hipBLASLt Datatypes Reference

hipblasltDatatype_t

enum hipblasltDatatype_t

Indicates the precision width of data stored in a blas type.

Values:

enumerator HIPBLASLT_R_16F

16 bit floating point, real

enumerator HIPBLASLT_R_32F

32 bit floating point, real

enumerator HIPBLASLT_R_64F

64 bit floating point, real

enumerator HIPBLASLT_C_16F

16 bit floating point, complex

enumerator HIPBLASLT_C_32F

32 bit floating point, complex

enumerator HIPBLASLT_C_64F

64 bit floating point, complex

enumerator HIPBLASLT_R_8I

8 bit signed integer, real

enumerator HIPBLASLT_R_8U

8 bit unsigned integer, real

enumerator HIPBLASLT_R_32I

32 bit signed integer, real

enumerator HIPBLASLT_R_32U

32 bit unsigned integer, real

enumerator HIPBLASLT_C_8I

8 bit signed integer, complex

enumerator HIPBLASLT_C_8U

8 bit unsigned integer, complex

enumerator HIPBLASLT_C_32I

32 bit signed integer, complex

enumerator HIPBLASLT_C_32U

32 bit unsigned integer, complex

enumerator HIPBLASLT_R_16B

16 bit bfloat, real

enumerator HIPBLASLT_C_16B

16 bit bfloat, complex

enumerator HIPBLASLT_R_8F_E4M3

8 bit floating point in E4M3 format, real

enumerator HIPBLASLT_R_8F_E5M2

8 bit floating point in E5M2 format, real

enumerator HIPBLASLT_DATATYPE_INVALID

Invalid datatype value, do not use

hipblasLtEpilogue_t

enum hipblasLtEpilogue_t

Specify the enum type to set the postprocessing options for the epilogue.

Values:

enumerator HIPBLASLT_EPILOGUE_DEFAULT

No special postprocessing, just scale and quantize the results if necessary.

enumerator HIPBLASLT_EPILOGUE_RELU

Apply ReLU point-wise transform to the results:(x:=max(x, 0))

enumerator HIPBLASLT_EPILOGUE_BIAS

Apply (broadcast) bias from the bias vector. Bias vector length must match matrix D rows, and it must be packed (such as stride between vector elements is 1). Bias vector is broadcast to all columns and added before applying the final postprocessing.

enumerator HIPBLASLT_EPILOGUE_RELU_BIAS

Apply bias and then ReLU transform.

enumerator HIPBLASLT_EPILOGUE_GELU

Apply GELU point-wise transform to the results (x:=GELU(x)).

enumerator HIPBLASLT_EPILOGUE_GELU_BIAS

Apply Bias and then GELU transform.

enumerator HIPBLASLT_EPILOGUE_GELU_AUX

Output GEMM results before applying GELU transform.

enumerator HIPBLASLT_EPILOGUE_GELU_AUX_BIAS

Output GEMM results after applying bias but before applying GELU transform.

enumerator HIPBLASLT_EPILOGUE_DGELU

Apply gradient GELU transform. Requires additional aux input.

enumerator HIPBLASLT_EPILOGUE_DGELU_BGRAD

Apply gradient GELU transform and bias gradient to the results. Requires additional aux input.

enumerator HIPBLASLT_EPILOGUE_BGRADA

Apply bias gradient to A and output gemm result.

enumerator HIPBLASLT_EPILOGUE_BGRADB

Apply bias gradient to B and output gemm result.

hipblasLtPointerMode_t

enum hipblasLtPointerMode_t

Pointer mode to use for alpha.

Values:

enumerator HIPBLASLT_POINTER_MODE_HOST

enumerator HIPBLASLT_POINTER_MODE_ALPHA_DEVICE_VECTOR_BETA_HOST

targets host memory

hipblasLtHandle_t

typedef void *hipblasLtHandle_t

Handle to the hipBLASLt library context queue.

The hipblasLtHandle_t type is a pointer type to an opaque structure holding the hipBLASLt library context. Use the following functions to manipulate this library context:

hipblasLtCreate(): To initialize the hipBLASLt library context and return a handle to an opaque structure holding the hipBLASLt library context. hipblasLtDestroy(): To destroy a previously created hipBLASLt library context descriptor and release the resources.

hipblasLtMatmulAlgo_t

struct hipblasLtMatmulAlgo_t

Description of the matrix multiplication algorithm.

