Go to the documentation of this file.
00001 *> \brief \b DCHKRFP
00002 *
00003 *  =========== DOCUMENTATION ===========
00004 *
00005 * Online html documentation available at 
00006 *            http://www.netlib.org/lapack/explore-html/ 
00007 *
00008 *  Definition:
00009 *  ===========
00010 *
00011 *       PROGRAM DCHKRFP
00012 * 
00013 *
00014 *> \par Purpose:
00015 *  =============
00016 *>
00017 *> \verbatim
00018 *>
00019 *> DCHKRFP is the main test program for the DOUBLE PRECISION linear
00020 *> equation routines with RFP storage format
00021 *>
00022 *> \endverbatim
00023 *
00024 *  Arguments:
00025 *  ==========
00026 *
00027 *> \verbatim
00028 *>  MAXIN   INTEGER
00029 *>          The number of different values that can be used for each of
00030 *>          M, N, or NB
00031 *>
00032 *>  MAXRHS  INTEGER
00033 *>          The maximum number of right hand sides
00034 *>
00035 *>  NTYPES  INTEGER
00036 *>
00037 *>  NMAX    INTEGER
00038 *>          The maximum allowable value for N.
00039 *>
00040 *>  NIN     INTEGER
00041 *>          The unit number for input
00042 *>
00043 *>  NOUT    INTEGER
00044 *>          The unit number for output
00045 *> \endverbatim
00046 *
00047 *  Authors:
00048 *  ========
00049 *
00050 *> \author Univ. of Tennessee 
00051 *> \author Univ. of California Berkeley 
00052 *> \author Univ. of Colorado Denver 
00053 *> \author NAG Ltd. 
00054 *
00055 *> \date November 2011
00056 *
00057 *> \ingroup double_lin
00058 *
00059 *  =====================================================================      PROGRAM DCHKRFP
00060 *
00061 *  -- LAPACK test routine (version 3.4.0) --
00062 *  -- LAPACK is a software package provided by Univ. of Tennessee,    --
00063 *  -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
00064 *     November 2011
00065 *
00066 *  =====================================================================
00067 *
00068 *     .. Parameters ..
00069       INTEGER            MAXIN
00070       PARAMETER          ( MAXIN = 12 )
00071       INTEGER            NMAX
00072       PARAMETER          ( NMAX =  50 )
00073       INTEGER            MAXRHS
00074       PARAMETER          ( MAXRHS = 16 )
00075       INTEGER            NTYPES
00076       PARAMETER          ( NTYPES = 9 )
00077       INTEGER            NIN, NOUT
00078       PARAMETER          ( NIN = 5, NOUT = 6 )
00079 *     ..
00080 *     .. Local Scalars ..
00081       LOGICAL            FATAL, TSTERR
00082       INTEGER            VERS_MAJOR, VERS_MINOR, VERS_PATCH
00083       INTEGER            I, NN, NNS, NNT
00084       DOUBLE PRECISION   EPS, S1, S2, THRESH
00085 
00086 *     ..
00087 *     .. Local Arrays ..
00088       INTEGER            NVAL( MAXIN ), NSVAL( MAXIN ), NTVAL( NTYPES )
00089       DOUBLE PRECISION   WORKA( NMAX, NMAX )
00090       DOUBLE PRECISION   WORKASAV( NMAX, NMAX )
00091       DOUBLE PRECISION   WORKB( NMAX, MAXRHS )
00092       DOUBLE PRECISION   WORKXACT( NMAX, MAXRHS )
00093       DOUBLE PRECISION   WORKBSAV( NMAX, MAXRHS )
00094       DOUBLE PRECISION   WORKX( NMAX, MAXRHS )
00095       DOUBLE PRECISION   WORKAFAC( NMAX, NMAX )
00096       DOUBLE PRECISION   WORKAINV( NMAX, NMAX )
00097       DOUBLE PRECISION   WORKARF( (NMAX*(NMAX+1))/2 )
00098       DOUBLE PRECISION   WORKAP( (NMAX*(NMAX+1))/2 )
00099       DOUBLE PRECISION   WORKARFINV( (NMAX*(NMAX+1))/2 )
00100       DOUBLE PRECISION   D_WORK_DLATMS( 3 * NMAX )
00101       DOUBLE PRECISION   D_WORK_DPOT01( NMAX )
00102       DOUBLE PRECISION   D_TEMP_DPOT02( NMAX, MAXRHS )
00103       DOUBLE PRECISION   D_TEMP_DPOT03( NMAX, NMAX )
00104       DOUBLE PRECISION   D_WORK_DLANSY( NMAX )
00105       DOUBLE PRECISION   D_WORK_DPOT02( NMAX )
00106       DOUBLE PRECISION   D_WORK_DPOT03( NMAX )
00107 *     ..
00108 *     .. External Functions ..
