pztrrfs.f

Go to the documentation of this file.

      SUBROUTINE pztrrfs( UPLO, TRANS, DIAG, N, NRHS, A, IA, JA, DESCA,
     $                    B, IB, JB, DESCB, X, IX, JX, DESCX, FERR,
     $                    BERR, WORK, LWORK, RWORK, LRWORK, INFO )
*
*  -- ScaLAPACK routine (version 1.7) --
*     University of Tennessee, Knoxville, Oak Ridge National Laboratory,
*     and University of California, Berkeley.
*     May 1, 1997
*
*     .. Scalar Arguments ..
      CHARACTER          DIAG, TRANS, UPLO
      INTEGER            INFO, IA, IB, IX, JA, JB, JX, LRWORK, LWORK,
     $                   n, nrhs
*     ..
*     .. Array Arguments ..
      INTEGER            DESCA( * ), DESCB( * ), DESCX( * )
      DOUBLE PRECISION   BERR( * ), FERR( * ), RWORK( * )
      COMPLEX*16         A( * ), B( * ), WORK( * ), X( * )
*     ..
*
*  Purpose
*  =======
*
*  PZTRRFS provides error bounds and backward error estimates for the
*  solution to a system of linear equations with a triangular
*  coefficient matrix.
*
*  The solution matrix X must be computed by PZTRTRS or some other
*  means before entering this routine.  PZTRRFS does not do iterative
*  refinement because doing so cannot improve the backward error.
*
*  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
*
*  In the following comments, sub( A ), sub( X ) and sub( B ) denote
*  respectively A(IA:IA+N-1,JA:JA+N-1), X(IX:IX+N-1,JX:JX+NRHS-1) and
*  B(IB:IB+N-1,JB:JB+NRHS-1).
*
*  Arguments
*  =========
*
*  UPLO    (global input) CHARACTER*1
*          = 'U':  sub( A ) is upper triangular;
*          = 'L':  sub( A ) is lower triangular.
*
*  TRANS   (global input) CHARACTER*1
*          Specifies the form of the system of equations.
*          = 'N': sub( A ) * sub( X ) = sub( B )          (No transpose)
*          = 'T': sub( A )**T * sub( X ) = sub( B )          (Transpose)
*          = 'C': sub( A )**H * sub( X ) = sub( B )
*                                                  (Conjugate transpose)
*
*  DIAG    (global input) CHARACTER*1
*          = 'N':  sub( A ) is non-unit triangular;
*          = 'U':  sub( A ) is unit triangular.
*
*  N       (global input) INTEGER
*          The order of the matrix sub( A ).  N >= 0.
*
*  NRHS    (global input) INTEGER
*          The number of right hand sides, i.e., the number of columns
*          of the matrices sub( B ) and sub( X ).  NRHS >= 0.
*
*  A       (local input) COMPLEX*16 pointer into the local memory
*          to an array of local dimension (LLD_A,LOCc(JA+N-1) ). This
*          array contains the local pieces of the original triangular
*          distributed matrix sub( A ).
*          If UPLO = 'U', the leading N-by-N upper triangular part of
*          sub( A ) contains the upper triangular part of the matrix,
*          and its strictly lower triangular part is not referenced.
*          If UPLO = 'L', the leading N-by-N lower triangular part of
*          sub( A ) contains the lower triangular part of the distribu-
*          ted matrix, and its strictly upper triangular part is not
*          referenced.
*          If DIAG = 'U', the diagonal elements of sub( A ) are also
*          not referenced and are assumed to be 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.
*
*  B       (local input) COMPLEX*16 pointer into the local memory
*          to an array of local dimension (LLD_B, LOCc(JB+NRHS-1) ).
*          On entry, this array contains the the local pieces of the
*          right hand sides sub( B ).
*
*  IB      (global input) INTEGER
*          The row index in the global array B indicating the first
*          row of sub( B ).
*
*  JB      (global input) INTEGER
*          The column index in the global array B indicating the
*          first column of sub( B ).
