hip.hipblas#
(No short description)
- Attributes:
- hipblasVersionMajor (
int): Macro constant.
- hipblaseVersionMinor (
int): Macro constant.
- hipblasVersionMinor (
int): Macro constant.
- hipblasVersionPatch (
int): Macro constant.
- FP_NAN:
(undocumented)
- FP_INFINITE:
(undocumented)
- FP_ZERO:
(undocumented)
- FP_SUBNORMAL:
(undocumented)
- FP_NORMAL:
(undocumented)
- FP_NAN:
(undocumented)
- FP_INFINITE:
(undocumented)
- FP_ZERO:
(undocumented)
- FP_SUBNORMAL:
(undocumented)
- FP_NORMAL:
(undocumented)
- FP_NAN:
(undocumented)
- FP_INFINITE:
(undocumented)
- FP_ZERO:
(undocumented)
- FP_SUBNORMAL:
(undocumented)
- FP_NORMAL:
(undocumented)
- FP_NAN:
(undocumented)
- FP_INFINITE:
(undocumented)
- FP_ZERO:
(undocumented)
- FP_SUBNORMAL:
(undocumented)
- FP_NORMAL:
(undocumented)
- FP_NAN:
(undocumented)
- FP_INFINITE:
(undocumented)
- FP_ZERO:
(undocumented)
- FP_SUBNORMAL:
(undocumented)
- FP_NORMAL:
(undocumented)
- HIP_SUCCESS:
(undocumented)
- HIP_ERROR_INVALID_VALUE:
(undocumented)
- HIP_ERROR_NOT_INITIALIZED:
(undocumented)
- HIP_ERROR_LAUNCH_OUT_OF_RESOURCES:
(undocumented)
- HIP_SUCCESS:
(undocumented)
- HIP_ERROR_INVALID_VALUE:
(undocumented)
- HIP_ERROR_NOT_INITIALIZED:
(undocumented)
- HIP_ERROR_LAUNCH_OUT_OF_RESOURCES:
(undocumented)
- HIP_SUCCESS:
(undocumented)
- HIP_ERROR_INVALID_VALUE:
(undocumented)
- HIP_ERROR_NOT_INITIALIZED:
(undocumented)
- HIP_ERROR_LAUNCH_OUT_OF_RESOURCES:
(undocumented)
- HIP_SUCCESS:
(undocumented)
- HIP_ERROR_INVALID_VALUE:
(undocumented)
- HIP_ERROR_NOT_INITIALIZED:
(undocumented)
- HIP_ERROR_LAUNCH_OUT_OF_RESOURCES:
(undocumented)
- hipblasVersionMajor (
- class hip.hipblas.hipblasBfloat16(*args, **kwargs)#
Bases:
PointerPython wrapper for cdef class chipblas.hipblasBfloat16.
Python wrapper for cdef class chipblas.hipblasBfloat16.
If this type is initialized via its
__init__method, it allocates a member of the underlying C type and destroys it again if the wrapper type is deallocated.This type also serves as adapter when appearing as argument type in a function signature. In this case, the type can further be initialized from a number of Python objects:
None:This will set the
self._ptrattribute toNULL.int:Interprets the integer value as pointer address and writes it to
self._ptr. No ownership is transferred.-
Takes the pointer address
pyobj.valueand writes it toself._ptr. No ownership is transferred. objectthat implements the CUDA Array Interface protocol:Takes the integer-valued pointer address, i.e. the first entry of the
datatuple frompyobj’s member__cuda_array_interface__and writes it toself._ptr.objectthat implements the Python buffer protocol:If the object represents a simple contiguous array, writes the
Py_bufferassociated withpyobjtoself._py_buffer, sets theself._py_buffer_acquiredflag toTrue, and writesself._py_buffer.bufto the data pointerself._ptr.-
Takes the pointer address
pyobj._ptrand writes it toself._ptr. No ownership is transferred.
Type checks are performed in the above order.
- C Attributes:
- _ptr (C type
void *, protected): Stores a pointer to the data of the original Python object.
- _is_ptr_owner (C type
bint, protected): If this wrapper is the owner of the underlying data.
- _py_buffer (C type ``Py_buffer`, protected):
Stores a pointer to the data of the original Python object.
- _py_buffer_acquired (C type
bint, protected): Stores a pointer to the data of the original Python object.
- _ptr (C type
- static PROPERTIES()#
- __getitem__(key, /)#
Return self[key].
- __init__()#
Constructor type hipblasBfloat16.
Constructor for type hipblasBfloat16.
- as_c_void_p(self)#
Returns the data’s address as
ctypes.c_void_pNote:Implemented as function to not collide with autogenerated property names.
- c_sizeof(self)#
Returns the size of the underlying C type in bytes. Note:
Implemented as function to not collide with autogenerated property names.
- createRef(self) Pointer#
Creates are reference to this pointer.
Returns a
Pointerthat stores the address of this `~.Pointer’s data pointer.- Note:
No ownership information is transferred.
- data#
(undocumented)
- static fromObj(pyobj)#
Creates a hipblasBfloat16 from a Python object.
Derives a hipblasBfloat16 from the given Python object
pyobj. In casepyobjis itself anhipblasBfloat16reference, this method returns it directly. No newhipblasBfloat16is created in this case.
- get_data(self, i)#
Get value
dataof(<chipblas.hipblasBfloat16*>self._ptr)[i].
- is_ptr_null#
If data pointer is NULL.
- set_data(self, i, unsigned short value)#
Set value
dataof(<chipblas.hipblasBfloat16*>self._ptr)[i].
- class hip.hipblas.hipblasComplex(*args, **kwargs)#
Bases:
PointerPython wrapper for cdef class chipblas.hipblasComplex.
Python wrapper for cdef class chipblas.hipblasComplex.
If this type is initialized via its
__init__method, it allocates a member of the underlying C type and destroys it again if the wrapper type is deallocated.This type also serves as adapter when appearing as argument type in a function signature. In this case, the type can further be initialized from a number of Python objects:
None:This will set the
self._ptrattribute toNULL.int:Interprets the integer value as pointer address and writes it to
self._ptr. No ownership is transferred.-
Takes the pointer address
pyobj.valueand writes it toself._ptr. No ownership is transferred. objectthat implements the CUDA Array Interface protocol:Takes the integer-valued pointer address, i.e. the first entry of the
datatuple frompyobj’s member__cuda_array_interface__and writes it toself._ptr.objectthat implements the Python buffer protocol:If the object represents a simple contiguous array, writes the
Py_bufferassociated withpyobjtoself._py_buffer, sets theself._py_buffer_acquiredflag toTrue, and writesself._py_buffer.bufto the data pointerself._ptr.-
Takes the pointer address
pyobj._ptrand writes it toself._ptr. No ownership is transferred.
Type checks are performed in the above order.
- C Attributes:
- _ptr (C type
void *, protected): Stores a pointer to the data of the original Python object.
- _is_ptr_owner (C type
bint, protected): If this wrapper is the owner of the underlying data.
- _py_buffer (C type ``Py_buffer`, protected):
Stores a pointer to the data of the original Python object.
- _py_buffer_acquired (C type
bint, protected): Stores a pointer to the data of the original Python object.
- _ptr (C type
- static PROPERTIES()#
- __getitem__(key, /)#
Return self[key].
- __init__()#
Constructor type hipblasComplex.
Constructor for type hipblasComplex.
- as_c_void_p(self)#
Returns the data’s address as
ctypes.c_void_pNote:Implemented as function to not collide with autogenerated property names.
- c_sizeof(self)#
Returns the size of the underlying C type in bytes. Note:
Implemented as function to not collide with autogenerated property names.
- createRef(self) Pointer#
Creates are reference to this pointer.
Returns a
Pointerthat stores the address of this `~.Pointer’s data pointer.- Note:
No ownership information is transferred.
- static fromObj(pyobj)#
Creates a hipblasComplex from a Python object.
Derives a hipblasComplex from the given Python object
pyobj. In casepyobjis itself anhipblasComplexreference, this method returns it directly. No newhipblasComplexis created in this case.
- get_x(self, i)#
Get value
xof(<chipblas.hipblasComplex*>self._ptr)[i].
- get_y(self, i)#
Get value
yof(<chipblas.hipblasComplex*>self._ptr)[i].
- is_ptr_null#
If data pointer is NULL.
- set_x(self, i, float value)#
Set value
xof(<chipblas.hipblasComplex*>self._ptr)[i].
- set_y(self, i, float value)#
Set value
yof(<chipblas.hipblasComplex*>self._ptr)[i].
- x#
(undocumented)
- y#
(undocumented)
- class hip.hipblas.hipblasDoubleComplex(*args, **kwargs)#
Bases:
PointerPython wrapper for cdef class chipblas.hipblasDoubleComplex.
Python wrapper for cdef class chipblas.hipblasDoubleComplex.
If this type is initialized via its
__init__method, it allocates a member of the underlying C type and destroys it again if the wrapper type is deallocated.This type also serves as adapter when appearing as argument type in a function signature. In this case, the type can further be initialized from a number of Python objects:
None:This will set the
self._ptrattribute toNULL.int:Interprets the integer value as pointer address and writes it to
self._ptr. No ownership is transferred.-
Takes the pointer address
pyobj.valueand writes it toself._ptr. No ownership is transferred. objectthat implements the CUDA Array Interface protocol:Takes the integer-valued pointer address, i.e. the first entry of the
datatuple frompyobj’s member__cuda_array_interface__and writes it toself._ptr.objectthat implements the Python buffer protocol:If the object represents a simple contiguous array, writes the
Py_bufferassociated withpyobjtoself._py_buffer, sets theself._py_buffer_acquiredflag toTrue, and writesself._py_buffer.bufto the data pointerself._ptr.-
Takes the pointer address
pyobj._ptrand writes it toself._ptr. No ownership is transferred.
Type checks are performed in the above order.
- C Attributes:
- _ptr (C type
void *, protected): Stores a pointer to the data of the original Python object.
- _is_ptr_owner (C type
bint, protected): If this wrapper is the owner of the underlying data.
- _py_buffer (C type ``Py_buffer`, protected):
Stores a pointer to the data of the original Python object.
- _py_buffer_acquired (C type
bint, protected): Stores a pointer to the data of the original Python object.
- _ptr (C type
- static PROPERTIES()#
- __getitem__(key, /)#
Return self[key].
- __init__()#
Constructor type hipblasDoubleComplex.
Constructor for type hipblasDoubleComplex.
- as_c_void_p(self)#
Returns the data’s address as
ctypes.c_void_pNote:Implemented as function to not collide with autogenerated property names.
- c_sizeof(self)#
Returns the size of the underlying C type in bytes. Note:
Implemented as function to not collide with autogenerated property names.
- createRef(self) Pointer#
Creates are reference to this pointer.
Returns a
Pointerthat stores the address of this `~.Pointer’s data pointer.- Note:
No ownership information is transferred.
- static fromObj(pyobj)#
Creates a hipblasDoubleComplex from a Python object.
Derives a hipblasDoubleComplex from the given Python object
pyobj. In casepyobjis itself anhipblasDoubleComplexreference, this method returns it directly. No newhipblasDoubleComplexis created in this case.
- get_x(self, i)#
Get value
xof(<chipblas.hipblasDoubleComplex*>self._ptr)[i].
- get_y(self, i)#
Get value
yof(<chipblas.hipblasDoubleComplex*>self._ptr)[i].
- is_ptr_null#
If data pointer is NULL.
- set_x(self, i, double value)#
Set value
xof(<chipblas.hipblasDoubleComplex*>self._ptr)[i].
- set_y(self, i, double value)#
Set value
yof(<chipblas.hipblasDoubleComplex*>self._ptr)[i].
- x#
(undocumented)
- y#
(undocumented)
- class hip.hipblas.hipblasStatus_t(value)#
Bases:
_hipblasStatus_t__Basehipblas status codes definition
- Attributes:
- HIPBLAS_STATUS_SUCCESS:
Function succeeds
- HIPBLAS_STATUS_NOT_INITIALIZED:
HIPBLAS library not initialized
- HIPBLAS_STATUS_ALLOC_FAILED:
resource allocation failed
- HIPBLAS_STATUS_INVALID_VALUE:
unsupported numerical value was passed to function
- HIPBLAS_STATUS_MAPPING_ERROR:
access to GPU memory space failed
- HIPBLAS_STATUS_EXECUTION_FAILED:
GPU program failed to execute
- HIPBLAS_STATUS_INTERNAL_ERROR:
an internal HIPBLAS operation failed
- HIPBLAS_STATUS_NOT_SUPPORTED:
function not implemented
- HIPBLAS_STATUS_ARCH_MISMATCH:
architecture mismatch
- HIPBLAS_STATUS_HANDLE_IS_NULLPTR:
hipBLAS handle is null pointer
- HIPBLAS_STATUS_INVALID_ENUM:
unsupported enum value was passed to function
- HIPBLAS_STATUS_UNKNOWN:
back-end returned an unsupported status code
- HIPBLAS_STATUS_SUCCESS = 0#
- HIPBLAS_STATUS_NOT_INITIALIZED = 1#
- HIPBLAS_STATUS_ALLOC_FAILED = 2#
- HIPBLAS_STATUS_INVALID_VALUE = 3#
- HIPBLAS_STATUS_MAPPING_ERROR = 4#
- HIPBLAS_STATUS_EXECUTION_FAILED = 5#
- HIPBLAS_STATUS_INTERNAL_ERROR = 6#
- HIPBLAS_STATUS_NOT_SUPPORTED = 7#
- HIPBLAS_STATUS_ARCH_MISMATCH = 8#
- HIPBLAS_STATUS_HANDLE_IS_NULLPTR = 9#
- HIPBLAS_STATUS_INVALID_ENUM = 10#
- HIPBLAS_STATUS_UNKNOWN = 11#
- static ctypes_type()#
The type of the enum constants as ctypes type.
- class hip.hipblas.hipblasPointerMode_t(value)#
Bases:
_hipblasPointerMode_t__BaseIndicates if scalar pointers are on host or device. This is used for scalars alpha and beta and for scalar function return values.
- Attributes:
- HIPBLAS_POINTER_MODE_HOST:
Scalar values affected by this variable will be located on the host.
- HIPBLAS_POINTER_MODE_DEVICE:
Scalar values affected by this variable will be located on the device.
- HIPBLAS_POINTER_MODE_HOST = 0#
- HIPBLAS_POINTER_MODE_DEVICE = 1#
- static ctypes_type()#
The type of the enum constants as ctypes type.
- class hip.hipblas.hipblasOperation_t(value)#
Bases:
_hipblasOperation_t__BaseUsed to specify whether the matrix is to be transposed or not.
- Attributes:
- HIPBLAS_OP_N:
Operate with the matrix.
- HIPBLAS_OP_T:
Operate with the transpose of the matrix.
- HIPBLAS_OP_C:
Operate with the conjugate transpose of the matrix.
- HIPBLAS_OP_N = 111#
- HIPBLAS_OP_T = 112#
- HIPBLAS_OP_C = 113#
- static ctypes_type()#
The type of the enum constants as ctypes type.
- class hip.hipblas.hipblasFillMode_t(value)#
Bases:
_hipblasFillMode_t__BaseUsed by the Hermitian, symmetric and triangular matrix routines to specify whether the upper or lower triangle is being referenced.
- Attributes:
- HIPBLAS_FILL_MODE_UPPER:
Upper triangle
- HIPBLAS_FILL_MODE_LOWER:
Lower triangle
- HIPBLAS_FILL_MODE_FULL:
(undocumented)
- HIPBLAS_FILL_MODE_UPPER = 121#
- HIPBLAS_FILL_MODE_LOWER = 122#
- HIPBLAS_FILL_MODE_FULL = 123#
- static ctypes_type()#
The type of the enum constants as ctypes type.
- class hip.hipblas.hipblasDiagType_t(value)#
Bases:
_hipblasDiagType_t__BaseIt is used by the triangular matrix routines to specify whether the matrix is unit triangular.
- Attributes:
- HIPBLAS_DIAG_NON_UNIT:
Non-unit triangular.
- HIPBLAS_DIAG_UNIT:
Unit triangular.
- HIPBLAS_DIAG_NON_UNIT = 131#
- HIPBLAS_DIAG_UNIT = 132#
- static ctypes_type()#
The type of the enum constants as ctypes type.
- class hip.hipblas.hipblasSideMode_t(value)#
Bases:
_hipblasSideMode_t__BaseIndicates the side matrix A is located relative to matrix B during multiplication.
- Attributes:
- HIPBLAS_SIDE_LEFT:
(undocumented)
- HIPBLAS_SIDE_RIGHT:
(undocumented)
- HIPBLAS_SIDE_BOTH:
(undocumented)
- HIPBLAS_SIDE_LEFT = 141#
- HIPBLAS_SIDE_RIGHT = 142#
- HIPBLAS_SIDE_BOTH = 143#
- static ctypes_type()#
The type of the enum constants as ctypes type.
- class hip.hipblas.hipblasDatatype_t(value)#
Bases:
_hipblasDatatype_t__BaseIndicates the precision of data used. hipblasDatatype_t is deprecated as of hipBLAS 2.0.0 and will be removed in a future release as generally replaced by hipDataType.
- Attributes:
- HIPBLAS_R_16F:
16 bit floating point, real
- HIPBLAS_R_32F:
32 bit floating point, real
- HIPBLAS_R_64F:
64 bit floating point, real
- HIPBLAS_C_16F:
16 bit floating point, complex
- HIPBLAS_C_32F:
32 bit floating point, complex
- HIPBLAS_C_64F:
64 bit floating point, complex
- HIPBLAS_R_8I:
8 bit signed integer, real
- HIPBLAS_R_8U:
8 bit unsigned integer, real
- HIPBLAS_R_32I:
32 bit signed integer, real
- HIPBLAS_R_32U:
32 bit unsigned integer, real
- HIPBLAS_C_8I:
8 bit signed integer, complex
- HIPBLAS_C_8U:
8 bit unsigned integer, complex
- HIPBLAS_C_32I:
32 bit signed integer, complex
- HIPBLAS_C_32U:
32 bit unsigned integer, complex
- HIPBLAS_R_16B:
16 bit bfloat, real
- HIPBLAS_C_16B:
16 bit bfloat, complex
- HIPBLAS_DATATYPE_INVALID:
Invalid datatype value, do not use
- HIPBLAS_R_16F = 150#
- HIPBLAS_R_32F = 151#
- HIPBLAS_R_64F = 152#
- HIPBLAS_C_16F = 153#
- HIPBLAS_C_32F = 154#
- HIPBLAS_C_64F = 155#
- HIPBLAS_R_8I = 160#
- HIPBLAS_R_8U = 161#
- HIPBLAS_R_32I = 162#
- HIPBLAS_R_32U = 163#
- HIPBLAS_C_8I = 164#
- HIPBLAS_C_8U = 165#
- HIPBLAS_C_32I = 166#
- HIPBLAS_C_32U = 167#
- HIPBLAS_R_16B = 168#
- HIPBLAS_C_16B = 169#
- HIPBLAS_DATATYPE_INVALID = 255#
- static ctypes_type()#
The type of the enum constants as ctypes type.
- class hip.hipblas.hipblasComputeType_t(value)#
Bases:
_hipblasComputeType_t__Base- Attributes:
- HIPBLAS_COMPUTE_16F:
compute will be at least 16-bit precision
- HIPBLAS_COMPUTE_16F_PEDANTIC:
compute will be exactly 16-bit precision
- HIPBLAS_COMPUTE_32F:
compute will be at least 32-bit precision
- HIPBLAS_COMPUTE_32F_PEDANTIC:
compute will be exactly 32-bit precision
- HIPBLAS_COMPUTE_32F_FAST_16F:
32-bit input can use 16-bit compute
- HIPBLAS_COMPUTE_32F_FAST_16BF:
32-bit input can is bf16 compute
- HIPBLAS_COMPUTE_32F_FAST_TF32:
(undocumented)
- HIPBLAS_COMPUTE_64F:
compute will be at least 64-bit precision
- HIPBLAS_COMPUTE_64F_PEDANTIC:
compute will be exactly 64-bit precision
- HIPBLAS_COMPUTE_32I:
compute will be at least 32-bit integer precision
- HIPBLAS_COMPUTE_32I_PEDANTIC:
compute will be exactly 32-bit integer precision
- HIPBLAS_COMPUTE_16F = 0#
- HIPBLAS_COMPUTE_16F_PEDANTIC = 1#
- HIPBLAS_COMPUTE_32F = 2#
- HIPBLAS_COMPUTE_32F_PEDANTIC = 3#
- HIPBLAS_COMPUTE_32F_FAST_16F = 4#
- HIPBLAS_COMPUTE_32F_FAST_16BF = 5#
- HIPBLAS_COMPUTE_32F_FAST_TF32 = 6#
- HIPBLAS_COMPUTE_64F = 7#
- HIPBLAS_COMPUTE_64F_PEDANTIC = 8#
- HIPBLAS_COMPUTE_32I = 9#
- HIPBLAS_COMPUTE_32I_PEDANTIC = 10#
- static ctypes_type()#
The type of the enum constants as ctypes type.
- class hip.hipblas.hipblasGemmAlgo_t(value)#
Bases:
_hipblasGemmAlgo_t__BaseIndicates if layer is active with bitmask.
- Attributes:
- HIPBLAS_GEMM_DEFAULT:
enumerator rocblas_gemm_algo_standard
- HIPBLAS_GEMM_DEFAULT = 160#
- static ctypes_type()#
The type of the enum constants as ctypes type.
- class hip.hipblas.hipblasAtomicsMode_t(value)#
Bases:
_hipblasAtomicsMode_t__BaseIndicates if atomics operations are allowed. Not allowing atomic operations may generally improve determinism and repeatability of results at a cost of performance. By default, the rocBLAS backend will allow atomic operations while the cuBLAS backend will disallow atomic operations. See backend documentation for more detail.
- Attributes:
- HIPBLAS_ATOMICS_NOT_ALLOWED:
Algorithms will refrain from atomics where applicable.
- HIPBLAS_ATOMICS_ALLOWED:
Algorithms will take advantage of atomics where applicable.
- HIPBLAS_ATOMICS_NOT_ALLOWED = 0#
- HIPBLAS_ATOMICS_ALLOWED = 1#
- static ctypes_type()#
The type of the enum constants as ctypes type.
- class hip.hipblas.hipblasGemmFlags_t(value)#
Bases:
_hipblasGemmFlags_t__BaseControl flags passed into gemm ex with flags algorithms. Only relevant with rocBLAS backend. See rocBLAS documentation for more information.
- Attributes:
- HIPBLAS_GEMM_FLAGS_NONE:
Default empty flags
- HIPBLAS_GEMM_FLAGS_USE_CU_EFFICIENCY:
(undocumented)
- HIPBLAS_GEMM_FLAGS_FP16_ALT_IMPL:
enumerator rocblas_gemm_flags_fp16_alt_impl
- HIPBLAS_GEMM_FLAGS_CHECK_SOLUTION_INDEX:
(undocumented)
- HIPBLAS_GEMM_FLAGS_FP16_ALT_IMPL_RNZ:
(undocumented)
- HIPBLAS_GEMM_FLAGS_NONE = 0#
- HIPBLAS_GEMM_FLAGS_USE_CU_EFFICIENCY = 2#
- HIPBLAS_GEMM_FLAGS_FP16_ALT_IMPL = 4#
- HIPBLAS_GEMM_FLAGS_CHECK_SOLUTION_INDEX = 8#
- HIPBLAS_GEMM_FLAGS_FP16_ALT_IMPL_RNZ = 16#
- static ctypes_type()#
The type of the enum constants as ctypes type.
- hip.hipblas.hipblasSetStream(handle, streamId)#
Set stream for handle */
- Args:
- handle (
Pointer/object): (undocumented)
- streamId (
ihipStream_t/object): (undocumented)
- handle (
- hip.hipblas.hipblasSetPointerMode(handle, mode)#
Set hipblas pointer mode */
- Args:
- handle (
Pointer/object): (undocumented)
- mode (
hipblasPointerMode_t): (undocumented)
- handle (
- hip.hipblas.hipblasSetVector(int n, int elemSize, x, int incx, y, int incy)#
copy vector from host to device
- Args:
- n (
int) – IN: [int] number of elements in the vector
- elemSize (
int) – IN: [int] Size of both vectors in bytes
- x (
Pointer/object) – IN: pointer to vector on the host
- incx (
int) – IN: [int] specifies the increment for the elements of the vector
- y (
Pointer/object) – OUT: pointer to vector on the device
- incy (
int) – IN: [int] specifies the increment for the elements of the vector
- n (
- hip.hipblas.hipblasGetVector(int n, int elemSize, x, int incx, y, int incy)#
copy vector from device to host
- Args:
- n (
int) – IN: [int] number of elements in the vector
- elemSize (
int) – IN: [int] Size of both vectors in bytes
- x (
Pointer/object) – IN: pointer to vector on the device
- incx (
int) – IN: [int] specifies the increment for the elements of the vector
- y (
Pointer/object) – OUT: pointer to vector on the host
- incy (
int) – IN: [int] specifies the increment for the elements of the vector
- n (
- hip.hipblas.hipblasSetMatrix(int rows, int cols, int elemSize, AP, int lda, BP, int ldb)#
copy matrix from host to device
- Args:
- rows (
int) – IN: [int] number of rows in matrices
- cols (
int) – IN: [int] number of columns in matrices
- elemSize (
int) – IN: [int] number of bytes per element in the matrix
- AP (
Pointer/object) – IN: pointer to matrix on the host
- lda (
int) – IN: [int] specifies the leading dimension of A, lda >= rows
- BP (
Pointer/object) – OUT: pointer to matrix on the GPU
- ldb (
int) – IN: [int] specifies the leading dimension of B, ldb >= rows
- rows (
- hip.hipblas.hipblasGetMatrix(int rows, int cols, int elemSize, AP, int lda, BP, int ldb)#
copy matrix from device to host
- Args:
- rows (
int) – IN: [int] number of rows in matrices
- cols (
int) – IN: [int] number of columns in matrices
- elemSize (
int) – IN: [int] number of bytes per element in the matrix
- AP (
Pointer/object) – IN: pointer to matrix on the GPU
- lda (
int) – IN: [int] specifies the leading dimension of A, lda >= rows
- BP (
Pointer/object) – OUT: pointer to matrix on the host
- ldb (
int) – IN: [int] specifies the leading dimension of B, ldb >= rows
- rows (
- hip.hipblas.hipblasSetVectorAsync(int n, int elemSize, x, int incx, y, int incy, stream)#
asynchronously copy vector from host to device
hipblasSetVectorAsync copies a vector from pinned host memory to device memory asynchronously. Memory on the host must be allocated with hipHostMalloc or the transfer will be synchronous.
- Args:
- n (
int) – IN: [int] number of elements in the vector
- elemSize (
int) – IN: [int] number of bytes per element in the matrix
- x (
Pointer/object) – IN: pointer to vector on the host
- incx (
int) – IN: [int] specifies the increment for the elements of the vector
- y (
Pointer/object) – OUT: pointer to vector on the device
- incy (
int) – IN: [int] specifies the increment for the elements of the vector
- stream (
ihipStream_t/object) – IN: specifies the stream into which this transfer request is queued
- n (
- hip.hipblas.hipblasGetVectorAsync(int n, int elemSize, x, int incx, y, int incy, stream)#
asynchronously copy vector from device to host
hipblasGetVectorAsync copies a vector from pinned host memory to device memory asynchronously. Memory on the host must be allocated with hipHostMalloc or the transfer will be synchronous.
- Args:
- n (
int) – IN: [int] number of elements in the vector
- elemSize (
int) – IN: [int] number of bytes per element in the matrix
- x (
Pointer/object) – IN: pointer to vector on the device
- incx (
int) – IN: [int] specifies the increment for the elements of the vector
- y (
Pointer/object) – OUT: pointer to vector on the host
- incy (
int) – IN: [int] specifies the increment for the elements of the vector
- stream (
ihipStream_t/object) – IN: specifies the stream into which this transfer request is queued
- n (
- hip.hipblas.hipblasSetMatrixAsync(int rows, int cols, int elemSize, AP, int lda, BP, int ldb, stream)#
asynchronously copy matrix from host to device
hipblasSetMatrixAsync copies a matrix from pinned host memory to device memory asynchronously. Memory on the host must be allocated with hipHostMalloc or the transfer will be synchronous.
- Args:
- rows (
int) – IN: [int] number of rows in matrices
- cols (
int) – IN: [int] number of columns in matrices
- elemSize (
int) – IN: [int] number of bytes per element in the matrix
- AP (
Pointer/object) – IN: pointer to matrix on the host
- lda (
int) – IN: [int] specifies the leading dimension of A, lda >= rows
- BP (
Pointer/object) – OUT: pointer to matrix on the GPU
- ldb (
int) – IN: [int] specifies the leading dimension of B, ldb >= rows
- stream (
ihipStream_t/object) – IN: specifies the stream into which this transfer request is queued
- rows (
- hip.hipblas.hipblasGetMatrixAsync(int rows, int cols, int elemSize, AP, int lda, BP, int ldb, stream)#
asynchronously copy matrix from device to host
hipblasGetMatrixAsync copies a matrix from device memory to pinned host memory asynchronously. Memory on the host must be allocated with hipHostMalloc or the transfer will be synchronous.
- Args:
- rows (
int) – IN: [int] number of rows in matrices
- cols (
int) – IN: [int] number of columns in matrices
- elemSize (
int) – IN: [int] number of bytes per element in the matrix
- AP (
Pointer/object) – IN: pointer to matrix on the GPU
- lda (
int) – IN: [int] specifies the leading dimension of A, lda >= rows
- BP (
Pointer/object) – OUT: pointer to matrix on the host
- ldb (
int) – IN: [int] specifies the leading dimension of B, ldb >= rows
- stream (
ihipStream_t/object) – IN: specifies the stream into which this transfer request is queued
- rows (
- hip.hipblas.hipblasSetAtomicsMode(handle, atomics_mode)#
Set hipblasSetAtomicsMode*/
- Args:
- handle (
Pointer/object): (undocumented)
- atomics_mode (
hipblasAtomicsMode_t): (undocumented)
- handle (
- hip.hipblas.hipblasGetAtomicsMode(handle, atomics_mode)#
Get hipblasSetAtomicsMode*/
- hip.hipblas.hipblasIsamax(handle, int n, x, int incx, result)#
BLAS Level 1 API
@{
amax finds the first index of the element of maximum magnitude of a vector x.
Supported precisions in rocBLAS : s,d,c,z.
Supported precisions in cuBLAS : s,d,c,z.
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x.
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of y.
- result (
ListOfInt/object) – INOUT: device pointer or host pointer to store the amax index. return is 0.0 if n, incx<=0.
- handle (
- hip.hipblas.hipblasIdamax(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIcamax(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIzamax(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIcamax_v2(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIzamax_v2(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIsamaxBatched(handle, int n, x, int incx, int batchCount, result)#
BLAS Level 1 API
@{
amaxBatched finds the first index of the element of maximum magnitude of each vector x_i in a batch, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z.
Supported precisions in cuBLAS : No support.
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in each vector x_i
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i. incx must be > 0.
- batchCount (
int) – IN: [int] number of instances in the batch, must be > 0.
- result (
ListOfInt/object) – OUT: device or host array of pointers of batchCount size for results. return is 0 if n, incx<=0.
- handle (
- hip.hipblas.hipblasIdamaxBatched(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIcamaxBatched(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIzamaxBatched(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIcamaxBatched_v2(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIzamaxBatched_v2(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIsamaxStridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
BLAS Level 1 API
@{
amaxStridedBatched finds the first index of the element of maximum magnitude of each vector x_i in a batch, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in each vector x_i
- x (
Pointer/object) – IN: device pointer to the first vector x_1.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i. incx must be > 0.
- stridex (
int) – IN: [hipblasStride] specifies the pointer increment between one x_i and the next x_(i + 1).
- batchCount (
int) – IN: [int] number of instances in the batch
- result (
ListOfInt/object) – OUT: device or host pointer for storing contiguous batchCount results. return is 0 if n <= 0, incx<=0.
- handle (
- hip.hipblas.hipblasIdamaxStridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIcamaxStridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIzamaxStridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIcamaxStridedBatched_v2(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIzamaxStridedBatched_v2(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIsamin(handle, int n, x, int incx, result)#
BLAS Level 1 API
@{
amin finds the first index of the element of minimum magnitude of a vector x.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x.
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of y.
- result (
ListOfInt/object) – INOUT: device pointer or host pointer to store the amin index. return is 0.0 if n, incx<=0.
- handle (
- hip.hipblas.hipblasIdamin(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIcamin(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIzamin(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIcamin_v2(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIzamin_v2(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIsaminBatched(handle, int n, x, int incx, int batchCount, result)#
BLAS Level 1 API
@{
aminBatched finds the first index of the element of minimum magnitude of each vector x_i in a batch, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in each vector x_i
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i. incx must be > 0.
- batchCount (
int) – IN: [int] number of instances in the batch, must be > 0.
- result (
ListOfInt/object) – OUT: device or host pointers to array of batchCount size for results. return is 0 if n, incx<=0.
- handle (
- hip.hipblas.hipblasIdaminBatched(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIcaminBatched(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIzaminBatched(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIcaminBatched_v2(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIzaminBatched_v2(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIsaminStridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
BLAS Level 1 API
@{
aminStridedBatched finds the first index of the element of minimum magnitude of each vector x_i in a batch, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in each vector x_i
- x (
Pointer/object) – IN: device pointer to the first vector x_1.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i. incx must be > 0.
- stridex (
int) – IN: [hipblasStride] specifies the pointer increment between one x_i and the next x_(i + 1)
- batchCount (
int) – IN: [int] number of instances in the batch
- result (
ListOfInt/object) – OUT: device or host pointer to array for storing contiguous batchCount results. return is 0 if n <= 0, incx<=0.
- handle (
- hip.hipblas.hipblasIdaminStridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIcaminStridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIzaminStridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIcaminStridedBatched_v2(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasIzaminStridedBatched_v2(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSasum(handle, int n, x, int incx, result)#
BLAS Level 1 API
@{
asum computes the sum of the magnitudes of elements of a real vector x, or the sum of magnitudes of the real and imaginary parts of elements if x is a complex vector.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x and y.
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x. incx must be > 0.
- result (
Pointer/object) – INOUT: device pointer or host pointer to store the asum product. return is 0.0 if n <= 0.
- handle (
- hip.hipblas.hipblasDasum(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasScasum(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDzasum(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasScasum_v2(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDzasum_v2(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSasumBatched(handle, int n, x, int incx, int batchCount, result)#
BLAS Level 1 API
@{
asumBatched computes the sum of the magnitudes of the elements in a batch of real vectors x_i, or the sum of magnitudes of the real and imaginary parts of elements if x_i is a complex vector, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in each vector x_i
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i. incx must be > 0.
- batchCount (
int) – IN: [int] number of instances in the batch.
- result (
Pointer/object) – OUT: device array or host array of batchCount size for results. return is 0.0 if n, incx<=0.
- handle (
- hip.hipblas.hipblasDasumBatched(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasScasumBatched(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDzasumBatched(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasScasumBatched_v2(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDzasumBatched_v2(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSasumStridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
BLAS Level 1 API
@{
asumStridedBatched computes the sum of the magnitudes of elements of a real vectors x_i, or the sum of magnitudes of the real and imaginary parts of elements if x_i is a complex vector, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in each vector x_i
- x (
Pointer/object) – IN: device pointer to the first vector x_1.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i. incx must be > 0.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1). There are no restrictions placed on stride_x, however the user should take care to ensure that stride_x is of appropriate size, for a typical case this means stride_x >= n * incx.
- batchCount (
int) – IN: [int] number of instances in the batch
- result (
Pointer/object) – OUT: device pointer or host pointer to array for storing contiguous batchCount results. return is 0.0 if n, incx<=0.
- handle (
- hip.hipblas.hipblasDasumStridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasScasumStridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDzasumStridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasScasumStridedBatched_v2(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDzasumStridedBatched_v2(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasHaxpy(handle, int n, alpha, x, int incx, y, int incy)#
BLAS Level 1 API
@{
axpy computes constant alpha multiplied by vector x, plus vector y
y := alpha * x + y
Supported precisions in rocBLAS : h,s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x and y.
- alpha (
Pointer/object) – IN: device pointer or host pointer to specify the scalar alpha.
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- y (
Pointer/object) – OUT: device pointer storing vector y.
- incy (
int) – INOUT: [int] specifies the increment for the elements of y.
- handle (
- hip.hipblas.hipblasSaxpy(handle, int n, alpha, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDaxpy(handle, int n, alpha, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCaxpy(handle, int n, alpha, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZaxpy(handle, int n, alpha, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCaxpy_v2(handle, int n, alpha, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZaxpy_v2(handle, int n, alpha, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasHaxpyBatched(handle, int n, alpha, x, int incx, y, int incy, int batchCount)#
BLAS Level 1 API
@{
axpyBatched compute y := alpha * x + y over a set of batched vectors.
Supported precisions in rocBLAS : h,s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x and y.
- alpha (
Pointer/object) – IN: specifies the scalar alpha.
- x (
ListOfPointer/object) – IN: pointer storing vector x on the GPU.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- y (
ListOfPointer/object) – OUT: pointer storing vector y on the GPU.
- incy (
int) – INOUT: [int] specifies the increment for the elements of y.
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasSaxpyBatched(handle, int n, alpha, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDaxpyBatched(handle, int n, alpha, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCaxpyBatched(handle, int n, alpha, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZaxpyBatched(handle, int n, alpha, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCaxpyBatched_v2(handle, int n, alpha, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZaxpyBatched_v2(handle, int n, alpha, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasHaxpyStridedBatched(handle, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
BLAS Level 1 API
@{
axpyStridedBatched compute y := alpha * x + y over a set of strided batched vectors.
Supported precisions in rocBLAS : h,s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int]
- alpha (
Pointer/object) – IN: specifies the scalar alpha.
- x (
Pointer/object) – IN: pointer storing vector x on the GPU.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- stridex (
int) – IN: [hipblasStride] specifies the increment between vectors of x.
- y (
Pointer/object) – OUT: pointer storing vector y on the GPU.
- incy (
int) – INOUT: [int] specifies the increment for the elements of y.
- stridey (
int) – IN: [hipblasStride] specifies the increment between vectors of y.
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasSaxpyStridedBatched(handle, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasDaxpyStridedBatched(handle, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCaxpyStridedBatched(handle, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZaxpyStridedBatched(handle, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCaxpyStridedBatched_v2(handle, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZaxpyStridedBatched_v2(handle, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasScopy(handle, int n, x, int incx, y, int incy)#
BLAS Level 1 API
@{
copy copies each element x[i] into y[i], for i = 1 , … , n
y := x,
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x to be copied to y.
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- y (
Pointer/object) – OUT: device pointer storing vector y.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- handle (
- hip.hipblas.hipblasDcopy(handle, int n, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCcopy(handle, int n, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZcopy(handle, int n, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCcopy_v2(handle, int n, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZcopy_v2(handle, int n, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasScopyBatched(handle, int n, x, int incx, y, int incy, int batchCount)#
BLAS Level 1 API
@{
copyBatched copies each element x_i[j] into y_i[j], for j = 1 , … , n; i = 1 , … , batchCount
y_i := x_i,
where (x_i, y_i) is the i-th instance of the batch. x_i and y_i are vectors.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in each x_i to be copied to y_i.
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each vector x_i.
- y (
ListOfPointer/object) – OUT: device array of device pointers storing each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of each vector y_i.
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDcopyBatched(handle, int n, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
ListOfPointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCcopyBatched(handle, int n, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZcopyBatched(handle, int n, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCcopyBatched_v2(handle, int n, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZcopyBatched_v2(handle, int n, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasScopyStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
BLAS Level 1 API
@{
copyStridedBatched copies each element x_i[j] into y_i[j], for j = 1 , … , n; i = 1 , … , batchCount
y_i := x_i,
where (x_i, y_i) is the i-th instance of the batch. x_i and y_i are vectors.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in each x_i to be copied to y_i.
- x (
Pointer/object) – IN: device pointer to the first vector (x_1) in the batch.
- incx (
int) – IN: [int] specifies the increments for the elements of vectors x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1). There are no restrictions placed on stride_x, however the user should take care to ensure that stride_x is of appropriate size, for a typical case this means stride_x >= n * incx.
- y (
Pointer/object) – OUT: device pointer to the first vector (y_1) in the batch.
- incy (
int) – IN: [int] specifies the increment for the elements of vectors y_i.
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) and the next one (y_i+1). There are no restrictions placed on stride_y, however the user should take care to ensure that stride_y is of appropriate size, for a typical case this means stride_y >= n * incy. stridey should be non zero.
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDcopyStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCcopyStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZcopyStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- hip.hipblas.hipblasCcopyStridedBatched_v2(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZcopyStridedBatched_v2(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasHdot(handle, int n, x, int incx, y, int incy, result)#
BLAS Level 1 API
@{
dot(u) performs the dot product of vectors x and y
result = x * y;
dotc performs the dot product of the conjugate of complex vector x and complex vector y
result = conjugate (x) * y;
Supported precisions in rocBLAS : h,bf,s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x and y.
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of y.
- y (
Pointer/object) – IN: device pointer storing vector y.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- result (
Pointer/object) – INOUT: device pointer or host pointer to store the dot product. return is 0.0 if n <= 0.
- handle (
- hip.hipblas.hipblasBfdot(handle, int n, x, int incx, y, int incy, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasBfloat16/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasBfloat16/object): (undocumented)
- incy (
int): (undocumented)
- result (
hipblasBfloat16/object): (undocumented)
- handle (
- hip.hipblas.hipblasSdot(handle, int n, x, int incx, y, int incy, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDdot(handle, int n, x, int incx, y, int incy, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCdotc(handle, int n, x, int incx, y, int incy, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- result (
hipblasComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasCdotu(handle, int n, x, int incx, y, int incy, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- result (
hipblasComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasZdotc(handle, int n, x, int incx, y, int incy, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- result (
hipblasDoubleComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasZdotu(handle, int n, x, int incx, y, int incy, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- result (
hipblasDoubleComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasCdotc_v2(handle, int n, x, int incx, y, int incy, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCdotu_v2(handle, int n, x, int incx, y, int incy, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZdotc_v2(handle, int n, x, int incx, y, int incy, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZdotu_v2(handle, int n, x, int incx, y, int incy, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasHdotBatched(handle, int n, x, int incx, y, int incy, int batchCount, result)#
BLAS Level 1 API
@{
dotBatched(u) performs a batch of dot products of vectors x and y
result_i = x_i * y_i;
dotcBatched performs a batch of dot products of the conjugate of complex vector x and complex vector y
result_i = conjugate (x_i) * y_i;
where (x_i, y_i) is the i-th instance of the batch. x_i and y_i are vectors, for i = 1, …, batchCount
Supported precisions in rocBLAS : h,bf,s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in each x_i and y_i.
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- y (
ListOfPointer/object) – IN: device array of device pointers storing each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- batchCount (
int) – IN: [int] number of instances in the batch
- result (
Pointer/object) – INOUT: device array or host array of batchCount size to store the dot products of each batch. return 0.0 for each element if n <= 0.
- handle (
- hip.hipblas.hipblasBfdotBatched(handle, int n, x, int incx, y, int incy, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSdotBatched(handle, int n, x, int incx, y, int incy, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDdotBatched(handle, int n, x, int incx, y, int incy, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCdotcBatched(handle, int n, x, int incx, y, int incy, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCdotuBatched(handle, int n, x, int incx, y, int incy, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZdotcBatched(handle, int n, x, int incx, y, int incy, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZdotuBatched(handle, int n, x, int incx, y, int incy, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCdotcBatched_v2(handle, int n, x, int incx, y, int incy, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCdotuBatched_v2(handle, int n, x, int incx, y, int incy, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZdotcBatched_v2(handle, int n, x, int incx, y, int incy, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZdotuBatched_v2(handle, int n, x, int incx, y, int incy, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasHdotStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount, result)#
BLAS Level 1 API
@{
dotStridedBatched(u) performs a batch of dot products of vectors x and y
result_i = x_i * y_i;
dotcStridedBatched performs a batch of dot products of the conjugate of complex vector x and complex vector y
result_i = conjugate (x_i) * y_i;
where (x_i, y_i) is the i-th instance of the batch. x_i and y_i are vectors, for i = 1, …, batchCount
Supported precisions in rocBLAS : h,bf,s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in each x_i and y_i.
- x (
Pointer/object) – IN: device pointer to the first vector (x_1) in the batch.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1)
- y (
Pointer/object) – IN: device pointer to the first vector (y_1) in the batch.
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) and the next one (y_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch
- result (
Pointer/object) – INOUT: device array or host array of batchCount size to store the dot products of each batch. return 0.0 for each element if n <= 0.
- handle (
- hip.hipblas.hipblasBfdotStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasBfloat16/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasBfloat16/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
hipblasBfloat16/object): (undocumented)
- handle (
- hip.hipblas.hipblasSdotStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
Pointer/object): (undocumented)
- handle (
- hip.hipblas.hipblasDdotStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
Pointer/object): (undocumented)
- handle (
- hip.hipblas.hipblasCdotcStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
hipblasComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasCdotuStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
hipblasComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasZdotcStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
hipblasDoubleComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasZdotuStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
hipblasDoubleComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasCdotcStridedBatched_v2(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
float2/object): (undocumented)
- handle (
- hip.hipblas.hipblasCdotuStridedBatched_v2(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
float2/object): (undocumented)
- handle (
- hip.hipblas.hipblasZdotcStridedBatched_v2(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
double2/object): (undocumented)
- handle (
- hip.hipblas.hipblasZdotuStridedBatched_v2(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount, result)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
double2/object): (undocumented)
- handle (
- hip.hipblas.hipblasSnrm2(handle, int n, x, int incx, result)#
BLAS Level 1 API
@{
nrm2 computes the euclidean norm of a real or complex vector
- result := sqrt( x’*x ) for real vectors
result := sqrt( x**H*x ) for complex vectors
Supported precisions in rocBLAS : s,d,c,z,sc,dz
Supported precisions in cuBLAS : s,d,sc,dz
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x.
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of y.
- result (
Pointer/object) – INOUT: device pointer or host pointer to store the nrm2 product. return is 0.0 if n, incx<=0.
- handle (
- hip.hipblas.hipblasDnrm2(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasScnrm2(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDznrm2(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasScnrm2_v2(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDznrm2_v2(handle, int n, x, int incx, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSnrm2Batched(handle, int n, x, int incx, int batchCount, result)#
BLAS Level 1 API
@{
nrm2Batched computes the euclidean norm over a batch of real or complex vectors
- result := sqrt( x_i’*x_i ) for real vectors x, for i = 1, …, batchCount
result := sqrt( x_i**H*x_i ) for complex vectors x, for i = 1, …, batchCount
Supported precisions in rocBLAS : s,d,c,z,sc,dz
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in each x_i.
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i. incx must be > 0.
- batchCount (
int) – IN: [int] number of instances in the batch
- result (
Pointer/object) – OUT: device pointer or host pointer to array of batchCount size for nrm2 results. return is 0.0 for each element if n <= 0, incx<=0.
- handle (
- hip.hipblas.hipblasDnrm2Batched(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasScnrm2Batched(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDznrm2Batched(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasScnrm2Batched_v2(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDznrm2Batched_v2(handle, int n, x, int incx, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSnrm2StridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
BLAS Level 1 API
@{
nrm2StridedBatched computes the euclidean norm over a batch of real or complex vectors
- = sqrt( x_i’*x_i ) for real vectors x, for i = 1, …, batchCount
:= sqrt( x_i**H*x_i ) for complex vectors, for i = 1, …, batchCount
Supported precisions in rocBLAS : s,d,c,z,sc,dz
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in each x_i.
