NAME

blas_dsort - sorts a real (double precision) vector X in increasing or decreasing order using quick sort algorithm

SYNOPSIS

  SUBROUTINE BLAS_DSORT( SORT, N, X, INCX)
  INTEGER SORT, N, INCX
  DOUBLE PRECISION X(*)

  SUBROUTINE BLAS_DSORT_64( SORT, N, X, INCX)
  INTEGER*8 SORT, N, INCX
  DOUBLE PRECISION X(*)

  SUBROUTINE SORT( [SORT], [N], X, [INCX])
  INTEGER :: SORT, N, INCX
  REAL(8), DIMENSION(:) :: X

  SUBROUTINE SORT_64( [SORT], [N], X, [INCX])
  INTEGER(8) :: SORT, N, INCX
  REAL(8), DIMENSION(:) :: X

#include <sunperf.h>

void blas_dsort(int sort, int n, double *x, int incx);

void blas_dsort_64(long sort, long n, double *x, long incx);

SORT (input)
SORT = 0, descending
SORT = 1, ascending

SORT = other value, error

SORT is default to 1 for F95 INTERFACE
N (input)
If N < = 1, the subroutine returns without trying to sort X.
X (input/output)
sorted
Minimum size (N-1)*|INCX|+1 is required
INCX (input)
INCX must not be zero. INCX could be negative. If INCX < 0, change the sorting direction defined by SORT. That is
If SORT = 0, let SORT = 1, INCX = |INCX|;

If SORT = 1, let SORT = 0, INCX = |INCX|.