SetBType

Function

Sets the position, data format, data type, and transpose status of matrix B. These settings must be consistent with those in the kernel.

Prototype

int32_t SetBType(TPosition pos, CubeFormat type, DataType dataType, bool isTrans = false)

Parameters

**Table 1** Parameters
Parameter	Input/Output	Description
pos	Input	Position in the buffer where matrix B is located, which can be TPosition::GM, TPosition::VECOUT, or TPosition::TSCM.
type	Input	Data format of matrix B, which can be CubeFormat::ND or CubeFormat::NZ.
dataType	Input	Data type of matrix B, which can be DataType::DT_FLOAT, DataType::DT_FLOAT16, DataType::DT_BFLOAT16, DataType::DT_INT8, or DataType::DT_INT4.
isTrans	Input	Whether matrix B should be transposed. Values: true: Matrix B is transposed. false: Matrix B is not transposed.

Returns

-1: setting failed; 0: setting succeeded.

Restrictions

None

Example

auto ascendcPlatform = platform_ascendc::PlatformAscendC(context->GetPlatformInfo());
matmul_tiling::MatmulApiTiling tiling(ascendcPlatform); 
tiling.SetAType(matmul_tiling::TPosition::GM, matmul_tiling::CubeFormat::ND, matmul_tiling::DataType::DT_FLOAT16); 
// Set matrix B, with the buffer located in the GM, data format of ND, data type of bfloat16, and transpose disabled by default.
tiling.SetBType(matmul_tiling::TPosition::GM, matmul_tiling::CubeFormat::ND, matmul_tiling::DataType::DT_FLOAT16);
tiling.SetCType(matmul_tiling::TPosition::GM, matmul_tiling::CubeFormat::ND, matmul_tiling::DataType::DT_FLOAT);   
tiling.SetBiasType(matmul_tiling::TPosition::GM, matmul_tiling::CubeFormat::ND, matmul_tiling::DataType::DT_FLOAT);   
tiling.SetShape(1024, 1024, 1024);   
tiling.SetOrgShape(1024, 1024, 1024);   
tiling.SetBias(true);   
tiling.SetBufferSpace(-1, -1, -1);
optiling::TCubeTiling tilingData;   
int ret = tiling.GetTiling(tilingData);

Parent topic: Matmul Tiling