F_i

PCGBTRF()

compute spike fill-in, l_i F_i = p_i b_{i-1
receive triangle b_{i-1} from previous processor

PCPBTRF()

l's off-diagonal block in conjugate transpose form.
{F_i}^c =  {h_i}{{b'}_i}^
copy matrix h_i (the last bw cols of g_i) to af storage

PCPTTRF()

l's off-diagonal block in conjugate transpose form.
{F_i}^c =  {h_i}{{b'}_i}^

PDGBTRF()

compute spike fill-in, l_i F_i = p_i b_{i-1
receive triangle b_{i-1} from previous processor

PDPBTRF()

l's off-diagonal block in transpose form.
{F_i}^t =  {h_i}{{b'}_i}^
copy matrix h_i (the last bw cols of g_i) to af storage

PDPTTRF()

l's off-diagonal block in transpose form.
{F_i}^t =  {h_i}{{b'}_i}^

PSGBTRF()

compute spike fill-in, l_i F_i = p_i b_{i-1
receive triangle b_{i-1} from previous processor

PSPBTRF()

l's off-diagonal block in transpose form.
{F_i}^t =  {h_i}{{b'}_i}^
copy matrix h_i (the last bw cols of g_i) to af storage

PSPTTRF()

l's off-diagonal block in transpose form.
{F_i}^t =  {h_i}{{b'}_i}^

PZGBTRF()

compute spike fill-in, l_i F_i = p_i b_{i-1
receive triangle b_{i-1} from previous processor

PZPBTRF()

l's off-diagonal block in conjugate transpose form.
{F_i}^c =  {h_i}{{b'}_i}^
copy matrix h_i (the last bw cols of g_i) to af storage

PZPTTRF()

l's off-diagonal block in conjugate transpose form.
{F_i}^c =  {h_i}{{b'}_i}^

fac

PCGESVX()

pcgesvx uses the lu factorization to compute the solution to 

PDGESVX()

pdgesvx uses the lu factorization to compute the solution to a rea

PSGESVX()

psgesvx uses the lu factorization to compute the solution to a rea

PZGESVX()

pzgesvx uses the lu factorization to compute the solution to 

FACT

PCGESVX()

pcgesvx uses the lu FACTorization to compute the solution to 

PCPOSVX()

pcposvx uses the cholesky FACTorization a = u**h*u or a = l*l**h t

PDGESVX()

pdgesvx uses the lu FACTorization to compute the solution to a rea

PDPOSVX()

pdposvx uses the cholesky FACTorization a = u**t*u or a = l*l**t t

PSGESVX()

psgesvx uses the lu FACTorization to compute the solution to a rea

PSPOSVX()

psposvx uses the cholesky FACTorization a = u**t*u or a = l*l**t t

PZGESVX()

pzgesvx uses the lu FACTorization to compute the solution to 

PZPOSVX()

pzposvx uses the cholesky FACTorization a = u**h*u or a = l*l**h t

facto

PCPOTRS()

hermitian positive definite distributed matrix using the cholesky
factorization sub( a ) = u**h*u or l*l**h computed by pcpotrf

PDPOTRS()

symmetric positive definite distributed matrix using the cholesky
factorization sub( a ) = u**t*u or l*l**t computed by pdpotrf

PSPOTRS()

symmetric positive definite distributed matrix using the cholesky
factorization sub( a ) = u**t*u or l*l**t computed by pspotrf

PZPOTRS()

hermitian positive definite distributed matrix using the cholesky
factorization sub( a ) = u**h*u or l*l**h computed by pzpotrf

Factor

CDBTF2()

cdbtrf computes an lu Factorization of a real m-by-n band matrix 

CDBTRF()

kv is the number of superdiagonals in the Factor 

CDTTRF()

cdttrf computes an lu Factorization of a complex tridiagonal matrix 

CPTTRSV()

u * x = b, or  u**h * x = b,
where l or u is the cholesky Factor of a hermitian positiv
a = u**h*d*u or a = l*d*l**h (computed by cpttrf).

DDBTF2()

ddbtrf computes an lu Factorization of a real m-by-n band matrix 

DDBTRF()

kv is the number of superdiagonals in the Factor 

DDTTRF()

ddttrf computes an lu Factorization of a complex tridiagonal matrix 

DPTTRSV()

l**t* x = b, or  l * x = b,
where l is the cholesky Factor of a hermitian positiv
a = l*d*l**h (computed by dpttrf).

PCDBSV()

gaussian elimination without pivoting
is used to Factor a reorderin

PCDBTRF()

offset to workspace for upper triangular Factor

PCDBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n complex

PCDBTRSV()

offset to workspace for upper triangular Factor

PCDTSV()

gaussian elimination without pivoting
is used to Factor a reorderin

PCDTTRF()

offset to workspace for upper triangular Factor

PCDTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n complex

PCDTTRSV()

offset to workspace for upper triangular Factor

PCGBSV()

gaussian elimination with pivoting
is used to Factor a reorderin

PCGBTRF()

transfer triangle b_i of local matrix to next processor
for fillin. overlap the send with the Factorization of a_i

PCGBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n complex

PCGEBD2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCGEBRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCGECON()

distributed complex matrix a(ia:ia+n-1,ja:ja+n-1), in either the
1-norm or the infinity-norm, using the lu Factorization computed b

PCGEEQU()

m-by-n distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja:ja+n-1) and
reduce its condition number.  r returns the row scale Factors and 
each row and column of the distributed matrix b with elements

PCGEHD2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCGEHRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCGELQ2()

pcgelq2 computes a lq Factorization of a complex distributed m-by-

PCGELQF()

pcgelqf computes a lq Factorization of a complex distributed m-by-

PCGELS()

systems involving an m-by-n matrix sub( a ) = a(ia:ia+m-1,ja:ja+n-1),
or its conjugate-transpose, using a qr or lq Factorization o

PCGEQL2()

pcgeql2 computes a ql Factorization of a complex distributed m-by-

PCGEQLF()

pcgeqlf computes a ql Factorization of a complex distributed m-by-

PCGEQPF()

pcgeqpf computes a qr Factorization with column pivoting of 

PCGEQR2()

pcgeqr2 computes a qr Factorization of a complex distributed m-by-

PCGEQRF()

pcgeqrf computes a qr Factorization of a complex distributed m-by-

PCGERFS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCGERQ2()

pcgerq2 computes a rq Factorization of a complex distributed m-by-

PCGERQF()

pcgerqf computes a rq Factorization of a complex distributed m-by-

PCGESV()

the lu decomposition with partial pivoting and row interchanges is
used to Factor sub( a ) as sub( a ) = p * l * u, where p is a permu
l and u are stored in sub( a ). the factored form of sub( a ) is then

PCGESVD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCGESVX()

pcgesvx uses the lu Factorization to compute the solution to 

PCGETF2()

pcgetf2 computes an lu Factorization of a general m-by-
partial pivoting with row interchanges.

PCGETRF()

pcgetrf computes an lu Factorization of a general m-by-n distribute
row interchanges.

PCGETRI()

pcgetri computes the inverse of a distributed matrix using the lu
Factorization computed by pcgetrf. this method inverts u and the
inva by solving the system inva*l = inv(u) for inva.

PCGETRS()

with a general n-by-n distributed matrix sub( a ) using the lu
Factorization computed by pcgetrf
and sub( b ) denotes b(ib:ib+n-1,jb:jb+nrhs-1).

PCGGQRF()

pcggqrf computes a generalized qr Factorization o
an n-by-p matrix sub( b ) = b(ib:ib+n-1,jb:jb+p-1):

PCGGRQF()

pcggrqf computes a generalized rq Factorization o
and a p-by-n matrix sub( b ) = b(ib:ib+p-1,jb:jb+n-1):

PCHEEV()

buted matrix a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCHEEVX()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCHEGS2()

sub( b ) must have been previously Factorized as u**h*u or l*l**h b

PCHEGST()

sub( b ) must have been previously Factorized as u**h*u or l*l**h b

PCHEGVX()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCHENGST()

sub( b ) must have been previously Factorized as u**h*u or l*l**h b

PCHENTRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCHETD2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCHETRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCHETTRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used t
nb_a   (global) desca( nb_ )   the blocking factor used to

PCLABRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLACGV()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLACON()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLACONSB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLACP2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLACP3()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLACPY()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLAEVSWP()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLANGE()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLAPIV()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLAPV2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLAQGE()

sub( a ) = a(ia:ia+m-1,ja:ja+n-1) using the row and scaling
Factors in the vectors r and c
notes

PCLAQSY()

pclaqsy equilibrates a symmetric distributed matrix
sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the scaling Factors in th

PCLARFB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLARFG()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLARFT()

pclarft forms the triangular Factor t of a complex block reflector 

PCLARZB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLARZT()

pclarzt forms the triangular Factor t of a complex block reflecto
reflectors as returned by pctzrzf.

PCLASCL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLASE2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLASET()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLASMSUB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLASSQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLASWP()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLATRA()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLATRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCLATRZ()

the upper trapezoidal matrix sub( a ) is Factored a
sub( a ) = ( r  0 ) * z,

PCLATTRS()

exit the loop if the growth Factor is too small

PCLAUU2()

pclauu2 computes the product u * u' or l' * l, where the triangular
Factor u or l is stored in the upper or lower triangular part o

PCLAUUM()

pclauum computes the product u * u' or l' * l, where the triangular
Factor u or l is stored in the upper or lower triangular part o

PCLAWIL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCMAX1()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCPBSV()

cholesky Factorization is used to factor a reordering o

PCPBTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to Factorization o
overlap the send with the factorization of a_i.

PCPBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n complex

PCPOCON()

1-norm) of a complex hermitian positive definite distributed matrix
using the cholesky Factorization a = u**h*u or a = l*l**h computed b

PCPOEQU()

(with respect to the two-norm).  sr and sc contain the scale
Factors, s(i) = 1/sqrt(a(i,i)), chosen so that the scaled distri
the  diagonal.  this choice of sr and sc puts the condition number

PCPORFS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCPOSV()

the cholesky decomposition is used to Factor sub( a ) a
sub( a ) = u**h * u,  if uplo = 'u', or

PCPOSVX()

pcposvx uses the cholesky Factorization a = u**h*u or a = l*l**h t

PCPOTF2()

pcpotf2 computes the cholesky Factorization of a complex hermitia

PCPOTRF()

pcpotrf computes the cholesky Factorization of an n-by-n comple
a(ia:ia+n-1, ja:ja+n-1).

PCPOTRI()

distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the
cholesky Factorization sub( a ) = u**h*u or l*l**h computed b

PCPOTRS()

hermitian positive definite distributed matrix using the cholesky
Factorization sub( a ) = u**h*u or l*l**h computed by pcpotrf

PCPTSV()

cholesky Factorization is used to factor a reordering o

PCPTTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to Factorization o
overlap the send with the factorization of a_i.

PCPTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n complex

PCSRSCL()

array a.
mb_a   (global) desca[ mb_ ]   the blocking Factor used to distribu
nb_a   (global) desca[ nb_ ]   the blocking factor used to distribu-

PCSTEIN()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCTRCON()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCTREVC()

products q*x and/or q*y, where q is an input unitary
matrix. if t was obtained from the schur Factorization of a
right or left eigenvectors of a.

PCTRRFS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCTRTI2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCTRTRI()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCTRTRS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCTZRZF()

the upper trapezoidal matrix sub( a ) is Factored a
sub( a ) = ( r  0 ) * z,

PCUNG2L()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNG2R()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNGL2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNGLQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNGQL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNGQR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNGR2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNGRQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNM2L()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNM2R()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNMBR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNMHR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNML2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNMLQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNMQL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNMQR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNMR2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNMR3()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNMRQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNMRZ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PCUNMTR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDDBSV()

gaussian elimination without pivoting
is used to Factor a reorderin

PDDBTRF()

offset to workspace for upper triangular Factor

PDDBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n real

PDDBTRSV()

offset to workspace for upper triangular Factor

PDDTSV()

gaussian elimination without pivoting
is used to Factor a reorderin

PDDTTRF()

offset to workspace for upper triangular Factor

PDDTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n real

PDDTTRSV()

offset to workspace for upper triangular Factor

PDGBSV()

gaussian elimination with pivoting
is used to Factor a reorderin

PDGBTRF()

transfer triangle b_i of local matrix to next processor
for fillin. overlap the send with the Factorization of a_i

PDGBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n real

PDGEBD2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDGEBRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDGECON()

distributed real matrix a(ia:ia+n-1,ja:ja+n-1), in either the 1-norm
or the infinity-norm, using the lu Factorization computed by pdgetrf
an estimate is obtained for norm(inv(a(ia:ia+n-1,ja:ja+n-1))), and

PDGEEQU()

m-by-n distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja:ja+n-1) and
reduce its condition number.  r returns the row scale Factors and 
each row and column of the distributed matrix b with elements

PDGEHD2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDGEHRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDGELQ2()

pdgelq2 computes a lq Factorization of a real distributed m-by-

PDGELQF()

pdgelqf computes a lq Factorization of a real distributed m-by-

PDGELS()

systems involving an m-by-n matrix sub( a ) = a(ia:ia+m-1,ja:ja+n-1),
or its transpose, using a qr or lq Factorization of sub( a ).  it i

PDGEQL2()

pdgeql2 computes a ql Factorization of a real distributed m-by-

PDGEQLF()

pdgeqlf computes a ql Factorization of a real distributed m-by-

PDGEQPF()

pdgeqpf computes a qr Factorization with column pivoting of 

PDGEQR2()

pdgeqr2 computes a qr Factorization of a real distributed m-by-

PDGEQRF()

pdgeqrf computes a qr Factorization of a real distributed m-by-

PDGERFS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDGERQ2()

pdgerq2 computes a rq Factorization of a real distributed m-by-

PDGERQF()

pdgerqf computes a rq Factorization of a real distributed m-by-

PDGESV()

the lu decomposition with partial pivoting and row interchanges is
used to Factor sub( a ) as sub( a ) = p * l * u, where p is a permu
l and u are stored in sub( a ). the factored form of sub( a ) is then

PDGESVD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDGESVX()

pdgesvx uses the lu Factorization to compute the solution to a rea

PDGETF2()

pdgetf2 computes an lu Factorization of a general m-by-
partial pivoting with row interchanges.

