如何为单个函数设置 ICC 属性“fp-model precision”，以防止关联优化？

我正在实施卡汉求和 http://en.wikipedia.org/wiki/Kahan_summation_algorithm，在支持 gcc47、gcc48、clang33、icc13 和 icc14 编译的项目中。

作为该算法的一部分，我想禁用利用实数加法的关联性的优化。 （浮点运算不具有关联性。）

我想禁用这些优化仅在相关功能中。我已经弄清楚如何在 gcc 下执行此操作，使用''无联想数学'' https://gcc.gnu.org/onlinedocs/gcc/Optimize-Options.html#Optimize-Options属性。我怎样才能在 icc 或 clang 中做到这一点？我搜索过但没有运气。

class KahanSummation
{
  // GCC declaration
  void operator()(float observation) __attribute__((__optimize__("no-associative-math")))
  {
    // Kahan summation implementation
  }
};

其他 GCC 属性意味着no-associative-math are no-unsafe-math-optimizations or no-fast-math.

看着一个英特尔演示文稿（PDF，幻灯片 8） https://software.intel.com/sites/default/files/article/326703/fp-control-2012-08.pdf或其他或其他（PDF，幻灯片 11） http://www.nccs.nasa.gov/images/FloatingPoint_consistency.pdf，我想在ICC中设置“fp-model precision”，仅用于此功能。我关心的编译器是 ICC 13 和 ICC 14。

class KahanSummation
{
  // ICC or clang declaration
  void operator()(float observation) __attribute__((????))
  {
    // Kahan summation implementation
  }
};

__attribute__ is a GCC extension. Clang also supports that syntax to maintain some GCC compatibility, but the only optimization-related attribute it seems to support is optnone http://clang.llvm.org/docs/AttributeReference.html#optnone-clang-optnone, which turns off all optimizations. ICC has a few optimization pragmas https://software.intel.com/sites/products/documentation/doclib/iss/2013/compiler/cpp-lin/GUID-346BAAA5-CF2D-4A26-9194-CA840BFB34E5.htm, but nothing that does what you want to do, from what I can tell. appears to support https://software.intel.com/sites/products/documentation/doclib/iss/2013/compiler/cpp-lin/GUID-0B88D6DE-D06F-4C57-A313-11291B202C39.htm #pragma float_control for VC++ compatibility, although I can't find anything on exactly how that pragma is supposed to be used in ICC's documentation, so you'll have to use VC++ http://msdn.microsoft.com/en-us/library/45ec64h6.aspx's.

不过，您可以做的是在单独的翻译单元（即 cpp 文件）中定义所需的函数：

// Your header
class KahanSummation
{
  // Declaration
  void operator()(float observation);
};

// Separate cpp file - implements only this function
void KahanSummation::operator()(float observation)
{
    // Kahan summation implementation
}

然后，您可以使用需要使用的任何编译器选项来编译单独的文件，并将生成的目标文件链接到程序的其余部分。