c++ - C++ 中转换为 simd 类型是未定义行为吗？

Question

这个问题已经有答案了:

Is `reinterpret_cast`ing between hardware SIMD vector pointer and the corresponding type an undefined behavior? (2 个回答)

已关闭 3 年前。

在 simd 教程中，我发现了以下代码片段。

void simd(float* a, int N)                                                                                                                                                                                        
{                      
// We assume N % 4 == 0.                                                                                                                                                                                        
 int nb_iters = N / 4;                                                                                                                                                                                         
 __m128* ptr = reinterpret_cast<__m128*>(a); // (*)                                                                                                                                                                                 

 for (int i = 0; i < nb_iters; ++i, ++ptr, a += 4)                                                                                                                                                              
     _mm_store_ps(a, _mm_sqrt_ps(*ptr));                                                                                                                                                                          
}   

现在我的问题是，带有 (*) 的行是否未定义行为？由于以下规范来自 ( https://en.cppreference.com/w/cpp/language/reinterpret_cast )

Whenever an attempt is made to read or modify the stored value of an object of type DynamicType through a glvalue of type AliasedType, the behavior is undefined unless one of the following is true:

• AliasedType and DynamicType are similar.
• AliasedType is the (possibly cv-qualified) signed or unsigned variant of DynamicType.
• AliasedType is std::byte, (since C++17)char, or unsigned char: this permits examination of the object representation of any object as an array of bytes.

在这种情况下，如何防止未定义的行为？我知道我可以 std::memcopy，但性能损失会使 simd 无用，还是我错了？

Answer 1

编辑:请查看拷贝中的答案(和/或此处彼得的答案)。我下面写的内容在技术上是正确的，但在实践中并不真正相关。

<小时/>

是的，根据 C++ 标准，这将是未定义的行为。您的编译器仍可能将其作为扩展正确处理(因为 SIMD 类型和内在函数一开始就不属于 C++ 标准的一部分)。

为了安全、正确地执行此操作而不影响速度，您可以使用内部函数将 4 个 float 直接从内存加载到 128 位寄存器中:

__m128 reg = _mm_load_ps(a);

请参阅Intel Intrinsics Guide对于重要的对齐约束:

__m128 _mm_load_ps (float const* mem_addr)

Load 128-bits (composed of 4 packed single-precision (32-bit) floating-point elements) from memory into dst. mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.