*
*  DESCB   (global and local input) INTEGER array of dimension DLEN_.
*          The array descriptor for the distributed matrix B.
*
*  X       (local input) COMPLEX*16 pointer into the local memory
*          to an array of local dimension (LLD_X, LOCc(JX+NRHS-1) ).
*          On entry, this array contains the the local pieces of the
*          solution vectors sub( X ).
*
*  IX      (global input) INTEGER
*          The row index in the global array X indicating the first
*          row of sub( X ).
*
*  JX      (global input) INTEGER
*          The column index in the global array X indicating the
*          first column of sub( X ).
*
*  DESCX   (global and local input) INTEGER array of dimension DLEN_.
*          The array descriptor for the distributed matrix X.
*
*  FERR    (local output) DOUBLE PRECISION array of local dimension
*          LOCc(JB+NRHS-1). The estimated forward error bounds for
*          each solution vector of sub( X ).  If XTRUE is the true
*          solution, FERR bounds the magnitude of the largest entry
*          in (sub( X ) - XTRUE) divided by the magnitude of the
*          largest entry in sub( X ).  The estimate is as reliable as
*          the estimate for RCOND, and is almost always a slight
*          overestimate of the true error.
*          This array is tied to the distributed matrix X.
*
*  BERR    (local output) DOUBLE PRECISION array of local dimension
*          LOCc(JB+NRHS-1). The componentwise relative backward
*          error of each solution vector (i.e., the smallest re-
*          lative change in any entry of sub( A ) or sub( B )
*          that makes sub( X ) an exact solution).
*          This array is tied to the distributed matrix X.
*
*  WORK    (local workspace/local output) COMPLEX*16 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
*          LWORK >= 2*LOCr( N + MOD( IA-1, MB_A ) ).
*
*          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.
*
*  RWORK   (local workspace/local output) DOUBLE PRECISION array,
*                                                 dimension (LRWORK)
*          On exit, RWORK(1) returns the minimal and optimal LRWORK.
*
*  LRWORK  (local or global input) INTEGER
*          The dimension of the array RWORK.
*          LRWORK is local input and must be at least
*          LRWORK >= LOCr( N + MOD( IB-1, MB_B ) ).
*
*          If LRWORK = -1, then LRWORK 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.
*
*  Notes
*  =====
*
*  This routine temporarily returns when N <= 1.
*
*  The distributed submatrices sub( X ) and sub( B ) should be
*  distributed the same way on the same processes.  These conditions
*  ensure that sub( X ) and sub( B ) are "perfectly" aligned.
*
*  Moreover, this routine requires the distributed submatrices sub( A ),
*  sub( X ), and sub( B ) to be aligned on a block boundary,
*  i.e., if f(x,y) = MOD( x-1, y ):
*  f( IA, DESCA( MB_ ) ) = f( JA, DESCA( NB_ ) ) = 0,
*  f( IB, DESCB( MB_ ) ) = f( JB, DESCB( NB_ ) ) = 0, and
*  f( IX, DESCX( MB_ ) ) = f( JX, DESCX( NB_ ) ) = 0.
*
*  =====================================================================
*
*     .. Parameters ..
      INTEGER            BLOCK_CYCLIC_2D, CSRC_, CTXT_, DLEN_, DTYPE_,
     $                   LLD_, MB_, M_, NB_, N_, RSRC_
      parameter( block_cyclic_2d = 1, dlen_ = 9, dtype_ = 1,
     $                     ctxt_ = 2, m_ = 3, n_ = 4, mb_ = 5, nb_ = 6,
     $                     rsrc_ = 7, csrc_ = 8, lld_ = 9 )
      DOUBLE PRECISION   ZERO, RONE
      PARAMETER          ( ZERO = 0.0d+0, rone = 1.0d+0 )
      COMPLEX*16         ONE
      parameter( one = ( 1.0d+0, 0.0d+0 ) )
*     ..