- x (
Pointer/object) – IN: device pointer to the first vector x_1.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i. incx must be > 0.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1). There are no restrictions placed on stride_x, however the user should take care to ensure that stride_x is of appropriate size, for a typical case this means stride_x >= n * incx.
- batchCount (
int) – IN: [int] number of instances in the batch
- result (
Pointer/object) – OUT: device pointer or host pointer to array for storing contiguous batchCount results. return is 0.0 for each element if n <= 0, incx<=0.
- handle (
- hip.hipblas.hipblasDnrm2StridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasScnrm2StridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDznrm2StridedBatched(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasScnrm2StridedBatched_v2(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasDznrm2StridedBatched_v2(handle, int n, x, int incx, long stridex, int batchCount, result)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSrot(handle, int n, x, int incx, y, int incy, c, s)#
BLAS Level 1 API
@{
rot applies the Givens rotation matrix defined by c=cos(alpha) and s=sin(alpha) to vectors x and y. Scalars c and s may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode.
Supported precisions in rocBLAS : s,d,c,z,sc,dz
Supported precisions in cuBLAS : s,d,c,z,cs,zd
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in the x and y vectors.
- x (
Pointer/object) – INOUT: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment between elements of x.
- y (
Pointer/object) – INOUT: device pointer storing vector y.
- incy (
int) – IN: [int] specifies the increment between elements of y.
- c (
Pointer/object) – IN: device pointer or host pointer storing scalar cosine component of the rotation matrix.
- s (
Pointer/object) – IN: device pointer or host pointer storing scalar sine component of the rotation matrix.
- handle (
- hip.hipblas.hipblasDrot(handle, int n, x, int incx, y, int incy, c, s)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCrot(handle, int n, x, int incx, y, int incy, c, s)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
hipblasComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasCsrot(handle, int n, x, int incx, y, int incy, c, s)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZrot(handle, int n, x, int incx, y, int incy, c, s)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
hipblasDoubleComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasZdrot(handle, int n, x, int incx, y, int incy, c, s)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCrot_v2(handle, int n, x, int incx, y, int incy, c, s)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCsrot_v2(handle, int n, x, int incx, y, int incy, c, s)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZrot_v2(handle, int n, x, int incx, y, int incy, c, s)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZdrot_v2(handle, int n, x, int incx, y, int incy, c, s)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSrotBatched(handle, int n, x, int incx, y, int incy, c, s, int batchCount)#
BLAS Level 1 API
@{
rotBatched applies the Givens rotation matrix defined by c=cos(alpha) and s=sin(alpha) to batched vectors x_i and y_i, for i = 1, …, batchCount. Scalars c and s may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode.
Supported precisions in rocBLAS : s,d,sc,dz
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in each x_i and y_i vectors.
- x (
ListOfPointer/object) – INOUT: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment between elements of each x_i.
- y (
ListOfPointer/object) – INOUT: device array of device pointers storing each vector y_i.
- incy (
int) – IN: [int] specifies the increment between elements of each y_i.
- c (
Pointer/object) – IN: device pointer or host pointer to scalar cosine component of the rotation matrix.
- s (
Pointer/object) – IN: device pointer or host pointer to scalar sine component of the rotation matrix.
- batchCount (
int) – IN: [int] the number of x and y arrays, i.e. the number of batches.
- handle (
- hip.hipblas.hipblasDrotBatched(handle, int n, x, int incx, y, int incy, c, s, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCrotBatched(handle, int n, x, int incx, y, int incy, c, s, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCsrotBatched(handle, int n, x, int incx, y, int incy, c, s, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZrotBatched(handle, int n, x, int incx, y, int incy, c, s, int batchCount)#
(No short description, might be part of a group.)
- Args:
- hip.hipblas.hipblasZdrotBatched(handle, int n, x, int incx, y, int incy, c, s, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCrotBatched_v2(handle, int n, x, int incx, y, int incy, c, s, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCsrotBatched_v2(handle, int n, x, int incx, y, int incy, c, s, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZrotBatched_v2(handle, int n, x, int incx, y, int incy, c, s, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZdrotBatched_v2(handle, int n, x, int incx, y, int incy, c, s, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSrotStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, c, s, int batchCount)#
BLAS Level 1 API
@{
rotStridedBatched applies the Givens rotation matrix defined by c=cos(alpha) and s=sin(alpha) to strided batched vectors x_i and y_i, for i = 1, …, batchCount. Scalars c and s may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode.
Supported precisions in rocBLAS : s,d,sc,dz
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in each x_i and y_i vectors.
- x (
Pointer/object) – INOUT: device pointer to the first vector x_1.
- incx (
int) – IN: [int] specifies the increment between elements of each x_i.
- stridex (
int) – IN: [hipblasStride] specifies the increment from the beginning of x_i to the beginning of x_(i+1)
- y (
Pointer/object) – INOUT: device pointer to the first vector y_1.
- incy (
int) – IN: [int] specifies the increment between elements of each y_i.
- stridey (
int) – IN: [hipblasStride] specifies the increment from the beginning of y_i to the beginning of y_(i+1)
- c (
Pointer/object) – IN: device pointer or host pointer to scalar cosine component of the rotation matrix.
- s (
Pointer/object) – IN: device pointer or host pointer to scalar sine component of the rotation matrix.
- batchCount (
int) – IN: [int] the number of x and y arrays, i.e. the number of batches.
- handle (
- hip.hipblas.hipblasDrotStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, c, s, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
Pointer/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCrotStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, c, s, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
hipblasComplex/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsrotStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, c, s, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
Pointer/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZrotStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, c, s, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
hipblasDoubleComplex/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZdrotStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, c, s, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
Pointer/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCrotStridedBatched_v2(handle, int n, x, int incx, long stridex, y, int incy, long stridey, c, s, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
float2/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsrotStridedBatched_v2(handle, int n, x, int incx, long stridex, y, int incy, long stridey, c, s, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
Pointer/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZrotStridedBatched_v2(handle, int n, x, int incx, long stridex, y, int incy, long stridey, c, s, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
double2/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZdrotStridedBatched_v2(handle, int n, x, int incx, long stridex, y, int incy, long stridey, c, s, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
Pointer/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSrotg(handle, a, b, c, s)#
BLAS Level 1 API
@{
rotg creates the Givens rotation matrix for the vector (a b). Scalars c and s and arrays a and b may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode. If the pointer mode is set to HIPBLAS_POINTER_MODE_HOST, this function blocks the CPU until the GPU has finished and the results are available in host memory. If the pointer mode is set to HIPBLAS_POINTER_MODE_DEVICE, this function returns immediately and synchronization is required to read the results.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- a (
Pointer/object) – INOUT: device pointer or host pointer to input vector element, overwritten with r.
- b (
Pointer/object) – INOUT: device pointer or host pointer to input vector element, overwritten with z.
- c (
Pointer/object) – INOUT: device pointer or host pointer to cosine element of Givens rotation.
- s (
Pointer/object) – INOUT: device pointer or host pointer sine element of Givens rotation.
- handle (
- hip.hipblas.hipblasDrotg(handle, a, b, c, s)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCrotg(handle, a, b, c, s)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- a (
hipblasComplex/object): (undocumented)
- b (
hipblasComplex/object): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
hipblasComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasZrotg(handle, a, b, c, s)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- a (
hipblasDoubleComplex/object): (undocumented)
- b (
hipblasDoubleComplex/object): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
hipblasDoubleComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasCrotg_v2(handle, a, b, c, s)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZrotg_v2(handle, a, b, c, s)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSrotgBatched(handle, a, b, c, s, int batchCount)#
BLAS Level 1 API
@{
rotgBatched creates the Givens rotation matrix for the batched vectors (a_i b_i), for i = 1, …, batchCount. a, b, c, and s may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode. If the pointer mode is set to HIPBLAS_POINTER_MODE_HOST, this function blocks the CPU until the GPU has finished and the results are available in host memory. If the pointer mode is set to HIPBLAS_POINTER_MODE_DEVICE, this function returns immediately and synchronization is required to read the results.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- a (
ListOfPointer/object) – INOUT: device array of device pointers storing each single input vector element a_i, overwritten with r_i.
- b (
ListOfPointer/object) – INOUT: device array of device pointers storing each single input vector element b_i, overwritten with z_i.
- c (
ListOfPointer/object) – INOUT: device array of device pointers storing each cosine element of Givens rotation for the batch.
- s (
ListOfPointer/object) – INOUT: device array of device pointers storing each sine element of Givens rotation for the batch.
- batchCount (
int) – IN: [int] number of batches (length of arrays a, b, c, and s).
- handle (
- hip.hipblas.hipblasDrotgBatched(handle, a, b, c, s, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- a (
ListOfPointer/object): (undocumented)
- b (
ListOfPointer/object): (undocumented)
- c (
ListOfPointer/object): (undocumented)
- s (
ListOfPointer/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCrotgBatched(handle, a, b, c, s, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZrotgBatched(handle, a, b, c, s, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCrotgBatched_v2(handle, a, b, c, s, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZrotgBatched_v2(handle, a, b, c, s, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSrotgStridedBatched(handle, a, long stridea, b, long strideb, c, long stridec, s, long strides, int batchCount)#
BLAS Level 1 API
@{
rotgStridedBatched creates the Givens rotation matrix for the strided batched vectors (a_i b_i), for i = 1, …, batchCount. a, b, c, and s may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode. If the pointer mode is set to HIPBLAS_POINTER_MODE_HOST, this function blocks the CPU until the GPU has finished and the results are available in host memory. If the pointer mode is set to HIPBLAS_POINTER_MODE_HOST, this function returns immediately and synchronization is required to read the results.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- a (
Pointer/object) – INOUT: device strided_batched pointer or host strided_batched pointer to first single input vector element a_1, overwritten with r.
- stridea (
int) – IN: [hipblasStride] distance between elements of a in batch (distance between a_i and a_(i + 1))
- b (
Pointer/object) – INOUT: device strided_batched pointer or host strided_batched pointer to first single input vector element b_1, overwritten with z.
- strideb (
int) – IN: [hipblasStride] distance between elements of b in batch (distance between b_i and b_(i + 1))
- c (
Pointer/object) – INOUT: device strided_batched pointer or host strided_batched pointer to first cosine element of Givens rotations c_1.
- stridec (
int) – IN: [hipblasStride] distance between elements of c in batch (distance between c_i and c_(i + 1))
- s (
Pointer/object) – INOUT: device strided_batched pointer or host strided_batched pointer to sine element of Givens rotations s_1.
- strides (
int) – IN: [hipblasStride] distance between elements of s in batch (distance between s_i and s_(i + 1))
- batchCount (
int) – IN: [int] number of batches (length of arrays a, b, c, and s).
- handle (
- hip.hipblas.hipblasDrotgStridedBatched(handle, a, long stridea, b, long strideb, c, long stridec, s, long strides, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- a (
Pointer/object): (undocumented)
- stridea (
int): (undocumented)
- b (
Pointer/object): (undocumented)
- strideb (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- stridec (
int): (undocumented)
- s (
Pointer/object): (undocumented)
- strides (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCrotgStridedBatched(handle, a, long stridea, b, long strideb, c, long stridec, s, long strides, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- a (
hipblasComplex/object): (undocumented)
- stridea (
int): (undocumented)
- b (
hipblasComplex/object): (undocumented)
- strideb (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- stridec (
int): (undocumented)
- s (
hipblasComplex/object): (undocumented)
- strides (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZrotgStridedBatched(handle, a, long stridea, b, long strideb, c, long stridec, s, long strides, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- a (
hipblasDoubleComplex/object): (undocumented)
- stridea (
int): (undocumented)
- b (
hipblasDoubleComplex/object): (undocumented)
- strideb (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- stridec (
int): (undocumented)
- s (
hipblasDoubleComplex/object): (undocumented)
- strides (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCrotgStridedBatched_v2(handle, a, long stridea, b, long strideb, c, long stridec, s, long strides, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- a (
float2/object): (undocumented)
- stridea (
int): (undocumented)
- b (
float2/object): (undocumented)
- strideb (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- stridec (
int): (undocumented)
- s (
float2/object): (undocumented)
- strides (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZrotgStridedBatched_v2(handle, a, long stridea, b, long strideb, c, long stridec, s, long strides, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- a (
double2/object): (undocumented)
- stridea (
int): (undocumented)
- b (
double2/object): (undocumented)
- strideb (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- stridec (
int): (undocumented)
- s (
double2/object): (undocumented)
- strides (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSrotm(handle, int n, x, int incx, y, int incy, param)#
BLAS Level 1 API
@{
rotm applies the modified Givens rotation matrix defined by param to vectors x and y.
Supported precisions in rocBLAS : s,d
Supported precisions in cuBLAS : s,d
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in the x and y vectors.
- x (
Pointer/object) – INOUT: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment between elements of x.
- y (
Pointer/object) – INOUT: device pointer storing vector y.
- incy (
int) – IN: [int] specifies the increment between elements of y.
- param (
Pointer/object) – IN: device vector or host vector of 5 elements defining the rotation. param[0] = flag param[1] = H11 param[2] = H21 param[3] = H12 param[4] = H22 The flag parameter defines the form of H: flag = -1 => H = ( H11 H12 H21 H22 ) flag = 0 => H = ( 1.0 H12 H21 1.0 ) flag = 1 => H = ( H11 1.0 -1.0 H22 ) flag = -2 => H = ( 1.0 0.0 0.0 1.0 ) param may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode.
- handle (
- hip.hipblas.hipblasDrotm(handle, int n, x, int incx, y, int incy, param)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSrotmBatched(handle, int n, x, int incx, y, int incy, param, int batchCount)#
BLAS Level 1 API
@{
rotmBatched applies the modified Givens rotation matrix defined by param_i to batched vectors x_i and y_i, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in the x and y vectors.
- x (
ListOfPointer/object) – INOUT: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment between elements of each x_i.
- y (
ListOfPointer/object) – INOUT: device array of device pointers storing each vector y_1.
- incy (
int) – IN: [int] specifies the increment between elements of each y_i.
- param (
Pointer/object) – IN: device array of device vectors of 5 elements defining the rotation. param[0] = flag param[1] = H11 param[2] = H21 param[3] = H12 param[4] = H22 The flag parameter defines the form of H: flag = -1 => H = ( H11 H12 H21 H22 ) flag = 0 => H = ( 1.0 H12 H21 1.0 ) flag = 1 => H = ( H11 1.0 -1.0 H22 ) flag = -2 => H = ( 1.0 0.0 0.0 1.0 ) param may ONLY be stored on the device for the batched version of this function.
- batchCount (
int) – IN: [int] the number of x and y arrays, i.e. the number of batches.
- handle (
- hip.hipblas.hipblasDrotmBatched(handle, int n, x, int incx, y, int incy, param, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSrotmStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, param, long strideParam, int batchCount)#
BLAS Level 1 API
@{
rotmStridedBatched applies the modified Givens rotation matrix defined by param_i to strided batched vectors x_i and y_i, for i = 1, …, batchCount
Supported precisions in rocBLAS : s,d
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in the x and y vectors.
- x (
Pointer/object) – INOUT: device pointer pointing to first strided batched vector x_1.
- incx (
int) – IN: [int] specifies the increment between elements of each x_i.
- stridex (
int) – IN: [hipblasStride] specifies the increment between the beginning of x_i and x_(i + 1)
- y (
Pointer/object) – INOUT: device pointer pointing to first strided batched vector y_1.
- incy (
int) – IN: [int] specifies the increment between elements of each y_i.
- stridey (
int) – IN: [hipblasStride] specifies the increment between the beginning of y_i and y_(i + 1)
- param (
Pointer/object) – IN: device pointer pointing to first array of 5 elements defining the rotation (param_1). param[0] = flag param[1] = H11 param[2] = H21 param[3] = H12 param[4] = H22 The flag parameter defines the form of H: flag = -1 => H = ( H11 H12 H21 H22 ) flag = 0 => H = ( 1.0 H12 H21 1.0 ) flag = 1 => H = ( H11 1.0 -1.0 H22 ) flag = -2 => H = ( 1.0 0.0 0.0 1.0 ) param may ONLY be stored on the device for the strided_batched version of this function.
- strideParam (
int) – IN: [hipblasStride] specifies the increment between the beginning of param_i and param_(i + 1)
- batchCount (
int) – IN: [int] the number of x and y arrays, i.e. the number of batches.
- handle (
- hip.hipblas.hipblasDrotmStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, param, long strideParam, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- param (
Pointer/object): (undocumented)
- strideParam (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSrotmg(handle, d1, d2, x1, y1, param)#
BLAS Level 1 API
@{
rotmg creates the modified Givens rotation matrix for the vector (d1 * x1, d2 * y1). Parameters may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode. If the pointer mode is set to HIPBLAS_POINTER_MODE_HOST, this function blocks the CPU until the GPU has finished and the results are available in host memory. If the pointer mode is set to HIPBLAS_POINTER_MODE_DEVICE, this function returns immediately and synchronization is required to read the results.
Supported precisions in rocBLAS : s,d
Supported precisions in cuBLAS : s,d
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- d1 (
Pointer/object) – INOUT: device pointer or host pointer to input scalar that is overwritten.
- d2 (
Pointer/object) – INOUT: device pointer or host pointer to input scalar that is overwritten.
- x1 (
Pointer/object) – INOUT: device pointer or host pointer to input scalar that is overwritten.
- y1 (
Pointer/object) – IN: device pointer or host pointer to input scalar.
- param (
Pointer/object) – OUT: device vector or host vector of 5 elements defining the rotation. param[0] = flag param[1] = H11 param[2] = H21 param[3] = H12 param[4] = H22 The flag parameter defines the form of H: flag = -1 => H = ( H11 H12 H21 H22 ) flag = 0 => H = ( 1.0 H12 H21 1.0 ) flag = 1 => H = ( H11 1.0 -1.0 H22 ) flag = -2 => H = ( 1.0 0.0 0.0 1.0 ) param may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode.
- handle (
- hip.hipblas.hipblasDrotmg(handle, d1, d2, x1, y1, param)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSrotmgBatched(handle, d1, d2, x1, y1, param, int batchCount)#
BLAS Level 1 API
@{
rotmgBatched creates the modified Givens rotation matrix for the batched vectors (d1_i * x1_i, d2_i * y1_i), for i = 1, …, batchCount. Parameters may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode. If the pointer mode is set to HIPBLAS_POINTER_MODE_HOST, this function blocks the CPU until the GPU has finished and the results are available in host memory. If the pointer mode is set to HIPBLAS_POINTER_MODE_DEVICE, this function returns immediately and synchronization is required to read the results.
Supported precisions in rocBLAS : s,d
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- d1 (
Pointer/object) – INOUT: device batched array or host batched array of input scalars that is overwritten.
- d2 (
Pointer/object) – INOUT: device batched array or host batched array of input scalars that is overwritten.
- x1 (
Pointer/object) – INOUT: device batched array or host batched array of input scalars that is overwritten.
- y1 (
Pointer/object) – IN: device batched array or host batched array of input scalars.
- param (
Pointer/object) – OUT: device batched array or host batched array of vectors of 5 elements defining the rotation. param[0] = flag param[1] = H11 param[2] = H21 param[3] = H12 param[4] = H22 The flag parameter defines the form of H: flag = -1 => H = ( H11 H12 H21 H22 ) flag = 0 => H = ( 1.0 H12 H21 1.0 ) flag = 1 => H = ( H11 1.0 -1.0 H22 ) flag = -2 => H = ( 1.0 0.0 0.0 1.0 ) param may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode.
- batchCount (
int) – IN: [int] the number of instances in the batch.
- handle (
- hip.hipblas.hipblasDrotmgBatched(handle, d1, d2, x1, y1, param, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSrotmgStridedBatched(handle, d1, long strided1, d2, long strided2, x1, long stridex1, y1, long stridey1, param, long strideParam, int batchCount)#
BLAS Level 1 API
@{
rotmgStridedBatched creates the modified Givens rotation matrix for the strided batched vectors (d1_i * x1_i, d2_i * y1_i), for i = 1, …, batchCount. Parameters may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode. If the pointer mode is set to HIPBLAS_POINTER_MODE_HOST, this function blocks the CPU until the GPU has finished and the results are available in host memory. If the pointer mode is set to HIPBLAS_POINTER_MODE_DEVICE, this function returns immediately and synchronization is required to read the results.
Supported precisions in rocBLAS : s,d
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- d1 (
Pointer/object) – INOUT: device strided_batched array or host strided_batched array of input scalars that is overwritten.
- strided1 (
int) – IN: [hipblasStride] specifies the increment between the beginning of d1_i and d1_(i+1)
- d2 (
Pointer/object) – INOUT: device strided_batched array or host strided_batched array of input scalars that is overwritten.
- strided2 (
int) – IN: [hipblasStride] specifies the increment between the beginning of d2_i and d2_(i+1)
- x1 (
Pointer/object) – INOUT: device strided_batched array or host strided_batched array of input scalars that is overwritten.
- stridex1 (
int) – IN: [hipblasStride] specifies the increment between the beginning of x1_i and x1_(i+1)
- y1 (
Pointer/object) – IN: device strided_batched array or host strided_batched array of input scalars.
- stridey1 (
int) – IN: [hipblasStride] specifies the increment between the beginning of y1_i and y1_(i+1)
- param (
Pointer/object) – OUT: device stridedBatched array or host stridedBatched array of vectors of 5 elements defining the rotation. param[0] = flag param[1] = H11 param[2] = H21 param[3] = H12 param[4] = H22 The flag parameter defines the form of H: flag = -1 => H = ( H11 H12 H21 H22 ) flag = 0 => H = ( 1.0 H12 H21 1.0 ) flag = 1 => H = ( H11 1.0 -1.0 H22 ) flag = -2 => H = ( 1.0 0.0 0.0 1.0 ) param may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode.
- strideParam (
int) – IN: [hipblasStride] specifies the increment between the beginning of param_i and param_(i + 1)
- batchCount (
int) – IN: [int] the number of instances in the batch.
- handle (
- hip.hipblas.hipblasDrotmgStridedBatched(handle, d1, long strided1, d2, long strided2, x1, long stridex1, y1, long stridey1, param, long strideParam, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- d1 (
Pointer/object): (undocumented)
- strided1 (
int): (undocumented)
- d2 (
Pointer/object): (undocumented)
- strided2 (
int): (undocumented)
- x1 (
Pointer/object): (undocumented)
- stridex1 (
int): (undocumented)
- y1 (
Pointer/object): (undocumented)
- stridey1 (
int): (undocumented)
- param (
Pointer/object): (undocumented)
- strideParam (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSscal(handle, int n, alpha, x, int incx)#
BLAS Level 1 API
@{
scal scales each element of vector x with scalar alpha.
x := alpha * x
Supported precisions in rocBLAS : s,d,c,z,cs,zd
Supported precisions in cuBLAS : s,d,c,z,cs,zd
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x.
- alpha (
Pointer/object) – IN: device pointer or host pointer for the scalar alpha.
- x (
Pointer/object) – INOUT: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- handle (
- hip.hipblas.hipblasDscal(handle, int n, alpha, x, int incx)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCscal(handle, int n, alpha, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsscal(handle, int n, alpha, x, int incx)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZscal(handle, int n, alpha, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZdscal(handle, int n, alpha, x, int incx)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCscal_v2(handle, int n, alpha, x, int incx)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCsscal_v2(handle, int n, alpha, x, int incx)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZscal_v2(handle, int n, alpha, x, int incx)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZdscal_v2(handle, int n, alpha, x, int incx)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSscalBatched(handle, int n, alpha, x, int incx, int batchCount)#
BLAS Level 1 API
@{
scalBatched scales each element of vector x_i with scalar alpha, for i = 1, … , batchCount.
x_i := alpha * x_i
where (x_i) is the i-th instance of the batch.
Supported precisions in rocBLAS : s,d,c,z,cs,zd
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in each x_i.
- alpha (
Pointer/object) – IN: host pointer or device pointer for the scalar alpha.
- x (
ListOfPointer/object) – INOUT: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- batchCount (
int) – IN: [int] specifies the number of batches in x.
- handle (
- hip.hipblas.hipblasDscalBatched(handle, int n, alpha, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCscalBatched(handle, int n, alpha, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZscalBatched(handle, int n, alpha, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCsscalBatched(handle, int n, alpha, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZdscalBatched(handle, int n, alpha, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCscalBatched_v2(handle, int n, alpha, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZscalBatched_v2(handle, int n, alpha, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCsscalBatched_v2(handle, int n, alpha, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZdscalBatched_v2(handle, int n, alpha, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSscalStridedBatched(handle, int n, alpha, x, int incx, long stridex, int batchCount)#
BLAS Level 1 API
@{
scalStridedBatched scales each element of vector x_i with scalar alpha, for i = 1, … , batchCount.
x_i := alpha * x_i ,
where (x_i) is the i-th instance of the batch.
Supported precisions in rocBLAS : s,d,c,z,cs,zd
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in each x_i.
- alpha (
Pointer/object) – IN: host pointer or device pointer for the scalar alpha.
- x (
Pointer/object) – INOUT: device pointer to the first vector (x_1) in the batch.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1). There are no restrictions placed on stride_x, however the user should take care to ensure that stride_x is of appropriate size, for a typical case this means stride_x >= n * incx.
- batchCount (
int) – IN: [int] specifies the number of batches in x.
- handle (
- hip.hipblas.hipblasDscalStridedBatched(handle, int n, alpha, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCscalStridedBatched(handle, int n, alpha, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZscalStridedBatched(handle, int n, alpha, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsscalStridedBatched(handle, int n, alpha, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZdscalStridedBatched(handle, int n, alpha, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCscalStridedBatched_v2(handle, int n, alpha, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZscalStridedBatched_v2(handle, int n, alpha, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCsscalStridedBatched_v2(handle, int n, alpha, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZdscalStridedBatched_v2(handle, int n, alpha, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSswap(handle, int n, x, int incx, y, int incy)#
BLAS Level 1 API
@{
swap interchanges vectors x and y.
y := x; x := y
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x and y.
- x (
Pointer/object) – INOUT: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- y (
Pointer/object) – INOUT: device pointer storing vector y.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- handle (
- hip.hipblas.hipblasDswap(handle, int n, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCswap(handle, int n, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZswap(handle, int n, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCswap_v2(handle, int n, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZswap_v2(handle, int n, x, int incx, y, int incy)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSswapBatched(handle, int n, x, int incx, y, int incy, int batchCount)#
BLAS Level 1 API
@{
swapBatched interchanges vectors x_i and y_i, for i = 1 , … , batchCount
y_i := x_i; x_i := y_i
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in each x_i and y_i.
- x (
ListOfPointer/object) – INOUT: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- y (
ListOfPointer/object) – INOUT: device array of device pointers storing each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDswapBatched(handle, int n, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
ListOfPointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCswapBatched(handle, int n, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZswapBatched(handle, int n, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCswapBatched_v2(handle, int n, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZswapBatched_v2(handle, int n, x, int incx, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSswapStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
BLAS Level 1 API
@{
swapStridedBatched interchanges vectors x_i and y_i, for i = 1 , … , batchCount
y_i := x_i; x_i := y_i
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in each x_i and y_i.
- x (
Pointer/object) – INOUT: device pointer to the first vector x_1.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1). There are no restrictions placed on stride_x, however the user should take care to ensure that stride_x is of appropriate size, for a typical case this means stride_x >= n * incx.
- y (
Pointer/object) – INOUT: device pointer to the first vector y_1.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) and the next one (y_i+1). There are no restrictions placed on stride_x, however the user should take care to ensure that stride_y is of appropriate size, for a typical case this means stride_y >= n * incy. stridey should be non zero.
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDswapStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCswapStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZswapStridedBatched(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- hip.hipblas.hipblasCswapStridedBatched_v2(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZswapStridedBatched_v2(handle, int n, x, int incx, long stridex, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSgbmv(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, x, int incx, beta, y, int incy)#
BLAS Level 2 API
@{
gbmv performs one of the matrix-vector operations
- y := alpha*A*x + beta*y, or
y := alpha*A**T*x + beta*y, or y := alpha*A**H*x + beta*y,
where alpha and beta are scalars, x and y are vectors and A is an m by n banded matrix with kl sub-diagonals and ku super-diagonals.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- trans (
hipblasOperation_t) – IN: [hipblasOperation_t] indicates whether matrix A is tranposed (conjugated) or not
- m (
int) – IN: [int] number of rows of matrix A
- n (
int) – IN: [int] number of columns of matrix A
- kl (
int) – IN: [int] number of sub-diagonals of A
- ku (
int) – IN: [int] number of super-diagonals of A
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- AP (
Pointer/object) – IN: device pointer storing banded matrix A. Leading (kl + ku + 1) by n part of the matrix contains the coefficients of the banded matrix. The leading diagonal resides in row (ku + 1) with the first super-diagonal above on the RHS of row ku. The first sub-diagonal resides below on the LHS of row ku + 2. This propagates up and down across sub/super-diagonals.
Ex: (m = n = 7; ku = 2, kl = 2) 1 2 3 0 0 0 0 0 0 3 3 3 3 3 4 1 2 3 0 0 0 0 2 2 2 2 2 2 5 4 1 2 3 0 0 —-> 1 1 1 1 1 1 1 0 5 4 1 2 3 0 4 4 4 4 4 4 0 0 0 5 4 1 2 0 5 5 5 5 5 0 0 0 0 0 5 4 1 2 0 0 0 0 0 0 0 0 0 0 0 5 4 1 0 0 0 0 0 0 0
Note that the empty elements which don’t correspond to data will not be referenced.
- lda (
int) – IN: [int] specifies the leading dimension of A. Must be >= (kl + ku + 1)
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- beta (
Pointer/object) – IN: device pointer or host pointer to scalar beta.
- y (
Pointer/object) – INOUT: device pointer storing vector y.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- handle (
- hip.hipblas.hipblasDgbmv(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- kl (
int): (undocumented)
- ku (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgbmv(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- kl (
int): (undocumented)
- ku (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgbmv(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- kl (
int): (undocumented)
- ku (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgbmv_v2(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- kl (
int): (undocumented)
- ku (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgbmv_v2(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- kl (
int): (undocumented)
- ku (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSgbmvBatched(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
BLAS Level 2 API
@{
gbmvBatched performs one of the matrix-vector operations
- y_i := alpha*A_i*x_i + beta*y_i, or
y_i := alpha*A_i**T*x_i + beta*y_i, or y_i := alpha*A_i**H*x_i + beta*y_i,
where (A_i, x_i, y_i) is the i-th instance of the batch. alpha and beta are scalars, x_i and y_i are vectors and A_i is an m by n banded matrix with kl sub-diagonals and ku super-diagonals, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- trans (
hipblasOperation_t) – IN: [hipblasOperation_t] indicates whether matrix A is tranposed (conjugated) or not
- m (
int) – IN: [int] number of rows of each matrix A_i
- n (
int) – IN: [int] number of columns of each matrix A_i
- kl (
int) – IN: [int] number of sub-diagonals of each A_i
- ku (
int) – IN: [int] number of super-diagonals of each A_i
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- AP (
Pointer/object) – IN: device array of device pointers storing each banded matrix A_i. Leading (kl + ku + 1) by n part of the matrix contains the coefficients of the banded matrix. The leading diagonal resides in row (ku + 1) with the first super-diagonal above on the RHS of row ku. The first sub-diagonal resides below on the LHS of row ku + 2. This propagates up and down across sub/super-diagonals.
Ex: (m = n = 7; ku = 2, kl = 2) 1 2 3 0 0 0 0 0 0 3 3 3 3 3 4 1 2 3 0 0 0 0 2 2 2 2 2 2 5 4 1 2 3 0 0 —-> 1 1 1 1 1 1 1 0 5 4 1 2 3 0 4 4 4 4 4 4 0 0 0 5 4 1 2 0 5 5 5 5 5 0 0 0 0 0 5 4 1 2 0 0 0 0 0 0 0 0 0 0 0 5 4 1 0 0 0 0 0 0 0
Note that the empty elements which don’t correspond to data will not be referenced.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i. Must be >= (kl + ku + 1)
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- beta (
Pointer/object) – IN: device pointer or host pointer to scalar beta.
- y (
ListOfPointer/object) – INOUT: device array of device pointers storing each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- batchCount (
int) – IN: [int] specifies the number of instances in the batch.
- handle (
- hip.hipblas.hipblasDgbmvBatched(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- kl (
int): (undocumented)
- ku (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- y (
ListOfPointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgbmvBatched(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- kl (
int): (undocumented)
- ku (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgbmvBatched(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- kl (
int): (undocumented)
- ku (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgbmvBatched_v2(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- kl (
int): (undocumented)
- ku (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgbmvBatched_v2(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- kl (
int): (undocumented)
- ku (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSgbmvStridedBatched(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
BLAS Level 2 API
@{
gbmvStridedBatched performs one of the matrix-vector operations
- y_i := alpha*A_i*x_i + beta*y_i, or
y_i := alpha*A_i**T*x_i + beta*y_i, or y_i := alpha*A_i**H*x_i + beta*y_i,
where (A_i, x_i, y_i) is the i-th instance of the batch. alpha and beta are scalars, x_i and y_i are vectors and A_i is an m by n banded matrix with kl sub-diagonals and ku super-diagonals, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- trans (
hipblasOperation_t) – IN: [hipblasOperation_t] indicates whether matrix A is tranposed (conjugated) or not
- m (
int) – IN: [int] number of rows of matrix A
- n (
int) – IN: [int] number of columns of matrix A
- kl (
int) – IN: [int] number of sub-diagonals of A
- ku (
int) – IN: [int] number of super-diagonals of A
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- AP (
Pointer/object) – IN: device pointer to first banded matrix (A_1). Leading (kl + ku + 1) by n part of the matrix contains the coefficients of the banded matrix. The leading diagonal resides in row (ku + 1) with the first super-diagonal above on the RHS of row ku. The first sub-diagonal resides below on the LHS of row ku + 2. This propagates up and down across sub/super-diagonals.
Ex: (m = n = 7; ku = 2, kl = 2) 1 2 3 0 0 0 0 0 0 3 3 3 3 3 4 1 2 3 0 0 0 0 2 2 2 2 2 2 5 4 1 2 3 0 0 —-> 1 1 1 1 1 1 1 0 5 4 1 2 3 0 4 4 4 4 4 4 0 0 0 5 4 1 2 0 5 5 5 5 5 0 0 0 0 0 5 4 1 2 0 0 0 0 0 0 0 0 0 0 0 5 4 1 0 0 0 0 0 0 0
Note that the empty elements which don’t correspond to data will not be referenced.
- lda (
int) – IN: [int] specifies the leading dimension of A. Must be >= (kl + ku + 1)
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- x (
Pointer/object) – IN: device pointer to first vector (x_1).
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1)
- beta (
Pointer/object) – IN: device pointer or host pointer to scalar beta.
- y (
Pointer/object) – INOUT: device pointer to first vector (y_1).
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) and the next one (x_i+1)
- batchCount (
int) – IN: [int] specifies the number of instances in the batch.
- handle (
- hip.hipblas.hipblasDgbmvStridedBatched(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- kl (
int): (undocumented)
- ku (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgbmvStridedBatched(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- kl (
int): (undocumented)
- ku (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgbmvStridedBatched(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- kl (
int): (undocumented)
- ku (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgbmvStridedBatched_v2(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- kl (
int): (undocumented)
- ku (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgbmvStridedBatched_v2(handle, trans, int m, int n, int kl, int ku, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- kl (
int): (undocumented)
- ku (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSgemv(handle, trans, int m, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
BLAS Level 2 API
@{
gemv performs one of the matrix-vector operations
- y := alpha*A*x + beta*y, or
y := alpha*A**T*x + beta*y, or y := alpha*A**H*x + beta*y,
where alpha and beta are scalars, x and y are vectors and A is an m by n matrix.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- trans (
hipblasOperation_t) – IN: [hipblasOperation_t] indicates whether matrix A is tranposed (conjugated) or not
- m (
int) – IN: [int] number of rows of matrix A
- n (
int) – IN: [int] number of columns of matrix A
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- AP (
Pointer/object) – IN: device pointer storing matrix A.
- lda (
int) – IN: [int] specifies the leading dimension of A.
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- beta (
Pointer/object) – IN: device pointer or host pointer to scalar beta.
- y (
Pointer/object) – INOUT: device pointer storing vector y.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- handle (
- hip.hipblas.hipblasDgemv(handle, trans, int m, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgemv(handle, trans, int m, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgemv(handle, trans, int m, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgemv_v2(handle, trans, int m, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgemv_v2(handle, trans, int m, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSgemvBatched(handle, trans, int m, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
BLAS Level 2 API
@{
gemvBatched performs a batch of matrix-vector operations
- y_i := alpha*A_i*x_i + beta*y_i, or
y_i := alpha*A_i**T*x_i + beta*y_i, or y_i := alpha*A_i**H*x_i + beta*y_i,
where (A_i, x_i, y_i) is the i-th instance of the batch. alpha and beta are scalars, x_i and y_i are vectors and A_i is an m by n matrix, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- trans (
hipblasOperation_t) – IN: [hipblasOperation_t] indicates whether matrices A_i are tranposed (conjugated) or not
- m (
int) – IN: [int] number of rows of each matrix A_i
- n (
int) – IN: [int] number of columns of each matrix A_i
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix A_i.
- lda (
int) – IN: [int] specifies the leading dimension of each matrix A_i.
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each vector x_i.
- beta (
Pointer/object) – IN: device pointer or host pointer to scalar beta.
- y (
ListOfPointer/object) – INOUT: device array of device pointers storing each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of each vector y_i.
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDgemvBatched(handle, trans, int m, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- y (
ListOfPointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgemvBatched(handle, trans, int m, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgemvBatched(handle, trans, int m, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgemvBatched_v2(handle, trans, int m, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgemvBatched_v2(handle, trans, int m, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSgemvStridedBatched(handle, transA, int m, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
BLAS Level 2 API
@{
gemvStridedBatched performs a batch of matrix-vector operations
- y_i := alpha*A_i*x_i + beta*y_i, or
y_i := alpha*A_i**T*x_i + beta*y_i, or y_i := alpha*A_i**H*x_i + beta*y_i,
where (A_i, x_i, y_i) is the i-th instance of the batch. alpha and beta are scalars, x_i and y_i are vectors and A_i is an m by n matrix, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] indicates whether matrices A_i are tranposed (conjugated) or not
- m (
int) – IN: [int] number of rows of matrices A_i
- n (
int) – IN: [int] number of columns of matrices A_i
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- AP (
Pointer/object) – IN: device pointer to the first matrix (A_1) in the batch.
- lda (
int) – IN: [int] specifies the leading dimension of matrices A_i.
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- x (
Pointer/object) – IN: device pointer to the first vector (x_1) in the batch.
- incx (
int) – IN: [int] specifies the increment for the elements of vectors x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1). There are no restrictions placed on stridex, however the user should take care to ensure that stridex is of appropriate size. When trans equals HIPBLAS_OP_N this typically means stridex >= n * incx, otherwise stridex >= m * incx.
- beta (
Pointer/object) – IN: device pointer or host pointer to scalar beta.
- y (
Pointer/object) – INOUT: device pointer to the first vector (y_1) in the batch.
- incy (
int) – IN: [int] specifies the increment for the elements of vectors y_i.
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) and the next one (y_i+1). There are no restrictions placed on stridey, however the user should take care to ensure that stridey is of appropriate size. When trans equals HIPBLAS_OP_N this typically means stridey >= m * incy, otherwise stridey >= n * incy. stridey should be non zero.
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDgemvStridedBatched(handle, transA, int m, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgemvStridedBatched(handle, transA, int m, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgemvStridedBatched(handle, transA, int m, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgemvStridedBatched_v2(handle, transA, int m, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgemvStridedBatched_v2(handle, transA, int m, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSger(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda)#
BLAS Level 2 API
@{
ger,geru,gerc performs the matrix-vector operations
- A := A + alpha*x*y**T , OR
A := A + alpha*x*y**H for gerc
where alpha is a scalar, x and y are vectors, and A is an m by n matrix.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- m (
int) – IN: [int] the number of rows of the matrix A.
- n (
int) – IN: [int] the number of columns of the matrix A.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- y (
Pointer/object) – IN: device pointer storing vector y.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- AP (
Pointer/object) – INOUT: device pointer storing matrix A.
- lda (
int) – IN: [int] specifies the leading dimension of A.
- handle (
- hip.hipblas.hipblasDger(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgeru(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgerc(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgeru(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgerc(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgeru_v2(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- hip.hipblas.hipblasCgerc_v2(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- hip.hipblas.hipblasZgeru_v2(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgerc_v2(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSgerBatched(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
BLAS Level 2 API
@{
gerBatched,geruBatched,gercBatched performs a batch of the matrix-vector operations
- A := A + alpha*x*y**T , OR
A := A + alpha*x*y**H for gerc
where (A_i, x_i, y_i) is the i-th instance of the batch. alpha is a scalar, x_i and y_i are vectors and A_i is an m by n matrix, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- m (
int) – IN: [int] the number of rows of each matrix A_i.
- n (
int) – IN: [int] the number of columns of eaceh matrix A_i.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each vector x_i.
- y (
ListOfPointer/object) – IN: device array of device pointers storing each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of each vector y_i.
- AP (
Pointer/object) – INOUT: device array of device pointers storing each matrix A_i.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i.
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDgerBatched(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
ListOfPointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgeruBatched(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgercBatched(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgeruBatched(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgercBatched(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgeruBatched_v2(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgercBatched_v2(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgeruBatched_v2(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgercBatched_v2(handle, int m, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSgerStridedBatched(handle, int m, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
BLAS Level 2 API
@{
gerStridedBatched,geruStridedBatched,gercStridedBatched performs the matrix-vector operations
- A_i := A_i + alpha*x_i*y_i**T, OR
A_i := A_i + alpha*x_i*y_i**H for gerc
where (A_i, x_i, y_i) is the i-th instance of the batch. alpha is a scalar, x_i and y_i are vectors and A_i is an m by n matrix, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- m (
int) – IN: [int] the number of rows of each matrix A_i.
- n (
int) – IN: [int] the number of columns of each matrix A_i.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
Pointer/object) – IN: device pointer to the first vector (x_1) in the batch.
- incx (
int) – IN: [int] specifies the increments for the elements of each vector x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1). There are no restrictions placed on stridex, however the user should take care to ensure that stridex is of appropriate size, for a typical case this means stridex >= m * incx.
- y (
Pointer/object) – INOUT: device pointer to the first vector (y_1) in the batch.
- incy (
int) – IN: [int] specifies the increment for the elements of each vector y_i.
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) and the next one (y_i+1). There are no restrictions placed on stridey, however the user should take care to ensure that stridey is of appropriate size, for a typical case this means stridey >= n * incy.
- AP (
Pointer/object) – INOUT: device pointer to the first matrix (A_1) in the batch.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i.
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDgerStridedBatched(handle, int m, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgeruStridedBatched(handle, int m, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgercStridedBatched(handle, int m, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgeruStridedBatched(handle, int m, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgercStridedBatched(handle, int m, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgeruStridedBatched_v2(handle, int m, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgercStridedBatched_v2(handle, int m, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgeruStridedBatched_v2(handle, int m, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgercStridedBatched_v2(handle, int m, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChbmv(handle, uplo, int n, int k, alpha, AP, int lda, x, int incx, beta, y, int incy)#
BLAS Level 2 API
@{
hbmv performs the matrix-vector operations
y := alpha*A*x + beta*y
where alpha and beta are scalars, x and y are n element vectors and A is an n by n Hermitian band matrix, with k super-diagonals.
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : c,z
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The leading (k + 1) by n part of A must contain the lower triangular band part of the Hermitian matrix, with the leading diagonal in row (1), the first sub-diagonal on the LHS of row 2, etc. The bottom right k by k triangle of A will not be referenced.
Ex (lower, lda = 2, n = 4, k = 1): A Represented matrix (1,0) (2,0) (3,0) (4,0) (1, 0) (5,-9) (0, 0) (0, 0) (5,9) (6,8) (7,7) (0,0) (5, 9) (2, 0) (6,-8) (0, 0)
(0, 0) (6, 8) (3, 0) (7,-7) (0, 0) (0, 0) (7, 7) (4, 0)
As a Hermitian matrix, the imaginary part of the main diagonal of A will not be referenced and is assumed to be == 0.
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: The upper triangular part of A is being supplied. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of A is being supplied.
- n (
int) – IN: [int] the order of the matrix A.
- k (
int) – IN: [int] the number of super-diagonals of the matrix A. Must be >= 0.
- alpha (
hipblasComplex/object) – IN: device pointer or host pointer to scalar alpha.
- AP (
hipblasComplex/object) – IN: device pointer storing matrix A. Of dimension (lda, n). if uplo == HIPBLAS_FILL_MODE_UPPER:
The leading (k + 1) by n part of A must contain the upper triangular band part of the Hermitian matrix, with the leading diagonal in row (k + 1), the first super-diagonal on the RHS of row k, etc. The top left k by x triangle of A will not be referenced.
Ex (upper, lda = n = 4, k = 1): A Represented matrix (0,0) (5,9) (6,8) (7,7) (1, 0) (5, 9) (0, 0) (0, 0) (1,0) (2,0) (3,0) (4,0) (5,-9) (2, 0) (6, 8) (0, 0) (0,0) (0,0) (0,0) (0,0) (0, 0) (6,-8) (3, 0) (7, 7) (0,0) (0,0) (0,0) (0,0) (0, 0) (0, 0) (7,-7) (4, 0)
- lda (
int) – IN: [int] specifies the leading dimension of A. must be >= k + 1
- x (
hipblasComplex/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- beta (
hipblasComplex/object) – IN: device pointer or host pointer to scalar beta.
- y (
hipblasComplex/object) – INOUT: device pointer storing vector y.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- handle (
- hip.hipblas.hipblasZhbmv(handle, uplo, int n, int k, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChbmv_v2(handle, uplo, int n, int k, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZhbmv_v2(handle, uplo, int n, int k, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChbmvBatched(handle, uplo, int n, int k, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
BLAS Level 2 API
@{
hbmvBatched performs one of the matrix-vector operations
y_i := alpha*A_i*x_i + beta*y_i
where alpha and beta are scalars, x_i and y_i are n element vectors and A_i is an n by n Hermitian band matrix with k super-diagonals, for each batch in i = [1, batchCount].
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The leading (k + 1) by n part of each A_i must contain the lower triangular band part of the Hermitian matrix, with the leading diagonal in row (1), the first sub-diagonal on the LHS of row 2, etc. The bottom right k by k triangle of each A_i will not be referenced.
Ex (lower, lda = 2, n = 4, k = 1): A Represented matrix (1,0) (2,0) (3,0) (4,0) (1, 0) (5,-9) (0, 0) (0, 0) (5,9) (6,8) (7,7) (0,0) (5, 9) (2, 0) (6,-8) (0, 0)
(0, 0) (6, 8) (3, 0) (7,-7) (0, 0) (0, 0) (7, 7) (4, 0)
As a Hermitian matrix, the imaginary part of the main diagonal of each A_i will not be referenced and is assumed to be == 0.
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: The upper triangular part of each A_i is being supplied. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of each A_i is being supplied.
- n (
int) – IN: [int] the order of each matrix A_i.
- k (
int) – IN: [int] the number of super-diagonals of each matrix A_i. Must be >= 0.
- alpha (
hipblasComplex/object) – IN: device pointer or host pointer to scalar alpha.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix_i A of dimension (lda, n). if uplo == HIPBLAS_FILL_MODE_UPPER:
The leading (k + 1) by n part of each A_i must contain the upper triangular band part of the Hermitian matrix, with the leading diagonal in row (k + 1), the first super-diagonal on the RHS of row k, etc. The top left k by x triangle of each A_i will not be referenced.
