Rate This Document

Findability

Accuracy

Completeness

Readability

Data Structures

This section describes the data types of functions in the KML_BLAS library.

Data Types

Data Type	Description
s	Single-precision floating-point type
d	Double-precision floating-point type
c	Single-precision complex type
z	Double-precision complex type
b	Half-precision floating-point type
h	Half-precision floating-point type

Macro Definition

1	#define CBLAS_INDEX size_t

Enumeration Types

Table 1 describes the enumeration types of KML_BLAS.

**Table 1** Enumeration types and syntax
Enumeration Type	Description	Syntax
CBLAS_ORDER	Defines whether the matrix is stored in row-major or column-major order.	enum CBLAS_ORDER { CblasRowMajor=101, CblasColMajor=102 }CBLAS_ORDER;
CBLAS_TRANSPOSE	Defines whether to transpose the original or conjugated matrix.	enum CBLAS_TRANSPOSE { CblasNoTrans=111, CblasTrans=112, CblasConjTrans=113, CblasConjNoTrans=114 } CBLAS_TRANSPOSE;
CBLAS_UPLO	Defines whether to use the upper triangle or lower triangle of the matrix.	enum CBLAS_UPLO { CblasUpper=121, CblasLower=122 } CBLAS_UPLO;
CBLAS_DIAG	Defines whether the matrix diagonal is a unit element.	enum CBLAS_DIAG { CblasNonUnit=131, CblasUnit=132 } CBLAS_DIAG;
CBLAS_SIDE	Defines the order (left or right) of Hermitian matrix multiplication.	enum CBLAS_SIDE { CblasLeft=141, CblasRight=142 } CBLAS_SIDE;
CBLAS_IDENTIFIER	Defines whether matrix A or B is used for pack execution.	typedef enum CBLAS_IDENTIFIER { CblasA = 151, CblasB = 152 } CBLAS_IDENTIFIER;

Parent topic: KML_BLAS Library Functions