*     .. Local Scalars ..
      LOGICAL            LQUERY, NOTRAN, NOUNIT, UPPER
      CHARACTER          TRANSN, TRANST
      INTEGER            IAROW, IXBCOL, IXBROW, IXCOL, IXROW, ICOFFA,
     $                   icoffb, icoffx, ictxt, icurcol, idum, ii, iixb,
     $                   iiw, ioffxb, ipb, ipr, ipv, iroffa, iroffb,
     $                   iroffx, iw, j, jbrhs, jj, jjfbe, jjxb, jn, jw,
     $                   k, kase, ldxb, lrwmin, lwmin, mycol, myrhs,
     $                   myrow, np, np0, npcol, npmod, nprow, nz
      DOUBLE PRECISION   EPS, EST, LSTRES, S, SAFE1, SAFE2, SAFMIN
      COMPLEX*16         ZDUM
*     ..
*     .. Local Arrays ..
      INTEGER            DESCW( DLEN_ ), IDUM1( 5 ), IDUM2( 5 )
*     ..
*     .. External Functions ..
      LOGICAL            LSAME
      INTEGER            ICEIL, INDXG2P, NUMROC
      DOUBLE PRECISION   PDLAMCH
      EXTERNAL           iceil, indxg2p, lsame, numroc, pdlamch
*     ..
*     .. External Subroutines ..
      EXTERNAL           blacs_gridinfo, chk1mat, descset, dgamx2d,
     $                   infog2l, pchk1mat, pchk2mat, pxerbla, pzatrmv,
     $                   pzaxpy, pzcopy, pzlacon, pztrmv,
     $                   pztrsv, zgebr2d, zgebs2d
*     ..
*     .. Intrinsic Functions ..
      INTRINSIC          abs, dble, dcmplx, dimag, ichar, max, min, mod
*     ..
*     .. Statement Functions ..
      DOUBLE PRECISION   CABS1
*     ..
*     .. Statement Function definitions ..
      cabs1( zdum ) = abs( dble( zdum ) ) + abs( dimag( zdum ) )
*     ..
*     .. Executable Statements ..
*
*     Get grid parameters
*
      ictxt = desca( ctxt_ )
      CALL blacs_gridinfo( ictxt, nprow, npcol, myrow, mycol )
*
*     Test the input parameters.
*
      info = 0
      IF( nprow.EQ.-1 ) THEN
         info = -( 900+ctxt_ )
      ELSE
         CALL chk1mat( n, 4, n, 4, ia, ja, desca, 9, info )
         CALL chk1mat( n, 4, nrhs, 5, ib, jb, descb, 13, info )
         CALL chk1mat( n, 4, nrhs, 5, ix, jx, descx, 17, info )
         IF( info.EQ.0 ) THEN
            upper = lsame( uplo, 'U' )
            notran = lsame( trans, 'N' )
            nounit = lsame( diag, 'N' )
            iroffa = mod( ia-1, desca( mb_ ) )
            icoffa = mod( ja-1, desca( nb_ ) )
            iroffb = mod( ib-1, descb( mb_ ) )
            icoffb = mod( jb-1, descb( nb_ ) )
            iroffx = mod( ix-1, descx( mb_ ) )
            icoffx = mod( jx-1, descx( nb_ ) )
            iarow = indxg2p( ia, desca( mb_ ), myrow, desca( rsrc_ ),
     $                       nprow )
            CALL infog2l( ib, jb, descb, nprow, npcol, myrow, mycol,
     $                    iixb, jjxb, ixbrow, ixbcol )
            ixrow = indxg2p( ix, descx( mb_ ), myrow, descx( rsrc_ ),
     $                       nprow )
            ixcol = indxg2p( jx, descx( nb_ ), mycol, descx( csrc_ ),
     $                       npcol )
            npmod = numroc( n+iroffa, desca( mb_ ), myrow, iarow,
     $                      nprow )
            lwmin = 2*npmod
            work( 1 ) = dble( lwmin )
            lrwmin = npmod
            rwork( 1 ) = dble( lrwmin )
            lquery = ( lwork.EQ.-1 .OR. lrwork.EQ.-1 )
*
            IF( .NOT.upper .AND. .NOT.lsame( uplo, 'L' ) ) THEN
               info = -1
            ELSE IF( .NOT.notran .AND. .NOT.lsame( trans, 'T' ) .AND.
