dpftrs (l) - Linux Manuals

dpftrs: solves a system of linear equations A*X = B with a symmetric positive definite matrix A using the Cholesky factorization A = U**T*U or A = L*L**T computed by DPFTRF

NAME

DPFTRS - solves a system of linear equations A*X = B with a symmetric positive definite matrix A using the Cholesky factorization A = U**T*U or A = L*L**T computed by DPFTRF

SYNOPSIS

SUBROUTINE DPFTRS(
TRANSR, UPLO, N, NRHS, A, B, LDB, INFO )

    
CHARACTER TRANSR, UPLO

    
INTEGER INFO, LDB, N, NRHS

    
DOUBLE PRECISION A( 0: * ), B( LDB, * )

PURPOSE

DPFTRS solves a system of linear equations A*X = B with a symmetric positive definite matrix A using the Cholesky factorization A = U**T*U or A = L*L**T computed by DPFTRF.

ARGUMENTS

TRANSR (input) CHARACTER
= aqNaq: The Normal TRANSR of RFP A is stored;
= aqTaq: The Transpose TRANSR of RFP A is stored.
UPLO (input) CHARACTER

= aqUaq: Upper triangle of RFP A is stored;
= aqLaq: Lower triangle of RFP A is stored.
N (input) INTEGER
The order of the matrix A. N >= 0.
NRHS (input) INTEGER
The number of right hand sides, i.e., the number of columns of the matrix B. NRHS >= 0.
A (input) DOUBLE PRECISION array, dimension ( N*(N+1)/2 ).
The triangular factor U or L from the Cholesky factorization of RFP A = U**T*U or RFP A = L*L**T, as computed by DPFTRF. See note below for more details about RFP A.
B (input/output) DOUBLE PRECISION array, dimension (LDB,NRHS)
On entry, the right hand side matrix B. On exit, the solution matrix X.
LDB (input) INTEGER
The leading dimension of the array B. LDB >= max(1,N).
INFO (output) INTEGER
= 0: successful exit
< 0: if INFO = -i, the i-th argument had an illegal value

FURTHER DETAILS

We first consider Rectangular Full Packed (RFP) Format when N is even. We give an example where N = 6.

 AP is Upper             AP is Lower

 00 01 02 03 04 05       00

 11 12 13 14 15       10 11

 22 23 24 25       20 21 22

    33 34 35       30 31 32 33

       44 45       40 41 42 43 44

          55       50 51 52 53 54 55
Let TRANSR = aqNaq. RFP holds AP as follows:
For UPLO = aqUaq the upper trapezoid A(0:5,0:2) consists of the last three columns of AP upper. The lower triangle A(4:6,0:2) consists of the transpose of the first three columns of AP upper.
For UPLO = aqLaq the lower trapezoid A(1:6,0:2) consists of the first three columns of AP lower. The upper triangle A(0:2,0:2) consists of the transpose of the last three columns of AP lower.
This covers the case N even and TRANSR = aqNaq.

 RFP A                   RFP A

03 04 05                33 43 53

13 14 15                00 44 54

23 24 25                10 11 55

33 34 35                20 21 22

00 44 45                30 31 32

01 11 55                40 41 42

02 12 22                50 51 52
Now let TRANSR = aqTaq. RFP A in both UPLO cases is just the transpose of RFP A above. One therefore gets:

   RFP A                   RFP A

03 13 23 33 00 01 02    33 00 10 20 30 40 50

04 14 24 34 44 11 12    43 44 11 21 31 41 51

05 15 25 35 45 55 22    53 54 55 22 32 42 52
We first consider Rectangular Full Packed (RFP) Format when N is odd. We give an example where N = 5.

AP is Upper                 AP is Lower

 00 01 02 03 04              00

 11 12 13 14              10 11

 22 23 24              20 21 22

    33 34              30 31 32 33

       44              40 41 42 43 44
Let TRANSR = aqNaq. RFP holds AP as follows:
For UPLO = aqUaq the upper trapezoid A(0:4,0:2) consists of the last three columns of AP upper. The lower triangle A(3:4,0:1) consists of the transpose of the first two columns of AP upper.
For UPLO = aqLaq the lower trapezoid A(0:4,0:2) consists of the first three columns of AP lower. The upper triangle A(0:1,1:2) consists of the transpose of the last two columns of AP lower.
This covers the case N odd and TRANSR = aqNaq.

 RFP A                   RFP A

02 03 04                00 33 43

12 13 14                10 11 44

22 23 24                20 21 22

00 33 34                30 31 32

01 11 44                40 41 42
Now let TRANSR = aqTaq. RFP A in both UPLO cases is just the transpose of RFP A above. One therefore gets:

   RFP A                   RFP A

02 12 22 00 01             00 10 20 30 40 50

03 13 23 33 11             33 11 21 31 41 51

04 14 24 34 44             43 44 22 32 42 52