如何：使用 Alloc 和 Free 提高内存性能
本文档演示如何使用 concurrency::Alloc 和 concurrency::Free 函数提高内存性能。 比较为三种不同类型（每种类型都指定 new 和 delete 运算符）并行地颠倒数组元素所需的时间。
多个线程频繁地同时调用 Alloc 和 Free 时，Alloc 和 Free 函数最有用。 运行时为每个线程保存单独的内存缓存；因此，运行时无需使用锁或内存屏障即可管理内存。
示例：指定 new 和 delete 运算符的类型
下面的示例演示三种类型，每一种都指定 new 和 delete 运算符。 new_delete 类使用 new 和 delete 全局运算符，malloc_free 类使用 C Runtime 的 malloc 和 free 函数，而 Alloc_Free 类使用并发运行时的 Alloc 和 Free 函数。
// A type that defines the new and delete operators. These operators
// call the global new and delete operators, respectively.
class new_delete
{
public:
static void* operator new(size_t size)
{
return ::operator new(size);
}
static void operator delete(void *p)
{
return ::operator delete(p);
}
int _data;
};
// A type that defines the new and delete operators. These operators
// call the C Runtime malloc and free functions, respectively.
class malloc_free
{
public:
static void* operator new(size_t size)
{
return malloc(size);
}
static void operator delete(void *p)
{
return free(p);
}
int _data;
};
// A type that defines the new and delete operators. These operators
// call the Concurrency Runtime Alloc and Free functions, respectively.
class Alloc_Free
{
public:
static void* operator new(size_t size)
{
return Alloc(size);
}
static void operator delete(void *p)
{
return Free(p);
}
int _data;
};
示例：swap 和 reverse_array 函数
下面的示例显示 swap 和 reverse_array 函数。 swap 函数在指定的索引处交换数组的内容。 它为临时变量分配堆中的内存。 reverse_array 函数创建一个大数组，并计算并行地多次颠倒该数组所需的时间。
// Exchanges the contents of a[index1] with a[index2].
template<class T>
void swap(T* a, int index1, int index2)
{
// For illustration, allocate memory from the heap.
// This is useful when sizeof(T) is large.
T* temp = new T;
*temp = a[index1];
a[index1] = a[index2];
a[index2] = *temp;
delete temp;
}
// Computes the time that it takes to reverse the elements of a
// large array of the specified type.
template <typename T>
__int64 reverse_array()
{
const int size = 5000000;
T* a = new T[size];
__int64 time = 0;
const int repeat = 11;
// Repeat the operation several times to amplify the time difference.
for (int i = 0; i < repeat; ++i)
{
time += time_call([&] {
parallel_for(0, size/2, [&](int index)
{
swap(a, index, size-index-1);
});
});
}
delete[] a;
return time;
}
示例：wmain 函数
以下示例演示 wmain 函数，该函数计算 reverse_array 函数作用于 new_delete、malloc_free 和 Alloc_Free 类型所需的时间，这三种类型都使用不同的内存分配方案。
int wmain()
{
// Compute the time that it takes to reverse large arrays of
// different types.
// new_delete
wcout << L"Took " << reverse_array<new_delete>()
<< " ms with new/delete." << endl;
// malloc_free
wcout << L"Took " << reverse_array<malloc_free>()
<< " ms with malloc/free." << endl;
// Alloc_Free
wcout << L"Took " << reverse_array<Alloc_Free>()
<< " ms with Alloc/Free." << endl;
}
完整代码示例
之后提供了完整示例。
// allocators.cpp
// compile with: /EHsc
#include <windows.h>
#include <ppl.h>
#include <iostream>
using namespace concurrency;
using namespace std;
// Calls the provided work function and returns the number of milliseconds
// that it takes to call that function.
template <class Function>
__int64 time_call(Function&& f)
{
__int64 begin = GetTickCount();
f();
return GetTickCount() - begin;
}
// A type that defines the new and delete operators. These operators
// call the global new and delete operators, respectively.
class new_delete
{
public:
static void* operator new(size_t size)
{
return ::operator new(size);
}
static void operator delete(void *p)
{
return ::operator delete(p);
}
int _data;
};
// A type that defines the new and delete operators. These operators
// call the C Runtime malloc and free functions, respectively.
class malloc_free
{
public:
static void* operator new(size_t size)
{
return malloc(size);
}
static void operator delete(void *p)
{
return free(p);
}
int _data;
};
// A type that defines the new and delete operators. These operators
// call the Concurrency Runtime Alloc and Free functions, respectively.
class Alloc_Free
{
public:
static void* operator new(size_t size)
{
return Alloc(size);
}
static void operator delete(void *p)
{
return Free(p);
}
int _data;
};
// Exchanges the contents of a[index1] with a[index2].
template<class T>
void swap(T* a, int index1, int index2)
{
// For illustration, allocate memory from the heap.
// This is useful when sizeof(T) is large.
T* temp = new T;
*temp = a[index1];
a[index1] = a[index2];
a[index2] = *temp;
delete temp;
}
// Computes the time that it takes to reverse the elements of a
// large array of the specified type.
template <typename T>
__int64 reverse_array()
{
const int size = 5000000;
T* a = new T[size];
__int64 time = 0;
const int repeat = 11;
// Repeat the operation several times to amplify the time difference.
for (int i = 0; i < repeat; ++i)
{
time += time_call([&] {
parallel_for(0, size/2, [&](int index)
{
swap(a, index, size-index-1);
});
});
}
delete[] a;
return time;
}
int wmain()
{
// Compute the time that it takes to reverse large arrays of
// different types.
// new_delete
wcout << L"Took " << reverse_array<new_delete>()
<< " ms with new/delete." << endl;
// malloc_free
wcout << L"Took " << reverse_array<malloc_free>()
<< " ms with malloc/free." << endl;
// Alloc_Free
wcout << L"Took " << reverse_array<Alloc_Free>()
<< " ms with Alloc/Free." << endl;
}
对于具有四个处理器的计算机，本示例生成以下示例输出。
Took 2031 ms with new/delete.
Took 1672 ms with malloc/free.
Took 656 ms with Alloc/Free.
在此示例中，使用 Alloc 和 Free 函数的类型提供最佳的内存性能，因为我们对 Alloc 和 Free 函数进行了优化，以便频繁地分配和释放来自多个线程的内存块。
编译代码
复制示例代码，并将它粘贴到 Visual Studio 项目中，或粘贴到名为 allocators.cpp 的文件中，再在 Visual Studio 命令提示符窗口中运行以下命令。
cl.exe /EHsc allocators.cpp