Ex (upper, lda = n = 4, k = 1): A Represented matrix (0,0) (5,9) (6,8) (7,7) (1, 0) (5, 9) (0, 0) (0, 0) (1,0) (2,0) (3,0) (4,0) (5,-9) (2, 0) (6, 8) (0, 0) (0,0) (0,0) (0,0) (0,0) (0, 0) (6,-8) (3, 0) (7, 7) (0,0) (0,0) (0,0) (0,0) (0, 0) (0, 0) (7,-7) (4, 0)
- lda (
int) – IN: [int] specifies the leading dimension of each A_i. must be >= max(1, n)
- x (
Pointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- beta (
hipblasComplex/object) – IN: device pointer or host pointer to scalar beta.
- y (
Pointer/object) – INOUT: device array of device pointers storing each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZhbmvBatched(handle, uplo, int n, int k, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChbmvBatched_v2(handle, uplo, int n, int k, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZhbmvBatched_v2(handle, uplo, int n, int k, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChbmvStridedBatched(handle, uplo, int n, int k, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
BLAS Level 2 API
@{
hbmvStridedBatched performs one of the matrix-vector operations
y_i := alpha*A_i*x_i + beta*y_i
where alpha and beta are scalars, x_i and y_i are n element vectors and A_i is an n by n Hermitian band matrix with k super-diagonals, for each batch in i = [1, batchCount].
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The leading (k + 1) by n part of each A_i must contain the lower triangular band part of the Hermitian matrix, with the leading diagonal in row (1), the first sub-diagonal on the LHS of row 2, etc. The bottom right k by k triangle of each A_i will not be referenced.
Ex (lower, lda = 2, n = 4, k = 1): A Represented matrix (1,0) (2,0) (3,0) (4,0) (1, 0) (5,-9) (0, 0) (0, 0) (5,9) (6,8) (7,7) (0,0) (5, 9) (2, 0) (6,-8) (0, 0)
(0, 0) (6, 8) (3, 0) (7,-7) (0, 0) (0, 0) (7, 7) (4, 0)
As a Hermitian matrix, the imaginary part of the main diagonal of each A_i will not be referenced and is assumed to be == 0.
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: The upper triangular part of each A_i is being supplied. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of each A_i is being supplied.
- n (
int) – IN: [int] the order of each matrix A_i.
- k (
int) – IN: [int] the number of super-diagonals of each matrix A_i. Must be >= 0.
- alpha (
hipblasComplex/object) – IN: device pointer or host pointer to scalar alpha.
- AP (
hipblasComplex/object) – IN: device array pointing to the first matrix A_1. Each A_i is of dimension (lda, n). if uplo == HIPBLAS_FILL_MODE_UPPER:
The leading (k + 1) by n part of each A_i must contain the upper triangular band part of the Hermitian matrix, with the leading diagonal in row (k + 1), the first super-diagonal on the RHS of row k, etc. The top left k by x triangle of each A_i will not be referenced.
Ex (upper, lda = n = 4, k = 1): A Represented matrix (0,0) (5,9) (6,8) (7,7) (1, 0) (5, 9) (0, 0) (0, 0) (1,0) (2,0) (3,0) (4,0) (5,-9) (2, 0) (6, 8) (0, 0) (0,0) (0,0) (0,0) (0,0) (0, 0) (6,-8) (3, 0) (7, 7) (0,0) (0,0) (0,0) (0,0) (0, 0) (0, 0) (7,-7) (4, 0)
- lda (
int) – IN: [int] specifies the leading dimension of each A_i. must be >= max(1, n)
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- x (
hipblasComplex/object) – IN: device array pointing to the first vector y_1.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1)
- beta (
hipblasComplex/object) – IN: device pointer or host pointer to scalar beta.
- y (
hipblasComplex/object) – INOUT: device array pointing to the first vector y_1.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) and the next one (y_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZhbmvStridedBatched(handle, uplo, int n, int k, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChbmvStridedBatched_v2(handle, uplo, int n, int k, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZhbmvStridedBatched_v2(handle, uplo, int n, int k, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChemv(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
BLAS Level 2 API
@{
hemv performs one of the matrix-vector operations
y := alpha*A*x + beta*y
where alpha and beta are scalars, x and y are n element vectors and A is an n by n Hermitian matrix.
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: the upper triangular part of the Hermitian matrix A is supplied. HIPBLAS_FILL_MODE_LOWER: the lower triangular part of the Hermitian matrix A is supplied.
- n (
int) – IN: [int] the order of the matrix A.
- alpha (
hipblasComplex/object) – IN: device pointer or host pointer to scalar alpha.
- AP (
hipblasComplex/object) – IN: device pointer storing matrix A. Of dimension (lda, n). if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular part of A must contain the upper triangular part of a Hermitian matrix. The lower triangular part of A will not be referenced.
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular part of A must contain the lower triangular part of a Hermitian matrix. The upper triangular part of A will not be referenced.
As a Hermitian matrix, the imaginary part of the main diagonal of A will not be referenced and is assumed to be == 0.
- lda (
int) – IN: [int] specifies the leading dimension of A. must be >= max(1, n)
- x (
hipblasComplex/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- beta (
hipblasComplex/object) – IN: device pointer or host pointer to scalar beta.
- y (
hipblasComplex/object) – INOUT: device pointer storing vector y.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- handle (
- hip.hipblas.hipblasZhemv(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChemv_v2(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZhemv_v2(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChemvBatched(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
BLAS Level 2 API
@{
hemvBatched performs one of the matrix-vector operations
y_i := alpha*A_i*x_i + beta*y_i
where alpha and beta are scalars, x_i and y_i are n element vectors and A_i is an n by n Hermitian matrix, for each batch in i = [1, batchCount].
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: the upper triangular part of the Hermitian matrix A is supplied. HIPBLAS_FILL_MODE_LOWER: the lower triangular part of the Hermitian matrix A is supplied.
- n (
int) – IN: [int] the order of each matrix A_i.
- alpha (
hipblasComplex/object) – IN: device pointer or host pointer to scalar alpha.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix A_i of dimension (lda, n). if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular part of each A_i must contain the upper triangular part of a Hermitian matrix. The lower triangular part of each A_i will not be referenced.
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular part of each A_i must contain the lower triangular part of a Hermitian matrix. The upper triangular part of each A_i will not be referenced.
As a Hermitian matrix, the imaginary part of the main diagonal of each A_i will not be referenced and is assumed to be == 0.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i. must be >= max(1, n)
- x (
Pointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- beta (
hipblasComplex/object) – IN: device pointer or host pointer to scalar beta.
- y (
Pointer/object) – INOUT: device array of device pointers storing each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZhemvBatched(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChemvBatched_v2(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZhemvBatched_v2(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChemvStridedBatched(handle, uplo, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
BLAS Level 2 API
@{
hemvStridedBatched performs one of the matrix-vector operations
y_i := alpha*A_i*x_i + beta*y_i
where alpha and beta are scalars, x_i and y_i are n element vectors and A_i is an n by n Hermitian matrix, for each batch in i = [1, batchCount].
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: the upper triangular part of the Hermitian matrix A is supplied. HIPBLAS_FILL_MODE_LOWER: the lower triangular part of the Hermitian matrix A is supplied.
- n (
int) – IN: [int] the order of each matrix A_i.
- alpha (
hipblasComplex/object) – IN: device pointer or host pointer to scalar alpha.
- AP (
hipblasComplex/object) – IN: device array of device pointers storing each matrix A_i of dimension (lda, n). if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular part of each A_i must contain the upper triangular part of a Hermitian matrix. The lower triangular part of each A_i will not be referenced.
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular part of each A_i must contain the lower triangular part of a Hermitian matrix. The upper triangular part of each A_i will not be referenced.
As a Hermitian matrix, the imaginary part of the main diagonal of each A_i will not be referenced and is assumed to be == 0.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i. must be >= max(1, n)
- strideA (
int) – IN: [hipblasStride] stride from the start of one (A_i) to the next (A_i+1)
- x (
hipblasComplex/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1).
- beta (
hipblasComplex/object) – IN: device pointer or host pointer to scalar beta.
- y (
hipblasComplex/object) – INOUT: device array of device pointers storing each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) and the next one (y_i+1).
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZhemvStridedBatched(handle, uplo, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChemvStridedBatched_v2(handle, uplo, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZhemvStridedBatched_v2(handle, uplo, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCher(handle, uplo, int n, alpha, x, int incx, AP, int lda)#
BLAS Level 2 API
@{
her performs the matrix-vector operations
A := A + alpha*x*x**H
where alpha is a real scalar, x is a vector, and A is an n by n Hermitian matrix.
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of A is supplied in A. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of A is supplied in A.
- n (
int) – IN: [int] the number of rows and columns of matrix A, must be at least 0.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
hipblasComplex/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- AP (
hipblasComplex/object) – INOUT: - device pointer storing the specified triangular portion of
the Hermitian matrix A. Of size (lda * n). if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of the Hermitian matrix A is supplied. The lower triangluar portion will not be touched.
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of the Hermitian matrix A is supplied. The upper triangular portion will not be touched.
Note that the imaginary part of the diagonal elements are not accessed and are assumed to be 0.
- lda (
int) – IN: [int] specifies the leading dimension of A. Must be at least max(1, n).
- handle (
- hip.hipblas.hipblasZher(handle, uplo, int n, alpha, x, int incx, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCher_v2(handle, uplo, int n, alpha, x, int incx, AP, int lda)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZher_v2(handle, uplo, int n, alpha, x, int incx, AP, int lda)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCherBatched(handle, uplo, int n, alpha, x, int incx, AP, int lda, int batchCount)#
BLAS Level 2 API
@{
herBatched performs the matrix-vector operations
A_i := A_i + alpha*x_i*x_i**H
where alpha is a real scalar, x_i is a vector, and A_i is an n by n symmetric matrix, for i = 1, …, batchCount.
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of each A_i is supplied in A. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of each A_i is supplied in A.
- n (
int) – IN: [int] the number of rows and columns of each matrix A_i, must be at least 0.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
Pointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- AP (
Pointer/object) – INOUT: - device array of device pointers storing the specified triangular portion of
each Hermitian matrix A_i of at least size ((n * (n + 1)) / 2). Array is of at least size batchCount. if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of each Hermitian matrix A_i is supplied. The lower triangular portion of each A_i will not be touched.
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of each Hermitian matrix A_i is supplied. The upper triangular portion of each A_i will not be touched.
Note that the imaginary part of the diagonal elements are not accessed and are assumed to be 0.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i. Must be at least max(1, n).
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZherBatched(handle, uplo, int n, alpha, x, int incx, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCherBatched_v2(handle, uplo, int n, alpha, x, int incx, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZherBatched_v2(handle, uplo, int n, alpha, x, int incx, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCherStridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, AP, int lda, long strideA, int batchCount)#
BLAS Level 2 API
@{
herStridedBatched performs the matrix-vector operations
A_i := A_i + alpha*x_i*x_i**H
where alpha is a real scalar, x_i is a vector, and A_i is an n by n Hermitian matrix, for i = 1, …, batchCount.
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of each A_i is supplied in A. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of each A_i is supplied in A.
- n (
int) – IN: [int] the number of rows and columns of each matrix A_i, must be at least 0.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
hipblasComplex/object) – IN: device pointer pointing to the first vector (x_1).
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1).
- AP (
hipblasComplex/object) – INOUT: - device array of device pointers storing the specified triangular portion of
each Hermitian matrix A_i. Points to the first matrix (A_1). if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of each Hermitian matrix A_i is supplied. The lower triangular portion of each A_i will not be touched.
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of each Hermitian matrix A_i is supplied. The upper triangular portion of each A_i will not be touched.
Note that the imaginary part of the diagonal elements are not accessed and are assumed to be 0.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i.
- strideA (
int) – IN: [hipblasStride] stride from the start of one (A_i) and the next (A_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZherStridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCherStridedBatched_v2(handle, uplo, int n, alpha, x, int incx, long stridex, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZherStridedBatched_v2(handle, uplo, int n, alpha, x, int incx, long stridex, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCher2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda)#
BLAS Level 2 API
@{
her2 performs the matrix-vector operations
A := A + alpha*x*y**H + conj(alpha)*y*x**H
where alpha is a complex scalar, x and y are vectors, and A is an n by n Hermitian matrix.
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of A is supplied. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of A is supplied.
- n (
int) – IN: [int] the number of rows and columns of matrix A, must be at least 0.
- alpha (
hipblasComplex/object) – IN: device pointer or host pointer to scalar alpha.
- x (
hipblasComplex/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- y (
hipblasComplex/object) – IN: device pointer storing vector y.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- AP (
hipblasComplex/object) – INOUT: - device pointer storing the specified triangular portion of
the Hermitian matrix A. Of size (lda, n). if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of the Hermitian matrix A is supplied. The lower triangular portion of A will not be touched.
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of the Hermitian matrix A is supplied. The upper triangular portion of A will not be touched.
Note that the imaginary part of the diagonal elements are not accessed and are assumed to be 0.
- lda (
int) – IN: [int] specifies the leading dimension of A. Must be at least max(lda, 1).
- handle (
- hip.hipblas.hipblasZher2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCher2_v2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZher2_v2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCher2Batched(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
BLAS Level 2 API
@{
her2Batched performs the matrix-vector operations
A_i := A_i + alpha*x_i*y_i**H + conj(alpha)*y_i*x_i**H
where alpha is a complex scalar, x_i and y_i are vectors, and A_i is an n by n Hermitian matrix for each batch in i = [1, batchCount].
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of each A_i is supplied. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of each A_i is supplied.
- n (
int) – IN: [int] the number of rows and columns of each matrix A_i, must be at least 0.
- alpha (
hipblasComplex/object) – IN: device pointer or host pointer to scalar alpha.
- x (
Pointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- y (
Pointer/object) – IN: device array of device pointers storing each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- AP (
Pointer/object) – INOUT: - device array of device pointers storing the specified triangular portion of
each Hermitian matrix A_i of size (lda, n). if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of each Hermitian matrix A_i is supplied. The lower triangular portion of each A_i will not be touched.
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of each Hermitian matrix A_i is supplied. The upper triangular portion of each A_i will not be touched.
Note that the imaginary part of the diagonal elements are not accessed and are assumed to be 0.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i. Must be at least max(lda, 1).
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZher2Batched(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCher2Batched_v2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZher2Batched_v2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCher2StridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
BLAS Level 2 API
@{
her2StridedBatched performs the matrix-vector operations
A_i := A_i + alpha*x_i*y_i**H + conj(alpha)*y_i*x_i**H
where alpha is a complex scalar, x_i and y_i are vectors, and A_i is an n by n Hermitian matrix for each batch in i = [1, batchCount].
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of each A_i is supplied. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of each A_i is supplied.
- n (
int) – IN: [int] the number of rows and columns of each matrix A_i, must be at least 0.
- alpha (
hipblasComplex/object) – IN: device pointer or host pointer to scalar alpha.
- x (
hipblasComplex/object) – IN: device pointer pointing to the first vector x_1.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] specifies the stride between the beginning of one vector (x_i) and the next (x_i+1).
- y (
hipblasComplex/object) – IN: device pointer pointing to the first vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- stridey (
int) – IN: [hipblasStride] specifies the stride between the beginning of one vector (y_i) and the next (y_i+1).
- AP (
hipblasComplex/object) – INOUT: - device pointer pointing to the first matrix (A_1). Stores the specified triangular portion of
each Hermitian matrix A_i. if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of each Hermitian matrix A_i is supplied. The lower triangular portion of each A_i will not be touched.
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of each Hermitian matrix A_i is supplied. The upper triangular portion of each A_i will not be touched.
Note that the imaginary part of the diagonal elements are not accessed and are assumed to be 0.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i. Must be at least max(lda, 1).
- strideA (
int) – IN: [hipblasStride] specifies the stride between the beginning of one matrix (A_i) and the next (A_i+1).
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZher2StridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCher2StridedBatched_v2(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZher2StridedBatched_v2(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChpmv(handle, uplo, int n, alpha, AP, x, int incx, beta, y, int incy)#
BLAS Level 2 API
@{
hpmv performs the matrix-vector operation
y := alpha*A*x + beta*y
where alpha and beta are scalars, x and y are n element vectors and A is an n by n Hermitian matrix, supplied in packed form (see description below).
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: the upper triangular part of the Hermitian matrix A is supplied in AP. HIPBLAS_FILL_MODE_LOWER: the lower triangular part of the Hermitian matrix A is supplied in AP.
- n (
int) – IN: [int] the order of the matrix A, must be >= 0.
- alpha (
hipblasComplex/object) – IN: device pointer or host pointer to scalar alpha.
- AP (
hipblasComplex/object) – IN: - device pointer storing the packed version of the specified triangular portion of
the Hermitian matrix A. Of at least size ((n * (n + 1)) / 2). if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of the Hermitian matrix A is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(0,1) AP(2) = A(1,1), etc.
- Ex: (HIPBLAS_FILL_MODE_UPPER; n = 3)
(1, 0) (2, 1) (3, 2) (2,-1) (4, 0) (5,-1) —–> [(1,0), (2,1), (4,0), (3,2), (5,-1), (6,0)] (3,-2) (5, 1) (6, 0)
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of the Hermitian matrix A is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(1,0) AP(2) = A(2,1), etc.
- Ex: (HIPBLAS_FILL_MODE_LOWER; n = 3)
(1, 0) (2, 1) (3, 2) (2,-1) (4, 0) (5,-1) —–> [(1,0), (2,-1), (3,-2), (4,0), (5,1), (6,0)] (3,-2) (5, 1) (6, 0)
Note that the imaginary part of the diagonal elements are not accessed and are assumed to be 0.
- x (
hipblasComplex/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- beta (
hipblasComplex/object) – IN: device pointer or host pointer to scalar beta.
- y (
hipblasComplex/object) – INOUT: device pointer storing vector y.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- handle (
- hip.hipblas.hipblasZhpmv(handle, uplo, int n, alpha, AP, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChpmv_v2(handle, uplo, int n, alpha, AP, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZhpmv_v2(handle, uplo, int n, alpha, AP, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChpmvBatched(handle, uplo, int n, alpha, AP, x, int incx, beta, y, int incy, int batchCount)#
BLAS Level 2 API
@{
hpmvBatched performs the matrix-vector operation
y_i := alpha*A_i*x_i + beta*y_i
where alpha and beta are scalars, x_i and y_i are n element vectors and A_i is an n by n Hermitian matrix, supplied in packed form (see description below), for each batch in i = [1, batchCount].
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: the upper triangular part of each Hermitian matrix A_i is supplied in AP. HIPBLAS_FILL_MODE_LOWER: the lower triangular part of each Hermitian matrix A_i is supplied in AP.
- n (
int) – IN: [int] the order of each matrix A_i.
- alpha (
hipblasComplex/object) – IN: device pointer or host pointer to scalar alpha.
- AP (
Pointer/object) – IN: - device pointer of device pointers storing the packed version of the specified triangular
portion of each Hermitian matrix A_i. Each A_i is of at least size ((n * (n + 1)) / 2). if uplo == HIPBLAS_FILL_MODE_UPPER: The upper triangular portion of each Hermitian matrix A_i is supplied. The matrix is compacted so that each AP_i contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(0,1) AP(2) = A(1,1), etc.
- Ex: (HIPBLAS_FILL_MODE_UPPER; n = 3)
(1, 0) (2, 1) (3, 2) (2,-1) (4, 0) (5,-1) —–> [(1,0), (2,1), (4,0), (3,2), (5,-1), (6,0)] (3,-2) (5, 1) (6, 0)
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of each Hermitian matrix A_i is supplied. The matrix is compacted so that each AP_i contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(1,0) AP(2) = A(2,1), etc.
- Ex: (HIPBLAS_FILL_MODE_LOWER; n = 3)
(1, 0) (2, 1) (3, 2) (2,-1) (4, 0) (5,-1) —–> [(1,0), (2,-1), (3,-2), (4,0), (5,1), (6,0)] (3,-2) (5, 1) (6, 0)
Note that the imaginary part of the diagonal elements are not accessed and are assumed to be 0.
- x (
Pointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- beta (
hipblasComplex/object) – IN: device pointer or host pointer to scalar beta.
- y (
Pointer/object) – INOUT: device array of device pointers storing each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZhpmvBatched(handle, uplo, int n, alpha, AP, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChpmvBatched_v2(handle, uplo, int n, alpha, AP, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZhpmvBatched_v2(handle, uplo, int n, alpha, AP, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChpmvStridedBatched(handle, uplo, int n, alpha, AP, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
BLAS Level 2 API
@{
hpmvStridedBatched performs the matrix-vector operation
y_i := alpha*A_i*x_i + beta*y_i
where alpha and beta are scalars, x_i and y_i are n element vectors and A_i is an n by n Hermitian matrix, supplied in packed form (see description below), for each batch in i = [1, batchCount].
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: the upper triangular part of each Hermitian matrix A_i is supplied in AP. HIPBLAS_FILL_MODE_LOWER: the lower triangular part of each Hermitian matrix A_i is supplied in AP.
- n (
int) – IN: [int] the order of each matrix A_i.
- alpha (
hipblasComplex/object) – IN: device pointer or host pointer to scalar alpha.
- AP (
hipblasComplex/object) – IN: - device pointer pointing to the beginning of the first matrix (AP_1). Stores the packed
version of the specified triangular portion of each Hermitian matrix AP_i of size ((n * (n + 1)) / 2). if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of each Hermitian matrix A_i is supplied. The matrix is compacted so that each AP_i contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(0,1) AP(2) = A(1,1), etc.
- Ex: (HIPBLAS_FILL_MODE_UPPER; n = 3)
(1, 0) (2, 1) (3, 2) (2,-1) (4, 0) (5,-1) —–> [(1,0), (2,1), (4,0), (3,2), (5,-1), (6,0)] (3,-2) (5, 1) (6, 0)
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of each Hermitian matrix A_i is supplied. The matrix is compacted so that each AP_i contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(1,0) AP(2) = A(2,1), etc.
- Ex: (HIPBLAS_FILL_MODE_LOWER; n = 3)
(1, 0) (2, 1) (3, 2) (2,-1) (4, 0) (5,-1) —–> [(1,0), (2,-1), (3,-2), (4,0), (5,1), (6,0)] (3,-2) (5, 1) (6, 0)
Note that the imaginary part of the diagonal elements are not accessed and are assumed to be 0.
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (AP_i) and the next one (AP_i+1).
- x (
hipblasComplex/object) – IN: device array pointing to the beginning of the first vector (x_1).
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1).
- beta (
hipblasComplex/object) – IN: device pointer or host pointer to scalar beta.
- y (
hipblasComplex/object) – INOUT: device array pointing to the beginning of the first vector (y_1).
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) and the next one (y_i+1).
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZhpmvStridedBatched(handle, uplo, int n, alpha, AP, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChpmvStridedBatched_v2(handle, uplo, int n, alpha, AP, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- strideA (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZhpmvStridedBatched_v2(handle, uplo, int n, alpha, AP, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- strideA (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChpr(handle, uplo, int n, alpha, x, int incx, AP)#
BLAS Level 2 API
@{
hpr performs the matrix-vector operations
A := A + alpha*x*x**H
where alpha is a real scalar, x is a vector, and A is an n by n Hermitian matrix, supplied in packed form.
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of A is supplied in AP. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of A is supplied in AP.
- n (
int) – IN: [int] the number of rows and columns of matrix A, must be at least 0.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
hipblasComplex/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- AP (
hipblasComplex/object) – INOUT: - device pointer storing the packed version of the specified triangular portion of
the Hermitian matrix A. Of at least size ((n * (n + 1)) / 2). if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of the Hermitian matrix A is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(0,1) AP(2) = A(1,1), etc.
- Ex: (HIPBLAS_FILL_MODE_UPPER; n = 3)
(1, 0) (2, 1) (4,9) (2,-1) (3, 0) (5,3) —–> [(1,0), (2,1), (3,0), (4,9), (5,3), (6,0)] (4,-9) (5,-3) (6,0)
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of the Hermitian matrix A is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(1,0) AP(2) = A(2,1), etc.
- Ex: (HIPBLAS_FILL_MODE_LOWER; n = 3)
(1, 0) (2, 1) (4,9) (2,-1) (3, 0) (5,3) —–> [(1,0), (2,-1), (4,-9), (3,0), (5,-3), (6,0)] (4,-9) (5,-3) (6,0)
Note that the imaginary part of the diagonal elements are not accessed and are assumed to be 0.
- handle (
- hip.hipblas.hipblasZhpr(handle, uplo, int n, alpha, x, int incx, AP)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasChpr_v2(handle, uplo, int n, alpha, x, int incx, AP)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZhpr_v2(handle, uplo, int n, alpha, x, int incx, AP)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasChprBatched(handle, uplo, int n, alpha, x, int incx, AP, int batchCount)#
BLAS Level 2 API
@{
hprBatched performs the matrix-vector operations
A_i := A_i + alpha*x_i*x_i**H
where alpha is a real scalar, x_i is a vector, and A_i is an n by n symmetric matrix, supplied in packed form, for i = 1, …, batchCount.
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of each A_i is supplied in AP. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of each A_i is supplied in AP.
- n (
int) – IN: [int] the number of rows and columns of each matrix A_i, must be at least 0.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
Pointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- AP (
Pointer/object) – INOUT: - device array of device pointers storing the packed version of the specified triangular portion of
each Hermitian matrix A_i of at least size ((n * (n + 1)) / 2). Array is of at least size batchCount. if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of each Hermitian matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(0,1) AP(2) = A(1,1), etc.
- Ex: (HIPBLAS_FILL_MODE_UPPER; n = 3)
(1, 0) (2, 1) (4,9) (2,-1) (3, 0) (5,3) —–> [(1,0), (2,1), (3,0), (4,9), (5,3), (6,0)] (4,-9) (5,-3) (6,0)
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of each Hermitian matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(1,0) AP(2) = A(2,1), etc.
- Ex: (HIPBLAS_FILL_MODE_LOWER; n = 3)
(1, 0) (2, 1) (4,9) (2,-1) (3, 0) (5,3) —–> [(1,0), (2,-1), (4,-9), (3,0), (5,-3), (6,0)] (4,-9) (5,-3) (6,0)
Note that the imaginary part of the diagonal elements are not accessed and are assumed to be 0.
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZhprBatched(handle, uplo, int n, alpha, x, int incx, AP, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasChprBatched_v2(handle, uplo, int n, alpha, x, int incx, AP, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZhprBatched_v2(handle, uplo, int n, alpha, x, int incx, AP, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasChprStridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, AP, long strideA, int batchCount)#
BLAS Level 2 API
@{
hprStridedBatched performs the matrix-vector operations
A_i := A_i + alpha*x_i*x_i**H
where alpha is a real scalar, x_i is a vector, and A_i is an n by n symmetric matrix, supplied in packed form, for i = 1, …, batchCount.
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of each A_i is supplied in AP. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of each A_i is supplied in AP.
- n (
int) – IN: [int] the number of rows and columns of each matrix A_i, must be at least 0.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
hipblasComplex/object) – IN: device pointer pointing to the first vector (x_1).
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1).
- AP (
hipblasComplex/object) – INOUT: - device array of device pointers storing the packed version of the specified triangular portion of
each Hermitian matrix A_i. Points to the first matrix (A_1). if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of each Hermitian matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(0,1) AP(2) = A(1,1), etc.
- Ex: (HIPBLAS_FILL_MODE_UPPER; n = 3)
(1, 0) (2, 1) (4,9) (2,-1) (3, 0) (5,3) —–> [(1,0), (2,1), (3,0), (4,9), (5,3), (6,0)] (4,-9) (5,-3) (6,0)
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of each Hermitian matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(1,0) AP(2) = A(2,1), etc.
- Ex: (HIPBLAS_FILL_MODE_LOWER; n = 3)
(1, 0) (2, 1) (4,9) (2,-1) (3, 0) (5,3) —–> [(1,0), (2,-1), (4,-9), (3,0), (5,-3), (6,0)] (4,-9) (5,-3) (6,0)
Note that the imaginary part of the diagonal elements are not accessed and are assumed to be 0.
- strideA (
int) – IN: [hipblasStride] stride from the start of one (A_i) and the next (A_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZhprStridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, AP, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChprStridedBatched_v2(handle, uplo, int n, alpha, x, int incx, long stridex, AP, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
float2/object): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZhprStridedBatched_v2(handle, uplo, int n, alpha, x, int incx, long stridex, AP, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
double2/object): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChpr2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP)#
BLAS Level 2 API
@{
hpr2 performs the matrix-vector operations
A := A + alpha*x*y**H + conj(alpha)*y*x**H
where alpha is a complex scalar, x and y are vectors, and A is an n by n Hermitian matrix, supplied in packed form.
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of A is supplied in AP. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of A is supplied in AP.
- n (
int) – IN: [int] the number of rows and columns of matrix A, must be at least 0.
- alpha (
hipblasComplex/object) – IN: device pointer or host pointer to scalar alpha.
- x (
hipblasComplex/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- y (
hipblasComplex/object) – IN: device pointer storing vector y.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- AP (
hipblasComplex/object) – INOUT: - device pointer storing the packed version of the specified triangular portion of
the Hermitian matrix A. Of at least size ((n * (n + 1)) / 2). if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of the Hermitian matrix A is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(0,1) AP(2) = A(1,1), etc.
- Ex: (HIPBLAS_FILL_MODE_UPPER; n = 3)
(1, 0) (2, 1) (4,9) (2,-1) (3, 0) (5,3) —–> [(1,0), (2,1), (3,0), (4,9), (5,3), (6,0)] (4,-9) (5,-3) (6,0)
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of the Hermitian matrix A is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(1,0) AP(2) = A(2,1), etc.
- Ex: (HIPBLAS_FILL_MODE_LOWER; n = 3)
(1, 0) (2, 1) (4,9) (2,-1) (3, 0) (5,3) —–> [(1,0), (2,-1), (4,-9), (3,0), (5,-3), (6,0)] (4,-9) (5,-3) (6,0)
Note that the imaginary part of the diagonal elements are not accessed and are assumed to be 0.
- handle (
- hip.hipblas.hipblasZhpr2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasChpr2_v2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZhpr2_v2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP)#
(No short description, might be part of a group.)
- Args:
- hip.hipblas.hipblasChpr2Batched(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int batchCount)#
BLAS Level 2 API
@{
hpr2Batched performs the matrix-vector operations
A_i := A_i + alpha*x_i*y_i**H + conj(alpha)*y_i*x_i**H
where alpha is a complex scalar, x_i and y_i are vectors, and A_i is an n by n symmetric matrix, supplied in packed form, for i = 1, …, batchCount.
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of each A_i is supplied in AP. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of each A_i is supplied in AP.
- n (
int) – IN: [int] the number of rows and columns of each matrix A_i, must be at least 0.
- alpha (
hipblasComplex/object) – IN: device pointer or host pointer to scalar alpha.
- x (
Pointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- y (
Pointer/object) – IN: device array of device pointers storing each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- AP (
Pointer/object) – INOUT: - device array of device pointers storing the packed version of the specified triangular portion of
each Hermitian matrix A_i of at least size ((n * (n + 1)) / 2). Array is of at least size batchCount. if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of each Hermitian matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(0,1) AP(2) = A(1,1), etc.
- Ex: (HIPBLAS_FILL_MODE_UPPER; n = 3)
(1, 0) (2, 1) (4,9) (2,-1) (3, 0) (5,3) —–> [(1,0), (2,1), (3,0), (4,9), (5,3), (6,0)] (4,-9) (5,-3) (6,0)
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of each Hermitian matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(1,0) AP(2) = A(2,1), etc.
- Ex: (HIPBLAS_FILL_MODE_LOWER; n = 3)
(1, 0) (2, 1) (4,9) (2,-1) (3, 0) (5,3) —–> [(1,0), (2,-1), (4,-9), (3,0), (5,-3), (6,0)] (4,-9) (5,-3) (6,0)
Note that the imaginary part of the diagonal elements are not accessed and are assumed to be 0.
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZhpr2Batched(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChpr2Batched_v2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZhpr2Batched_v2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChpr2StridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, long strideA, int batchCount)#
BLAS Level 2 API
@{
hpr2StridedBatched performs the matrix-vector operations
A_i := A_i + alpha*x_i*y_i**H + conj(alpha)*y_i*x_i**H
where alpha is a complex scalar, x_i and y_i are vectors, and A_i is an n by n symmetric matrix, supplied in packed form, for i = 1, …, batchCount.
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of each A_i is supplied in AP. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of each A_i is supplied in AP.
- n (
int) – IN: [int] the number of rows and columns of each matrix A_i, must be at least 0.
- alpha (
hipblasComplex/object) – IN: device pointer or host pointer to scalar alpha.
- x (
hipblasComplex/object) – IN: device pointer pointing to the first vector (x_1).
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1).
- y (
hipblasComplex/object) – IN: device pointer pointing to the first vector (y_1).
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) and the next one (y_i+1).
- AP (
hipblasComplex/object) – INOUT: - device array of device pointers storing the packed version of the specified triangular portion of
each Hermitian matrix A_i. Points to the first matrix (A_1). if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of each Hermitian matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(0,1) AP(2) = A(1,1), etc.
- Ex: (HIPBLAS_FILL_MODE_UPPER; n = 3)
(1, 0) (2, 1) (4,9) (2,-1) (3, 0) (5,3) —–> [(1,0), (2,1), (3,0), (4,9), (5,3), (6,0)] (4,-9) (5,-3) (6,0)
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of each Hermitian matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(1,0) AP(2) = A(2,1), etc.
- Ex: (HIPBLAS_FILL_MODE_LOWER; n = 3)
(1, 0) (2, 1) (4,9) (2,-1) (3, 0) (5,3) —–> [(1,0), (2,-1), (4,-9), (3,0), (5,-3), (6,0)] (4,-9) (5,-3) (6,0)
Note that the imaginary part of the diagonal elements are not accessed and are assumed to be 0.
- strideA (
int) – IN: [hipblasStride] stride from the start of one (A_i) and the next (A_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZhpr2StridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChpr2StridedBatched_v2(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
float2/object): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZhpr2StridedBatched_v2(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
double2/object): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsbmv(handle, uplo, int n, int k, alpha, AP, int lda, x, int incx, beta, y, int incy)#
BLAS Level 2 API
@{
sbmv performs the matrix-vector operation:
y := alpha*A*x + beta*y,
where alpha and beta are scalars, x and y are n element vectors and A should contain an upper or lower triangular n by n symmetric banded matrix.
Supported precisions in rocBLAS : s,d
Supported precisions in cuBLAS : s,d
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- n (
int) – IN: [int]
- k (
int) – IN: [int] specifies the number of sub- and super-diagonals
- alpha (
Pointer/object) – IN: specifies the scalar alpha
- AP (
Pointer/object) – IN: pointer storing matrix A on the GPU
- lda (
int) – IN: [int] specifies the leading dimension of matrix A
- x (
Pointer/object) – IN: pointer storing vector x on the GPU
- incx (
int) – IN: [int] specifies the increment for the elements of x
- beta (
Pointer/object) – IN: specifies the scalar beta
- y (
Pointer/object) – OUT: pointer storing vector y on the GPU
- incy (
int) – IN: [int] specifies the increment for the elements of y
- handle (
- hip.hipblas.hipblasDsbmv(handle, uplo, int n, int k, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsbmvBatched(handle, uplo, int n, int k, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
BLAS Level 2 API
@{
sbmvBatched performs the matrix-vector operation:
y_i := alpha*A_i*x_i + beta*y_i,
where (A_i, x_i, y_i) is the i-th instance of the batch. alpha and beta are scalars, x_i and y_i are vectors and A_i is an n by n symmetric banded matrix, for i = 1, …, batchCount. A should contain an upper or lower triangular n by n symmetric banded matrix.
Supported precisions in rocBLAS : s,d
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- n (
int) – IN: [int] number of rows and columns of each matrix A_i
- k (
int) – IN: [int] specifies the number of sub- and super-diagonals
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix A_i
- lda (
int) – IN: [int] specifies the leading dimension of each matrix A_i
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i
- incx (
int) – IN: [int] specifies the increment for the elements of each vector x_i
- beta (
Pointer/object) – IN: device pointer or host pointer to scalar beta
- y (
ListOfPointer/object) – OUT: device array of device pointers storing each vector y_i
- incy (
int) – IN: [int] specifies the increment for the elements of each vector y_i
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDsbmvBatched(handle, uplo, int n, int k, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- y (
ListOfPointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsbmvStridedBatched(handle, uplo, int n, int k, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
BLAS Level 2 API
@{
sbmvStridedBatched performs the matrix-vector operation:
y_i := alpha*A_i*x_i + beta*y_i,
where (A_i, x_i, y_i) is the i-th instance of the batch. alpha and beta are scalars, x_i and y_i are vectors and A_i is an n by n symmetric banded matrix, for i = 1, …, batchCount. A should contain an upper or lower triangular n by n symmetric banded matrix.
Supported precisions in rocBLAS : s,d
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- n (
int) – IN: [int] number of rows and columns of each matrix A_i
- k (
int) – IN: [int] specifies the number of sub- and super-diagonals
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha
- AP (
Pointer/object) – IN: Device pointer to the first matrix A_1 on the GPU
- lda (
int) – IN: [int] specifies the leading dimension of each matrix A_i
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- x (
Pointer/object) – IN: Device pointer to the first vector x_1 on the GPU
- incx (
int) – IN: [int] specifies the increment for the elements of each vector x_i
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1). There are no restrictions placed on stridex, however the user should take care to ensure that stridex is of appropriate size. This typically means stridex >= n * incx. stridex should be non zero.
- beta (
Pointer/object) – IN: device pointer or host pointer to scalar beta
- y (
Pointer/object) – OUT: Device pointer to the first vector y_1 on the GPU
- incy (
int) – IN: [int] specifies the increment for the elements of each vector y_i
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) and the next one (y_i+1). There are no restrictions placed on stridey, however the user should take care to ensure that stridey is of appropriate size. This typically means stridey >= n * incy. stridey should be non zero.
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDsbmvStridedBatched(handle, uplo, int n, int k, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSspmv(handle, uplo, int n, alpha, AP, x, int incx, beta, y, int incy)#
BLAS Level 2 API
@{
spmv performs the matrix-vector operation:
y := alpha*A*x + beta*y,
where alpha and beta are scalars, x and y are n element vectors and A should contain an upper or lower triangular n by n packed symmetric matrix.
Supported precisions in rocBLAS : s,d
Supported precisions in cuBLAS : s,d
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- n (
int) – IN: [int]
- alpha (
Pointer/object) – IN: specifies the scalar alpha
- AP (
Pointer/object) – IN: pointer storing matrix A on the GPU
- x (
Pointer/object) – IN: pointer storing vector x on the GPU
- incx (
int) – IN: [int] specifies the increment for the elements of x
- beta (
Pointer/object) – IN: specifies the scalar beta
- y (
Pointer/object) – OUT: pointer storing vector y on the GPU
- incy (
int) – IN: [int] specifies the increment for the elements of y
- handle (
- hip.hipblas.hipblasDspmv(handle, uplo, int n, alpha, AP, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSspmvBatched(handle, uplo, int n, alpha, AP, x, int incx, beta, y, int incy, int batchCount)#
BLAS Level 2 API
@{
spmvBatched performs the matrix-vector operation:
y_i := alpha*AP_i*x_i + beta*y_i,
where (A_i, x_i, y_i) is the i-th instance of the batch. alpha and beta are scalars, x_i and y_i are vectors and A_i is an n by n symmetric matrix, for i = 1, …, batchCount. A should contain an upper or lower triangular n by n packed symmetric matrix.
Supported precisions in rocBLAS : s,d
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- n (
int) – IN: [int] number of rows and columns of each matrix A_i
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix A_i
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i
- incx (
int) – IN: [int] specifies the increment for the elements of each vector x_i
- beta (
Pointer/object) – IN: device pointer or host pointer to scalar beta
- y (
ListOfPointer/object) – OUT: device array of device pointers storing each vector y_i
- incy (
int) – IN: [int] specifies the increment for the elements of each vector y_i
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDspmvBatched(handle, uplo, int n, alpha, AP, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- y (
ListOfPointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSspmvStridedBatched(handle, uplo, int n, alpha, AP, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
BLAS Level 2 API
@{
spmvStridedBatched performs the matrix-vector operation:
y_i := alpha*A_i*x_i + beta*y_i,
where (A_i, x_i, y_i) is the i-th instance of the batch. alpha and beta are scalars, x_i and y_i are vectors and A_i is an n by n symmetric matrix, for i = 1, …, batchCount. A should contain an upper or lower triangular n by n packed symmetric matrix.
Supported precisions in rocBLAS : s,d
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- n (
int) – IN: [int] number of rows and columns of each matrix A_i
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha
- AP (
Pointer/object) – IN: Device pointer to the first matrix A_1 on the GPU
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- x (
Pointer/object) – IN: Device pointer to the first vector x_1 on the GPU
- incx (
int) – IN: [int] specifies the increment for the elements of each vector x_i
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1). There are no restrictions placed on stridex, however the user should take care to ensure that stridex is of appropriate size. This typically means stridex >= n * incx. stridex should be non zero.
- beta (
Pointer/object) – IN: device pointer or host pointer to scalar beta
- y (
Pointer/object) – OUT: Device pointer to the first vector y_1 on the GPU
- incy (
int) – IN: [int] specifies the increment for the elements of each vector y_i
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) and the next one (y_i+1). There are no restrictions placed on stridey, however the user should take care to ensure that stridey is of appropriate size. This typically means stridey >= n * incy. stridey should be non zero.
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDspmvStridedBatched(handle, uplo, int n, alpha, AP, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- strideA (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSspr(handle, uplo, int n, alpha, x, int incx, AP)#
BLAS Level 2 API
@{
spr performs the matrix-vector operations
A := A + alpha*x*x**T
where alpha is a scalar, x is a vector, and A is an n by n symmetric matrix, supplied in packed form.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of A is supplied in AP. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of A is supplied in AP.
- n (
int) – IN: [int] the number of rows and columns of matrix A, must be at least 0.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- AP (
Pointer/object) – INOUT: - device pointer storing the packed version of the specified triangular portion of
the symmetric matrix A. Of at least size ((n * (n + 1)) / 2). if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of the symmetric matrix A is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(0,1) AP(2) = A(1,1), etc.
- Ex: (HIPBLAS_FILL_MODE_UPPER; n = 4)
1 2 4 7 2 3 5 8 —–> [1, 2, 3, 4, 5, 6, 7, 8, 9, 0] 4 5 6 9 7 8 9 0
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of the symmetric matrix A is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(1,0) AP(2) = A(2,1), etc.
- Ex: (HIPBLAS_FILL_MODE_LOWER; n = 4)
1 2 3 4 2 5 6 7 —–> [1, 2, 3, 4, 5, 6, 7, 8, 9, 0] 3 6 8 9 4 7 9 0
- handle (
- hip.hipblas.hipblasDspr(handle, uplo, int n, alpha, x, int incx, AP)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCspr(handle, uplo, int n, alpha, x, int incx, AP)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasZspr(handle, uplo, int n, alpha, x, int incx, AP)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- handle (
- hip.hipblas.hipblasCspr_v2(handle, uplo, int n, alpha, x, int incx, AP)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZspr_v2(handle, uplo, int n, alpha, x, int incx, AP)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSsprBatched(handle, uplo, int n, alpha, x, int incx, AP, int batchCount)#
BLAS Level 2 API
@{
sprBatched performs the matrix-vector operations
A_i := A_i + alpha*x_i*x_i**T
where alpha is a scalar, x_i is a vector, and A_i is an n by n symmetric matrix, supplied in packed form, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of each A_i is supplied in AP. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of each A_i is supplied in AP.
- n (
int) – IN: [int] the number of rows and columns of each matrix A_i, must be at least 0.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- AP (
Pointer/object) – INOUT: - device array of device pointers storing the packed version of the specified triangular portion of
each symmetric matrix A_i of at least size ((n * (n + 1)) / 2). Array is of at least size batchCount. if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of each symmetric matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(0,1) AP(2) = A(1,1), etc.
- Ex: (HIPBLAS_FILL_MODE_UPPER; n = 4)
1 2 4 7 2 3 5 8 —–> [1, 2, 3, 4, 5, 6, 7, 8, 9, 0] 4 5 6 9 7 8 9 0
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of each symmetric matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(1,0) AP(2) = A(2,1), etc.
- Ex: (HIPBLAS_FILL_MODE_LOWER; n = 4)
1 2 3 4 2 5 6 7 —–> [1, 2, 3, 4, 5, 6, 7, 8, 9, 0] 3 6 8 9 4 7 9 0
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDsprBatched(handle, uplo, int n, alpha, x, int incx, AP, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCsprBatched(handle, uplo, int n, alpha, x, int incx, AP, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZsprBatched(handle, uplo, int n, alpha, x, int incx, AP, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCsprBatched_v2(handle, uplo, int n, alpha, x, int incx, AP, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZsprBatched_v2(handle, uplo, int n, alpha, x, int incx, AP, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSsprStridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, AP, long strideA, int batchCount)#
BLAS Level 2 API
@{
sprStridedBatched performs the matrix-vector operations
A_i := A_i + alpha*x_i*x_i**T
where alpha is a scalar, x_i is a vector, and A_i is an n by n symmetric matrix, supplied in packed form, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of each A_i is supplied in AP. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of each A_i is supplied in AP.
- n (
int) – IN: [int] the number of rows and columns of each matrix A_i, must be at least 0.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
Pointer/object) – IN: device pointer pointing to the first vector (x_1).
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1).
- AP (
Pointer/object) – INOUT: - device pointer storing the packed version of the specified triangular portion of
each symmetric matrix A_i. Points to the first A_1. if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of each symmetric matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(0,1) AP(2) = A(1,1), etc.
- Ex: (HIPBLAS_FILL_MODE_UPPER; n = 4)
1 2 4 7 2 3 5 8 —–> [1, 2, 3, 4, 5, 6, 7, 8, 9, 0] 4 5 6 9 7 8 9 0
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of each symmetric matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(1,0) AP(2) = A(2,1), etc.
- Ex: (HIPBLAS_FILL_MODE_LOWER; n = 4)
1 2 3 4 2 5 6 7 —–> [1, 2, 3, 4, 5, 6, 7, 8, 9, 0] 3 6 8 9 4 7 9 0
- strideA (
int) – IN: [hipblasStride] stride from the start of one (A_i) and the next (A_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDsprStridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, AP, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsprStridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, AP, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsprStridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, AP, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsprStridedBatched_v2(handle, uplo, int n, alpha, x, int incx, long stridex, AP, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
float2/object): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsprStridedBatched_v2(handle, uplo, int n, alpha, x, int incx, long stridex, AP, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
double2/object): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSspr2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP)#
BLAS Level 2 API
@{
spr2 performs the matrix-vector operation
A := A + alpha*x*y**T + alpha*y*x**T
where alpha is a scalar, x and y are vectors, and A is an n by n symmetric matrix, supplied in packed form.
Supported precisions in rocBLAS : s,d
Supported precisions in cuBLAS : s,d
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of A is supplied in AP. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of A is supplied in AP.
- n (
int) – IN: [int] the number of rows and columns of matrix A, must be at least 0.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- y (
Pointer/object) – IN: device pointer storing vector y.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- AP (
Pointer/object) – INOUT: - device pointer storing the packed version of the specified triangular portion of
the symmetric matrix A. Of at least size ((n * (n + 1)) / 2). if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of the symmetric matrix A is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(0,1) AP(2) = A(1,1), etc.