This is an opaque structure holding the description of the matrix multiplication algorithm. This structure can be trivially serialized and later restored for use with the same version of hipBLASLt library to save on selecting the right configuration again.

hipblasLtMatmulDesc_t

typedef hipblasLtMatmulDescOpaque_t *hipblasLtMatmulDesc_t

Descriptor of the matrix multiplication operation.

This is a pointer to an opaque structure holding the description of the matrix multiplication operation hipblasLtMatmul(). Use the following functions to manipulate this descriptor: hipblasLtMatmulDescCreate(): To create one instance of the descriptor. hipblasLtMatmulDescDestroy(): To destroy a previously created descriptor and release the resources.

hipblasLtMatmulDescAttributes_t

enum hipblasLtMatmulDescAttributes_t

Specify the attributes that define the specifics of the matrix multiply operation.

Values:

enumerator HIPBLASLT_MATMUL_DESC_TRANSA

Specifies the type of transformation operation that should be performed on matrix A. Default value is HIPBLAS_OP_N (for example, non-transpose operation). See hipblasOperation_t. Data Type:int32_t

enumerator HIPBLASLT_MATMUL_DESC_TRANSB

Specifies the type of transformation operation that should be performed on matrix B. Default value is HIPBLAS_OP_N (for example, non-transpose operation). See hipblasOperation_t. Data Type:int32_t

enumerator HIPBLASLT_MATMUL_DESC_EPILOGUE

Epilogue function. See hipblasLtEpilogue_t. Default value is: HIPBLASLT_EPILOGUE_DEFAULT. Data Type: uint32_t

enumerator HIPBLASLT_MATMUL_DESC_BIAS_POINTER

Bias or Bias gradient vector pointer in the device memory. Data Type:void* /const void*

enumerator HIPBLASLT_MATMUL_DESC_BIAS_DATA_TYPE

Type of the bias vector in the device memory. Can be set same as D matrix type or Scale type. Bias case: see HIPBLASLT_EPILOGUE_BIAS. Data Type:int32_t based on hipblasltDatatype_t

enumerator HIPBLASLT_MATMUL_DESC_A_SCALE_POINTER

Device pointer to the scale factor value that converts data in matrix A to the compute data type range. The scaling factor must have the same type as the compute type. If not specified, or set to NULL, the scaling factor is assumed to be 1. If set for an unsupported matrix data, scale, and compute type combination, calling hipblasLtMatmul() will return HIPBLAS_INVALID_VALUE. Default value: NULL Data Type: void* /const void*

enumerator HIPBLASLT_MATMUL_DESC_B_SCALE_POINTER

Equivalent to HIPBLASLT_MATMUL_DESC_A_SCALE_POINTER for matrix B. Default value: NULL Type: void* /const void*

enumerator HIPBLASLT_MATMUL_DESC_C_SCALE_POINTER

Equivalent to HIPBLASLT_MATMUL_DESC_A_SCALE_POINTER for matrix C. Default value: NULL Type: void* /const void*

enumerator HIPBLASLT_MATMUL_DESC_D_SCALE_POINTER

Equivalent to HIPBLASLT_MATMUL_DESC_A_SCALE_POINTER for matrix D. Default value: NULL Type: void* /const void*

enumerator HIPBLASLT_MATMUL_DESC_EPILOGUE_AUX_SCALE_POINTER

Equivalent to HIPBLASLT_MATMUL_DESC_A_SCALE_POINTER for matrix AUX. Default value: NULL Type: void* /const void*

enumerator HIPBLASLT_MATMUL_DESC_EPILOGUE_AUX_POINTER

Epilogue auxiliary buffer pointer in the device memory. Data Type:void* /const void*

enumerator HIPBLASLT_MATMUL_DESC_EPILOGUE_AUX_LD

The leading dimension of the epilogue auxiliary buffer pointer in the device memory. Data Type:int64_t