PDGETRF()

pdgetrf computes an lu Factorization of a general m-by-n distribute
row interchanges.

PDGETRI()

pdgetri computes the inverse of a distributed matrix using the lu
Factorization computed by pdgetrf. this method inverts u and the
inva by solving the system inva*l = inv(u) for inva.

PDGETRS()

with a general n-by-n distributed matrix sub( a ) using the lu
Factorization computed by pdgetrf
sub( b ) denotes b(ib:ib+n-1,jb:jb+nrhs-1).

PDGGQRF()

pdggqrf computes a generalized qr Factorization o
an n-by-p matrix sub( b ) = b(ib:ib+n-1,jb:jb+p-1):

PDGGRQF()

pdggrqf computes a generalized rq Factorization o
and a p-by-n matrix sub( b ) = b(ib:ib+p-1,jb:jb+n-1):

PDLABRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLACON()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLACONSB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLACP2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLACP3()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLACPY()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLAED2()

nb      (global input) integer
the blocking Factor used to distribute the columns of th

PDLAED3()

nb      (global input) integer
the blocking Factor used to distribute the columns of th

PDLAEVSWP()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLANGE()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLAPIV()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLAPV2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLAQGE()

sub( a ) = a(ia:ia+m-1,ja:ja+n-1) using the row and scaling
Factors in the vectors r and c
notes

PDLAQSY()

pdlaqsy equilibrates a symmetric distributed matrix
sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the scaling Factors in th

PDLARED1D()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLARED2D()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLARFB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLARFG()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLARFT()

pdlarft forms the triangular Factor t of a real block reflector 

PDLARZB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLARZT()

pdlarzt forms the triangular Factor t of a real block reflecto
reflectors as returned by pdtzrzf.

PDLASCL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLASE2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLASET()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLASMSUB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLASSQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLASWP()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLATRA()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLATRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDLATRZ()

the upper trapezoidal matrix sub( a ) is Factored a
sub( a ) = ( r  0 ) * z,

PDLAUU2()

pdlauu2 computes the product u * u' or l' * l, where the triangular
Factor u or l is stored in the upper or lower triangular part o

PDLAUUM()

pdlauum computes the product u * u' or l' * l, where the triangular
Factor u or l is stored in the upper or lower triangular part o

PDLAWIL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORG2L()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORG2R()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORGL2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORGLQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORGQL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORGQR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORGR2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORGRQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORM2L()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORM2R()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORMBR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORMHR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORML2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORMLQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORMQL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORMQR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORMR2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORMR3()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORMRQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORMRZ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDORMTR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDPBSV()

cholesky Factorization is used to factor a reordering o

PDPBTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to Factorization o
overlap the send with the factorization of a_i.

PDPBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n real

PDPOCON()

1-norm) of a real symmetric positive definite distributed matrix
using the cholesky Factorization a = u**t*u or a = l*l**t computed b

PDPOEQU()

(with respect to the two-norm).  sr and sc contain the scale
Factors, s(i) = 1/sqrt(a(i,i)), chosen so that the scaled distri
the  diagonal.  this choice of sr and sc puts the condition number

PDPORFS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDPOSV()

the cholesky decomposition is used to Factor sub( a ) a
sub( a ) = u**t * u,  if uplo = 'u', or

PDPOSVX()

pdposvx uses the cholesky Factorization a = u**t*u or a = l*l**t t

PDPOTF2()

pdpotf2 computes the cholesky Factorization of a real symmetri

PDPOTRF()

pdpotrf computes the cholesky Factorization of an n-by-n rea
a(ia:ia+n-1, ja:ja+n-1).

PDPOTRI()

distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the
cholesky Factorization sub( a ) = u**t*u or l*l**t computed b

PDPOTRS()

symmetric positive definite distributed matrix using the cholesky
Factorization sub( a ) = u**t*u or l*l**t computed by pdpotrf

PDPTSV()

cholesky Factorization is used to factor a reordering o

PDPTTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to Factorization o
overlap the send with the factorization of a_i.

PDPTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n real

PDRSCL()

array a.
mb_a   (global) desca[ mb_ ]   the blocking Factor used to distribu
nb_a   (global) desca[ nb_ ]   the blocking factor used to distribu-

PDSTEBZ()

fudge   double precision, default = 2.0
a "fudge Factor" to widen the gershgorin intervals.  ideally
arithmetic, this needs to be larger.  the default for

PDSTEIN()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDSYEV()

buted matrix a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDSYEVX()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDSYGS2()

sub( b ) must have been previously Factorized as u**t*u or l*l**t b

PDSYGST()

sub( b ) must have been previously Factorized as u**t*u or l*l**t b

PDSYGVX()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDSYNGST()

sub( b ) must have been previously Factorized as u**h*u or l*l**h b

PDSYNTRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDSYTD2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDSYTRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDSYTTRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used t
nb_a   (global) desca( nb_ )   the blocking factor used to

PDTRCON()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDTRRFS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDTRTI2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDTRTRI()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDTRTRS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PDTZRZF()

the upper trapezoidal matrix sub( a ) is Factored a
sub( a ) = ( r  0 ) * z,

PDZSUM1()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PJLAENV()

= 1: the data layout blocksize;
= 2: the panel blocking Factor
= 4: execution path control;

PSCSUM1()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSDBSV()

gaussian elimination without pivoting
is used to Factor a reorderin

PSDBTRF()

offset to workspace for upper triangular Factor

PSDBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n real

PSDBTRSV()

offset to workspace for upper triangular Factor

PSDTSV()

gaussian elimination without pivoting
is used to Factor a reorderin

PSDTTRF()

offset to workspace for upper triangular Factor

PSDTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n real

PSDTTRSV()

offset to workspace for upper triangular Factor

PSGBSV()

gaussian elimination with pivoting
is used to Factor a reorderin

PSGBTRF()

transfer triangle b_i of local matrix to next processor
for fillin. overlap the send with the Factorization of a_i

PSGBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n real

PSGEBD2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSGEBRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSGECON()

distributed real matrix a(ia:ia+n-1,ja:ja+n-1), in either the 1-norm
or the infinity-norm, using the lu Factorization computed by psgetrf
an estimate is obtained for norm(inv(a(ia:ia+n-1,ja:ja+n-1))), and

PSGEEQU()

m-by-n distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja:ja+n-1) and
reduce its condition number.  r returns the row scale Factors and 
each row and column of the distributed matrix b with elements

PSGEHD2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSGEHRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSGELQ2()

psgelq2 computes a lq Factorization of a real distributed m-by-

PSGELQF()

psgelqf computes a lq Factorization of a real distributed m-by-

PSGELS()

systems involving an m-by-n matrix sub( a ) = a(ia:ia+m-1,ja:ja+n-1),
or its transpose, using a qr or lq Factorization of sub( a ).  it i

PSGEQL2()

psgeql2 computes a ql Factorization of a real distributed m-by-

PSGEQLF()

psgeqlf computes a ql Factorization of a real distributed m-by-

PSGEQPF()

psgeqpf computes a qr Factorization with column pivoting of 

PSGEQR2()

psgeqr2 computes a qr Factorization of a real distributed m-by-

PSGEQRF()

psgeqrf computes a qr Factorization of a real distributed m-by-

PSGERFS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSGERQ2()

psgerq2 computes a rq Factorization of a real distributed m-by-

PSGERQF()

psgerqf computes a rq Factorization of a real distributed m-by-

PSGESV()

the lu decomposition with partial pivoting and row interchanges is
used to Factor sub( a ) as sub( a ) = p * l * u, where p is a permu
l and u are stored in sub( a ). the factored form of sub( a ) is then

PSGESVD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSGESVX()

psgesvx uses the lu Factorization to compute the solution to a rea

PSGETF2()

psgetf2 computes an lu Factorization of a general m-by-
partial pivoting with row interchanges.

PSGETRF()

psgetrf computes an lu Factorization of a general m-by-n distribute
row interchanges.

PSGETRI()

psgetri computes the inverse of a distributed matrix using the lu
Factorization computed by psgetrf. this method inverts u and the
inva by solving the system inva*l = inv(u) for inva.

PSGETRS()

with a general n-by-n distributed matrix sub( a ) using the lu
Factorization computed by psgetrf
sub( b ) denotes b(ib:ib+n-1,jb:jb+nrhs-1).

PSGGQRF()

psggqrf computes a generalized qr Factorization o
an n-by-p matrix sub( b ) = b(ib:ib+n-1,jb:jb+p-1):

PSGGRQF()

psggrqf computes a generalized rq Factorization o
and a p-by-n matrix sub( b ) = b(ib:ib+p-1,jb:jb+n-1):

PSLABRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLACON()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLACONSB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLACP2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLACP3()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLACPY()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLAED2()

nb      (global input) integer
the blocking Factor used to distribute the columns of th

PSLAED3()

nb      (global input) integer
the blocking Factor used to distribute the columns of th

PSLAEVSWP()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLANGE()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLAPIV()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLAPV2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLAQGE()

sub( a ) = a(ia:ia+m-1,ja:ja+n-1) using the row and scaling
Factors in the vectors r and c
notes

PSLAQSY()

pslaqsy equilibrates a symmetric distributed matrix
sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the scaling Factors in th

PSLARED1D()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLARED2D()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLARFB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLARFG()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLARFT()

pslarft forms the triangular Factor t of a real block reflector 

PSLARZB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLARZT()

pslarzt forms the triangular Factor t of a real block reflecto
reflectors as returned by pstzrzf.

PSLASCL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLASE2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLASET()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLASMSUB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLASSQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLASWP()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLATRA()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLATRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSLATRZ()

the upper trapezoidal matrix sub( a ) is Factored a
sub( a ) = ( r  0 ) * z,

PSLAUU2()

pslauu2 computes the product u * u' or l' * l, where the triangular
Factor u or l is stored in the upper or lower triangular part o

PSLAUUM()

pslauum computes the product u * u' or l' * l, where the triangular
Factor u or l is stored in the upper or lower triangular part o

PSLAWIL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORG2L()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORG2R()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORGL2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORGLQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORGQL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORGQR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORGR2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORGRQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORM2L()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORM2R()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORMBR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORMHR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORML2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORMLQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORMQL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORMQR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORMR2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORMR3()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORMRQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORMRZ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSORMTR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSPBSV()

cholesky Factorization is used to factor a reordering o

PSPBTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to Factorization o
overlap the send with the factorization of a_i.

PSPBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n real

PSPOCON()

1-norm) of a real symmetric positive definite distributed matrix
using the cholesky Factorization a = u**t*u or a = l*l**t computed b

PSPOEQU()

(with respect to the two-norm).  sr and sc contain the scale
Factors, s(i) = 1/sqrt(a(i,i)), chosen so that the scaled distri
the  diagonal.  this choice of sr and sc puts the condition number

PSPORFS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSPOSV()

the cholesky decomposition is used to Factor sub( a ) a
sub( a ) = u**t * u,  if uplo = 'u', or

PSPOSVX()

psposvx uses the cholesky Factorization a = u**t*u or a = l*l**t t

PSPOTF2()

pspotf2 computes the cholesky Factorization of a real symmetri

PSPOTRF()

pspotrf computes the cholesky Factorization of an n-by-n rea
a(ia:ia+n-1, ja:ja+n-1).

PSPOTRI()

distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the
cholesky Factorization sub( a ) = u**t*u or l*l**t computed b

PSPOTRS()

symmetric positive definite distributed matrix using the cholesky
Factorization sub( a ) = u**t*u or l*l**t computed by pspotrf

PSPTSV()

cholesky Factorization is used to factor a reordering o

PSPTTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to Factorization o
overlap the send with the factorization of a_i.

PSPTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n real

PSRSCL()

array a.
mb_a   (global) desca[ mb_ ]   the blocking Factor used to distribu
nb_a   (global) desca[ nb_ ]   the blocking factor used to distribu-

PSSTEBZ()

fudge   real, default = 2.0
a "fudge Factor" to widen the gershgorin intervals.  ideally
arithmetic, this needs to be larger.  the default for

PSSTEIN()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSSYEV()

buted matrix a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSSYEVX()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSSYGS2()

sub( b ) must have been previously Factorized as u**t*u or l*l**t b

PSSYGST()

sub( b ) must have been previously Factorized as u**t*u or l*l**t b

PSSYGVX()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSSYNGST()

sub( b ) must have been previously Factorized as u**h*u or l*l**h b

PSSYNTRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSSYTD2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSSYTRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSSYTTRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used t
nb_a   (global) desca( nb_ )   the blocking factor used to

PSTRCON()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSTRRFS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSTRTI2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSTRTRI()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSTRTRS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PSTZRZF()

the upper trapezoidal matrix sub( a ) is Factored a
sub( a ) = ( r  0 ) * z,

PZDBSV()

gaussian elimination without pivoting
is used to Factor a reorderin

PZDBTRF()

offset to workspace for upper triangular Factor

PZDBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n complex

PZDBTRSV()

offset to workspace for upper triangular Factor

PZDRSCL()

array a.
mb_a   (global) desca[ mb_ ]   the blocking Factor used to distribu
nb_a   (global) desca[ nb_ ]   the blocking factor used to distribu-

PZDTSV()

gaussian elimination without pivoting
is used to Factor a reorderin

PZDTTRF()

offset to workspace for upper triangular Factor

PZDTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n complex

PZDTTRSV()

offset to workspace for upper triangular Factor

PZGBSV()

gaussian elimination with pivoting
is used to Factor a reorderin

PZGBTRF()

transfer triangle b_i of local matrix to next processor
for fillin. overlap the send with the Factorization of a_i

PZGBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n complex

PZGEBD2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZGEBRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZGECON()

distributed complex matrix a(ia:ia+n-1,ja:ja+n-1), in either the
1-norm or the infinity-norm, using the lu Factorization computed b

PZGEEQU()

m-by-n distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja:ja+n-1) and
reduce its condition number.  r returns the row scale Factors and 
each row and column of the distributed matrix b with elements

PZGEHD2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZGEHRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZGELQ2()

pzgelq2 computes a lq Factorization of a complex distributed m-by-

PZGELQF()

pzgelqf computes a lq Factorization of a complex distributed m-by-

PZGELS()

systems involving an m-by-n matrix sub( a ) = a(ia:ia+m-1,ja:ja+n-1),
or its conjugate-transpose, using a qr or lq Factorization o

PZGEQL2()

pzgeql2 computes a ql Factorization of a complex distributed m-by-

PZGEQLF()

pzgeqlf computes a ql Factorization of a complex distributed m-by-

PZGEQPF()

pzgeqpf computes a qr Factorization with column pivoting of 

PZGEQR2()

pzgeqr2 computes a qr Factorization of a complex distributed m-by-

PZGEQRF()

pzgeqrf computes a qr Factorization of a complex distributed m-by-

PZGERFS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZGERQ2()

pzgerq2 computes a rq Factorization of a complex distributed m-by-

PZGERQF()

pzgerqf computes a rq Factorization of a complex distributed m-by-

PZGESV()

the lu decomposition with partial pivoting and row interchanges is
used to Factor sub( a ) as sub( a ) = p * l * u, where p is a permu
l and u are stored in sub( a ). the factored form of sub( a ) is then

PZGESVD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZGESVX()

pzgesvx uses the lu Factorization to compute the solution to 

PZGETF2()

pzgetf2 computes an lu Factorization of a general m-by-
partial pivoting with row interchanges.

PZGETRF()

pzgetrf computes an lu Factorization of a general m-by-n distribute
row interchanges.

PZGETRI()

pzgetri computes the inverse of a distributed matrix using the lu
Factorization computed by pzgetrf. this method inverts u and the
inva by solving the system inva*l = inv(u) for inva.

PZGETRS()

with a general n-by-n distributed matrix sub( a ) using the lu
Factorization computed by pzgetrf
and sub( b ) denotes b(ib:ib+n-1,jb:jb+nrhs-1).

PZGGQRF()

pzggqrf computes a generalized qr Factorization o
an n-by-p matrix sub( b ) = b(ib:ib+n-1,jb:jb+p-1):

PZGGRQF()

pzggrqf computes a generalized rq Factorization o
and a p-by-n matrix sub( b ) = b(ib:ib+p-1,jb:jb+n-1):

PZHEEV()

buted matrix a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZHEEVX()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZHEGS2()

sub( b ) must have been previously Factorized as u**h*u or l*l**h b

PZHEGST()

sub( b ) must have been previously Factorized as u**h*u or l*l**h b

PZHEGVX()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZHENGST()

sub( b ) must have been previously Factorized as u**h*u or l*l**h b

PZHENTRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZHETD2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZHETRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZHETTRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used t
nb_a   (global) desca( nb_ )   the blocking factor used to

PZLABRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLACGV()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLACON()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLACONSB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLACP2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLACP3()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLACPY()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLAEVSWP()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLANGE()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLAPIV()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLAPV2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLAQGE()

sub( a ) = a(ia:ia+m-1,ja:ja+n-1) using the row and scaling
Factors in the vectors r and c
notes

PZLAQSY()

pzlaqsy equilibrates a symmetric distributed matrix
sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the scaling Factors in th

PZLARFB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLARFG()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLARFT()

pzlarft forms the triangular Factor t of a complex block reflector 

PZLARZB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLARZT()

pzlarzt forms the triangular Factor t of a complex block reflecto
reflectors as returned by pztzrzf.

PZLASCL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLASE2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLASET()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLASMSUB()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLASSQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLASWP()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLATRA()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLATRD()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZLATRZ()

the upper trapezoidal matrix sub( a ) is Factored a
sub( a ) = ( r  0 ) * z,

PZLATTRS()

exit the loop if the growth Factor is too small

PZLAUU2()

pzlauu2 computes the product u * u' or l' * l, where the triangular
Factor u or l is stored in the upper or lower triangular part o

PZLAUUM()

pzlauum computes the product u * u' or l' * l, where the triangular
Factor u or l is stored in the upper or lower triangular part o

PZLAWIL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZMAX1()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZPBSV()

cholesky Factorization is used to factor a reordering o

PZPBTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to Factorization o
overlap the send with the factorization of a_i.

PZPBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n complex

PZPOCON()

1-norm) of a complex hermitian positive definite distributed matrix
using the cholesky Factorization a = u**h*u or a = l*l**h computed b

PZPOEQU()

(with respect to the two-norm).  sr and sc contain the scale
Factors, s(i) = 1/sqrt(a(i,i)), chosen so that the scaled distri
the  diagonal.  this choice of sr and sc puts the condition number

PZPORFS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZPOSV()

the cholesky decomposition is used to Factor sub( a ) a
sub( a ) = u**h * u,  if uplo = 'u', or

PZPOSVX()

pzposvx uses the cholesky Factorization a = u**h*u or a = l*l**h t

PZPOTF2()

pzpotf2 computes the cholesky Factorization of a complex hermitia

PZPOTRF()

pzpotrf computes the cholesky Factorization of an n-by-n comple
a(ia:ia+n-1, ja:ja+n-1).

PZPOTRI()

distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the
cholesky Factorization sub( a ) = u**h*u or l*l**h computed b

PZPOTRS()

hermitian positive definite distributed matrix using the cholesky
Factorization sub( a ) = u**h*u or l*l**h computed by pzpotrf

PZPTSV()

cholesky Factorization is used to factor a reordering o

PZPTTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to Factorization o
overlap the send with the factorization of a_i.

PZPTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the Factor
a(1:n, ja:ja+n-1) is an n-by-n complex

PZSTEIN()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZTRCON()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZTREVC()

products q*x and/or q*y, where q is an input unitary
matrix. if t was obtained from the schur Factorization of a
right or left eigenvectors of a.

PZTRRFS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZTRTI2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZTRTRI()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZTRTRS()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZTZRZF()

the upper trapezoidal matrix sub( a ) is Factored a
sub( a ) = ( r  0 ) * z,

PZUNG2L()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNG2R()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNGL2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNGLQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNGQL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNGQR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNGR2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNGRQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNM2L()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNM2R()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNMBR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNMHR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNML2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNMLQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNMQL()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNMQR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNMR2()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNMR3()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNMRQ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNMRZ()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

PZUNMTR()

array a.
mb_a   (global) desca( mb_ )   the blocking Factor used to distribut
nb_a   (global) desca( nb_ )   the blocking factor used to distribute

SDBTF2()

sdbtrf computes an lu Factorization of a real m-by-n band matrix 

SDBTRF()

kv is the number of superdiagonals in the Factor 

SDTTRF()

sdttrf computes an lu Factorization of a complex tridiagonal matrix 

SPTTRSV()

l**t* x = b, or  l * x = b,
where l is the cholesky Factor of a hermitian positiv
a = l*d*l**h (computed by spttrf).

ZDBTF2()

zdbtrf computes an lu Factorization of a real m-by-n band matrix 

ZDBTRF()

kv is the number of superdiagonals in the Factor 

ZDTTRF()

zdttrf computes an lu Factorization of a complex tridiagonal matrix 

ZPTTRSV()

u * x = b, or  u**h * x = b,
where l or u is the cholesky Factor of a hermitian positiv
a = u**h*d*u or a = l*d*l**h (computed by zpttrf).

factorable

PCDBSV()

processors was not
stably factorable wo/interchanges

PCDTSV()

processors was not
stably factorable wo/interchanges

PDDBSV()

processors was not
stably factorable wo/interchanges

PDDTSV()

processors was not
stably factorable wo/interchanges

PSDBSV()

processors was not
stably factorable wo/interchanges

PSDTSV()

processors was not
stably factorable wo/interchanges

PZDBSV()

processors was not
stably factorable wo/interchanges

PZDTSV()

processors was not
stably factorable wo/interchanges

factored

PCDBSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PCDBTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PCDTSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PCDTTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PCGBSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PCGBTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PCGELQ2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, the elements o
lower trapezoidal matrix l (l is lower triangular if m <= n);

PCGELQF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, the elements o
lower trapezoidal matrix l (l is lower triangular if m <= n);

PCGEQL2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m >= n, th
a( ia+m-n:ia+m-1, ja:ja+n-1 ) contains the n-by-n lower

PCGEQLF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m >= n, th
a( ia+m-n:ia+m-1, ja:ja+n-1 ) contains the n-by-n lower

PCGEQPF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if m >= n);

PCGEQR2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored.  on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if m >= n);

PCGEQRF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored.  on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if m >= n);

PCGERQ2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m <= n, th
m by m upper triangular matrix r; if m >= n, the elements on

PCGERQF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m <= n, th
m by m upper triangular matrix r; if m >= n, the elements on

PCGESV()

tation matrix, l is unit lower triangular, and u is upper triangular.
l and u are stored in sub( a ). the factored form of sub( a ) is the

PCGESVX()

3. the factored form of a is used to estimate the condition numbe
less than machine precision, steps 4-6 are skipped.

PCGETRF()

on entry, this array contains the local pieces of the m-by-n
distributed matrix sub( a ) to be factored. on exit, thi
the factorization sub( a ) = p*l*u; the unit diagonal ele-

PCGGQRF()

on entry, the local pieces of the n-by-m distributed matrix
sub( a ) which is to be factored.  on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if n >= m);

PCGGRQF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m <= n, th
m by m upper triangular matrix r; if m >= n, the elements on

PCHEGS2()

= 'u':  upper triangle of sub( a ) is stored and sub( b ) is
factored as u**h*u
factored as l*l**h.

PCHEGST()

= 'u':  upper triangle of sub( a ) is stored and sub( b ) is
factored as u**h*u
factored as l*l**h.

PCHENGST()

= 'u':  upper triangle of sub( a ) is stored and sub( b ) is
factored as u**h*u
factored as l*l**h.

PCLASSQ()

on exit, sumsq is overwritten with  smsq , the basic sum of
squares from which  scl  has been factored out
=====================================================================

PCLATRZ()

the upper trapezoidal matrix sub( a ) is factored a
sub( a ) = ( r  0 ) * z,

PCPBSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PCPBTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PCPORFS()

this array contains the local pieces of the n-by-n hermitian
distributed matrix sub( a ) to be factored
sub( a ) contains the upper triangular part of the matrix,

PCPOSV()

where u is an upper triangular matrix and l is a lower triangular
matrix.  the factored form of sub( a ) is then used to solve th

PCPOSVX()

3. the factored form of a is used to estimate the condition numbe
less than machine precision, steps 4-6 are skipped.

PCPOTF2()

on entry, this array contains the local pieces of the
n-by-n symmetric distributed matrix sub( a ) to be factored
sub( a ) contains the upper triangular part of the matrix,

PCPOTRF()

on entry, this array contains the local pieces of the
n-by-n hermitian distributed matrix sub( a ) to be factored
sub( a ) contains the upper triangular part of the matrix,

PCPTSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PCPTTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PCTZRZF()

the upper trapezoidal matrix sub( a ) is factored a
sub( a ) = ( r  0 ) * z,

PDDBSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PDDBTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PDDTSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PDDTTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PDGBSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PDGBTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PDGELQ2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, the elements o
lower trapezoidal matrix l (l is lower triangular if m <= n);

PDGELQF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, the elements o
lower trapezoidal matrix l (l is lower triangular if m <= n);

PDGEQL2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m >= n, th
a( ia+m-n:ia+m-1, ja:ja+n-1 ) contains the n-by-n lower

PDGEQLF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m >= n, th
a( ia+m-n:ia+m-1, ja:ja+n-1 ) contains the n-by-n lower

PDGEQPF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if m >= n);

PDGEQR2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored.  on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if m >= n);

PDGEQRF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored.  on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if m >= n);

PDGERQ2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m <= n, th
m by m upper triangular matrix r; if m >= n, the elements on

PDGERQF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m <= n, th
m by m upper triangular matrix r; if m >= n, the elements on

PDGESV()

tation matrix, l is unit lower triangular, and u is upper triangular.
l and u are stored in sub( a ). the factored form of sub( a ) is the

PDGESVX()

3. the factored form of a is used to estimate the condition numbe
less than machine precision, steps 4-6 are skipped.

PDGETRF()

on entry, this array contains the local pieces of the m-by-n
distributed matrix sub( a ) to be factored. on exit, thi
the factorization sub( a ) = p*l*u; the unit diagonal ele-

PDGGQRF()

on entry, the local pieces of the n-by-m distributed matrix
sub( a ) which is to be factored.  on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if n >= m);

PDGGRQF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m <= n, th
m by m upper triangular matrix r; if m >= n, the elements on

PDLASSQ()

on exit, sumsq is overwritten with  smsq , the basic sum of
squares from which  scl  has been factored out
=====================================================================

PDLATRZ()

the upper trapezoidal matrix sub( a ) is factored a
sub( a ) = ( r  0 ) * z,

PDPBSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PDPBTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PDPORFS()

this array contains the local pieces of the n-by-n symmetric
distributed matrix sub( a ) to be factored
sub( a ) contains the upper triangular part of the matrix,

PDPOSV()

where u is an upper triangular matrix and l is a lower triangular
matrix.  the factored form of sub( a ) is then used to solve th

PDPOSVX()

3. the factored form of a is used to estimate the condition numbe
less than machine precision, steps 4-6 are skipped.