- Ex: (HIPBLAS_FILL_MODE_UPPER; n = 4)
1 2 4 7 2 3 5 8 —–> [1, 2, 3, 4, 5, 6, 7, 8, 9, 0] 4 5 6 9 7 8 9 0
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of the symmetric matrix A is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(1,0) AP(n) = A(2,1), etc.
- Ex: (HIPBLAS_FILL_MODE_LOWER; n = 4)
1 2 3 4 2 5 6 7 —–> [1, 2, 3, 4, 5, 6, 7, 8, 9, 0] 3 6 8 9 4 7 9 0
- handle (
- hip.hipblas.hipblasDspr2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP)#
(No short description, might be part of a group.)
- Args:
- hip.hipblas.hipblasSspr2Batched(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int batchCount)#
BLAS Level 2 API
@{
spr2Batched performs the matrix-vector operation
A_i := A_i + alpha*x_i*y_i**T + alpha*y_i*x_i**T
where alpha is a scalar, x_i and y_i are vectors, and A_i is an n by n symmetric matrix, supplied in packed form, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of each A_i is supplied in AP. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of each A_i is supplied in AP.
- n (
int) – IN: [int] the number of rows and columns of each matrix A_i, must be at least 0.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- y (
ListOfPointer/object) – IN: device array of device pointers storing each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- AP (
Pointer/object) – INOUT: - device array of device pointers storing the packed version of the specified triangular portion of
each symmetric matrix A_i of at least size ((n * (n + 1)) / 2). Array is of at least size batchCount. if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of each symmetric matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(0,1) AP(2) = A(1,1), etc.
- Ex: (HIPBLAS_FILL_MODE_UPPER; n = 4)
1 2 4 7 2 3 5 8 —–> [1, 2, 3, 4, 5, 6, 7, 8, 9, 0] 4 5 6 9 7 8 9 0
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of each symmetric matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(1,0) AP(n) = A(2,1), etc.
- Ex: (HIPBLAS_FILL_MODE_LOWER; n = 4)
1 2 3 4 2 5 6 7 —–> [1, 2, 3, 4, 5, 6, 7, 8, 9, 0] 3 6 8 9 4 7 9 0
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDspr2Batched(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
ListOfPointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSspr2StridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, long strideA, int batchCount)#
BLAS Level 2 API
@{
spr2StridedBatched performs the matrix-vector operation
A_i := A_i + alpha*x_i*y_i**T + alpha*y_i*x_i**T
where alpha is a scalar, x_i amd y_i are vectors, and A_i is an n by n symmetric matrix, supplied in packed form, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ HIPBLAS_FILL_MODE_UPPER: The upper triangular part of each A_i is supplied in AP. HIPBLAS_FILL_MODE_LOWER: The lower triangular part of each A_i is supplied in AP.
- n (
int) – IN: [int] the number of rows and columns of each matrix A_i, must be at least 0.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
Pointer/object) – IN: device pointer pointing to the first vector (x_1).
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1).
- y (
Pointer/object) – IN: device pointer pointing to the first vector (y_1).
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) and the next one (y_i+1).
- AP (
Pointer/object) – INOUT: - device pointer storing the packed version of the specified triangular portion of
each symmetric matrix A_i. Points to the first A_1. if uplo == HIPBLAS_FILL_MODE_UPPER:
The upper triangular portion of each symmetric matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(0,1) AP(2) = A(1,1), etc.
- Ex: (HIPBLAS_FILL_MODE_UPPER; n = 4)
1 2 4 7 2 3 5 8 —–> [1, 2, 3, 4, 5, 6, 7, 8, 9, 0] 4 5 6 9 7 8 9 0
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The lower triangular portion of each symmetric matrix A_i is supplied. The matrix is compacted so that AP contains the triangular portion column-by-column so that: AP(0) = A(0,0) AP(1) = A(1,0) AP(n) = A(2,1), etc.
- Ex: (HIPBLAS_FILL_MODE_LOWER; n = 4)
1 2 3 4 2 5 6 7 —–> [1, 2, 3, 4, 5, 6, 7, 8, 9, 0] 3 6 8 9 4 7 9 0
- strideA (
int) – IN: [hipblasStride] stride from the start of one (A_i) and the next (A_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDspr2StridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsymv(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
BLAS Level 2 API
@{
symv performs the matrix-vector operation:
y := alpha*A*x + beta*y,
where alpha and beta are scalars, x and y are n element vectors and A should contain an upper or lower triangular n by n symmetric matrix.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- n (
int) – IN: [int]
- alpha (
Pointer/object) – IN: specifies the scalar alpha
- AP (
Pointer/object) – IN: pointer storing matrix A on the GPU
- lda (
int) – IN: [int] specifies the leading dimension of A
- x (
Pointer/object) – IN: pointer storing vector x on the GPU
- incx (
int) – IN: [int] specifies the increment for the elements of x
- beta (
Pointer/object) – IN: specifies the scalar beta
- y (
Pointer/object) – OUT: pointer storing vector y on the GPU
- incy (
int) – IN: [int] specifies the increment for the elements of y
- handle (
- hip.hipblas.hipblasDsymv(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsymv(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsymv(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsymv_v2(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsymv_v2(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsymvBatched(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
BLAS Level 2 API
@{
symvBatched performs the matrix-vector operation:
y_i := alpha*A_i*x_i + beta*y_i,
where (A_i, x_i, y_i) is the i-th instance of the batch. alpha and beta are scalars, x_i and y_i are vectors and A_i is an n by n symmetric matrix, for i = 1, …, batchCount. A a should contain an upper or lower triangular symmetric matrix and the opposing triangular part of A is not referenced
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- n (
int) – IN: [int] number of rows and columns of each matrix A_i
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix A_i
- lda (
int) – IN: [int] specifies the leading dimension of each matrix A_i
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i
- incx (
int) – IN: [int] specifies the increment for the elements of each vector x_i
- beta (
Pointer/object) – IN: device pointer or host pointer to scalar beta
- y (
ListOfPointer/object) – OUT: device array of device pointers storing each vector y_i
- incy (
int) – IN: [int] specifies the increment for the elements of each vector y_i
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDsymvBatched(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- y (
ListOfPointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsymvBatched(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsymvBatched(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsymvBatched_v2(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsymvBatched_v2(handle, uplo, int n, alpha, AP, int lda, x, int incx, beta, y, int incy, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsymvStridedBatched(handle, uplo, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
BLAS Level 2 API
@{
symvStridedBatched performs the matrix-vector operation:
y_i := alpha*A_i*x_i + beta*y_i,
where (A_i, x_i, y_i) is the i-th instance of the batch. alpha and beta are scalars, x_i and y_i are vectors and A_i is an n by n symmetric matrix, for i = 1, …, batchCount. A a should contain an upper or lower triangular symmetric matrix and the opposing triangular part of A is not referenced
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- n (
int) – IN: [int] number of rows and columns of each matrix A_i
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha
- AP (
Pointer/object) – IN: Device pointer to the first matrix A_1 on the GPU
- lda (
int) – IN: [int] specifies the leading dimension of each matrix A_i
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- x (
Pointer/object) – IN: Device pointer to the first vector x_1 on the GPU
- incx (
int) – IN: [int] specifies the increment for the elements of each vector x_i
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1). There are no restrictions placed on stridex, however the user should take care to ensure that stridex is of appropriate size. This typically means stridex >= n * incx. stridex should be non zero.
- beta (
Pointer/object) – IN: device pointer or host pointer to scalar beta
- y (
Pointer/object) – OUT: Device pointer to the first vector y_1 on the GPU
- incy (
int) – IN: [int] specifies the increment for the elements of each vector y_i
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) and the next one (y_i+1). There are no restrictions placed on stridey, however the user should take care to ensure that stridey is of appropriate size. This typically means stridey >= n * incy. stridey should be non zero.
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDsymvStridedBatched(handle, uplo, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsymvStridedBatched(handle, uplo, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsymvStridedBatched(handle, uplo, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsymvStridedBatched_v2(handle, uplo, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
float2/object): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsymvStridedBatched_v2(handle, uplo, int n, alpha, AP, int lda, long strideA, x, int incx, long stridex, beta, y, int incy, long stridey, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- beta (
double2/object): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsyr(handle, uplo, int n, alpha, x, int incx, AP, int lda)#
BLAS Level 2 API
@{
syr performs the matrix-vector operations
A := A + alpha*x*x**T
where alpha is a scalar, x is a vector, and A is an n by n symmetric matrix.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- n (
int) – IN: [int] the number of rows and columns of matrix A.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- AP (
Pointer/object) – INOUT: device pointer storing matrix A.
- lda (
int) – IN: [int] specifies the leading dimension of A.
- handle (
- hip.hipblas.hipblasDsyr(handle, uplo, int n, alpha, x, int incx, AP, int lda)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCsyr(handle, uplo, int n, alpha, x, int incx, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyr(handle, uplo, int n, alpha, x, int incx, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyr_v2(handle, uplo, int n, alpha, x, int incx, AP, int lda)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZsyr_v2(handle, uplo, int n, alpha, x, int incx, AP, int lda)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSsyrBatched(handle, uplo, int n, alpha, x, int incx, AP, int lda, int batchCount)#
BLAS Level 2 API
@{
syrBatched performs a batch of matrix-vector operations
A[i] := A[i] + alpha*x[i]*x[i]**T
where alpha is a scalar, x is an array of vectors, and A is an array of n by n symmetric matrices, for i = 1 , … , batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- n (
int) – IN: [int] the number of rows and columns of matrix A.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- AP (
Pointer/object) – INOUT: device array of device pointers storing each matrix A_i.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i.
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDsyrBatched(handle, uplo, int n, alpha, x, int incx, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- hip.hipblas.hipblasCsyrBatched(handle, uplo, int n, alpha, x, int incx, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- hip.hipblas.hipblasZsyrBatched(handle, uplo, int n, alpha, x, int incx, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyrBatched_v2(handle, uplo, int n, alpha, x, int incx, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZsyrBatched_v2(handle, uplo, int n, alpha, x, int incx, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSsyrStridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, AP, int lda, long strideA, int batchCount)#
BLAS Level 2 API
@{
syrStridedBatched performs the matrix-vector operations
A[i] := A[i] + alpha*x[i]*x[i]**T
where alpha is a scalar, vectors, and A is an array of n by n symmetric matrices, for i = 1 , … , batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- n (
int) – IN: [int] the number of rows and columns of each matrix A.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
Pointer/object) – IN: device pointer to the first vector x_1.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] specifies the pointer increment between vectors (x_i) and (x_i+1).
- AP (
Pointer/object) – INOUT: device pointer to the first matrix A_1.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i.
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDsyrStridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyrStridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyrStridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyrStridedBatched_v2(handle, uplo, int n, alpha, x, int incx, long stridex, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyrStridedBatched_v2(handle, uplo, int n, alpha, x, int incx, long stridex, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsyr2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda)#
BLAS Level 2 API
@{
syr2 performs the matrix-vector operations
A := A + alpha*x*y**T + alpha*y*x**T
where alpha is a scalar, x and y are vectors, and A is an n by n symmetric matrix.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- n (
int) – IN: [int] the number of rows and columns of matrix A.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- y (
Pointer/object) – IN: device pointer storing vector y.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- AP (
Pointer/object) – INOUT: device pointer storing matrix A.
- lda (
int) – IN: [int] specifies the leading dimension of A.
- handle (
- hip.hipblas.hipblasDsyr2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyr2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyr2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyr2_v2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyr2_v2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsyr2Batched(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
BLAS Level 2 API
@{
syr2Batched performs a batch of matrix-vector operations
A[i] := A[i] + alpha*x[i]*y[i]**T + alpha*y[i]*x[i]**T
where alpha is a scalar, x[i] and y[i] are vectors, and A[i] is a n by n symmetric matrix, for i = 1 , … , batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- n (
int) – IN: [int] the number of rows and columns of matrix A.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- y (
ListOfPointer/object) – IN: device array of device pointers storing each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- AP (
Pointer/object) – INOUT: device array of device pointers storing each matrix A_i.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i.
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDsyr2Batched(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
ListOfPointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyr2Batched(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyr2Batched(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyr2Batched_v2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyr2Batched_v2(handle, uplo, int n, alpha, x, int incx, y, int incy, AP, int lda, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsyr2StridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
BLAS Level 2 API
@{
syr2StridedBatched the matrix-vector operations
A[i] := A[i] + alpha*x[i]*y[i]**T + alpha*y[i]*x[i]**T
where alpha is a scalar, x[i] and y[i] are vectors, and A[i] is a n by n symmetric matrices, for i = 1 , … , batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- n (
int) – IN: [int] the number of rows and columns of each matrix A.
- alpha (
Pointer/object) – IN: device pointer or host pointer to scalar alpha.
- x (
Pointer/object) – IN: device pointer to the first vector x_1.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] specifies the pointer increment between vectors (x_i) and (x_i+1).
- y (
Pointer/object) – IN: device pointer to the first vector y_1.
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- stridey (
int) – IN: [hipblasStride] specifies the pointer increment between vectors (y_i) and (y_i+1).
- AP (
Pointer/object) – INOUT: device pointer to the first matrix A_1.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i.
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDsyr2StridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyr2StridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyr2StridedBatched(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
hipblasDoubleComplex/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyr2StridedBatched_v2(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
float2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyr2StridedBatched_v2(handle, uplo, int n, alpha, x, int incx, long stridex, y, int incy, long stridey, AP, int lda, long strideA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
double2/object): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStbmv(handle, uplo, transA, diag, int m, int k, AP, int lda, x, int incx)#
BLAS Level 2 API
@{
tbmv performs one of the matrix-vector operations
- x := A*x or
x := A**T*x or x := A**H*x,
x is a vectors and A is a banded m by m matrix (see description below).
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A is an upper banded triangular matrix. HIPBLAS_FILL_MODE_LOWER: A is a lower banded triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] indicates whether matrix A is tranposed (conjugated) or not.
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: The main diagonal of A is assumed to consist of only
1’s and is not referenced.
HIPBLAS_DIAG_NON_UNIT: No assumptions are made of A’s main diagonal.
- m (
int) – IN: [int] the number of rows and columns of the matrix represented by A.
- k (
int) – IN: [int] if uplo == HIPBLAS_FILL_MODE_UPPER, k specifies the number of super-diagonals of the matrix A. if uplo == HIPBLAS_FILL_MODE_LOWER, k specifies the number of sub-diagonals of the matrix A. k must satisfy k > 0 && k < lda.
- AP (
Pointer/object) – IN: device pointer storing banded triangular matrix A. if uplo == HIPBLAS_FILL_MODE_UPPER:
The matrix represented is an upper banded triangular matrix with the main diagonal and k super-diagonals, everything else can be assumed to be 0. The matrix is compacted so that the main diagonal resides on the k’th row, the first super diagonal resides on the RHS of the k-1’th row, etc, with the k’th diagonal on the RHS of the 0’th row.
- Ex: (HIPBLAS_FILL_MODE_UPPER; m = 5; k = 2)
1 6 9 0 0 0 0 9 8 7 0 2 7 8 0 0 6 7 8 9 0 0 3 8 7 —-> 1 2 3 4 5 0 0 0 4 9 0 0 0 0 0 0 0 0 0 5 0 0 0 0 0
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The matrix represnted is a lower banded triangular matrix with the main diagonal and k sub-diagonals, everything else can be assumed to be 0. The matrix is compacted so that the main diagonal resides on the 0’th row, working up to the k’th diagonal residing on the LHS of the k’th row.
- Ex: (HIPBLAS_FILL_MODE_LOWER; m = 5; k = 2)
1 0 0 0 0 1 2 3 4 5 6 2 0 0 0 6 7 8 9 0 9 7 3 0 0 —-> 9 8 7 0 0 0 8 8 4 0 0 0 0 0 0 0 0 7 9 5 0 0 0 0 0
- lda (
int) – IN: [int] specifies the leading dimension of A. lda must satisfy lda > k.
- x (
Pointer/object) – INOUT: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- handle (
- hip.hipblas.hipblasDtbmv(handle, uplo, transA, diag, int m, int k, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- k (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtbmv(handle, uplo, transA, diag, int m, int k, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- k (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtbmv(handle, uplo, transA, diag, int m, int k, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- k (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtbmv_v2(handle, uplo, transA, diag, int m, int k, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- k (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtbmv_v2(handle, uplo, transA, diag, int m, int k, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- k (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStbmvBatched(handle, uplo, transA, diag, int m, int k, AP, int lda, x, int incx, int batchCount)#
BLAS Level 2 API
@{
tbmvBatched performs one of the matrix-vector operations
- x_i := A_i*x_i or
x_i := A_i**T*x_i or x_i := A_i**H*x_i,
where (A_i, x_i) is the i-th instance of the batch. x_i is a vector and A_i is an m by m matrix, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: each A_i is an upper banded triangular matrix. HIPBLAS_FILL_MODE_LOWER: each A_i is a lower banded triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] indicates whether each matrix A_i is tranposed (conjugated) or not.
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: The main diagonal of each A_i is assumed to consist of only
1’s and is not referenced.
HIPBLAS_DIAG_NON_UNIT: No assumptions are made of each A_i’s main diagonal.
- m (
int) – IN: [int] the number of rows and columns of the matrix represented by each A_i.
- k (
int) – IN: [int] if uplo == HIPBLAS_FILL_MODE_UPPER, k specifies the number of super-diagonals of each matrix A_i. if uplo == HIPBLAS_FILL_MODE_LOWER, k specifies the number of sub-diagonals of each matrix A_i. k must satisfy k > 0 && k < lda.
- AP (
Pointer/object) – IN: device array of device pointers storing each banded triangular matrix A_i. if uplo == HIPBLAS_FILL_MODE_UPPER:
The matrix represented is an upper banded triangular matrix with the main diagonal and k super-diagonals, everything else can be assumed to be 0. The matrix is compacted so that the main diagonal resides on the k’th row, the first super diagonal resides on the RHS of the k-1’th row, etc, with the k’th diagonal on the RHS of the 0’th row.
- Ex: (HIPBLAS_FILL_MODE_UPPER; m = 5; k = 2)
1 6 9 0 0 0 0 9 8 7 0 2 7 8 0 0 6 7 8 9 0 0 3 8 7 —-> 1 2 3 4 5 0 0 0 4 9 0 0 0 0 0 0 0 0 0 5 0 0 0 0 0
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The matrix represnted is a lower banded triangular matrix with the main diagonal and k sub-diagonals, everything else can be assumed to be 0. The matrix is compacted so that the main diagonal resides on the 0’th row, working up to the k’th diagonal residing on the LHS of the k’th row.
- Ex: (HIPBLAS_FILL_MODE_LOWER; m = 5; k = 2)
1 0 0 0 0 1 2 3 4 5 6 2 0 0 0 6 7 8 9 0 9 7 3 0 0 —-> 9 8 7 0 0 0 8 8 4 0 0 0 0 0 0 0 0 7 9 5 0 0 0 0 0
- lda (
int) – IN: [int] specifies the leading dimension of each A_i. lda must satisfy lda > k.
- x (
ListOfPointer/object) – INOUT: device array of device pointer storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDtbmvBatched(handle, uplo, transA, diag, int m, int k, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- k (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtbmvBatched(handle, uplo, transA, diag, int m, int k, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- k (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtbmvBatched(handle, uplo, transA, diag, int m, int k, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- k (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtbmvBatched_v2(handle, uplo, transA, diag, int m, int k, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- k (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtbmvBatched_v2(handle, uplo, transA, diag, int m, int k, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- k (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStbmvStridedBatched(handle, uplo, transA, diag, int m, int k, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
BLAS Level 2 API
@{
tbmvStridedBatched performs one of the matrix-vector operations
- x_i := A_i*x_i or
x_i := A_i**T*x_i or x_i := A_i**H*x_i,
where (A_i, x_i) is the i-th instance of the batch. x_i is a vector and A_i is an m by m matrix, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: each A_i is an upper banded triangular matrix. HIPBLAS_FILL_MODE_LOWER: each A_i is a lower banded triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] indicates whether each matrix A_i is tranposed (conjugated) or not.
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: The main diagonal of each A_i is assumed to consist of only
1’s and is not referenced.
HIPBLAS_DIAG_NON_UNIT: No assumptions are made of each A_i’s main diagonal.
- m (
int) – IN: [int] the number of rows and columns of the matrix represented by each A_i.
- k (
int) – IN: [int] if uplo == HIPBLAS_FILL_MODE_UPPER, k specifies the number of super-diagonals of each matrix A_i. if uplo == HIPBLAS_FILL_MODE_LOWER, k specifies the number of sub-diagonals of each matrix A_i. k must satisfy k > 0 && k < lda.
- AP (
Pointer/object) – IN: device array to the first matrix A_i of the batch. Stores each banded triangular matrix A_i. if uplo == HIPBLAS_FILL_MODE_UPPER:
The matrix represented is an upper banded triangular matrix with the main diagonal and k super-diagonals, everything else can be assumed to be 0. The matrix is compacted so that the main diagonal resides on the k’th row, the first super diagonal resides on the RHS of the k-1’th row, etc, with the k’th diagonal on the RHS of the 0’th row.
- Ex: (HIPBLAS_FILL_MODE_UPPER; m = 5; k = 2)
1 6 9 0 0 0 0 9 8 7 0 2 7 8 0 0 6 7 8 9 0 0 3 8 7 —-> 1 2 3 4 5 0 0 0 4 9 0 0 0 0 0 0 0 0 0 5 0 0 0 0 0
- if uplo == HIPBLAS_FILL_MODE_LOWER:
The matrix represnted is a lower banded triangular matrix with the main diagonal and k sub-diagonals, everything else can be assumed to be 0. The matrix is compacted so that the main diagonal resides on the 0’th row, working up to the k’th diagonal residing on the LHS of the k’th row.
- Ex: (HIPBLAS_FILL_MODE_LOWER; m = 5; k = 2)
1 0 0 0 0 1 2 3 4 5 6 2 0 0 0 6 7 8 9 0 9 7 3 0 0 —-> 9 8 7 0 0 0 8 8 4 0 0 0 0 0 0 0 0 7 9 5 0 0 0 0 0
- lda (
int) – IN: [int] specifies the leading dimension of each A_i. lda must satisfy lda > k.
- strideA (
int) – IN: [hipblasStride] stride from the start of one A_i matrix to the next A_(i + 1).
- x (
Pointer/object) – INOUT: device array to the first vector x_i of the batch.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one x_i matrix to the next x_(i + 1).
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDtbmvStridedBatched(handle, uplo, transA, diag, int m, int k, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- k (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtbmvStridedBatched(handle, uplo, transA, diag, int m, int k, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- k (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtbmvStridedBatched(handle, uplo, transA, diag, int m, int k, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- k (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtbmvStridedBatched_v2(handle, uplo, transA, diag, int m, int k, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- k (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtbmvStridedBatched_v2(handle, uplo, transA, diag, int m, int k, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- k (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStbsv(handle, uplo, transA, diag, int n, int k, AP, int lda, x, int incx)#
BLAS Level 2 API
@{
tbsv solves
A*x = b or A**T*x = b or A**H*x = b,
where x and b are vectors and A is a banded triangular matrix.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_N: Solves A*x = b HIPBLAS_OP_T: Solves A**T*x = b HIPBLAS_OP_C: Solves A**H*x = b
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: A is assumed to be unit triangular (i.e. the diagonal elements
of A are not used in computations).
HIPBLAS_DIAG_NON_UNIT: A is not assumed to be unit triangular.
- n (
int) – IN: [int] n specifies the number of rows of b. n >= 0.
- k (
int) – IN: [int] if(uplo == HIPBLAS_FILL_MODE_UPPER)
k specifies the number of super-diagonals of A.
- if(uplo == HIPBLAS_FILL_MODE_LOWER)
k specifies the number of sub-diagonals of A.
k >= 0.
- AP (
Pointer/object) – IN: device pointer storing the matrix A in banded format.
- lda (
int) – IN: [int] specifies the leading dimension of A. lda >= (k + 1).
- x (
Pointer/object) – INOUT: device pointer storing input vector b. Overwritten by the output vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- handle (
- hip.hipblas.hipblasDtbsv(handle, uplo, transA, diag, int n, int k, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtbsv(handle, uplo, transA, diag, int n, int k, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtbsv(handle, uplo, transA, diag, int n, int k, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtbsv_v2(handle, uplo, transA, diag, int n, int k, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtbsv_v2(handle, uplo, transA, diag, int n, int k, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStbsvBatched(handle, uplo, transA, diag, int n, int k, AP, int lda, x, int incx, int batchCount)#
BLAS Level 2 API
@{
tbsvBatched solves
A_i*x_i = b_i or A_i**T*x_i = b_i or A_i**H*x_i = b_i,
where x_i and b_i are vectors and A_i is a banded triangular matrix, for i = [1, batchCount].
The input vectors b_i are overwritten by the output vectors x_i.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A_i is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A_i is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_N: Solves A_i*x_i = b_i HIPBLAS_OP_T: Solves A_i**T*x_i = b_i HIPBLAS_OP_C: Solves A_i**H*x_i = b_i
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: each A_i is assumed to be unit triangular (i.e. the diagonal elements
of each A_i are not used in computations).
HIPBLAS_DIAG_NON_UNIT: each A_i is not assumed to be unit triangular.
- n (
int) – IN: [int] n specifies the number of rows of each b_i. n >= 0.
- k (
int) – IN: [int] if(uplo == HIPBLAS_FILL_MODE_UPPER)
k specifies the number of super-diagonals of each A_i.
- if(uplo == HIPBLAS_FILL_MODE_LOWER)
k specifies the number of sub-diagonals of each A_i.
k >= 0.
- AP (
Pointer/object) – IN: device vector of device pointers storing each matrix A_i in banded format.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i. lda >= (k + 1).
- x (
ListOfPointer/object) – INOUT: device vector of device pointers storing each input vector b_i. Overwritten by each output vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDtbsvBatched(handle, uplo, transA, diag, int n, int k, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtbsvBatched(handle, uplo, transA, diag, int n, int k, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtbsvBatched(handle, uplo, transA, diag, int n, int k, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtbsvBatched_v2(handle, uplo, transA, diag, int n, int k, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtbsvBatched_v2(handle, uplo, transA, diag, int n, int k, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStbsvStridedBatched(handle, uplo, transA, diag, int n, int k, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
BLAS Level 2 API
@{
tbsvStridedBatched solves
A_i*x_i = b_i or A_i**T*x_i = b_i or A_i**H*x_i = b_i,
where x_i and b_i are vectors and A_i is a banded triangular matrix, for i = [1, batchCount].
The input vectors b_i are overwritten by the output vectors x_i.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A_i is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A_i is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_N: Solves A_i*x_i = b_i HIPBLAS_OP_T: Solves A_i**T*x_i = b_i HIPBLAS_OP_C: Solves A_i**H*x_i = b_i
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: each A_i is assumed to be unit triangular (i.e. the diagonal elements
of each A_i are not used in computations).
HIPBLAS_DIAG_NON_UNIT: each A_i is not assumed to be unit triangular.
- n (
int) – IN: [int] n specifies the number of rows of each b_i. n >= 0.
- k (
int) – IN: [int] if(uplo == HIPBLAS_FILL_MODE_UPPER)
k specifies the number of super-diagonals of each A_i.
- if(uplo == HIPBLAS_FILL_MODE_LOWER)
k specifies the number of sub-diagonals of each A_i.
k >= 0.
- AP (
Pointer/object) – IN: device pointer pointing to the first banded matrix A_1.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i. lda >= (k + 1).
- strideA (
int) – IN: [hipblasStride] specifies the distance between the start of one matrix (A_i) and the next (A_i+1).
- x (
Pointer/object) – INOUT: device pointer pointing to the first input vector b_1. Overwritten by output vectors x.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] specifies the distance between the start of one vector (x_i) and the next (x_i+1).
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDtbsvStridedBatched(handle, uplo, transA, diag, int n, int k, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtbsvStridedBatched(handle, uplo, transA, diag, int n, int k, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtbsvStridedBatched(handle, uplo, transA, diag, int n, int k, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtbsvStridedBatched_v2(handle, uplo, transA, diag, int n, int k, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtbsvStridedBatched_v2(handle, uplo, transA, diag, int n, int k, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStpmv(handle, uplo, transA, diag, int m, AP, x, int incx)#
BLAS Level 2 API
@{
tpmv performs one of the matrix-vector operations
x = A*x or x = A**T*x,
where x is an n element vector and A is an n by n unit, or non-unit, upper or lower triangular matrix, supplied in the pack form.
The vector x is overwritten.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t]
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: A is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: A is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of A. m >= 0.
- AP (
Pointer/object) – IN: - device pointer storing matrix A,
of dimension at least ( m * ( m + 1 ) / 2 ).
Before entry with uplo = HIPBLAS_FILL_MODE_UPPER, the array A must contain the upper triangular matrix packed sequentially, column by column, so that A[0] contains a_{0,0}, A[1] and A[2] contain a_{0,1} and a_{1, 1} respectively, and so on. Before entry with uplo = HIPBLAS_FILL_MODE_LOWER, the array A must contain the lower triangular matrix packed sequentially, column by column, so that A[0] contains a_{0,0}, A[1] and A[2] contain a_{1,0} and a_{2,0} respectively, and so on. Note that when DIAG = HIPBLAS_DIAG_UNIT, the diagonal elements of A are not referenced, but are assumed to be unity.
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x. incx must not be zero.
- handle (
- hip.hipblas.hipblasDtpmv(handle, uplo, transA, diag, int m, AP, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtpmv(handle, uplo, transA, diag, int m, AP, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtpmv(handle, uplo, transA, diag, int m, AP, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtpmv_v2(handle, uplo, transA, diag, int m, AP, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
float2/object): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtpmv_v2(handle, uplo, transA, diag, int m, AP, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
double2/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStpmvBatched(handle, uplo, transA, diag, int m, AP, x, int incx, int batchCount)#
BLAS Level 2 API
@{
tpmvBatched performs one of the matrix-vector operations
x_i = A_i*x_i or x_i = A**T*x_i, 0 le i < batchCount
where x_i is an n element vector and A_i is an n by n (unit, or non-unit, upper or lower triangular matrix)
The vectors x_i are overwritten.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A_i is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A_i is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t]
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: A_i is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: A_i is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of matrices A_i. m >= 0.
- AP (
Pointer/object) – IN: device pointer storing pointer of matrices A_i, of dimension ( lda, m )
- x (
ListOfPointer/object) – IN: device pointer storing vectors x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of vectors x_i.
- batchCount (
int) – IN: [int] The number of batched matrices/vectors.
- handle (
- hip.hipblas.hipblasDtpmvBatched(handle, uplo, transA, diag, int m, AP, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtpmvBatched(handle, uplo, transA, diag, int m, AP, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtpmvBatched(handle, uplo, transA, diag, int m, AP, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtpmvBatched_v2(handle, uplo, transA, diag, int m, AP, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtpmvBatched_v2(handle, uplo, transA, diag, int m, AP, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStpmvStridedBatched(handle, uplo, transA, diag, int m, AP, long strideA, x, int incx, long stridex, int batchCount)#
BLAS Level 2 API
@{
tpmvStridedBatched performs one of the matrix-vector operations
x_i = A_i*x_i or x_i = A**T*x_i, 0 le i < batchCount
where x_i is an n element vector and A_i is an n by n (unit, or non-unit, upper or lower triangular matrix) with strides specifying how to retrieve $x_i$ (resp. $A_i$) from $x_{i-1}$ (resp. $A_i$).
The vectors x_i are overwritten.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A_i is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A_i is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t]
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: A_i is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: A_i is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of matrices A_i. m >= 0.
- AP (
Pointer/object) – IN: device pointer of the matrix A_0, of dimension ( lda, m )
- strideA (
int) – IN: [hipblasStride] stride from the start of one A_i matrix to the next A_{i + 1}
- x (
Pointer/object) – IN: device pointer storing the vector x_0.
- incx (
int) – IN: [int] specifies the increment for the elements of one vector x.
- stridex (
int) – IN: [hipblasStride] stride from the start of one x_i vector to the next x_{i + 1}
- batchCount (
int) – IN: [int] The number of batched matrices/vectors.
- handle (
- hip.hipblas.hipblasDtpmvStridedBatched(handle, uplo, transA, diag, int m, AP, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- strideA (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtpmvStridedBatched(handle, uplo, transA, diag, int m, AP, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtpmvStridedBatched(handle, uplo, transA, diag, int m, AP, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtpmvStridedBatched_v2(handle, uplo, transA, diag, int m, AP, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
float2/object): (undocumented)
- strideA (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtpmvStridedBatched_v2(handle, uplo, transA, diag, int m, AP, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
double2/object): (undocumented)
- strideA (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStpsv(handle, uplo, transA, diag, int m, AP, x, int incx)#
BLAS Level 2 API
@{
tpsv solves
A*x = b or A**T*x = b, or A**H*x = b,
where x and b are vectors and A is a triangular matrix stored in the packed format.
The input vector b is overwritten by the output vector x.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_N: Solves A*x = b HIPBLAS_OP_T: Solves A**T*x = b HIPBLAS_OP_C: Solves A**H*x = b
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: A is assumed to be unit triangular (i.e. the diagonal elements
of A are not used in computations).
HIPBLAS_DIAG_NON_UNIT: A is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of b. m >= 0.
- AP (
Pointer/object) – IN: device pointer storing the packed version of matrix A, of dimension >= (n * (n + 1) / 2)
- x (
Pointer/object) – INOUT: device pointer storing vector b on input, overwritten by x on output.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- handle (
- hip.hipblas.hipblasDtpsv(handle, uplo, transA, diag, int m, AP, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtpsv(handle, uplo, transA, diag, int m, AP, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtpsv(handle, uplo, transA, diag, int m, AP, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtpsv_v2(handle, uplo, transA, diag, int m, AP, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
float2/object): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtpsv_v2(handle, uplo, transA, diag, int m, AP, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
double2/object): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStpsvBatched(handle, uplo, transA, diag, int m, AP, x, int incx, int batchCount)#
BLAS Level 2 API
@{
tpsvBatched solves
A_i*x_i = b_i or A_i**T*x_i = b_i, or A_i**H*x_i = b_i,
where x_i and b_i are vectors and A_i is a triangular matrix stored in the packed format, for i in [1, batchCount].
The input vectors b_i are overwritten by the output vectors x_i.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: each A_i is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: each A_i is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_N: Solves A*x = b HIPBLAS_OP_T: Solves A**T*x = b HIPBLAS_OP_C: Solves A**H*x = b
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: each A_i is assumed to be unit triangular (i.e. the diagonal elements
of each A_i are not used in computations).
HIPBLAS_DIAG_NON_UNIT: each A_i is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of each b_i. m >= 0.
- AP (
Pointer/object) – IN: device array of device pointers storing the packed versions of each matrix A_i, of dimension >= (n * (n + 1) / 2)
- x (
ListOfPointer/object) – INOUT: device array of device pointers storing each input vector b_i, overwritten by x_i on output.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- batchCount (
int) – IN: [int] specifies the number of instances in the batch.
- handle (
- hip.hipblas.hipblasDtpsvBatched(handle, uplo, transA, diag, int m, AP, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtpsvBatched(handle, uplo, transA, diag, int m, AP, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtpsvBatched(handle, uplo, transA, diag, int m, AP, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtpsvBatched_v2(handle, uplo, transA, diag, int m, AP, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtpsvBatched_v2(handle, uplo, transA, diag, int m, AP, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStpsvStridedBatched(handle, uplo, transA, diag, int m, AP, long strideA, x, int incx, long stridex, int batchCount)#
BLAS Level 2 API
@{
tpsvStridedBatched solves
A_i*x_i = b_i or A_i**T*x_i = b_i, or A_i**H*x_i = b_i,
where x_i and b_i are vectors and A_i is a triangular matrix stored in the packed format, for i in [1, batchCount].
The input vectors b_i are overwritten by the output vectors x_i.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: each A_i is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: each A_i is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_N: Solves A*x = b HIPBLAS_OP_T: Solves A**T*x = b HIPBLAS_OP_C: Solves A**H*x = b
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: each A_i is assumed to be unit triangular (i.e. the diagonal elements
of each A_i are not used in computations).
HIPBLAS_DIAG_NON_UNIT: each A_i is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of each b_i. m >= 0.
- AP (
Pointer/object) – IN: device pointer pointing to the first packed matrix A_1, of dimension >= (n * (n + 1) / 2)
- strideA (
int) – IN: [hipblasStride] stride from the beginning of one packed matrix (AP_i) and the next (AP_i+1).
- x (
Pointer/object) – INOUT: device pointer pointing to the first input vector b_1. Overwritten by each x_i on output.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] stride from the beginning of one vector (x_i) and the next (x_i+1).
- batchCount (
int) – IN: [int] specifies the number of instances in the batch.
- handle (
- hip.hipblas.hipblasDtpsvStridedBatched(handle, uplo, transA, diag, int m, AP, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- strideA (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtpsvStridedBatched(handle, uplo, transA, diag, int m, AP, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtpsvStridedBatched(handle, uplo, transA, diag, int m, AP, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtpsvStridedBatched_v2(handle, uplo, transA, diag, int m, AP, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
float2/object): (undocumented)
- strideA (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtpsvStridedBatched_v2(handle, uplo, transA, diag, int m, AP, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
double2/object): (undocumented)
- strideA (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStrmv(handle, uplo, transA, diag, int m, AP, int lda, x, int incx)#
BLAS Level 2 API
@{
trmv performs one of the matrix-vector operations
x = A*x or x = A**T*x,
where x is an n element vector and A is an n by n unit, or non-unit, upper or lower triangular matrix.
The vector x is overwritten.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t]
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: A is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: A is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of A. m >= 0.
- AP (
Pointer/object) – IN: device pointer storing matrix A, of dimension ( lda, m )
- lda (
int) – IN: [int] specifies the leading dimension of A. lda = max( 1, m ).
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- handle (
- hip.hipblas.hipblasDtrmv(handle, uplo, transA, diag, int m, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrmv(handle, uplo, transA, diag, int m, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrmv(handle, uplo, transA, diag, int m, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrmv_v2(handle, uplo, transA, diag, int m, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrmv_v2(handle, uplo, transA, diag, int m, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStrmvBatched(handle, uplo, transA, diag, int m, AP, int lda, x, int incx, int batchCount)#
BLAS Level 2 API
@{
trmvBatched performs one of the matrix-vector operations
x_i = A_i*x_i or x_i = A**T*x_i, 0 le i < batchCount
where x_i is an n element vector and A_i is an n by n (unit, or non-unit, upper or lower triangular matrix)
The vectors x_i are overwritten.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A_i is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A_i is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t]
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: A_i is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: A_i is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of matrices A_i. m >= 0.
- AP (
Pointer/object) – IN: device pointer storing pointer of matrices A_i, of dimension ( lda, m )
- lda (
int) – IN: [int] specifies the leading dimension of A_i. lda >= max( 1, m ).
- x (
ListOfPointer/object) – IN: device pointer storing vectors x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of vectors x_i.
- batchCount (
int) – IN: [int] The number of batched matrices/vectors.
- handle (
- hip.hipblas.hipblasDtrmvBatched(handle, uplo, transA, diag, int m, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrmvBatched(handle, uplo, transA, diag, int m, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrmvBatched(handle, uplo, transA, diag, int m, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrmvBatched_v2(handle, uplo, transA, diag, int m, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrmvBatched_v2(handle, uplo, transA, diag, int m, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStrmvStridedBatched(handle, uplo, transA, diag, int m, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
BLAS Level 2 API
@{
trmvStridedBatched performs one of the matrix-vector operations
x_i = A_i*x_i or x_i = A**T*x_i, 0 le i < batchCount
where x_i is an n element vector and A_i is an n by n (unit, or non-unit, upper or lower triangular matrix) with strides specifying how to retrieve $x_i$ (resp. $A_i$) from $x_{i-1}$ (resp. $A_i$).
The vectors x_i are overwritten.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A_i is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A_i is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t]
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: A_i is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: A_i is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of matrices A_i. m >= 0.
- AP (
Pointer/object) – IN: device pointer of the matrix A_0, of dimension ( lda, m )
- lda (
int) – IN: [int] specifies the leading dimension of A_i. lda >= max( 1, m ).
- strideA (
int) – IN: [hipblasStride] stride from the start of one A_i matrix to the next A_{i + 1}
- x (
Pointer/object) – IN: device pointer storing the vector x_0.
- incx (
int) – IN: [int] specifies the increment for the elements of one vector x.
- stridex (
int) – IN: [hipblasStride] stride from the start of one x_i vector to the next x_{i + 1}
- batchCount (
int) – IN: [int] The number of batched matrices/vectors.
- handle (
- hip.hipblas.hipblasDtrmvStridedBatched(handle, uplo, transA, diag, int m, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrmvStridedBatched(handle, uplo, transA, diag, int m, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrmvStridedBatched(handle, uplo, transA, diag, int m, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrmvStridedBatched_v2(handle, uplo, transA, diag, int m, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrmvStridedBatched_v2(handle, uplo, transA, diag, int m, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStrsv(handle, uplo, transA, diag, int m, AP, int lda, x, int incx)#
BLAS Level 2 API
@{
trsv solves
A*x = b or A**T*x = b,
where x and b are vectors and A is a triangular matrix.
The vector x is overwritten on b.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t]
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: A is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: A is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of b. m >= 0.
- AP (
Pointer/object) – IN: device pointer storing matrix A, of dimension ( lda, m )
- lda (
int) – IN: [int] specifies the leading dimension of A. lda = max( 1, m ).
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- handle (
- hip.hipblas.hipblasDtrsv(handle, uplo, transA, diag, int m, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrsv(handle, uplo, transA, diag, int m, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrsv(handle, uplo, transA, diag, int m, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrsv_v2(handle, uplo, transA, diag, int m, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrsv_v2(handle, uplo, transA, diag, int m, AP, int lda, x, int incx)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStrsvBatched(handle, uplo, transA, diag, int m, AP, int lda, x, int incx, int batchCount)#
BLAS Level 2 API
@{
trsvBatched solves
A_i*x_i = b_i or A_i**T*x_i = b_i,
where (A_i, x_i, b_i) is the i-th instance of the batch. x_i and b_i are vectors and A_i is an m by m triangular matrix.
The vector x is overwritten on b.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t]
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: A is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: A is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of b. m >= 0.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix A_i.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i. lda = max(1, m)
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDtrsvBatched(handle, uplo, transA, diag, int m, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrsvBatched(handle, uplo, transA, diag, int m, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrsvBatched(handle, uplo, transA, diag, int m, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrsvBatched_v2(handle, uplo, transA, diag, int m, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrsvBatched_v2(handle, uplo, transA, diag, int m, AP, int lda, x, int incx, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStrsvStridedBatched(handle, uplo, transA, diag, int m, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
BLAS Level 2 API
@{
trsvStridedBatched solves
A_i*x_i = b_i or A_i**T*x_i = b_i,
where (A_i, x_i, b_i) is the i-th instance of the batch. x_i and b_i are vectors and A_i is an m by m triangular matrix, for i = 1, …, batchCount.
The vector x is overwritten on b.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t]
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: A is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: A is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of each b_i. m >= 0.
- AP (
Pointer/object) – IN: device pointer to the first matrix (A_1) in the batch, of dimension ( lda, m )
- lda (
int) – IN: [int] specifies the leading dimension of each A_i. lda = max( 1, m ).
- strideA (
int) – IN: [hipblasStride] stride from the start of one A_i matrix to the next A_(i + 1)
- x (
Pointer/object) – IN,OUT: device pointer to the first vector (x_1) in the batch.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one x_i vector to the next x_(i + 1)
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasDtrsvStridedBatched(handle, uplo, transA, diag, int m, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrsvStridedBatched(handle, uplo, transA, diag, int m, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrsvStridedBatched(handle, uplo, transA, diag, int m, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrsvStridedBatched_v2(handle, uplo, transA, diag, int m, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrsvStridedBatched_v2(handle, uplo, transA, diag, int m, AP, int lda, long strideA, x, int incx, long stridex, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasHgemm(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
BLAS Level 3 API
@{
gemm performs one of the matrix-matrix operations
C = alpha*op( A )*op( B ) + beta*C,
where op( X ) is one of
op( X ) = X or op( X ) = X**T or op( X ) = X**H,
alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C an m by n matrix.
Supported precisions in rocBLAS : h,s,d,c,z
Supported precisions in cuBLAS : h,s,d,c,z
.
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t]
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( A )
- transB (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( B )
- m (
int) – IN: [int] number or rows of matrices op( A ) and C
- n (
int) – IN: [int] number of columns of matrices op( B ) and C
- k (
int) – IN: [int] number of columns of matrix op( A ) and number of rows of matrix op( B )
- alpha (
Pointer/object) – IN: device pointer or host pointer specifying the scalar alpha.
- AP (
Pointer/object) – IN: device pointer storing matrix A.
- lda (
int) – IN: [int] specifies the leading dimension of A.
- BP (
Pointer/object) – IN: device pointer storing matrix B.
- ldb (
int) – IN: [int] specifies the leading dimension of B.
- beta (
Pointer/object) – IN: device pointer or host pointer specifying the scalar beta.
- CP (
Pointer/object) – IN,OUT: device pointer storing matrix C on the GPU.
- ldc (
int) – IN: [int] specifies the leading dimension of C.
- handle (
- hip.hipblas.hipblasSgemm(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasDgemm(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgemm(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- BP (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- CP (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgemm(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgemm_v2(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgemm_v2(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasHgemmBatched(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
BLAS Level 3 API
@{
- gemmBatched performs one of the batched matrix-matrix operations
C_i = alpha*op( A_i )*op( B_i ) + beta*C_i, for i = 1, …, batchCount.
- where op( X ) is one of
op( X ) = X or
op( X ) = X**T or op( X ) = X**H,
alpha and beta are scalars, and A, B and C are strided batched matrices, with
op( A ) an m by k by batchCount strided_batched matrix, op( B ) an k by n by batchCount strided_batched matrix and C an m by n by batchCount strided_batched matrix.
Supported precisions in rocBLAS : h,s,d,c,z
Supported precisions in cuBLAS : h,s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( A )
- transB (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( B )
- m (
int) – IN: [int] matrix dimension m.
- n (
int) – IN: [int] matrix dimension n.
- k (
int) – IN: [int] matrix dimension k.
- alpha (
Pointer/object) – IN: device pointer or host pointer specifying the scalar alpha.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix A_i.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i.
- BP (
Pointer/object) – IN: device array of device pointers storing each matrix B_i.
- ldb (
int) – IN: [int] specifies the leading dimension of each B_i.
- beta (
Pointer/object) – IN: device pointer or host pointer specifying the scalar beta.
- CP (
Pointer/object) – IN,OUT: device array of device pointers storing each matrix C_i.
- ldc (
int) – IN: [int] specifies the leading dimension of each C_i.
- batchCount (
int) – IN: [int] number of gemm operations in the batch
- handle (
- hip.hipblas.hipblasSgemmBatched(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasDgemmBatched(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgemmBatched(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgemmBatched(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgemmBatched_v2(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgemmBatched_v2(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasHgemmStridedBatched(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, long long strideA, BP, int ldb, long long strideB, beta, CP, int ldc, long long strideC, int batchCount)#
BLAS Level 3 API
@{
gemmStridedBatched performs one of the strided batched matrix-matrix operations
C_i = alpha*op( A_i )*op( B_i ) + beta*C_i, for i = 1, …, batchCount.
where op( X ) is one of
op( X ) = X or op( X ) = X**T or op( X ) = X**H,
alpha and beta are scalars, and A, B and C are strided batched matrices, with op( A ) an m by k by batchCount strided_batched matrix, op( B ) an k by n by batchCount strided_batched matrix and C an m by n by batchCount strided_batched matrix.
Supported precisions in rocBLAS : h,s,d,c,z
Supported precisions in cuBLAS : h,s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( A )
- transB (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( B )
- m (
int) – IN: [int] matrix dimension m.
- n (
int) – IN: [int] matrix dimension n.
- k (
int) – IN: [int] matrix dimension k.