     $         .NOT.lsame( trans, 'C' ) ) THEN
               info = -2
            ELSE IF( .NOT.nounit .AND. .NOT.lsame( diag, 'U' ) ) THEN
               info = -3
            ELSE IF( n.LT.0 ) THEN
               info = -4
            ELSE IF( nrhs.LT.0 ) THEN
               info = -5
            ELSE IF( iroffa.NE.0 ) THEN
               info = -7
            ELSE IF( icoffa.NE.0 ) THEN
               info = -8
            ELSE IF( desca( mb_ ).NE.desca( nb_ ) ) THEN
               info = -( 900+nb_ )
            ELSE IF( iroffa.NE.iroffb .OR. iarow.NE.ixbrow ) THEN
               info = -11
            ELSE IF( desca( mb_ ).NE.descb( mb_ ) ) THEN
               info = -( 1300+mb_ )
            ELSE IF( ictxt.NE.descb( ctxt_ ) ) THEN
               info = -( 1300+ctxt_ )
            ELSE IF( iroffx.NE.0 .OR. ixbrow.NE.ixrow ) THEN
               info = -15
            ELSE IF( icoffb.NE.icoffx .OR. ixbcol.NE.ixcol ) THEN
               info = -16
            ELSE IF( descb( mb_ ).NE.descx( mb_ ) ) THEN
               info = -( 1700+mb_ )
            ELSE IF( descb( nb_ ).NE.descx( nb_ ) ) THEN
               info = -( 1700+nb_ )
            ELSE IF( ictxt.NE.descx( ctxt_ ) ) THEN
               info = -( 1700+ctxt_ )
            ELSE IF( lwork.LT.lwmin .AND. .NOT.lquery ) THEN
               info = -21
            ELSE IF( lrwork.LT.lrwmin .AND. .NOT.lquery ) THEN
               info = -23
            END IF
         END IF
*
         IF( upper ) THEN
            idum1( 1 ) = ichar( 'U' )
         ELSE
            idum1( 1 ) = ichar( 'L' )
         END IF
         idum2( 1 ) = 1
         IF( notran ) THEN
            idum1( 2 ) = ichar( 'N' )
         ELSE IF( lsame( trans, 'T' ) ) THEN
            idum1( 2 ) = ichar( 'T' )
         ELSE
            idum1( 2 ) = ichar( 'C' )
         END IF
         idum2( 2 ) = 2
         IF( nounit ) THEN
            idum1( 3 ) = ichar( 'N' )
         ELSE
            idum1( 3 ) = ichar( 'U' )
         END IF
         idum2( 3 ) = 3
         IF( lwork.EQ.-1 ) THEN
            idum1( 4 ) = -1
         ELSE
            idum1( 4 ) = 1
         END IF
         idum2( 4 ) = 21
         IF( lrwork.EQ.-1 ) THEN
            idum1( 5 ) = -1
         ELSE
            idum1( 5 ) = 1
         END IF
         idum2( 5 ) = 23
         CALL pchk1mat( n, 4, n, 4, ia, ja, desca, 9, 0, idum1, idum2,
     $                  info )
         CALL pchk2mat( n, 4, nrhs, 5, ib, jb, descb, 13, n, 4, nrhs, 5,
     $                  ix, jx, descx, 17, 5, idum1, idum2, info )
      END IF
      IF( info.NE.0 ) THEN
         CALL pxerbla( ictxt, 'PZTRRFS', -info )
         RETURN
      ELSE IF( lquery ) THEN
         RETURN
      END IF
*
      jjfbe = jjxb
      myrhs = numroc( jb+nrhs-1, descb( nb_ ), mycol, descb( csrc_ ),
     $                npcol )
*
*     Quick return if possible
*
      IF( n.LE.1 .OR. nrhs.EQ.0 ) THEN
         DO 10 jj = jjfbe, myrhs
            ferr( jj ) = zero
            berr( jj ) = zero
   10    CONTINUE
         RETURN
      END IF
*
      IF( notran ) THEN
         transn = 'N'
         transt = 'C'
      ELSE
         transn = 'C'
         transt = 'N'
      END IF
*
      np0 = numroc( n+iroffb, descb( mb_ ), myrow, ixbrow, nprow )
      CALL descset( descw, n+iroffb, 1, desca( mb_ ), 1, ixbrow, ixbcol,
     $              ictxt, max( 1, np0 ) )
