Compute Library
 22.08
ClGemmLowpMatrixMultiplyNativeKernel.cpp
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2019-2021 Arm Limited.
3  *
4  * SPDX-License-Identifier: MIT
5  *
6  * Permission is hereby granted, free of charge, to any person obtaining a copy
7  * of this software and associated documentation files (the "Software"), to
8  * deal in the Software without restriction, including without limitation the
9  * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
10  * sell copies of the Software, and to permit persons to whom the Software is
11  * furnished to do so, subject to the following conditions:
12  *
13  * The above copyright notice and this permission notice shall be included in all
14  * copies or substantial portions of the Software.
15  *
16  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
17  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
18  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
19  * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
20  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
21  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
22  * SOFTWARE.
23  */
25 
32 #include "arm_compute/core/Utils.h"
35 
39 
40 #include "support/Cast.h"
41 #include "support/StringSupport.h"
42 
43 namespace arm_compute
44 {
45 namespace opencl
46 {
47 namespace kernels
48 {
49 namespace
50 {
51 using ElementsProcessed = Steps;
52 
53 Status validate_arguments(const ITensorInfo *src0, const ITensorInfo *src1, const ITensorInfo *dst, const GEMMLHSMatrixInfo &lhs_info, const GEMMRHSMatrixInfo &rhs_info,
54  const GEMMReshapeInfo &gemm_info)
55 {
56  ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(src0, src1, dst);
58  if(src0->data_type() == DataType::QASYMM8)
59  {
61  }
62  else
63  {
65  }
66  ARM_COMPUTE_RETURN_ERROR_ON_MSG(src0->num_dimensions() > 4, "The number of dimensions for the LHS matrix must be <= 4");
67  ARM_COMPUTE_RETURN_ERROR_ON_MSG(src1->num_dimensions() > 3, "The number of dimensions for the RHS matrix must be <= 3");
68  ARM_COMPUTE_RETURN_ERROR_ON(lhs_info.k0 != rhs_info.k0);
69  ARM_COMPUTE_RETURN_ERROR_ON_MSG(((lhs_info.k0 & (lhs_info.k0 - 1)) && lhs_info.k0 != 3), "Only 2,3,4,8,16 are supported for k0");
70  ARM_COMPUTE_RETURN_ERROR_ON(lhs_info.k0 > 16);
71  ARM_COMPUTE_RETURN_ERROR_ON(lhs_info.m0 < 1 || lhs_info.m0 > 8);
72  ARM_COMPUTE_RETURN_ERROR_ON_MSG(((rhs_info.n0 & (rhs_info.n0 - 1)) && rhs_info.n0 != 3), "Only 2,3,4,8,16 are supported for n0");
73  ARM_COMPUTE_RETURN_ERROR_ON_MSG(rhs_info.export_to_cl_image, "Export to CLImage not supported for quantized GEMM");
74 
75  const int m = gemm_info.m();
76  const int n = gemm_info.n();
77  const int k = gemm_info.k();
78 
82 
83  ARM_COMPUTE_RETURN_ERROR_ON(src0->dimension(0) != static_cast<unsigned int>(k));
84  ARM_COMPUTE_RETURN_ERROR_ON(src1->dimension(0) != static_cast<unsigned int>(n));
85  ARM_COMPUTE_RETURN_ERROR_ON(src1->dimension(1) != static_cast<unsigned int>(k));
86  if(gemm_info.reinterpret_input_as_3d())
87  {
88  ARM_COMPUTE_RETURN_ERROR_ON(src0->dimension(1) * src0->dimension(2) != static_cast<unsigned int>(m));
89  }
90  else
91  {
92  ARM_COMPUTE_RETURN_ERROR_ON(src0->dimension(1) != static_cast<unsigned int>(m));
93  }
94 
95  if(dst->total_size() != 0)
96  {
97  const TensorInfo tensor_info_dst = dst->clone()->set_tensor_shape(misc::shape_calculator::compute_mm_shape(*src0, *src1, gemm_info));
100  }
101 
102  return Status{};
103 }
104 
105 std::pair<Status, Window> validate_and_configure_window(const ITensorInfo *src0, ITensorInfo *src1, ITensorInfo *dst, const GEMMLHSMatrixInfo &lhs_info, const GEMMRHSMatrixInfo &rhs_info,
106  const GEMMReshapeInfo &gemm_info, ElementsProcessed &num_elements_processed)
107 {
108  unsigned int &num_elems_processed_per_iteration_x = num_elements_processed[0];
109  unsigned int &num_elems_processed_per_iteration_y = num_elements_processed[1];
110  bool reinterpret_input_as_3d = gemm_info.reinterpret_input_as_3d();
111  bool reinterpret_dst_as_3d = (gemm_info.depth_output_gemm3d() != 0);
112 
113  Window win{};
114  bool window_changed = false;
115 
116  // In case both input and dst have to be reinterpreted as 3D tensors,
117  // force reinterpret_dst_as_3d to be false.
