;
int na_cols, na_rows;
int np_cols, np_rows;
int my_prow, my_pcol;
int mpi_comm;
int info, *sc_desc;
/* blacs */
int my_blacs_ctxt, sc_desc[9], info;
int na_rows, na_cols;
double startVal;
#ifdef DOUBLE_PRECISION_REAL
double *a, *z, *as, *ev;
#else
float *a, *z, *as, *ev;
#endif
/* The Matrix */
MATRIX_TYPE *a, *as, *z;
EV_TYPE *ev;
int success;
int error, status;
elpa_t handle;
elpa_autotune_t autotune_handle;
int value, error, unfinished, i;
int i, unfinished;
int value;
#ifdef WITH_MPI
MPI_Init(&argc, &argv);
MPI_Comm_size(MPI_COMM_WORLD, &nprocs);
MPI_Comm_rank(MPI_COMM_WORLD, &myid);
#else
nprocs = 1;
myid=0;
MPI_COMM_WORLD=1;
myid = 0;
#endif
na = 100;
nev = 50;
nblk = 16;
if (myid == 0) {
printf("This is the c version of an ELPA test-programm\n");
printf("\n");
printf("It will call the 2stage ELPA real solver for an\n");
printf("matrix of size %d. It will compute %d eigenvalues\n",na,nev);
printf("and uses a blocksize of %d\n",nblk);
printf("\n");
printf("This is an example program with much less functionality\n");
printf("as it's Fortran counterpart. It's only purpose is to show how \n");
printf("to evoke ELPA2 from a c programm\n");
printf("\n");
#ifdef DOUBLE_PRECISION_REAL
printf(" Double precision version of ELPA2 is used. \n");
#else
printf(" Single precision version of ELPA2 is used. \n");
#endif
if (argc == 4) {
na = atoi(argv[1]);
nev = atoi(argv[2]);
nblk = atoi(argv[3]);
} else {
na = 1000;
nev = 500;
nblk = 16;
}
status = 0;
startVal = sqrt((double) nprocs);
np_colsStart = (int) round(startVal);
for (np_cols=np_colsStart;np_cols>1;np_cols--){
if (nprocs %np_cols ==0){
break;
for (np_cols = (int) sqrt((double) nprocs); np_cols > 1; np_cols--) {
if (nprocs % np_cols == 0) {
break;
}
}
np_rows = nprocs/np_cols;
if (myid == 0) {
printf("\n");
printf("Number of processor rows %d, cols %d, total %d \n",np_rows,np_cols,nprocs);
}
/* set up blacs */
/* convert communicators before */
#ifdef WITH_MPI
my_mpi_comm_world = MPI_Comm_c2f(MPI_COMM_WORLD);
mpi_comm = MPI_Comm_c2f(MPI_COMM_WORLD);
#else
my_mpi_comm_world = 1;
mpi_comm = 0;
#endif
set_up_blacsgrid_f(my_mpi_comm_world, np_rows, np_cols, 'C', &my_blacs_ctxt, &my_prow, &my_pcol);
if (myid == 0) {
printf("\n");
printf("Past BLACS_Gridinfo...\n");
printf("\n");
}
sc_desc = malloc(9*sizeof(int));
set_up_blacsgrid_f(mpi_comm, np_rows, np_cols, 'C', &my_blacs_ctxt, &my_prow, &my_pcol);
set_up_blacs_descriptor_f(na, nblk, my_prow, my_pcol, np_rows, np_cols, &na_rows, &na_cols, sc_desc, my_blacs_ctxt, &info);
if (myid == 0) {
printf("\n");
printf("Past scalapack descriptor setup...\n");
printf("\n");
}
/* allocate the matrices needed for elpa */
if (myid == 0) {
printf("\n");
printf("Allocating matrices with na_rows=%d and na_cols=%d\n",na_rows, na_cols);
printf("\n");
}
#ifdef DOUBLE_PRECISION_REAL
a = malloc(na_rows*na_cols*sizeof(double));
z = malloc(na_rows*na_cols*sizeof(double));