PDPOTF2()

on entry, this array contains the local pieces of the
n-by-n symmetric distributed matrix sub( a ) to be factored
sub( a ) contains the upper triangular part of the matrix,

PDPOTRF()

on entry, this array contains the local pieces of the
n-by-n symmetric distributed matrix sub( a ) to be factored
sub( a ) contains the upper triangular part of the matrix,

PDPTSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PDPTTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PDSYGS2()

= 'u':  upper triangle of sub( a ) is stored and sub( b ) is
factored as u**t*u
factored as l*l**t.

PDSYGST()

= 'u':  upper triangle of sub( a ) is stored and sub( b ) is
factored as u**t*u
factored as l*l**t.

PDSYNGST()

= 'u':  upper triangle of sub( a ) is stored and sub( b ) is
factored as u**h*u
factored as l*l**h.

PDTZRZF()

the upper trapezoidal matrix sub( a ) is factored a
sub( a ) = ( r  0 ) * z,

PSDBSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PSDBTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PSDTSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PSDTTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PSGBSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PSGBTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PSGELQ2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, the elements o
lower trapezoidal matrix l (l is lower triangular if m <= n);

PSGELQF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, the elements o
lower trapezoidal matrix l (l is lower triangular if m <= n);

PSGEQL2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m >= n, th
a( ia+m-n:ia+m-1, ja:ja+n-1 ) contains the n-by-n lower

PSGEQLF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m >= n, th
a( ia+m-n:ia+m-1, ja:ja+n-1 ) contains the n-by-n lower

PSGEQPF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if m >= n);

PSGEQR2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored.  on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if m >= n);

PSGEQRF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored.  on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if m >= n);

PSGERQ2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m <= n, th
m by m upper triangular matrix r; if m >= n, the elements on

PSGERQF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m <= n, th
m by m upper triangular matrix r; if m >= n, the elements on

PSGESV()

tation matrix, l is unit lower triangular, and u is upper triangular.
l and u are stored in sub( a ). the factored form of sub( a ) is the

PSGESVX()

3. the factored form of a is used to estimate the condition numbe
less than machine precision, steps 4-6 are skipped.

PSGETRF()

on entry, this array contains the local pieces of the m-by-n
distributed matrix sub( a ) to be factored. on exit, thi
the factorization sub( a ) = p*l*u; the unit diagonal ele-

PSGGQRF()

on entry, the local pieces of the n-by-m distributed matrix
sub( a ) which is to be factored.  on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if n >= m);

PSGGRQF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m <= n, th
m by m upper triangular matrix r; if m >= n, the elements on

PSLASSQ()

on exit, sumsq is overwritten with  smsq , the basic sum of
squares from which  scl  has been factored out
=====================================================================

PSLATRZ()

the upper trapezoidal matrix sub( a ) is factored a
sub( a ) = ( r  0 ) * z,

PSPBSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PSPBTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PSPORFS()

this array contains the local pieces of the n-by-n symmetric
distributed matrix sub( a ) to be factored
sub( a ) contains the upper triangular part of the matrix,

PSPOSV()

where u is an upper triangular matrix and l is a lower triangular
matrix.  the factored form of sub( a ) is then used to solve th

PSPOSVX()

3. the factored form of a is used to estimate the condition numbe
less than machine precision, steps 4-6 are skipped.

PSPOTF2()

on entry, this array contains the local pieces of the
n-by-n symmetric distributed matrix sub( a ) to be factored
sub( a ) contains the upper triangular part of the matrix,

PSPOTRF()

on entry, this array contains the local pieces of the
n-by-n symmetric distributed matrix sub( a ) to be factored
sub( a ) contains the upper triangular part of the matrix,

PSPTSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PSPTTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PSSYGS2()

= 'u':  upper triangle of sub( a ) is stored and sub( b ) is
factored as u**t*u
factored as l*l**t.

PSSYGST()

= 'u':  upper triangle of sub( a ) is stored and sub( b ) is
factored as u**t*u
factored as l*l**t.

PSSYNGST()

= 'u':  upper triangle of sub( a ) is stored and sub( b ) is
factored as u**h*u
factored as l*l**h.

PSTZRZF()

the upper trapezoidal matrix sub( a ) is factored a
sub( a ) = ( r  0 ) * z,

PZDBSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PZDBTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PZDTSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PZDTTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PZGBSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PZGBTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PZGELQ2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, the elements o
lower trapezoidal matrix l (l is lower triangular if m <= n);

PZGELQF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, the elements o
lower trapezoidal matrix l (l is lower triangular if m <= n);

PZGEQL2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m >= n, th
a( ia+m-n:ia+m-1, ja:ja+n-1 ) contains the n-by-n lower

PZGEQLF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m >= n, th
a( ia+m-n:ia+m-1, ja:ja+n-1 ) contains the n-by-n lower

PZGEQPF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if m >= n);

PZGEQR2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored.  on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if m >= n);

PZGEQRF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored.  on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if m >= n);

PZGERQ2()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m <= n, th
m by m upper triangular matrix r; if m >= n, the elements on

PZGERQF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m <= n, th
m by m upper triangular matrix r; if m >= n, the elements on

PZGESV()

tation matrix, l is unit lower triangular, and u is upper triangular.
l and u are stored in sub( a ). the factored form of sub( a ) is the

PZGESVX()

3. the factored form of a is used to estimate the condition numbe
less than machine precision, steps 4-6 are skipped.

PZGETRF()

on entry, this array contains the local pieces of the m-by-n
distributed matrix sub( a ) to be factored. on exit, thi
the factorization sub( a ) = p*l*u; the unit diagonal ele-

PZGGQRF()

on entry, the local pieces of the n-by-m distributed matrix
sub( a ) which is to be factored.  on exit, the elements o
upper trapezoidal matrix r (r is upper triangular if n >= m);

PZGGRQF()

on entry, the local pieces of the m-by-n distributed matrix
sub( a ) which is to be factored. on exit, if m <= n, th
m by m upper triangular matrix r; if m >= n, the elements on

PZHEGS2()

= 'u':  upper triangle of sub( a ) is stored and sub( b ) is
factored as u**h*u
factored as l*l**h.

PZHEGST()

= 'u':  upper triangle of sub( a ) is stored and sub( b ) is
factored as u**h*u
factored as l*l**h.

PZHENGST()

= 'u':  upper triangle of sub( a ) is stored and sub( b ) is
factored as u**h*u
factored as l*l**h.

PZLASSQ()

on exit, sumsq is overwritten with  smsq , the basic sum of
squares from which  scl  has been factored out
=====================================================================

PZLATRZ()

the upper trapezoidal matrix sub( a ) is factored a
sub( a ) = ( r  0 ) * z,

PZPBSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PZPBTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PZPORFS()

this array contains the local pieces of the n-by-n hermitian
distributed matrix sub( a ) to be factored
sub( a ) contains the upper triangular part of the matrix,

PZPOSV()

where u is an upper triangular matrix and l is a lower triangular
matrix.  the factored form of sub( a ) is then used to solve th

PZPOSVX()

3. the factored form of a is used to estimate the condition numbe
less than machine precision, steps 4-6 are skipped.

PZPOTF2()

on entry, this array contains the local pieces of the
n-by-n symmetric distributed matrix sub( a ) to be factored
sub( a ) contains the upper triangular part of the matrix,

PZPOTRF()

on entry, this array contains the local pieces of the
n-by-n hermitian distributed matrix sub( a ) to be factored
sub( a ) contains the upper triangular part of the matrix,

PZPTSV()

> 0:  if info = k<=nprocs, the submatrix stored on processor
info and factored locally was no
the factorization was not completed.

PZPTTRS()

1) local phase:
the individual pieces are factored independently and i
fillin, which is stored in a non-inspectable way in auxiliary

PZTZRZF()

the upper trapezoidal matrix sub( a ) is factored a
sub( a ) = ( r  0 ) * z,

factori

PCPOSV()

referenced. on exit, if info = 0, this array contains the
local pieces of the factor u or l from the cholesky factori

PDPOSV()

referenced. on exit, if info = 0, this array contains the
local pieces of the factor u or l from the cholesky factori

PSPOSV()

referenced. on exit, if info = 0, this array contains the
local pieces of the factor u or l from the cholesky factori

PZPOSV()

referenced. on exit, if info = 0, this array contains the
local pieces of the factor u or l from the cholesky factori

factoring

PCDBSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PCDBTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PCDTSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PCDTTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PCGBSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PCGBTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PCPBSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PCPBTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PCPTSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PCPTTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PDDBSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PDDBTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PDDTSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PDDTTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PDGBSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PDGBTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PDPBSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PDPBTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PDPTSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PDPTTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PSDBSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PSDBTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PSDTSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PSDTTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PSGBSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PSGBTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PSPBSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PSPBTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PSPTSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PSPTTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PZDBSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PZDBTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PZDTSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PZDTTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PZGBSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PZGBTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PZPBSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PZPBTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PZPTSV()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

PZPTTRS()

space af. mathematically, this is equivalent to reordering
the matrix a as p a p^t and then factoring the principa
the matrices factored on each processor. the factors of

factoriza

PCGETF2()

pcgetf2 computes an lu factorization of a general m-by-
partial pivoting with row interchanges.

PDGETF2()

pdgetf2 computes an lu factorization of a general m-by-
partial pivoting with row interchanges.

PSGETF2()

psgetf2 computes an lu factorization of a general m-by-
partial pivoting with row interchanges.

PZGETF2()

pzgetf2 computes an lu factorization of a general m-by-
partial pivoting with row interchanges.

factorization

CDBTF2()

cdbtrf computes an lu factorization of a real m-by-n band matrix 

CDBTRF()

ju is the index of the last column affected by the current
stage of the factorization

CDTTRF()

cdttrf computes an lu factorization of a complex tridiagonal matrix 

CDTTRSV()

u * x = b,  u**t * x = b,  or  u**h * x = b,
with factors of the tridiagonal matrix a from the lu factorization

CLANV2()

clanv2 computes the schur factorization of a complex 2-by-

CPTTRSV()

of the tridiagonal matrix a is stored and the form of the
factorization
= 'l':  e is the subdiagonal of l, and a = l*d*l'.

DDBTF2()

ddbtrf computes an lu factorization of a real m-by-n band matrix 

DDBTRF()

ju is the index of the last column affected by the current
stage of the factorization

DDTTRF()

ddttrf computes an lu factorization of a complex tridiagonal matrix 

DDTTRSV()

u * x = b,  u**t * x = b,  or  u**h * x = b,
with factors of the tridiagonal matrix a from the lu factorization

DPTTRSV()

the n diagonal elements of the diagonal matrix d from the
factorization computed by dpttrf
e       (input) complex array, dimension (n-1)

DSTEIN2()

copy the matrix t so it won't be destroyed in factorization

PCDBSV()

on exit, this array contains information containing details
of the factorization
the factors returned are different from those returned

PCDBTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PCDBTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using pcdbtrs after the factorization

PCDBTRSV()

use factorization of odd-even connection block to modif

PCDTSV()

diagonally dominant-like,  and
the factorization was not completed
info-nprocs representing interactions with other

PCDTTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PCDTTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using pcdttrs after the factorization

PCDTTRSV()

use factorization of odd-even connection block to modif

PCGBSV()

on exit, this array contains information containing details
of the factorization
the factors returned are different from those returned

PCGBTRF()

transfer triangle b_i of local matrix to next processor
for fillin. overlap the send with the factorization of a_i

PCGBTRS()

ipiv    (local output) integer array, dimension >= desca( nb ).
pivot indices for local factorizations
factorization and solve.

PCGECON()

distributed complex matrix a(ia:ia+n-1,ja:ja+n-1), in either the
1-norm or the infinity-norm, using the lu factorization computed b

PCGELQ2()

pcgelq2 computes a lq factorization of a complex distributed m-by-

PCGELQF()

pcgelqf computes a lq factorization of a complex distributed m-by-

PCGELS()

systems involving an m-by-n matrix sub( a ) = a(ia:ia+m-1,ja:ja+n-1),
or its conjugate-transpose, using a qr or lq factorization o

PCGEQL2()

pcgeql2 computes a ql factorization of a complex distributed m-by-

PCGEQLF()

pcgeqlf computes a ql factorization of a complex distributed m-by-

PCGEQPF()

pcgeqpf computes a qr factorization with column pivoting of 

PCGEQR2()

pcgeqr2 computes a qr factorization of a complex distributed m-by-

PCGEQRF()

pcgeqrf computes a qr factorization of a complex distributed m-by-

PCGERQ2()

pcgerq2 computes a rq factorization of a complex distributed m-by-

PCGERQF()

pcgerqf computes a rq factorization of a complex distributed m-by-

PCGESV()

sub( a ) to be factored. on exit, this array contains the
local pieces of the factors l and u from the factorization
stored.

PCGESVX()

pcgesvx uses the lu factorization to compute the solution to 

PCGETF2()

pcgetf2 computes an lu factorization of a general m-by-
partial pivoting with row interchanges.

PCGETRF()

pcgetrf computes an lu factorization of a general m-by-n distribute
row interchanges.

PCGETRI()

pcgetri computes the inverse of a distributed matrix using the lu
factorization computed by pcgetrf. this method inverts u and the
inva by solving the system inva*l = inv(u) for inva.

PCGETRS()

with a general n-by-n distributed matrix sub( a ) using the lu
factorization computed by pcgetrf
and sub( b ) denotes b(ib:ib+n-1,jb:jb+nrhs-1).