118  if(reinterpret_input_as_3d == reinterpret_dst_as_3d)
119  {
120  reinterpret_dst_as_3d = false;
121  }
122 
123  // dst tensor auto initialization if not yet initialized
124  auto_init_if_empty(*dst, src0->clone()->set_tensor_shape(misc::shape_calculator::compute_mm_shape(*src0, *src1, gemm_info)).set_data_type(DataType::S32));
125 
126  TensorInfo tmp_info(*dst);
127 
128  if(reinterpret_dst_as_3d)
129  {
130  // Since the dst tensor has to be reinterpreted as 3D and the execute window is based on a 2D GEMM,
131  // the window needs to be constructed on the 2D collapsed version of the tensor
132  TensorShape tmp_shape(dst->tensor_shape());
133  tmp_shape.collapse(2U, 1U);
134  tmp_info.set_tensor_shape(tmp_shape);
135  }
136 
137  // Configure kernel window
138  num_elems_processed_per_iteration_x = rhs_info.n0;
139  num_elems_processed_per_iteration_y = lhs_info.m0;
140 
141  win = calculate_max_window(tmp_info, Steps(num_elems_processed_per_iteration_x, num_elems_processed_per_iteration_y));
142 
143  // RHS matrix still needs padding on the X
144  AccessWindowStatic src1_access(src1, 0, 0,
145  ceil_to_multiple(src1->dimension(0), num_elems_processed_per_iteration_x),
146  src1->dimension(1));
147 
148  window_changed = update_window_and_padding(win, src1_access); // window used by the execute_window_loop
149 
150  // Collapse along the Z direction
151  // This collapse needs to be here in order to tune the Z dimension of LWS
152  Window collapsed = win;
153  const unsigned int dimension_to_collapse = std::min(static_cast<unsigned int>(dst->num_dimensions()), 2u);
154  collapsed = win.collapse(win, dimension_to_collapse);
155 
156  Status err = (window_changed) ? ARM_COMPUTE_CREATE_ERROR(ErrorCode::RUNTIME_ERROR, "Insufficient Padding!") : Status{};
157  return std::make_pair(err, collapsed);
158 }
159 } // namespace
160 
162 {
163  _type = CLKernelType::GEMM;
164 }
165 
167  const GEMMLHSMatrixInfo &lhs_info, const GEMMRHSMatrixInfo &rhs_info, const GEMMReshapeInfo &gemm_info)
168 {
169  ARM_COMPUTE_ERROR_ON_NULLPTR(src0, src1, dst);
170 
171  ARM_COMPUTE_ERROR_THROW_ON(validate_arguments(src0, src1, dst, lhs_info, rhs_info, gemm_info));
172 
173  _reinterpret_input_as_3d = gemm_info.reinterpret_input_as_3d();
174  _reinterpret_output_as_3d = (gemm_info.depth_output_gemm3d() != 0);
175  _use_dummy_work_items = preferred_dummy_work_items_support(CLKernelLibrary::get().get_device());
176 
177  // We still need padding on the X dimension for the RHS matrix
178  auto padding_info = get_padding_info({ src0, dst });
179 
180  // In case both input and dst have to be reinterpreted as 3D tensors,
181  // force reinterpret_input_as_3d and reinterpret_dst_as_3d to be false.
182  if(_reinterpret_input_as_3d == _reinterpret_output_as_3d)
183  {
184  _reinterpret_input_as_3d = false;
185  _reinterpret_output_as_3d = false;
186  }
187 
188  // Check if we need to slide the matrix B
189  const unsigned int num_dimensions_src0 = src0->num_dimensions();
190  _slide_matrix_b = (src1->num_dimensions() >= num_dimensions_src0);
191 
192  ElementsProcessed num_elements_processed{};
193 
194  // Configure kernel window
195  auto win_config = validate_and_configure_window(src0, src1, dst, lhs_info, rhs_info, gemm_info, num_elements_processed);
196  ARM_COMPUTE_ERROR_THROW_ON(win_config.first);
197  ICLKernel::configure_internal(win_config.second);
198 
199  // If _reinterpret_input_as_3d = _reinterpret_output_as_3d = true,
200  // we will dispatch a batched-GEMM to reduce the complexity of the address calculation within the OpenCL kernel.