enumerator HIPBLASLT_MATMUL_DESC_EPILOGUE_AUX_BATCH_STRIDE

The batch stride of the epilogue auxiliary buffer pointer in the device memory. Data Type:int64_t

enumerator HIPBLASLT_MATMUL_DESC_POINTER_MODE

Specifies alpha and beta are passed by reference, whether they are scalars on the host or on the device, or device vectors. Default value is: HIPBLASLT_POINTER_MODE_HOST (i.e., on the host). Data Type: int32_t based on hipblasLtPointerMode_t

enumerator HIPBLASLT_MATMUL_DESC_MAX

hipblasLtMatmulHeuristicResult_t

struct hipblasLtMatmulHeuristicResult_t

Description of the matrix multiplication algorithm.

This is a descriptor that holds the configured matrix multiplication algorithm descriptor and its runtime properties. This structure can be trivially serialized and later restored for use with the same version of hipBLASLt library to save on selecting the right configuration again.

hipblasLtMatmulPreference_t

typedef hipblasLtMatmulPreferenceOpaque_t *hipblasLtMatmulPreference_t

Descriptor of the matrix multiplication preference.

This is a pointer to an opaque structure holding the description of the preferences for hipblasLtMatmulAlgoGetHeuristic() configuration. Use the following functions to manipulate this descriptor: hipblasLtMatmulPreferenceCreate(): To create one instance of the descriptor. hipblasLtMatmulPreferenceDestroy(): To destroy a previously created descriptor and release the resources.

hipblasLtMatmulPreferenceAttributes_t

enum hipblasLtMatmulPreferenceAttributes_t

It is an enumerated type used to apply algorithm search preferences while fine-tuning the heuristic function.

Values:

enumerator HIPBLASLT_MATMUL_PREF_SEARCH_MODE

Search mode. Data Type: uint32_t

enumerator HIPBLASLT_MATMUL_PREF_MAX_WORKSPACE_BYTES

Maximum allowed workspace memory. Default is 0 (no workspace memory allowed). Data Type: uint64_t

enumerator HIPBLASLT_MATMUL_PREF_MAX

hipblasLtMatrixLayout_t

typedef hipblasLtMatrixLayoutOpaque_t *hipblasLtMatrixLayout_t

Descriptor of the matrix layout.

This is a pointer to an opaque structure holding the description of a matrix layout. Use the following functions to manipulate this descriptor: hipblasLtMatrixLayoutCreate(): To create one instance of the descriptor. hipblasLtMatrixLayoutDestroy(): To destroy a previously created descriptor and release the resources.

hipblasLtMatrixLayoutAttribute_t

enum hipblasLtMatrixLayoutAttribute_t

Specify the attributes that define the details of the matrix.

Values:

enumerator HIPBLASLT_MATRIX_LAYOUT_BATCH_COUNT

Number of batch of this matrix. Default value is 1. Data Type: int32_t

enumerator HIPBLASLT_MATRIX_LAYOUT_STRIDED_BATCH_OFFSET

Stride (in elements) to the next matrix for the strided batch operation. Default value is 0. Data Type: int64_t

enumerator HIPBLASLT_MATRIX_LAYOUT_TYPE

Data type, see hipblasltDataType_t.

uint32_t

enumerator HIPBLASLT_MATRIX_LAYOUT_ORDER

Memory order of the data, see cublasLtOrder_t.

int32_t, default: HIPBLASLT_ORDER_COL

enumerator HIPBLASLT_MATRIX_LAYOUT_ROWS

Number of rows.

Usually only values that can be expressed as int32_t are supported.

uint64_t

enumerator HIPBLASLT_MATRIX_LAYOUT_COLS

Number of columns.

Usually only values that can be expressed as int32_t are supported.

uint64_t

enumerator HIPBLASLT_MATRIX_LAYOUT_LD

Matrix leading dimension.

For HIPBLASLT_ORDER_COL this is stride (in elements) of matrix column, for more details and documentation for other memory orders see documentation for cublasLtOrder_t values.

Currently only non-negative values are supported, must be large enough so that matrix memory locations are not overlapping (e.g. greater or equal to HIPBLASLT_MATRIX_LAYOUT_ROWS in case of HIPBLASLT_ORDER_COL).

int64_t;