- alpha (
Pointer/object) – IN: device pointer or host pointer specifying the scalar alpha.
- AP (
Pointer/object) – IN: device pointer pointing to the first matrix A_1.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i.
- strideA (
int) – IN: [hipblasStride] stride from the start of one A_i matrix to the next A_(i + 1).
- BP (
Pointer/object) – IN: device pointer pointing to the first matrix B_1.
- ldb (
int) – IN: [int] specifies the leading dimension of each B_i.
- strideB (
int) – IN: [hipblasStride] stride from the start of one B_i matrix to the next B_(i + 1).
- beta (
Pointer/object) – IN: device pointer or host pointer specifying the scalar beta.
- CP (
Pointer/object) – IN,OUT: device pointer pointing to the first matrix C_1.
- ldc (
int) – IN: [int] specifies the leading dimension of each C_i.
- strideC (
int) – IN: [hipblasStride] stride from the start of one C_i matrix to the next C_(i + 1).
- batchCount (
int) – IN: [int] number of gemm operatons in the batch
- handle (
- hip.hipblas.hipblasSgemmStridedBatched(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, long long strideA, BP, int ldb, long long strideB, beta, CP, int ldc, long long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasDgemmStridedBatched(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, long long strideA, BP, int ldb, long long strideB, beta, CP, int ldc, long long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgemmStridedBatched(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, long long strideA, BP, int ldb, long long strideB, beta, CP, int ldc, long long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- CP (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgemmStridedBatched(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, long long strideA, BP, int ldb, long long strideB, beta, CP, int ldc, long long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgemmStridedBatched_v2(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, long long strideA, BP, int ldb, long long strideB, beta, CP, int ldc, long long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgemmStridedBatched_v2(handle, transA, transB, int m, int n, int k, alpha, AP, int lda, long long strideA, BP, int ldb, long long strideB, beta, CP, int ldc, long long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCherk(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc)#
BLAS Level 3 API
@{
herk performs one of the matrix-matrix operations for a Hermitian rank-k update
C := alpha*op( A )*op( A )^H + beta*C
where alpha and beta are scalars, op(A) is an n by k matrix, and C is a n x n Hermitian matrix stored as either upper or lower.
op( A ) = A, and A is n by k if transA == HIPBLAS_OP_N op( A ) = A^H and A is k by n if transA == HIPBLAS_OP_C
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_C: op(A) = A^H HIPBLAS_ON_N: op(A) = A
- n (
int) – IN: [int] n specifies the number of rows and columns of C. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
hipblasComplex/object) – IN: pointer storing matrix A on the GPU. Martrix dimension is ( lda, k ) when if transA = HIPBLAS_OP_N, otherwise (lda, n) only the upper/lower triangular part is accessed.
- lda (
int) – IN: [int] lda specifies the first dimension of A. if transA = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
hipblasComplex/object) – IN: pointer storing matrix C on the GPU. The imaginary component of the diagonal elements are not used but are set to zero unless quick return.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- handle (
- hip.hipblas.hipblasZherk(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCherk_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZherk_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCherkBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc, int batchCount)#
BLAS Level 3 API
@{
herkBatched performs a batch of the matrix-matrix operations for a Hermitian rank-k update
C_i := alpha*op( A_i )*op( A_i )^H + beta*C_i
where alpha and beta are scalars, op(A) is an n by k matrix, and C_i is a n x n Hermitian matrix stored as either upper or lower.
op( A_i ) = A_i, and A_i is n by k if transA == HIPBLAS_OP_N op( A_i ) = A_i^H and A_i is k by n if transA == HIPBLAS_OP_C
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C_i is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_C: op(A) = A^H HIPBLAS_OP_N: op(A) = A
- n (
int) – IN: [int] n specifies the number of rows and columns of C_i. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix_i A of dimension (lda, k) when transA is HIPBLAS_OP_N, otherwise of dimension (lda, n)
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if transA = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
Pointer/object) – IN: device array of device pointers storing each matrix C_i on the GPU. The imaginary component of the diagonal elements are not used but are set to zero unless quick return.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZherkBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCherkBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZherkBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCherkStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, beta, CP, int ldc, long strideC, int batchCount)#
BLAS Level 3 API
@{
herkStridedBatched performs a batch of the matrix-matrix operations for a Hermitian rank-k update
C_i := alpha*op( A_i )*op( A_i )^H + beta*C_i
where alpha and beta are scalars, op(A) is an n by k matrix, and C_i is a n x n Hermitian matrix stored as either upper or lower.
op( A_i ) = A_i, and A_i is n by k if transA == HIPBLAS_OP_N op( A_i ) = A_i^H and A_i is k by n if transA == HIPBLAS_OP_C
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C_i is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_C: op(A) = A^H HIPBLAS_OP_N: op(A) = A
- n (
int) – IN: [int] n specifies the number of rows and columns of C_i. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
hipblasComplex/object) – IN: Device pointer to the first matrix A_1 on the GPU of dimension (lda, k) when transA is HIPBLAS_OP_N, otherwise of dimension (lda, n)
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if transA = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
hipblasComplex/object) – IN: Device pointer to the first matrix C_1 on the GPU. The imaginary component of the diagonal elements are not used but are set to zero unless quick return.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- strideC (
int) – INOUT: [hipblasStride] stride from the start of one matrix (C_i) and the next one (C_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZherkStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCherkStridedBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZherkStridedBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCherkx(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
BLAS Level 3 API
@{
herkx performs one of the matrix-matrix operations for a Hermitian rank-k update
C := alpha*op( A )*op( B )^H + beta*C
where alpha and beta are scalars, op(A) and op(B) are n by k matrices, and C is a n x n Hermitian matrix stored as either upper or lower. This routine should only be used when the caller can guarantee that the result of op( A )*op( B )^T will be Hermitian.
op( A ) = A, op( B ) = B, and A and B are n by k if trans == HIPBLAS_OP_N op( A ) = A^H, op( B ) = B^H, and A and B are k by n if trans == HIPBLAS_OP_C
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_C: op( A ) = A^H, op( B ) = B^H HIPBLAS_OP_N: op( A ) = A, op( B ) = B
- n (
int) – IN: [int] n specifies the number of rows and columns of C. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
hipblasComplex/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
hipblasComplex/object) – IN: pointer storing matrix A on the GPU. Martrix dimension is ( lda, k ) when if trans = HIPBLAS_OP_N, otherwise (lda, n) only the upper/lower triangular part is accessed.
- lda (
int) – IN: [int] lda specifies the first dimension of A. if trans = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- BP (
hipblasComplex/object) – IN: pointer storing matrix B on the GPU. Martrix dimension is ( ldb, k ) when if trans = HIPBLAS_OP_N, otherwise (ldb, n) only the upper/lower triangular part is accessed.
- ldb (
int) – IN: [int] ldb specifies the first dimension of B. if trans = HIPBLAS_OP_N, ldb >= max( 1, n ), otherwise ldb >= max( 1, k ).
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
hipblasComplex/object) – IN: pointer storing matrix C on the GPU. The imaginary component of the diagonal elements are not used but are set to zero unless quick return.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- handle (
- hip.hipblas.hipblasZherkx(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCherkx_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZherkx_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCherkxBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
BLAS Level 3 API
@{
herkxBatched performs a batch of the matrix-matrix operations for a Hermitian rank-k update
C_i := alpha*op( A_i )*op( B_i )^H + beta*C_i
where alpha and beta are scalars, op(A_i) and op(B_i) are n by k matrices, and C_i is a n x n Hermitian matrix stored as either upper or lower. This routine should only be used when the caller can guarantee that the result of op( A )*op( B )^T will be Hermitian.
op( A_i ) = A_i, op( B_i ) = B_i, and A_i and B_i are n by k if trans == HIPBLAS_OP_N op( A_i ) = A_i^H, op( B_i ) = B_i^H, and A_i and B_i are k by n if trans == HIPBLAS_OP_C
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C_i is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_C: op(A) = A^H HIPBLAS_OP_N: op(A) = A
- n (
int) – IN: [int] n specifies the number of rows and columns of C_i. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
hipblasComplex/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix_i A of dimension (lda, k) when trans is HIPBLAS_OP_N, otherwise of dimension (lda, n)
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if trans = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- BP (
Pointer/object) – IN: device array of device pointers storing each matrix_i B of dimension (ldb, k) when trans is HIPBLAS_OP_N, otherwise of dimension (ldb, n)
- ldb (
int) – IN: [int] ldb specifies the first dimension of B_i. if trans = HIPBLAS_OP_N, ldb >= max( 1, n ), otherwise ldb >= max( 1, k ).
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
Pointer/object) – IN: device array of device pointers storing each matrix C_i on the GPU. The imaginary component of the diagonal elements are not used but are set to zero unless quick return.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZherkxBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCherkxBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZherkxBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCherkxStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
BLAS Level 3 API
@{
herkxStridedBatched performs a batch of the matrix-matrix operations for a Hermitian rank-k update
C_i := alpha*op( A_i )*op( B_i )^H + beta*C_i
where alpha and beta are scalars, op(A_i) and op(B_i) are n by k matrices, and C_i is a n x n Hermitian matrix stored as either upper or lower. This routine should only be used when the caller can guarantee that the result of op( A )*op( B )^T will be Hermitian.
op( A_i ) = A_i, op( B_i ) = B_i, and A_i and B_i are n by k if trans == HIPBLAS_OP_N op( A_i ) = A_i^H, op( B_i ) = B_i^H, and A_i and B_i are k by n if trans == HIPBLAS_OP_C
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C_i is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_C: op( A_i ) = A_i^H, op( B_i ) = B_i^H HIPBLAS_OP_N: op( A_i ) = A_i, op( B_i ) = B_i
- n (
int) – IN: [int] n specifies the number of rows and columns of C_i. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
hipblasComplex/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
hipblasComplex/object) – IN: Device pointer to the first matrix A_1 on the GPU of dimension (lda, k) when trans is HIPBLAS_OP_N, otherwise of dimension (lda, n)
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if trans = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- BP (
hipblasComplex/object) – IN: Device pointer to the first matrix B_1 on the GPU of dimension (ldb, k) when trans is HIPBLAS_OP_N, otherwise of dimension (ldb, n)
- ldb (
int) – IN: [int] ldb specifies the first dimension of B_i. if trans = HIPBLAS_OP_N, ldb >= max( 1, n ), otherwise ldb >= max( 1, k ).
- strideB (
int) – IN: [hipblasStride] stride from the start of one matrix (B_i) and the next one (B_i+1)
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
hipblasComplex/object) – IN: Device pointer to the first matrix C_1 on the GPU. The imaginary component of the diagonal elements are not used but are set to zero unless quick return.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- strideC (
int) – INOUT: [hipblasStride] stride from the start of one matrix (C_i) and the next one (C_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZherkxStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCherkxStridedBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZherkxStridedBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCher2k(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
BLAS Level 3 API
@{
her2k performs one of the matrix-matrix operations for a Hermitian rank-2k update
C := alpha*op( A )*op( B )^H + conj(alpha)*op( B )*op( A )^H + beta*C
where alpha and beta are scalars, op(A) and op(B) are n by k matrices, and C is a n x n Hermitian matrix stored as either upper or lower.
op( A ) = A, op( B ) = B, and A and B are n by k if trans == HIPBLAS_OP_N op( A ) = A^H, op( B ) = B^H, and A and B are k by n if trans == HIPBLAS_OP_C
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_C: op( A ) = A^H, op( B ) = B^H HIPBLAS_OP_N: op( A ) = A, op( B ) = B
- n (
int) – IN: [int] n specifies the number of rows and columns of C. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
hipblasComplex/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
hipblasComplex/object) – IN: pointer storing matrix A on the GPU. Martrix dimension is ( lda, k ) when if trans = HIPBLAS_OP_N, otherwise (lda, n) only the upper/lower triangular part is accessed.
- lda (
int) – IN: [int] lda specifies the first dimension of A. if trans = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- BP (
hipblasComplex/object) – IN: pointer storing matrix B on the GPU. Martrix dimension is ( ldb, k ) when if trans = HIPBLAS_OP_N, otherwise (ldb, n) only the upper/lower triangular part is accessed.
- ldb (
int) – IN: [int] ldb specifies the first dimension of B. if trans = HIPBLAS_OP_N, ldb >= max( 1, n ), otherwise ldb >= max( 1, k ).
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
hipblasComplex/object) – IN: pointer storing matrix C on the GPU. The imaginary component of the diagonal elements are not used but are set to zero unless quick return.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- handle (
- hip.hipblas.hipblasZher2k(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCher2k_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZher2k_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCher2kBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
BLAS Level 3 API
@{
her2kBatched performs a batch of the matrix-matrix operations for a Hermitian rank-2k update
C_i := alpha*op( A_i )*op( B_i )^H + conj(alpha)*op( B_i )*op( A_i )^H + beta*C_i
where alpha and beta are scalars, op(A_i) and op(B_i) are n by k matrices, and C_i is a n x n Hermitian matrix stored as either upper or lower.
op( A_i ) = A_i, op( B_i ) = B_i, and A_i and B_i are n by k if trans == HIPBLAS_OP_N op( A_i ) = A_i^H, op( B_i ) = B_i^H, and A_i and B_i are k by n if trans == HIPBLAS_OP_C
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C_i is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_C: op(A) = A^H HIPBLAS_OP_N: op(A) = A
- n (
int) – IN: [int] n specifies the number of rows and columns of C_i. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
hipblasComplex/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix_i A of dimension (lda, k) when trans is HIPBLAS_OP_N, otherwise of dimension (lda, n)
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if trans = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- BP (
Pointer/object) – IN: device array of device pointers storing each matrix_i B of dimension (ldb, k) when trans is HIPBLAS_OP_N, otherwise of dimension (ldb, n)
- ldb (
int) – IN: [int] ldb specifies the first dimension of B_i. if trans = HIPBLAS_OP_N, ldb >= max( 1, n ), otherwise ldb >= max( 1, k ).
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
Pointer/object) – IN: device array of device pointers storing each matrix C_i on the GPU. The imaginary component of the diagonal elements are not used but are set to zero unless quick return.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZher2kBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCher2kBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZher2kBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCher2kStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
BLAS Level 3 API
@{
her2kStridedBatched performs a batch of the matrix-matrix operations for a Hermitian rank-2k update
C_i := alpha*op( A_i )*op( B_i )^H + conj(alpha)*op( B_i )*op( A_i )^H + beta*C_i
where alpha and beta are scalars, op(A_i) and op(B_i) are n by k matrices, and C_i is a n x n Hermitian matrix stored as either upper or lower.
op( A_i ) = A_i, op( B_i ) = B_i, and A_i and B_i are n by k if trans == HIPBLAS_OP_N op( A_i ) = A_i^H, op( B_i ) = B_i^H, and A_i and B_i are k by n if trans == HIPBLAS_OP_C
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C_i is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_C: op( A_i ) = A_i^H, op( B_i ) = B_i^H HIPBLAS_OP_N: op( A_i ) = A_i, op( B_i ) = B_i
- n (
int) – IN: [int] n specifies the number of rows and columns of C_i. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
hipblasComplex/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
hipblasComplex/object) – IN: Device pointer to the first matrix A_1 on the GPU of dimension (lda, k) when trans is HIPBLAS_OP_N, otherwise of dimension (lda, n)
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if trans = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- BP (
hipblasComplex/object) – IN: Device pointer to the first matrix B_1 on the GPU of dimension (ldb, k) when trans is HIPBLAS_OP_N, otherwise of dimension (ldb, n)
- ldb (
int) – IN: [int] ldb specifies the first dimension of B_i. if trans = HIPBLAS_OP_N, ldb >= max( 1, n ), otherwise ldb >= max( 1, k ).
- strideB (
int) – IN: [hipblasStride] stride from the start of one matrix (B_i) and the next one (B_i+1)
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
hipblasComplex/object) – IN: Device pointer to the first matrix C_1 on the GPU. The imaginary component of the diagonal elements are not used but are set to zero unless quick return.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- strideC (
int) – INOUT: [hipblasStride] stride from the start of one matrix (C_i) and the next one (C_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZher2kStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCher2kStridedBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZher2kStridedBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsymm(handle, side, uplo, int m, int n, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
BLAS Level 3 API
@{
symm performs one of the matrix-matrix operations:
C := alpha*A*B + beta*C if side == HIPBLAS_SIDE_LEFT, C := alpha*B*A + beta*C if side == HIPBLAS_SIDE_RIGHT,
where alpha and beta are scalars, B and C are m by n matrices, and A is a symmetric matrix stored as either upper or lower.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] HIPBLAS_SIDE_LEFT: C := alpha*A*B + beta*C HIPBLAS_SIDE_RIGHT: C := alpha*B*A + beta*C
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: A is a lower triangular matrix
- m (
int) – IN: [int] m specifies the number of rows of B and C. m >= 0.
- n (
int) – IN: [int] n specifies the number of columns of B and C. n >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A and B are not referenced.
- AP (
Pointer/object) – IN: pointer storing matrix A on the GPU. A is m by m if side == HIPBLAS_SIDE_LEFT A is n by n if side == HIPBLAS_SIDE_RIGHT only the upper/lower triangular part is accessed.
- lda (
int) – IN: [int] lda specifies the first dimension of A. if side = HIPBLAS_SIDE_LEFT, lda >= max( 1, m ), otherwise lda >= max( 1, n ).
- BP (
Pointer/object) – IN: pointer storing matrix B on the GPU. Matrix dimension is m by n
- ldb (
int) – IN: [int] ldb specifies the first dimension of B. ldb >= max( 1, m )
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
Pointer/object) – IN: pointer storing matrix C on the GPU. Matrix dimension is m by n
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, m )
- handle (
- hip.hipblas.hipblasDsymm(handle, side, uplo, int m, int n, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsymm(handle, side, uplo, int m, int n, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- BP (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- CP (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsymm(handle, side, uplo, int m, int n, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsymm_v2(handle, side, uplo, int m, int n, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsymm_v2(handle, side, uplo, int m, int n, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsymmBatched(handle, side, uplo, int m, int n, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
BLAS Level 3 API
@{
symmBatched performs a batch of the matrix-matrix operations:
C_i := alpha*A_i*B_i + beta*C_i if side == HIPBLAS_SIDE_LEFT, C_i := alpha*B_i*A_i + beta*C_i if side == HIPBLAS_SIDE_RIGHT,
where alpha and beta are scalars, B_i and C_i are m by n matrices, and A_i is a symmetric matrix stored as either upper or lower.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] HIPBLAS_SIDE_LEFT: C_i := alpha*A_i*B_i + beta*C_i HIPBLAS_SIDE_RIGHT: C_i := alpha*B_i*A_i + beta*C_i
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: A_i is a lower triangular matrix
- m (
int) – IN: [int] m specifies the number of rows of B_i and C_i. m >= 0.
- n (
int) – IN: [int] n specifies the number of columns of B_i and C_i. n >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A_i and B_i are not referenced.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix A_i on the GPU. A_i is m by m if side == HIPBLAS_SIDE_LEFT A_i is n by n if side == HIPBLAS_SIDE_RIGHT only the upper/lower triangular part is accessed.
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if side = HIPBLAS_SIDE_LEFT, lda >= max( 1, m ), otherwise lda >= max( 1, n ).
- BP (
Pointer/object) – IN: device array of device pointers storing each matrix B_i on the GPU. Matrix dimension is m by n
- ldb (
int) – IN: [int] ldb specifies the first dimension of B_i. ldb >= max( 1, m )
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C_i need not be set before entry.
- CP (
Pointer/object) – IN: device array of device pointers storing each matrix C_i on the GPU. Matrix dimension is m by n
- ldc (
int) – IN: [int] ldc specifies the first dimension of C_i. ldc >= max( 1, m )
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDsymmBatched(handle, side, uplo, int m, int n, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsymmBatched(handle, side, uplo, int m, int n, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsymmBatched(handle, side, uplo, int m, int n, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsymmBatched_v2(handle, side, uplo, int m, int n, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsymmBatched_v2(handle, side, uplo, int m, int n, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsymmStridedBatched(handle, side, uplo, int m, int n, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
BLAS Level 3 API
@{
symmStridedBatched performs a batch of the matrix-matrix operations:
C_i := alpha*A_i*B_i + beta*C_i if side == HIPBLAS_SIDE_LEFT, C_i := alpha*B_i*A_i + beta*C_i if side == HIPBLAS_SIDE_RIGHT,
where alpha and beta are scalars, B_i and C_i are m by n matrices, and A_i is a symmetric matrix stored as either upper or lower.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] HIPBLAS_SIDE_LEFT: C_i := alpha*A_i*B_i + beta*C_i HIPBLAS_SIDE_RIGHT: C_i := alpha*B_i*A_i + beta*C_i
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: A_i is a lower triangular matrix
- m (
int) – IN: [int] m specifies the number of rows of B_i and C_i. m >= 0.
- n (
int) – IN: [int] n specifies the number of columns of B_i and C_i. n >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A_i and B_i are not referenced.
- AP (
Pointer/object) – IN: device pointer to first matrix A_1 A_i is m by m if side == HIPBLAS_SIDE_LEFT A_i is n by n if side == HIPBLAS_SIDE_RIGHT only the upper/lower triangular part is accessed.
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if side = HIPBLAS_SIDE_LEFT, lda >= max( 1, m ), otherwise lda >= max( 1, n ).
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- BP (
Pointer/object) – IN: device pointer to first matrix B_1 of dimension (ldb, n) on the GPU.
- ldb (
int) – IN: [int] ldb specifies the first dimension of B_i. ldb >= max( 1, m )
- strideB (
int) – IN: [hipblasStride] stride from the start of one matrix (B_i) and the next one (B_i+1)
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
Pointer/object) – IN: device pointer to first matrix C_1 of dimension (ldc, n) on the GPU.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, m ).
- strideC (
int) – INOUT: [hipblasStride] stride from the start of one matrix (C_i) and the next one (C_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDsymmStridedBatched(handle, side, uplo, int m, int n, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsymmStridedBatched(handle, side, uplo, int m, int n, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- CP (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsymmStridedBatched(handle, side, uplo, int m, int n, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsymmStridedBatched_v2(handle, side, uplo, int m, int n, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsymmStridedBatched_v2(handle, side, uplo, int m, int n, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsyrk(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc)#
BLAS Level 3 API
@{
syrk performs one of the matrix-matrix operations for a symmetric rank-k update
C := alpha*op( A )*op( A )^T + beta*C
where alpha and beta are scalars, op(A) is an n by k matrix, and C is a symmetric n x n matrix stored as either upper or lower.
op( A ) = A, and A is n by k if transA == HIPBLAS_OP_N op( A ) = A^T and A is k by n if transA == HIPBLAS_OP_T
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
HIPBLAS_OP_C is not supported for complex types, see cherk and zherk.
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_T: op(A) = A^T HIPBLAS_OP_N: op(A) = A HIPBLAS_OP_C: op(A) = A^T
- n (
int) – IN: [int] n specifies the number of rows and columns of C. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
Pointer/object) – IN: pointer storing matrix A on the GPU. Martrix dimension is ( lda, k ) when if transA = HIPBLAS_OP_N, otherwise (lda, n) only the upper/lower triangular part is accessed.
- lda (
int) – IN: [int] lda specifies the first dimension of A. if transA = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
Pointer/object) – IN: pointer storing matrix C on the GPU.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- handle (
- hip.hipblas.hipblasDsyrk(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyrk(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- CP (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyrk(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyrk_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyrk_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsyrkBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc, int batchCount)#
BLAS Level 3 API
@{
syrkBatched performs a batch of the matrix-matrix operations for a symmetric rank-k update
C_i := alpha*op( A_i )*op( A_i )^T + beta*C_i
where alpha and beta are scalars, op(A_i) is an n by k matrix, and C_i is a symmetric n x n matrix stored as either upper or lower.
op( A_i ) = A_i, and A_i is n by k if transA == HIPBLAS_OP_N op( A_i ) = A_i^T and A_i is k by n if transA == HIPBLAS_OP_T
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
HIPBLAS_OP_C is not supported for complex types, see cherk and zherk.
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C_i is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_T: op(A) = A^T HIPBLAS_OP_N: op(A) = A HIPBLAS_OP_C: op(A) = A^T
- n (
int) – IN: [int] n specifies the number of rows and columns of C_i. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix_i A of dimension (lda, k) when transA is HIPBLAS_OP_N, otherwise of dimension (lda, n)
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if transA = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
Pointer/object) – IN: device array of device pointers storing each matrix C_i on the GPU.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDsyrkBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyrkBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyrkBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyrkBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyrkBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsyrkStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, beta, CP, int ldc, long strideC, int batchCount)#
BLAS Level 3 API
@{
syrkStridedBatched performs a batch of the matrix-matrix operations for a symmetric rank-k update
C_i := alpha*op( A_i )*op( A_i )^T + beta*C_i
where alpha and beta are scalars, op(A_i) is an n by k matrix, and C_i is a symmetric n x n matrix stored as either upper or lower.
op( A_i ) = A_i, and A_i is n by k if transA == HIPBLAS_OP_N op( A_i ) = A_i^T and A_i is k by n if transA == HIPBLAS_OP_T
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
HIPBLAS_OP_C is not supported for complex types, see cherk and zherk.
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C_i is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_T: op(A) = A^T HIPBLAS_OP_N: op(A) = A HIPBLAS_OP_C: op(A) = A^T
- n (
int) – IN: [int] n specifies the number of rows and columns of C_i. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
Pointer/object) – IN: Device pointer to the first matrix A_1 on the GPU of dimension (lda, k) when transA is HIPBLAS_OP_N, otherwise of dimension (lda, n)
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if transA = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
Pointer/object) – IN: Device pointer to the first matrix C_1 on the GPU. on the GPU.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- strideC (
int) – INOUT: [hipblasStride] stride from the start of one matrix (C_i) and the next one (C_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDsyrkStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyrkStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- CP (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyrkStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyrkStridedBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyrkStridedBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsyr2k(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
BLAS Level 3 API
@{
syr2k performs one of the matrix-matrix operations for a symmetric rank-2k update
C := alpha*(op( A )*op( B )^T + op( B )*op( A )^T) + beta*C
where alpha and beta are scalars, op(A) and op(B) are n by k matrix, and C is a symmetric n x n matrix stored as either upper or lower.
op( A ) = A, op( B ) = B, and A and B are n by k if trans == HIPBLAS_OP_N op( A ) = A^T, op( B ) = B^T, and A and B are k by n if trans == HIPBLAS_OP_T
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_T: op( A ) = A^T, op( B ) = B^T HIPBLAS_OP_N: op( A ) = A, op( B ) = B
- n (
int) – IN: [int] n specifies the number of rows and columns of C. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A) and op(B). k >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
Pointer/object) – IN: pointer storing matrix A on the GPU. Martrix dimension is ( lda, k ) when if trans = HIPBLAS_OP_N, otherwise (lda, n) only the upper/lower triangular part is accessed.
- lda (
int) – IN: [int] lda specifies the first dimension of A. if trans = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- BP (
Pointer/object) – IN: pointer storing matrix B on the GPU. Martrix dimension is ( ldb, k ) when if trans = HIPBLAS_OP_N, otherwise (ldb, n) only the upper/lower triangular part is accessed.
- ldb (
int) – IN: [int] ldb specifies the first dimension of B. if trans = HIPBLAS_OP_N, ldb >= max( 1, n ), otherwise ldb >= max( 1, k ).
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
Pointer/object) – IN: pointer storing matrix C on the GPU.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- handle (
- hip.hipblas.hipblasDsyr2k(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyr2k(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- BP (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- CP (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyr2k(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyr2k_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyr2k_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsyr2kBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
BLAS Level 3 API
@{
syr2kBatched performs a batch of the matrix-matrix operations for a symmetric rank-2k update
C_i := alpha*(op( A_i )*op( B_i )^T + op( B_i )*op( A_i )^T) + beta*C_i
where alpha and beta are scalars, op(A_i) and op(B_i) are n by k matrix, and C_i is a symmetric n x n matrix stored as either upper or lower.
op( A_i ) = A_i, op( B_i ) = B_i, and A_i and B_i are n by k if trans == HIPBLAS_OP_N op( A_i ) = A_i^T, op( B_i ) = B_i^T, and A_i and B_i are k by n if trans == HIPBLAS_OP_T
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C_i is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_T: op( A_i ) = A_i^T, op( B_i ) = B_i^T HIPBLAS_OP_N: op( A_i ) = A_i, op( B_i ) = B_i
- n (
int) – IN: [int] n specifies the number of rows and columns of C_i. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix_i A of dimension (lda, k) when trans is HIPBLAS_OP_N, otherwise of dimension (lda, n)
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if trans = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- BP (
Pointer/object) – IN: device array of device pointers storing each matrix_i B of dimension (ldb, k) when trans is HIPBLAS_OP_N, otherwise of dimension (ldb, n)
- ldb (
int) – IN: [int] ldb specifies the first dimension of B. if trans = HIPBLAS_OP_N, ldb >= max( 1, n ), otherwise ldb >= max( 1, k ).
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
Pointer/object) – IN: device array of device pointers storing each matrix C_i on the GPU.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDsyr2kBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyr2kBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyr2kBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyr2kBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyr2kBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsyr2kStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
BLAS Level 3 API
@{
syr2kStridedBatched performs a batch of the matrix-matrix operations for a symmetric rank-2k update
C_i := alpha*(op( A_i )*op( B_i )^T + op( B_i )*op( A_i )^T) + beta*C_i
where alpha and beta are scalars, op(A_i) and op(B_i) are n by k matrix, and C_i is a symmetric n x n matrix stored as either upper or lower.
op( A_i ) = A_i, op( B_i ) = B_i, and A_i and B_i are n by k if trans == HIPBLAS_OP_N op( A_i ) = A_i^T, op( B_i ) = B_i^T, and A_i and B_i are k by n if trans == HIPBLAS_OP_T
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C_i is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_T: op( A_i ) = A_i^T, op( B_i ) = B_i^T HIPBLAS_OP_N: op( A_i ) = A_i, op( B_i ) = B_i
- n (
int) – IN: [int] n specifies the number of rows and columns of C_i. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
Pointer/object) – IN: Device pointer to the first matrix A_1 on the GPU of dimension (lda, k) when trans is HIPBLAS_OP_N, otherwise of dimension (lda, n)
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if trans = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- BP (
Pointer/object) – IN: Device pointer to the first matrix B_1 on the GPU of dimension (ldb, k) when trans is HIPBLAS_OP_N, otherwise of dimension (ldb, n)
- ldb (
int) – IN: [int] ldb specifies the first dimension of B_i. if trans = HIPBLAS_OP_N, ldb >= max( 1, n ), otherwise ldb >= max( 1, k ).
- strideB (
int) – IN: [hipblasStride] stride from the start of one matrix (B_i) and the next one (B_i+1)
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
Pointer/object) – IN: Device pointer to the first matrix C_1 on the GPU.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- strideC (
int) – INOUT: [hipblasStride] stride from the start of one matrix (C_i) and the next one (C_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDsyr2kStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyr2kStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- CP (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyr2kStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyr2kStridedBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyr2kStridedBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsyrkx(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
BLAS Level 3 API
@{
syrkx performs one of the matrix-matrix operations for a symmetric rank-k update
C := alpha*op( A )*op( B )^T + beta*C
where alpha and beta are scalars, op(A) and op(B) are n by k matrix, and C is a symmetric n x n matrix stored as either upper or lower. This routine should only be used when the caller can guarantee that the result of op( A )*op( B )^T will be symmetric.
op( A ) = A, op( B ) = B, and A and B are n by k if trans == HIPBLAS_OP_N op( A ) = A^T, op( B ) = B^T, and A and B are k by n if trans == HIPBLAS_OP_T
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_T: op( A ) = A^T, op( B ) = B^T HIPBLAS_OP_N: op( A ) = A, op( B ) = B
- n (
int) – IN: [int] n specifies the number of rows and columns of C. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A) and op(B). k >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
Pointer/object) – IN: pointer storing matrix A on the GPU. Martrix dimension is ( lda, k ) when if trans = HIPBLAS_OP_N, otherwise (lda, n) only the upper/lower triangular part is accessed.
- lda (
int) – IN: [int] lda specifies the first dimension of A. if trans = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- BP (
Pointer/object) – IN: pointer storing matrix B on the GPU. Martrix dimension is ( ldb, k ) when if trans = HIPBLAS_OP_N, otherwise (ldb, n) only the upper/lower triangular part is accessed.
- ldb (
int) – IN: [int] ldb specifies the first dimension of B. if trans = HIPBLAS_OP_N, ldb >= max( 1, n ), otherwise ldb >= max( 1, k ).
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
Pointer/object) – IN: pointer storing matrix C on the GPU.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- handle (
- hip.hipblas.hipblasDsyrkx(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyrkx(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- BP (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- CP (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyrkx(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyrkx_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyrkx_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsyrkxBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
BLAS Level 3 API
@{
syrkxBatched performs a batch of the matrix-matrix operations for a symmetric rank-k update
C_i := alpha*op( A_i )*op( B_i )^T + beta*C_i
where alpha and beta are scalars, op(A_i) and op(B_i) are n by k matrix, and C_i is a symmetric n x n matrix stored as either upper or lower. This routine should only be used when the caller can guarantee that the result of op( A_i )*op( B_i )^T will be symmetric.
op( A_i ) = A_i, op( B_i ) = B_i, and A_i and B_i are n by k if trans == HIPBLAS_OP_N op( A_i ) = A_i^T, op( B_i ) = B_i^T, and A_i and B_i are k by n if trans == HIPBLAS_OP_T
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C_i is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_T: op( A_i ) = A_i^T, op( B_i ) = B_i^T HIPBLAS_OP_N: op( A_i ) = A_i, op( B_i ) = B_i
- n (
int) – IN: [int] n specifies the number of rows and columns of C_i. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix_i A of dimension (lda, k) when trans is HIPBLAS_OP_N, otherwise of dimension (lda, n)
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if trans = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- BP (
Pointer/object) – IN: device array of device pointers storing each matrix_i B of dimension (ldb, k) when trans is HIPBLAS_OP_N, otherwise of dimension (ldb, n)
- ldb (
int) – IN: [int] ldb specifies the first dimension of B. if trans = HIPBLAS_OP_N, ldb >= max( 1, n ), otherwise ldb >= max( 1, k ).
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
Pointer/object) – IN: device array of device pointers storing each matrix C_i on the GPU.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDsyrkxBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyrkxBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyrkxBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyrkxBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyrkxBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSsyrkxStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
BLAS Level 3 API
@{
syrkxStridedBatched performs a batch of the matrix-matrix operations for a symmetric rank-k update
C_i := alpha*op( A_i )*op( B_i )^T + beta*C_i
where alpha and beta are scalars, op(A_i) and op(B_i) are n by k matrix, and C_i is a symmetric n x n matrix stored as either upper or lower. This routine should only be used when the caller can guarantee that the result of op( A_i )*op( B_i )^T will be symmetric.
op( A_i ) = A_i, op( B_i ) = B_i, and A_i and B_i are n by k if trans == HIPBLAS_OP_N op( A_i ) = A_i^T, op( B_i ) = B_i^T, and A_i and B_i are k by n if trans == HIPBLAS_OP_T
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: C_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: C_i is a lower triangular matrix
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_T: op( A_i ) = A_i^T, op( B_i ) = B_i^T HIPBLAS_OP_N: op( A_i ) = A_i, op( B_i ) = B_i
- n (
int) – IN: [int] n specifies the number of rows and columns of C_i. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of op(A). k >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and A need not be set before entry.
- AP (
Pointer/object) – IN: Device pointer to the first matrix A_1 on the GPU of dimension (lda, k) when trans is HIPBLAS_OP_N, otherwise of dimension (lda, n)
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if trans = HIPBLAS_OP_N, lda >= max( 1, n ), otherwise lda >= max( 1, k ).
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- BP (
Pointer/object) – IN: Device pointer to the first matrix B_1 on the GPU of dimension (ldb, k) when trans is HIPBLAS_OP_N, otherwise of dimension (ldb, n)
- ldb (
int) – IN: [int] ldb specifies the first dimension of B_i. if trans = HIPBLAS_OP_N, ldb >= max( 1, n ), otherwise ldb >= max( 1, k ).
- strideB (
int) – IN: [hipblasStride] stride from the start of one matrix (B_i) and the next one (B_i+1)
- beta (
Pointer/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
Pointer/object) – IN: Device pointer to the first matrix C_1 on the GPU.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, n ).
- strideC (
int) – INOUT: [hipblasStride] stride from the start of one matrix (C_i) and the next one (C_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDsyrkxStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyrkxStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- CP (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyrkxStridedBatched(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCsyrkxStridedBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZsyrkxStridedBatched_v2(handle, uplo, transA, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSgeam(handle, transA, transB, int m, int n, alpha, AP, int lda, beta, BP, int ldb, CP, int ldc)#
BLAS Level 3 API
@{
geam performs one of the matrix-matrix operations
C = alpha*op( A ) + beta*op( B ),
where op( X ) is one of
op( X ) = X or op( X ) = X**T or op( X ) = X**H,
alpha and beta are scalars, and A, B and C are matrices, with op( A ) an m by n matrix, op( B ) an m by n matrix, and C an m by n matrix.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( A )
- transB (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( B )
- m (
int) – IN: [int] matrix dimension m.
- n (
int) – IN: [int] matrix dimension n.
- alpha (
Pointer/object) – IN: device pointer or host pointer specifying the scalar alpha.
- AP (
Pointer/object) – IN: device pointer storing matrix A.
- lda (
int) – IN: [int] specifies the leading dimension of A.
- beta (
Pointer/object) – IN: device pointer or host pointer specifying the scalar beta.
- BP (
Pointer/object) – IN: device pointer storing matrix B.
- ldb (
int) – IN: [int] specifies the leading dimension of B.
- CP (
Pointer/object) – IN,OUT: device pointer storing matrix C.
- ldc (
int) – IN: [int] specifies the leading dimension of C.
- handle (
- hip.hipblas.hipblasDgeam(handle, transA, transB, int m, int n, alpha, AP, int lda, beta, BP, int ldb, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgeam(handle, transA, transB, int m, int n, alpha, AP, int lda, beta, BP, int ldb, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- BP (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- CP (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgeam(handle, transA, transB, int m, int n, alpha, AP, int lda, beta, BP, int ldb, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgeam_v2(handle, transA, transB, int m, int n, alpha, AP, int lda, beta, BP, int ldb, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- beta (
float2/object): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgeam_v2(handle, transA, transB, int m, int n, alpha, AP, int lda, beta, BP, int ldb, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- beta (
double2/object): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSgeamBatched(handle, transA, transB, int m, int n, alpha, AP, int lda, beta, BP, int ldb, CP, int ldc, int batchCount)#
BLAS Level 3 API
@{
geamBatched performs one of the batched matrix-matrix operations
C_i = alpha*op( A_i ) + beta*op( B_i ) for i = 0, 1, … batchCount - 1
where alpha and beta are scalars, and op(A_i), op(B_i) and C_i are m by n matrices and op( X ) is one of
op( X ) = X or op( X ) = X**T
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( A )
- transB (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( B )
- m (
int) – IN: [int] matrix dimension m.
- n (
int) – IN: [int] matrix dimension n.
- alpha (
Pointer/object) – IN: device pointer or host pointer specifying the scalar alpha.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix A_i on the GPU. Each A_i is of dimension ( lda, k ), where k is m when transA == HIPBLAS_OP_N and is n when transA == HIPBLAS_OP_T.
- lda (
int) – IN: [int] specifies the leading dimension of A.
- beta (
Pointer/object) – IN: device pointer or host pointer specifying the scalar beta.
- BP (
Pointer/object) – IN: device array of device pointers storing each matrix B_i on the GPU. Each B_i is of dimension ( ldb, k ), where k is m when transB == HIPBLAS_OP_N and is n when transB == HIPBLAS_OP_T.
- ldb (
int) – IN: [int] specifies the leading dimension of B.
- CP (
Pointer/object) – IN,OUT: device array of device pointers storing each matrix C_i on the GPU. Each C_i is of dimension ( ldc, n ).
- ldc (
int) – IN: [int] specifies the leading dimension of C.
- batchCount (
int) – IN: [int] number of instances i in the batch.
- handle (
- hip.hipblas.hipblasDgeamBatched(handle, transA, transB, int m, int n, alpha, AP, int lda, beta, BP, int ldb, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgeamBatched(handle, transA, transB, int m, int n, alpha, AP, int lda, beta, BP, int ldb, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgeamBatched(handle, transA, transB, int m, int n, alpha, AP, int lda, beta, BP, int ldb, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgeamBatched_v2(handle, transA, transB, int m, int n, alpha, AP, int lda, beta, BP, int ldb, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- beta (
float2/object): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgeamBatched_v2(handle, transA, transB, int m, int n, alpha, AP, int lda, beta, BP, int ldb, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- beta (
double2/object): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSgeamStridedBatched(handle, transA, transB, int m, int n, alpha, AP, int lda, long strideA, beta, BP, int ldb, long strideB, CP, int ldc, long strideC, int batchCount)#
BLAS Level 3 API
@{
geamStridedBatched performs one of the batched matrix-matrix operations
C_i = alpha*op( A_i ) + beta*op( B_i ) for i = 0, 1, … batchCount - 1
where alpha and beta are scalars, and op(A_i), op(B_i) and C_i are m by n matrices and op( X ) is one of
op( X ) = X or op( X ) = X**T
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( A )
- transB (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( B )
- m (
int) – IN: [int] matrix dimension m.
- n (
int) – IN: [int] matrix dimension n.
- alpha (
Pointer/object) – IN: device pointer or host pointer specifying the scalar alpha.
- AP (
Pointer/object) – IN: device pointer to the first matrix A_0 on the GPU. Each A_i is of dimension ( lda, k ), where k is m when transA == HIPBLAS_OP_N and is n when transA == HIPBLAS_OP_T.
- lda (
int) – IN: [int] specifies the leading dimension of A.
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- beta (
Pointer/object) – IN: device pointer or host pointer specifying the scalar beta.
- BP (
Pointer/object) – IN: pointer to the first matrix B_0 on the GPU. Each B_i is of dimension ( ldb, k ), where k is m when transB == HIPBLAS_OP_N and is n when transB == HIPBLAS_OP_T.
- ldb (
int) – IN: [int] specifies the leading dimension of B.
- strideB (
int) – IN: [hipblasStride] stride from the start of one matrix (B_i) and the next one (B_i+1)
- CP (
Pointer/object) – IN,OUT: pointer to the first matrix C_0 on the GPU. Each C_i is of dimension ( ldc, n ).
- ldc (
int) – IN: [int] specifies the leading dimension of C.
- strideC (
int) – IN: [hipblasStride] stride from the start of one matrix (C_i) and the next one (C_i+1)
- batchCount (
int) – IN: [int] number of instances i in the batch.
- handle (
- hip.hipblas.hipblasDgeamStridedBatched(handle, transA, transB, int m, int n, alpha, AP, int lda, long strideA, beta, BP, int ldb, long strideB, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgeamStridedBatched(handle, transA, transB, int m, int n, alpha, AP, int lda, long strideA, beta, BP, int ldb, long strideB, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- beta (
hipblasComplex/object): (undocumented)
- BP (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- CP (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgeamStridedBatched(handle, transA, transB, int m, int n, alpha, AP, int lda, long strideA, beta, BP, int ldb, long strideB, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgeamStridedBatched_v2(handle, transA, transB, int m, int n, alpha, AP, int lda, long strideA, beta, BP, int ldb, long strideB, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- beta (
float2/object): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgeamStridedBatched_v2(handle, transA, transB, int m, int n, alpha, AP, int lda, long strideA, beta, BP, int ldb, long strideB, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- beta (
double2/object): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChemm(handle, side, uplo, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
BLAS Level 3 API
@{
hemm performs one of the matrix-matrix operations:
C := alpha*A*B + beta*C if side == HIPBLAS_SIDE_LEFT, C := alpha*B*A + beta*C if side == HIPBLAS_SIDE_RIGHT,
where alpha and beta are scalars, B and C are m by n matrices, and A is a Hermitian matrix stored as either upper or lower.
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] HIPBLAS_SIDE_LEFT: C := alpha*A*B + beta*C HIPBLAS_SIDE_RIGHT: C := alpha*B*A + beta*C
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: A is a lower triangular matrix
- n (
int) – IN: [int] n specifies the number of rows of B and C. n >= 0.
- k (
int) – IN: [int] n specifies the number of columns of B and C. k >= 0.
- alpha (
hipblasComplex/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A and B are not referenced.
- AP (
hipblasComplex/object) – IN: pointer storing matrix A on the GPU. A is m by m if side == HIPBLAS_SIDE_LEFT A is n by n if side == HIPBLAS_SIDE_RIGHT Only the upper/lower triangular part is accessed. The imaginary component of the diagonal elements is not used.
- lda (
int) – IN: [int] lda specifies the first dimension of A. if side = HIPBLAS_SIDE_LEFT, lda >= max( 1, m ), otherwise lda >= max( 1, n ).
- BP (
hipblasComplex/object) – IN: pointer storing matrix B on the GPU. Matrix dimension is m by n
- ldb (
int) – IN: [int] ldb specifies the first dimension of B. ldb >= max( 1, m )
- beta (
hipblasComplex/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
hipblasComplex/object) – IN: pointer storing matrix C on the GPU. Matrix dimension is m by n
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, m )
- handle (
- hip.hipblas.hipblasZhemm(handle, side, uplo, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChemm_v2(handle, side, uplo, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZhemm_v2(handle, side, uplo, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChemmBatched(handle, side, uplo, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
BLAS Level 3 API
@{
hemmBatched performs a batch of the matrix-matrix operations:
C_i := alpha*A_i*B_i + beta*C_i if side == HIPBLAS_SIDE_LEFT, C_i := alpha*B_i*A_i + beta*C_i if side == HIPBLAS_SIDE_RIGHT,
where alpha and beta are scalars, B_i and C_i are m by n matrices, and A_i is a Hermitian matrix stored as either upper or lower.
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] HIPBLAS_SIDE_LEFT: C_i := alpha*A_i*B_i + beta*C_i HIPBLAS_SIDE_RIGHT: C_i := alpha*B_i*A_i + beta*C_i
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: A_i is a lower triangular matrix
- n (
int) – IN: [int] n specifies the number of rows of B_i and C_i. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of B_i and C_i. k >= 0.
- alpha (
hipblasComplex/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A_i and B_i are not referenced.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix A_i on the GPU. A_i is m by m if side == HIPBLAS_SIDE_LEFT A_i is n by n if side == HIPBLAS_SIDE_RIGHT Only the upper/lower triangular part is accessed. The imaginary component of the diagonal elements is not used.
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if side = HIPBLAS_SIDE_LEFT, lda >= max( 1, m ), otherwise lda >= max( 1, n ).
- BP (
Pointer/object) – IN: device array of device pointers storing each matrix B_i on the GPU. Matrix dimension is m by n
- ldb (
int) – IN: [int] ldb specifies the first dimension of B_i. ldb >= max( 1, m )
- beta (
hipblasComplex/object) – IN: beta specifies the scalar beta. When beta is zero then C_i need not be set before entry.
- CP (
Pointer/object) – IN: device array of device pointers storing each matrix C_i on the GPU. Matrix dimension is m by n
- ldc (
int) – IN: [int] ldc specifies the first dimension of C_i. ldc >= max( 1, m )
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasZhemmBatched(handle, side, uplo, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChemmBatched_v2(handle, side, uplo, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZhemmBatched_v2(handle, side, uplo, int n, int k, alpha, AP, int lda, BP, int ldb, beta, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChemmStridedBatched(handle, side, uplo, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
BLAS Level 3 API
@{
hemmStridedBatched performs a batch of the matrix-matrix operations:
C_i := alpha*A_i*B_i + beta*C_i if side == HIPBLAS_SIDE_LEFT, C_i := alpha*B_i*A_i + beta*C_i if side == HIPBLAS_SIDE_RIGHT,
where alpha and beta are scalars, B_i and C_i are m by n matrices, and A_i is a Hermitian matrix stored as either upper or lower.