201  // This means that the actual m used by the kernel is given by dst->info()->dimension(1) and not by gemm_info.m
202  const unsigned int internal_m = _reinterpret_output_as_3d ? gemm_info.m() : dst->dimension(1);
203  // Calculate partial (store instead of load) M0 and partial N0 for the partial blocks at the end of a row/column if any. This is to avoid padding.
204  const unsigned int partial_store_m0 = internal_m % lhs_info.m0;
205  const unsigned int partial_store_n0 = gemm_info.n() % rhs_info.n0;
206 
207  // Shrink M0 to be always <= M (internal_m) to prevent out-of-bounds reads.
208  // NOTE: This might have implications on heuristics and performance
209  const unsigned int internal_m0 = std::min(internal_m, lhs_info.m0);
210 
211  // Create build options
212  CLBuildOptions build_opts;
213  build_opts.add_option_if(_reinterpret_input_as_3d, "-DREINTERPRET_INPUT_AS_3D");
214  build_opts.add_option_if(_reinterpret_output_as_3d, "-DREINTERPRET_OUTPUT_AS_3D");
215  build_opts.add_option_if(_reinterpret_input_as_3d || _reinterpret_output_as_3d, "-DHEIGHT_GEMM3D=" + support::cpp11::to_string(dst->dimension(1)));
216  build_opts.add_option_if(_reinterpret_input_as_3d || _reinterpret_output_as_3d, "-DDEPTH_GEMM3D=" + support::cpp11::to_string(dst->dimension(2)));
217  build_opts.add_option_if(!_slide_matrix_b, "-DMATRIX_B_DEPTH=" + support::cpp11::to_string(src1->dimension(2)));
218  build_opts.add_option_if(_use_dummy_work_items, "-DDUMMY_WORK_ITEMS");
219  build_opts.add_option("-DM=" + support::cpp11::to_string(src0->dimension(1)));
220  build_opts.add_option("-DN=" + support::cpp11::to_string(gemm_info.n()));
221  build_opts.add_option("-DK=" + support::cpp11::to_string(gemm_info.k()));
222  build_opts.add_option("-DM0=" + support::cpp11::to_string(internal_m0));
223  build_opts.add_option("-DN0=" + support::cpp11::to_string(rhs_info.n0));
224  build_opts.add_option("-DK0=" + support::cpp11::to_string(rhs_info.k0));
225  build_opts.add_option("-DDATA_TYPE=" + get_cl_type_from_data_type(src0->data_type()));
226  build_opts.add_option("-DACC_DATA_TYPE=" + get_cl_dot8_acc_type_from_data_type(src0->data_type()));
227  build_opts.add_option("-DPARTIAL_STORE_M0=" + support::cpp11::to_string(partial_store_m0));
228  build_opts.add_option("-DPARTIAL_STORE_N0=" + support::cpp11::to_string(partial_store_n0));
229  std::string kernel_name("gemmlowp_mm_native");
230 
231  // Create kernel
232  _kernel = create_kernel(compile_context, kernel_name, build_opts.options());
233 
234  // Set config_id for enabling LWS tuning
235  _config_id = kernel_name;
236  _config_id += "_";
237  _config_id += dot8_supported(CLKernelLibrary::get().get_device()) ? "_dot8" : "";
238  _config_id += "_";
239  _config_id += (_reinterpret_input_as_3d ? "3di_" : "");
240  _config_id += (_reinterpret_output_as_3d ? "3do_" : "");
241  _config_id += support::cpp11::to_string(dst->dimension(1));
242  _config_id += "_";
243  _config_id += support::cpp11::to_string(dst->dimension(0));
244  _config_id += "_";
245  _config_id += support::cpp11::to_string(gemm_info.k());
246  _config_id += "_";
247  _config_id += support::cpp11::to_string(dst->dimension(2));
248  _config_id += "_";
249  _config_id += support::cpp11::to_string(lhs_info.m0);
250  _config_id += "_";
251  _config_id += support::cpp11::to_string(rhs_info.n0);
252  _config_id += "_";
253  _config_id += support::cpp11::to_string(lhs_info.k0);
254 
256 }
257 