PCGGQRF()

pcggqrf computes a generalized qr factorization o
an n-by-p matrix sub( b ) = b(ib:ib+n-1,jb:jb+p-1):

PCGGRQF()

pcggrqf computes a generalized rq factorization o
and a p-by-n matrix sub( b ) = b(ib:ib+p-1,jb:jb+n-1):

PCHEGS2()

this array contains the local pieces of the triangular factor
from the cholesky factorization of sub( b ), as returned b

PCHEGST()

this array contains the local pieces of the triangular factor
from the cholesky factorization of sub( b ), as returned b

PCHEGVX()

matrix is overwritten by the triangular factor u or l from
the cholesky factorization sub( b ) = u**h*u o

PCHENGST()

this array contains the local pieces of the triangular factor
from the cholesky factorization of sub( b ), as returned b

PCLATRZ()

the  factorization is obtained by householder's method.  the kt
introduce zeros into the (m - k + 1)th row of sub( a ), is given in

PCPBSV()

cholesky factorization is used to factor a reordering o

PCPBTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PCPBTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using pcpbtrs after the factorization

PCPBTRSV()

use factorization of odd-even connection block to modif

PCPOCON()

1-norm) of a complex hermitian positive definite distributed matrix
using the cholesky factorization a = u**h*u or a = l*l**h computed b

PCPORFS()

on entry, this array contains the factors l or u from the
cholesky factorization sub( a ) = l*l**h or u**h*u, a

PCPOSV()

a(ia:ia+k-1,ja:ja+k-1) is not positive definite, and
the factorization could not be completed, and th

PCPOSVX()

pcposvx uses the cholesky factorization a = u**h*u or a = l*l**h t

PCPOTF2()

pcpotf2 computes the cholesky factorization of a complex hermitia

PCPOTRF()

pcpotrf computes the cholesky factorization of an n-by-n comple
a(ia:ia+n-1, ja:ja+n-1).

PCPOTRI()

distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the
cholesky factorization sub( a ) = u**h*u or l*l**h computed b

PCPOTRS()

hermitian positive definite distributed matrix using the cholesky
factorization sub( a ) = u**h*u or l*l**h computed by pcpotrf

PCPTSV()

cholesky factorization is used to factor a reordering o

PCPTTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PCPTTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using pcpttrs after the factorization

PCPTTRSV()

use factorization of odd-even connection block to modif

PCTREVC()

products q*x and/or q*y, where q is an input unitary
matrix. if t was obtained from the schur factorization of a
right or left eigenvectors of a.

PCTZRZF()

the  factorization is obtained by householder's method.  the kt
introduce zeros into the (m - k + 1)th row of sub( a ), is given in

PDDBSV()

on exit, this array contains information containing details
of the factorization
the factors returned are different from those returned

PDDBTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PDDBTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using pddbtrs after the factorization

PDDBTRSV()

use factorization of odd-even connection block to modif

PDDTSV()

diagonally dominant-like,  and
the factorization was not completed
info-nprocs representing interactions with other

PDDTTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PDDTTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using pddttrs after the factorization

PDDTTRSV()

use factorization of odd-even connection block to modif

PDGBSV()

on exit, this array contains information containing details
of the factorization
the factors returned are different from those returned

PDGBTRF()

transfer triangle b_i of local matrix to next processor
for fillin. overlap the send with the factorization of a_i

PDGBTRS()

ipiv    (local output) integer array, dimension >= desca( nb ).
pivot indices for local factorizations
factorization and solve.

PDGECON()

distributed real matrix a(ia:ia+n-1,ja:ja+n-1), in either the 1-norm
or the infinity-norm, using the lu factorization computed by pdgetrf
an estimate is obtained for norm(inv(a(ia:ia+n-1,ja:ja+n-1))), and

PDGELQ2()

pdgelq2 computes a lq factorization of a real distributed m-by-

PDGELQF()

pdgelqf computes a lq factorization of a real distributed m-by-

PDGELS()

systems involving an m-by-n matrix sub( a ) = a(ia:ia+m-1,ja:ja+n-1),
or its transpose, using a qr or lq factorization of sub( a ).  it i

PDGEQL2()

pdgeql2 computes a ql factorization of a real distributed m-by-

PDGEQLF()

pdgeqlf computes a ql factorization of a real distributed m-by-

PDGEQPF()

pdgeqpf computes a qr factorization with column pivoting of 

PDGEQR2()

pdgeqr2 computes a qr factorization of a real distributed m-by-

PDGEQRF()

pdgeqrf computes a qr factorization of a real distributed m-by-

PDGERQ2()

pdgerq2 computes a rq factorization of a real distributed m-by-

PDGERQF()

pdgerqf computes a rq factorization of a real distributed m-by-

PDGESV()

sub( a ) to be factored. on exit, this array contains the
local pieces of the factors l and u from the factorization
stored.

PDGESVX()

pdgesvx uses the lu factorization to compute the solution to a rea

PDGETF2()

pdgetf2 computes an lu factorization of a general m-by-
partial pivoting with row interchanges.

PDGETRF()

pdgetrf computes an lu factorization of a general m-by-n distribute
row interchanges.

PDGETRI()

pdgetri computes the inverse of a distributed matrix using the lu
factorization computed by pdgetrf. this method inverts u and the
inva by solving the system inva*l = inv(u) for inva.

PDGETRS()

with a general n-by-n distributed matrix sub( a ) using the lu
factorization computed by pdgetrf
sub( b ) denotes b(ib:ib+n-1,jb:jb+nrhs-1).

PDGGQRF()

pdggqrf computes a generalized qr factorization o
an n-by-p matrix sub( b ) = b(ib:ib+n-1,jb:jb+p-1):

PDGGRQF()

pdggrqf computes a generalized rq factorization o
and a p-by-n matrix sub( b ) = b(ib:ib+p-1,jb:jb+n-1):

PDLATRZ()

the  factorization is obtained by householder's method.  the kt
the (m - k + 1)th row of sub( a ), is given in the form

PDPBSV()

cholesky factorization is used to factor a reordering o

PDPBTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PDPBTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using pdpbtrs after the factorization

PDPBTRSV()

use factorization of odd-even connection block to modif

PDPOCON()

1-norm) of a real symmetric positive definite distributed matrix
using the cholesky factorization a = u**t*u or a = l*l**t computed b

PDPORFS()

on entry, this array contains the factors l or u from the
cholesky factorization sub( a ) = l*l**t or u**t*u, a

PDPOSV()

a(ia:ia+k-1,ja:ja+k-1) is not positive definite, and
the factorization could not be completed, and th

PDPOSVX()

pdposvx uses the cholesky factorization a = u**t*u or a = l*l**t t

PDPOTF2()

pdpotf2 computes the cholesky factorization of a real symmetri

PDPOTRF()

pdpotrf computes the cholesky factorization of an n-by-n rea
a(ia:ia+n-1, ja:ja+n-1).

PDPOTRI()

distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the
cholesky factorization sub( a ) = u**t*u or l*l**t computed b

PDPOTRS()

symmetric positive definite distributed matrix using the cholesky
factorization sub( a ) = u**t*u or l*l**t computed by pdpotrf

PDPTSV()

cholesky factorization is used to factor a reordering o

PDPTTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PDPTTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using pdpttrs after the factorization

PDPTTRSV()

use factorization of odd-even connection block to modif

PDSYGS2()

this array contains the local pieces of the triangular factor
from the cholesky factorization of sub( b ), as returned b

PDSYGST()

this array contains the local pieces of the triangular factor
from the cholesky factorization of sub( b ), as returned b

PDSYGVX()

matrix is overwritten by the triangular factor u or l from
the cholesky factorization sub( b ) = u**t*u o

PDSYNGST()

this array contains the local pieces of the triangular factor
from the cholesky factorization of sub( b ), as returned b

PDTZRZF()

the  factorization is obtained by householder's method.  the kt
the (m - k + 1)th row of sub( a ), is given in the form

PSDBSV()

on exit, this array contains information containing details
of the factorization
the factors returned are different from those returned

PSDBTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PSDBTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using psdbtrs after the factorization

PSDBTRSV()

use factorization of odd-even connection block to modif

PSDTSV()

diagonally dominant-like,  and
the factorization was not completed
info-nprocs representing interactions with other

PSDTTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PSDTTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using psdttrs after the factorization

PSDTTRSV()

use factorization of odd-even connection block to modif

PSGBSV()

on exit, this array contains information containing details
of the factorization
the factors returned are different from those returned

PSGBTRF()

transfer triangle b_i of local matrix to next processor
for fillin. overlap the send with the factorization of a_i

PSGBTRS()

ipiv    (local output) integer array, dimension >= desca( nb ).
pivot indices for local factorizations
factorization and solve.

PSGECON()

distributed real matrix a(ia:ia+n-1,ja:ja+n-1), in either the 1-norm
or the infinity-norm, using the lu factorization computed by psgetrf
an estimate is obtained for norm(inv(a(ia:ia+n-1,ja:ja+n-1))), and

PSGELQ2()

psgelq2 computes a lq factorization of a real distributed m-by-

PSGELQF()

psgelqf computes a lq factorization of a real distributed m-by-

PSGELS()

systems involving an m-by-n matrix sub( a ) = a(ia:ia+m-1,ja:ja+n-1),
or its transpose, using a qr or lq factorization of sub( a ).  it i

PSGEQL2()

psgeql2 computes a ql factorization of a real distributed m-by-

PSGEQLF()

psgeqlf computes a ql factorization of a real distributed m-by-

PSGEQPF()

psgeqpf computes a qr factorization with column pivoting of 

PSGEQR2()

psgeqr2 computes a qr factorization of a real distributed m-by-

PSGEQRF()

psgeqrf computes a qr factorization of a real distributed m-by-

PSGERQ2()

psgerq2 computes a rq factorization of a real distributed m-by-

PSGERQF()

psgerqf computes a rq factorization of a real distributed m-by-

PSGESV()

sub( a ) to be factored. on exit, this array contains the
local pieces of the factors l and u from the factorization
stored.

PSGESVX()

psgesvx uses the lu factorization to compute the solution to a rea

PSGETF2()

psgetf2 computes an lu factorization of a general m-by-
partial pivoting with row interchanges.

PSGETRF()

psgetrf computes an lu factorization of a general m-by-n distribute
row interchanges.

PSGETRI()

psgetri computes the inverse of a distributed matrix using the lu
factorization computed by psgetrf. this method inverts u and the
inva by solving the system inva*l = inv(u) for inva.

PSGETRS()

with a general n-by-n distributed matrix sub( a ) using the lu
factorization computed by psgetrf
sub( b ) denotes b(ib:ib+n-1,jb:jb+nrhs-1).

PSGGQRF()

psggqrf computes a generalized qr factorization o
an n-by-p matrix sub( b ) = b(ib:ib+n-1,jb:jb+p-1):

PSGGRQF()

psggrqf computes a generalized rq factorization o
and a p-by-n matrix sub( b ) = b(ib:ib+p-1,jb:jb+n-1):

PSLATRZ()

the  factorization is obtained by householder's method.  the kt
the (m - k + 1)th row of sub( a ), is given in the form

PSPBSV()

cholesky factorization is used to factor a reordering o

PSPBTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PSPBTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using pspbtrs after the factorization

PSPBTRSV()

use factorization of odd-even connection block to modif

PSPOCON()

1-norm) of a real symmetric positive definite distributed matrix
using the cholesky factorization a = u**t*u or a = l*l**t computed b

PSPORFS()

on entry, this array contains the factors l or u from the
cholesky factorization sub( a ) = l*l**t or u**t*u, a

PSPOSV()

a(ia:ia+k-1,ja:ja+k-1) is not positive definite, and
the factorization could not be completed, and th

PSPOSVX()

psposvx uses the cholesky factorization a = u**t*u or a = l*l**t t

PSPOTF2()

pspotf2 computes the cholesky factorization of a real symmetri

PSPOTRF()

pspotrf computes the cholesky factorization of an n-by-n rea
a(ia:ia+n-1, ja:ja+n-1).

PSPOTRI()

distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the
cholesky factorization sub( a ) = u**t*u or l*l**t computed b

PSPOTRS()

symmetric positive definite distributed matrix using the cholesky
factorization sub( a ) = u**t*u or l*l**t computed by pspotrf

PSPTSV()

cholesky factorization is used to factor a reordering o

PSPTTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PSPTTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using pspttrs after the factorization

PSPTTRSV()

use factorization of odd-even connection block to modif

PSSYGS2()

this array contains the local pieces of the triangular factor
from the cholesky factorization of sub( b ), as returned b

PSSYGST()

this array contains the local pieces of the triangular factor
from the cholesky factorization of sub( b ), as returned b

PSSYGVX()

matrix is overwritten by the triangular factor u or l from
the cholesky factorization sub( b ) = u**t*u o

PSSYNGST()

this array contains the local pieces of the triangular factor
from the cholesky factorization of sub( b ), as returned b

PSTZRZF()

the  factorization is obtained by householder's method.  the kt
the (m - k + 1)th row of sub( a ), is given in the form

PZDBSV()

on exit, this array contains information containing details
of the factorization
the factors returned are different from those returned

PZDBTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PZDBTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using pzdbtrs after the factorization

PZDBTRSV()

use factorization of odd-even connection block to modif

PZDTSV()

diagonally dominant-like,  and
the factorization was not completed
info-nprocs representing interactions with other

PZDTTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PZDTTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using pzdttrs after the factorization

PZDTTRSV()

use factorization of odd-even connection block to modif

PZGBSV()

on exit, this array contains information containing details
of the factorization
the factors returned are different from those returned

PZGBTRF()

transfer triangle b_i of local matrix to next processor
for fillin. overlap the send with the factorization of a_i

PZGBTRS()

ipiv    (local output) integer array, dimension >= desca( nb ).
pivot indices for local factorizations
factorization and solve.

