Clang __m128 won't work with DirectXMath

[walbourn]

In a few places in the DirectXMath library, individual vector element access is done by V.m128_f32[i] / V.n128_u32[i] and V.m128_u32[i] / V.n128_u32[i] which lets the compiler do the 'spill' to memory automatically. Clang doesn't support this MSVC/Intel compiler-specific behavior in how __m128 is defined so won't build.

This impacts the following functions:

• XMVectorGetByIndex

• XMVectorGetByIndexPtr

• XMVectorGetIntByIndex

• XMVectorGetIntByIndexPtr

• XMVectorSetByIndex

• XMVectorSetByIndexPtr

• XMVectorSetIntByIndex

[walbourn]

The portable solution is to use an explicit union:

XMVECTORF32 U;
U.v = V;
return U.f[i];

Which should result in the same codegen on Visual C++. XMVectorSetIntByIndexPtr is already using this pattern.

See Get member of __m128 by index?

[walbourn]

Fixed in master branch. See this commit

Will be included in a future release of DirectXMath