Supported precisions in rocBLAS : c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] HIPBLAS_SIDE_LEFT: C_i := alpha*A_i*B_i + beta*C_i HIPBLAS_SIDE_RIGHT: C_i := alpha*B_i*A_i + beta*C_i
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A_i is an upper triangular matrix HIPBLAS_FILL_MODE_LOWER: A_i is a lower triangular matrix
- n (
int) – IN: [int] n specifies the number of rows of B_i and C_i. n >= 0.
- k (
int) – IN: [int] k specifies the number of columns of B_i and C_i. k >= 0.
- alpha (
hipblasComplex/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A_i and B_i are not referenced.
- AP (
hipblasComplex/object) – IN: device pointer to first matrix A_1 A_i is m by m if side == HIPBLAS_SIDE_LEFT A_i is n by n if side == HIPBLAS_SIDE_RIGHT Only the upper/lower triangular part is accessed. The imaginary component of the diagonal elements is not used.
- lda (
int) – IN: [int] lda specifies the first dimension of A_i. if side = HIPBLAS_SIDE_LEFT, lda >= max( 1, m ), otherwise lda >= max( 1, n ).
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- BP (
hipblasComplex/object) – IN: device pointer to first matrix B_1 of dimension (ldb, n) on the GPU
- ldb (
int) – IN: [int] ldb specifies the first dimension of B_i. if side = HIPBLAS_OP_N, ldb >= max( 1, m ), otherwise ldb >= max( 1, n ).
- strideB (
int) – IN: [hipblasStride] stride from the start of one matrix (B_i) and the next one (B_i+1)
- beta (
hipblasComplex/object) – IN: beta specifies the scalar beta. When beta is zero then C need not be set before entry.
- CP (
hipblasComplex/object) – IN: device pointer to first matrix C_1 of dimension (ldc, n) on the GPU.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, m )
- strideC (
int) – INOUT: [hipblasStride] stride from the start of one matrix (C_i) and the next one (C_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch
- handle (
- hip.hipblas.hipblasZhemmStridedBatched(handle, side, uplo, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
hipblasDoubleComplex/object): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasChemmStridedBatched_v2(handle, side, uplo, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
float2/object): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZhemmStridedBatched_v2(handle, side, uplo, int n, int k, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, beta, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
double2/object): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStrmm(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, C, int ldc)#
BLAS Level 3 API
@{
trmm performs one of the matrix-matrix operations
C := alpha*op( A )*B, or C := alpha*B*op( A )
where alpha is a scalar, B and C are an m by n matrices, A is a unit, or non-unit, upper or lower triangular matrix and op( A ) is one of
op( A ) = A or op( A ) = A^T or op( A ) = A^H.
Note that trmm can provide in-place functionality by passing in the same address for both matrices B and C and by setting ldb equal to ldc.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
When uplo == HIPBLAS_FILL_MODE_UPPER the leading k by k upper triangular part of the array A must contain the upper triangular matrix and the strictly lower triangular part of A is not referenced.
When uplo == HIPBLAS_FILL_MODE_LOWER the leading k by k lower triangular part of the array A must contain the lower triangular matrix and the strictly upper triangular part of A is not referenced.
Note that when diag == HIPBLAS_DIAG_UNIT the diagonal elements of A are not referenced either, but are assumed to be unity.
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] Specifies whether op(A) multiplies B from the left or right as follows: HIPBLAS_SIDE_LEFT: C := alpha*op( A )*B. HIPBLAS_SIDE_RIGHT: C := alpha*B*op( A ).
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] Specifies whether the matrix A is an upper or lower triangular matrix as follows: HIPBLAS_FILL_MODE_UPPER: A is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] Specifies the form of op(A) to be used in the matrix multiplication as follows: HIPBLAS_OP_N: op(A) = A. HIPBLAS_OP_T: op(A) = A^T. HIPBLAS_OP_C: op(A) = A^H.
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] Specifies whether or not A is unit triangular as follows: HIPBLAS_DIAG_UNIT: A is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: A is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of B and C. m >= 0.
- n (
int) – IN: [int] n specifies the number of columns of B and C. n >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A is not referenced and B need not be set before entry.
- A (
Pointer/object) – IN: Device pointer to matrix A on the GPU. A has dimension ( lda, k ), where k is m when side == HIPBLAS_SIDE_LEFT and is n when side == HIPBLAS_SIDE_RIGHT.
- lda (
int) – IN: [int] lda specifies the first dimension of A. if side == HIPBLAS_SIDE_LEFT, lda >= max( 1, m ), if side == HIPBLAS_SIDE_RIGHT, lda >= max( 1, n ).
- B (
Pointer/object) – INOUT: Device pointer to the matrix B of dimension (ldb, n) on the GPU.
- ldb (
int) – IN: [int] ldb specifies the first dimension of B. ldb >= max( 1, m ).
- C (
Pointer/object) – IN: Device pointer to the matrix C of dimension (ldc, n) on the GPU. Users can pass in the same matrix B to parameter C to achieve in-place functionality of trmm.
- ldc (
int) – IN: [int] ldc specifies the first dimension of C. ldc >= max( 1, m ).
- handle (
- hip.hipblas.hipblasDtrmm(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, C, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- C (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrmm(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, C, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- A (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- B (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- C (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrmm(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, C, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- A (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- B (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- C (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrmm_v2(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, C, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- A (
float2/object): (undocumented)
- lda (
int): (undocumented)
- B (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- C (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrmm_v2(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, C, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- A (
double2/object): (undocumented)
- lda (
int): (undocumented)
- B (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- C (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStrmmBatched(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, C, int ldc, int batchCount)#
BLAS Level 3 API
@{
trmmBatched performs one of the batched matrix-matrix operations
C_i := alpha*op( A_i )*B_i, or C_i := alpha*B_i*op( A_i ) for i = 0, 1, … batchCount -1
where alpha is a scalar, B_i and C_i are m by n matrices, A_i is a unit, or non-unit, upper or lower triangular matrix and op( A_i ) is one of
op( A_i ) = A_i or op( A_i ) = A_i^T or op( A_i ) = A_i^H.
Note that trmmBatched can provide in-place functionality by passing in the same address for both matrices B and C and by setting ldb equal to ldc.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
When uplo == HIPBLAS_FILL_MODE_UPPER the leading k by k upper triangular part of the array A must contain the upper triangular matrix and the strictly lower triangular part of A is not referenced.
When uplo == HIPBLAS_FILL_MODE_LOWER the leading k by k lower triangular part of the array A must contain the lower triangular matrix and the strictly upper triangular part of A is not referenced.
Note that when diag == HIPBLAS_DIAG_UNIT the diagonal elements of A_i are not referenced either, but are assumed to be unity.
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] Specifies whether op(A_i) multiplies B_i from the left or right as follows: HIPBLAS_SIDE_LEFT: B_i := alpha*op( A_i )*B_i. HIPBLAS_SIDE_RIGHT: B_i := alpha*B_i*op( A_i ).
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] Specifies whether the matrix A is an upper or lower triangular matrix as follows: HIPBLAS_FILL_MODE_UPPER: A is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] Specifies the form of op(A_i) to be used in the matrix multiplication as follows: HIPBLAS_OP_N: op(A_i) = A_i. HIPBLAS_OP_T: op(A_i) = A_i^T. HIPBLAS_OP_C: op(A_i) = A_i^H.
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] Specifies whether or not A_i is unit triangular as follows: HIPBLAS_DIAG_UNIT: A_i is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: A_i is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of B_i and C_i. m >= 0.
- n (
int) – IN: [int] n specifies the number of columns of B_i and C_i. n >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A_i is not referenced and B_i need not be set before entry.
- A (
ListOfPointer/object) – IN: Device array of device pointers storing each matrix A_i on the GPU. Each A_i is of dimension ( lda, k ), where k is m when side == HIPBLAS_SIDE_LEFT and is n when side == HIPBLAS_SIDE_RIGHT.
- lda (
int) – IN: [int] lda specifies the first dimension of A. if side == HIPBLAS_SIDE_LEFT, lda >= max( 1, m ), if side == HIPBLAS_SIDE_RIGHT, lda >= max( 1, n ).
- B (
ListOfPointer/object) – INOUT: device array of device pointers storing each matrix B_i of dimension (ldb, n) on the GPU.
- ldb (
int) – IN: [int] ldb specifies the first dimension of B_i. ldb >= max( 1, m ).
- C (
ListOfPointer/object) – IN: device array of device pointers storing each matrix C_i of dimension (ldc, n) on the GPU. Users can pass in the same matrices B to parameter C to achieve in-place functionality of trmmBatched.
- ldc (
int) – IN: lec specifies the first dimension of C_i. ldc >= max( 1, m ).
- batchCount (
int) – IN: [int] number of instances i in the batch.
- handle (
- hip.hipblas.hipblasDtrmmBatched(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, C, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- A (
ListOfPointer/object): (undocumented)
- lda (
int): (undocumented)
- B (
ListOfPointer/object): (undocumented)
- ldb (
int): (undocumented)
- C (
ListOfPointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrmmBatched(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, C, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- C (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrmmBatched(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, C, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- C (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrmmBatched_v2(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, C, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- C (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrmmBatched_v2(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, C, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- C (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStrmmStridedBatched(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, long strideA, B, int ldb, long strideB, C, int ldc, long strideC, int batchCount)#
BLAS Level 3 API
@{
trmmStridedBatched performs one of the strided_batched matrix-matrix operations
C_i := alpha*op( A_i )*B_i, or C_i := alpha*B_i*op( A_i ) for i = 0, 1, … batchCount -1
where alpha is a scalar, B_i and C_i are m by n matrices, A_i is a unit, or non-unit, upper or lower triangular matrix and op( A_i ) is one of
op( A_i ) = A_i or op( A_i ) = A_i^T or op( A_i ) = A_i^H.
Note that trmmStridedBatched can provide in-place functionality by passing in the same address for both matrices B and C and by setting ldb equal to ldc.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
When uplo == HIPBLAS_FILL_MODE_UPPER the leading k by k upper triangular part of the array A must contain the upper triangular matrix and the strictly lower triangular part of A is not referenced.
When uplo == HIPBLAS_FILL_MODE_LOWER the leading k by k lower triangular part of the array A must contain the lower triangular matrix and the strictly upper triangular part of A is not referenced.
Note that when diag == HIPBLAS_DIAG_UNIT the diagonal elements of A_i are not referenced either, but are assumed to be unity.
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] Specifies whether op(A_i) multiplies B_i from the left or right as follows: HIPBLAS_SIDE_LEFT: C_i := alpha*op( A_i )*B_i. HIPBLAS_SIDE_RIGHT: C_i := alpha*B_i*op( A_i ).
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] Specifies whether the matrix A is an upper or lower triangular matrix as follows: HIPBLAS_FILL_MODE_UPPER: A is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] Specifies the form of op(A_i) to be used in the matrix multiplication as follows: HIPBLAS_OP_N: op(A_i) = A_i. HIPBLAS_OP_T: op(A_i) = A_i^T. HIPBLAS_OP_C: op(A_i) = A_i^H.
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] Specifies whether or not A_i is unit triangular as follows: HIPBLAS_DIAG_UNIT: A_i is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: A_i is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of B_i and C_i. m >= 0.
- n (
int) – IN: [int] n specifies the number of columns of B_i and C_i. n >= 0.
- alpha (
Pointer/object) – IN: alpha specifies the scalar alpha. When alpha is zero then A_i is not referenced and B_i need not be set before entry.
- A (
Pointer/object) – IN: Device pointer to the first matrix A_0 on the GPU. Each A_i is of dimension ( lda, k ), where k is m when side == HIPBLAS_SIDE_LEFT and is n when side == HIPBLAS_SIDE_RIGHT.
- lda (
int) – IN: [int] lda specifies the first dimension of A. if side == HIPBLAS_SIDE_LEFT, lda >= max( 1, m ), if side == HIPBLAS_SIDE_RIGHT, lda >= max( 1, n ).
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- B (
Pointer/object) – INOUT: Device pointer to the first matrix B_0 on the GPU. Each B_i is of dimension ( ldb, n )
- ldb (
int) – IN: [int] ldb specifies the first dimension of B_i. ldb >= max( 1, m ).
- strideB (
int) – IN: [hipblasStride] stride from the start of one matrix (B_i) and the next one (B_i+1)
- C (
Pointer/object) – IN: Device pointer to the first matrix C_0 on the GPU. Each C_i is of dimension ( ldc, n ).
- ldc (
int) – IN: [int] ldc specifies the first dimension of C_i. ldc >= max( 1, m ).
- strideC (
int) – IN: [hipblasStride] stride from the start of one matrix (C_i) and the next one (C_i+1)
- batchCount (
int) – IN: [int] number of instances i in the batch.
- handle (
- hip.hipblas.hipblasDtrmmStridedBatched(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, long strideA, B, int ldb, long strideB, C, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- C (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrmmStridedBatched(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, long strideA, B, int ldb, long strideB, C, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- A (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- B (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- C (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrmmStridedBatched(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, long strideA, B, int ldb, long strideB, C, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- A (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- B (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- C (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrmmStridedBatched_v2(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, long strideA, B, int ldb, long strideB, C, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- A (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- B (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- C (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrmmStridedBatched_v2(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, long strideA, B, int ldb, long strideB, C, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- A (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- B (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- C (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStrsm(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, BP, int ldb)#
BLAS Level 3 API
@{
trsm solves
op(A)*X = alpha*B or X*op(A) = alpha*B,
where alpha is a scalar, X and B are m by n matrices, A is triangular matrix and op(A) is one of
op( A ) = A or op( A ) = A^T or op( A ) = A^H.
The matrix X is overwritten on B.
Note about memory allocation: When trsm is launched with a k evenly divisible by the internal block size of 128, and is no larger than 10 of these blocks, the API takes advantage of utilizing pre-allocated memory found in the handle to increase overall performance. This memory can be managed by using the environment variable WORKBUF_TRSM_B_CHNK. When this variable is not set the device memory used for temporary storage will default to 1 MB and may result in chunking, which in turn may reduce performance. Under these circumstances it is recommended that WORKBUF_TRSM_B_CHNK be set to the desired chunk of right hand sides to be used at a time.
(where k is m when HIPBLAS_SIDE_LEFT and is n when HIPBLAS_SIDE_RIGHT)
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] HIPBLAS_SIDE_LEFT: op(A)*X = alpha*B. HIPBLAS_SIDE_RIGHT: X*op(A) = alpha*B.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_N: op(A) = A. HIPBLAS_OP_T: op(A) = A^T. HIPBLAS_OP_C: op(A) = A^H.
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: A is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: A is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of B. m >= 0.
- n (
int) – IN: [int] n specifies the number of columns of B. n >= 0.
- alpha (
Pointer/object) – IN: device pointer or host pointer specifying the scalar alpha. When alpha is &zero then A is not referenced and B need not be set before entry.
- AP (
Pointer/object) – IN: device pointer storing matrix A. of dimension ( lda, k ), where k is m when HIPBLAS_SIDE_LEFT and is n when HIPBLAS_SIDE_RIGHT only the upper/lower triangular part is accessed.
- lda (
int) – IN: [int] lda specifies the first dimension of A. if side = HIPBLAS_SIDE_LEFT, lda >= max( 1, m ), if side = HIPBLAS_SIDE_RIGHT, lda >= max( 1, n ).
- BP (
Pointer/object) – IN,OUT: device pointer storing matrix B.
- ldb (
int) – IN: [int] ldb specifies the first dimension of B. ldb >= max( 1, m ).
- handle (
- hip.hipblas.hipblasDtrsm(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, BP, int ldb)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrsm(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, BP, int ldb)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- BP (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrsm(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, BP, int ldb)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrsm_v2(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, BP, int ldb)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrsm_v2(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, BP, int ldb)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStrsmBatched(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, BP, int ldb, int batchCount)#
BLAS Level 3 API
@{
trsmBatched performs the following batched operation:
op(A_i)*X_i = alpha*B_i or X_i*op(A_i) = alpha*B_i, for i = 1, …, batchCount.
where alpha is a scalar, X and B are batched m by n matrices, A is triangular batched matrix and op(A) is one of
op( A ) = A or op( A ) = A^T or op( A ) = A^H.
Each matrix X_i is overwritten on B_i for i = 1, …, batchCount.
Note about memory allocation: When trsm is launched with a k evenly divisible by the internal block size of 128, and is no larger than 10 of these blocks, the API takes advantage of utilizing pre-allocated memory found in the handle to increase overall performance. This memory can be managed by using the environment variable WORKBUF_TRSM_B_CHNK. When this variable is not set the device memory used for temporary storage will default to 1 MB and may result in chunking, which in turn may reduce performance. Under these circumstances it is recommended that WORKBUF_TRSM_B_CHNK be set to the desired chunk of right hand sides to be used at a time. (where k is m when HIPBLAS_SIDE_LEFT and is n when HIPBLAS_SIDE_RIGHT)
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] HIPBLAS_SIDE_LEFT: op(A)*X = alpha*B. HIPBLAS_SIDE_RIGHT: X*op(A) = alpha*B.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: each A_i is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: each A_i is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_N: op(A) = A. HIPBLAS_OP_T: op(A) = A^T. HIPBLAS_OP_C: op(A) = A^H.
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: each A_i is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: each A_i is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of each B_i. m >= 0.
- n (
int) – IN: [int] n specifies the number of columns of each B_i. n >= 0.
- alpha (
Pointer/object) – IN: device pointer or host pointer specifying the scalar alpha. When alpha is &zero then A is not referenced and B need not be set before entry.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix A_i on the GPU. Matricies are of dimension ( lda, k ), where k is m when HIPBLAS_SIDE_LEFT and is n when HIPBLAS_SIDE_RIGHT only the upper/lower triangular part is accessed.
- lda (
int) – IN: [int] lda specifies the first dimension of each A_i. if side = HIPBLAS_SIDE_LEFT, lda >= max( 1, m ), if side = HIPBLAS_SIDE_RIGHT, lda >= max( 1, n ).
- BP (
Pointer/object) – IN,OUT: device array of device pointers storing each matrix B_i on the GPU.
- ldb (
int) – IN: [int] ldb specifies the first dimension of each B_i. ldb >= max( 1, m ).
- batchCount (
int) – IN: [int] number of trsm operatons in the batch.
- handle (
- hip.hipblas.hipblasDtrsmBatched(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, BP, int ldb, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrsmBatched(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, BP, int ldb, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrsmBatched(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, BP, int ldb, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrsmBatched_v2(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, BP, int ldb, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrsmBatched_v2(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, BP, int ldb, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStrsmStridedBatched(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, int batchCount)#
BLAS Level 3 API
@{
trsmSridedBatched performs the following strided batched operation:
op(A_i)*X_i = alpha*B_i or X_i*op(A_i) = alpha*B_i, for i = 1, …, batchCount.
where alpha is a scalar, X and B are strided batched m by n matrices, A is triangular strided batched matrix and op(A) is one of
op( A ) = A or op( A ) = A^T or op( A ) = A^H.
Each matrix X_i is overwritten on B_i for i = 1, …, batchCount.
Note about memory allocation: When trsm is launched with a k evenly divisible by the internal block size of 128, and is no larger than 10 of these blocks, the API takes advantage of utilizing pre-allocated memory found in the handle to increase overall performance. This memory can be managed by using the environment variable WORKBUF_TRSM_B_CHNK. When this variable is not set the device memory used for temporary storage will default to 1 MB and may result in chunking, which in turn may reduce performance. Under these circumstances it is recommended that WORKBUF_TRSM_B_CHNK be set to the desired chunk of right hand sides to be used at a time. (where k is m when HIPBLAS_SIDE_LEFT and is n when HIPBLAS_SIDE_RIGHT)
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] HIPBLAS_SIDE_LEFT: op(A)*X = alpha*B. HIPBLAS_SIDE_RIGHT: X*op(A) = alpha*B.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: each A_i is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: each A_i is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_N: op(A) = A. HIPBLAS_OP_T: op(A) = A^T. HIPBLAS_OP_C: op(A) = A^H.
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: each A_i is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: each A_i is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of each B_i. m >= 0.
- n (
int) – IN: [int] n specifies the number of columns of each B_i. n >= 0.
- alpha (
Pointer/object) – IN: device pointer or host pointer specifying the scalar alpha. When alpha is &zero then A is not referenced and B need not be set before entry.
- AP (
Pointer/object) – IN: device pointer pointing to the first matrix A_1. of dimension ( lda, k ), where k is m when HIPBLAS_SIDE_LEFT and is n when HIPBLAS_SIDE_RIGHT only the upper/lower triangular part is accessed.
- lda (
int) – IN: [int] lda specifies the first dimension of each A_i. if side = HIPBLAS_SIDE_LEFT, lda >= max( 1, m ), if side = HIPBLAS_SIDE_RIGHT, lda >= max( 1, n ).
- strideA (
int) – IN: [hipblasStride] stride from the start of one A_i matrix to the next A_(i + 1).
- BP (
Pointer/object) – IN,OUT: device pointer pointing to the first matrix B_1.
- ldb (
int) – IN: [int] ldb specifies the first dimension of each B_i. ldb >= max( 1, m ).
- strideB (
int) – IN: [hipblasStride] stride from the start of one B_i matrix to the next B_(i + 1).
- batchCount (
int) – IN: [int] number of trsm operatons in the batch.
- handle (
- hip.hipblas.hipblasDtrsmStridedBatched(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrsmStridedBatched(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasComplex/object): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrsmStridedBatched(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
hipblasDoubleComplex/object): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrsmStridedBatched_v2(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
float2/object): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrsmStridedBatched_v2(handle, side, uplo, transA, diag, int m, int n, alpha, AP, int lda, long strideA, BP, int ldb, long strideB, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
double2/object): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- BP (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasStrtri(handle, uplo, diag, int n, AP, int lda, invA, int ldinvA)#
BLAS Level 3 API
@{
trtri compute the inverse of a matrix A, namely, invA
and write the result into invA;
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’ if HIPBLAS_FILL_MODE_UPPER, the lower part of A is not referenced if HIPBLAS_FILL_MODE_LOWER, the upper part of A is not referenced
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] = ‘HIPBLAS_DIAG_NON_UNIT’, A is non-unit triangular; = ‘HIPBLAS_DIAG_UNIT’, A is unit triangular;
- n (
int) – IN: [int] size of matrix A and invA
- AP (
Pointer/object) – IN: device pointer storing matrix A.
- lda (
int) – IN: [int] specifies the leading dimension of A.
- invA (
Pointer/object) – OUT: device pointer storing matrix invA.
- ldinvA (
int) – IN: [int] specifies the leading dimension of invA.
- handle (
- hip.hipblas.hipblasDtrtri(handle, uplo, diag, int n, AP, int lda, invA, int ldinvA)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCtrtri(handle, uplo, diag, int n, AP, int lda, invA, int ldinvA)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- invA (
hipblasComplex/object): (undocumented)
- ldinvA (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrtri(handle, uplo, diag, int n, AP, int lda, invA, int ldinvA)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- invA (
hipblasDoubleComplex/object): (undocumented)
- ldinvA (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrtri_v2(handle, uplo, diag, int n, AP, int lda, invA, int ldinvA)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZtrtri_v2(handle, uplo, diag, int n, AP, int lda, invA, int ldinvA)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasStrtriBatched(handle, uplo, diag, int n, AP, int lda, invA, int ldinvA, int batchCount)#
BLAS Level 3 API
@{
trtriBatched compute the inverse of A_i and write into invA_i where A_i and invA_i are the i-th matrices in the batch, for i = 1, …, batchCount.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] = ‘HIPBLAS_DIAG_NON_UNIT’, A is non-unit triangular; = ‘HIPBLAS_DIAG_UNIT’, A is unit triangular;
- n (
int) – IN: [int]
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix A_i.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i.
- invA (
Pointer/object) – OUT: device array of device pointers storing the inverse of each matrix A_i. Partial inplace operation is supported, see below. If UPLO = ‘U’, the leading N-by-N upper triangular part of the invA will store the inverse of the upper triangular matrix, and the strictly lower triangular part of invA is cleared. If UPLO = ‘L’, the leading N-by-N lower triangular part of the invA will store the inverse of the lower triangular matrix, and the strictly upper triangular part of invA is cleared.
- ldinvA (
int) – IN: [int] specifies the leading dimension of each invA_i.
- batchCount (
int) – IN: [int] numbers of matrices in the batch
- handle (
- hip.hipblas.hipblasDtrtriBatched(handle, uplo, diag, int n, AP, int lda, invA, int ldinvA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- hip.hipblas.hipblasCtrtriBatched(handle, uplo, diag, int n, AP, int lda, invA, int ldinvA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- hip.hipblas.hipblasZtrtriBatched(handle, uplo, diag, int n, AP, int lda, invA, int ldinvA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- hip.hipblas.hipblasCtrtriBatched_v2(handle, uplo, diag, int n, AP, int lda, invA, int ldinvA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- hip.hipblas.hipblasZtrtriBatched_v2(handle, uplo, diag, int n, AP, int lda, invA, int ldinvA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- hip.hipblas.hipblasStrtriStridedBatched(handle, uplo, diag, int n, AP, int lda, long strideA, invA, int ldinvA, long stride_invA, int batchCount)#
BLAS Level 3 API
@{
trtriStridedBatched compute the inverse of A_i and write into invA_i where A_i and invA_i are the i-th matrices in the batch, for i = 1, …, batchCount
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] specifies whether the upper ‘HIPBLAS_FILL_MODE_UPPER’ or lower ‘HIPBLAS_FILL_MODE_LOWER’
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] = ‘HIPBLAS_DIAG_NON_UNIT’, A is non-unit triangular; = ‘HIPBLAS_DIAG_UNIT’, A is unit triangular;
- n (
int) – IN: [int]
- AP (
Pointer/object) – IN: device pointer pointing to address of first matrix A_1.
- lda (
int) – IN: [int] specifies the leading dimension of each A.
- strideA (
int) – IN: [hipblasStride] “batch stride a”: stride from the start of one A_i matrix to the next A_(i + 1).
- invA (
Pointer/object) – OUT: device pointer storing the inverses of each matrix A_i. Partial inplace operation is supported, see below. If UPLO = ‘U’, the leading N-by-N upper triangular part of the invA will store the inverse of the upper triangular matrix, and the strictly lower triangular part of invA is cleared. If UPLO = ‘L’, the leading N-by-N lower triangular part of the invA will store the inverse of the lower triangular matrix, and the strictly upper triangular part of invA is cleared.
- ldinvA (
int) – IN: [int] specifies the leading dimension of each invA_i.
- stride_invA (
int) – IN: [hipblasStride] “batch stride invA”: stride from the start of one invA_i matrix to the next invA_(i + 1).
- batchCount (
int) – IN: [int] numbers of matrices in the batch
- handle (
- hip.hipblas.hipblasDtrtriStridedBatched(handle, uplo, diag, int n, AP, int lda, long strideA, invA, int ldinvA, long stride_invA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- invA (
Pointer/object): (undocumented)
- ldinvA (
int): (undocumented)
- stride_invA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrtriStridedBatched(handle, uplo, diag, int n, AP, int lda, long strideA, invA, int ldinvA, long stride_invA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- invA (
hipblasComplex/object): (undocumented)
- ldinvA (
int): (undocumented)
- stride_invA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrtriStridedBatched(handle, uplo, diag, int n, AP, int lda, long strideA, invA, int ldinvA, long stride_invA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- invA (
hipblasDoubleComplex/object): (undocumented)
- ldinvA (
int): (undocumented)
- stride_invA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCtrtriStridedBatched_v2(handle, uplo, diag, int n, AP, int lda, long strideA, invA, int ldinvA, long stride_invA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- invA (
float2/object): (undocumented)
- ldinvA (
int): (undocumented)
- stride_invA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZtrtriStridedBatched_v2(handle, uplo, diag, int n, AP, int lda, long strideA, invA, int ldinvA, long stride_invA, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- n (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- invA (
double2/object): (undocumented)
- ldinvA (
int): (undocumented)
- stride_invA (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSdgmm(handle, side, int m, int n, AP, int lda, x, int incx, CP, int ldc)#
BLAS Level 3 API
@{
dgmm performs one of the matrix-matrix operations
- C = A * diag(x) if side == HIPBLAS_SIDE_RIGHT
C = diag(x) * A if side == HIPBLAS_SIDE_LEFT
where C and A are m by n dimensional matrices. diag( x ) is a diagonal matrix and x is vector of dimension n if side == HIPBLAS_SIDE_RIGHT and dimension m if side == HIPBLAS_SIDE_LEFT.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : s,d,c,z
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] specifies the side of diag(x)
- m (
int) – IN: [int] matrix dimension m.
- n (
int) – IN: [int] matrix dimension n.
- AP (
Pointer/object) – IN: device pointer storing matrix A.
- lda (
int) – IN: [int] specifies the leading dimension of A.
- x (
Pointer/object) – IN: device pointer storing vector x.
- incx (
int) – IN: [int] specifies the increment between values of x
- CP (
Pointer/object) – IN,OUT: device pointer storing matrix C.
- ldc (
int) – IN: [int] specifies the leading dimension of C.
- handle (
- hip.hipblas.hipblasDdgmm(handle, side, int m, int n, AP, int lda, x, int incx, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCdgmm(handle, side, int m, int n, AP, int lda, x, int incx, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- CP (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZdgmm(handle, side, int m, int n, AP, int lda, x, int incx, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCdgmm_v2(handle, side, int m, int n, AP, int lda, x, int incx, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZdgmm_v2(handle, side, int m, int n, AP, int lda, x, int incx, CP, int ldc)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSdgmmBatched(handle, side, int m, int n, AP, int lda, x, int incx, CP, int ldc, int batchCount)#
BLAS Level 3 API
@{
dgmmBatched performs one of the batched matrix-matrix operations
- C_i = A_i * diag(x_i) for i = 0, 1, … batchCount-1 if side == HIPBLAS_SIDE_RIGHT
C_i = diag(x_i) * A_i for i = 0, 1, … batchCount-1 if side == HIPBLAS_SIDE_LEFT
where C_i and A_i are m by n dimensional matrices. diag(x_i) is a diagonal matrix and x_i is vector of dimension n if side == HIPBLAS_SIDE_RIGHT and dimension m if side == HIPBLAS_SIDE_LEFT.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] specifies the side of diag(x)
- m (
int) – IN: [int] matrix dimension m.
- n (
int) – IN: [int] matrix dimension n.
- AP (
Pointer/object) – IN: device array of device pointers storing each matrix A_i on the GPU. Each A_i is of dimension ( lda, n )
- lda (
int) – IN: [int] specifies the leading dimension of A_i.
- x (
ListOfPointer/object) – IN: device array of device pointers storing each vector x_i on the GPU. Each x_i is of dimension n if side == HIPBLAS_SIDE_RIGHT and dimension m if side == HIPBLAS_SIDE_LEFT
- incx (
int) – IN: [int] specifies the increment between values of x_i
- CP (
Pointer/object) – IN,OUT: device array of device pointers storing each matrix C_i on the GPU. Each C_i is of dimension ( ldc, n ).
- ldc (
int) – IN: [int] specifies the leading dimension of C_i.
- batchCount (
int) – IN: [int] number of instances in the batch.
- handle (
- hip.hipblas.hipblasDdgmmBatched(handle, side, int m, int n, AP, int lda, x, int incx, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
ListOfPointer/object): (undocumented)
- incx (
int): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCdgmmBatched(handle, side, int m, int n, AP, int lda, x, int incx, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZdgmmBatched(handle, side, int m, int n, AP, int lda, x, int incx, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCdgmmBatched_v2(handle, side, int m, int n, AP, int lda, x, int incx, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZdgmmBatched_v2(handle, side, int m, int n, AP, int lda, x, int incx, CP, int ldc, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSdgmmStridedBatched(handle, side, int m, int n, AP, int lda, long strideA, x, int incx, long stridex, CP, int ldc, long strideC, int batchCount)#
BLAS Level 3 API
@{
dgmmStridedBatched performs one of the batched matrix-matrix operations
- C_i = A_i * diag(x_i) if side == HIPBLAS_SIDE_RIGHT for i = 0, 1, … batchCount-1
C_i = diag(x_i) * A_i if side == HIPBLAS_SIDE_LEFT for i = 0, 1, … batchCount-1
where C_i and A_i are m by n dimensional matrices. diag(x_i) is a diagonal matrix and x_i is vector of dimension n if side == HIPBLAS_SIDE_RIGHT and dimension m if side == HIPBLAS_SIDE_LEFT.
Supported precisions in rocBLAS : s,d,c,z
Supported precisions in cuBLAS : No support
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] specifies the side of diag(x)
- m (
int) – IN: [int] matrix dimension m.
- n (
int) – IN: [int] matrix dimension n.
- AP (
Pointer/object) – IN: device pointer to the first matrix A_0 on the GPU. Each A_i is of dimension ( lda, n )
- lda (
int) – IN: [int] specifies the leading dimension of A.
- strideA (
int) – IN: [hipblasStride] stride from the start of one matrix (A_i) and the next one (A_i+1)
- x (
Pointer/object) – IN: pointer to the first vector x_0 on the GPU. Each x_i is of dimension n if side == HIPBLAS_SIDE_RIGHT and dimension m if side == HIPBLAS_SIDE_LEFT
- incx (
int) – IN: [int] specifies the increment between values of x
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector(x_i) and the next one (x_i+1)
- CP (
Pointer/object) – IN,OUT: device pointer to the first matrix C_0 on the GPU. Each C_i is of dimension ( ldc, n ).
- ldc (
int) – IN: [int] specifies the leading dimension of C.
- strideC (
int) – IN: [hipblasStride] stride from the start of one matrix (C_i) and the next one (C_i+1)
- batchCount (
int) – IN: [int] number of instances i in the batch.
- handle (
- hip.hipblas.hipblasDdgmmStridedBatched(handle, side, int m, int n, AP, int lda, long strideA, x, int incx, long stridex, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- AP (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- CP (
Pointer/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCdgmmStridedBatched(handle, side, int m, int n, AP, int lda, long strideA, x, int incx, long stridex, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- AP (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- CP (
hipblasComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZdgmmStridedBatched(handle, side, int m, int n, AP, int lda, long strideA, x, int incx, long stridex, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- AP (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
hipblasDoubleComplex/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- CP (
hipblasDoubleComplex/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCdgmmStridedBatched_v2(handle, side, int m, int n, AP, int lda, long strideA, x, int incx, long stridex, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- AP (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
float2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- CP (
float2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZdgmmStridedBatched_v2(handle, side, int m, int n, AP, int lda, long strideA, x, int incx, long stridex, CP, int ldc, long strideC, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- AP (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- x (
double2/object): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- CP (
double2/object): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSgetrf(handle, int n, A, int lda, ipiv, info)#
SOLVER API
@{
getrf computes the LU factorization of a general n-by-n matrix A using partial pivoting with row interchanges. The LU factorization can be done without pivoting if ipiv is passed as a nullptr.
In the case that ipiv is not null, the factorization has the form:
\[A = PLU\]where P is a permutation matrix, L is lower triangular with unit diagonal elements, and U is upper triangular.
In the case that ipiv is null, the factorization is done without pivoting:
\[A = LU\]- Args:
- handle (
Pointer/object) – IN: hipblasHandle_t.
- n (
int) – IN: int. n >= 0.
The number of columns and rows of the matrix A.
- A (
Pointer/object) – INOUT: pointer to type. Array on the GPU of dimension lda*n.
On entry, the n-by-n matrix A to be factored. On exit, the factors L and U from the factorization. The unit diagonal elements of L are not stored.
- lda (
int) – IN: int. lda >= n.
Specifies the leading dimension of A.
- ipiv (
ListOfInt/object) – OUT: pointer to int. Array on the GPU of dimension n.
The vector of pivot indices. Elements of ipiv are 1-based indices. For 1 <= i <= n, the row i of the matrix was interchanged with row ipiv[i]. Matrix P of the factorization can be derived from ipiv. The factorization here can be done without pivoting if ipiv is passed in as a nullptr.
- info (
ListOfInt/object) – OUT: pointer to a int on the GPU.
If info = 0, successful exit. If info = j > 0, U is singular. U[j,j] is the first zero pivot.
- handle (
- hip.hipblas.hipblasDgetrf(handle, int n, A, int lda, ipiv, info)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCgetrf(handle, int n, A, int lda, ipiv, info)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZgetrf(handle, int n, A, int lda, ipiv, info)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCgetrf_v2(handle, int n, A, int lda, ipiv, info)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZgetrf_v2(handle, int n, A, int lda, ipiv, info)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSgetrfBatched(handle, int n, A, int lda, ipiv, info, int batchCount)#
SOLVER API
@{
getrfBatched computes the LU factorization of a batch of general n-by-n matrices using partial pivoting with row interchanges. The LU factorization can be done without pivoting if ipiv is passed as a nullptr.
In the case that ipiv is not null, the factorization of matrix \(A_i\) in the batch has the form:
\[A_i = P_iL_iU_i\]where \(P_i\) is a permutation matrix, \(L_i\) is lower triangular with unit diagonal elements, and \(U_i\) is upper triangular.
In the case that ipiv is null, the factorization is done without pivoting:
\[A_i = L_iU_i\]- Args:
- handle (
Pointer/object) – IN: hipblasHandle_t.
- n (
int) – IN: int. n >= 0.
The number of columns and rows of all matrices A_i in the batch.
- A (
ListOfPointer/object) – INOUT: array of pointers to type. Each pointer points to an array on the GPU of dimension lda*n.
On entry, the n-by-n matrices A_i to be factored. On exit, the factors L_i and U_i from the factorizations. The unit diagonal elements of L_i are not stored.
- lda (
int) – IN: int. lda >= n.
Specifies the leading dimension of matrices A_i.
- ipiv (
ListOfInt/object) – OUT: pointer to int. Array on the GPU.
Contains the vectors of pivot indices ipiv_i (corresponding to A_i). Dimension of ipiv_i is n. Elements of ipiv_i are 1-based indices. For each instance A_i in the batch and for 1 <= j <= n, the row j of the matrix A_i was interchanged with row ipiv_i[j]. Matrix P_i of the factorization can be derived from ipiv_i. The factorization here can be done without pivoting if ipiv is passed in as a nullptr.
- info (
ListOfInt/object) – OUT: pointer to int. Array of batchCount integers on the GPU.
If info[i] = 0, successful exit for factorization of A_i. If info[i] = j > 0, U_i is singular. U_i[j,j] is the first zero pivot.
- batchCount (
int) – IN: int. batchCount >= 0.
Number of matrices in the batch.
- handle (
- hip.hipblas.hipblasDgetrfBatched(handle, int n, A, int lda, ipiv, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCgetrfBatched(handle, int n, A, int lda, ipiv, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZgetrfBatched(handle, int n, A, int lda, ipiv, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCgetrfBatched_v2(handle, int n, A, int lda, ipiv, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZgetrfBatched_v2(handle, int n, A, int lda, ipiv, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSgetrfStridedBatched(handle, int n, A, int lda, long strideA, ipiv, long strideP, info, int batchCount)#
SOLVER API
@{
getrfStridedBatched computes the LU factorization of a batch of general n-by-n matrices using partial pivoting with row interchanges. The LU factorization can be done without pivoting if ipiv is passed as a nullptr.
In the case that ipiv is not null, the factorization of matrix \(A_i\) in the batch has the form:
\[A_i = P_iL_iU_i\]where \(P_i\) is a permutation matrix, \(L_i\) is lower triangular with unit diagonal elements, and \(U_i\) is upper triangular.
In the case that ipiv is null, the factorization is done without pivoting:
\[A_i = L_iU_i\]- Args:
- handle (
Pointer/object) – IN: hipblasHandle_t.
- n (
int) – IN: int. n >= 0.
The number of columns and rows of all matrices A_i in the batch.
- A (
Pointer/object) – INOUT: pointer to type. Array on the GPU (the size depends on the value of strideA).
On entry, the n-by-n matrices A_i to be factored. On exit, the factors L_i and U_i from the factorization. The unit diagonal elements of L_i are not stored.
- lda (
int) – IN: int. lda >= n.
Specifies the leading dimension of matrices A_i.
- strideA (
int) – IN: hipblasStride.
Stride from the start of one matrix A_i to the next one A_(i+1). There is no restriction for the value of strideA. Normal use case is strideA >= lda*n
- ipiv (
ListOfInt/object) – OUT: pointer to int. Array on the GPU (the size depends on the value of strideP).
Contains the vectors of pivots indices ipiv_i (corresponding to A_i). Dimension of ipiv_i is n. Elements of ipiv_i are 1-based indices. For each instance A_i in the batch and for 1 <= j <= n, the row j of the matrix A_i was interchanged with row ipiv_i[j]. Matrix P_i of the factorization can be derived from ipiv_i. The factorization here can be done without pivoting if ipiv is passed in as a nullptr.
- strideP (
int) – IN: hipblasStride.
Stride from the start of one vector ipiv_i to the next one ipiv_(i+1). There is no restriction for the value of strideP. Normal use case is strideP >= n.
- info (
ListOfInt/object) – OUT: pointer to int. Array of batchCount integers on the GPU.
If info[i] = 0, successful exit for factorization of A_i. If info[i] = j > 0, U_i is singular. U_i[j,j] is the first zero pivot.
- batchCount (
int) – IN: int. batchCount >= 0.
Number of matrices in the batch.
- handle (
- hip.hipblas.hipblasDgetrfStridedBatched(handle, int n, A, int lda, long strideA, ipiv, long strideP, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCgetrfStridedBatched(handle, int n, A, int lda, long strideA, ipiv, long strideP, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- hip.hipblas.hipblasZgetrfStridedBatched(handle, int n, A, int lda, long strideA, ipiv, long strideP, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- A (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- strideP (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgetrfStridedBatched_v2(handle, int n, A, int lda, long strideA, ipiv, long strideP, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZgetrfStridedBatched_v2(handle, int n, A, int lda, long strideA, ipiv, long strideP, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSgetrs(handle, trans, int n, int nrhs, A, int lda, ipiv, B, int ldb, info)#
SOLVER API
@{
getrs solves a system of n linear equations on n variables in its factorized form.
It solves one of the following systems, depending on the value of trans:
\[\begin{split}\begin{array}{cl} A X = B & \: \text{not transposed,}\\ A^T X = B & \: \text{transposed, or}\\ A^H X = B & \: \text{conjugate transposed.} \end{array}\end{split}\]- Args:
- handle (
Pointer/object) – IN: hipblasHandle_t.
- trans (
hipblasOperation_t) – IN: hipblasOperation_t.
Specifies the form of the system of equations.
- n (
int) – IN: int. n >= 0.
The order of the system, i.e. the number of columns and rows of A.
- nrhs (
int) – IN: int. nrhs >= 0.
The number of right hand sides, i.e., the number of columns of the matrix B.
- A (
Pointer/object) – IN: pointer to type. Array on the GPU of dimension lda*n.
The factors L and U of the factorization A = P*L*U returned by
hipblasSgetrf.- lda (
int) – IN: int. lda >= n.
The leading dimension of A.
- ipiv (
ListOfInt/object) – IN: pointer to int. Array on the GPU of dimension n.
The pivot indices returned by
hipblasSgetrf.- B (
Pointer/object) – IN,OUT: pointer to type. Array on the GPU of dimension ldb*nrhs.
On entry, the right hand side matrix B. On exit, the solution matrix X.
- ldb (
int) – IN: int. ldb >= n.
The leading dimension of B.
- info (
ListOfInt/object) – OUT: pointer to a int on the host.
If info = 0, successful exit. If info = j < 0, the argument at position -j is invalid.
- handle (
- hip.hipblas.hipblasDgetrs(handle, trans, int n, int nrhs, A, int lda, ipiv, B, int ldb, info)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- handle (
- hip.hipblas.hipblasCgetrs(handle, trans, int n, int nrhs, A, int lda, ipiv, B, int ldb, info)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- B (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- handle (
- hip.hipblas.hipblasZgetrs(handle, trans, int n, int nrhs, A, int lda, ipiv, B, int ldb, info)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- B (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- handle (
- hip.hipblas.hipblasCgetrs_v2(handle, trans, int n, int nrhs, A, int lda, ipiv, B, int ldb, info)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
float2/object): (undocumented)
- lda (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- B (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- handle (
- hip.hipblas.hipblasZgetrs_v2(handle, trans, int n, int nrhs, A, int lda, ipiv, B, int ldb, info)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
double2/object): (undocumented)
- lda (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- B (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- handle (
- hip.hipblas.hipblasSgetrsBatched(handle, trans, int n, int nrhs, A, int lda, ipiv, B, int ldb, info, int batchCount)#
SOLVER API
@{
getrsBatched solves a batch of systems of n linear equations on n variables in its factorized forms.
For each instance i in the batch, it solves one of the following systems, depending on the value of trans:
\[\begin{split}\begin{array}{cl} A_i X_i = B_i & \: \text{not transposed,}\\ A_i^T X_i = B_i & \: \text{transposed, or}\\ A_i^H X_i = B_i & \: \text{conjugate transposed.} \end{array}\end{split}\]- Args:
- handle (
Pointer/object) – IN: hipblasHandle_t.
- trans (
hipblasOperation_t) – IN: hipblasOperation_t.
Specifies the form of the system of equations of each instance in the batch.
- n (
int) – IN: int. n >= 0.
The order of the system, i.e. the number of columns and rows of all A_i matrices.
- nrhs (
int) – IN: int. nrhs >= 0.
The number of right hand sides, i.e., the number of columns of all the matrices B_i.
- A (
ListOfPointer/object) – IN: Array of pointers to type. Each pointer points to an array on the GPU of dimension lda*n.
The factors L_i and U_i of the factorization A_i = P_i*L_i*U_i returned by
hipblasSgetrfBatched.- lda (
int) – IN: int. lda >= n.
The leading dimension of matrices A_i.
- ipiv (
ListOfInt/object) – IN: pointer to int. Array on the GPU.
Contains the vectors ipiv_i of pivot indices returned by
hipblasSgetrfBatched.- B (
ListOfPointer/object) – IN,OUT: Array of pointers to type. Each pointer points to an array on the GPU of dimension ldb*nrhs.
On entry, the right hand side matrices B_i. On exit, the solution matrix X_i of each system in the batch.
- ldb (
int) – IN: int. ldb >= n.
The leading dimension of matrices B_i.
- info (
ListOfInt/object) – OUT: pointer to a int on the host.
If info = 0, successful exit. If info = j < 0, the argument at position -j is invalid.
- batchCount (
int) – IN: int. batchCount >= 0.
Number of instances (systems) in the batch.
- handle (
- hip.hipblas.hipblasDgetrsBatched(handle, trans, int n, int nrhs, A, int lda, ipiv, B, int ldb, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
ListOfPointer/object): (undocumented)
- lda (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- B (
ListOfPointer/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgetrsBatched(handle, trans, int n, int nrhs, A, int lda, ipiv, B, int ldb, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgetrsBatched(handle, trans, int n, int nrhs, A, int lda, ipiv, B, int ldb, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgetrsBatched_v2(handle, trans, int n, int nrhs, A, int lda, ipiv, B, int ldb, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgetrsBatched_v2(handle, trans, int n, int nrhs, A, int lda, ipiv, B, int ldb, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSgetrsStridedBatched(handle, trans, int n, int nrhs, A, int lda, long strideA, ipiv, long strideP, B, int ldb, long strideB, info, int batchCount)#
SOLVER API
@{
getrsStridedBatched solves a batch of systems of n linear equations on n variables in its factorized forms.
