Compute Library
CpuGemmMatrixMultiplyKernel Class Reference

Kernel to multiply two input matrices "A" and "B". More...

#include <CpuGemmMatrixMultiplyKernel.h>

Collaboration diagram for CpuGemmMatrixMultiplyKernel:

Data Structures

struct  GemmMatrixMulKernel

Public Member Functions

 CpuGemmMatrixMultiplyKernel ()=default
void configure (const ITensorInfo *lhs, const ITensorInfo *rhs, ITensorInfo *dst, float alpha, bool is_interleaved, const GEMMReshapeInfo &reshape_info=GEMMReshapeInfo())
 Initialise the kernel's input and output. More...
void run_op (ITensorPack &tensors, const Window &window, const ThreadInfo &info) override
 Execute the kernel on the passed window. More...
const char * name () const override
 Name of the kernel. More...
- Public Member Functions inherited from ICPPKernel
virtual ~ICPPKernel ()=default
 Default destructor. More...
virtual void run (const Window &window, const ThreadInfo &info)
 Execute the kernel on the passed window. More...
virtual void run_nd (const Window &window, const ThreadInfo &info, const Window &thread_locator)
 legacy compatibility layer for implemantions which do not support thread_locator In these cases we simply narrow the interface down the legacy version More...
virtual size_t get_mws (const CPUInfo &platform, size_t thread_count) const
 Return minimum workload size of the relevant kernel. More...
- Public Member Functions inherited from IKernel
 IKernel ()
 Constructor. More...
virtual ~IKernel ()=default
 Destructor. More...
virtual bool is_parallelisable () const
 Indicates whether or not the kernel is parallelisable. More...
virtual BorderSize border_size () const
 The size of the border for that kernel. More...
const Windowwindow () const
 The maximum window the kernel can be executed on. More...
bool is_window_configured () const
 Function to check if the embedded window of this kernel has been configured. More...

Static Public Member Functions

static Status validate (const ITensorInfo *lhs, const ITensorInfo *rhs, const ITensorInfo *dst, float alpha, bool is_interleaved, const GEMMReshapeInfo &reshape_info)
 Static function to check if given info will lead to a valid configuration of CpuGemmMatrixMultiplyKernel. More...
static const std::vector< GemmMatrixMulKernel > & get_available_kernels ()
- Static Public Member Functions inherited from ICpuKernel< CpuGemmMatrixMultiplyKernel >
static const auto * get_implementation (const SelectorType &selector, KernelSelectionType selection_type=KernelSelectionType::Supported)
 Micro-kernel selector. More...

Additional Inherited Members

- Static Public Attributes inherited from ICPPKernel
static constexpr size_t default_mws = 1

Detailed Description

Kernel to multiply two input matrices "A" and "B".

All elements of the output matrix/vector will be multiplied by alpha after the matrix multiplication

If the output tensor is a matrix, the implementation assumes that the input tensors lhs and rhs are both matrices and reshaped respectively with CpuGemmInterleave4x4Kernel" and CpuGemmTranspose1xWKernel
If the output tensor is a vector and the data type is F32, the implementation assumes that the first input tensor lhs is a vector and the second input tensor rhs a matrix. The implementation also assumes that both tensors have not been reshaped

Definition at line 42 of file CpuGemmMatrixMultiplyKernel.h.

Constructor & Destructor Documentation

◆ CpuGemmMatrixMultiplyKernel()

Member Function Documentation



◆ configure()

void configure ( const ITensorInfo lhs,
const ITensorInfo rhs,
ITensorInfo dst,
float  alpha,
bool  is_interleaved,
const GEMMReshapeInfo reshape_info = GEMMReshapeInfo() 

Initialise the kernel's input and output.