      ipb = 1
      ipr = 1
      ipv = ipr + np0
      IF( myrow.EQ.ixbrow ) THEN
         iiw = 1 + iroffb
         np = np0 - iroffb
      ELSE
         iiw = 1
         np = np0
      END IF
      iw = 1 + iroffb
      jw = 1
      ldxb = descb( lld_ )
      ioffxb = ( jjxb-1 )*ldxb
*
*     NZ = maximum number of nonzero entries in each row of A, plus 1
*
      nz = n + 1
      eps = pdlamch( ictxt, 'Epsilon' )
      safmin = pdlamch( ictxt, 'Safe minimum' )
      safe1 = nz*safmin
      safe2 = safe1 / eps
      jn = min( iceil( jb, descb( nb_ ) )*descb( nb_ ), jb+nrhs-1 )
*
*     Handle first block separately
*
      jbrhs = jn - jb + 1
      DO 90 k = 0, jbrhs - 1
*
*        Compute residual R = B - op(A) * X,
*        where op(A) = A or A', depending on TRANS.
*
         CALL pzcopy( n, x, ix, jx+k, descx, 1, work( ipr ), iw, jw,
     $                descw, 1 )
         CALL pztrmv( uplo, trans, diag, n, a, ia, ja, desca,
     $                work( ipr ), iw, jw, descw, 1 )
         CALL pzaxpy( n, -one, b, ib, jb+k, descb, 1, work( ipr ), iw,
     $                jw, descw, 1 )
*
*        Compute componentwise relative backward error from formula
*
*        max(i) ( abs(R(i)) / ( abs(op(A))*abs(X) + abs(B) )(i) )
*
*        where abs(Z) is the componentwise absolute value of the matrix
*        or vector Z.  If the i-th component of the denominator is less
*        than SAFE2, then SAFE1 is added to the i-th components of the
*        numerator and denominator before dividing.
*
         IF( mycol.EQ.ixbcol ) THEN
            IF( np.GT.0 ) THEN
               DO 20 ii = iixb, iixb + np - 1
                  rwork( iiw+ii-iixb ) = cabs1( b( ii+ioffxb ) )
   20          CONTINUE
            END IF
         END IF
*
         CALL pzatrmv( uplo, trans, diag, n, rone, a, ia, ja, desca, x,
     $                 ix, jx+k, descx, 1, rone, rwork( ipb ), iw, jw,
     $                 descw, 1 )
*
         s = zero
         IF( mycol.EQ.ixbcol ) THEN
            IF( np.GT.0 ) THEN
               DO 30 ii = iiw - 1, iiw + np - 2
                  IF( rwork( ipb+ii ).GT.safe2 ) THEN
                     s = max( s, cabs1( work( ipr+ii ) ) /
     $                           rwork( ipb+ii ) )
                  ELSE
                     s = max( s, ( cabs1( work( ipr+ii ) )+safe1 ) /
     $                           ( rwork( ipb+ii )+safe1 ) )
                  END IF
   30          CONTINUE
            END IF
         END IF
*
         CALL dgamx2d( ictxt, 'All', ' ', 1, 1, s, 1, idum, idum, 1,
     $                 -1, mycol )
         IF( mycol.EQ.ixbcol )
     $      berr( jjfbe ) = s
*
*        Bound error from formula
*
*        norm(X - XTRUE) / norm(X) .le. FERR =
*        norm( abs(inv(op(A)))*
*           ( abs(R) + NZ*EPS*( abs(op(A))*abs(X)+abs(B) ))) / norm(X)
*
*        where
*          norm(Z) is the magnitude of the largest component of Z
*          inv(op(A)) is the inverse of op(A)
*          abs(Z) is the componentwise absolute value of the matrix or
*             vector Z
*          NZ is the maximum number of nonzeros in any row of A, plus 1
*          EPS is machine epsilon
*
*        The i-th component of abs(R)+NZ*EPS*(abs(op(A))*abs(X)+abs(B))
*        is incremented by SAFE1 if the i-th component of
*        abs(op(A))*abs(X) + abs(B) is less than SAFE2.