For each instance i in the batch, it solves one of the following systems, depending on the value of trans:
\[\begin{split}\begin{array}{cl} A_i X_i = B_i & \: \text{not transposed,}\\ A_i^T X_i = B_i & \: \text{transposed, or}\\ A_i^H X_i = B_i & \: \text{conjugate transposed.} \end{array}\end{split}\]- Args:
- handle (
Pointer/object) – IN: hipblasHandle_t.
- trans (
hipblasOperation_t) – IN: hipblasOperation_t.
Specifies the form of the system of equations of each instance in the batch.
- n (
int) – IN: int. n >= 0.
The order of the system, i.e. the number of columns and rows of all A_i matrices.
- nrhs (
int) – IN: int. nrhs >= 0.
The number of right hand sides, i.e., the number of columns of all the matrices B_i.
- A (
Pointer/object) – IN: pointer to type. Array on the GPU (the size depends on the value of strideA).
The factors L_i and U_i of the factorization A_i = P_i*L_i*U_i returned by
hipblasSgetrfStridedBatched.- lda (
int) – IN: int. lda >= n.
The leading dimension of matrices A_i.
- strideA (
int) – IN: hipblasStride.
Stride from the start of one matrix A_i to the next one A_(i+1). There is no restriction for the value of strideA. Normal use case is strideA >= lda*n.
- ipiv (
ListOfInt/object) – IN: pointer to int. Array on the GPU (the size depends on the value of strideP).
Contains the vectors ipiv_i of pivot indices returned by
hipblasSgetrfStridedBatched.- strideP (
int) – IN: hipblasStride.
Stride from the start of one vector ipiv_i to the next one ipiv_(i+1). There is no restriction for the value of strideP. Normal use case is strideP >= n.
- B (
Pointer/object) – IN,OUT: pointer to type. Array on the GPU (size depends on the value of strideB).
On entry, the right hand side matrices B_i. On exit, the solution matrix X_i of each system in the batch.
- ldb (
int) – IN: int. ldb >= n.
The leading dimension of matrices B_i.
- strideB (
int) – IN: hipblasStride.
Stride from the start of one matrix B_i to the next one B_(i+1). There is no restriction for the value of strideB. Normal use case is strideB >= ldb*nrhs.
- info (
ListOfInt/object) – OUT: pointer to a int on the host.
If info = 0, successful exit. If info = j < 0, the argument at position -j is invalid.
- batchCount (
int) – IN: int. batchCount >= 0.
Number of instances (systems) in the batch.
- handle (
- hip.hipblas.hipblasDgetrsStridedBatched(handle, trans, int n, int nrhs, A, int lda, long strideA, ipiv, long strideP, B, int ldb, long strideB, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- strideP (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgetrsStridedBatched(handle, trans, int n, int nrhs, A, int lda, long strideA, ipiv, long strideP, B, int ldb, long strideB, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- strideP (
int): (undocumented)
- B (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgetrsStridedBatched(handle, trans, int n, int nrhs, A, int lda, long strideA, ipiv, long strideP, B, int ldb, long strideB, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- strideP (
int): (undocumented)
- B (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgetrsStridedBatched_v2(handle, trans, int n, int nrhs, A, int lda, long strideA, ipiv, long strideP, B, int ldb, long strideB, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- strideP (
int): (undocumented)
- B (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgetrsStridedBatched_v2(handle, trans, int n, int nrhs, A, int lda, long strideA, ipiv, long strideP, B, int ldb, long strideB, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- strideP (
int): (undocumented)
- B (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSgetriBatched(handle, int n, A, int lda, ipiv, C, int ldc, info, int batchCount)#
SOLVER API
@{
getriBatched computes the inverse \(C_i = A_i^{-1}\) of a batch of general n-by-n matrices \(A_i\).
The inverse is computed by solving the linear system
\[A_i C_i = I\]- Args:
- handle (
Pointer/object) – IN: hipblasHandle_t.
- n (
int) – IN: int. n >= 0.
The number of rows and columns of all matrices A_i in the batch.
- A (
ListOfPointer/object) – IN: array of pointers to type. Each pointer points to an array on the GPU of dimension lda*n.
The factors L_i and U_i of the factorization A_i = P_i*L_i*U_i returned by
hipblasSgetrfBatched.- lda (
int) – IN: int. lda >= n.
Specifies the leading dimension of matrices A_i.
- ipiv (
ListOfInt/object) – IN: pointer to int. Array on the GPU (the size depends on the value of strideP).
The pivot indices returned by
hipblasSgetrfBatched. ipiv can be passed in as a nullptr, this will assume that getrfBatched was called without partial pivoting.- C (
ListOfPointer/object) – OUT: array of pointers to type. Each pointer points to an array on the GPU of dimension ldc*n.
If info[i] = 0, the inverse of matrices A_i. Otherwise, undefined.
- ldc (
int) – IN: int. ldc >= n.
Specifies the leading dimension of C_i.
- info (
ListOfInt/object) – OUT: pointer to int. Array of batchCount integers on the GPU.
If info[i] = 0, successful exit for inversion of A_i. If info[i] = j > 0, U_i is singular. U_i[j,j] is the first zero pivot.
- batchCount (
int) – IN: int. batchCount >= 0.
Number of matrices in the batch.
- handle (
- hip.hipblas.hipblasDgetriBatched(handle, int n, A, int lda, ipiv, C, int ldc, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- A (
ListOfPointer/object): (undocumented)
- lda (
int): (undocumented)
- ipiv (
ListOfInt/object): (undocumented)
- C (
ListOfPointer/object): (undocumented)
- ldc (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgetriBatched(handle, int n, A, int lda, ipiv, C, int ldc, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZgetriBatched(handle, int n, A, int lda, ipiv, C, int ldc, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCgetriBatched_v2(handle, int n, A, int lda, ipiv, C, int ldc, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZgetriBatched_v2(handle, int n, A, int lda, ipiv, C, int ldc, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSgels(handle, trans, int m, int n, int nrhs, A, int lda, B, int ldb, info, deviceInfo)#
GELS solves an overdetermined (or underdetermined) linear system defined by an m-by-n matrix A, and a corresponding matrix B, using the QR factorization computed by
hipblasSgeqrf(or the LQ factorization computed by “GELQF”).@{
Depending on the value of trans, the problem solved by this function is either of the form
\[\begin{split}\begin{array}{cl} A X = B & \: \text{not transposed, or}\\ A' X = B & \: \text{transposed if real, or conjugate transposed if complex} \end{array}\end{split}\]If m >= n (or m < n in the case of transpose/conjugate transpose), the system is overdetermined and a least-squares solution approximating X is found by minimizing
\[|| B - A X || \quad \text{(or} \: || B - A' X ||\text{)}\]- Args:
- handle (
Pointer/object) – IN: hipblasHandle_t.
- trans (
hipblasOperation_t) – IN: hipblasOperation_t.
Specifies the form of the system of equations.
- m (
int) – IN: int. m >= 0.
The number of rows of matrix A.
- n (
int) – IN: int. n >= 0.
The number of columns of matrix A.
- nrhs (
int) – IN: int. nrhs >= 0.
The number of columns of matrices B and X; i.e., the columns on the right hand side.
- A (
Pointer/object) – INOUT: pointer to type. Array on the GPU of dimension lda*n.
On entry, the matrix A. On exit, the QR (or LQ) factorization of A as returned by “GEQRF” (or “GELQF”).
- lda (
int) – IN: int. lda >= m.
Specifies the leading dimension of matrix A.
- B (
Pointer/object) – INOUT: pointer to type. Array on the GPU of dimension ldb*nrhs.
On entry, the matrix B. On exit, when info = 0, B is overwritten by the solution vectors (and the residuals in the overdetermined cases) stored as columns.
- ldb (
int) – IN: int. ldb >= max(m,n).
Specifies the leading dimension of matrix B.
- info (
ListOfInt/object) – OUT: pointer to an int on the host.
If info = 0, successful exit. If info = j < 0, the argument at position -j is invalid.
- deviceInfo (
ListOfInt/object) – OUT: pointer to int on the GPU.
If info = 0, successful exit. If info = i > 0, the solution could not be computed because input matrix A is rank deficient; the i-th diagonal element of its triangular factor is zero.
- handle (
- hip.hipblas.hipblasDgels(handle, trans, int m, int n, int nrhs, A, int lda, B, int ldb, info, deviceInfo)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- deviceInfo (
ListOfInt/object): (undocumented)
- handle (
- hip.hipblas.hipblasCgels(handle, trans, int m, int n, int nrhs, A, int lda, B, int ldb, info, deviceInfo)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- B (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- deviceInfo (
ListOfInt/object): (undocumented)
- handle (
- hip.hipblas.hipblasZgels(handle, trans, int m, int n, int nrhs, A, int lda, B, int ldb, info, deviceInfo)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- B (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- deviceInfo (
ListOfInt/object): (undocumented)
- handle (
- hip.hipblas.hipblasCgels_v2(handle, trans, int m, int n, int nrhs, A, int lda, B, int ldb, info, deviceInfo)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
float2/object): (undocumented)
- lda (
int): (undocumented)
- B (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- deviceInfo (
ListOfInt/object): (undocumented)
- handle (
- hip.hipblas.hipblasZgels_v2(handle, trans, int m, int n, int nrhs, A, int lda, B, int ldb, info, deviceInfo)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
double2/object): (undocumented)
- lda (
int): (undocumented)
- B (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- deviceInfo (
ListOfInt/object): (undocumented)
- handle (
- hip.hipblas.hipblasSgelsBatched(handle, trans, int m, int n, int nrhs, A, int lda, B, int ldb, info, deviceInfo, int batchCount)#
gelsBatched solves a batch of overdetermined (or underdetermined) linear systems defined by a set of m-by-n matrices \(A_j\), and corresponding matrices \(B_j\), using the QR factorizations computed by “GEQRF_BATCHED” (or the LQ factorizations computed by “GELQF_BATCHED”).
@{
For each instance in the batch, depending on the value of trans, the problem solved by this function is either of the form
\[\begin{split}\begin{array}{cl} A_j X_j = B_j & \: \text{not transposed, or}\\ A_j' X_j = B_j & \: \text{transposed if real, or conjugate transposed if complex} \end{array}\end{split}\]If m >= n (or m < n in the case of transpose/conjugate transpose), the system is overdetermined and a least-squares solution approximating X_j is found by minimizing
\[|| B_j - A_j X_j || \quad \text{(or} \: || B_j - A_j' X_j ||\text{)}\]- Args:
- handle (
Pointer/object) – IN: hipblasHandle_t.
- trans (
hipblasOperation_t) – IN: hipblasOperation_t.
Specifies the form of the system of equations.
- m (
int) – IN: int. m >= 0.
The number of rows of all matrices A_j in the batch.
- n (
int) – IN: int. n >= 0.
The number of columns of all matrices A_j in the batch.
- nrhs (
int) – IN: int. nrhs >= 0.
The number of columns of all matrices B_j and X_j in the batch; i.e., the columns on the right hand side.
- A (
ListOfPointer/object) – INOUT: array of pointer to type. Each pointer points to an array on the GPU of dimension lda*n.
On entry, the matrices A_j. On exit, the QR (or LQ) factorizations of A_j as returned by “GEQRF_BATCHED” (or “GELQF_BATCHED”).
- lda (
int) – IN: int. lda >= m.
Specifies the leading dimension of matrices A_j.
- B (
ListOfPointer/object) – INOUT: array of pointer to type. Each pointer points to an array on the GPU of dimension ldb*nrhs.
On entry, the matrices B_j. On exit, when info[j] = 0, B_j is overwritten by the solution vectors (and the residuals in the overdetermined cases) stored as columns.
- ldb (
int) – IN: int. ldb >= max(m,n).
Specifies the leading dimension of matrices B_j.
- info (
ListOfInt/object) – OUT: pointer to an int on the host.
If info = 0, successful exit. If info = j < 0, the argument at position -j is invalid.
- deviceInfo (
ListOfInt/object) – OUT: pointer to int. Array of batchCount integers on the GPU.
If deviceInfo[j] = 0, successful exit for solution of A_j. If deviceInfo[j] = i > 0, the solution of A_j could not be computed because input matrix A_j is rank deficient; the i-th diagonal element of its triangular factor is zero.
- batchCount (
int) – IN: int. batchCount >= 0.
Number of matrices in the batch.
- handle (
- hip.hipblas.hipblasDgelsBatched(handle, trans, int m, int n, int nrhs, A, int lda, B, int ldb, info, deviceInfo, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
ListOfPointer/object): (undocumented)
- lda (
int): (undocumented)
- B (
ListOfPointer/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- deviceInfo (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgelsBatched(handle, trans, int m, int n, int nrhs, A, int lda, B, int ldb, info, deviceInfo, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- deviceInfo (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgelsBatched(handle, trans, int m, int n, int nrhs, A, int lda, B, int ldb, info, deviceInfo, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- deviceInfo (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgelsBatched_v2(handle, trans, int m, int n, int nrhs, A, int lda, B, int ldb, info, deviceInfo, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- deviceInfo (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgelsBatched_v2(handle, trans, int m, int n, int nrhs, A, int lda, B, int ldb, info, deviceInfo, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- deviceInfo (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSgelsStridedBatched(handle, trans, int m, int n, int nrhs, A, int lda, long strideA, B, int ldb, long strideB, info, deviceInfo, int batchCount)#
gelsStridedBatched solves a batch of overdetermined (or underdetermined) linear systems defined by a set of m-by-n matrices \(A_j\), and corresponding matrices \(B_j\), using the QR factorizations computed by “GEQRF_STRIDED_BATCHED” (or the LQ factorizations computed by “GELQF_STRIDED_BATCHED”).
@{
For each instance in the batch, depending on the value of trans, the problem solved by this function is either of the form
\[\begin{split}\begin{array}{cl} A_j X_j = B_j & \: \text{not transposed, or}\\ A_j' X_j = B_j & \: \text{transposed if real, or conjugate transposed if complex} \end{array}\end{split}\]If m >= n (or m < n in the case of transpose/conjugate transpose), the system is overdetermined and a least-squares solution approximating X_j is found by minimizing
\[|| B_j - A_j X_j || \quad \text{(or} \: || B_j - A_j' X_j ||\text{)}\]- Args:
- handle (
Pointer/object) – IN: hipblasHandle_t.
- trans (
hipblasOperation_t) – IN: hipblasOperation_t.
Specifies the form of the system of equations.
- m (
int) – IN: int. m >= 0.
The number of rows of all matrices A_j in the batch.
- n (
int) – IN: int. n >= 0.
The number of columns of all matrices A_j in the batch.
- nrhs (
int) – IN: int. nrhs >= 0.
The number of columns of all matrices B_j and X_j in the batch; i.e., the columns on the right hand side.
- A (
Pointer/object) – INOUT: pointer to type. Array on the GPU (the size depends on the value of strideA).
On entry, the matrices A_j. On exit, the QR (or LQ) factorizations of A_j as returned by “GEQRF_STRIDED_BATCHED” (or “GELQF_STRIDED_BATCHED”).
- lda (
int) – IN: int. lda >= m.
Specifies the leading dimension of matrices A_j.
- strideA (
int) – IN: hipblasStride.
Stride from the start of one matrix A_j to the next one A_(j+1). There is no restriction for the value of strideA. Normal use case is strideA >= lda*n
- B (
Pointer/object) – INOUT: pointer to type. Array on the GPU (the size depends on the value of strideB).
On entry, the matrices B_j. On exit, when info[j] = 0, each B_j is overwritten by the solution vectors (and the residuals in the overdetermined cases) stored as columns.
- ldb (
int) – IN: int. ldb >= max(m,n).
Specifies the leading dimension of matrices B_j.
- strideB (
int) – IN: hipblasStride.
Stride from the start of one matrix B_j to the next one B_(j+1). There is no restriction for the value of strideB. Normal use case is strideB >= ldb*nrhs
- info (
ListOfInt/object) – OUT: pointer to an int on the host.
If info = 0, successful exit. If info = j < 0, the argument at position -j is invalid.
- deviceInfo (
ListOfInt/object) – OUT: pointer to int. Array of batchCount integers on the GPU.
If deviceInfo[j] = 0, successful exit for solution of A_j. If deviceInfo[j] = i > 0, the solution of A_j could not be computed because input matrix A_j is rank deficient; the i-th diagonal element of its triangular factor is zero.
- batchCount (
int) – IN: int. batchCount >= 0.
Number of matrices in the batch.
- handle (
- hip.hipblas.hipblasDgelsStridedBatched(handle, trans, int m, int n, int nrhs, A, int lda, long strideA, B, int ldb, long strideB, info, deviceInfo, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- deviceInfo (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgelsStridedBatched(handle, trans, int m, int n, int nrhs, A, int lda, long strideA, B, int ldb, long strideB, info, deviceInfo, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- B (
hipblasComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- deviceInfo (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgelsStridedBatched(handle, trans, int m, int n, int nrhs, A, int lda, long strideA, B, int ldb, long strideB, info, deviceInfo, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- B (
hipblasDoubleComplex/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- deviceInfo (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgelsStridedBatched_v2(handle, trans, int m, int n, int nrhs, A, int lda, long strideA, B, int ldb, long strideB, info, deviceInfo, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- B (
float2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- deviceInfo (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgelsStridedBatched_v2(handle, trans, int m, int n, int nrhs, A, int lda, long strideA, B, int ldb, long strideB, info, deviceInfo, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- trans (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- nrhs (
int): (undocumented)
- A (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- B (
double2/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- deviceInfo (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasSgeqrf(handle, int m, int n, A, int lda, ipiv, info)#
SOLVER API
@{
geqrf computes a QR factorization of a general m-by-n matrix A.
The factorization has the form
\[\begin{split}A = Q\left[\begin{array}{c} R\\ 0 \end{array}\right]\end{split}\]where R is upper triangular (upper trapezoidal if m < n), and Q is a m-by-m orthogonal/unitary matrix represented as the product of Householder matrices
\[Q = H_1H_2\cdots H_k, \quad \text{with} \: k = \text{min}(m,n)\]Each Householder matrix \(H_i\) is given by
\[H_i = I - \text{ipiv}[i] \cdot v_i v_i'\]- Args:
- handle (
Pointer/object) – IN: hipblasHandle_t.
- m (
int) – IN: int. m >= 0.
The number of rows of the matrix A.
- n (
int) – IN: int. n >= 0.
The number of columns of the matrix A.
- A (
Pointer/object) – INOUT: pointer to type. Array on the GPU of dimension lda*n.
On entry, the m-by-n matrix to be factored. On exit, the elements on and above the diagonal contain the factor R; the elements below the diagonal are the last m - i elements of Householder vector v_i.
- lda (
int) – IN: int. lda >= m.
Specifies the leading dimension of A.
- ipiv (
Pointer/object) – OUT: pointer to type. Array on the GPU of dimension min(m,n).
The Householder scalars.
- info (
ListOfInt/object) – OUT: pointer to a int on the host.
If info = 0, successful exit. If info = j < 0, the argument at position -j is invalid.
- handle (
- hip.hipblas.hipblasDgeqrf(handle, int m, int n, A, int lda, ipiv, info)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCgeqrf(handle, int m, int n, A, int lda, ipiv, info)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZgeqrf(handle, int m, int n, A, int lda, ipiv, info)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCgeqrf_v2(handle, int m, int n, A, int lda, ipiv, info)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZgeqrf_v2(handle, int m, int n, A, int lda, ipiv, info)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSgeqrfBatched(handle, int m, int n, A, int lda, ipiv, info, int batchCount)#
SOLVER API
@{
geqrfBatched computes the QR factorization of a batch of general m-by-n matrices.
The factorization of matrix \(A_i\) in the batch has the form
\[\begin{split}A_i = Q_i\left[\begin{array}{c} R_i\\ 0 \end{array}\right]\end{split}\]where \(R_i\) is upper triangular (upper trapezoidal if m < n), and \(Q_i\) is a m-by-m orthogonal/unitary matrix represented as the product of Householder matrices
\[Q_i = H_{i_1}H_{i_2}\cdots H_{i_k}, \quad \text{with} \: k = \text{min}(m,n)\]Each Householder matrix \(H_{i_j}\) is given by
\[H_{i_j} = I - \text{ipiv}_i[j] \cdot v_{i_j} v_{i_j}'\]- Args:
- handle (
Pointer/object) – IN: hipblasHandle_t.
- m (
int) – IN: int. m >= 0.
The number of rows of all the matrices A_i in the batch.
- n (
int) – IN: int. n >= 0.
The number of columns of all the matrices A_i in the batch.
- A (
ListOfPointer/object) – INOUT: Array of pointers to type. Each pointer points to an array on the GPU of dimension lda*n.
On entry, the m-by-n matrices A_i to be factored. On exit, the elements on and above the diagonal contain the factor R_i. The elements below the diagonal are the last m - j elements of Householder vector v_(i_j).
- lda (
int) – IN: int. lda >= m.
Specifies the leading dimension of matrices A_i.
- ipiv (
Pointer/object) – OUT: array of pointers to type. Each pointer points to an array on the GPU of dimension min(m, n).
Contains the vectors ipiv_i of corresponding Householder scalars.
- info (
ListOfInt/object) – OUT: pointer to a int on the host.
If info = 0, successful exit. If info = j < 0, the argument at position -j is invalid.
- batchCount (
int) – IN: int. batchCount >= 0.
Number of matrices in the batch.
- handle (
- hip.hipblas.hipblasDgeqrfBatched(handle, int m, int n, A, int lda, ipiv, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCgeqrfBatched(handle, int m, int n, A, int lda, ipiv, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZgeqrfBatched(handle, int m, int n, A, int lda, ipiv, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasCgeqrfBatched_v2(handle, int m, int n, A, int lda, ipiv, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasZgeqrfBatched_v2(handle, int m, int n, A, int lda, ipiv, info, int batchCount)#
(No short description, might be part of a group.)
- hip.hipblas.hipblasSgeqrfStridedBatched(handle, int m, int n, A, int lda, long strideA, ipiv, long strideP, info, int batchCount)#
SOLVER API
@{
geqrfStridedBatched computes the QR factorization of a batch of general m-by-n matrices.
The factorization of matrix \(A_i\) in the batch has the form
\[\begin{split}A_i = Q_i\left[\begin{array}{c} R_i\\ 0 \end{array}\right]\end{split}\]where \(R_i\) is upper triangular (upper trapezoidal if m < n), and \(Q_i\) is a m-by-m orthogonal/unitary matrix represented as the product of Householder matrices
\[Q_i = H_{i_1}H_{i_2}\cdots H_{i_k}, \quad \text{with} \: k = \text{min}(m,n)\]Each Householder matrix \(H_{i_j}\) is given by
\[H_{i_j} = I - \text{ipiv}_j[j] \cdot v_{i_j} v_{i_j}'\]- Args:
- handle (
Pointer/object) – IN: hipblasHandle_t.
- m (
int) – IN: int. m >= 0.
The number of rows of all the matrices A_i in the batch.
- n (
int) – IN: int. n >= 0.
The number of columns of all the matrices A_i in the batch.
- A (
Pointer/object) – INOUT: pointer to type. Array on the GPU (the size depends on the value of strideA).
On entry, the m-by-n matrices A_i to be factored. On exit, the elements on and above the diagonal contain the factor R_i. The elements below the diagonal are the last m - j elements of Householder vector v_(i_j).
- lda (
int) – IN: int. lda >= m.
Specifies the leading dimension of matrices A_i.
- strideA (
int) – IN: hipblasStride.
Stride from the start of one matrix A_i to the next one A_(i+1). There is no restriction for the value of strideA. Normal use case is strideA >= lda*n.
- ipiv (
Pointer/object) – OUT: pointer to type. Array on the GPU (the size depends on the value of strideP).
Contains the vectors ipiv_i of corresponding Householder scalars.
- strideP (
int) – IN: hipblasStride.
Stride from the start of one vector ipiv_i to the next one ipiv_(i+1). There is no restriction for the value of strideP. Normal use is strideP >= min(m,n).
- info (
ListOfInt/object) – OUT: pointer to a int on the host.
If info = 0, successful exit. If info = j < 0, the argument at position -j is invalid.
- batchCount (
int) – IN: int. batchCount >= 0.
Number of matrices in the batch.
- handle (
- hip.hipblas.hipblasDgeqrfStridedBatched(handle, int m, int n, A, int lda, long strideA, ipiv, long strideP, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- ipiv (
Pointer/object): (undocumented)
- strideP (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgeqrfStridedBatched(handle, int m, int n, A, int lda, long strideA, ipiv, long strideP, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- A (
hipblasComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- ipiv (
hipblasComplex/object): (undocumented)
- strideP (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgeqrfStridedBatched(handle, int m, int n, A, int lda, long strideA, ipiv, long strideP, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- A (
hipblasDoubleComplex/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- ipiv (
hipblasDoubleComplex/object): (undocumented)
- strideP (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasCgeqrfStridedBatched_v2(handle, int m, int n, A, int lda, long strideA, ipiv, long strideP, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- A (
float2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- ipiv (
float2/object): (undocumented)
- strideP (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasZgeqrfStridedBatched_v2(handle, int m, int n, A, int lda, long strideA, ipiv, long strideP, info, int batchCount)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- A (
double2/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- ipiv (
double2/object): (undocumented)
- strideP (
int): (undocumented)
- info (
ListOfInt/object): (undocumented)
- batchCount (
int): (undocumented)
- handle (
- hip.hipblas.hipblasGemmEx(handle, transA, transB, int m, int n, int k, alpha, A, aType, int lda, B, bType, int ldb, beta, C, cType, int ldc, computeType, algo)#
BLAS EX API
gemmEx performs one of the matrix-matrix operations
C = alpha*op( A )*op( B ) + beta*C,
where op( X ) is one of
op( X ) = X or op( X ) = X**T or op( X ) = X**H,
alpha and beta are scalars, and A, B, and C are matrices, with op( A ) an m by k matrix, op( B ) a k by n matrix and C is a m by n matrix.
Supported types are determined by the backend. See cuBLAS documentation for cuBLAS backend. For rocBLAS backend, conversion from hipblasComputeType_t to rocblas_datatype_t happens within hipBLAS. Supported types are as follows:
aType | bType | cType | computeType |———- | ———- | ———- | ——————- |HIP_R_16F | HIP_R_16F | HIP_R_16F | HIPBLAS_COMPUTE_16F |HIP_R_16F | HIP_R_16F | HIP_R_16F | HIPBLAS_COMPUTE_32F |HIP_R_16F | HIP_R_16F | HIP_R_32F | HIPBLAS_COMPUTE_32F |HIP_R_16BF | HIP_R_16BF | HIP_R_16BF | HIPBLAS_COMPUTE_32F |HIP_R_16BF | HIP_R_16BF | HIP_R_32F | HIPBLAS_COMPUTE_32F |HIP_R_32F | HIP_R_32F | HIP_R_32F | HIPBLAS_COMPUTE_32F |HIP_R_64F | HIP_R_64F | HIP_R_64F | HIPBLAS_COMPUTE_64F |HIP_R_8I | HIP_R_8I | HIP_R_32I | HIPBLAS_COMPUTE_32I |HIP_C_32F | HIP_C_32F | HIP_C_32F | HIPBLAS_COMPUTE_32F |HIP_C_64F | HIP_C_64F | HIP_C_64F | HIPBLAS_COMPUTE_64F |
hipblasGemmExWithFlags is also available which is identical to hipblasGemmEx with the addition of a “flags” parameter which controls flags used in Tensile to control gemm algorithms with the rocBLAS backend. When using a cuBLAS backend this parameter is ignored.
With HIPBLAS_V2 define, hipblasGemmEx accepts hipDataType for aType, bType, and cType. It also accepts hipblasComputeType_t for computeType. hipblasGemmEx will no longer support hipblasDataType_t for these parameters in a future release. hipblasGemmEx follows the same convention.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasGemmEx(hipblasHandle_t handle,
hipblasOperation_t transA, hipblasOperation_t transB, int m, int n, int k, const void* alpha, const void* A, hipDataType aType, int lda, const void* B, hipDataType bType, int ldb, const void* beta, void* C, hipDataType cType, int ldc, hipblasComputeType_t computeType, hipblasGemmAlgo_t algo)
- hipblasStatus_t hipblasGemmExWithFlags(hipblasHandle_t handle,
hipblasOperation_t transA, hipblasOperation_t transB, int m, int n, int k, const void* alpha, const void* A, hipDataType aType, int lda, const void* B, hipDataType bType, int ldb, const void* beta, void* C, hipDataType cType, int ldc, hipblasComputeType_t computeType, hipblasGemmAlgo_t algo, hipblasGemmFlags_t flags)
else// [DEPRECATED]- hipblasStatus_t hipblasGemmEx(hipblasHandle_t handle,
hipblasOperation_t transA, hipblasOperation_t transB, int m, int n, int k, const void* alpha, const void* A, hipblasDatatype_t aType, int lda, const void* B, hipblasDatatype_t bType, int ldb, const void* beta, void* C, hipblasDatatype_t cType, int ldc, hipblasDatatype_t computeType, hipblasGemmAlgo_t algo)
- hipblasStatus_t hipblasGemmExWithFlags(hipblasHandle_t handle,
hipblasOperation_t transA, hipblasOperation_t transB, int m, int n, int k, const void* alpha, const void* A, hipblasDatatype_t aType, int lda, const void* B, hipblasDatatype_t bType, int ldb, const void* beta, void* C, hipblasDatatype_t cType, int ldc, hipblasDatatype_t computeType, hipblasGemmAlgo_t algo, hipblasGemmFlags_t flags)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( A ).
- transB (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( B ).
- m (
int) – IN: [int] matrix dimension m.
- n (
int) – IN: [int] matrix dimension n.
- k (
int) – IN: [int] matrix dimension k.
- alpha (
Pointer/object) – IN: [const void *] device pointer or host pointer specifying the scalar alpha. Same datatype as computeType.
- A (
Pointer/object) – IN: [void *] device pointer storing matrix A.
- aType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of matrix A.
- [hipDataType]
specifies the datatype of matrix A.
- lda (
int) – IN: [int] specifies the leading dimension of A.
- B (
Pointer/object) – IN: [void *] device pointer storing matrix B.
- bType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of matrix B.
- [hipDataType]
specifies the datatype of matrix B.
- ldb (
int) – IN: [int] specifies the leading dimension of B.
- beta (
Pointer/object) – IN: [const void *] device pointer or host pointer specifying the scalar beta. Same datatype as computeType.
- C (
Pointer/object) – IN: [void *] device pointer storing matrix C.
- cType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of matrix C.
- [hipDataType]
specifies the datatype of matrix C.
- ldc (
int) – IN: [int] specifies the leading dimension of C.
- computeType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipblasComputeType_t]
specifies the datatype of computation.
- algo (
hipblasGemmAlgo_t) – IN: [hipblasGemmAlgo_t] enumerant specifying the algorithm type.
- handle (
- hip.hipblas.hipblasGemmEx_v2(handle, transA, transB, int m, int n, int k, alpha, A, aType, int lda, B, bType, int ldb, beta, C, cType, int ldc, computeType, algo)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- A (
Pointer/object): (undocumented)
- aType (
hipDataType): (undocumented)
- lda (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- bType (
hipDataType): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- C (
Pointer/object): (undocumented)
- cType (
hipDataType): (undocumented)
- ldc (
int): (undocumented)
- computeType (
hipblasComputeType_t): (undocumented)
- algo (
hipblasGemmAlgo_t): (undocumented)
- handle (
- hip.hipblas.hipblasGemmExWithFlags(handle, transA, transB, int m, int n, int k, alpha, A, aType, int lda, B, bType, int ldb, beta, C, cType, int ldc, computeType, algo, flags)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- A (
Pointer/object): (undocumented)
- aType (
hipblasDatatype_t): (undocumented)
- lda (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- bType (
hipblasDatatype_t): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- C (
Pointer/object): (undocumented)
- cType (
hipblasDatatype_t): (undocumented)
- ldc (
int): (undocumented)
- computeType (
hipblasDatatype_t): (undocumented)
- algo (
hipblasGemmAlgo_t): (undocumented)
- flags (
hipblasGemmFlags_t): (undocumented)
- handle (
- hip.hipblas.hipblasGemmExWithFlags_v2(handle, transA, transB, int m, int n, int k, alpha, A, aType, int lda, B, bType, int ldb, beta, C, cType, int ldc, computeType, algo, flags)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- A (
Pointer/object): (undocumented)
- aType (
hipDataType): (undocumented)
- lda (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- bType (
hipDataType): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- C (
Pointer/object): (undocumented)
- cType (
hipDataType): (undocumented)
- ldc (
int): (undocumented)
- computeType (
hipblasComputeType_t): (undocumented)
- algo (
hipblasGemmAlgo_t): (undocumented)
- flags (
hipblasGemmFlags_t): (undocumented)
- handle (
- hip.hipblas.hipblasGemmBatchedEx(handle, transA, transB, int m, int n, int k, alpha, A, aType, int lda, B, bType, int ldb, beta, C, cType, int ldc, int batchCount, computeType, algo)#
BLAS EX API
- gemmBatchedEx performs one of the batched matrix-matrix operations
C_i = alpha*op(A_i)*op(B_i) + beta*C_i, for i = 1, …, batchCount.
- where op( X ) is one of
op( X ) = X or op( X ) = X**T or op( X ) = X**H,
alpha and beta are scalars, and A, B, and C are batched pointers to matrices, with op( A ) an m by k by batchCount batched matrix, op( B ) a k by n by batchCount batched matrix and C a m by n by batchCount batched matrix. The batched matrices are an array of pointers to matrices. The number of pointers to matrices is batchCount.
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
hipblasGemmBatchedExWithFlags is also available which is identical to hipblasGemmBatchedEx with the addition of a “flags” parameter which controls flags used in Tensile to control gemm algorithms with the rocBLAS backend. When using a cuBLAS backend this parameter is ignored.
With HIPBLAS_V2 define, hipblasGemmBatchedEx accepts hipDataType for aType, bType, and cType. It also accepts hipblasComputeType_t for computeType. hipblasGemmBatchedEx will no longer support hipblasDataType_t for these parameters in a future release. hipblasGemmBatchedExWithFlags follows the same convention.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasGemmBatchedEx(hipblasHandle_t handle,
hipblasOperation_t transA, hipblasOperation_t transB, int m, int n, int k, const void* alpha, const void* A[], hipDataType aType, int lda, const void* B[], hipDataType bType, int ldb, const void* beta, void* C[], hipDataType cType, int ldc, int batchCount, hipblasComputeType_t computeType, hipblasGemmAlgo_t algo)
- hipblasStatus_t hipblasGemmBatchedExWithFlags(hipblasHandle_t handle,
hipblasOperation_t transA, hipblasOperation_t transB, int m, int n, int k, const void* alpha, const void* A[], hipDataType aType, int lda, const void* B[], hipDataType bType, int ldb, const void* beta, void* C[], hipDataType cType, int ldc, int batchCount, hipblasComputeType_t computeType, hipblasGemmAlgo_t algo, hipblasGemmFlags_t flags)
else// [DEPRECATED]- hipblasStatus_t hipblasGemmBatchedEx(hipblasHandle_t handle,
hipblasOperation_t transA, hipblasOperation_t transB, int m, int n, int k, const void* alpha, const void* A[], hipblasDatatype_t aType, int lda, const void* B[], hipblasDatatype_t bType, int ldb, const void* beta, void* C[], hipblasDatatype_t cType, int ldc, int batchCount, hipblasDatatype_t computeType, hipblasGemmAlgo_t algo)
- hipblasStatus_t hipblasGemmBatchedExWithFlags(hipblasHandle_t handle,
hipblasOperation_t transA, hipblasOperation_t transB, int m, int n, int k, const void* alpha, const void* A[], hipblasDatatype_t aType, int lda, const void* B[], hipblasDatatype_t bType, int ldb, const void* beta, void* C[], hipblasDatatype_t cType, int ldc, int batchCount, hipblasDatatype_t computeType, hipblasGemmAlgo_t algo, hipblasGemmFlags_t flags)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( A ).
- transB (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( B ).
- m (
int) – IN: [int] matrix dimension m.
- n (
int) – IN: [int] matrix dimension n.
- k (
int) – IN: [int] matrix dimension k.
- alpha (
Pointer/object) – IN: [const void *] device pointer or host pointer specifying the scalar alpha. Same datatype as computeType.
- A (
ListOfPointer/object) – IN: [void *] device pointer storing array of pointers to each matrix A_i.
- aType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each matrix A_i.
- [hipDataType]
specifies the datatype of each matrix A_i.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i.
- B (
ListOfPointer/object) – IN: [void *] device pointer storing array of pointers to each matrix B_i.
- bType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each matrix B_i.
- [hipDataType]
specifies the datatype of each matrix B_i.
- ldb (
int) – IN: [int] specifies the leading dimension of each B_i.
- beta (
Pointer/object) – IN: [const void *] device pointer or host pointer specifying the scalar beta. Same datatype as computeType.
- C (
ListOfPointer/object) – IN: [void *] device array of device pointers to each matrix C_i.
- cType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each matrix C_i.
- [hipDataType]
specifies the datatype of each matrix C_i.
- ldc (
int) – IN: [int] specifies the leading dimension of each C_i.
- batchCount (
int) – IN: [int] number of gemm operations in the batch.
- computeType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipblasComputeType_t]
specifies the datatype of computation.
- algo (
hipblasGemmAlgo_t) – IN: [hipblasGemmAlgo_t] enumerant specifying the algorithm type.
- handle (
- hip.hipblas.hipblasGemmBatchedEx_v2(handle, transA, transB, int m, int n, int k, alpha, A, aType, int lda, B, bType, int ldb, beta, C, cType, int ldc, int batchCount, computeType, algo)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- A (
ListOfPointer/object): (undocumented)
- aType (
hipDataType): (undocumented)
- lda (
int): (undocumented)
- B (
ListOfPointer/object): (undocumented)
- bType (
hipDataType): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- C (
ListOfPointer/object): (undocumented)
- cType (
hipDataType): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- computeType (
hipblasComputeType_t): (undocumented)
- algo (
hipblasGemmAlgo_t): (undocumented)
- handle (
- hip.hipblas.hipblasGemmBatchedExWithFlags(handle, transA, transB, int m, int n, int k, alpha, A, aType, int lda, B, bType, int ldb, beta, C, cType, int ldc, int batchCount, computeType, algo, flags)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- A (
ListOfPointer/object): (undocumented)
- aType (
hipblasDatatype_t): (undocumented)
- lda (
int): (undocumented)
- B (
ListOfPointer/object): (undocumented)
- bType (
hipblasDatatype_t): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- C (
ListOfPointer/object): (undocumented)
- cType (
hipblasDatatype_t): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- computeType (
hipblasDatatype_t): (undocumented)
- algo (
hipblasGemmAlgo_t): (undocumented)
- flags (
hipblasGemmFlags_t): (undocumented)
- handle (
- hip.hipblas.hipblasGemmBatchedExWithFlags_v2(handle, transA, transB, int m, int n, int k, alpha, A, aType, int lda, B, bType, int ldb, beta, C, cType, int ldc, int batchCount, computeType, algo, flags)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- A (
ListOfPointer/object): (undocumented)
- aType (
hipDataType): (undocumented)
- lda (
int): (undocumented)
- B (
ListOfPointer/object): (undocumented)
- bType (
hipDataType): (undocumented)
- ldb (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- C (
ListOfPointer/object): (undocumented)
- cType (
hipDataType): (undocumented)
- ldc (
int): (undocumented)
- batchCount (
int): (undocumented)
- computeType (
hipblasComputeType_t): (undocumented)
- algo (
hipblasGemmAlgo_t): (undocumented)
- flags (
hipblasGemmFlags_t): (undocumented)
- handle (
- hip.hipblas.hipblasGemmStridedBatchedEx(handle, transA, transB, int m, int n, int k, alpha, A, aType, int lda, long strideA, B, bType, int ldb, long strideB, beta, C, cType, int ldc, long strideC, int batchCount, computeType, algo)#
BLAS EX API
gemmStridedBatchedEx performs one of the strided_batched matrix-matrix operations
C_i = alpha*op(A_i)*op(B_i) + beta*C_i, for i = 1, …, batchCount
where op( X ) is one of
op( X ) = X or op( X ) = X**T or op( X ) = X**H,
alpha and beta are scalars, and A, B, and C are strided_batched matrices, with op( A ) an m by k by batchCount strided_batched matrix, op( B ) a k by n by batchCount strided_batched matrix and C a m by n by batchCount strided_batched matrix.
The strided_batched matrices are multiple matrices separated by a constant stride. The number of matrices is batchCount.
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
hipblasGemmStridedBatchedExWithFlags is also available which is identical to hipblasStridedBatchedGemmEx with the addition of a “flags” parameter which controls flags used in Tensile to control gemm algorithms with the rocBLAS backend. When using a cuBLAS backend this parameter is ignored.
With HIPBLAS_V2 define, hipblasGemmStridedBatchedEx accepts hipDataType for aType, bType, and cType. It also accepts hipblasComputeType_t for computeType. hipblasGemmStridedBatchedEx will no longer support hipblasDataType_t for these parameters in a future release. hipblasGemmStridedBatchedExWithFlags follows the same convention.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasGemmStridedBatchedEx(hipblasHandle_t handle,
hipblasOperation_t transA, hipblasOperation_t transB, int m, int n, int k, const void* alpha, const void* A, hipDataType aType, int lda, hipblasStride strideA, const void* B, hipDataType bType, int ldb, hipblasStride strideB, const void* beta, void* C, hipDataType cType, int ldc, hipblasStride strideC, int batchCount, hipblasComputeType_t computeType, hipblasGemmAlgo_t algo)
- hipblasStatus_t hipblasGemmStridedBatchedExWithFlags(hipblasHandle_t handle,
hipblasOperation_t transA, hipblasOperation_t transB, int m, int n, int k, const void* alpha, const void* A, hipDataType aType, int lda, hipblasStride strideA, const void* B, hipDataType bType, int ldb, hipblasStride strideB, const void* beta, void* C, hipDataType cType, int ldc, hipblasStride strideC, int batchCount, hipblasComputeType_t computeType, hipblasGemmAlgo_t algo, hipblasGemmFlags_t flags)
else// [DEPRECATED]- hipblasStatus_t hipblasGemmStridedBatchedEx(hipblasHandle_t handle,
hipblasOperation_t transA, hipblasOperation_t transB, int m, int n, int k, const void* alpha, const void* A, hipblasDatatype_t aType, int lda, hipblasStride strideA, const void* B, hipblasDatatype_t bType, int ldb, hipblasStride strideB, const void* beta, void* C, hipblasDatatype_t cType, int ldc, hipblasStride strideC, int batchCount, hipblasDatatype_t computeType, hipblasGemmAlgo_t algo)
- hipblasStatus_t hipblasGemmStridedBatchedExWithFlags(hipblasHandle_t handle,
hipblasOperation_t transA, hipblasOperation_t transB, int m, int n, int k, const void* alpha, const void* A, hipblasDatatype_t aType, int lda, hipblasStride strideA, const void* B, hipblasDatatype_t bType, int ldb, hipblasStride strideB, const void* beta, void* C, hipblasDatatype_t cType, int ldc, hipblasStride strideC, int batchCount, hipblasDatatype_t computeType, hipblasGemmAlgo_t algo, hipblasGemmFlags_t flags)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( A ).
- transB (
hipblasOperation_t) – IN: [hipblasOperation_t] specifies the form of op( B ).
- m (
int) – IN: [int] matrix dimension m.
- n (
int) – IN: [int] matrix dimension n.
- k (
int) – IN: [int] matrix dimension k.
- alpha (
Pointer/object) – IN: [const void *] device pointer or host pointer specifying the scalar alpha. Same datatype as computeType.
- A (
Pointer/object) – IN: [void *] device pointer pointing to first matrix A_1.
- aType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each matrix A_i.
- [hipDataType]
specifies the datatype of each matrix A_i.
- lda (
int) – IN: [int] specifies the leading dimension of each A_i.
- strideA (
int) – IN: [hipblasStride] specifies stride from start of one A_i matrix to the next A_(i + 1).
- B (
Pointer/object) – IN: [void *] device pointer pointing to first matrix B_1.
- bType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each matrix B_i.
- [hipDataType]
specifies the datatype of each matrix B_i.
- ldb (
int) – IN: [int] specifies the leading dimension of each B_i.
- strideB (
int) – IN: [hipblasStride] specifies stride from start of one B_i matrix to the next B_(i + 1).
- beta (
Pointer/object) – IN: [const void *] device pointer or host pointer specifying the scalar beta. Same datatype as computeType.
- C (
Pointer/object) – IN: [void *] device pointer pointing to first matrix C_1.
- cType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each matrix C_i.
- [hipDataType]
specifies the datatype of each matrix C_i.
- ldc (
int) – IN: [int] specifies the leading dimension of each C_i.
- strideC (
int) – IN: [hipblasStride] specifies stride from start of one C_i matrix to the next C_(i + 1).
- batchCount (
int) – IN: [int] number of gemm operations in the batch.
- computeType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipblasComputeType_t]
specifies the datatype of computation.
- algo (
hipblasGemmAlgo_t) – IN: [hipblasGemmAlgo_t] enumerant specifying the algorithm type.
- handle (
- hip.hipblas.hipblasGemmStridedBatchedEx_v2(handle, transA, transB, int m, int n, int k, alpha, A, aType, int lda, long strideA, B, bType, int ldb, long strideB, beta, C, cType, int ldc, long strideC, int batchCount, computeType, algo)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- A (
Pointer/object): (undocumented)
- aType (
hipDataType): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- bType (
hipDataType): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- C (
Pointer/object): (undocumented)
- cType (
hipDataType): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- computeType (
hipblasComputeType_t): (undocumented)
- algo (
hipblasGemmAlgo_t): (undocumented)
- handle (
- hip.hipblas.hipblasGemmStridedBatchedExWithFlags(handle, transA, transB, int m, int n, int k, alpha, A, aType, int lda, long strideA, B, bType, int ldb, long strideB, beta, C, cType, int ldc, long strideC, int batchCount, computeType, algo, flags)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- A (
Pointer/object): (undocumented)
- aType (
hipblasDatatype_t): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- bType (
hipblasDatatype_t): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- C (
Pointer/object): (undocumented)
- cType (
hipblasDatatype_t): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- computeType (
hipblasDatatype_t): (undocumented)
- algo (
hipblasGemmAlgo_t): (undocumented)
- flags (
hipblasGemmFlags_t): (undocumented)
- handle (
- hip.hipblas.hipblasGemmStridedBatchedExWithFlags_v2(handle, transA, transB, int m, int n, int k, alpha, A, aType, int lda, long strideA, B, bType, int ldb, long strideB, beta, C, cType, int ldc, long strideC, int batchCount, computeType, algo, flags)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- transB (
hipblasOperation_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- k (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- A (
Pointer/object): (undocumented)
- aType (
hipDataType): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- bType (
hipDataType): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- beta (
Pointer/object): (undocumented)
- C (
Pointer/object): (undocumented)
- cType (
hipDataType): (undocumented)
- ldc (
int): (undocumented)
- strideC (
int): (undocumented)
- batchCount (
int): (undocumented)
- computeType (
hipblasComputeType_t): (undocumented)
- algo (
hipblasGemmAlgo_t): (undocumented)
- flags (
hipblasGemmFlags_t): (undocumented)
- handle (
- hip.hipblas.hipblasTrsmEx(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, invA, int invAsize, computeType)#
(No short description, might be part of a group.)
BLAS EX API
trsmEx solves
op(A)*X = alpha*B or X*op(A) = alpha*B,
where alpha is a scalar, X and B are m by n matrices, A is triangular matrix and op(A) is one of
op( A ) = A or op( A ) = A^T or op( A ) = A^H.
The matrix X is overwritten on B.
This function gives the user the ability to reuse the invA matrix between runs. If invA == NULL, hipblasTrsmEx will automatically calculate invA on every run.
Setting up invA: The accepted invA matrix consists of the packed 128x128 inverses of the diagonal blocks of matrix A, followed by any smaller diagonal block that remains. To set up invA it is recommended that hipblasTrtriBatched be used with matrix A as the input.
