Compute Library
 21.08
OMPScheduler.cpp
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2017-2021 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 
27 #include "arm_compute/core/Error.h"
29 #include "arm_compute/core/Utils.h"
30 #include <omp.h>
31 
32 namespace arm_compute
33 {
35  : _num_threads(omp_get_max_threads())
36 {
37 }
38 
39 unsigned int OMPScheduler::num_threads() const
40 {
41  return _num_threads;
42 }
43 
45 {
46  const unsigned int num_cores = omp_get_max_threads();
47  _num_threads = (num_threads == 0) ? num_cores : num_threads;
48 }
49 
50 void OMPScheduler::schedule(ICPPKernel *kernel, const Hints &hints)
51 {
52  ITensorPack tensors;
53  schedule_common(kernel, hints, kernel->window(), tensors);
54 }
55 
56 void OMPScheduler::schedule_op(ICPPKernel *kernel, const Hints &hints, const Window &window, ITensorPack &tensors)
57 {
58  ARM_COMPUTE_ERROR_ON_MSG(!kernel, "The child class didn't set the kernel");
60  "Dynamic scheduling is not supported in OMPScheduler");
61 
62  const Window &max_window = window;
63  const unsigned int num_iterations = max_window.num_iterations(hints.split_dimension());
64  const unsigned int num_threads = std::min(num_iterations, _num_threads);
65 
66  if(!kernel->is_parallelisable() || num_threads == 1)
67  {
69  info.cpu_info = &cpu_info();
70  kernel->run_op(tensors, max_window, info);
71  }
72  else
73  {
74  const unsigned int num_windows = num_threads;
75  std::vector<IScheduler::Workload> workloads(num_windows);
76  for(unsigned int t = 0; t < num_windows; t++)
77  {
78  //Capture 't' by copy, all the other variables by reference:
79  workloads[t] = [t, &hints, &max_window, &num_windows, &kernel, &tensors](const ThreadInfo & info)
80  {
81  Window win = max_window.split_window(hints.split_dimension(), t, num_windows);
82  win.validate();
83  kernel->run_op(tensors, win, info);
84  };
85  }
86  run_workloads(workloads);
87  }
88 }
89 #ifndef DOXYGEN_SKIP_THIS
90 void OMPScheduler::run_workloads(std::vector<arm_compute::IScheduler::Workload> &workloads)
91 {
92  const unsigned int num_threads = std::min(_num_threads, static_cast<unsigned int>(workloads.size()));
93  if(num_threads < 1)
94  {
95  return;
96  }
97 
99  info.cpu_info = &cpu_info();
100  info.num_threads = num_threads;
101  #pragma omp parallel firstprivate(info) num_threads(num_threads)
102  {
103  const int tid = omp_get_thread_num();
104  info.thread_id = tid;
105  workloads[tid](info);
106  }
107 }
108 #endif /* DOXYGEN_SKIP_THIS */
109 } // namespace arm_compute
const Window & window() const
The maximum window the kernel can be executed on.
Definition: IKernel.cpp:28
void schedule_op(ICPPKernel *kernel, const Hints &hints, const Window &window, ITensorPack &tensors) override
Multithread the execution of the passed kernel if possible.
OMPScheduler()
Constructor.
Common interface for all kernels implemented in C++.
Definition: ICPPKernel.h:38
void set_num_threads(unsigned int num_threads) override
Sets the number of threads the scheduler will use to run the kernels.
Window split_window(size_t dimension, size_t id, size_t total) const
Split a window into a set of sub windows along a given dimension.
Definition: Window.inl:189
virtual void run_op(ITensorPack &tensors, const Window &window, const ThreadInfo &info)
Execute the kernel on the passed window.
Definition: ICPPKernel.h:86
CPUInfo & cpu_info()
Get CPU info.
Definition: IScheduler.cpp:40
void validate() const
Will validate all the window&#39;s dimensions&#39; values when asserts are enabled.
Definition: Window.inl:173
constexpr size_t num_iterations(size_t dimension) const
Return the number of iterations needed to iterate through a given dimension.
Definition: Window.inl:182
OpenMP scheduler (Alternative to the CPPScheduler).
Copyright (c) 2017-2021 Arm Limited.
Split the workload dynamically using a bucket system.
StrategyHint strategy() const
Return the prefered strategy to use to split workload.
Definition: IScheduler.h:116
virtual bool is_parallelisable() const
Indicates whether or not the kernel is parallelisable.
Definition: IKernel.cpp:41
#define ARM_COMPUTE_ERROR_ON_MSG(cond, msg)
Definition: Error.h:456
void schedule(ICPPKernel *kernel, const Hints &hints) override
Multithread the execution of the passed kernel if possible.
ScaleKernelInfo info(interpolation_policy, default_border_mode, PixelValue(), sampling_policy, false)
Information about executing thread and CPU.
Definition: CPPTypes.h:158
unsigned int num_threads() const override
Returns the number of threads that the OMPScheduler has in its pool.
Tensor packing service.
Definition: ITensorPack.h:39
unsigned int split_dimension() const
Return the prefered split dimension.
Definition: IScheduler.h:96
const CPUInfo * cpu_info
Definition: CPPTypes.h:162
Describe a multidimensional execution window.
Definition: Window.h:39