*
*        Use PZLACON to estimate the infinity-norm of the matrix
*           inv(op(A)) * diag(W),
*        where W = abs(R) + NZ*EPS*( abs(op(A))*abs(X)+abs(B) )))
*
         IF( mycol.EQ.ixbcol ) THEN
            IF( np.GT.0 ) THEN
               DO 40 ii = iiw - 1, iiw + np - 2
                  IF( rwork( ipb+ii ).GT.safe2 ) THEN
                     rwork( ipb+ii ) = cabs1( work( ipr+ii ) ) +
     $                                 nz*eps*rwork( ipb+ii )
                  ELSE
                     rwork( ipb+ii ) = cabs1( work( ipr+ii ) ) +
     $                                 nz*eps*rwork( ipb+ii ) + safe1
                  END IF
   40          CONTINUE
            END IF
         END IF
*
         kase = 0
   50    CONTINUE
         IF( mycol.EQ.ixbcol ) THEN
            CALL zgebs2d( ictxt, 'Rowwise', ' ', np, 1, work( ipr ),
     $                    descw( lld_ ) )
         ELSE
            CALL zgebr2d( ictxt, 'Rowwise', ' ', np, 1, work( ipr ),
     $                    descw( lld_ ), myrow, ixbcol )
         END IF
         descw( csrc_ ) = mycol
         CALL pzlacon( n, work( ipv ), iw, jw, descw, work( ipr ),
     $                 iw, jw, descw, est, kase )
         descw( csrc_ ) = ixbcol
*
         IF( kase.NE.0 ) THEN
            IF( kase.EQ.1 ) THEN
*
*              Multiply by diag(W)*inv(op(A)').
*
               CALL pztrsv( uplo, transt, diag, n, a, ia, ja, desca,
     $                      work( ipr ), iw, jw, descw, 1 )
               IF( mycol.EQ.ixbcol ) THEN
                  IF( np.GT.0 ) THEN
                     DO 60 ii = iiw - 1, iiw + np - 2
                        work( ipr+ii ) = rwork( ipb+ii )*work( ipr+ii )
   60                CONTINUE
                  END IF
               END IF
            ELSE
*
*              Multiply by inv(op(A))*diag(W).
*
               IF( mycol.EQ.ixbcol ) THEN
                  IF( np.GT.0 ) THEN
                     DO 70 ii = iiw - 1, iiw + np - 2
                        work( ipr+ii ) = rwork( ipb+ii )*work( ipr+ii )
   70                CONTINUE
                  END IF
               END IF
               CALL pztrsv( uplo, transn, diag, n, a, ia, ja, desca,
     $                      work( ipr ), iw, jw, descw, 1 )
            END IF
            GO TO 50
         END IF
*
*        Normalize error.
