Routine: PDORMQR()  File: SRC\pdormqr.f

 
 
# lines: 450
  # code: 450
  # comment: 0
  # blank:0
# Variables:64
# Callers:5
# Callings:3
# Words:239
# Keywords:158
 

 

..
     .. Array Arguments ..
     ..
  Purpose
  =======
  PDORMQR overwrites the general real M-by-N distributed matrix
  sub( C ) = C(IC:IC+M-1,JC:JC+N-1) with
                      SIDE = 'L'            SIDE = 'R'
  TRANS = 'N':      Q * sub( C )          sub( C ) * Q
  TRANS = 'T':      Q**T * sub( C )       sub( C ) * Q**T
  where Q is a real orthogonal distributed matrix defined as the
  product of k elementary reflectors
        Q = H(1) H(2) . . . H(k)
  as returned by PDGEQRF. Q is of order M if SIDE = 'L' and of order N
  if SIDE = 'R'.
  Notes
  =====
  Each global data object is described by an associated description
  vector.  This vector stores the information required to establish
  the mapping between an object element and its corresponding process
  and memory location.
  Let A be a generic term for any 2D block cyclicly distributed array.
  Such a global array has an associated description vector DESCA.
  In the following comments, the character _ should be read as
  "of the global array".
  NOTATION        STORED IN      EXPLANATION
  --------------- -------------- --------------------------------------
  DTYPE_A(global) DESCA( DTYPE_ )The descriptor type.  In this case,
                                 DTYPE_A = 1.
  CTXT_A (global) DESCA( CTXT_ ) The BLACS context handle, indicating
                                 the BLACS process grid A is distribu-
                                 ted over. The context itself is glo-
                                 bal, but the handle (the integer
                                 value) may vary.
  M_A    (global) DESCA( M_ )    The number of rows in the global
                                 array A.
  N_A    (global) DESCA( N_ )    The number of columns in the global
                                 array A.
  MB_A   (global) DESCA( MB_ )   The blocking factor used to distribute
                                 the rows of the array.
  NB_A   (global) DESCA( NB_ )   The blocking factor used to distribute
                                 the columns of the array.
  RSRC_A (global) DESCA( RSRC_ ) The process row over which the first
                                 row of the array A is distributed.
  CSRC_A (global) DESCA( CSRC_ ) The process column over which the
                                 first column of the array A is
                                 distributed.
  LLD_A  (local)  DESCA( LLD_ )  The leading dimension of the local
                                 array.  LLD_A >= MAX(1,LOCr(M_A)).
  Let K be the number of rows or columns of a distributed matrix,
  and assume that its process grid has dimension p x q.
  LOCr( K ) denotes the number of elements of K that a process
  would receive if K were distributed over the p processes of its
  process column.
  Similarly, LOCc( K ) denotes the number of elements of K that a
  process would receive if K were distributed over the q processes of
  its process row.
  The values of LOCr() and LOCc() may be determined via a call to the
  ScaLAPACK tool function, NUMROC:
          LOCr( M ) = NUMROC( M, MB_A, MYROW, RSRC_A, NPROW ),
          LOCc( N ) = NUMROC( N, NB_A, MYCOL, CSRC_A, NPCOL ).
  An upper bound for these quantities may be computed by:
          LOCr( M ) <= ceil( ceil(M/MB_A)/NPROW )*MB_A
          LOCc( N ) <= ceil( ceil(N/NB_A)/NPCOL )*NB_A
  Arguments
  =========
  SIDE    (global input) CHARACTER
          = 'L': apply Q or Q**T from the Left;
          = 'R': apply Q or Q**T from the Right.
  TRANS   (global input) CHARACTER
          = 'N':  No transpose, apply Q;
          = 'T':  Transpose, apply Q**T.
  M       (global input) INTEGER
          The number of rows to be operated on i.e the number of rows
          of the distributed submatrix sub( C ). M >= 0.
  N       (global input) INTEGER
          The number of columns to be operated on i.e the number of
          columns of the distributed submatrix sub( C ). N >= 0.
  K       (global input) INTEGER
          The number of elementary reflectors whose product defines the
          matrix Q.  If SIDE = 'L', M >= K >= 0, if SIDE = 'R',
          N >= K >= 0.
  A       (local input) DOUBLE PRECISION pointer into the local memory
          to an array of dimension (LLD_A,LOCc(JA+K-1)). On entry, the
          j-th column must contain the vector which defines the elemen-
          tary reflector H(j), JA <= j <= JA+K-1, as returned by
          PDGEQRF in the K columns of its distributed matrix
          argument A(IA:*,JA:JA+K-1). A(IA:*,JA:JA+K-1) is modified by
          the routine but restored on exit.
