aclblasS8gemm

Applicability

Product	Supported
Atlas 350 Accelerator Card	√
Atlas A3 training product / Atlas A3 inference product	√
Atlas A2 training product / Atlas A2 inference product	√
Atlas 200I/500 A2 inference product	√
Atlas inference product	√
Atlas training product	√

Function Usage

Computes matrix-matrix multiplication: C = αAB + βC. The input data type is int8_t, and the output data type is int32_t. This API is asynchronous.

Prototype

      
       
         
         
           aclError aclblasS8gemm(aclTransType transA,
aclTransType transB,
aclTransType transC,
int m,
int n,
int k,
const int32_t *alpha,
const int8_t *matrixA,
int lda,
const int8_t *matrixB,
int ldb,
const int32_t *beta,
int32_t *matrixC,
int ldc,
aclComputeType type,
aclrtStream stream)

          

        

      
     

Parameters

Parameter	Input/Output	Description
transA	Input	Transpose flag of matrix A. For details about the type definition, see aclTransType.
transB	Input	Transpose flag of matrix B. For details about the type definition, see aclTransType.
transC	Input	Transpose flag of matrix C. For details about the type definition, see aclTransType. Currently, the value can only be ACL_TRANS_N.
m	Input	Number of rows in matrix A and matrix C.
n	Input	Number of columns in matrix B and matrix C.
k	Input	Number of columns in matrix A and number of rows in matrix B.
alpha	Input	Pointer to scalar α used for executing the multiplication operation.
matrixA	Input	Pointer to matrix A.
lda	Input	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.
matrixB	Input	Pointer to matrix B.
ldb	Input	Leading dimension of matrix B. For row-major transpose, ldb is the number of columns of matrix B. This parameter is reserved and can only be set to –1.
beta	Input	Pointer to scalar β used for executing the multiplication operation.
matrixC	Input and output	Pointer to matrix C.
ldc	Input	Leading dimension of matrix C. This parameter is reserved and can only be set to –1.
type	Input	Computation precision type (high precision by default). For details about the type definition, see aclComputeType.
stream	Input	Stream where the operator is executed. For details about the type definition, see aclrtStream.

Returns

0 on success; else, failure. For details, see aclError.

Parent topic: CBLAS APIs