minresblas.f90 File Reference

BLAS routines for MINRES. More...

Go to the source code of this file.

Functions/Subroutines
subroutine	daxpy (n, da, dx, incx, dy, incy)
	Constant times a vector plus a vector.
subroutine	dcopy (n, dx, incx, dy, incy)
	Copies a vector, x, to a vector, y.
DOUBLE PRECISION function	ddot (n, dx, incx, dy, incy)
	Forms the dot product of two vectors.
DOUBLE PRECISION function	dnrm2 (n, x, incx)
	Euclidean vector norm.
subroutine	daxpy2 (n, a, x, y, z)
	Set z = a*x + y.
subroutine	dload2 (n, const, x)
	Set x = constant.
subroutine	dscal2 (n, a, x, y)
	Set y = a*x.

---

Detailed Description

BLAS routines for MINRES.

     minresblas.f

     This file contains Level 1 BLAS from netlib, Thu May 16 1991
     (with declarations of the form dx(1) changed to dx(*)):
        daxpy    dcopy    ddot
     Also
        dnrm2    (from NAG,I think).

     Also a few utilities to avoid some of the
     loops in MINRES (so the debugger can step past them quickly):
        daxpy2   dload2   dscal2

 15 Jul 2003: dnrm2  is now the NAG version.

Definition in file minresblas.f90.

---

Function/Subroutine Documentation

subroutine daxpy	(	integer	n,
		double precision	da,
		double precision, dimension(*)	dx,
		integer	incx,
		double precision, dimension(*)	dy,
		integer	incy
	)

Constant times a vector plus a vector.

Uses unrolled loops for increments equal to one. (jack dongarra, linpack, 3/11/78)

Parameters:

[in]	n	size of vectors
[in]	da	scalar constant
[in]	dx	input vector
[in]	incx	increment for dx
[out]	dy	output vector
[in]	incy	increment for dy

Definition at line 35 of file minresblas.f90.

Referenced by minres().

subroutine daxpy2	(	integer	n,
		double precision	a,
		double precision, dimension(n)	x,
		double precision, dimension(n)	y,
		double precision, dimension(n)	z
	)

Set z = a*x + y.

31 May 1999: First version written for MINRES.

Parameters:

[in]	n	size of vectors
[in]	a	scalar constant
[in]	x	input vector
[in]	y	input vector
[out]	z	output vector

Definition at line 274 of file minresblas.f90.

subroutine dcopy	(	integer	n,
		double precision, dimension(*)	dx,
		integer	incx,
		double precision, dimension(*)	dy,
		integer	incy
	)

Copies a vector, x, to a vector, y.

Uses unrolled loops for increments equal to one. (jack dongarra, linpack, 3/11/78)

Parameters:

[in]	n	size of vectors
[in]	dx	input vector
[in]	incx	increment for dx
[out]	dy	output vector
[in]	incy	increment for dy

Definition at line 92 of file minresblas.f90.

Referenced by minres().

DOUBLE PRECISION function ddot	(	integer	n,
		double precision, dimension(*)	dx,
		integer	incx,
		double precision, dimension(*)	dy,
		integer	incy
	)

Forms the dot product of two vectors.

Uses unrolled loops for increments equal to one. (jack dongarra, linpack, 3/11/78)

Parameters:

[in]	n	size of vectors
[in]	dx	input vector
[in]	incx	increment for dx
[in]	dy	input vector
[in]	incy	increment for dy

Returns:

dot porduct dx*dy

Definition at line 152 of file minresblas.f90.

Referenced by minres().

subroutine dload2	(	integer	n,
		double precision	const,
		double precision, dimension(n)	x
	)

Set x = constant.

Load all elements of x with const.

Parameters:

[in]	n	size of vectors
[in]	const	scalar constant
[out]	x	output vector

Definition at line 297 of file minresblas.f90.

Referenced by minres().

DOUBLE PRECISION function dnrm2	(	integer	n,
		double precision, dimension(*)	x,
		integer	incx
	)

Euclidean vector norm.

dnrm2 returns the Euclidean norm of a vector via the function name, so that dnrm2 := sqrt( x'*x ).

15 Jul 2003: dnrm2 obtained from SNOPT src (probably from NAG). s1flmx replaced by safe large number.

Parameters:

[in]	n	size of vectors
[in]	x	input vector
[in]	incx	increment for x

Returns:

vector norm

Definition at line 212 of file minresblas.f90.

Referenced by minres().

subroutine dscal2	(	integer	n,
		double precision	a,
		double precision, dimension(n)	x,
		double precision, dimension(n)	y
	)

Set y = a*x.

Parameters:

[in]	n	size of vectors
[in]	a	scalar constant
[in]	x	input vector
[out]	y	output vector

Definition at line 324 of file minresblas.f90.

Referenced by minres().