Function: gemv_ex

aclError aclblasGemvEx(aclTransType transA,

int m,

int n,

const void *alpha,

const void *a,

int lda,

aclDataType dataTypeA,

const void *x,

int incx,

aclDataType dataTypeX,

const void *beta,

void *y,

int incy,

aclDataType dataTypeY,

aclComputeType type,

aclrtStream stream);

ret = acl.blas.gemv_ex(trans_a, m, n,alpha, a, lda, data_type_a, x, incx, data_type_x, beta, y, incy, data_type_y, type, stream )

Computes matrix-vector multiplication: y = αAx + βy. The data types of the input and output data are set using arguments. This API is asynchronous.

trans_a: int, whether matrix A is transposed.

m: int, number of rows in matrix A (row-major first in data storing).

n: int, number of columns in matrix A.

alpha: int, data pointer address of Scalar α used for executing the multiplication operation.

a: int, data pointer address of matrix A.

lda: int, leading dimension of matrix A. For row-major transpose, lda is the number of columns of matrix A. This parameter is reserved and can only be set to –1.

data_type_a: int, data type of matrix A.

x: int, data pointer address of vector x.

incx: int, increment between successive x elements. This parameter is reserved and can only be set to –1.

data_type_x: int, data type of vector x.

beta: int, data pointer address of Scalar β used for executing the multiplication operation.

y: int, data pointer address of vector y.

incy: int, increment between successive y elements. This parameter is reserved and can only be set to –1.

data_type_y: int, data type of vector y.

type: int, compute precision. High precision is defaulted.

stream: int, stream where the operator is executed.

ret: int, error code.

• 0 indicates success.
• Other values indicate failure.

Only the following data types of A, x and y are supported, and data type of α and β is the same as that of y.

Data Type of A Data Type of x Data Type of y

aclFloat16 aclFloat16 aclFloat16

aclFloat16 aclFloat16 float(float32)

int8_t int8_t float(float32)

int8_t int8_t int32_t

For details about the API call sequence, see API Call Sequence.