PZGECON()

distributed complex matrix a(ia:ia+n-1,ja:ja+n-1), in either the
1-norm or the infinity-norm, using the lu factorization computed b

PZGELQ2()

pzgelq2 computes a lq factorization of a complex distributed m-by-

PZGELQF()

pzgelqf computes a lq factorization of a complex distributed m-by-

PZGELS()

systems involving an m-by-n matrix sub( a ) = a(ia:ia+m-1,ja:ja+n-1),
or its conjugate-transpose, using a qr or lq factorization o

PZGEQL2()

pzgeql2 computes a ql factorization of a complex distributed m-by-

PZGEQLF()

pzgeqlf computes a ql factorization of a complex distributed m-by-

PZGEQPF()

pzgeqpf computes a qr factorization with column pivoting of 

PZGEQR2()

pzgeqr2 computes a qr factorization of a complex distributed m-by-

PZGEQRF()

pzgeqrf computes a qr factorization of a complex distributed m-by-

PZGERQ2()

pzgerq2 computes a rq factorization of a complex distributed m-by-

PZGERQF()

pzgerqf computes a rq factorization of a complex distributed m-by-

PZGESV()

sub( a ) to be factored. on exit, this array contains the
local pieces of the factors l and u from the factorization
stored.

PZGESVX()

pzgesvx uses the lu factorization to compute the solution to 

PZGETF2()

pzgetf2 computes an lu factorization of a general m-by-
partial pivoting with row interchanges.

PZGETRF()

pzgetrf computes an lu factorization of a general m-by-n distribute
row interchanges.

PZGETRI()

pzgetri computes the inverse of a distributed matrix using the lu
factorization computed by pzgetrf. this method inverts u and the
inva by solving the system inva*l = inv(u) for inva.

PZGETRS()

with a general n-by-n distributed matrix sub( a ) using the lu
factorization computed by pzgetrf
and sub( b ) denotes b(ib:ib+n-1,jb:jb+nrhs-1).

PZGGQRF()

pzggqrf computes a generalized qr factorization o
an n-by-p matrix sub( b ) = b(ib:ib+n-1,jb:jb+p-1):

PZGGRQF()

pzggrqf computes a generalized rq factorization o
and a p-by-n matrix sub( b ) = b(ib:ib+p-1,jb:jb+n-1):

PZHEGS2()

this array contains the local pieces of the triangular factor
from the cholesky factorization of sub( b ), as returned b

PZHEGST()

this array contains the local pieces of the triangular factor
from the cholesky factorization of sub( b ), as returned b

PZHEGVX()

matrix is overwritten by the triangular factor u or l from
the cholesky factorization sub( b ) = u**h*u o

PZHENGST()

this array contains the local pieces of the triangular factor
from the cholesky factorization of sub( b ), as returned b

PZLATRZ()

the  factorization is obtained by householder's method.  the kt
introduce zeros into the (m - k + 1)th row of sub( a ), is given in

PZPBSV()

cholesky factorization is used to factor a reordering o

PZPBTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PZPBTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using pzpbtrs after the factorization

PZPBTRSV()

use factorization of odd-even connection block to modif

PZPOCON()

1-norm) of a complex hermitian positive definite distributed matrix
using the cholesky factorization a = u**h*u or a = l*l**h computed b

PZPORFS()

on entry, this array contains the factors l or u from the
cholesky factorization sub( a ) = l*l**h or u**h*u, a

PZPOSV()

a(ia:ia+k-1,ja:ja+k-1) is not positive definite, and
the factorization could not be completed, and th

PZPOSVX()

pzposvx uses the cholesky factorization a = u**h*u or a = l*l**h t

PZPOTF2()

pzpotf2 computes the cholesky factorization of a complex hermitia

PZPOTRF()

pzpotrf computes the cholesky factorization of an n-by-n comple
a(ia:ia+n-1, ja:ja+n-1).

PZPOTRI()

distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the
cholesky factorization sub( a ) = u**h*u or l*l**h computed b

PZPOTRS()

hermitian positive definite distributed matrix using the cholesky
factorization sub( a ) = u**h*u or l*l**h computed by pzpotrf

PZPTSV()

cholesky factorization is used to factor a reordering o

PZPTTRF()

transfer last triangle d_i of local matrix to next processor
which needs it to calculate fillin due to factorization o
overlap the send with the factorization of a_i.

PZPTTRS()

auxiliary fillin space.
fillin is created during the factorization routin
is to be solved using pzpttrs after the factorization

PZPTTRSV()

use factorization of odd-even connection block to modif

PZTREVC()

products q*x and/or q*y, where q is an input unitary
matrix. if t was obtained from the schur factorization of a
right or left eigenvectors of a.

PZTZRZF()

the  factorization is obtained by householder's method.  the kt
introduce zeros into the (m - k + 1)th row of sub( a ), is given in

SDBTF2()

sdbtrf computes an lu factorization of a real m-by-n band matrix 

SDBTRF()

ju is the index of the last column affected by the current
stage of the factorization

SDTTRF()

sdttrf computes an lu factorization of a complex tridiagonal matrix 

SDTTRSV()

u * x = b,  u**t * x = b,  or  u**h * x = b,
with factors of the tridiagonal matrix a from the lu factorization

SPTTRSV()

the n diagonal elements of the diagonal matrix d from the
factorization computed by spttrf
e       (input) complex array, dimension (n-1)

SSTEIN2()

copy the matrix t so it won't be destroyed in factorization

ZDBTF2()

zdbtrf computes an lu factorization of a real m-by-n band matrix 

ZDBTRF()

ju is the index of the last column affected by the current
stage of the factorization

ZDTTRF()

zdttrf computes an lu factorization of a complex tridiagonal matrix 

ZDTTRSV()

u * x = b,  u**t * x = b,  or  u**h * x = b,
with factors of the tridiagonal matrix a from the lu factorization

ZLANV2()

zlanv2 computes the schur factorization of a complex 2-by-

ZPTTRSV()

of the tridiagonal matrix a is stored and the form of the
factorization
= 'l':  e is the subdiagonal of l, and a = l*d*l'.

factorizations

PCGBSV()

ipiv    (local output) integer array, dimension >= desca( nb ).
pivot indices for local factorizations
factorization and solve.

PCGBTRS()

ipiv    (local output) integer array, dimension >= desca( nb ).
pivot indices for local factorizations
factorization and solve.

PDGBSV()

ipiv    (local output) integer array, dimension >= desca( nb ).
pivot indices for local factorizations
factorization and solve.

PDGBTRS()

ipiv    (local output) integer array, dimension >= desca( nb ).
pivot indices for local factorizations
factorization and solve.

PSGBSV()

ipiv    (local output) integer array, dimension >= desca( nb ).
pivot indices for local factorizations
factorization and solve.

PSGBTRS()

ipiv    (local output) integer array, dimension >= desca( nb ).
pivot indices for local factorizations
factorization and solve.

PZGBSV()

ipiv    (local output) integer array, dimension >= desca( nb ).
pivot indices for local factorizations
factorization and solve.

PZGBTRS()

ipiv    (local output) integer array, dimension >= desca( nb ).
pivot indices for local factorizations
factorization and solve.

Factorize

CDBTRF()

here a11, a21 and a31 denote the current block of jb columns
which is about to be Factorized. the number of rows in th
of columns are jb, j2, j3. the superdiagonal elements of a13

DDBTRF()

here a11, a21 and a31 denote the current block of jb columns
which is about to be Factorized. the number of rows in th
of columns are jb, j2, j3. the superdiagonal elements of a13

SDBTRF()

here a11, a21 and a31 denote the current block of jb columns
which is about to be Factorized. the number of rows in th
of columns are jb, j2, j3. the superdiagonal elements of a13

ZDBTRF()

here a11, a21 and a31 denote the current block of jb columns
which is about to be Factorized. the number of rows in th
of columns are jb, j2, j3. the superdiagonal elements of a13

factorized

CDBTRF()

here a11, a21 and a31 denote the current block of jb columns
which is about to be factorized. the number of rows in th
of columns are jb, j2, j3. the superdiagonal elements of a13

DDBTRF()

here a11, a21 and a31 denote the current block of jb columns
which is about to be factorized. the number of rows in th
of columns are jb, j2, j3. the superdiagonal elements of a13

PCHEGS2()

sub( b ) must have been previously factorized as u**h*u or l*l**h b

PCHEGST()

sub( b ) must have been previously factorized as u**h*u or l*l**h b

PCHENGST()

sub( b ) must have been previously factorized as u**h*u or l*l**h b

PDSYGS2()

sub( b ) must have been previously factorized as u**t*u or l*l**t b

PDSYGST()

sub( b ) must have been previously factorized as u**t*u or l*l**t b

PDSYNGST()

sub( b ) must have been previously factorized as u**h*u or l*l**h b

PSSYGS2()

sub( b ) must have been previously factorized as u**t*u or l*l**t b

PSSYGST()

sub( b ) must have been previously factorized as u**t*u or l*l**t b

PSSYNGST()

sub( b ) must have been previously factorized as u**h*u or l*l**h b

PZHEGS2()

sub( b ) must have been previously factorized as u**h*u or l*l**h b

PZHEGST()

sub( b ) must have been previously factorized as u**h*u or l*l**h b

PZHENGST()

sub( b ) must have been previously factorized as u**h*u or l*l**h b

SDBTRF()

here a11, a21 and a31 denote the current block of jb columns
which is about to be factorized. the number of rows in th
of columns are jb, j2, j3. the superdiagonal elements of a13

ZDBTRF()

here a11, a21 and a31 denote the current block of jb columns
which is about to be factorized. the number of rows in th
of columns are jb, j2, j3. the superdiagonal elements of a13

factors

CDTTRSV()

u * x = b,  u**t * x = b,  or  u**h * x = b,
with factors of the tridiagonal matrix a from the lu factorizatio

DDTTRSV()

u * x = b,  u**t * x = b,  or  u**h * x = b,
with factors of the tridiagonal matrix a from the lu factorizatio

DSTEIN2()

compute lu factors with partial pivoting  ( pt = lu 

PCDBSV()

note that permutations are performed on the matrix, so that
the factors returned are different from those returne

PCDBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n complex

PCDTSV()

on exit, this array contains information containing the
factors of the matrix
d       (local input/local output) complex pointer to local

PCDTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n complex

PCGBSV()

note that permutations are performed on the matrix, so that
the factors returned are different from those returne

PCGBTRF()

use partial factors to update remainde

PCGBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n complex

PCGEBD2()

tauq    (local output) complex array dimension
locc(ja+min(m,n)-1). the scalar factors of the elementar
tied to the distributed matrix a. see further details.

PCGEBRD()

tauq    (local output) complex array dimension
locc(ja+min(m,n)-1). the scalar factors of the elementar
tied to the distributed matrix a. see further details.

PCGECON()

to an array of dimension ( lld_a, locc(ja+n-1) ). on entry,
this array contains the local pieces of the factors l and 
unit diagonal elements of l are not stored.

PCGEEQU()

m-by-n distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja:ja+n-1) and
reduce its condition number.  r returns the row scale factors and 
each row and column of the distributed matrix b with elements

PCGEHD2()

tau     (local output) complex array, dimension locc(ja+n-2)
the scalar factors of the elementary reflectors (see furthe
set to zero. tau is tied to the distributed matrix a.

PCGEHRD()

tau     (local output) complex array, dimension locc(ja+n-2)
the scalar factors of the elementary reflectors (see furthe
set to zero. tau is tied to the distributed matrix a.

PCGELQ2()

tau     (local output) complex, array, dimension
locr(ia+min(m,n)-1).  this array contains the scalar factors
matrix a.

PCGELQF()

tau     (local output) complex, array, dimension
locr(ia+min(m,n)-1).  this array contains the scalar factors
matrix a.

PCGEQL2()

tau     (local output) complex, array, dimension locc(ja+n-1)
this array contains the scalar factors of the elementar

PCGEQLF()

tau     (local output) complex, array, dimension locc(ja+n-1)
this array contains the scalar factors of the elementar

PCGEQPF()

tau     (local output) complex, array, dimension
locc(ja+min(m,n)-1). this array contains the scalar factors
distributed matrix a.

PCGEQR2()

tau     (local output) complex, array, dimension
locc(ja+min(m,n)-1). this array contains the scalar factors
distributed matrix a.

PCGEQRF()

tau     (local output) complex, array, dimension
locc(ja+min(m,n)-1). this array contains the scalar factors
distributed matrix a.

PCGERFS()

this array contains the local pieces of the distributed
factors of the matrix sub( a ) = p * l * u as computed b

PCGERQ2()

tau     (local output) complex, array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PCGERQF()

tau     (local output) complex, array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PCGESV()

sub( a ) to be factored. on exit, this array contains the
local pieces of the factors l and u from the factorizatio
stored.

PCGESVX()

1. if fact = 'e', real scaling factors are computed to equilibrat
trans = 'n':  diag(r)*a*diag(c)     *inv(diag(c))*x = diag(r)*b

PCGETF2()

distributed matrix sub( a ). on exit, this array contains
the local pieces of the factors l and u from the factoriza
not stored.

PCGETRF()

distributed matrix sub( a ) to be factored. on exit, this
array contains the local pieces of the factors l and u fro
ments of l are not stored.

PCGETRS()

memory to an array of dimension (lld_a, locc(ja+n-1)).
on entry, this array contains the local pieces of the factors
diagonal elements of l are not stored.

PCGGQRF()

taua    (local output) complex, array, dimension
locc(ja+min(n,m)-1). this array contains the scalar factors
matrix q. taua is tied to the distributed matrix a. (see

PCGGRQF()

taua    (local output) complex, array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar
tied to the distributed matrix a (see further details).

PCHEEVX()

for clustersize > n/sqrt(nprow*npcol) execution time will
grow as the square of the cluster size, all other factors
workspace means less reorthogonalization but faster

PCHEGVX()

for clustersize > n/sqrt(nprow*npcol) execution time will
grow as the square of the cluster size, all other factors
workspace means less reorthogonalization but faster

PCHENTRD()

tau     (local output) complex, array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PCHETD2()

tau     (local output) complex, array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PCHETRD()

tau     (local output) complex, array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PCHETTRD()

tau     (local output) complex, array, dimension
locq(ja+n-1). this array contains the scalar factors tau o
matrix a.

PCLABRD()

tauq    (local output) complex array dimension
locc(ja+min(m,n)-1). the scalar factors of the elementar
tied to the distributed matrix a. see further details.