          If SIDE = 'L', LLD_A >= MAX( 1, LOCr(IA+M-1) );
          if SIDE = 'R', LLD_A >= MAX( 1, LOCr(IA+N-1) ).
  IA      (global input) INTEGER
          The row index in the global array A indicating the first
          row of sub( A ).
  JA      (global input) INTEGER
          The column index in the global array A indicating the
          first column of sub( A ).
  DESCA   (global and local input) INTEGER array of dimension DLEN_.
          The array descriptor for the distributed matrix A.
  TAU     (local input) DOUBLE PRECISION array, dimension LOCc(JA+K-1).
          This array contains the scalar factors TAU(j) of the
          elementary reflectors H(j) as returned by PDGEQRF.
          TAU is tied to the distributed matrix A.
  C       (local input/local output) DOUBLE PRECISION pointer into the
          local memory to an array of dimension (LLD_C,LOCc(JC+N-1)).
          On entry, the local pieces of the distributed matrix sub(C).
          On exit, sub( C ) is overwritten by Q*sub( C ) or Q'*sub( C )
          or sub( C )*Q' or sub( C )*Q.
  IC      (global input) INTEGER
          The row index in the global array C indicating the first
          row of sub( C ).
  JC      (global input) INTEGER
          The column index in the global array C indicating the
          first column of sub( C ).
  DESCC   (global and local input) INTEGER array of dimension DLEN_.
          The array descriptor for the distributed matrix C.
  WORK    (local workspace/local output) DOUBLE PRECISION array,
                                                     dimension (LWORK)
          On exit, WORK(1) returns the minimal and optimal LWORK.
  LWORK   (local or global input) INTEGER
          The dimension of the array WORK.
          LWORK is local input and must be at least
          If SIDE = 'L',
            LWORK >= MAX( (NB_A*(NB_A-1))/2, (NqC0 + MpC0)*NB_A ) +
                     NB_A * NB_A
          else if SIDE = 'R',
            LWORK >= MAX( (NB_A*(NB_A-1))/2, ( NqC0 + MAX( NpA0 +
                     NUMROC( NUMROC( N+ICOFFC, NB_A, 0, 0, NPCOL ),
                             NB_A, 0, 0, LCMQ ), MpC0 ) )*NB_A ) +
                     NB_A * NB_A
          end if
          where LCMQ = LCM / NPCOL with LCM = ICLM( NPROW, NPCOL ),
          IROFFA = MOD( IA-1, MB_A ), ICOFFA = MOD( JA-1, NB_A ),
          IAROW = INDXG2P( IA, MB_A, MYROW, RSRC_A, NPROW ),
          NpA0 = NUMROC( N+IROFFA, MB_A, MYROW, IAROW, NPROW ),
          IROFFC = MOD( IC-1, MB_C ), ICOFFC = MOD( JC-1, NB_C ),
          ICROW = INDXG2P( IC, MB_C, MYROW, RSRC_C, NPROW ),
          ICCOL = INDXG2P( JC, NB_C, MYCOL, CSRC_C, NPCOL ),
          MpC0 = NUMROC( M+IROFFC, MB_C, MYROW, ICROW, NPROW ),
          NqC0 = NUMROC( N+ICOFFC, NB_C, MYCOL, ICCOL, NPCOL ),
          ILCM, INDXG2P and NUMROC are ScaLAPACK tool functions;
          MYROW, MYCOL, NPROW and NPCOL can be determined by calling
          the subroutine BLACS_GRIDINFO.
          If LWORK = -1, then LWORK is global input and a workspace
          query is assumed; the routine only calculates the minimum
          and optimal size for all work arrays. Each of these
          values is returned in the first entry of the corresponding
          work array, and no error message is issued by PXERBLA.
  INFO    (global output) INTEGER
          = 0:  successful exit
          < 0:  If the i-th argument is an array and the j-entry had
                an illegal value, then INFO = -(i*100+j), if the i-th
                argument is a scalar and had an illegal value, then
                INFO = -i.
  Alignment requirements
  ======================
  The distributed submatrices A(IA:*, JA:*) and C(IC:IC+M-1,JC:JC+N-1)
  must verify some alignment properties, namely the following
  expressions should be true:
  If SIDE = 'L',
    ( MB_A.EQ.MB_C .AND. IROFFA.EQ.IROFFC .AND. IAROW.EQ.ICROW )
  If SIDE = 'R',
    ( MB_A.EQ.NB_C .AND. IROFFA.EQ.ICOFFC )
  =====================================================================
     .. Parameters ..