Device memory of size 128 x k should be allocated for invA ahead of time, where k is m when HIPBLAS_SIDE_LEFT and is n when HIPBLAS_SIDE_RIGHT. The actual number of elements in invA should be passed as invAsize.
To begin, hipblasTrtriBatched must be called on the full 128x128 sized diagonal blocks of matrix A. Below are the restricted parameters:
n = 128
ldinvA = 128
stride_invA = 128x128
batchCount = k / 128,
- Then any remaining block may be added:
n = k % 128
invA = invA + stride_invA * previousBatchCount
ldinvA = 128
batchCount = 1
With HIPBLAS_V2 define, hipblasTrsmEx accepts hipDataType for computeType rather than hipblasDatatype_t. hipblasTrsmEx will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasTrsmEx(hipblasHandle_t handle,
hipblasSideMode_t side, hipblasFillMode_t uplo, hipblasOperation_t transA, hipblasDiagType_t diag, int m, int n, const void* alpha, void* A, int lda, void* B, int ldb, const void* invA, int invAsize, hipDataType computeType)
else// [DEPRECATED]- hipblasStatus_t hipblasTrsmEx(hipblasHandle_t handle,
hipblasSideMode_t side, hipblasFillMode_t uplo, hipblasOperation_t transA, hipblasDiagType_t diag, int m, int n, const void* alpha, void* A, int lda, void* B, int ldb, const void* invA, int invAsize, hipblasDatatype_t computeType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] HIPBLAS_SIDE_LEFT: op(A)*X = alpha*B. HIPBLAS_SIDE_RIGHT: X*op(A) = alpha*B.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: A is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: A is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_N: op(A) = A. HIPBLAS_OP_T: op(A) = A^T. HIPBLAS_ON_C: op(A) = A^H.
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: A is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: A is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of B. m >= 0.
- n (
int) – IN: [int] n specifies the number of columns of B. n >= 0.
- alpha (
Pointer/object) – IN: [void *] device pointer or host pointer specifying the scalar alpha. When alpha is &zero then A is not referenced, and B need not be set before entry.
- A (
Pointer/object) – IN: [void *] device pointer storing matrix A. of dimension ( lda, k ), where k is m when HIPBLAS_SIDE_LEFT and is n when HIPBLAS_SIDE_RIGHT only the upper/lower triangular part is accessed.
- lda (
int) – IN: [int] lda specifies the first dimension of A. if side = HIPBLAS_SIDE_LEFT, lda >= max( 1, m ), if side = HIPBLAS_SIDE_RIGHT, lda >= max( 1, n ).
- B (
Pointer/object) – IN,OUT: [void *] device pointer storing matrix B. B is of dimension ( ldb, n ). Before entry, the leading m by n part of the array B must contain the right-hand side matrix B, and on exit is overwritten by the solution matrix X.
- ldb (
int) – IN: [int] ldb specifies the first dimension of B. ldb >= max( 1, m ).
- invA (
Pointer/object) – IN: [void *] device pointer storing the inverse diagonal blocks of A. invA is of dimension ( ld_invA, k ), where k is m when HIPBLAS_SIDE_LEFT and is n when HIPBLAS_SIDE_RIGHT. ld_invA must be equal to 128.
- invAsize (
int) – IN: [int] invAsize specifies the number of elements of device memory in invA.
- computeType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasTrsmEx_v2(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, invA, int invAsize, computeType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- invA (
Pointer/object): (undocumented)
- invAsize (
int): (undocumented)
- computeType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasTrsmBatchedEx(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, int batchCount, invA, int invAsize, computeType)#
(No short description, might be part of a group.)
BLAS EX API
trsmBatchedEx solves
op(A_i)*X_i = alpha*B_i or X_i*op(A_i) = alpha*B_i,
for i = 1, …, batchCount; and where alpha is a scalar, X and B are arrays of m by n matrices, A is an array of triangular matrix and each op(A_i) is one of
op( A_i ) = A_i or op( A_i ) = A_i^T or op( A_i ) = A_i^H.
Each matrix X_i is overwritten on B_i.
This function gives the user the ability to reuse the invA matrix between runs. If invA == NULL, hipblasTrsmBatchedEx will automatically calculate each invA_i on every run.
Setting up invA: Each accepted invA_i matrix consists of the packed 128x128 inverses of the diagonal blocks of matrix A_i, followed by any smaller diagonal block that remains. To set up each invA_i it is recommended that hipblasTrtriBatched be used with matrix A_i as the input. invA is an array of pointers of batchCount length holding each invA_i.
Device memory of size 128 x k should be allocated for each invA_i ahead of time, where k is m when HIPBLAS_SIDE_LEFT and is n when HIPBLAS_SIDE_RIGHT. The actual number of elements in each invA_i should be passed as invAsize.
To begin, hipblasTrtriBatched must be called on the full 128x128 sized diagonal blocks of each matrix A_i. Below are the restricted parameters:
n = 128
ldinvA = 128
stride_invA = 128x128
batchCount = k / 128,
- Then any remaining block may be added:
n = k % 128
invA = invA + stride_invA * previousBatchCount
ldinvA = 128
batchCount = 1
With HIPBLAS_V2 define, hipblasTrsmBatchedEx accepts hipDataType for computeType rather than hipblasDatatype_t. hipblasTrsmBatchedEx will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasTrsmBatchedEx(hipblasHandle_t handle,
hipblasSideMode_t side, hipblasFillMode_t uplo, hipblasOperation_t transA, hipblasDiagType_t diag, int m, int n, const void* alpha, void* A, int lda, void* B, int ldb, int batchCount, const void* invA, int invAsize, hipDataType computeType)
else// [DEPRECATED]- hipblasStatus_t hipblasTrsmBatchedEx(hipblasHandle_t handle,
hipblasSideMode_t side, hipblasFillMode_t uplo, hipblasOperation_t transA, hipblasDiagType_t diag, int m, int n, const void* alpha, void* A, int lda, void* B, int ldb, int batchCount, const void* invA, int invAsize, hipblasDatatype_t computeType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] HIPBLAS_SIDE_LEFT: op(A)*X = alpha*B. HIPBLAS_SIDE_RIGHT: X*op(A) = alpha*B.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: each A_i is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: each A_i is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_N: op(A) = A. HIPBLAS_OP_T: op(A) = A^T. HIPBLAS_OP_C: op(A) = A^H.
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: each A_i is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: each A_i is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of each B_i. m >= 0.
- n (
int) – IN: [int] n specifies the number of columns of each B_i. n >= 0.
- alpha (
Pointer/object) – IN: [void *] device pointer or host pointer alpha specifying the scalar alpha. When alpha is &zero then A is not referenced, and B need not be set before entry.
- A (
Pointer/object) – IN: [void *] device array of device pointers storing each matrix A_i. each A_i is of dimension ( lda, k ), where k is m when HIPBLAS_SIDE_LEFT and is n when HIPBLAS_SIDE_RIGHT only the upper/lower triangular part is accessed.
- lda (
int) – IN: [int] lda specifies the first dimension of each A_i. if side = HIPBLAS_SIDE_LEFT, lda >= max( 1, m ), if side = HIPBLAS_SIDE_RIGHT, lda >= max( 1, n ).
- B (
Pointer/object) – IN,OUT: [void *] device array of device pointers storing each matrix B_i. each B_i is of dimension ( ldb, n ). Before entry, the leading m by n part of the array B_i must contain the right-hand side matrix B_i, and on exit is overwritten by the solution matrix X_i
- ldb (
int) – IN: [int] ldb specifies the first dimension of each B_i. ldb >= max( 1, m ).
- batchCount (
int) – IN: [int] specifies how many batches.
- invA (
Pointer/object) – IN: [void *] device array of device pointers storing the inverse diagonal blocks of each A_i. each invA_i is of dimension ( ld_invA, k ), where k is m when HIPBLAS_SIDE_LEFT and is n when HIPBLAS_SIDE_RIGHT. ld_invA must be equal to 128.
- invAsize (
int) – IN: [int] invAsize specifies the number of elements of device memory in each invA_i.
- computeType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasTrsmBatchedEx_v2(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, B, int ldb, int batchCount, invA, int invAsize, computeType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- batchCount (
int): (undocumented)
- invA (
Pointer/object): (undocumented)
- invAsize (
int): (undocumented)
- computeType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasTrsmStridedBatchedEx(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, long strideA, B, int ldb, long strideB, int batchCount, invA, int invAsize, long strideInvA, computeType)#
(No short description, might be part of a group.)
BLAS EX API
trsmStridedBatchedEx solves
op(A_i)*X_i = alpha*B_i or X_i*op(A_i) = alpha*B_i,
for i = 1, …, batchCount; and where alpha is a scalar, X and B are strided batched m by n matrices, A is a strided batched triangular matrix and op(A_i) is one of
op( A_i ) = A_i or op( A_i ) = A_i^T or op( A_i ) = A_i^H.
Each matrix X_i is overwritten on B_i.
This function gives the user the ability to reuse each invA_i matrix between runs. If invA == NULL, hipblasTrsmStridedBatchedEx will automatically calculate each invA_i on every run.
Setting up invA: Each accepted invA_i matrix consists of the packed 128x128 inverses of the diagonal blocks of matrix A_i, followed by any smaller diagonal block that remains. To set up invA_i it is recommended that hipblasTrtriBatched be used with matrix A_i as the input. invA is a contiguous piece of memory holding each invA_i.
Device memory of size 128 x k should be allocated for each invA_i ahead of time, where k is m when HIPBLAS_SIDE_LEFT and is n when HIPBLAS_SIDE_RIGHT. The actual number of elements in each invA_i should be passed as invAsize.
To begin, hipblasTrtriBatched must be called on the full 128x128 sized diagonal blocks of each matrix A_i. Below are the restricted parameters:
n = 128
ldinvA = 128
stride_invA = 128x128
batchCount = k / 128,
- Then any remaining block may be added:
n = k % 128
invA = invA + stride_invA * previousBatchCount
ldinvA = 128
batchCount = 1
With HIPBLAS_V2 define, hipblasStridedBatchedTrsmEx accepts hipDataType for computeType rather than hipblasDatatype_t. hipblasTrsmStridedBatchedEx will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasTrsmStridedBatchedEx(hipblasHandle_t handle,
hipblasSideMode_t side, hipblasFillMode_t uplo, hipblasOperation_t transA, hipblasDiagType_t diag, int m, int n, const void* alpha, void* A, int lda, hipblasStride strideA, void* B, int ldb, hipblasStride strideB, int batchCount, const void* invA, int invAsize, hipblasStride strideInvA, hipDataType computeType);
else// [DEPRECATED]- hipblasStatus_t hipblasTrsmStridedBatchedEx(hipblasHandle_t handle,
hipblasSideMode_t side, hipblasFillMode_t uplo, hipblasOperation_t transA, hipblasDiagType_t diag, int m, int n, const void* alpha, void* A, int lda, hipblasStride strideA, void* B, int ldb, hipblasStride strideB, int batchCount, const void* invA, int invAsize, hipblasStride strideInvA, hipblasDatatype_t computeType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- side (
hipblasSideMode_t) – IN: [hipblasSideMode_t] HIPBLAS_SIDE_LEFT: op(A)*X = alpha*B. HIPBLAS_SIDE_RIGHT: X*op(A) = alpha*B.
- uplo (
hipblasFillMode_t) – IN: [hipblasFillMode_t] HIPBLAS_FILL_MODE_UPPER: each A_i is an upper triangular matrix. HIPBLAS_FILL_MODE_LOWER: each A_i is a lower triangular matrix.
- transA (
hipblasOperation_t) – IN: [hipblasOperation_t] HIPBLAS_OP_N: op(A) = A. HIPBLAS_OP_T: op(A) = A^T. HIPBLAS_OP_C: op(A) = A^H.
- diag (
hipblasDiagType_t) – IN: [hipblasDiagType_t] HIPBLAS_DIAG_UNIT: each A_i is assumed to be unit triangular. HIPBLAS_DIAG_NON_UNIT: each A_i is not assumed to be unit triangular.
- m (
int) – IN: [int] m specifies the number of rows of each B_i. m >= 0.
- n (
int) – IN: [int] n specifies the number of columns of each B_i. n >= 0.
- alpha (
Pointer/object) – IN: [void *] device pointer or host pointer specifying the scalar alpha. When alpha is &zero then A is not referenced, and B need not be set before entry.
- A (
Pointer/object) – IN: [void *] device pointer storing matrix A. of dimension ( lda, k ), where k is m when HIPBLAS_SIDE_LEFT and is n when HIPBLAS_SIDE_RIGHT only the upper/lower triangular part is accessed.
- lda (
int) – IN: [int] lda specifies the first dimension of A. if side = HIPBLAS_SIDE_LEFT, lda >= max( 1, m ), if side = HIPBLAS_SIDE_RIGHT, lda >= max( 1, n ).
- strideA (
int) – IN: [hipblasStride] The stride between each A matrix.
- B (
Pointer/object) – IN,OUT: [void *] device pointer pointing to first matrix B_i. each B_i is of dimension ( ldb, n ). Before entry, the leading m by n part of each array B_i must contain the right-hand side of matrix B_i, and on exit is overwritten by the solution matrix X_i.
- ldb (
int) – IN: [int] ldb specifies the first dimension of each B_i. ldb >= max( 1, m ).
- strideB (
int) – IN: [hipblasStride] The stride between each B_i matrix.
- batchCount (
int) – IN: [int] specifies how many batches.
- invA (
Pointer/object) – IN: [void *] device pointer storing the inverse diagonal blocks of each A_i. invA points to the first invA_1. each invA_i is of dimension ( ld_invA, k ), where k is m when HIPBLAS_SIDE_LEFT and is n when HIPBLAS_SIDE_RIGHT. ld_invA must be equal to 128.
- invAsize (
int) – IN: [int] invAsize specifies the number of elements of device memory in each invA_i.
- strideInvA (
int) – IN: [hipblasStride] The stride between each invA matrix.
- computeType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasTrsmStridedBatchedEx_v2(handle, side, uplo, transA, diag, int m, int n, alpha, A, int lda, long strideA, B, int ldb, long strideB, int batchCount, invA, int invAsize, long strideInvA, computeType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- side (
hipblasSideMode_t): (undocumented)
- uplo (
hipblasFillMode_t): (undocumented)
- transA (
hipblasOperation_t): (undocumented)
- diag (
hipblasDiagType_t): (undocumented)
- m (
int): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- A (
Pointer/object): (undocumented)
- lda (
int): (undocumented)
- strideA (
int): (undocumented)
- B (
Pointer/object): (undocumented)
- ldb (
int): (undocumented)
- strideB (
int): (undocumented)
- batchCount (
int): (undocumented)
- invA (
Pointer/object): (undocumented)
- invAsize (
int): (undocumented)
- strideInvA (
int): (undocumented)
- computeType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasAxpyEx(handle, int n, alpha, alphaType, x, xType, int incx, y, yType, int incy, executionType)#
BLAS EX API
axpyEx computes constant alpha multiplied by vector x, plus vector y
y := alpha * x + y
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
With HIPBLAS_V2 define, hipblasAxpyEx accepts hipDataType for alphaType, xType, yType, and executionType rather than hipblasDatatype_t. hipblasAxpyEx will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasAxpyEx(hipblasHandle_t handle,
int n, const void* alpha, hipDataType alphaType, const void* x, hipDataType xType, int incx, void* y, hipDataType yType, int incy, hipDataType executionType)
else// [DEPRECATED]- hipblasStatus_t hipblasAxpyEx(hipblasHandle_t handle,
int n, const void* alpha, hipblasDatatype_t alphaType, const void* x, hipblasDatatype_t xType, int incx, void* y, hipblasDatatype_t yType, int incy, hipblasDatatype_t executionType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x and y.
- alpha (
Pointer/object) – IN: device pointer or host pointer to specify the scalar alpha.
- alphaType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of alpha.
- [hipDataType]
specifies the datatype of alpha.
- x (
Pointer/object) – IN: device pointer storing vector x.
- xType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of vector x.
- [hipDataType]
specifies the datatype of vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- y (
Pointer/object) – INOUT: device pointer storing vector y.
- yType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of vector y.
- [hipDataType]
specifies the datatype of vector y.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- executionType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasAxpyEx_v2(handle, int n, alpha, alphaType, x, xType, int incx, y, yType, int incy, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- alphaType (
hipDataType): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- yType (
hipDataType): (undocumented)
- incy (
int): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasAxpyBatchedEx(handle, int n, alpha, alphaType, x, xType, int incx, y, yType, int incy, int batchCount, executionType)#
BLAS EX API
axpyBatchedEx computes constant alpha multiplied by vector x, plus vector y over a set of batched vectors.
y := alpha * x + y
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
With HIPBLAS_V2 define, hipblasAxpyBatchedEx accepts hipDataType for alphaType, xType, yType, and executionType rather than hipblasDatatype_t. hipblasAxpyBatchedEx will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasAxpyBatchedEx(hipblasHandle_t handle,
int n, const void* alpha, hipDataType alphaType, const void* x, hipDataType xType, int incx, void* y, hipDataType yType, int incy, int batchCount, hipDataType executionType)
else// [DEPRECATED]- hipblasStatus_t hipblasAxpyBatchedEx(hipblasHandle_t handle,
int n, const void* alpha, hipblasDatatype_t alphaType, const void* x, hipblasDatatype_t xType, int incx, void* y, hipblasDatatype_t yType, int incy, int batchCount, hipblasDatatype_t executionType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in each x_i and y_i.
- alpha (
Pointer/object) – IN: device pointer or host pointer to specify the scalar alpha.
- alphaType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of alpha.
- [hipDataType]
specifies the datatype of alpha.
- x (
Pointer/object) – IN: device array of device pointers storing each vector x_i.
- xType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector x_i.
- [hipDataType]
specifies the datatype of each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- y (
Pointer/object) – INOUT: device array of device pointers storing each vector y_i.
- yType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector y_i.
- [hipDataType]
specifies the datatype of each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- batchCount (
int) – IN: [int] number of instances in the batch.
- executionType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasAxpyBatchedEx_v2(handle, int n, alpha, alphaType, x, xType, int incx, y, yType, int incy, int batchCount, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- alphaType (
hipDataType): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- yType (
hipDataType): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasAxpyStridedBatchedEx(handle, int n, alpha, alphaType, x, xType, int incx, long stridex, y, yType, int incy, long stridey, int batchCount, executionType)#
BLAS EX API
axpyStridedBatchedEx computes constant alpha multiplied by vector x, plus vector y over a set of strided batched vectors.
y := alpha * x + y
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
With HIPBLAS_V2 define, hipblasAxpyStridedBatchedEx accepts hipDataType for alphaType, xType, yType, and executionType rather than hipblasDatatype_t. hipblasAxpyStridedBatchedEx will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasAxpyStridedBatchedEx(hipblasHandle_t handle,
int n, const void* alpha, hipDataType alphaType, const void* x, hipDataType xType, int incx, hipblasStride stridex, void* y, hipDataType yType, int incy, hipblasStride stridey, int batchCount, hipDataType executionType)
else// [DEPRECATED]- hipblasStatus_t hipblasAxpyStridedBatchedEx(hipblasHandle_t handle,
int n, const void* alpha, hipblasDatatype_t alphaType, const void* x, hipblasDatatype_t xType, int incx, hipblasStride stridex, void* y, hipblasDatatype_t yType, int incy, hipblasStride stridey, int batchCount, hipblasDatatype_t executionType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in each x_i and y_i.
- alpha (
Pointer/object) – IN: device pointer or host pointer to specify the scalar alpha.
- alphaType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of alpha.
- [hipDataType]
specifies the datatype of alpha.
- x (
Pointer/object) – IN: device pointer to the first vector x_1.
- xType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector x_i.
- [hipDataType]
specifies the datatype of each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) to the next one (x_i+1). There are no restrictions placed on stridex, however the user should take care to ensure that stridex is of appropriate size, for a typical case this means stridex >= n * incx.
- y (
Pointer/object) – INOUT: device pointer to the first vector y_1.
- yType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector y_i.
- [hipDataType]
specifies the datatype of each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) to the next one (y_i+1). There are no restrictions placed on stridey, however the user should take care to ensure that stridey is of appropriate size, for a typical case this means stridey >= n * incy.
- batchCount (
int) – IN: [int] number of instances in the batch.
- executionType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasAxpyStridedBatchedEx_v2(handle, int n, alpha, alphaType, x, xType, int incx, long stridex, y, yType, int incy, long stridey, int batchCount, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- alphaType (
hipDataType): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- yType (
hipDataType): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasDotEx(handle, int n, x, xType, int incx, y, yType, int incy, result, resultType, executionType)#
BLAS EX API
@{
dotEx performs the dot product of vectors x and y
result = x * y;
dotcEx performs the dot product of the conjugate of complex vector x and complex vector y
result = conjugate (x) * y;
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
With HIPBLAS_V2 define, hipblasDot(c)Ex accepts hipDataType for xType, yType, resultType, and executionType rather than hipblasDatatype_t. hipblasDot(c)Ex will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasDotEx(hipblasHandle_t handle,
int n, const void* x, hipDataType xType, int incx, const void* y, hipDataType yType, int incy, void* result, hipDataType resultType, hipDataType executionType)
else// [DEPRECATED]- hipblasStatus_t hipblasDotEx(hipblasHandle_t handle,
int n, const void* x, hipblasDatatype_t xType, int incx, const void* y, hipblasDatatype_t yType, int incy, void* result, hipblasDatatype_t resultType, hipblasDatatype_t executionType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x and y.
- x (
Pointer/object) – IN: device pointer storing vector x.
- xType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of vector x.
- [hipDataType]
specifies the datatype of vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of y.
- y (
Pointer/object) – IN: device pointer storing vector y.
- yType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of vector y.
- [hipDataType]
specifies the datatype of vector y.
- incy (
int) – IN: [int] specifies the increment for the elements of y.
- result (
Pointer/object) – INOUT: device pointer or host pointer to store the dot product. return is 0.0 if n <= 0.
- resultType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of the result.
- [hipDataType]
specifies the datatype of the result.
- executionType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasDotcEx(handle, int n, x, xType, int incx, y, yType, int incy, result, resultType, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipblasDatatype_t): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- yType (
hipblasDatatype_t): (undocumented)
- incy (
int): (undocumented)
- result (
Pointer/object): (undocumented)
- resultType (
hipblasDatatype_t): (undocumented)
- executionType (
hipblasDatatype_t): (undocumented)
- handle (
- hip.hipblas.hipblasDotEx_v2(handle, int n, x, xType, int incx, y, yType, int incy, result, resultType, executionType)#
(No short description, might be part of a group.)
@}
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- yType (
hipDataType): (undocumented)
- incy (
int): (undocumented)
- result (
Pointer/object): (undocumented)
- resultType (
hipDataType): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasDotcEx_v2(handle, int n, x, xType, int incx, y, yType, int incy, result, resultType, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- yType (
hipDataType): (undocumented)
- incy (
int): (undocumented)
- result (
Pointer/object): (undocumented)
- resultType (
hipDataType): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasDotBatchedEx(handle, int n, x, xType, int incx, y, yType, int incy, int batchCount, result, resultType, executionType)#
BLAS EX API
@{
dotBatchedEx performs a batch of dot products of vectors x and y
result_i = x_i * y_i;
dotcBatchedEx performs a batch of dot products of the conjugate of complex vector x and complex vector y
result_i = conjugate (x_i) * y_i;
where (x_i, y_i) is the i-th instance of the batch. x_i and y_i are vectors, for i = 1, …, batchCount
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
With HIPBLAS_V2 define, hipblasDot(c)BatchedEx accepts hipDataType for xType, yType, resultType, and executionType rather than hipblasDatatype_t. hipblasDot(c)BatchedEx will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasDotBatchedEx(hipblasHandle_t handle,
int n, const void* x, hipDataType xType, int incx, const void* y, hipDataType yType, int incy, int batchCount, void* result, hipDataType resultType, hipDataType executionType)
else// [DEPRECATED]- hipblasStatus_t hipblasDotBatchedEx(hipblasHandle_t handle,
int n, const void* x, hipblasDatatype_t xType, int incx, const void* y, hipblasDatatype_t yType, int incy, int batchCount, void* result, hipblasDatatype_t resultType, hipblasDatatype_t executionType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in each x_i and y_i.
- x (
Pointer/object) – IN: device array of device pointers storing each vector x_i.
- xType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector x_i.
- [hipDataType]
specifies the datatype of each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- y (
Pointer/object) – IN: device array of device pointers storing each vector y_i.
- yType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector y_i.
- [hipDataType]
specifies the datatype of each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- batchCount (
int) – IN: [int] number of instances in the batch
- result (
Pointer/object) – INOUT: device array or host array of batchCount size to store the dot products of each batch. return 0.0 for each element if n <= 0.
- resultType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of the result.
- [hipDataType]
specifies the datatype of the result.
- executionType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasDotcBatchedEx(handle, int n, x, xType, int incx, y, yType, int incy, int batchCount, result, resultType, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipblasDatatype_t): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- yType (
hipblasDatatype_t): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
Pointer/object): (undocumented)
- resultType (
hipblasDatatype_t): (undocumented)
- executionType (
hipblasDatatype_t): (undocumented)
- handle (
- hip.hipblas.hipblasDotBatchedEx_v2(handle, int n, x, xType, int incx, y, yType, int incy, int batchCount, result, resultType, executionType)#
(No short description, might be part of a group.)
@}
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- yType (
hipDataType): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
Pointer/object): (undocumented)
- resultType (
hipDataType): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasDotcBatchedEx_v2(handle, int n, x, xType, int incx, y, yType, int incy, int batchCount, result, resultType, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- yType (
hipDataType): (undocumented)
- incy (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
Pointer/object): (undocumented)
- resultType (
hipDataType): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasDotStridedBatchedEx(handle, int n, x, xType, int incx, long stridex, y, yType, int incy, long stridey, int batchCount, result, resultType, executionType)#
BLAS EX API
@{
dotStridedBatchedEx performs a batch of dot products of vectors x and y
result_i = x_i * y_i;
dotc_strided_batched_ex performs a batch of dot products of the conjugate of complex vector x and complex vector y
result_i = conjugate (x_i) * y_i;
where (x_i, y_i) is the i-th instance of the batch. x_i and y_i are vectors, for i = 1, …, batchCount
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
With HIPBLAS_V2 define, hipblasDot(c)StridedBatchedEx accepts hipDataType for xType, yType, resultType, and executionType rather than hipblasDatatype_t. hipblasDot(c)StridedBatchedEx will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasDotStridedBatchedEx(hipblasHandle_t handle,
int n, const void* x, hipDataType xType, int incx, hipblasStride stridex, const void* y, hipDataType yType, int incy, hipblasStride stridey, int batchCount, void* result, hipDataType resultType, hipDataType executionType)
else// [DEPRECATED]- hipblasStatus_t hipblasDotStridedBatchedEx(hipblasHandle_t handle,
int n, const void* x, hipblasDatatype_t xType, int incx, hipblasStride stridex, const void* y, hipblasDatatype_t yType, int incy, hipblasStride stridey, int batchCount, void* result, hipblasDatatype_t resultType, hipblasDatatype_t executionType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in each x_i and y_i.
- x (
Pointer/object) – IN: device pointer to the first vector (x_1) in the batch.
- xType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector x_i.
- [hipDataType]
specifies the datatype of each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1)
- y (
Pointer/object) – IN: device pointer to the first vector (y_1) in the batch.
- yType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector y_i.
- [hipDataType]
specifies the datatype of each vector y_i.
- incy (
int) – IN: [int] specifies the increment for the elements of each y_i.
- stridey (
int) – IN: [hipblasStride] stride from the start of one vector (y_i) and the next one (y_i+1)
- batchCount (
int) – IN: [int] number of instances in the batch
- result (
Pointer/object) – INOUT: device array or host array of batchCount size to store the dot products of each batch. return 0.0 for each element if n <= 0.
- resultType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of the result.
- [hipDataType]
specifies the datatype of the result.
- executionType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasDotcStridedBatchedEx(handle, int n, x, xType, int incx, long stridex, y, yType, int incy, long stridey, int batchCount, result, resultType, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipblasDatatype_t): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- yType (
hipblasDatatype_t): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
Pointer/object): (undocumented)
- resultType (
hipblasDatatype_t): (undocumented)
- executionType (
hipblasDatatype_t): (undocumented)
- handle (
- hip.hipblas.hipblasDotStridedBatchedEx_v2(handle, int n, x, xType, int incx, long stridex, y, yType, int incy, long stridey, int batchCount, result, resultType, executionType)#
(No short description, might be part of a group.)
@}
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- yType (
hipDataType): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
Pointer/object): (undocumented)
- resultType (
hipDataType): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasDotcStridedBatchedEx_v2(handle, int n, x, xType, int incx, long stridex, y, yType, int incy, long stridey, int batchCount, result, resultType, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- yType (
hipDataType): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
Pointer/object): (undocumented)
- resultType (
hipDataType): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasNrm2Ex(handle, int n, x, xType, int incx, result, resultType, executionType)#
BLAS_EX API
nrm2Ex computes the euclidean norm of a real or complex vector
- result := sqrt( x’*x ) for real vectors
result := sqrt( x**H*x ) for complex vectors
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
With HIPBLAS_V2 define, hipblasNrm2Ex accepts hipDataType for xType, resultType, and executionType rather than hipblasDatatype_t. hipblasNrm2Ex will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasNrm2Ex(hipblasHandle_t handle,
int n, const void* x, hipDataType xType, int incx, void* result, hipDataType resultType, hipDataType executionType)
else// [DEPRECATED]- hipblasStatus_t hipblasNrm2Ex(hipblasHandle_t handle,
int n, const void* x, hipblasDatatype_t xType, int incx, void* result, hipblasDatatype_t resultType, hipblasDatatype_t executionType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x.
- x (
Pointer/object) – IN: device pointer storing vector x.
- xType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of the vector x.
- [hipDataType]
specifies the datatype of the vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of y.
- result (
Pointer/object) – INOUT: device pointer or host pointer to store the nrm2 product. return is 0.0 if n, incx<=0.
- resultType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of the result.
- [hipDataType]
specifies the datatype of the result.
- executionType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasNrm2Ex_v2(handle, int n, x, xType, int incx, result, resultType, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- result (
Pointer/object): (undocumented)
- resultType (
hipDataType): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasNrm2BatchedEx(handle, int n, x, xType, int incx, int batchCount, result, resultType, executionType)#
BLAS_EX API
nrm2BatchedEx computes the euclidean norm over a batch of real or complex vectors
- result := sqrt( x_i’*x_i ) for real vectors x, for i = 1, …, batchCount
result := sqrt( x_i**H*x_i ) for complex vectors x, for i = 1, …, batchCount
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
With HIPBLAS_V2 define, hipblasNrm2BatchedEx accepts hipDataType for xType, resultType, and executionType rather than hipblasDatatype_t. hipblasNrm2BatchedEx will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasNrm2BatchedEx(hipblasHandle_t handle,
int n, const void* x, hipDataType xType, int incx, int batchCount, void* result, hipDataType resultType, hipDataType executionType)
else// [DEPRECATED]- hipblasStatus_t hipblasNrm2BatchedEx(hipblasHandle_t handle,
int n, const void* x, hipblasDatatype_t xType, int incx, int batchCount, void* result, hipblasDatatype_t resultType, hipblasDatatype_t executionType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in each x_i.
- x (
Pointer/object) – IN: device array of device pointers storing each vector x_i.
- xType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector x_i.
- [hipDataType]
specifies the datatype of each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i. incx must be > 0.
- batchCount (
int) – IN: [int] number of instances in the batch
- result (
Pointer/object) – OUT: device pointer or host pointer to array of batchCount size for nrm2 results. return is 0.0 for each element if n <= 0, incx<=0.
- resultType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of the result.
- [hipDataType]
specifies the datatype of the result.
- executionType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasNrm2BatchedEx_v2(handle, int n, x, xType, int incx, int batchCount, result, resultType, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
Pointer/object): (undocumented)
- resultType (
hipDataType): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasNrm2StridedBatchedEx(handle, int n, x, xType, int incx, long stridex, int batchCount, result, resultType, executionType)#
BLAS_EX API
nrm2StridedBatchedEx computes the euclidean norm over a batch of real or complex vectors
- = sqrt( x_i’*x_i ) for real vectors x, for i = 1, …, batchCount
:= sqrt( x_i**H*x_i ) for complex vectors, for i = 1, …, batchCount
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
With HIPBLAS_V2 define, hipblasNrm2StridedBatchedEx accepts hipDataType for xType, resultType, and executionType rather than hipblasDatatype_t. hipblasNrm2StridedBatchedEx will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasNrm2StridedBatchedEx(hipblasHandle_t handle,
int n, const void* x, hipDataType xType, int incx, hipblasStride stridex, int batchCount, void* result, hipDataType resultType, hipDataType executionType)
else// [DEPRECATED]- hipblasStatus_t hipblasNrm2StridedBatchedEx(hipblasHandle_t handle,
int n, const void* x, hipblasDatatype_t xType, int incx, hipblasStride stridex, int batchCount, void* result, hipblasDatatype_t resultType, hipblasDatatype_t executionType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in each x_i.
- x (
Pointer/object) – IN: device pointer to the first vector x_1.
- xType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector x_i.
- [hipDataType]
specifies the datatype of each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i. incx must be > 0.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) and the next one (x_i+1). There are no restrictions placed on stride_x, however the user should take care to ensure that stride_x is of appropriate size, for a typical case this means stride_x >= n * incx.
- batchCount (
int) – IN: [int] number of instances in the batch
- result (
Pointer/object) – OUT: device pointer or host pointer to array for storing contiguous batchCount results. return is 0.0 for each element if n <= 0, incx<=0.
- resultType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of the result.
- [hipDataType]
specifies the datatype of the result.
- executionType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasNrm2StridedBatchedEx_v2(handle, int n, x, xType, int incx, long stridex, int batchCount, result, resultType, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- result (
Pointer/object): (undocumented)
- resultType (
hipDataType): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasRotEx(handle, int n, x, xType, int incx, y, yType, int incy, c, s, csType, executionType)#
BLAS EX API
rotEx applies the Givens rotation matrix defined by c=cos(alpha) and s=sin(alpha) to vectors x and y. Scalars c and s may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode.
- In the case where cs_type is real:
- x := c * x + s * y
y := c * y - s * x
- In the case where cs_type is complex, the imaginary part of c is ignored:
- x := real(c) * x + s * y
y := real(c) * y - conj(s) * x
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
With HIPBLAS_V2 define, hipblasRotEx accepts hipDataType for xType, yType, csType, and executionType rather than hipblasDatatype_t. hipblasRotEx will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasRotEx(hipblasHandle_t handle,
int n, void* x, hipDataType xType, int incx, void* y, hipDataType yType, int incy, const void* c, const void* s, hipDataType csType, hipDataType executionType)
else// [DEPRECATED]- hipblasStatus_t hipblasRotEx(hipblasHandle_t handle,
int n, void* x, hipblasDatatype_t xType, int incx, void* y, hipblasDatatype_t yType, int incy, const void* c, const void* s, hipblasDatatype_t csType, hipblasDatatype_t executionType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in the x and y vectors.
- x (
Pointer/object) – INOUT: device pointer storing vector x.
- xType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of vector x.
- [hipDataType]
specifies the datatype of vector x.
- incx (
int) – IN: [int] specifies the increment between elements of x.
- y (
Pointer/object) – INOUT: device pointer storing vector y.
- yType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of vector y.
- [hipDataType]
specifies the datatype of vector y.
- incy (
int) – IN: [int] specifies the increment between elements of y.
- c (
Pointer/object) – IN: device pointer or host pointer storing scalar cosine component of the rotation matrix.
- s (
Pointer/object) – IN: device pointer or host pointer storing scalar sine component of the rotation matrix.
- csType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of c and s.
- [hipDataType]
specifies the datatype of c and s.
- executionType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasRotEx_v2(handle, int n, x, xType, int incx, y, yType, int incy, c, s, csType, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- yType (
hipDataType): (undocumented)
- incy (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
Pointer/object): (undocumented)
- csType (
hipDataType): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasRotBatchedEx(handle, int n, x, xType, int incx, y, yType, int incy, c, s, csType, int batchCount, executionType)#
BLAS EX API
rotBatchedEx applies the Givens rotation matrix defined by c=cos(alpha) and s=sin(alpha) to batched vectors x_i and y_i, for i = 1, …, batchCount. Scalars c and s may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode.
- In the case where cs_type is real:
x := c * x + s * y y := c * y - s * x
- In the case where cs_type is complex, the imaginary part of c is ignored:
x := real(c) * x + s * y y := real(c) * y - conj(s) * x
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
With HIPBLAS_V2 define, hipblasRotBatchedEx accepts hipDataType for xType, yType, csType, and executionType rather than hipblasDatatype_t. hipblasRotBatchedEx will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasRotBatchedEx(hipblasHandle_t handle,
int n, void* x, hipDataType xType, int incx, void* y, hipDataType yType, int incy, const void* c, const void* s, hipDataType csType, int batchCount, hipDataType executionType)
else// [DEPRECATED]- hipblasStatus_t hipblasRotBatchedEx(hipblasHandle_t handle,
int n, void* x, hipblasDatatype_t xType, int incx, void* y, hipblasDatatype_t yType, int incy, const void* c, const void* s, hipblasDatatype_t csType, int batchCount, hipblasDatatype_t executionType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in each x_i and y_i vectors.
- x (
Pointer/object) – INOUT: device array of device pointers storing each vector x_i.
- xType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector x_i.
- [hipDataType]
specifies the datatype of each vector x_i.
- incx (
int) – IN: [int] specifies the increment between elements of each x_i.
- y (
Pointer/object) – INOUT: device array of device pointers storing each vector y_i.
- yType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector y_i.
- [hipDataType]
specifies the datatype of each vector y_i.
- incy (
int) – IN: [int] specifies the increment between elements of each y_i.
- c (
Pointer/object) – IN: device pointer or host pointer to scalar cosine component of the rotation matrix.
- s (
Pointer/object) – IN: device pointer or host pointer to scalar sine component of the rotation matrix.
- csType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of c and s.
- [hipDataType]
specifies the datatype of c and s.
- batchCount (
int) – IN: [int] the number of x and y arrays, i.e. the number of batches.
- executionType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasRotBatchedEx_v2(handle, int n, x, xType, int incx, y, yType, int incy, c, s, csType, int batchCount, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- yType (
hipDataType): (undocumented)
- incy (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
Pointer/object): (undocumented)
- csType (
hipDataType): (undocumented)
- batchCount (
int): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasRotStridedBatchedEx(handle, int n, x, xType, int incx, long stridex, y, yType, int incy, long stridey, c, s, csType, int batchCount, executionType)#
BLAS Level 1 API
rotStridedBatchedEx applies the Givens rotation matrix defined by c=cos(alpha) and s=sin(alpha) to strided batched vectors x_i and y_i, for i = 1, …, batchCount. Scalars c and s may be stored in either host or device memory, location is specified by calling hipblasSetPointerMode.
- In the case where cs_type is real:
x := c * x + s * y y := c * y - s * x
- In the case where cs_type is complex, the imaginary part of c is ignored:
x := real(c) * x + s * y y := real(c) * y - conj(s) * x
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
With HIPBLAS_V2 define, hipblasRotStridedBatchedEx accepts hipDataType for xType, yType, csType, and executionType rather than hipblasDatatype_t. hipblasRotStridedBatchedEx will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasRotStridedBatchedEx(hipblasHandle_t handle,
int n, void* x, hipDataType xType, int incx, hipblasStride stridex, void* y, hipDataType yType, int incy, hipblasStride stridey, const void* c, const void* s, hipDataType csType, int batchCount, hipDataType executionType)
else// [DEPRECATED]- hipblasStatus_t hipblasRotStridedBatchedEx(hipblasHandle_t handle,
int n, void* x, hipblasDatatype_t xType, int incx, hipblasStride stridex, void* y, hipblasDatatype_t yType, int incy, hipblasStride stridey, const void* c, const void* s, hipblasDatatype_t csType, int batchCount, hipblasDatatype_t executionType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] number of elements in each x_i and y_i vectors.
- x (
Pointer/object) – INOUT: device pointer to the first vector x_1.
- xType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector x_i.
- [hipDataType]
specifies the datatype of each vector x_i.
- incx (
int) – IN: [int] specifies the increment between elements of each x_i.
- stridex (
int) – IN: [hipblasStride] specifies the increment from the beginning of x_i to the beginning of x_(i+1)
- y (
Pointer/object) – INOUT: device pointer to the first vector y_1.
- yType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector y_i.
- [hipDataType]
specifies the datatype of each vector y_i.
- incy (
int) – IN: [int] specifies the increment between elements of each y_i.
- stridey (
int) – IN: [hipblasStride] specifies the increment from the beginning of y_i to the beginning of y_(i+1)
- c (
Pointer/object) – IN: device pointer or host pointer to scalar cosine component of the rotation matrix.
- s (
Pointer/object) – IN: device pointer or host pointer to scalar sine component of the rotation matrix.
- csType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of c and s.
- [hipDataType]
specifies the datatype of c and s.
- batchCount (
int) – IN: [int] the number of x and y arrays, i.e. the number of batches.
- executionType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasRotStridedBatchedEx_v2(handle, int n, x, xType, int incx, long stridex, y, yType, int incy, long stridey, c, s, csType, int batchCount, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- y (
Pointer/object): (undocumented)
- yType (
hipDataType): (undocumented)
- incy (
int): (undocumented)
- stridey (
int): (undocumented)
- c (
Pointer/object): (undocumented)
- s (
Pointer/object): (undocumented)
- csType (
hipDataType): (undocumented)
- batchCount (
int): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasScalEx(handle, int n, alpha, alphaType, x, xType, int incx, executionType)#
BLAS EX API
scalEx scales each element of vector x with scalar alpha.
x := alpha * x
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
With HIPBLAS_V2 define, hipblasScalEx accepts hipDataType for alphaType, xType, and executionType rather than hipblasDatatype_t. hipblasScalEx will only accept hipDataType in a future release.
ifdefHIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2- hipblasStatus_t hipblasScalEx(hipblasHandle_t handle,a
int n, const void* alpha, hipDataType alphaType, void* x, hipDataType xType, int incx, hipDataType executionType)
else// [DEPRECATED]- hipblasStatus_t hipblasScalEx(hipblasHandle_t handle,
int n, const void* alpha, hipblasDatatype_t alphaType, void* x, hipblasDatatype_t xType, int incx, hipblasDatatype_t executionType)
endif- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x.
- alpha (
Pointer/object) – IN: device pointer or host pointer for the scalar alpha.
- alphaType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of alpha.
- [hipDataType]
specifies the datatype of alpha.
- x (
Pointer/object) – INOUT: device pointer storing vector x.
- xType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of vector x.
- [hipDataType]
specifies the datatype of vector x.
- incx (
int) – IN: [int] specifies the increment for the elements of x.
- executionType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasScalEx_v2(handle, int n, alpha, alphaType, x, xType, int incx, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- alphaType (
hipDataType): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasScalBatchedEx(handle, int n, alpha, alphaType, x, xType, int incx, int batchCount, executionType)#
BLAS EX API
scalBatchedEx scales each element of each vector x_i with scalar alpha.
x_i := alpha * x_i
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
With HIPBLAS_V2 define, hipblasScalBatchedEx accepts hipDataType for alphaType, xType, and executionType rather than hipblasDatatype_t. hipblasScalBatchedEx will only accept hipDataType in a future release.
#ifdef HIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2 hipblasStatus_t hipblasScalBatchedEx(hipblasHandle_t handle, int n, const void* alpha, hipDataType alphaType, void* x, hipDataType xType, int incx, int batchCount, hipDataType executionType) #else // [DEPRECATED] hipblasStatus_t hipblasScalBatchedEx(hipblasHandle_t handle, int n, const void* alpha, hipblasDatatype_t alphaType, void* x, hipblasDatatype_t xType, int incx, int batchCount, hipblasDatatype_t executionType) #endif
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x.
- alpha (
Pointer/object) – IN: device pointer or host pointer for the scalar alpha.
- alphaType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of alpha.
- [hipDataType]
specifies the datatype of alpha.
- x (
Pointer/object) – INOUT: device array of device pointers storing each vector x_i.
- xType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector x_i.
- [hipDataType]
specifies the datatype of each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- batchCount (
int) – IN: [int] number of instances in the batch.
- executionType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasScalBatchedEx_v2(handle, int n, alpha, alphaType, x, xType, int incx, int batchCount, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- alphaType (
hipDataType): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- batchCount (
int): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasScalStridedBatchedEx(handle, int n, alpha, alphaType, x, xType, int incx, long stridex, int batchCount, executionType)#
BLAS EX API
scalStridedBatchedEx scales each element of vector x with scalar alpha over a set of strided batched vectors.
x := alpha * x
Supported types are determined by the backend. See rocBLAS/cuBLAS documentation.
With HIPBLAS_V2 define, hipblasScalStridedBatchedEx accepts hipDataType for alphaType, xType, and executionType rather than hipblasDatatype_t. hipblasScalStridedBatchedEx will only accept hipDataType in a future release.
#ifdef HIPBLAS_V2 // available in hipBLAS version 2.0.0 and later with -DHIPBLAS_V2 hipblasStatus_t hipblasScalStridedBatchedEx(hipblasHandle_t handle, int n, const void* alpha, hipDataType alphaType, void* x, hipDataType xType, int incx, hipblasStride stridex, int batchCount, hipDataType executionType) #else // [DEPRECATED] hipblasStatus_t hipblasScalStridedBatchedEx(hipblasHandle_t handle, int n, const void* alpha, hipblasDatatype_t alphaType, void* x, hipblasDatatype_t xType, int incx, hipblasStride stridex, int batchCount, hipblasDatatype_t executionType) #endif
- Args:
- handle (
Pointer/object) – IN: [hipblasHandle_t] handle to the hipblas library context queue.
- n (
int) – IN: [int] the number of elements in x.
- alpha (
Pointer/object) – IN: device pointer or host pointer for the scalar alpha.
- alphaType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of alpha.
- [hipDataType]
specifies the datatype of alpha.
- x (
Pointer/object) – INOUT: device pointer to the first vector x_1.
- xType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of each vector x_i.
- [hipDataType]
specifies the datatype of each vector x_i.
- incx (
int) – IN: [int] specifies the increment for the elements of each x_i.
- stridex (
int) – IN: [hipblasStride] stride from the start of one vector (x_i) to the next one (x_i+1). There are no restrictions placed on stridex, however the user should take care to ensure that stridex is of appropriate size, for a typical case this means stridex >= n * incx.
- batchCount (
int) – IN: [int] number of instances in the batch.
- executionType (
hipblasDatatype_t) – IN: - [hipblasDatatype_t] [DEPRECATED]
specifies the datatype of computation.
- [hipDataType]
specifies the datatype of computation.
- handle (
- hip.hipblas.hipblasScalStridedBatchedEx_v2(handle, int n, alpha, alphaType, x, xType, int incx, long stridex, int batchCount, executionType)#
(No short description, might be part of a group.)
- Args:
- handle (
Pointer/object): (undocumented)
- n (
int): (undocumented)
- alpha (
Pointer/object): (undocumented)
- alphaType (
hipDataType): (undocumented)
- x (
Pointer/object): (undocumented)
- xType (
hipDataType): (undocumented)
- incx (
int): (undocumented)
- stridex (
int): (undocumented)
- batchCount (
int): (undocumented)
- executionType (
hipDataType): (undocumented)
- handle (
- hip.hipblas.hipblasStatusToString(status)#
(No short description, might be part of a group.)
HIPBLAS Auxiliary API
hipblasStatusToString
Returns string representing hipblasStatus_t value
- Args:
- status (
hipblasStatus_t) – IN: [hipblasStatus_t] hipBLAS status to convert to string
- status (