00109       DOUBLE PRECISION   DLAMCH, DSECND
00110       EXTERNAL           DLAMCH, DSECND
00111 *     ..
00112 *     .. External Subroutines ..
00113       EXTERNAL           ILAVER, DDRVRFP, DDRVRF1, DDRVRF2, DDRVRF3,
00114      +                   DDRVRF4
00115 *     ..
00116 *     .. Executable Statements ..
00117 *
00118       S1 = DSECND( )
00119       FATAL = .FALSE.
00120 *
00121 *     Read a dummy line.
00122 *
00123       READ( NIN, FMT = * )
00124 *
00125 *     Report LAPACK version tag (e.g. LAPACK-3.2.0)
00126 *
00127       CALL ILAVER( VERS_MAJOR, VERS_MINOR, VERS_PATCH )
00128       WRITE( NOUT, FMT = 9994 ) VERS_MAJOR, VERS_MINOR, VERS_PATCH
00129 *
00130 *     Read the values of N
00131 *
00132       READ( NIN, FMT = * )NN
00133       IF( NN.LT.1 ) THEN
00134          WRITE( NOUT, FMT = 9996 )' NN ', NN, 1
00135          NN = 0
00136          FATAL = .TRUE.
00137       ELSE IF( NN.GT.MAXIN ) THEN
00138          WRITE( NOUT, FMT = 9995 )' NN ', NN, MAXIN
00139          NN = 0
00140          FATAL = .TRUE.
00141       END IF
00142       READ( NIN, FMT = * )( NVAL( I ), I = 1, NN )
00143       DO 10 I = 1, NN
00144          IF( NVAL( I ).LT.0 ) THEN
00145             WRITE( NOUT, FMT = 9996 )' M  ', NVAL( I ), 0
00146             FATAL = .TRUE.
00147          ELSE IF( NVAL( I ).GT.NMAX ) THEN
00148             WRITE( NOUT, FMT = 9995 )' M  ', NVAL( I ), NMAX
00149             FATAL = .TRUE.
00150          END IF
00151    10 CONTINUE
00152       IF( NN.GT.0 )
00153      $   WRITE( NOUT, FMT = 9993 )'N   ', ( NVAL( I ), I = 1, NN )
00154 *
00155 *     Read the values of NRHS
00156 *
00157       READ( NIN, FMT = * )NNS
00158       IF( NNS.LT.1 ) THEN
00159          WRITE( NOUT, FMT = 9996 )' NNS', NNS, 1
00160          NNS = 0
00161          FATAL = .TRUE.
00162       ELSE IF( NNS.GT.MAXIN ) THEN
00163          WRITE( NOUT, FMT = 9995 )' NNS', NNS, MAXIN
00164          NNS = 0
00165          FATAL = .TRUE.
00166       END IF
00167       READ( NIN, FMT = * )( NSVAL( I ), I = 1, NNS )
00168       DO 30 I = 1, NNS
00169          IF( NSVAL( I ).LT.0 ) THEN
00170             WRITE( NOUT, FMT = 9996 )'NRHS', NSVAL( I ), 0
00171             FATAL = .TRUE.
00172          ELSE IF( NSVAL( I ).GT.MAXRHS ) THEN
00173             WRITE( NOUT, FMT = 9995 )'NRHS', NSVAL( I ), MAXRHS
00174             FATAL = .TRUE.
00175          END IF
00176    30 CONTINUE
00177       IF( NNS.GT.0 )
00178      $   WRITE( NOUT, FMT = 9993 )'NRHS', ( NSVAL( I ), I = 1, NNS )
00179 *
00180 *     Read the matrix types
00181 *
00182       READ( NIN, FMT = * )NNT
00183       IF( NNT.LT.1 ) THEN
00184          WRITE( NOUT, FMT = 9996 )' NMA', NNT, 1
00185          NNT = 0
00186          FATAL = .TRUE.
00187       ELSE IF( NNT.GT.NTYPES ) THEN
00188          WRITE( NOUT, FMT = 9995 )' NMA', NNT, NTYPES
00189          NNT = 0
00190          FATAL = .TRUE.
00191       END IF
00192       READ( NIN, FMT = * )( NTVAL( I ), I = 1, NNT )
00193       DO 320 I = 1, NNT
00194          IF( NTVAL( I ).LT.0 ) THEN
00195             WRITE( NOUT, FMT = 9996 )'TYPE', NTVAL( I ), 0
00196             FATAL = .TRUE.
00197          ELSE IF( NTVAL( I ).GT.NTYPES ) THEN
00198             WRITE( NOUT, FMT = 9995 )'TYPE', NTVAL( I ), NTYPES
00199             FATAL = .TRUE.
00200          END IF
00201   320 CONTINUE
00202       IF( NNT.GT.0 )
00203      $   WRITE( NOUT, FMT = 9993 )'TYPE', ( NTVAL( I ), I = 1, NNT )
00204 *
00205 *     Read the threshold value for the test ratios.