PCLAHRD()

tau     (local output) complex array, dimension locc(ja+n-2)
the scalar factors of the elementary reflectors (see furthe

PCLAQGE()

sub( a ) = a(ia:ia+m-1,ja:ja+n-1) using the row and scaling
factors in the vectors r and c
notes

PCLAQSY()

pclaqsy equilibrates a symmetric distributed matrix
sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the scaling factors in th

PCLATRD()

tau     (local output) complex, array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PCLATRZ()

tau     (local output) complex, array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PCPBSV()

note that permutations are performed on the matrix, so that
the factors returned are different from those returne

PCPBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n complex

PCPOCON()

an array of dimension ( lld_a, locc(ja+n-1) ). on entry, this
array contains the local pieces of the factors l or u fro
l*l', as computed by pcpotrf.

PCPOEQU()

(with respect to the two-norm).  sr and sc contain the scale
factors, s(i) = 1/sqrt(a(i,i)), chosen so that the scaled distri
the  diagonal.  this choice of sr and sc puts the condition number

PCPORFS()

to an array of local dimension (lld_af,locc(ja+n-1)).
on entry, this array contains the factors l or u from th
computed by pcpotrf.

PCPOSVX()

1. if fact = 'e', real scaling factors are computed to equilibrat
diag(sr) * a * diag(sc) * inv(diag(sc)) * x = diag(sr) * b

PCPOTRS()

an array of dimension (lld_a, locc(ja+n-1)). on entry, this
array contains the factors l or u from the cholesky facto

PCPTSV()

on exit, this array contains information containing the
factors of the matrix

PCPTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n complex

PCTZRZF()

tau     (local output) complex, array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PCUNG2L()

tau     (local input) complex, array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PCUNG2R()

tau     (local input) complex, array, dimension locc(ja+k-1).
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PCUNGL2()

tau     (local input) complex, array, dimension locr(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PCUNGLQ()

tau     (local input) complex, array, dimension locr(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PCUNGQL()

tau     (local input) complex, array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PCUNGQR()

tau     (local input) complex, array, dimension locc(ja+k-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PCUNGR2()

tau     (local input) complex, array, dimension locr(ia+m-1)
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PCUNGRQ()

tau     (local input) complex, array, dimension locr(ia+m-1)
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PCUNM2L()

tau     (local input) complex, array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PCUNM2R()

tau     (local input) complex, array, dimension locc(ja+k-1).
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PCUNMHR()

if side = 'l', and locc(ja+n-2) if side = 'r'. this array
contains the scalar factors tau(j) of the elementar
the distributed matrix a.

PCUNML2()

tau     (local input) complex, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PCUNMLQ()

tau     (local input) complex, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PCUNMQL()

tau     (local input) complex, array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PCUNMQR()

tau     (local input) complex, array, dimension locc(ja+k-1).
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PCUNMR2()

tau     (local input) complex, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PCUNMR3()

tau     (local input) complex, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PCUNMRQ()

tau     (local input) complex, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PCUNMRZ()

tau     (local input) complex, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PDDBSV()

note that permutations are performed on the matrix, so that
the factors returned are different from those returne

PDDBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n real

PDDTSV()

on exit, this array contains information containing the
factors of the matrix
d       (local input/local output) double precision pointer to local

PDDTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n real

PDGBSV()

note that permutations are performed on the matrix, so that
the factors returned are different from those returne

PDGBTRF()

use partial factors to update remainde

PDGBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n real

PDGEBD2()

tauq    (local output) double precision array dimension
locc(ja+min(m,n)-1). the scalar factors of the elementar
is tied to the distributed matrix a. see further details.

PDGEBRD()

tauq    (local output) double precision array dimension
locc(ja+min(m,n)-1). the scalar factors of the elementar
is tied to the distributed matrix a. see further details.

PDGECON()

to an array of dimension ( lld_a, locc(ja+n-1) ). on entry,
this array contains the local pieces of the factors l and 
unit diagonal elements of l are not stored.

PDGEEQU()

m-by-n distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja:ja+n-1) and
reduce its condition number.  r returns the row scale factors and 
each row and column of the distributed matrix b with elements

PDGEHD2()

tau     (local output) double precision array, dimension locc(ja+n-2)
the scalar factors of the elementary reflectors (see furthe
set to zero. tau is tied to the distributed matrix a.

PDGEHRD()

tau     (local output) double precision array, dimension locc(ja+n-2)
the scalar factors of the elementary reflectors (see furthe
set to zero. tau is tied to the distributed matrix a.

PDGELQ2()

tau     (local output) double precision array, dimension
locr(ia+min(m,n)-1).  this array contains the scalar factors
matrix a.

PDGELQF()

tau     (local output) double precision array, dimension
locr(ia+min(m,n)-1).  this array contains the scalar factors
matrix a.

PDGEQL2()

tau     (local output) double precision array, dimension locc(ja+n-1)
this array contains the scalar factors of the elementar

PDGEQLF()

tau     (local output) double precision array, dimension locc(ja+n-1)
this array contains the scalar factors of the elementar

PDGEQPF()

tau     (local output) double precision array, dimension
locc(ja+min(m,n)-1). this array contains the scalar factors
distributed matrix a.

PDGEQR2()

tau     (local output) double precision array, dimension
locc(ja+min(m,n)-1). this array contains the scalar factors
distributed matrix a.

PDGEQRF()

tau     (local output) double precision array, dimension
locc(ja+min(m,n)-1). this array contains the scalar factors
distributed matrix a.

PDGERFS()

this array contains the local pieces of the distributed
factors of the matrix sub( a ) = p * l * u as computed b

PDGERQ2()

tau     (local output) double precision array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PDGERQF()

tau     (local output) double precision array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PDGESV()

sub( a ) to be factored. on exit, this array contains the
local pieces of the factors l and u from the factorizatio
stored.

PDGESVX()

1. if fact = 'e', real scaling factors are computed to equilibrat
trans = 'n':  diag(r)*a*diag(c)     *inv(diag(c))*x = diag(r)*b

PDGETF2()

distributed matrix sub( a ). on exit, this array contains
the local pieces of the factors l and u from the factoriza
not stored.

PDGETRF()

distributed matrix sub( a ) to be factored. on exit, this
array contains the local pieces of the factors l and u fro
ments of l are not stored.

PDGETRS()

memory to an array of dimension (lld_a, locc(ja+n-1)).
on entry, this array contains the local pieces of the factors
diagonal elements of l are not stored.

PDGGQRF()

taua    (local output) double precision array, dimension
locc(ja+min(n,m)-1). this array contains the scalar factors
orthogonal matrix q. taua is tied to the distributed matrix

PDGGRQF()

taua    (local output) double precision array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar
taua is tied to the distributed matrix a (see further

PDLABRD()

tauq    (local output) double precision array dimension
locc(ja+min(m,n)-1). the scalar factors of the elementar
is tied to the distributed matrix a. see further details.

PDLAHRD()

tau     (local output) double precision array, dimension locc(ja+n-2)
the scalar factors of the elementary reflectors (see furthe

PDLAQGE()

sub( a ) = a(ia:ia+m-1,ja:ja+n-1) using the row and scaling
factors in the vectors r and c
notes

PDLAQSY()

pdlaqsy equilibrates a symmetric distributed matrix
sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the scaling factors in th

PDLATRD()

tau     (local output) double precision array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PDLATRZ()

tau     (local output) double precision array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PDORG2L()

tau     (local input) double precision array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PDORG2R()

tau     (local input) double precision array, dimension locc(ja+k-1).
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PDORGL2()

tau     (local input) double precision array, dimension locr(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PDORGLQ()

tau     (local input) double precision array, dimension locr(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PDORGQL()

tau     (local input) double precision array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PDORGQR()

tau     (local input) double precision array, dimension locc(ja+k-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PDORGR2()

tau     (local input) double precision array, dimension locr(ia+m-1)
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PDORGRQ()

tau     (local input) double precision array, dimension locr(ia+m-1)
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PDORM2L()

tau     (local input) double precision array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PDORM2R()

tau     (local input) double precision array, dimension locc(ja+k-1).
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PDORMHR()

if side = 'l', and locc(ja+n-2) if side = 'r'. this array
contains the scalar factors tau(j) of the elementar
the distributed matrix a.

PDORML2()

tau     (local input) double precision array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PDORMLQ()

tau     (local input) double precision array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PDORMQL()

tau     (local input) double precision array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PDORMQR()

tau     (local input) double precision array, dimension locc(ja+k-1).
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PDORMR2()

tau     (local input) double precision array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PDORMR3()

tau     (local input) double precision array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PDORMRQ()

tau     (local input) double precision array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PDORMRZ()

tau     (local input) double precision array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PDPBSV()

note that permutations are performed on the matrix, so that
the factors returned are different from those returne

PDPBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n real

PDPOCON()

to an array of dimension ( lld_a, locc(ja+n-1) ). on entry,
this array contains the local pieces of the factors l or 
or l*l', as computed by pdpotrf.

PDPOEQU()

(with respect to the two-norm).  sr and sc contain the scale
factors, s(i) = 1/sqrt(a(i,i)), chosen so that the scaled distri
the  diagonal.  this choice of sr and sc puts the condition number

PDPORFS()

to an array of local dimension (lld_af,locc(ja+n-1)).
on entry, this array contains the factors l or u from th
computed by pdpotrf.

PDPOSVX()

1. if fact = 'e', real scaling factors are computed to equilibrat
diag(sr) * a * diag(sc) * inv(diag(sc)) * x = diag(sr) * b

PDPOTRS()

an array of dimension (lld_a, locc(ja+n-1)). on entry, this
array contains the factors l or u from the cholesky facto

PDPTSV()

on exit, this array contains information containing the
factors of the matrix

PDPTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n real

PDSYEVX()

for clustersize > n/sqrt(nprow*npcol) execution time will
grow as the square of the cluster size, all other factors
workspace means less reorthogonalization but faster

PDSYGVX()

for clustersize > n/sqrt(nprow*npcol) execution time will
grow as the square of the cluster size, all other factors
workspace means less reorthogonalization but faster

PDSYNTRD()

tau     (local output) double precision array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PDSYTD2()

tau     (local output) double precision array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PDSYTRD()

tau     (local output) double precision array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PDSYTTRD()

tau     (local output) double precision array, dimension
locq(ja+n-1). this array contains the scalar factors tau o
matrix a.

PDTZRZF()

tau     (local output) double precision array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PSDBSV()

note that permutations are performed on the matrix, so that
the factors returned are different from those returne

PSDBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n real

PSDTSV()

on exit, this array contains information containing the
factors of the matrix
d       (local input/local output) real pointer to local

PSDTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n real

PSGBSV()

note that permutations are performed on the matrix, so that
the factors returned are different from those returne

PSGBTRF()

use partial factors to update remainde

PSGBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n real

PSGEBD2()

tauq    (local output) real array dimension
locc(ja+min(m,n)-1). the scalar factors of the elementar
is tied to the distributed matrix a. see further details.

PSGEBRD()

tauq    (local output) real array dimension
locc(ja+min(m,n)-1). the scalar factors of the elementar
is tied to the distributed matrix a. see further details.

PSGECON()

to an array of dimension ( lld_a, locc(ja+n-1) ). on entry,
this array contains the local pieces of the factors l and 
unit diagonal elements of l are not stored.

PSGEEQU()

m-by-n distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja:ja+n-1) and
reduce its condition number.  r returns the row scale factors and 
each row and column of the distributed matrix b with elements

PSGEHD2()

tau     (local output) real array, dimension locc(ja+n-2)
the scalar factors of the elementary reflectors (see furthe
set to zero. tau is tied to the distributed matrix a.

PSGEHRD()

tau     (local output) real array, dimension locc(ja+n-2)
the scalar factors of the elementary reflectors (see furthe
set to zero. tau is tied to the distributed matrix a.

PSGELQ2()

tau     (local output) real, array, dimension
locr(ia+min(m,n)-1).  this array contains the scalar factors
matrix a.

PSGELQF()

tau     (local output) real, array, dimension
locr(ia+min(m,n)-1).  this array contains the scalar factors
matrix a.

PSGEQL2()

tau     (local output) real, array, dimension locc(ja+n-1)
this array contains the scalar factors of the elementar

PSGEQLF()

tau     (local output) real, array, dimension locc(ja+n-1)
this array contains the scalar factors of the elementar

PSGEQPF()

tau     (local output) real, array, dimension
locc(ja+min(m,n)-1). this array contains the scalar factors
distributed matrix a.

PSGEQR2()

tau     (local output) real, array, dimension
locc(ja+min(m,n)-1). this array contains the scalar factors
distributed matrix a.

PSGEQRF()

tau     (local output) real, array, dimension
locc(ja+min(m,n)-1). this array contains the scalar factors
distributed matrix a.

PSGERFS()

this array contains the local pieces of the distributed
factors of the matrix sub( a ) = p * l * u as computed b

PSGERQ2()

tau     (local output) real, array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PSGERQF()

tau     (local output) real, array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PSGESV()

sub( a ) to be factored. on exit, this array contains the
local pieces of the factors l and u from the factorizatio
stored.

PSGESVX()

1. if fact = 'e', real scaling factors are computed to equilibrat
trans = 'n':  diag(r)*a*diag(c)     *inv(diag(c))*x = diag(r)*b

PSGETF2()

distributed matrix sub( a ). on exit, this array contains
the local pieces of the factors l and u from the factoriza
not stored.

PSGETRF()

distributed matrix sub( a ) to be factored. on exit, this
array contains the local pieces of the factors l and u fro
ments of l are not stored.

PSGETRS()

memory to an array of dimension (lld_a, locc(ja+n-1)).
on entry, this array contains the local pieces of the factors
diagonal elements of l are not stored.