259  const GEMMRHSMatrixInfo &rhs_info, const GEMMReshapeInfo &gemm_info)
260 {
261  ElementsProcessed num_elements_processed{};
262  ARM_COMPUTE_RETURN_ON_ERROR(validate_arguments(src0, src1, dst, lhs_info, rhs_info, gemm_info));
264  src1->clone().get(),
265  dst->clone().get(),
266  lhs_info,
267  rhs_info,
268  gemm_info,
269  num_elements_processed)
270  .first);
271 
272  return Status{};
273 }
274 
275 void ClGemmLowpMatrixMultiplyNativeKernel::run_op(ITensorPack &tensors, const Window &window, cl::CommandQueue &queue)
276 {
279 
280  const auto src0 = utils::cast::polymorphic_downcast<const ICLTensor *>(tensors.get_const_tensor(TensorType::ACL_SRC_0));
281  const auto src1 = utils::cast::polymorphic_downcast<const ICLTensor *>(tensors.get_const_tensor(TensorType::ACL_SRC_1));
282  auto dst = utils::cast::polymorphic_downcast<ICLTensor *>(tensors.get_tensor(TensorType::ACL_DST));
283 
284  if(src1->info()->num_dimensions() < 3)
285  {
286  // The stride_z for matrix B must be zero if we do not slice
287  ARM_COMPUTE_ERROR_ON(src1->info()->strides_in_bytes()[3] != 0);
288  }
289 
291  Window slice_matrix_b = slice;
292 
293  slice_matrix_b.set(Window::DimX, Window::Dimension(0, 1, 1));
294  slice_matrix_b.set(Window::DimY, Window::Dimension(0, 1, 1));
295 
296  if(_reinterpret_input_as_3d)
297  {
298  // Pass bottom paddings to the kernel if the input has to be reinterpreted as 3D tensor
299  const unsigned int idx0 = 3 * num_arguments_per_2D_tensor() + 3;
300  const unsigned int total_cross_plane_pad = src0->info()->padding().top + src0->info()->padding().bottom;
301  _kernel.setArg<cl_uint>(idx0, static_cast<unsigned int>(total_cross_plane_pad));
302  }
303 
304  if(_reinterpret_output_as_3d)
305  {
306  // Pass bottom paddings to the kernel if the output has to be reinterpreted as 3D tensor
307  const unsigned int idx0 = 3 * num_arguments_per_2D_tensor() + 3 + (_reinterpret_input_as_3d ? 1 : 0);
308  const unsigned int total_cross_plane_pad = dst->info()->padding().top + dst->info()->padding().bottom;
309  _kernel.setArg<cl_uint>(idx0, static_cast<unsigned int>(total_cross_plane_pad));
310  }
311 
312  do
313  {
314  Window slice_b = slice;
315  // Don't slice matrix B along the z dimension if matrix B has just 2 dimensions and matrix A more than 2
316  // This scenario can happen when the matrix multiplication is used to perform a convolution operation
317  if(!_slide_matrix_b)
318  {
319  slice_b = slice_matrix_b;
320  }
321 
322  unsigned int idx = 0;
323  add_2D_tensor_argument(idx, src0, slice);
324  add_2D_tensor_argument(idx, src1, slice_b);
325  add_2D_tensor_argument(idx, dst, slice);
326  _kernel.setArg<cl_uint>(idx++, static_cast<unsigned int>(src0->info()->strides_in_bytes()[2]));
327  _kernel.setArg<cl_uint>(idx++, static_cast<unsigned int>(src1->info()->strides_in_bytes()[2]));
328  _kernel.setArg<cl_uint>(idx++, static_cast<unsigned int>(dst->info()->strides_in_bytes()[2]));
329  enqueue(queue, *this, slice, lws_hint(), _use_dummy_work_items);
330  }
331  while(window.slide_window_slice_3D(slice));
332 }
333 } // namespace kernels
334 } // namespace opencl
335 } // namespace arm_compute
virtual size_t num_dimensions() const =0
The number of dimensions of the tensor (rank)
Window calculate_max_window(const ValidRegion &valid_region, const Steps &steps, bool skip_border, BorderSize border_size)
const Window & window() const
The maximum window the kernel can be executed on.
