|
PLASMA
2.4.5
PLASMA - Parallel Linear Algebra for Scalable Multi-core Architectures
|

Go to the source code of this file.
Functions | |
| int | CORE_sormqr (int side, int trans, int M, int N, int K, int IB, float *A, int LDA, float *T, int LDT, float *C, int LDC, float *WORK, int LDWORK) |
| void | QUARK_CORE_sormqr (Quark *quark, Quark_Task_Flags *task_flags, int side, int trans, int m, int n, int k, int ib, int nb, float *A, int lda, float *T, int ldt, float *C, int ldc) |
| void | CORE_sormqr_quark (Quark *quark) |
PLASMA core_blas kernel PLASMA is a software package provided by Univ. of Tennessee, Univ. of California Berkeley and Univ. of Colorado Denver
Definition in file core_sormqr.c.
| int CORE_sormqr | ( | int | side, |
| int | trans, | ||
| int | M, | ||
| int | N, | ||
| int | K, | ||
| int | IB, | ||
| float * | A, | ||
| int | LDA, | ||
| float * | T, | ||
| int | LDT, | ||
| float * | C, | ||
| int | LDC, | ||
| float * | WORK, | ||
| int | LDWORK | ||
| ) |
CORE_sormqr overwrites the general complex M-by-N tile C with
SIDE = 'L' SIDE = 'R'
TRANS = 'N': Q * C C * Q TRANS = 'C': Q**T * C C * Q**T
where Q is a complex unitary matrix defined as the product of k elementary reflectors
Q = H(1) H(2) . . . H(k)
as returned by CORE_sgeqrt. Q is of order M if SIDE = 'L' and of order N if SIDE = 'R'.
| [in] | side |
|
| [in] | trans |
|
| [in] | M | The number of rows of the tile C. M >= 0. |
| [in] | N | The number of columns of the tile C. N >= 0. |
| [in] | K | The number of elementary reflectors whose product defines the matrix Q. If SIDE = PlasmaLeft, M >= K >= 0; if SIDE = PlasmaRight, N >= K >= 0. |
| [in] | IB | The inner-blocking size. IB >= 0. |
| [in] | A | Dimension: (LDA,K) The i-th column must contain the vector which defines the elementary reflector H(i), for i = 1,2,...,k, as returned by CORE_sgeqrt in the first k columns of its array argument A. |
| [in] | LDA | The leading dimension of the array A. If SIDE = PlasmaLeft, LDA >= max(1,M); if SIDE = PlasmaRight, LDA >= max(1,N). |
| [out] | T | The IB-by-K triangular factor T of the block reflector. T is upper triangular by block (economic storage); The rest of the array is not referenced. |
| [in] | LDT | The leading dimension of the array T. LDT >= IB. |
| [in,out] | C | On entry, the M-by-N tile C. On exit, C is overwritten by Q*C or Q**T*C or C*Q**T or C*Q. |
| [in] | LDC | The leading dimension of the array C. LDC >= max(1,M). |
| [in,out] | WORK | On exit, if INFO = 0, WORK(1) returns the optimal LDWORK. |
| [in] | LDWORK | The dimension of the array WORK. If SIDE = PlasmaLeft, LDWORK >= max(1,N); if SIDE = PlasmaRight, LDWORK >= max(1,M). |
| PLASMA_SUCCESS | successful exit |
| <0 | if -i, the i-th argument had an illegal value |
Definition at line 108 of file core_sormqr.c.
References coreblas_error, lapack_const, max, min, PLASMA_SUCCESS, PlasmaColumnwise, PlasmaForward, PlasmaLeft, PlasmaNoTrans, PlasmaRight, and PlasmaTrans.

| void CORE_sormqr_quark | ( | Quark * | quark | ) |
Definition at line 257 of file core_sormqr.c.
References A, C, CORE_sormqr(), quark_unpack_args_14, side, T, and trans.


| void QUARK_CORE_sormqr | ( | Quark * | quark, |
| Quark_Task_Flags * | task_flags, | ||
| int | side, | ||
| int | trans, | ||
| int | m, | ||
| int | n, | ||
| int | k, | ||
| int | ib, | ||
| int | nb, | ||
| float * | A, | ||
| int | lda, | ||
| float * | T, | ||
| int | ldt, | ||
| float * | C, | ||
| int | ldc | ||
| ) |
Definition at line 224 of file core_sormqr.c.
References CORE_sormqr_quark(), DAG_CORE_UNMQR, INOUT, INPUT, QUARK_Insert_Task(), QUARK_REGION_L, SCRATCH, and VALUE.