PSGGQRF()

taua    (local output) real, array, dimension
locc(ja+min(n,m)-1). this array contains the scalar factors
orthogonal matrix q. taua is tied to the distributed matrix

PSGGRQF()

taua    (local output) real, array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar
taua is tied to the distributed matrix a (see further

PSLABRD()

tauq    (local output) real array dimension
locc(ja+min(m,n)-1). the scalar factors of the elementar
is tied to the distributed matrix a. see further details.

PSLAHRD()

tau     (local output) real array, dimension locc(ja+n-2)
the scalar factors of the elementary reflectors (see furthe

PSLAQGE()

sub( a ) = a(ia:ia+m-1,ja:ja+n-1) using the row and scaling
factors in the vectors r and c
notes

PSLAQSY()

pslaqsy equilibrates a symmetric distributed matrix
sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the scaling factors in th

PSLATRD()

tau     (local output) real, array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PSLATRZ()

tau     (local output) real, array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PSORG2L()

tau     (local input) real, array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PSORG2R()

tau     (local input) real, array, dimension locc(ja+k-1).
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PSORGL2()

tau     (local input) real, array, dimension locr(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PSORGLQ()

tau     (local input) real, array, dimension locr(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PSORGQL()

tau     (local input) real, array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PSORGQR()

tau     (local input) real, array, dimension locc(ja+k-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PSORGR2()

tau     (local input) real, array, dimension locr(ia+m-1)
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PSORGRQ()

tau     (local input) real, array, dimension locr(ia+m-1)
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PSORM2L()

tau     (local input) real, array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PSORM2R()

tau     (local input) real, array, dimension locc(ja+k-1).
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PSORMHR()

if side = 'l', and locc(ja+n-2) if side = 'r'. this array
contains the scalar factors tau(j) of the elementar
the distributed matrix a.

PSORML2()

tau     (local input) real, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PSORMLQ()

tau     (local input) real, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PSORMQL()

tau     (local input) real, array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PSORMQR()

tau     (local input) real, array, dimension locc(ja+k-1).
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PSORMR2()

tau     (local input) real, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PSORMR3()

tau     (local input) real, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PSORMRQ()

tau     (local input) real, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PSORMRZ()

tau     (local input) real, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PSPBSV()

note that permutations are performed on the matrix, so that
the factors returned are different from those returne

PSPBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n real

PSPOCON()

an array of dimension ( lld_a, locc(ja+n-1) ). on entry, this
array contains the local pieces of the factors l or u fro
l*l', as computed by pspotrf.

PSPOEQU()

(with respect to the two-norm).  sr and sc contain the scale
factors, s(i) = 1/sqrt(a(i,i)), chosen so that the scaled distri
the  diagonal.  this choice of sr and sc puts the condition number

PSPORFS()

to an array of local dimension (lld_af,locc(ja+n-1)).
on entry, this array contains the factors l or u from th
computed by pspotrf.

PSPOSVX()

1. if fact = 'e', real scaling factors are computed to equilibrat
diag(sr) * a * diag(sc) * inv(diag(sc)) * x = diag(sr) * b

PSPOTRS()

an array of dimension (lld_a, locc(ja+n-1)). on entry, this
array contains the factors l or u from the cholesky facto

PSPTSV()

on exit, this array contains information containing the
factors of the matrix

PSPTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n real

PSSYEVX()

for clustersize > n/sqrt(nprow*npcol) execution time will
grow as the square of the cluster size, all other factors
workspace means less reorthogonalization but faster

PSSYGVX()

for clustersize > n/sqrt(nprow*npcol) execution time will
grow as the square of the cluster size, all other factors
workspace means less reorthogonalization but faster

PSSYNTRD()

tau     (local output) real, array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PSSYTD2()

tau     (local output) real, array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PSSYTRD()

tau     (local output) real, array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PSSYTTRD()

tau     (local output) real, array, dimension
locq(ja+n-1). this array contains the scalar factors tau o
matrix a.

PSTZRZF()

tau     (local output) real, array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PZDBSV()

note that permutations are performed on the matrix, so that
the factors returned are different from those returne

PZDBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n complex

PZDTSV()

on exit, this array contains information containing the
factors of the matrix
d       (local input/local output) complex*16 pointer to local

PZDTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n complex

PZGBSV()

note that permutations are performed on the matrix, so that
the factors returned are different from those returne

PZGBTRF()

use partial factors to update remainde

PZGBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n complex

PZGEBD2()

tauq    (local output) complex*16 array dimension
locc(ja+min(m,n)-1). the scalar factors of the elementar
tied to the distributed matrix a. see further details.

PZGEBRD()

tauq    (local output) complex*16 array dimension
locc(ja+min(m,n)-1). the scalar factors of the elementar
tied to the distributed matrix a. see further details.

PZGECON()

to an array of dimension ( lld_a, locc(ja+n-1) ). on entry,
this array contains the local pieces of the factors l and 
unit diagonal elements of l are not stored.

PZGEEQU()

m-by-n distributed matrix sub( a ) = a(ia:ia+n-1,ja:ja:ja+n-1) and
reduce its condition number.  r returns the row scale factors and 
each row and column of the distributed matrix b with elements

PZGEHD2()

tau     (local output) complex*16 array, dimension locc(ja+n-2)
the scalar factors of the elementary reflectors (see furthe
set to zero. tau is tied to the distributed matrix a.

PZGEHRD()

tau     (local output) complex*16 array, dimension locc(ja+n-2)
the scalar factors of the elementary reflectors (see furthe
set to zero. tau is tied to the distributed matrix a.

PZGELQ2()

tau     (local output) complex*16, array, dimension
locr(ia+min(m,n)-1).  this array contains the scalar factors
matrix a.

PZGELQF()

tau     (local output) complex*16, array, dimension
locr(ia+min(m,n)-1).  this array contains the scalar factors
matrix a.

PZGEQL2()

tau     (local output) complex*16, array, dimension locc(ja+n-1)
this array contains the scalar factors of the elementar

PZGEQLF()

tau     (local output) complex*16, array, dimension locc(ja+n-1)
this array contains the scalar factors of the elementar

PZGEQPF()

tau     (local output) complex*16, array, dimension
locc(ja+min(m,n)-1). this array contains the scalar factors
distributed matrix a.

PZGEQR2()

tau     (local output) complex*16, array, dimension
locc(ja+min(m,n)-1). this array contains the scalar factors
distributed matrix a.

PZGEQRF()

tau     (local output) complex*16, array, dimension
locc(ja+min(m,n)-1). this array contains the scalar factors
distributed matrix a.

PZGERFS()

this array contains the local pieces of the distributed
factors of the matrix sub( a ) = p * l * u as computed b

PZGERQ2()

tau     (local output) complex*16, array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PZGERQF()

tau     (local output) complex*16, array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PZGESV()

sub( a ) to be factored. on exit, this array contains the
local pieces of the factors l and u from the factorizatio
stored.

PZGESVX()

1. if fact = 'e', real scaling factors are computed to equilibrat
trans = 'n':  diag(r)*a*diag(c)     *inv(diag(c))*x = diag(r)*b

PZGETF2()

distributed matrix sub( a ). on exit, this array contains
the local pieces of the factors l and u from the factoriza
not stored.

PZGETRF()

distributed matrix sub( a ) to be factored. on exit, this
array contains the local pieces of the factors l and u fro
ments of l are not stored.

PZGETRS()

memory to an array of dimension (lld_a, locc(ja+n-1)).
on entry, this array contains the local pieces of the factors
diagonal elements of l are not stored.

PZGGQRF()

taua    (local output) complex*16, array, dimension
locc(ja+min(n,m)-1). this array contains the scalar factors
matrix q. taua is tied to the distributed matrix a. (see

PZGGRQF()

taua    (local output) complex*16, array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar
tied to the distributed matrix a (see further details).

PZHEEVX()

for clustersize > n/sqrt(nprow*npcol) execution time will
grow as the square of the cluster size, all other factors
workspace means less reorthogonalization but faster

PZHEGVX()

for clustersize > n/sqrt(nprow*npcol) execution time will
grow as the square of the cluster size, all other factors
workspace means less reorthogonalization but faster

PZHENTRD()

tau     (local output) complex*16, array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PZHETD2()

tau     (local output) complex*16, array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PZHETRD()

tau     (local output) complex*16, array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PZHETTRD()

tau     (local output) complex*16, array, dimension
locq(ja+n-1). this array contains the scalar factors tau o
matrix a.

PZLABRD()

tauq    (local output) complex*16 array dimension
locc(ja+min(m,n)-1). the scalar factors of the elementar
tied to the distributed matrix a. see further details.

PZLAHRD()

tau     (local output) complex*16 array, dimension locc(ja+n-2)
the scalar factors of the elementary reflectors (see furthe

PZLAQGE()

sub( a ) = a(ia:ia+m-1,ja:ja+n-1) using the row and scaling
factors in the vectors r and c
notes

PZLAQSY()

pzlaqsy equilibrates a symmetric distributed matrix
sub( a ) = a(ia:ia+n-1,ja:ja+n-1) using the scaling factors in th

PZLATRD()

tau     (local output) complex*16, array, dimension
locc(ja+n-1). this array contains the scalar factors tau o
matrix a.

PZLATRZ()

tau     (local output) complex*16, array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PZPBSV()

note that permutations are performed on the matrix, so that
the factors returned are different from those returne

PZPBTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n complex

PZPOCON()

an array of dimension ( lld_a, locc(ja+n-1) ). on entry, this
array contains the local pieces of the factors l or u fro
l*l', as computed by pzpotrf.

PZPOEQU()

(with respect to the two-norm).  sr and sc contain the scale
factors, s(i) = 1/sqrt(a(i,i)), chosen so that the scaled distri
the  diagonal.  this choice of sr and sc puts the condition number

PZPORFS()

to an array of local dimension (lld_af,locc(ja+n-1)).
on entry, this array contains the factors l or u from th
computed by pzpotrf.

PZPOSVX()

1. if fact = 'e', real scaling factors are computed to equilibrat
diag(sr) * a * diag(sc) * inv(diag(sc)) * x = diag(sr) * b

PZPOTRS()

an array of dimension (lld_a, locc(ja+n-1)). on entry, this
array contains the factors l or u from the cholesky facto

PZPTSV()

on exit, this array contains information containing the
factors of the matrix

PZPTTRS()

where a(1:n, ja:ja+n-1) is the matrix used to produce the factors
a(1:n, ja:ja+n-1) is an n-by-n complex

PZTZRZF()

tau     (local output) complex*16, array, dimension locr(ia+m-1)
this array contains the scalar factors of the elementar

PZUNG2L()

tau     (local input) complex*16, array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PZUNG2R()

tau     (local input) complex*16, array, dimension locc(ja+k-1).
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PZUNGL2()

tau     (local input) complex*16, array, dimension locr(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PZUNGLQ()

tau     (local input) complex*16, array, dimension locr(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PZUNGQL()

tau     (local input) complex*16, array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PZUNGQR()

tau     (local input) complex*16, array, dimension locc(ja+k-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PZUNGR2()

tau     (local input) complex*16, array, dimension locr(ia+m-1)
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PZUNGRQ()

tau     (local input) complex*16, array, dimension locr(ia+m-1)
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PZUNM2L()

tau     (local input) complex*16, array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PZUNM2R()

tau     (local input) complex*16, array, dimension locc(ja+k-1).
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PZUNMHR()

if side = 'l', and locc(ja+n-2) if side = 'r'. this array
contains the scalar factors tau(j) of the elementar
the distributed matrix a.

PZUNML2()

tau     (local input) complex*16, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PZUNMLQ()

tau     (local input) complex*16, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PZUNMQL()

tau     (local input) complex*16, array, dimension locc(ja+n-1)
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PZUNMQR()

tau     (local input) complex*16, array, dimension locc(ja+k-1).
this array contains the scalar factors tau(j) of th
tau is tied to the distributed matrix a.

PZUNMR2()

tau     (local input) complex*16, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PZUNMR3()

tau     (local input) complex*16, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PZUNMRQ()

tau     (local input) complex*16, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

PZUNMRZ()

tau     (local input) complex*16, array, dimension locc(ia+k-1).
this array contains the scalar factors tau(i) of th
tau is tied to the distributed matrix a.

SDTTRSV()

u * x = b,  u**t * x = b,  or  u**h * x = b,
with factors of the tridiagonal matrix a from the lu factorizatio

SSTEIN2()

compute lu factors with partial pivoting  ( pt = lu 

ZDTTRSV()

u * x = b,  u**t * x = b,  or  u**h * x = b,
with factors of the tridiagonal matrix a from the lu factorizatio

Fahey

CLAMSH()

implemented by:  m. Fahey, may 28, 199
=====================================================================

CLANV2()

implemented by mark r. Fahey, may 28, 199
=====================================================================

CLAREF()

implemented by:  m. Fahey, may 28, 199
=====================================================================

PCLACONSB()

implemented by:  m. Fahey, may 28, 199
=====================================================================

PCLACP3()

implemented by:  m. Fahey, may 28, 199
=====================================================================

PCLASMSUB()

implemented by:  m. Fahey, may 28, 199
=====================================================================

PCLAWIL()

implemented by:  m. Fahey, may 28, 199
=====================================================================

PCTREVC()

implemented by mark r. Fahey, june, 200
=====================================================================

PZLACONSB()

implemented by:  m. Fahey, may 28, 199
=====================================================================

PZLACP3()

implemented by:  m. Fahey, may 28, 199
=====================================================================

PZLASMSUB()

implemented by:  m. Fahey, may 28, 199
=====================================================================

PZLAWIL()

implemented by:  m. Fahey, may 28, 199
=====================================================================

PZTREVC()

implemented by mark r. Fahey, june, 200
=====================================================================

ZLAMSH()

implemented by:  m. Fahey, may 28, 199
=====================================================================

ZLANV2()

implemented by mark r. Fahey, may 28, 199
=====================================================================

ZLAREF()

implemented by:  m. Fahey, may 28, 199
=====================================================================

fail