as = malloc(na_rows*na_cols*sizeof(double));
ev = malloc(na*sizeof(double));
#else
a = malloc(na_rows*na_cols*sizeof(float));
z = malloc(na_rows*na_cols*sizeof(float));
as = malloc(na_rows*na_cols*sizeof(float));
ev = malloc(na*sizeof(float));
#endif
#ifdef DOUBLE_PRECISION_REAL
a = calloc(na_rows*na_cols, sizeof(MATRIX_TYPE));
z = calloc(na_rows*na_cols, sizeof(MATRIX_TYPE));
as = calloc(na_rows*na_cols, sizeof(MATRIX_TYPE));
ev = calloc(na, sizeof(EV_TYPE));
#ifdef TEST_REAL
#ifdef TEST_DOUBLE
prepare_matrix_random_real_double_f(na, myid, na_rows, na_cols, sc_desc, a, z, as);
#else
prepare_matrix_random_real_single_f(na, myid, na_rows, na_cols, sc_desc, a, z, as);
#endif
#else
#ifdef TEST_DOUBLE
prepare_matrix_random_complex_double_f(na, myid, na_rows, na_cols, sc_desc, a, z, as);
#else
prepare_matrix_random_complex_single_f(na, myid, na_rows, na_cols, sc_desc, a, z, as);
#endif
#endif
if (elpa_init(CURRENT_API_VERSION) != ELPA_OK) {
......@@ -207,6 +184,9 @@ int main(int argc, char** argv) {
elpa_set(handle, "nev", nev, &error);
assert_elpa_ok(error);
if (myid == 0) {
printf("Setting the matrix parameters na=%d, nev=%d \n",na,nev);
}
elpa_set(handle, "local_nrows", na_rows, &error);
assert_elpa_ok(error);
......@@ -230,16 +210,11 @@ int main(int argc, char** argv) {
/* Setup */
assert_elpa_ok(elpa_setup(handle));
/* Set tunables */
elpa_set(handle, "gpu", 0, &error);
assert_elpa_ok(error);
#ifdef WITH_MPI
mpierr = MPI_Barrier(MPI_COMM_WORLD);
#endif
autotune_handle = elpa_autotune_setup(handle, ELPA_AUTOTUNE_FAST, ELPA_AUTOTUNE_DOMAIN_REAL);
/* mimic 10 scf steps */
/* mimic 20 scf steps */
for (i=0; i < 20; i++) {
......@@ -256,12 +231,23 @@ int main(int argc, char** argv) {
assert_elpa_ok(error);
/* check the results */
#ifdef DOUBLE_PRECISION_REAL
#ifdef TEST_REAL
#ifdef TEST_DOUBLE
status = check_correctness_evp_numeric_residuals_real_double_f(na, nev, na_rows, na_cols, as, z, ev, sc_desc, myid);
memcpy(a, as, na_rows*na_cols*sizeof(double));
#else
status = check_correctness_evp_numeric_residuals_real_single_f(na, nev, na_rows, na_cols, as, z, ev, sc_desc, myid);
memcpy(a, as, na_rows*na_cols*sizeof(float));
#endif
#else
#ifdef TEST_DOUBLE
status = check_correctness_evp_numeric_residuals_complex_double_f(na, nev, na_rows, na_cols, as, z, ev, sc_desc, myid);
memcpy(a, as, na_rows*na_cols*sizeof(complex double));
#else
status = check_correctness_evp_numeric_residuals_complex_single_f(na, nev, na_rows, na_cols, as, z, ev, sc_desc, myid);
memcpy(a, as, na_rows*na_cols*sizeof(complex float));
#endif
#endif
if (status !=0){
......@@ -296,7 +282,6 @@ int main(int argc, char** argv) {
}
}
free(sc_desc);
free(a);
free(z);
free(as);
......@@ -305,5 +290,6 @@ int main(int argc, char** argv) {
#ifdef WITH_MPI
MPI_Finalize();
#endif
return 0;
return !!status;
}
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment