Compute Library
 22.05
ClGemmMatrixMultiplyReshapedKernel.cpp
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2018-2021 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 #include "src/core/CL/CLUtils.h"
36 #include "src/core/CL/CLValidate.h"
42 #include "support/Cast.h"
43 #include "support/StringSupport.h"
44 
45 namespace arm_compute
46 {
47 namespace opencl
48 {
49 namespace kernels
50 {
51 namespace
52 {
53 using ElementsProcessed = Steps;
54 
55 const auto post_op_utils = experimental::PostOpCLKernelUtils(
56 {
57  // PostOp sequence -> {Kernel Postfix, PostOp Slots}
58  { {}, { "", {} } },
59  { { experimental::PostOpType::Activation }, { "", { 1 } } },
60 
61  { { experimental::PostOpType::Eltwise_Add }, { "_post_act_eltwise_op_act", { 2 } } },
62  { { experimental::PostOpType::Eltwise_PRelu }, { "_post_act_eltwise_op_act", { 2 } } },
63 
64  { { experimental::PostOpType::Activation, experimental::PostOpType::Eltwise_Add }, { "_post_act_eltwise_op_act", { 1, 2 } } },
65  { { experimental::PostOpType::Activation, experimental::PostOpType::Eltwise_PRelu }, { "_post_act_eltwise_op_act", { 1, 2 } } },
66 
67  { { experimental::PostOpType::Eltwise_Add, experimental::PostOpType::Activation }, { "_post_act_eltwise_op_act", { 2, 3 } } },
68  { { experimental::PostOpType::Eltwise_PRelu, experimental::PostOpType::Activation }, { "_post_act_eltwise_op_act", { 2, 3 } } },
69 
72 });
73 
74 Status validate_arguments(const ITensorInfo *src0, const ITensorInfo *src1, const ITensorInfo *src2, const ITensorInfo *dst, float alpha, float beta, const GEMMLHSMatrixInfo &lhs_info,
75  const GEMMRHSMatrixInfo &rhs_info,
76  const GEMMKernelInfo &gemm_info)
77 {
78  ARM_COMPUTE_UNUSED(alpha);
79  ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(src0, src1, dst);
83  ARM_COMPUTE_RETURN_ERROR_ON_MSG(src0->num_dimensions() > 4, "The number of dimensions for the LHS matrix must be <= 4");
84  ARM_COMPUTE_RETURN_ERROR_ON_MSG(src1->num_dimensions() > 3, "The number of dimensions for the RHS matrix must be <= 3");
85  ARM_COMPUTE_RETURN_ERROR_ON(lhs_info.k0 != rhs_info.k0);
86  ARM_COMPUTE_RETURN_ERROR_ON(lhs_info.transpose == rhs_info.transpose);
87  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");
88  ARM_COMPUTE_RETURN_ERROR_ON(lhs_info.k0 > 16);
89  ARM_COMPUTE_RETURN_ERROR_ON(lhs_info.m0 < 2 || lhs_info.m0 > 8);
90  ARM_COMPUTE_RETURN_ERROR_ON_MSG((lhs_info.transpose) && ((lhs_info.m0 & (lhs_info.m0 - 1)) && lhs_info.m0 != 3), "Only 2,3,4,8,16 are supported for m0");
91  ARM_COMPUTE_RETURN_ERROR_ON_MSG((rhs_info.transpose) && ((rhs_info.n0 & (rhs_info.n0 - 1)) && rhs_info.n0 != 3), "Only 2,3,4,8,16 are supported for n0");
92  ARM_COMPUTE_RETURN_ERROR_ON_MSG((gemm_info.reinterpret_input_as_3d || gemm_info.depth_output_gemm3d != 0) && (src2 != nullptr)
93  && (!gemm_info.broadcast_bias),
94  "Bias addition only supported with broadcast mode in case the input or dst has to be reinterpreted as 3D");
95  ARM_COMPUTE_RETURN_ERROR_ON_MSG(gemm_info.fp_mixed_precision && (src0->data_type() == DataType::F32), "Mixed precision only supported for F16 data type");
97  ARM_COMPUTE_RETURN_ERROR_ON_MSG(!post_op_utils.is_post_op_sequence_supported(gemm_info.post_ops), "The sequence of Post Ops is not supported");
98 
99  const unsigned int m = gemm_info.m;
100  const unsigned int n = gemm_info.n;
101  const unsigned int k = gemm_info.k;
102 
103  TensorShape tensor_shape0{ src0->tensor_shape() };