00206 *
00207       READ( NIN, FMT = * )THRESH
00208       WRITE( NOUT, FMT = 9992 )THRESH
00209 *
00210 *     Read the flag that indicates whether to test the error exits.
00211 *
00212       READ( NIN, FMT = * )TSTERR
00213 *
00214       IF( FATAL ) THEN
00215          WRITE( NOUT, FMT = 9999 )
00216          STOP
00217       END IF
00218 *
00219       IF( FATAL ) THEN
00220          WRITE( NOUT, FMT = 9999 )
00221          STOP
00222       END IF
00223 *
00224 *     Calculate and print the machine dependent constants.
00225 *
00226       EPS = DLAMCH( 'Underflow threshold' )
00227       WRITE( NOUT, FMT = 9991 )'underflow', EPS
00228       EPS = DLAMCH( 'Overflow threshold' )
00229       WRITE( NOUT, FMT = 9991 )'overflow ', EPS
00230       EPS = DLAMCH( 'Epsilon' )
00231       WRITE( NOUT, FMT = 9991 )'precision', EPS
00232       WRITE( NOUT, FMT = * )
00233 *
00234 *     Test the error exit of:
00235 *
00236       IF( TSTERR )
00237      $   CALL DERRRFP( NOUT )
00238 *
00239 *     Test the routines: dpftrf, dpftri, dpftrs (as in DDRVPO).
00240 *     This also tests the routines: dtfsm, dtftri, dtfttr, dtrttf.
00241 *
00242       CALL DDRVRFP( NOUT, NN, NVAL, NNS, NSVAL, NNT, NTVAL, THRESH,
00243      $              WORKA, WORKASAV, WORKAFAC, WORKAINV, WORKB,
00244      $              WORKBSAV, WORKXACT, WORKX, WORKARF, WORKARFINV,
00245      $              D_WORK_DLATMS, D_WORK_DPOT01, D_TEMP_DPOT02,
00246      $              D_TEMP_DPOT03, D_WORK_DLANSY, D_WORK_DPOT02,
00247      $              D_WORK_DPOT03 )
00248 *
00249 *     Test the routine: dlansf
00250 *
00251       CALL DDRVRF1( NOUT, NN, NVAL, THRESH, WORKA, NMAX, WORKARF,
00252      +              D_WORK_DLANSY )
00253 *
00254 *     Test the convertion routines:
00255 *       dtfttp, dtpttf, dtfttr, dtrttf, dtrttp and dtpttr.
00256 *
00257       CALL DDRVRF2( NOUT, NN, NVAL, WORKA, NMAX, WORKARF,
00258      +              WORKAP, WORKASAV )
00259 *
00260 *     Test the routine: dtfsm
00261 *
00262       CALL DDRVRF3( NOUT, NN, NVAL, THRESH, WORKA, NMAX, WORKARF,
00263      +              WORKAINV, WORKAFAC, D_WORK_DLANSY,
00264      +              D_WORK_DPOT03, D_WORK_DPOT01 )
00265 *
00266 *
00267 *     Test the routine: dsfrk
00268 *
00269       CALL DDRVRF4( NOUT, NN, NVAL, THRESH, WORKA, WORKAFAC, NMAX,
00270      +              WORKARF, WORKAINV, NMAX, D_WORK_DLANSY)
00271 *
00272       CLOSE ( NIN )
00273       S2 = DSECND( )
00274       WRITE( NOUT, FMT = 9998 )
00275       WRITE( NOUT, FMT = 9997 )S2 - S1
00276 *
00277  9999 FORMAT( / ' Execution not attempted due to input errors' )
00278  9998 FORMAT( / ' End of tests' )
00279  9997 FORMAT( ' Total time used = ', F12.2, ' seconds', / )
00280  9996 FORMAT( ' !! Invalid input value: ', A4, '=', I6, '; must be >=',
00281      $      I6 )
00282  9995 FORMAT( ' !! Invalid input value: ', A4, '=', I6, '; must be <=',
00283      $      I6 )
00284  9994 FORMAT( /  ' Tests of the DOUBLE PRECISION LAPACK RFP routines ',
00285      $      / ' LAPACK VERSION ', I1, '.', I1, '.', I1,
00286      $      / / ' The following parameter values will be used:' )
00287  9993 FORMAT( 4X, A4, ':  ', 10I6, / 11X, 10I6 )
00288  9992 FORMAT( / ' Routines pass computational tests if test ratio is ',
00289      $      'less than', F8.2, / )
00290  9991 FORMAT( ' Relative machine ', A, ' is taken to be', D16.6 )
00291 *
00292 *     End of DCHKRFP
00293 *
00294       END