Definition: IKernel.cpp:28
bool dot8_supported(const cl::Device &device)
Helper function to check whether the cl_arm_integer_dot_product_int8 extension is supported...
Definition: CLHelpers.cpp:241
virtual size_t dimension(size_t index) const =0
Return the size of the requested dimension.
void enqueue(cl::CommandQueue &queue, ICLKernel &kernel, const Window &window, const cl::NDRange &lws_hint=CLKernelLibrary::get().default_ndrange(), bool use_dummy_work_items=false)
Add the kernel to the command queue with the given window.
Definition: ICLKernel.cpp:32
const StringSet & options() const
Gets the current options list set.
cl::NDRange lws_hint() const
Return the Local-Workgroup-Size hint.
Definition: ICLKernel.h:384
bool preferred_dummy_work_items_support(const cl::Device &device)
Helper function to check if "dummy work-items" are preferred to have a power of two NDRange In case d...
Definition: CLHelpers.cpp:367
GEMM reshape information class.
Definition: Types.h:2092
std::string get_cl_dot8_acc_type_from_data_type(const DataType &dt)
Translates a tensor data type to the appropriate OpenCL dot8 accumulator type.
Definition: CLHelpers.cpp:175
#define ARM_COMPUTE_RETURN_ON_ERROR(status)
Checks if a status contains an error and returns it.
Definition: Error.h:204
std::string to_string(T &&value)
Convert integer and float values to string.
virtual DataType data_type() const =0
Data type used for each element of the tensor.
TensorShape compute_mm_shape(const ITensorInfo &input0, const ITensorInfo &input1, bool is_interleaved_transposed, const GEMMReshapeInfo &reshape_info)
Calculate the matrix multiplication output shape of two tensors.
#define ARM_COMPUTE_ERROR_ON(cond)
If the condition is true then an error message is printed and an exception thrown.
Definition: Error.h:466
static CLKernelLibrary & get()
Access the KernelLibrary singleton.
GEMM LHS (Left Hand Side) matrix information.
Definition: Types.h:2236
Store the tensor&#39;s metadata.
Definition: ITensorInfo.h:40
#define ARM_COMPUTE_ERROR_THROW_ON(status)
Definition: Error.h:455
Describe one of the image&#39;s dimensions with a start, end and step.
Definition: Window.h:79
Manages all the OpenCL kernels compilation and caching, provides accessors for the OpenCL Context...
Status class.
Definition: Error.h:52
Status validate_arguments(const ITensorInfo *src, const ITensorInfo *weights, const ITensorInfo *dst, const PadStrideInfo &conv_info)
#define ARM_COMPUTE_RETURN_ERROR_ON(cond)
If the condition is true, an error is returned.
Definition: Error.h:296
Copyright (c) 2017-2022 Arm Limited.
#define ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(...)
Definition: Validate.h:159
1 channel, 1 S32 per channel
void add_option(std::string option)
Adds option to the existing build option list.
int n() const
Number of matrix B columns.
Definition: Types.h:2130
const ITensor * get_const_tensor(int id) const
Get constant tensor of a given id.
Definition: ITensorPack.cpp:54
unsigned int k0
Number of partial accumulations performed by the matrix multiplication.
Definition: Types.h:2259
cl::Kernel create_kernel(const CLCompileContext &ctx, const std::string &kernel_name, const std::set< std::string > &build_opts=std::set< std::string >())
Creates an opencl kernel using a compile context.
Definition: CLHelpers.cpp:404
static constexpr size_t DimX
Alias for dimension 0 also known as X dimension.
Definition: Window.h:43
bool update_window_and_padding(Window &win, Ts &&... patterns)
Update window and padding size for each of the access patterns.
Definition: WindowHelpers.h:46
#define ARM_COMPUTE_UNUSED(...)
To avoid unused variables warnings.
Definition: Error.h:152
GEMM RHS (Right Hand Side) matrix information.
Definition: Types.h:2251
auto ceil_to_multiple(S value, T divisor) -> decltype(((value+divisor - 1)/divisor) *divisor)
Computes the smallest number larger or equal to value that is a multiple of divisor.
Definition: Utils.h:71
unsigned int n0
Number of columns processed by the matrix multiplication.