104  tensor_shape0.set(0, k);
105  tensor_shape0.set(1, m);
106 
107  TensorShape tensor_shape1{ src1->tensor_shape() };
108  tensor_shape1.set(0, n);
109  tensor_shape1.set(1, k);
110 
111  if(src2 != nullptr && !(helpers::float_ops::is_zero(beta)))
112  {
113  const unsigned int src2_dim0 = src2->dimension(0);
114  const unsigned int src2_dim1 = src2->dimension(1);
115 
117  if(gemm_info.broadcast_bias)
118  {
119  ARM_COMPUTE_RETURN_ERROR_ON_MSG((src2_dim1 != 1 || src2_dim0 != n), "Incorrect dimension of bias matrix which is to be broadcasted");
120  }
121  else
122  {
123  ARM_COMPUTE_RETURN_ERROR_ON_MSG((src2_dim0 != n || src2_dim1 != m), "Incorrect dimension of bias matrix");
124  }
125  }
126 
127  const TensorInfo tensor_info0 = src0->clone()->set_tensor_shape(tensor_shape0);
128  const TensorInfo tensor_info1 = src1->clone()->set_tensor_shape(tensor_shape1);
129 
130  const TensorInfo tensor_info_reshaped0 = src0->clone()->set_tensor_shape(misc::shape_calculator::compute_lhs_reshaped_shape(tensor_info0, lhs_info));
131  const TensorInfo tensor_info_reshaped1 = src1->clone()->set_tensor_shape(misc::shape_calculator::compute_rhs_reshaped_shape(tensor_info1, rhs_info));
132 
133  ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(src0, &tensor_info_reshaped0);
134  ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(src1, &tensor_info_reshaped1);
135 
136  if(dst->total_size() != 0)
137  {
138  const TensorInfo tensor_info_dst = dst->clone()->set_tensor_shape(misc::shape_calculator::compute_mm_shape(*src0, *src1, gemm_info));
139  ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(dst, &tensor_info_dst);
141  ARM_COMPUTE_RETURN_ERROR_ON_MSG(!post_op_utils.are_post_op_shapes_compliant(dst, gemm_info.post_ops), "The Post Op shapes are not compliant");
142  }
143 
144  return Status{};
145 }
146 
147 std::pair<Status, Window> validate_and_configure_window(ITensorInfo *src0, ITensorInfo *src1, ITensorInfo *src2, ITensorInfo *dst, const GEMMLHSMatrixInfo &lhs_info,
148  const GEMMRHSMatrixInfo &rhs_info,
149  const GEMMKernelInfo &gemm_info, ElementsProcessed &num_elements_processed)
150 {
151  ARM_COMPUTE_UNUSED(src0, src1, src2);
152  unsigned int &num_elems_processed_per_iteration_x = num_elements_processed[0];
153  unsigned int &num_elems_processed_per_iteration_y = num_elements_processed[1];
154  bool reinterpret_output_as_3d = gemm_info.depth_output_gemm3d != 0;
155 
156  TensorInfo tmp_info(*dst);
157 
158  if(reinterpret_output_as_3d)
159  {
160  // Since the dst tensor has to be reinterpreted as 3D and the execute window is based on a 2D GEMM,
161  // the window needs to be constructed on the 2D collapsed version of the tensor
162  TensorShape tmp_shape(dst->tensor_shape());
163  tmp_shape.collapse(2U, 1U);
164  tmp_info.set_tensor_shape(tmp_shape);
165  }
166 
167  // Configure kernel window
168  num_elems_processed_per_iteration_x = rhs_info.n0;
169  num_elems_processed_per_iteration_y = lhs_info.m0;
170 
171  Window win = calculate_max_window(tmp_info, Steps(num_elems_processed_per_iteration_x, num_elems_processed_per_iteration_y));
172 
173  // Collapse along the Z direction
174  // This collapse needs to be here in order to tune the Z dimension of LWS
175  Window collapsed = win;
176  const unsigned int dimension_to_collapse = std::min(static_cast<unsigned int>(dst->num_dimensions()), 2u);
177  collapsed = win.collapse(win, dimension_to_collapse);
