Files
igl/external/embree/kernels/algorithms/parallel_for.h
T

90 lines
3.5 KiB
C++
Executable File

// ======================================================================== //
// Copyright 2009-2014 Intel Corporation //
// //
// Licensed under the Apache License, Version 2.0 (the "License"); //
// you may not use this file except in compliance with the License. //
// You may obtain a copy of the License at //
// //
// http://www.apache.org/licenses/LICENSE-2.0 //
// //
// Unless required by applicable law or agreed to in writing, software //
// distributed under the License is distributed on an "AS IS" BASIS, //
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. //
// See the License for the specific language governing permissions and //
// limitations under the License. //
// ======================================================================== //
#pragma once
#include "range.h"
namespace embree
{
template<typename Index, typename Func>
class ParallelForTask
{
public:
__forceinline ParallelForTask (const Index taskCount, const Func& func)
: func(func)
{
#if 0
for (size_t taskIndex=0; taskIndex<taskCount; taskIndex++)
func(taskIndex);
#else
if (taskCount == 0) return;
else if (taskCount == 1) func(0);
else LockStepTaskScheduler::instance()->dispatchTaskSet(task_set,this,taskCount);
#endif
}
static void task_set(void* data, const size_t threadIndex, const size_t threadCount, const size_t taskIndex, const size_t taskCount) {
((ParallelForTask*)data)->func(taskIndex);
}
private:
const Func& func;
};
/* simple parallel_for without range optimization (similar to a task set) */
template<typename Index, typename Func>
__forceinline void parallel_for( const Index N, const Func& func)
{
ParallelForTask<Index,Func>(N,func);
}
/* sequential for with range optimization */
template<typename Index, typename Func>
__forceinline void sequential_for( const Index first, const Index last, const Func& func)
{
func(range<Index>(first,last));
}
/* sequential for with range optimization and minimal granularity per thread */
template<typename Index, typename Func>
__forceinline void sequential_for( const Index first, const Index last, const Index minStepSize, const Func& func)
{
func(range<Index>(first,last));
}
/* parallel for with range optimization */
template<typename Index, typename Func>
__forceinline void parallel_for( const Index first, const Index last, const Index minStepSize, const Func& func)
{
size_t taskCount = (last-first+minStepSize-1)/minStepSize;
if (taskCount > 1) taskCount = min(taskCount,LockStepTaskScheduler::instance()->getNumThreads());
parallel_for(taskCount, [&](const size_t taskIndex) {
const size_t k0 = first+(taskIndex+0)*(last-first)/taskCount;
const size_t k1 = first+(taskIndex+1)*(last-first)/taskCount;
func(range<Index>(k0,k1));
});
}
/* parallel for with range optimization and minimal granularity per thread */
template<typename Index, typename Func>
__forceinline void parallel_for( const Index first, const Index last, const Func& func)
{
parallel_for(first,last,(Index)1,func);
}
}