*
         lstres = zero
         IF( mycol.EQ.ixbcol ) THEN
            IF( np.GT.0 ) THEN
               DO 80 ii = iixb, iixb + np - 1
                  lstres = max( lstres, cabs1( x( ioffxb+ii ) ) )
   80          CONTINUE
            END IF
            CALL dgamx2d( ictxt, 'Column', ' ', 1, 1, lstres, 1, idum,
     $                    idum, 1, -1, mycol )
            IF( lstres.NE.zero )
     $         ferr( jjfbe ) = est / lstres
*
            jjxb = jjxb + 1
            jjfbe = jjfbe + 1
            ioffxb = ioffxb + ldxb
*
         END IF
*
   90 CONTINUE
*
      icurcol = mod( ixbcol+1, npcol )
*
*     Do for each right hand side
*
      DO 180 j = jn + 1, jb + nrhs - 1, descb( nb_ )
         jbrhs = min( jb+nrhs-j, descb( nb_ ) )
         descw( csrc_ ) = icurcol
*
         DO 170 k = 0, jbrhs - 1
*
*           Compute residual R = B - op(A) * X,
*           where op(A) = A or A', depending on TRANS.
*
            CALL pzcopy( n, x, ix, j+k, descx, 1, work( ipr ), iw, jw,
     $                   descw, 1 )
            CALL pztrmv( uplo, trans, diag, n, a, ia, ja, desca,
     $                   work( ipr ), iw, jw, descw, 1 )
            CALL pzaxpy( n, -one, b, ib, j+k, descb, 1, work( ipr ),
     $                   iw, jw, descw, 1 )
*
*           Compute componentwise relative backward error from formula
*
*           max(i) ( abs(R(i)) / ( abs(op(A))*abs(X) + abs(B) )(i) )
*
*           where abs(Z) is the componentwise absolute value of the
*           matrix or vector Z.  If the i-th component of the
*           denominator is less than SAFE2, then SAFE1 is added to the
*           i-th components of the numerator and denominator before
*           dividing.
*
            IF( mycol.EQ.ixbcol ) THEN
               IF( np.GT.0 ) THEN
                  DO 100 ii = iixb, iixb + np - 1
                     rwork( iiw+ii-iixb ) = cabs1( b( ii+ioffxb ) )
  100             CONTINUE
               END IF
            END IF
*
            CALL pzatrmv( uplo, trans, diag, n, rone, a, ia, ja, desca,
     $                    x, ix, j+k, descx, 1, rone, rwork( ipb ), iw,
     $                    jw, descw, 1 )
*
            s = zero
            IF( mycol.EQ.ixbcol ) THEN
               IF( np.GT.0 ) THEN
                  DO 110 ii = iiw - 1, iiw + np - 2
                     IF( rwork( ipb+ii ).GT.safe2 ) THEN
                        s = max( s, cabs1( work( ipr+ii ) ) /
     $                              rwork( ipb+ii ) )
                     ELSE
                        s = max( s, ( cabs1( work( ipr+ii ) )+safe1 ) /
     $                              ( rwork( ipb+ii )+safe1 ) )
                     END IF
  110             CONTINUE
               END IF
            END IF
*
            CALL dgamx2d( ictxt, 'All', ' ', 1, 1, s, 1, idum, idum, 1,
     $                    -1, mycol )
            IF( mycol.EQ.ixbcol )
     $         berr( jjfbe ) = s
*
*           Bound error from formula
*
*           norm(X - XTRUE) / norm(X) .le. FERR =
*           norm( abs(inv(op(A)))*
*              ( abs(R) + NZ*EPS*( abs(op(A))*abs(X)+abs(B) )))/norm(X)
*
*           where
*             norm(Z) is the magnitude of the largest component of Z
*             inv(op(A)) is the inverse of op(A)
*             abs(Z) is the componentwise absolute value of the matrix
*                or vector Z
*             NZ is the maximum number of nonzeros in any row of A,
*                plus 1
*             EPS is machine epsilon
*
*           The i-th component of
*              abs(R)+NZ*EPS*(abs(op(A))*abs(X)+abs(B))
*           is incremented by SAFE1 if the i-th component of
*           abs(op(A))*abs(X) + abs(B) is less than SAFE2.