178 
179  return std::make_pair(Status{}, collapsed);
180 }
181 } // namespace
182 
184 {
185  _type = CLKernelType::GEMM;
186 }
187 
189  const ITensorInfo *src0, const ITensorInfo *src1, const ITensorInfo *src2, ITensorInfo *dst, float alpha, float beta,
190  const GEMMLHSMatrixInfo &lhs_info, const GEMMRHSMatrixInfo &rhs_info, const GEMMKernelInfo &gemm_info)
191 {
192  ARM_COMPUTE_ERROR_ON_NULLPTR(src0, src1, dst);
193 
194  // dst tensor auto initialization if not yet initialized
195  auto_init_if_empty(*dst, src0->clone()->set_tensor_shape(misc::shape_calculator::compute_mm_shape(*src0, *src1, gemm_info)));
196 
197  ARM_COMPUTE_ERROR_THROW_ON(validate_arguments(src0, src1, src2, dst, alpha, beta, lhs_info, rhs_info, gemm_info));
198 
199  auto padding_info = get_padding_info({ src0, src1, src2, dst });
200  _reinterpret_output_as_3d = gemm_info.depth_output_gemm3d != 0;
201  _use_dummy_work_items = preferred_dummy_work_items_support(CLKernelLibrary::get().get_device());
202  _add_bias = src2 != nullptr;
203  _export_to_cl_image = rhs_info.export_to_cl_image;
204  _num_post_op_args = gemm_info.post_ops.total_num_arguments();
205 
206  // Check if we need to slide the matrix B
207  const unsigned int num_dimensions_src0 = src0->num_dimensions();
208  _slide_matrix_b = (src1->num_dimensions() >= num_dimensions_src0);
209 
210  ElementsProcessed num_elements_processed{};
211 
212  // Configure kernel window
213  auto win_config = validate_and_configure_window(src0->clone().get(),
214  src1->clone().get(),
215  (src2 != nullptr) ? src2->clone().get() : nullptr,
216  dst->clone().get(),
217  lhs_info,
218  rhs_info,
219  gemm_info,
220  num_elements_processed);
221  ARM_COMPUTE_ERROR_THROW_ON(win_config.first);
222  ICLKernel::configure_internal(win_config.second);
223 
224  const bool enable_mixed_precision = gemm_info.fp_mixed_precision;
225  const DataType data_type = src0->data_type();
226 
227  // 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.
228  const unsigned int internal_m = _reinterpret_output_as_3d ? gemm_info.m : dst->dimension(1);
229 
230  const unsigned int partial_store_m0 = internal_m % lhs_info.m0;
231  const unsigned int partial_store_n0 = gemm_info.n % rhs_info.n0;
232  _m = gemm_info.m;
233  _n = gemm_info.n;
234  _k = gemm_info.k;
235 
236  // Create build options
237  CLBuildOptions build_opts;
238  build_opts.add_option_if(!(helpers::float_ops::is_one(alpha)), "-DALPHA=" + float_to_string_with_full_precision(alpha));
239  build_opts.add_option_if(src2 != nullptr, "-DBETA=" + float_to_string_with_full_precision(beta));
240  build_opts.add_option_if(helpers::float_ops::is_one(beta), "-DUNIT_BETA");
241  build_opts.add_option_if(_reinterpret_output_as_3d, "-DREINTERPRET_OUTPUT_AS_3D");
242  build_opts.add_option_if(_reinterpret_output_as_3d, "-DHEIGHT_GEMM3D=" + support::cpp11::to_string(dst->dimension(1)));
243  build_opts.add_option_if(_reinterpret_output_as_3d, "-DDEPTH_GEMM3D=" + support::cpp11::to_string(dst->dimension(2)));
244  build_opts.add_option_if(gemm_info.broadcast_bias, "-DBROADCAST_BIAS");
245  build_opts.add_option_if(!_slide_matrix_b, "-DMATRIX_B_DEPTH=" + support::cpp11::to_string(src1->dimension(2)));
246  build_opts.add_option_if(lhs_info.interleave, "-DLHS_INTERLEAVE");
247  build_opts.add_option_if(rhs_info.interleave, "-DRHS_INTERLEAVE");
248  build_opts.add_option_if(lhs_info.transpose, "-DLHS_TRANSPOSE");
249  build_opts.add_option_if(_use_dummy_work_items, "-DDUMMY_WORK_ITEMS");
250  build_opts.add_option_if(enable_mixed_precision, "-DMIXED_PRECISION");
251  build_opts.add_option_if(rhs_info.export_to_cl_image, "-DOPENCL_IMAGE_SUPPORT");
252  build_opts.add_option("-DRHS_HEIGHT=" + support::cpp11::to_string(src1->dimension(1)));
253  build_opts.add_option("-DDATA_TYPE=" + get_cl_type_from_data_type(data_type));
254  build_opts.add_option("-DDATA_TYPE_ACCUMULATOR=" + (enable_mixed_precision ? get_cl_type_from_data_type(DataType::F32) : get_cl_type_from_data_type(data_type)));
255  build_opts.add_option("-DM0=" + support::cpp11::to_string(lhs_info.m0));
256  build_opts.add_option("-DN0=" + support::cpp11::to_string(rhs_info.n0));
257  build_opts.add_option("-DK0=" + support::cpp11::to_string(lhs_info.k0));
258  build_opts.add_option("-DV0=" + support::cpp11::to_string(lhs_info.v0));
259  build_opts.add_option("-DH0=" + support::cpp11::to_string(rhs_info.h0));
260  build_opts.add_option("-DPARTIAL_STORE_M0=" + support::cpp11::to_string(partial_store_m0));
261  build_opts.add_option("-DPARTIAL_STORE_N0=" + support::cpp11::to_string(partial_store_n0));
262  // If post_ops are used, then we disable the use of gemm_info.activation_info
263  if(gemm_info.post_ops.size() > 0)
264  {
265  post_op_utils.set_post_ops_cl_build_options(build_opts, gemm_info.post_ops);
266  }
267  else
268  {
269  build_opts.add_option_if(gemm_info.activation_info.enabled(), "-DACTIVATION_TYPE=" + lower_string(string_from_activation_func(gemm_info.activation_info.activation())));
270  build_opts.add_option_if(gemm_info.activation_info.enabled(), "-DA_VAL=" + float_to_string_with_full_precision(gemm_info.activation_info.a()));
271  build_opts.add_option_if(gemm_info.activation_info.enabled(), "-DB_VAL=" + float_to_string_with_full_precision(gemm_info.activation_info.b()));
272  }
273 
274  std::string kernel_name("gemm_mm_reshaped_");
275  kernel_name += lhs_info.transpose ? "lhs_t_" : "lhs_nt_";
276  kernel_name += rhs_info.transpose ? "rhs_t" : "rhs_nt";
277  kernel_name += rhs_info.export_to_cl_image ? "_texture" : "";
278  post_op_utils.set_post_ops_cl_kernel_name(kernel_name, gemm_info.post_ops);
279 
280  // A macro guard to compile ONLY the kernel of interest
281  build_opts.add_option("-D" + upper_string(kernel_name));
282 
283  // Create kernel
284  _kernel = create_kernel(compile_context, kernel_name, build_opts.options());
285 
286  // Set config_id for enabling LWS tuning
287  _config_id = kernel_name;
288  _config_id += "_";
289  _config_id += (_add_bias ? "add_bias_" : "");
290  _config_id += (gemm_info.broadcast_bias ? "broadcast_bias_" : "");
291  _config_id += (_reinterpret_output_as_3d ? "3do_" : "");
292  _config_id += (gemm_info.activation_info.enabled() ? "fused_activation_" : "");
293  _config_id += lower_string(string_from_data_type(src0->data_type()));
294  _config_id += "_";
295  _config_id += (enable_mixed_precision ? "mixed_precision_" : "");
296  _config_id += support::cpp11::to_string(dst->dimension(1));
297  _config_id += "_";
298  _config_id += support::cpp11::to_string(dst->dimension(0));
299  _config_id += "_";
300  _config_id += support::cpp11::to_string(gemm_info.k);
301  _config_id += "_";
302  _config_id += support::cpp11::to_string(dst->dimension(2));
303  _config_id += "_";
304  _config_id += support::cpp11::to_string(lhs_info.m0);
305  _config_id += "_";