If the output tensor is a matrix, the input matrices lhs and rhs should be the output of the kernels: CpuGemmInterleave4x4Kernel and CpuGemmTranspose1xWKernel These two kernels change the layout of the original matrices to be more cache-friendly.
[in]lhsLeft-handside tensor info containing the interleaved Matrix A or the vector A. Data types supported: F16/F32
[in]rhsRight-handside tensor info containing the transposed Matrix B if the first input tensor A is not a vector. If the output tensor is a vector, rhs must contain the matrix B not reshaped. Data type supported: same as lhs
[out]dstOutput tensor to store the result of matrix multiplication. Data type supported: same as lhs.
[in]alphaWeight of the matrix product
[in]is_interleaved(Optional) True if lhs and rhs have been reshaped respectively using CpuGemmInterleave4x4Kernel and CpuGemmTranspose1xWKernel
[in]reshape_info(Optional) GEMM reshape info. If is_interleaved_transposed = true, this object must contain the information to understand how lhs and rhs have been reshaped

Definition at line 128 of file CpuGemmMatrixMultiplyKernel.cpp.

129 {
132  // dst tensor auto inizialitation if not yet initialized
133  TensorShape tensor_shape{ lhs->tensor_shape() };
134  tensor_shape.set(0, is_interleaved ? reshape_info.n() : rhs->dimension(0));
135  tensor_shape.set(1, is_interleaved ? reshape_info.m() : lhs->dimension(1));
137  auto_init_if_empty(*dst, lhs->clone()->set_tensor_shape(tensor_shape));
139  // Perform validate step
140  ARM_COMPUTE_ERROR_THROW_ON(validate_arguments(lhs, rhs, dst, alpha, is_interleaved, reshape_info));
142  _alpha = alpha;
144  // Configure kernel window
145  Window win{};
147  // Check if the dst tensor is a vector. If so,the kernel runs the vector-matrix multiplication
148  const bool is_dst_vector = (dst->dimension(1) == 1);
149  if(is_dst_vector)
150  {
151  const unsigned int num_elems_processed_per_iteration_x = (lhs->data_type() == DataType::F32) ? 16 : 32;
153  win = calculate_max_window(*dst, Steps(num_elems_processed_per_iteration_x));
154  }
155  else
156  {
157  constexpr unsigned int num_elems_processed_per_iteration_x = 8;
158  constexpr unsigned int num_elems_processed_per_iteration_y = 4;
160  win = calculate_max_window(*dst, Steps(num_elems_processed_per_iteration_x, num_elems_processed_per_iteration_y));
161  }
163  const auto uk = CpuGemmMatrixMultiplyKernel::get_implementation(DataTypeISASelectorData{ lhs->data_type(), CPUInfo::get().get_isa() });
165  _func = uk->ukernel;
167  ICPPKernel::configure(win);
168 }

References ARM_COMPUTE_ERROR_ON_NULLPTR, ARM_COMPUTE_ERROR_THROW_ON, arm_compute::auto_init_if_empty(), arm_compute::calculate_max_window(), ICloneable< T >::clone(), ITensorInfo::data_type(), ITensorInfo::dimension(), arm_compute::test::validation::dst, arm_compute::F32, CPUInfo::get(), ICpuKernel< CpuGemmMatrixMultiplyKernel >::get_implementation(), GEMMReshapeInfo::m(), GEMMReshapeInfo::n(), TensorShape::set(), ITensorInfo::tensor_shape(), and arm_compute::cpu::kernels::validate_arguments().

◆ get_available_kernels()

const std::vector< CpuGemmMatrixMultiplyKernel::GemmMatrixMulKernel > & get_available_kernels ( )

Definition at line 198 of file CpuGemmMatrixMultiplyKernel.cpp.

199 {
200  return available_kernels;
201 }

◆ name()

const char * name ( ) const

Name of the kernel.

Kernel name

Implements ICPPKernel.

Definition at line 193 of file CpuGemmMatrixMultiplyKernel.cpp.

194 {
195  return "CpuGemmMatrixMultiplyKernel";
196 }

◆ run_op()

void run_op ( ITensorPack tensors,
const Window window,
const ThreadInfo info 

Execute the kernel on the passed window.