 
Display dynamic version Find AutoScroll Reload FontSize: - + Hide Comments Hide Blanks Frame FullScreen MailPrint

 
001        SUBROUTINE PDORMQR( SIDE , TRANS , M , N , K , A , IA , JA , DESCA , TAU ,
002       $C , IC , JC , DESCC , WORK , LWORK , INFO )
003  
004  *     -- ScaLAPACK routine(version 1.7) --
005  *     University of Tennessee , Knoxville , Oak Ridge National Laboratory ,
006  *     and University of California , Berkeley.
007  *     May 25 , 2001
008  
009  *     .. Scalar Arguments ..
010        CHARACTER SIDE , TRANS
011        INTEGER IA , IC , INFO , JA , JC , K , LWORK , M , N
012        INTEGER BLOCK_CYCLIC_2D , CSRC_ , CTXT_ , DLEN_ , DTYPE_ ,
013       $LLD_ , MB_ , M_ , NB_ , N_ , RSRC_
014        PARAMETER( BLOCK_CYCLIC_2D = 1 , DLEN_ = 9 , DTYPE_ = 1 ,
015       $CTXT_ = 2 , M_ = 3 , N_ = 4 , MB_ = 5 , NB_ = 6 ,
016       $RSRC_ = 7 , CSRC_ = 8 , LLD_ = 9 )
017  *     ..
018  *     .. Local Scalars ..
019        LOGICAL LEFT , LQUERY , NOTRAN
020        CHARACTER COLBTOP , ROWBTOP
021        INTEGER IAROW , ICC , ICCOL , ICOFFC , ICROW , ICTXT , IINFO ,
022       $IPW , IROFFA , IROFFC , J , J1 , J2 , J3 , JB , JCC ,
023       $LCM , LCMQ , LWMIN , MI , MPC0 , MYCOL , MYROW , NI ,
024       $NPA0 , NPCOL , NPROW , NQ , NQC0
025  *     ..
026  *     .. Local Arrays ..
027        INTEGER IDUM1( 4 ) , IDUM2( 4 )
028  *     ..
029  *     .. External Subroutines ..
030        EXTERNAL BLACS_GRIDINFO , CHK1MAT , PCHK2MAT , PDLARFB ,
031       $PDLARFT , PDORM2R , PB_TOPGET , PB_TOPSET , PXERBLA
032  *     ..
033  *     .. External Functions ..
034        LOGICAL LSAME
035        INTEGER ICEIL , ILCM , INDXG2P , NUMROC
036        EXTERNAL ICEIL , ILCM , INDXG2P , LSAME , NUMROC
037  *     ..
038  *     .. Intrinsic Functions ..
039        INTRINSIC DBLE , ICHAR , MAX , MIN , MOD
040  *     ..
041  *     .. Executable Statements ..
042  
043  *     Get grid parameters
044  
045        ICTXT = DESCA( CTXT_ )
046        CALL BLACS_GRIDINFO( ICTXT , NPROW , NPCOL , MYROW , MYCOL )
047  
048  *     Test the input parameters
049  
050        INFO = 0
051        IF( NPROW.EQ. - 1 ) THEN
052            INFO = - (900 + CTXT_)
053        ELSE
054            LEFT = LSAME( SIDE , 'L' )
055            NOTRAN = LSAME( TRANS , 'N' )
056  
057  *         NQ is the order of Q
058  
059            IF( LEFT ) THEN
060                NQ = M
061                CALL CHK1MAT( M , 3 , K , 5 , IA , JA , DESCA , 9 , INFO )
062            ELSE
063                NQ = N
064                CALL CHK1MAT( N , 4 , K , 5 , IA , JA , DESCA , 9 , INFO )
065            END IF
066            CALL CHK1MAT( M , 3 , N , 4 , IC , JC , DESCC , 14 , INFO )
067            IF( INFO.EQ.0 ) THEN
068                IROFFA = MOD( IA - 1 , DESCA( MB_ ) )
069                IROFFC = MOD( IC - 1 , DESCC( MB_ ) )
070                ICOFFC = MOD( JC - 1 , DESCC( NB_ ) )
071                IAROW = INDXG2P( IA , DESCA( MB_ ) , MYROW , DESCA( RSRC_ ) ,
072       $        NPROW )
073                ICROW = INDXG2P( IC , DESCC( MB_ ) , MYROW , DESCC( RSRC_ ) ,
074       $        NPROW )
075                ICCOL = INDXG2P( JC , DESCC( NB_ ) , MYCOL , DESCC( CSRC_ ) ,
076       $        NPCOL )