306  _config_id += support::cpp11::to_string(rhs_info.n0);
307  _config_id += "_";
308  _config_id += support::cpp11::to_string(lhs_info.k0);
309  _config_id += "_";
310  _config_id += support::cpp11::to_string(lhs_info.v0);
311  _config_id += "_";
312  _config_id += support::cpp11::to_string(rhs_info.h0);
313  _config_id += "_";
314  _config_id += support::cpp11::to_string(lhs_info.interleave);
315  _config_id += "_";
316  _config_id += support::cpp11::to_string(rhs_info.interleave);
317 
319 }
320 
321 Status ClGemmMatrixMultiplyReshapedKernel::validate(const ITensorInfo *src0, const ITensorInfo *src1, const ITensorInfo *src2, const ITensorInfo *dst, float alpha, float beta,
322  const GEMMLHSMatrixInfo &lhs_info,
323  const GEMMRHSMatrixInfo &rhs_info, const GEMMKernelInfo &gemm_info)
324 {
325  ARM_COMPUTE_RETURN_ON_ERROR(validate_arguments(src0, src1, src2, dst, alpha, beta, lhs_info, rhs_info, gemm_info));
326  return Status{};
327 }
328 
329 void ClGemmMatrixMultiplyReshapedKernel::run_op(ITensorPack &tensors, const Window &window, cl::CommandQueue &queue)
330 {
333 
334  const auto src0 = utils::cast::polymorphic_downcast<const ICLTensor *>(tensors.get_const_tensor(TensorType::ACL_SRC_0));
335  const auto src1 = utils::cast::polymorphic_downcast<const ICLTensor *>(tensors.get_const_tensor(TensorType::ACL_SRC_1));
336  const auto src2 = utils::cast::polymorphic_downcast<const ICLTensor *>(tensors.get_const_tensor(TensorType::ACL_SRC_2));
337  auto dst = utils::cast::polymorphic_downcast<ICLTensor *>(tensors.get_tensor(TensorType::ACL_DST));
338 
339  ARM_COMPUTE_ERROR_ON_NULLPTR(src0, src1, dst);
340  ARM_COMPUTE_ERROR_ON(_add_bias && src2 == nullptr);
341 
342  if(src1->info()->num_dimensions() < 3)
343  {
344  // The stride_z for matrix B must be zero if we do not slice
345  ARM_COMPUTE_ERROR_ON(src1->info()->strides_in_bytes()[3] != 0);
346  }
347 
349  Window slice_matrix_b = slice;
350 
351  slice_matrix_b.set(Window::DimX, Window::Dimension(0, 1, 1));
352  slice_matrix_b.set(Window::DimY, Window::Dimension(0, 1, 1));
353 
354  const unsigned int total_cross_plane_pad = dst->info()->padding().top + dst->info()->padding().bottom;
355 
356  cl::Image2D src1_image2d;
357 
358  if(_export_to_cl_image)
359  {
360  const TensorShape shape2d(src1->info()->dimension(0) / 4, src1->info()->dimension(1) * src1->info()->dimension(2));
361  const size_t image_row_pitch = src1->info()->strides_in_bytes()[1];
362 
363  src1_image2d = create_image2d_from_buffer(CLKernelLibrary::get().context(), src1->cl_buffer(), shape2d, src1->info()->data_type(), image_row_pitch);
364  }
365 
366  do
367  {
368  Window slice_b = slice;
369  // Don't slice matrix B along the z dimension if matrix B has just 2 dimensions and matrix A more than 2
370  // This scenario can happen when the matrix multiplication is used to perform a convolution operation
371  if(!_slide_matrix_b)
372  {
373  slice_b = slice_matrix_b;
374  }
375 
376  unsigned int idx = 0;
377 
378  // LHS buffer
379  add_2D_tensor_argument(idx, src0, slice);
380 
381  // RHS buffer or RHS OpenCL image (_export_to_cl_image == true)
382  if(_export_to_cl_image)
383  {
384  _kernel.setArg(idx++, src1_image2d);
385  }
386  else
387  {
388  add_2D_tensor_argument(idx, src1, slice_b);
389  }
390 
391  // Bias buffer (_add_bias == true)
392  add_2D_tensor_argument_if(_add_bias, idx, src2, slice);
393 
394  // dst buffer