Definition: Types.h:2258
quantized, asymmetric fixed-point 8-bit number unsigned
void configure(const CLCompileContext &compile_context, const ITensorInfo *src0, ITensorInfo *src1, ITensorInfo *dst, const GEMMLHSMatrixInfo &lhs_info, const GEMMRHSMatrixInfo &rhs_info, const GEMMReshapeInfo &gemm_info)
Initialise the kernel&#39;s input and dst.
std::string get_cl_type_from_data_type(const DataType &dt)
Translates a tensor data type to the appropriate OpenCL type.
Definition: CLHelpers.cpp:39
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...
virtual std::unique_ptr< T > clone() const =0
Provide a clone of the current object of class T.
void run_op(ITensorPack &tensors, const Window &window, cl::CommandQueue &queue) override
Enqueue the OpenCL kernel to process the given window on the passed OpenCL command queue...
int k() const
Number of matrix A columns or matrix B rows.
Definition: Types.h:2138
void add_option_if(bool cond, std::string option)
Adds option if a given condition is true;.
void set(size_t dimension, const Dimension &dim)
Set the values of a given dimension.
Definition: Window.inl:49
static constexpr unsigned int num_arguments_per_2D_tensor()
Returns the number of arguments enqueued per 2D tensor object.
Definition: ICLKernel.h:306
bool slide_window_slice_3D(Window &slice) const
Slide the passed 3D window slice.
Definition: Window.h:349
#define ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(k)
Definition: Validate.h:915
bool has_padding_changed(const std::unordered_map< const ITensorInfo *, PaddingSize > &padding_map)
Check if the previously stored padding info has changed after configuring a kernel.
Definition: Utils.cpp:601
quantized, symmetric fixed-point 8-bit number
CLCompileContext class.
quantized, symmetric per channel fixed-point 8-bit number
static constexpr size_t DimY
Alias for dimension 1 also known as Y dimension.
Definition: Window.h:45
void add_2D_tensor_argument(unsigned int &idx, const ICLTensor *tensor, const Window &window)
Add the passed 2D tensor&#39;s parameters to the object&#39;s kernel&#39;s arguments starting from the index idx...
Definition: ICLKernel.h:203
std::pair< Status, Window > validate_and_configure_window(ITensorInfo *src, ITensorInfo *dst)
ITensor * get_tensor(int id)
Get tensor of a given id from the pac.
Definition: ITensorPack.cpp:64
#define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(...)
Definition: Validate.h:439
#define ARM_COMPUTE_CREATE_ERROR(error_code, msg)
Creates an error with a given message.
Definition: Error.h:159
int m() const
Number of matrix A rows.
Definition: Types.h:2122
static Status validate(const ITensorInfo *src0, const ITensorInfo *src1, const ITensorInfo *dst, const GEMMLHSMatrixInfo &lhs_info, const GEMMRHSMatrixInfo &rhs_info, const GEMMReshapeInfo &gemm_info)
Static function to check if given info will lead to a valid configuration.
#define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(...)
Definition: Validate.h:541
#define ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(t, c,...)
Definition: Validate.h:788
std::unordered_map< const ITensorInfo *, PaddingSize > get_padding_info(std::initializer_list< const ITensorInfo *> infos)
Stores padding information before configuring a kernel.
Definition: Utils.cpp:586
Wrapper to configure the Khronos OpenCL C++ header.
int depth_output_gemm3d() const
Depth (third dimension) of the output tensor to be used with the GEMM3D kernel.
Definition: Types.h:2165
#define ARM_COMPUTE_RETURN_ERROR_ON_MSG(cond, msg)
If the condition is true, an error is returned.
Definition: Error.h:244
Tensor packing service.
Definition: ITensorPack.h:39
#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)
Definition: Validate.h:157
unsigned int k0
Number of partial accumulations performed by the matrix multiplication.
Definition: Types.h:2244
unsigned int m0
Number of rows processed by the matrix multiplication.
Definition: Types.h:2243
quantized, asymmetric fixed-point 8-bit number signed
Window first_slice_window_3D() const
First 3D slice of the window.
Definition: Window.h:305
std::string kernel_name
bool reinterpret_input_as_3d() const
Flag which specifies if the input tensor has to be reinterpreted as 3D.
Definition: Types.h:2173
Describe a multidimensional execution window.
Definition: Window.h:39
Convolution using GEMM.
#define ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(f, s)
Definition: Validate.h:201
SimpleTensor< T > slice(const SimpleTensor< T > &src, Coordinates starts, Coordinates ends)