077                MPC0 = NUMROC( M + IROFFC , DESCC( MB_ ) , MYROW , ICROW , NPROW )
078                NQC0 = NUMROC( N + ICOFFC , DESCC( NB_ ) , MYCOL , ICCOL , NPCOL )
079  
080                IF( LEFT ) THEN
081                    LWMIN = MAX(( DESCA( NB_ ) * ( DESCA( NB_ ) - 1 ) ) / 2 ,
082       $( MPC0 + NQC0 ) * DESCA( NB_ ) ) +
083       $            DESCA( NB_ ) * DESCA( NB_ )
084                ELSE
085                    NPA0 = NUMROC( N + IROFFA , DESCA( MB_ ) , MYROW , IAROW ,
086       $            NPROW )
087                    LCM = ILCM( NPROW , NPCOL )
088                    LCMQ = LCM / NPCOL
089                    LWMIN = MAX(( DESCA( NB_ ) * ( DESCA( NB_ ) - 1 ) )
090       $            / 2 ,( NQC0 + MAX( NPA0 + NUMROC( NUMROC(
091       $            N + ICOFFC , DESCA( NB_ ) , 0 , 0 , NPCOL ) ,
092       $            DESCA( NB_ ) , 0 , 0 , LCMQ ) , MPC0 ) ) *
093       $            DESCA( NB_ ) ) + DESCA( NB_ ) * DESCA( NB_ )
094                END IF
095  
096                WORK( 1 ) = DBLE( LWMIN )
097                LQUERY =( LWORK.EQ. - 1 )
098                IF( .NOT.LEFT .AND. .NOT.LSAME( SIDE , 'R' ) ) THEN
099                    INFO = - 1
100                ELSE IF( .NOT.NOTRAN .AND. .NOT.LSAME( TRANS , 'T' ) ) THEN
101                    INFO = - 2
102                ELSE IF( K.LT.0 .OR. K.GT.NQ ) THEN
103                    INFO = - 5
104                ELSE IF( .NOT.LEFT .AND. DESCA( MB_ ).NE.DESCC( NB_ ) ) THEN
105                    INFO = - (900 + NB_)
106                ELSE IF( LEFT .AND. IROFFA.NE.IROFFC ) THEN
107                    INFO = - 12
108                ELSE IF( LEFT .AND. IAROW.NE.ICROW ) THEN
109                    INFO = - 12
110                ELSE IF( .NOT.LEFT .AND. IROFFA.NE.ICOFFC ) THEN
111                    INFO = - 13
112                ELSE IF( LEFT .AND. DESCA( MB_ ).NE.DESCC( MB_ ) ) THEN
113                    INFO = - (1400 + MB_)
114                ELSE IF( ICTXT.NE.DESCC( CTXT_ ) ) THEN
115                    INFO = - (1400 + CTXT_)
116                ELSE IF( LWORK.LT.LWMIN .AND. .NOT.LQUERY ) THEN
117                    INFO = - 16
118                END IF
119            END IF
120  
121            IF( LEFT ) THEN
122                IDUM1( 1 ) = ICHAR( 'L' )
123            ELSE
124                IDUM1( 1 ) = ICHAR( 'R' )
125            END IF
126            IDUM2( 1 ) = 1
127            IF( NOTRAN ) THEN
128                IDUM1( 2 ) = ICHAR( 'N' )
129            ELSE
130                IDUM1( 2 ) = ICHAR( 'T' )
131            END IF
132            IDUM2( 2 ) = 2
133            IDUM1( 3 ) = K
134            IDUM2( 3 ) = 5
135            IF( LWORK.EQ. - 1 ) THEN
136                IDUM1( 4 ) = - 1
137            ELSE
138                IDUM1( 4 ) = 1
139            END IF
140            IDUM2( 4 ) = 16
141            IF( LEFT ) THEN
142                CALL PCHK2MAT( M , 3 , K , 5 , IA , JA , DESCA , 9 , M , 3 , N , 4 , IC ,
143       $        JC , DESCC , 14 , 4 , IDUM1 , IDUM2 , INFO )
144            ELSE
145                CALL PCHK2MAT( N , 4 , K , 5 , IA , JA , DESCA , 9 , M , 3 , N , 4 , IC ,
146       $        JC , DESCC , 14 , 4 , IDUM1 , IDUM2 , INFO )