395  add_2D_tensor_argument(idx, dst, slice);
396 
397  // post op argument buffers
398  for(size_t i = 0; i < _num_post_op_args; ++i)
399  {
400  const auto post_op_arg = utils::cast::polymorphic_downcast<const ICLTensor *>(tensors.get_const_tensor(experimental::get_post_op_arg_type(i)));
401  add_2D_tensor_argument(idx, post_op_arg, slice);
402  }
403 
404  // LHS stride_z
405  _kernel.setArg<cl_uint>(idx++, static_cast<unsigned int>(src0->info()->strides_in_bytes()[2]));
406 
407  // RHS stride_z (not used if _export_to_cl_image == true)
408  _kernel.setArg<cl_uint>(idx++, static_cast<unsigned int>(src1->info()->strides_in_bytes()[2]));
409 
410  // Bias stride_z (if _add_bias == true)
411  if(_add_bias)
412  {
413  _kernel.setArg<cl_uint>(idx++, static_cast<unsigned int>(src2->info()->strides_in_bytes()[2]));
414  }
415 
416  // dst stride_z
417  _kernel.setArg<cl_uint>(idx++, static_cast<unsigned int>(dst->info()->strides_in_bytes()[2]));
418 
419  // post op argument stride_z
420  for(size_t i = 0; i < _num_post_op_args; ++i)
421  {
422  const auto post_op_arg = utils::cast::polymorphic_downcast<const ICLTensor *>(tensors.get_const_tensor(experimental::get_post_op_arg_type(i)));
423  _kernel.setArg<cl_uint>(idx++, static_cast<unsigned int>(post_op_arg->info()->strides_in_bytes()[2]));
424  }
425  // Cross-plan padding (if _reinterpret_output_as_3d = true)
426  if(_reinterpret_output_as_3d)
427  {
428  _kernel.setArg<cl_uint>(idx++, static_cast<unsigned int>(total_cross_plane_pad));
429  }
430 
431  // Pass m, n and k at runtime
432  _kernel.setArg<cl_int>(idx++, _m);
433  _kernel.setArg<cl_int>(idx++, _n);
434 
435  // K dimension (not used if _export_to_cl_image == true)
436  _kernel.setArg<cl_int>(idx++, _k);
437 
438  // Dispatch kernel
439  enqueue(queue, *this, slice, lws_hint(), _use_dummy_work_items);
440  }
441  while(window.slide_window_slice_3D(slice));
442 }
443 } // namespace kernels
444 } // namespace opencl
445 } // namespace arm_compute
virtual size_t num_dimensions() const =0
The number of dimensions of the tensor (rank)
bool is_one(float a, float epsilon=0.00001f)
Checks if the input floating point number is 1.0f checking if the difference is within a range define...
Definition: float_ops.h:97
Window calculate_max_window(const ValidRegion &valid_region, const Steps &steps, bool skip_border, BorderSize border_size)
#define ARM_COMPUTE_RETURN_ERROR_ON_F16_UNSUPPORTED(tensor)
Definition: CLValidate.h:35
const Window & window() const
The maximum window the kernel can be executed on.
Definition: IKernel.cpp:28
Shape of a tensor.
Definition: TensorShape.h:39
Descriptor used by the GEMM kernels.
void add_2D_tensor_argument_if(bool cond, 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:214
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.
unsigned int v0
Number of vertical blocks of size (m0xk0) stored on the same output row.
Definition: Types.h:2063
unsigned int depth_output_gemm3d
Depth of the output tensor in case is reinterpreted as 3D.
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:363
#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.
1 channel, 1 F32 per channel
#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
const std::string & string_from_activation_func(ActivationLayerInfo::ActivationFunction act)
Translates a given activation function to a string.
Definition: Utils.cpp:163
static CLKernelLibrary & get()
Access the KernelLibrary singleton.