If is_parallelisable() returns false then the passed window must be equal to window()
The window has to be a region within the window returned by the window() method
The width of the window has to be a multiple of num_elems_processed_per_iteration().
[in]tensorsA vector containing the tensors to operate on.
[in]windowRegion on which to execute the kernel. (Must be a region of the window returned by window())
[in]infoInfo about executing thread and CPU.

Reimplemented from ICPPKernel.

Definition at line 178 of file CpuGemmMatrixMultiplyKernel.cpp.

179 {
182  ARM_COMPUTE_ERROR_ON(tensors.empty());
183  ARM_COMPUTE_ERROR_ON(_func == nullptr);
185  const ITensor *lhs = tensors.get_const_tensor(TensorType::ACL_SRC_0);
186  const ITensor *rhs = tensors.get_const_tensor(TensorType::ACL_SRC_1);
187  ITensor *dst = tensors.get_tensor(TensorType::ACL_DST);
189  const bool is_dst_vector = (dst->info()->dimension(1) == 1);
190  (*_func)(lhs, rhs, dst, window, info, _alpha, is_dst_vector);
191 }

References arm_compute::ACL_DST, arm_compute::ACL_SRC_0, arm_compute::ACL_SRC_1, ARM_COMPUTE_ERROR_ON, ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW, ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL, arm_compute::test::validation::dst, ITensorPack::empty(), ITensorPack::get_const_tensor(), ITensorPack::get_tensor(), arm_compute::test::validation::info, and IKernel::window().

◆ validate()

Status validate ( const ITensorInfo lhs,
const ITensorInfo rhs,
const ITensorInfo dst,
float  alpha,
bool  is_interleaved,
const GEMMReshapeInfo reshape_info 

Static function to check if given info will lead to a valid configuration of CpuGemmMatrixMultiplyKernel.

Similar to CpuGemmMatrixMultiplyKernel::configure()

a status

Definition at line 170 of file CpuGemmMatrixMultiplyKernel.cpp.

172 {
173  ARM_COMPUTE_RETURN_ON_ERROR(validate_arguments(lhs, rhs, dst, alpha, is_interleaved, reshape_info));
175  return Status{};
176 }

References ARM_COMPUTE_RETURN_ON_ERROR, arm_compute::test::validation::dst, and arm_compute::cpu::kernels::validate_arguments().

Referenced by CpuGemm::validate().

The documentation for this class was generated from the following files:
Window calculate_max_window(const ValidRegion &valid_region, const Steps &steps, bool skip_border, BorderSize border_size)
Definition: WindowHelpers.cpp:28
auto dst
Definition: DFT.cpp:170
Status validate_arguments(const ITensorInfo *src, const ITensorInfo *weights, const ITensorInfo *dst, const PadStrideInfo &conv_info)
Definition: CpuDirectConv2dKernel.cpp:60
Definition: Validate.h:1004
static CPUInfo & get()
Access the KernelLibrary singleton.
Definition: CPPTypes.cpp:40
Definition: Types.h:45
Definition: Types.h:46
Checks if a status contains an error and returns it.
Definition: Error.h:204
Definition: Validate.h:161
#define ARM_COMPUTE_ERROR_ON(cond)
If the condition is true then an error message is printed and an exception thrown.
Definition: Error.h:467
Definition: Error.h:456
Definition: Types.h:55
bool auto_init_if_empty(ITensorInfo &info, const TensorShape &shape, int num_channels, DataType data_type, QuantizationInfo quantization_info=QuantizationInfo())
Auto initialize the tensor info (shape, number of channels and data type) if the current assignment i...
Definition: AutoConfiguration.h:43
Definition: Validate.h:205
arm_compute::cpu::ICpuKernel< CpuGemmMatrixMultiplyKernel >::get_implementation
static const auto * get_implementation(const SelectorType &selector, KernelSelectionType selection_type=KernelSelectionType::Supported)
Micro-kernel selector.
Definition: ICpuKernel.h:53
const Window & window() const
The maximum window the kernel can be executed on.
Definition: IKernel.cpp:28
@ F32
32-bit floating-point number
ScaleKernelInfo info(interpolation_policy, default_border_mode, PixelValue(), sampling_policy, false)