147            END IF
148        END IF
149  
150        IF( INFO.NE.0 ) THEN
151            CALL PXERBLA( ICTXT , 'PDORMQR' , - INFO )
152            RETURN
153        ELSE IF( LQUERY ) THEN
154            RETURN
155        END IF
156  
157  *     Quick return if possible
158  
159        IF( M.EQ.0 .OR. N.EQ.0 .OR. K.EQ.0 )
160       $    RETURN
161  
162            CALL PB_TOPGET( ICTXT , 'Broadcast' , 'Rowwise' , ROWBTOP )
163            CALL PB_TOPGET( ICTXT , 'Broadcast' , 'Columnwise' , COLBTOP )
164  
165            IF(( LEFT .AND. .NOT.NOTRAN ) .OR.
166       $( .NOT.LEFT .AND. NOTRAN ) ) THEN
167            J1 = MIN( ICEIL( JA , DESCA( NB_ ) ) * DESCA( NB_ ) , JA + K - 1 )
168       $    + 1
169            J2 = JA + K - 1
170            J3 = DESCA( NB_ )
171        ELSE
172            J1 = MAX(((JA + K - 2) / DESCA( NB_ ) ) * DESCA( NB_ ) + 1 , JA )
173            J2 = MIN( ICEIL( JA , DESCA( NB_ ) ) * DESCA( NB_ ) , JA + K - 1 )
174       $    + 1
175            J3 = - DESCA( NB_ )
176        END IF
177  
178        IF( LEFT ) THEN
179            NI = N
180            JCC = JC
181            IF( NOTRAN ) THEN
182                CALL PB_TOPSET( ICTXT , 'Broadcast' , 'Rowwise' , 'D - ring' )
183            ELSE
184                CALL PB_TOPSET( ICTXT , 'Broadcast' , 'Rowwise' , 'I - ring' )
185            END IF
186            CALL PB_TOPSET( ICTXT , 'Broadcast' , 'Columnwise' , ' ' )
187        ELSE
188            MI = M
189            ICC = IC
190        END IF
191  
192  *     Use unblocked code for the first block if necessary
193  
194        IF(( LEFT .AND. .NOT.NOTRAN ) .OR.( .NOT.LEFT .AND. NOTRAN ) )
195       $    CALL PDORM2R ( SIDE , TRANS , M , N , J1 - JA , A , IA , JA , DESCA , TAU ,
196       $    C , IC , JC , DESCC , WORK , LWORK , IINFO )
197  
198            IPW = DESCA( NB_ ) * DESCA( NB_ ) + 1
199            DO 10 J = J1 , J2 , J3
200                JB = MIN( DESCA( NB_ ) , K - J + JA )
201  
202  *             Form the triangular factor of the block reflector
203  *             H = H(j) H(j + 1) . . . H(j + jb - 1)
204  
205                CALL PDLARFT ( 'Forward' , 'Columnwise' , NQ - J + JA , JB , A ,
206       $        IA + J - JA , J , DESCA , TAU , WORK , WORK( IPW ) )
207                IF( LEFT ) THEN
208  
209  *                 H or H' is applied to C(ic + j - ja : ic + m - 1 , jc : jc + n - 1)
210  
211                    MI = M - J + JA
212                    ICC = IC + J - JA
213                ELSE
214  
215  *                 H or H' is applied to C(ic : ic + m - 1 , jc + j - ja : jc + n - 1)
216  
217                    NI = N - J + JA
218                    JCC = JC + J - JA
219                END IF
220  
221  *             Apply H or H'
222  
223                CALL PDLARFB ( SIDE , TRANS , 'Forward' , 'Columnwise' , MI , NI ,
224       $        JB , A , IA + J - JA , J , DESCA , WORK , C , ICC , JCC ,
225       $        DESCC , WORK( IPW ) )
226     10     CONTINUE
227  
228  *         Use unblocked code for the last block if necessary
229  
230            IF(( LEFT .AND. NOTRAN ) .OR.( .NOT.LEFT .AND. .NOT.NOTRAN ) )
231       $        CALL PDORM2R ( SIDE , TRANS , M , N , J2 - JA , A , IA , JA , DESCA , TAU ,
232       $        C , IC , JC , DESCC , WORK , LWORK , IINFO )
233  
234                CALL PB_TOPSET( ICTXT , 'Broadcast' , 'Rowwise' , ROWBTOP )
235                CALL PB_TOPSET( ICTXT , 'Broadcast' , 'Columnwise' , COLBTOP )
236  
237                WORK( 1 ) = DBLE( LWMIN )
238  
239                RETURN
240  
241  *             End of PDORMQR
242  
243            END