GEMM LHS (Left Hand Side) matrix information.
Definition: Types.h:2054
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
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...
std::string lower_string(const std::string &val)
Lower a given string.
Definition: Utils.cpp:351
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
void configure(const ClCompileContext &compile_context, const ITensorInfo *src0, const ITensorInfo *src1, const ITensorInfo *src2, ITensorInfo *dst, float alpha, float beta, const GEMMLHSMatrixInfo &lhs_info, const GEMMRHSMatrixInfo &rhs_info, const GEMMKernelInfo &gemm_info)
Initialise the kernel&#39;s input and output.
bool interleave
True if the v0 (m0xk0) blocks have to be interleaved in the output row.
Definition: Types.h:2065
bool export_to_cl_image
True if the reshaped rhs has to be exported to cl_image.
Definition: Types.h:2081
Copyright (c) 2017-2022 Arm Limited.
1 channel, 1 F16 per channel
TensorType get_post_op_arg_type(size_t index)
Get post op argument TensorType from post op argument index in a flattened, ordered post op argument ...
Definition: PostOpUtils.h:79
#define ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(...)
Definition: Validate.h:159
void add_option(std::string option)
Adds option to the existing build option list.
bool transpose
True if the (m0xk0) block has to be transposed before been stored.
Definition: Types.h:2064
const ITensor * get_const_tensor(int id) const
Get constant tensor of a given id.
Definition: ITensorPack.cpp:54
static Status validate(const ITensorInfo *src0, const ITensorInfo *src1, const ITensorInfo *src2, const ITensorInfo *dst, float alpha, float beta, const GEMMLHSMatrixInfo &lhs_info, const GEMMRHSMatrixInfo &rhs_info, const GEMMKernelInfo &gemm_info)
Static function to check if given info will lead to a valid configuration.
std::string upper_string(const std::string &val)
Raise a given string to upper case.
Definition: Utils.cpp:358
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:391
const std::string & string_from_data_type(DataType dt)
Convert a data type identity into a string.
Definition: Utils.cpp:135
static constexpr size_t DimX
Alias for dimension 0 also known as X dimension.
Definition: Window.h:43
#define ARM_COMPUTE_UNUSED(...)
To avoid unused variables warnings.
Definition: Error.h:152
TensorShape compute_lhs_reshaped_shape(const ITensorInfo &a, const GEMMLHSMatrixInfo &lhs_info, bool reinterpret_input_as_3d=false)
Calculate the Left Hand Side matrix reshaped shape.
Status validate_image2d_support_on_rhs(const ITensorInfo &tensor_reshaped_info, const GEMMRHSMatrixInfo &rhs_info)
Utility function to validate the image2d OpenCL object support on the RHS reshaped matrix...
GEMM RHS (Right Hand Side) matrix information.
Definition: Types.h:2069
std::string float_to_string_with_full_precision(float val)
Create a string with the float in full precision.
Definition: Utils.h:1124
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
TensorShape compute_rhs_reshaped_shape(const ITensorInfo &a, const GEMMRHSMatrixInfo &rhs_info)
Calculate the Right Hand Side matrix reshaped shape.
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 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
bool slide_window_slice_3D(Window &slice) const
Slide the passed 3D window slice.
Definition: Window.h:348
#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
CLCompileContext class.
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_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.
bool is_zero(float a, float epsilon=0.00001f)
Checks if the input floating point number is 0.0f checking if the difference is within a range define...
Definition: float_ops.h:109
experimental::PostOpList< ITensorInfo * > post_ops
(EXPERIMENTAL_POST_OPS) Specifies a list of post ops to be fused after the main op.
#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:2062
cl::Image2D create_image2d_from_buffer(const cl::Context &ctx, const cl::Buffer &buffer, const TensorShape &shape2d, DataType data_type, size_t image_row_pitch)
Create a cl::Image2D object from an OpenCL buffer.
Definition: CLUtils.cpp:35
unsigned int m0
Number of rows processed by the matrix multiplication.
Definition: Types.h:2061
Window first_slice_window_3D() const
First 3D slice of the window.
Definition: Window.h:304
std::string kernel_name
DataType
Available data types.
Definition: Types.h:79
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)