*
*           Use PZLACON to estimate the infinity-norm of the matrix
*              inv(op(A)) * diag(W),
*           where W = abs(R) + NZ*EPS*( abs(op(A))*abs(X)+abs(B) )))
*
            IF( mycol.EQ.ixbcol ) THEN
               IF( np.GT.0 ) THEN
                  DO 120 ii = iiw - 1, iiw + np - 2
                     IF( rwork( ipb+ii ).GT.safe2 ) THEN
                        rwork( ipb+ii ) = cabs1( work( ipr+ii ) ) +
     $                                    nz*eps*rwork( ipb+ii )
                     ELSE
                        rwork( ipb+ii ) = cabs1( work( ipr+ii ) ) +
     $                                    nz*eps*rwork( ipb+ii ) + safe1
                     END IF
  120             CONTINUE
               END IF
            END IF
*
            kase = 0
  130       CONTINUE
            IF( mycol.EQ.ixbcol ) THEN
               CALL zgebs2d( ictxt, 'Rowwise', ' ', np, 1, work( ipr ),
     $                       descw( lld_ ) )
            ELSE
               CALL zgebr2d( ictxt, 'Rowwise', ' ', np, 1, work( ipr ),
     $                       descw( lld_ ), myrow, ixbcol )
            END IF
            descw( csrc_ ) = mycol
            CALL pzlacon( n, work( ipv ), iw, jw, descw, work( ipr ),
     $                    iw, jw, descw, est, kase )
            descw( csrc_ ) = ixbcol
*
            IF( kase.NE.0 ) THEN
               IF( kase.EQ.1 ) THEN
*
*                 Multiply by diag(W)*inv(op(A)').
*
                  CALL pztrsv( uplo, transt, diag, n, a, ia, ja, desca,
     $                         work( ipr ), iw, jw, descw, 1 )
                  IF( mycol.EQ.ixbcol ) THEN
                     IF( np.GT.0 ) THEN
                        DO 140 ii = iiw - 1, iiw + np - 2
                           work( ipr+ii ) = rwork( ipb+ii )*
     $                                      work( ipr+ii )
  140                   CONTINUE
                     END IF
                  END IF
               ELSE
*
*                 Multiply by inv(op(A))*diag(W).
*
                  IF( mycol.EQ.ixbcol ) THEN
                     IF( np.GT.0 ) THEN
                        DO 150 ii = iiw - 1, iiw + np - 2
                           work( ipr+ii ) = rwork( ipb+ii )*
     $                                      work( ipr+ii )
  150                   CONTINUE
                     END IF
                  END IF
                  CALL pztrsv( uplo, transn, diag, n, a, ia, ja, desca,
     $                         work( ipr ), iw, jw, descw, 1 )
               END IF
               GO TO 130
            END IF
*
*           Normalize error.
*
            lstres = zero
            IF( mycol.EQ.ixbcol ) THEN
               IF( np.GT.0 ) THEN
                  DO 160 ii = iixb, iixb + np - 1
                     lstres = max( lstres, cabs1( x( ioffxb+ii ) ) )
  160             CONTINUE
               END IF
               CALL dgamx2d( ictxt, 'Column', ' ', 1, 1, lstres, 1,
     $                       idum, idum, 1, -1, mycol )
               IF( lstres.NE.zero )
     $            ferr( jjfbe ) = est / lstres
*
               jjxb = jjxb + 1
               jjfbe = jjfbe + 1
               ioffxb = ioffxb + ldxb
*
            END IF
*
  170    CONTINUE
*
         icurcol = mod( icurcol+1, npcol )
*
  180 CONTINUE
*
      work( 1 ) = dcmplx( dble( lwmin ) )
      rwork( 1 ) = dble( lrwmin )
*
      RETURN
*
*     End of PZTRRFS
*
      END