laed6#
Functions
-
void slaed6(const INT kniter, const INT orgati, const f32 rho, const f32 *restrict D, const f32 *restrict Z, const f32 finit, f32 *tau, INT *info)#
SLAED6 computes the positive or negative root (closest to the origin) of z(0) z(1) z(2) f(x) = rho + ——— + ——-— + ——— d(0)-x d(1)-x d(2)-x.
It is assumed that
This routine will be called by SLAED4 when necessary. In most cases, the root sought is the smallest in magnitude, though it might not be in some extremely rare situations.if orgati is true the root is between d[1] and d[2]; otherwise it is between d[0] and d[1]
Parameters
inkniterRefer to SLAED4 for its significance.
inorgatiIf orgati is nonzero (true), the needed root is between d[1] and d[2]; otherwise it is between d[0] and d[1]. See SLAED4 for further details.
inrhoThe scalar in the equation f(x) above.
inDDouble precision array, dimension (3). D satisfies d[0] < d[1] < d[2].
inZDouble precision array, dimension (3). Each of the elements in Z must be positive.
infinitThe value of f at 0. It is more accurate than the one evaluated inside this routine.
outtauThe root of the equation f(x).
outinfo= 0: successful exit
> 0: if info = 1, failure to converge
void slaed6(
const INT kniter,
const INT orgati,
const f32 rho,
const f32* restrict D,
const f32* restrict Z,
const f32 finit,
f32* tau,
INT* info
);
Functions
-
void dlaed6(const INT kniter, const INT orgati, const f64 rho, const f64 *restrict D, const f64 *restrict Z, const f64 finit, f64 *tau, INT *info)#
DLAED6 computes the positive or negative root (closest to the origin) of z(0) z(1) z(2) f(x) = rho + ——— + ——-— + ——— d(0)-x d(1)-x d(2)-x.
It is assumed that
This routine will be called by DLAED4 when necessary. In most cases, the root sought is the smallest in magnitude, though it might not be in some extremely rare situations.if orgati is true the root is between d[1] and d[2]; otherwise it is between d[0] and d[1]
Parameters
inkniterRefer to DLAED4 for its significance.
inorgatiIf orgati is nonzero (true), the needed root is between d[1] and d[2]; otherwise it is between d[0] and d[1]. See DLAED4 for further details.
inrhoThe scalar in the equation f(x) above.
inDDouble precision array, dimension (3). D satisfies d[0] < d[1] < d[2].
inZDouble precision array, dimension (3). Each of the elements in Z must be positive.
infinitThe value of f at 0. It is more accurate than the one evaluated inside this routine.
outtauThe root of the equation f(x).
outinfo= 0: successful exit
> 0: if info = 1, failure to converge
void dlaed6(
const INT kniter,
const INT orgati,
const f64 rho,
const f64* restrict D,
const f64* restrict Z,
const f64 finit,
f64* tau,
INT* info
);