1solve_evp_complex(3) Library Functions Manual solve_evp_complex(3)
2
3
4
6 solve_evp_complex - solve the double-precision complex eigenvalue
7 problem with the 1-stage ELPA solver. This interface is old and
8 deprecated. It is recommended to use solve_evp_complex_1stage_double(3)
9
10
12 FORTRAN INTERFACE
13 use elpa1
14 success = solve_evp_complex (na, nev, a(lda,matrixCols), ev(nev),
15 q(ldq, matrixCols), ldq, nblk, matrixCols, mpi_comm_rows,
16 mpi_comm_cols, mpi_comm_all, useGPU)
17
18 With the definitions of the input and output variables:
19
20 integer, intent(in) na: global dimension of quadratic
21 matrix a to solve
22 integer, intent(in) nev: number of eigenvalues to be
23 computed; the first nev eigenvalules are calculated
24 complex*16, intent(inout) a: locally distributed part
25 of the matrix a. The local dimensions are lda x matrixCols
26 integer, intent(in) lda: leading dimension of
27 locally distributed matrix a
28 real*8, intent(inout) ev: on output the first nev
29 computed eigenvalues
30 complex*16, intent(inout) q: on output the first nev
31 computed eigenvectors
32 integer, intent(in) ldq: leading dimension of
33 matrix q which stores the eigenvectors
34 integer, intent(in) nblk: blocksize of block cyclic
35 distributin, must be the same in both directions
36 integer, intent(in) matrixCols: number of columns of
37 locally distributed matrices a and q
38 integer, intent(in) mpi_comm_rows: communicator for
39 communication in rows. Constructed with elpa_get_communicators(3)
40 integer, intent(in) mpi_comm_cols: communicator for
41 communication in colums. Constructed with elpa_get_communicators(3)
42 integer, intent(in) mpi_comm_all: communicator for all MPI
43 process used in ELPA
44 logical, optional, intent(in) useGPU: decide whether GPUs should
45 be used or not
46
47 logical success: return value indicating
48 success or failure
49
51 Old, deprecated interface, which will be deleted at some point. Use
52 solve_evp_complex_1stage(3) or elpa_solve_evp_complex(3). Solve the
53 complex eigenvalue problem with the 1-stage solver. The ELPA
54 communicators mpi_comm_rows and mpi_comm_cols are obtained with the
55 elpa_get_communicators(3) function. The distributed quadratic marix a
56 has global dimensions na x na, and a local size lda x matrixCols. The
57 solver will compute the first nev eigenvalues, which will be stored on
58 exit in ev. The eigenvectors corresponding to the eigenvalues will be
59 stored in q. All memory of the arguments must be allocated outside the
60 call to the solver.
61
63 elpa_get_communicators(3) elpa_solve_evp_real_double(3)
64 elpa_solve_evp_real_single(3) elpa_solve_evp_complex_double(3)
65 elpa_solve_evp_complex_single(3) elpa_solve_evp_real_1stage_double(3)
66 elpa_solve_evp_real_1stage_single(3)
67 elpa_solve_evp_complex_1stage_single(3)
68 elpa_solve_evp_real_2stage_double(3)
69 elpa_solve_evp_real_2stage_single(3)
70 elpa_solve_evp_complex_2stage_double(3)
71 elpa_solve_evp_complex_2stage_single(3) elpa2_print_kernels(1)
72
73
74
75ELPA Thu Mar 17 2016 solve_evp_complex(3)